·No text on this page.

·Synovial Fluid Changes
Joint Disease
in

·For approximately a quarter of a cen
tury THE COMMONWEALTH FUND
through its Division of Publica
tions sponsored edited produced
and distributed books and pamphlets
germane to its purposes and opera
tions as a philanthropic foundation
On July 1 1951 the Fund en
tered into an arrangement by which
HARVARD UNIVERSITY PRESS became
the publisher of Commonwealth
Fund books
assuming responsi
bility for their production and dis
tribution The Fund continues to
sponsor and edit its books and co
operates with the Press in all phases
of manufacture and distribution

·Synovial Fluid Changes
in Joint Disease
DI MARIAN W ROPES, MD
Associate Physician
Massachusetts General Hospital
Assistant Clinical Professor
of Medicine
Harvard Medical Sc
AND WALTER BAUER, MD
Chlef of Medical Services
Massachusetts General Hospital
Jackson Professor of Clinical Medicine
and Director of
Robert W Lovett Memorial Foundation
for the Study of Crippling Disease
Harvard Medical School
(ef
PUBLISHED FOR THE COMMONWEALTH FUND
by
HARVARD UNIVERSITY PRESS
CAMBRIDGE, MASSACHUSETTS
1953

·Copyright, 1953, by
The Commonwealth Fund
Published for The Commonwealth Fund
by Harvard University Press
PUBI ISHED IN GREAT BRITAIN
BY GEOFFREY CUMBERLEGE
OXFORD UNIVERSITY PRESS
LONDON
LIBRARY OF CONGRESS CATALOG CARD NUMBER 52 12252
PANTED IN TIE UNITED STATES OF AMERICA

·Preface The studies on synovial fluid reported in this volume were initiated twenty years ago in the hope that knowledge of the changes in the fluid in joint disease would aid in better understanding and differential diagnosis of the various arthritides This project has been part of a larger program, the aim of which has been to care for patients with joint disease and to gain knowledge of the anatomy and physiology of articular structures, the diseases which affect them and the individuals so afflicted This program, which has been camed out by the Arthritis Group at the Massachusetts General Hospital, has been supported by the Robert W Lovett Memorial Foundation for the Study of Crippling Disease, Harvard Medical School since its inception, and by the Commonwealth Fund and the Massachusetts Department of Public Health subsequently From the outset it was apparent that study of the origin and properties of normal synovial fruid was essential as a basis for comparison and evaluation of the changes produced by disease Normal fluid was obtained post mortem from patients who had had no signs or symptoms of joint disease Because sufficient fluid for complete analysis cannot be aspirated from normal human joints fluid was also obtained from normal cattle immediately after slaughter In addition to the acquisition of knowledge of the characteristics of normal synovial fluid, studies were undertaken to add to the existing information concerning the exchange of substances across the blood-joint barrier Fluids from pathological effusions were obtained from the large group of patients with joint disease seen by the Lovett Foundation Group on the wards and in the clinics of the Massachusetts ·Vi PREFACE General Hospital The alterations in the characteristics of the fluid with disease were found to reflect the degree of change in the synovial tissues and to be of value in diagnosis or to aid in interpreting alterations in the permeability of the blood joint barrier and in intra-articular metabolism On the basis of the findings in the first 800 fluids it became apparent that pathological effusions could be divided into two major groups the first, Group I, which contains fluids with relatively slight abnormalities, was found to include diseases in which the inflammation was trau matıc in orgin, whereas the second, Group II, includes rheuma toid arthritis and infectious arthritis In this volume the present concept of the nature of synovial fluid and the characteristics of normal fluid will be presented first The findings in respect to each individual characteristic of joint fluid under conditions of disease will then be described and related, when possible, to the underlying physiological alterations The abnormalities found in fluids from the various types of joint disease and their diagnostic significance will be discussed It is only by studies of this type that one can learn more of the physiological and metabolic alterations in articular tissues in joint disease and obtain clues as to the pathogenic mechanisms involved In diseases, such as rheumatoid arthritis, in which the etiology and pathogenicity are unknown, elucidation of under lying mechanisms becomes of major importance The solution of many of these problems must await the acquisition of basic chemical knowledge This is well exemplified in the case of mucin In this instance we are in need of information concerning the chemical composition of the polysaccharides of connective tissue, their structural configuration, the enzyme sys tems concerned with their synthesis and degradation and the intermediate compounds As this knowledge comes to hand the metabolism of mucin in articular diseases can be investigated again The results obtained by new methods should reveal dif ferences in articular disease in terms of cellular derangements and enzymatic systems It is hoped that the alterations observed can be translated in terms of disease mechanism Such an ap proach to the study of chronic rheumatic diseases is enhanced by constant interplay between a group of workers trained in various disciplines ·V11 PRETACE The solution of many of thc problems that arise at the bedside lies in the laboratory Likewise the validity of a given hypothesis which has its origin in the laboratory can only be tested in terms of disease significance at the bedside It is this crossing and re crossing from ward to laboratory which is so vital to our primary objectives, namely the amelioration, alleviation and prevention of the ravages of a group of chronic disabling diseases Acluevement of these goals, be it partial or complete, can come only from a better understanding of the course and nature of the underlying disease mechanisms and the testing of possible means of arrest and prevention We are indebted to Dr Howard C Coggeshall for the initial cytological studies, including the use of the supravital technique, and for his assistance in the study of the initial case material We also are indebted to Dr J Peter Kulka for the review and description of the pathological material and to Dr G E Perlmann for the reported electrophoretic studies which were carried out under her supervision We wish to thank all past and present members of the Lovett Foundation Group who have given generously of time and thought and furnished us with case material It would have been impossible to accumulate the mass of data if it had not been for the assistance of the following, to whom we also wish to acknowledge our debt Miss Elsie C Rossmeisl, Mrs R Barbara Peabody, Miss Elizabeth Manning, Mrs Dorothy Slobodkın, Miss Olivia B Swaim, Miss Barbara Kraft and Miss Katherine L Brine We wish to express our appreciation to Miss Hope Richardson for bibliographic and editorial assistance and to Mrs Magdalene Evers for secretarial assistance in preparation of the manuscript M WR W B November 1952

·No text on this page.

·Contents V Preface X111 Figures Tables ху 1 2 10 10 10 11 12 12 13 14 16 I Introduction Normal Synovial Fluid Synovial Fluid in Patients with Edema Methods Aspiration of Fluid Examination of Fluid Gross Appearance Bacteriology Cytology Physical and Chemical Data Indications for Aspiration II Variations in Characteristics of Synovial Fluid with Disease Intra articular Pressure Amount of Fluid Gross Appearance of Fluid Bacteriology and Immunology Cytology Leukocytes Erythrocytes Specific Gravity and Total Solids Protein Fibrinogen Mucin and Viscosity Non electrolytes 18 19 19 20 21 23 23 38 38 39 52 53 64 ·CONTENTS Non protein Nitrogen and Unc Acid 61 Bilirubin 64 Sulfonamides 65 Penicillin 65 Chloromycetin 65 Sugar 66 Lactate 72 Electrolytes 73 Thiocyanate 74 Hydrogen Ion Concentration 74 Lipids 76 Enzymes 77 III Findings in Fluids in Group I, Diseases of Traumatic Origin 78 Traumatic Arthritis 78 Osteochondritis Dissecans 82 Osteochondromatosis 82 Neuroarthropathy (Charcot Joints) 83 Degenerative Joint Disease 85 Summary of Findings in Group I Fluids 88 IV Findings in Fluids Resembling Both Groups in Some Respects 89 Hypertrophic Pulmonary Osteoarthropathy 89 Hemophilic Arthritis 90 Scurvy 90 Lupus Erythematosus Disseminatus 91 Gout 92 Rheumatic Fever 93 Arthritis Due to Hypersensitivity 94 Erythema Nodosum 95 Hemorrhagic Villous Synovitis 95 Neoplasms 96 98 98 V Findings in Fluids in Group II Rheumatoid Arthritis and Diseases of Infectious Origin Specific Infectious Arthritides Septic Arthritis (Due to Streptococci Staphylococci Pneumococci and E coli) Gonococcal Arthritis Meningococcal Arthritis Haverhill Fever Arthritis Associated with Bacillary Dysentery 100 102 104 105 105

·COVTENTS
X1
100
106
106
109
110
112
Bacteroides Arthritis
German Micaslcs
Tuberculous Arthritis
Syphilitic Arthritis (Congenital)
Reiters Syndrome
Rheumatoid Arthritis
Effect of Synov cctomy on Synovial Fluid Tindings in
Rhcumatoid Arthritis
Intermittent Hydrarthrosis
Bursitis and "Bakers Cysts"
VI Diagnostic Valuc of Joint Aspiration
VII Physiological Considerations
Bibliography
Index
110
118
120
122
128
131
145

·No text on this page.

·Figures I A natural sized drawing of the interior of the left knce joint of a normal dog facing 4 2 Camera lucida drawing of low magnification (*100) of a mi croscopic section taken from the right kncc of the same dog as in Figure 1 facing 5 3 Relationship between the log of the viscosity and the square root of the concentration of mucin glucosamine in synovial fluid and in mucin and polysacchande solutions 7 4 The appearance of the cells of normal synovial flud with two methods of staining 15 5 Total number of leukocytes per cu mm in synovial flud from various joint diseases 28 6 Absolute number of polymorphonuclears per cu mm in synovial fluid from various joint diseases 30 7 Cytological findings and concentration of glucose in series of fuids obtained darly from a patient with generalized rheumatoid arthritis who had bilateral intermittent knee effusions recurring at regular intervals 35 8 Cytological findings and concentration of glucose in a series of fluids obtained daily from a patient with intermittent hydrarthrosis 37 9 Concentration of total protein (exclusive of mucın) in synovial fluid from various joint diseases 10 Concentration of globulin in synovial fluid from various joint diseases 11 Concentration of mucin nitrogen in synovial fluid from various joint diseases 54 12 Relative viscosity at 38° C of synovial fluid from various joint diseases 60 13 Types of mucın precipitate obtained with acetic acid facing 62 48

·YIV
FIGURES
14 The difference between the concentrations of glucose in serum
and synovial fluid in various joint diseases
15 The relationship between the lactate and glucose concentra
tions in synovial fluids and sera in various joint diseases
68
75

·Tables I Normal Human Synovial Fluid 3 II Comparison of the Distribution Ratios for Synovial Fluid, Edema Fluid, the “In Vivo" Dialysate, Ascitic Fluid and Lymph and the Thcorctical Donnan Ratios 5 III Vanations in Cytology of Synovial Fluid with Duration of Effusion 25 IV Comparson of 9 Fluids with Positive Cultures and 33 Fluids with Negative Cultures in Gonococcal Arthritis 27 V Relation of Cytological and Chemical Findings in Synovial Fluid and Serum to Duration of Effusion (Rheumatoid Arthritis) 32 VI Total Leukocyte and Polymorphonuclear Counts in 24 Fluids Aspirated from knees of a Patient with Rheumatoid Arthritis Over the Course of 16 Months 34 VII Relation of Cytological and Chemical Fındıngs in Synovial Fluid and Serum to Seventy of Disease (Rheumatoid Arthritis) 42 VIII Relation of Cytological and Chemical Fındıngs in Synovial Fluid and Serum to Duration of Disease (Rheumatoid Ar thritis) 42 IX Variations in Concentrations of Albumin Globulin and Mu cin in Intermittent Bilateral Knee Effusions in a Patient with Rheumatoid Arthritis 45 X Comparison of Different Methods of Estimating the Con concentration and State of Mucin in Various Types of Joint Disease 63 XI Concentrations of Glucose in Synovial Fluids Aspirated 3 to 4 Hours After Eating 66 XII The Relationship of Intra articular Pressure and Concen tration of Glucose in Synovial Fluid 71

·ХИ
TABLES
XIII Distribution Ratios for Chloride, Calcium and Phosphorus
in Various Diseases
73
XIV Synovial Fluid Findings in Two Patients with Rheumatoid
Arthritis Before and After Synovectomy
118
Chart Synovial Fluid Findings in Joint Disease inside back cover

·Synovial Fluid Changes in Joint Disease

·No text on this page.

·I Introduction PARACELSUS (1493 1541) is credited with having first noted the presence of viscous flud in the articular cavities 148 The fluid is contained in the space formed by the opposing cartilage covered epiphyses and the connecting joint capsule It is now generally agreed that the highly vascular synovial tissue lining the capsule is slightly modified connective tissue, varying in different parts of the joint as described by Key,108 and not an actual membrane Many recent workers consider the articular cavities as connective tissue spaces synovial Aud as tissue fluid and the embodied mucin as the equivalent of the polysaccharide protein component of other connective tissue (see reviewsl. 78 ) The contents of the articular cavity are presumably in direct communication with the intercellular fluid of the synovialis From available data, it has been concluded that synovial fluid is_a_protem containing dialysate of blood plasma to which mucin, secreted by the synovial cells, is added as the plasma water diffuses through the synovial tissue spaces into the larger tissue space, the joint cavity 12 Synovial fluid, as might be expected from the hypothesis that it is tissue fluid, shows changes with disease which reflect altera tions in the synovial tissues and intra articular metabolism The normal exchange of substances between the vascular and lymphatic systems may be disturbed or there may be altered utilization, formation or destruction of various constituents within the joint Examination of synovial fluid should therefore yield information concerning the type and degree of change in the joint Analysis of fluid has, however, been of limited value in the diagnosis and prognosis of joint disorders The knowledge of the 1 ·2 SYNOVIAL FLUID CHANGES physical and chemical characteristics of normal synovial Aud that is necessary to make sound evaluation of pathological fluids has accumulated slowly Furthermore, although a few studies on small groups of fuds from various joint diseases have been re ported, no extensive investigations have been carried out, so information on the variations in the fluid under conditions of disease has been lacking The present study, in which the findings in 1,500 fuds from various diseases are compared with the normal and, insofar as possible, interpreted in terms of the underlying pathological physiological changes, indicates the value of joint aspiration NORMAL SYNOVIAL FLUID The following summary of the characteristics of normal synovial fud is accompanied by a discussion of the exchange of substances through normal synovial tissues, a process which must be understood in order to evaluate the altered permeability produced by disease The details of the studies leading to the conclusions outlined in this chapter have previously been re ported la 38 187 171 Normal human synovial kurd was obtained post mortem from patients who had had no signs or symptoms of joint disease Because sullicient fluid for complete analysis cannot be aspirated from normal human joints, many of the data were obtained from cattle fud, which is apparently, essentially the same as human Normal human synovial fud is a clear, pale yellow, viscous flud which does not clot As might be expected in a fluid that serves chiefly for nounshment of avascular cartilage and for lubrcation, it contains very few cells The cytology is comparable to that of other tissue fluids substantiating the concept of the joint cavity as a connective tissue space The amount of flud obtainable from a normal human joint varies from 0 13 to 35 cc (Table I) The volume of fluid present is determined normally by two opposing forces capillary pressure on one hand, and the difference in osmotic pressure between plasma and fuid on the other The slightly negative intra articular pressure, which varies with motion, also appears to play a role in thus equlbrum Elholm and Norbache found an increase in the volume of fiud in rabbits joints after exercise ·3 INTRODUCTION TABLE I Normal Human Synovial Fluid Range Average 25 78 - 71 - - 21 Amount per knee joint, cc. 0 13 -35 11 Relapve viscosity at 25 C 57 - 1160 235 pH 7.29 -7 45 7 39 Total cell count, cells per cu mm 13 180 Differential-per cent Polymorphonuclears 0 8.5 Lymphocytes 0 24 6 Monocytes 0 479 Clastnatocytes 0 101 Unclassified phagocytes 0 49 Synovial cells 0 43 Total solids gm. per 100 gm 240 -4 83 341 Total albumin and globulin gm per 100 cc. 107 -2 13 172 Albunun gm per 100 cc 1 02 Globulin gm per 100 CC 05° Mucin nitrogen gm per 100 cc. O 06Q 0 135 0 204 Mucio glucosamine gm per 100 cc 0012-0 132 0074 Fibrinogen 0 0 Glucose Approximately same concentration as in plasma Non protein nitrogent Approximately same concentration as in plasma Electrolytest Approximately same concentration as in plasma dialysate Analysus has been made on only I fuid from a patient without edema. The averages for total albumin and globulin include 9 other fluds See Table II Substances entering the joint must pass the capillary endothelium and diffuse through the intercellular spaces of the synoviales Comparison of the composition of joint fud with that of plasma indicates that non electrolytes diffuse readily in either direction between blood and synovial fluid and that electrolytes are distributed in accord with the Gibbs Donnan theory of membrane equilibrium?87 (Table II) The concentration of nonelectrolytes (glucose, non protein nitrogen, unic acid) is approximately equal in fluid and serum Chloride and bicarbonate are present in higher concentration in the fluid than in the serum while the concentrations of sodium potassium, calcium and mag. nesium are lower in the fud The low distribution ratios obtained for potassium, calcium and magnesium indicate that part of each of these cations is held in the serum, presumably bound to protein It is of interest that Joseph, Reed, Steck, Folk and Kaplan102 found the mobility of many ions in synovial tissues the same as in water but that the bivalent cations had lower apparent mobilities in the tissues than in water Thus the distribution of electro- ·4 SYNOVIAL FLUD CHANGES lytes and non electrolytes between serum and fluid is in accord with the concept that synovial fluid is a dialysate in equilibrium with blood plasma This relationship is also suggested by the marked vascularity of the synovial intima whose numerous capil laries and venules are in close proximity to the articular lumen 12 79 (See Figures 1 and 2 ) Such a theory explains all that is known of the physical and chemical composition of synovial Auid except the presence of albumin, globulin and mucın Slight capillary permeability to protein presumably explains the entrance of albumin and globulin into the joint cavity The high albumin/ globulin ratio of the fluid compared with that of the serum, which accords with the findings of other investigators in lymph, edem? and ascıtıc fluids (see review12), indicates greater permea bility to albumin than to globulin, corresponding to the difference in molecular weights The presence of mucın, whether it be formed by the surrounding connective tissue, as is most lihely, 17 39 03 99 107 110 201 or by cartilage, in no way invalidates the theory The properties which distinguish synovial fluid from other fluids which are also dialysates of blood plasma are attributable to muçin Mucın, the characteristic component of synovial fluid, is an easily dissociable protein-polysaccharide complex The exact nature of the components has not been determined It is known however, that the polysaccharide contains equimolar amounts of acetyl glucosamine and a uronic acid 132 and that the protein component consists of at least an albumin and several globulin fractions 170 Jeanloze has shown that the polysaccharide is a non ramified polymer of d glucuronic acid and N acetyl D gluco samine residues connected through B 1-3 linkages Glucosamine constitutes 40 per cent of the polysaccharide and 8 per cent of the mucın 187 The degree of dissociation of the protein polysaccharide complex in vivo has not been determined with certainty 12 65 79 131 139 170 The hypothesis that mucın is formed by the cells of the connective tissue surrounding the joint is supported by good evidence l. 8a 1180 131 A substance similar to synovial fluid muçin has been found in the subcutaneous tissues of rabbits 184 the sexual skin of monkeys 138 cocks combs83 and the connective tissue lining the astragalotibial joints of cattle 164 In addition the en ·F FIGURE 1 A natural sized drawing of the interior of the left knee joint of a normal dog The blood vessels of the rear extremities had been perfused with a suspen sion of graphite It will be noted that the subsynovial tissues are very vascular particulrly in the region of the infrapatellar fat pid and immediately adjacent to the patella

·No text on this page.

·INTRODUCTON 5 TADLE II Comparison of the Distribution Ratios for Synovial Muid, Edema Fluid tho "In Vivo Dialysato Ascitic Fluid, and Lymph and the Thcorctical Donnan Ratios (Sce Review) Synovial fuld Cartie Tlumen 10 vtro dialyzatok dos Edema Guld Axcitia fuld Lymph 99 99 97 93 08 05 C! CIC HCO, HCO, PO. PO 04 00 07 97 1 00 103 1 17 1 12 107 Na, 93 97 01 Na 75 73 78 Ca 83 87 80 76 84 Mge 88 68 86 93 93 96 93 97 Mg Theoretical Donnan NPN, NPN Uric acid, Uric acid 1 87 91 84 largement of the Golgi apparatus of the synovial cells in response to trauma3" 110 and the associated accumulation of intercellular mucoid material demonstrated by King110 offer support for this theory The mode of destruction of muçin is not known but in view of the difficulty with which large molecules leave the joint cavity it is probable that breakdown occurs in the synovial fluid or tissues Many bacterial and tissue enzymes that produce in vitro degradation of mucın have been recognized but similar enzymatic activity has not been demonstrated in synovial fluid or tissues 131 170 Ascorbic acid, which also causes degradation of muçin in vitro, has not been shown to have hyaluronidase activity in vivo Testicular extract, however has been found to maintain 13 ·(1 FIGURE 2 Camera lucida drawing ot low magnification (*100) of a micro scopic section taken from the right knee joint of the same dog as in Figure 1 The rich subsynovial vascular system is well illustrated In some instances the blood cipıll iries are separated from the interior of the joint by not more than one or two lavers of synovial lining cells ·10,000 3,000 6000 spoo 4000- 4000 pooh 1000 Boo! 6004 500 400- 300 RELATIVE VISCOSITY H + 아 VGlucosamine in mg per 100 CC FIGURE 3 Relationship between the log of the viscosity and the square root of the concentration of mucin glucosamine in synovial fluid and in mucin and polysaccharide solutions Viscosity measurements were made at 38 C # Bursal fluid (human) in 0 9 per cent NaCI Normal carpometacarpal and astragalotibial joint fluid (cattle) Muan (cattle) in M/15 Na HPO, Mucin (cattle) in 05 per cent NaHCO - Mucin (cattle) in joint Aud (cattle) Mucin (cattle) in 05 per cent Na,Co, • Astragalotibial joint fluid (cattle) in 0 9 per cent NaCl • Mucın (cattle) in serum (human) diluted with Ringer s phosphate solution Mucin (cattle) in M/15 Na acetate Polysaccharide (cattle) in 0 9 per cent NaCl Figure 3 originally printed in Ropes 1 Y Robertson W T B Rossmeisl EC Peabody RB And Bauer Ý Synovial Fluid Mucin Acta Med Scandinav 1947 Suppl 196 700 ·6 SYNOVIAL FLUID CHANGES its hyaluronidase activity when injected into joints of patients having rheumatoid arthritis 158 Exposure of synovial fluid or hyaluronic acid solutions to drays in vitro has been shown to cause marked reduction of viscosity 167 The high viscosity of normal synovial fluid is due to the presence of mucin 44 131 170 This is shown by the fact that the viscosity of the fluid approaches that of water after the mucın is precipi tated or destroyed enzymatically Evidence that the long chain molecule of the polysaccharide component causes the viscosity is given by the persistence of the viscosity of mucin solutions after tryptic digestion of the proteins Gunter80 noted similar persistence of the fiber-forming capacity of cattle synovial fud after tryptic digestion A proportional relationship has been found to exist between the log of the viscosity of a mucın or polysaccharide solution or synovial fluid and the square root of the mucin glucosamine con centration*70 (Figure 3) This proportionality, which apparently remains constant even after partial splitting of the polysaccharide, provides ? possible means of estimating the degree of degradation of mucın in pathological effusions The most reliable indication of the state of mucın in any fluid, however, is the nature of the precipitate obtained with acetic acid Further evidence of the state of mucin may be obtained in some cases by the nitrogen/ glucosamine ratio (a comparison of the methods is given on P 62) The properties of mucin suggest its participation in various physiological processes It is presumably of great importance in the lubrication of joints 79 97 170 Because of its high base binding power it apparently aids in regulation of calcium equilibrium 47 107 It also affects the transfer of water between plasma and syno vial fluid 167 Its effect in holding water in the intercellular tissue spaces is suggested particularly in myxedema, a disease in which the amount of mucin in synovial and other connective tissues is greatly increased Ta 8. 11. 2052 213 The permeability of the synovial tissues to substances of large molecular size such as proteins is of special significance in the physiology of joints because of the effect of such substances on osmotic pressure and thereby on water exchange Possibly it also explains the frequent localization of disease processes in the ·NTNODUCTION 9 Absorption is greatly increased from joints which are crcrcised and from those in which severe, acute inflammation has been produced, and is decrcased with chronic inflammation of joints" i 51 108 115 185 186 Cfskind," however, found reduced absorption of perabrodil and indigo carminc in hemorrhagic traumatic arthritis as well as rheumatoid arthritis Edlunds' ascribed the increased absorption after cxercise to a "breaking of the resistance of the tissues and concluded that the mechanism of absorption was the same in synovial tissues as in that described for other connective tissues by McMaster (see revicw7o) Edlund and Linderholm noted incrcased absorption of water and hemoglobin after administration of salyrgan and a decrease following production of subcutancous inflammation or contralateral intra articular burns co Hoffheinz87 concluded that blocking or cutting the nerve supply to a joint decreased absorption of sodium iodide from the joint Seifter, Baeder and Begamy181 reported that hyaluronidase and desoxycorticosterone acetate increased absorption of phenolsulfonphthalein from joints Cortisone and 21 acetoxypregnenolone decreased the absorption from joints and counteracted the increase produced by hyaluronidase 182 Edlund,69 however, concluded that desorycorticosterone acetate does not affect absorption from normal joints but decreases absorption from mildly injured joints and increases it from severely injured joints The only experimental evidence of the presence of immune bodies in normal synovial fluid is found in the work of Poehl- He studied 100 fluids from patients having various stages of syphilis with and without joint symptoms of the patients with positive blood Wassermann reactions, 25 of 36 untreated secondary cases, 2 of 12 latent cases, and 3 of 9 tertiary cases had positive reactions in the fluid He never found a positive Wassermann in the fluid when the reaction in the blood was negative Relatively little is known of the enzymes of normal synovial fluid Podkaminskylul noted the presence of an amylase, protease and lipase but no catalase We have found, however, that normal fluid (bovine and human) causes breakdown of hydrogen peroxide The activity of alkaline phosphatase in normal cattle fluid is higher than that in the serum but varies markedly from fluid to mann 152 ·8 SYNOVIAL FLUID CHANGES articulations Compared with true membranes, synovial tissue has a markedly greater permeability 12 The factors affecting permeability are only partially under stood Engel® has concluded, from the results of experiments in animals on which unilateral lumbosacral sympathectomy had been performed that loss of activity of the sympathetic system decreases capillary permeabilty Cheng $$ however, has been unable to corroborate Engels findings in a comparison of acid and alkaline dyes, Tan.200 found that 36 of 50 acid dyes and 16 of 26 basic dyes entered the joint Engel, however, noted that almost all acd dyes but no alkaline dyes appeared in the joint perfusate He explamed the difference by the hypothesis that the Lipoid soluble alkaline dyes were removed from the blood stream by the cells and tissues within a few minutes, whereas the acid dyes remained in the blood for hours Permeability has been found to be increased in inflamned joints by von Seemen 203 Burman and Kling, at and Bondi and Jacoby, 21 although Engel63 found decreased permeability to acid dyes and Scott Thobum and Hanzlık??s and Frohlich and Singer noted no increased entrance of salıcylates into inflamed joints Tanı"98 reported greater entrance of both acid and alkalıne dyes in in flammation of short duration but a decrease after 11 to 14 days The entrance of particulate matter into the joint has been less ughly investigated Kuhns and Weatherford 118 found that carbon particles injected into the abdominal wall of rats appeared in the histiocytes of the synovial membrane but in smaller quantities than in the liver spleen, kidney and bone marrow Experimental studies have shown that living bacteria gain access to synovial fluid more readily than to spinal fluid, aqueous humor or urne (see review!) Resorption of small molecules takes place almost entirely by way of the subsynovial blood capillaries and only to a slight degree through the lymphatics 12 79 Proteins in contrast to small, readily diffusible substances are removed from joints only by way of the lymphatics 12 The removal of particulate matter in cluding the cellular debris of wear and tear, is slower and 35 achieved by a combination of several processes in which the mononuclear phagocytes, the predominant cellular constituents of synovial fluid, play the chief role la 79 ·INTRODUCTION 11 substances as sugar to be cstablished between serum and fluid In some cases the paticnts were not fasting or were under anesthesia at the time of aspiration The concentrations of sugar and phosphate in such cases were not of value for compinson with the findings in other fluids The skin over the joint to be aspirated was prepared by shaving and washing with green soap, alcohol and other 12 hours before aspiration and again washing with 1odine and alcohol at the time of aspiration The skin and the subcutaneous and subsynovial tissues were infiltrated with novocaine at the site of puncture The majonty of fluids were obtained from knees, although in a few cases, shoulder, elbow, hip or ankle joints were aspirated On the rare occasions when fluid was aspirated simultaneously from one or both knees and another joint, it was similar in all The usual site of puncture in the knee was slightly above and approximately 2 cm lateral to the patella From this position one could enter readily either the quadriceps pouch or the region of the joint cavity under the patella An 18 or 19 gauge needle with stylet, of the type employed for lumbar puncture, was used for withdrawing the fluid Intra articular pressure was determined by means of an Ayer manometer In order to obtain comparable and consistent pressure readings, the patients were recumbent and quiet, with the leg extended in a position of comfort If any motion of the leg occurred while the pressure was being determined, a nse in pressure, which subsided slowly to the original level, resulted In order to facilitate the passage of extremely viscous fluids into the manometer, it was often necessary to rotate the manometer so that it was parallel to the leg and therefore level with the joint After rotating the manometer back to its normal position, readings were not made until the fluid level had been constant for several minutes Blood was obtained from the median cubital vein at the time of aspiration Following aspiration an elastic bandage was applied and the patient allowed to walk immediately if necessary Examination of Fluid The first sample of flud was collected for bacteriological studies in a stenle tube, usually containing 2 to 3 mg of sodium citrate per cc of fluid In the sample used for cytological studies and ·10 SYNOVIAL FLUID CHANGES fluid 167 The activity in normal human fluid is much lower than that in the serum SYNOVIAL FLUID IN PATIENTS WITH EDEMA The synovial fluid from patients with edema but no joint disease provides additional evidence in support of the hypothesis that synovial fluid is a plasma dialysate to which mucın is added as the plasma water diffuses through the synovialis This synovial fluid is comparable to the edema fluid obtained from other con nective tissues A disturbance of the equilibrium regulating the formation of tissue fluid, which is usually due to increased capillary hydrostatic pressure and/or decreased osmotic pressure difference between plasma and fluid, accounts for the increased volume of fluid in both cases All the findings in the synovia in such patients indicate dilution of normal fluid In the series studied in this laboratory, the amount obtained from the knee was occasionally as much as 40 cc 171 It was clear, very pale yellow and much less viscous than normal with an average relative viscosity of 53 in contrast to 235 Ragan255 in 5 fluids found volumes from 12 to 30 cc with viscosities ranging from 3 to 622 The total nucleated cell count also was lower than normal, ranging from 0 to 128 38 119 1 8 The specific gravity varied from 1 008 to 1015 90 171 The average content of total solids was reduced from 341 gm per 100 gm to 2 23 The pH remained essentially normal, with a range from 7 29 to 7 45 23 171 The average total protein, exclusive of mucin was 0 63 gm per 100 cc in contrast to 172 gm in fluid from patients without edema The globulin fraction was very low, ranging from 0 to 0 33 gm per 100 cc 31 171 The average concentration of mucın nitrogen in our series was reduced from the normal value of 0 104 gm nitrogen per 100 cc to 0061 gm 171 The lowered mucin content associated with the increased diffusion of water into joints in patients with edema suggests that the rate of formation of mucın is not so accelerated as the rate of transudation of water METHODS Aspiration of Fluid The majonty of the patients fasted for at least 12 hours before the joints were aspirated in order to allow equilibrium for such ·INTRODUCTION 13 plun infusion broth or thioglycollate broth were adequatc For gonococci, meningococci, Streptobacillus moniliformis or pleuro pneumonia like organisms, boiled blood ascitic agir plates or broth were used For bacteroides, thioglycollate broth was used and cultures were cpt 2 uccks Thc percentage of positive results in infectious arthritis was increased if the cultures were planted at the bedside They were incubated at 37° C crobically and under 10 per cent Co, In most cases, cultures were made also in anaerobic meat tubes All cultures were kept for 4 days and many for 2 weeks Smears stained with Gram's stain were cramined in all cases in which cultures were made some of the fluids from rheumatoid arthritis were inoculated into chick embryos, and were injected into mice intravenously, intra articularly or intracerebrally Guinea pig inoculation was necessary in tuberculous arthritis In a few cases, cultures for acid fast bacıllı were made on Duboss medium 60 70 Cytology Cytological methods were essentially the same as those used in the study of normal human fluid 38 The erythrocytes and leukocytes were counted simultaneously in a standard hemocytometer In all cases except occasional traumatic effusions, it was necessary to dilute the fluid in a standard white cell counting pipette Normal saline was used, since the routine diluents cause precipitation of mucin with resultant reduction in the number of cells in the supernatant fluid Addition of methylene blue to the salıne aided in recognition of nucleated cells Differential counts were made either on thin fixed smears stained with Wrights stain or on fresh preparations stained supravitally with neutral red and Janus green 11 205 In many cases counts were obtained by both methods and showed essentially the same percentage of cell types The chief differences appeared in the distribution of mononuclear cells, which at times are relatively hard to differentiate with Wright s stain in fixed smears However, differentiation of one mononuclear cell from another is not absolutely essential for diagnostic purposes The relative proportion of polymorphonuclear to mononuclear cells gives the best indication of the degree and type of inflammation in the synovial tissues The total number of phagocytes is of course, of additional value in determining the degree of tissue irritation and injury In general differential counts on pathological effusions can be made ·12 SYNOVIAL FLUID CHANGES determination of fibrinogen content, clotting was prevented by the addition of 2 mg of potassium oxalate per cc of flud For the other analyses, no anticoagulant was added The specimens of fud and blood to be used for determination of pH were collected under oil The determinations which were made in each case depended on the amount of fluid obtained A few drops were sufficient for determination of the gross appearance, total nucleated cell count differential count, viscosity and the type of mucın precipitate The total amount necessary for the other most significant analyses determinations of the sugar, total protein, globulin and mucin concentrations pH, bacterial and immunological examinationswas at least 8 cc * In the first 400 pathological effusions examined much more complete studies were made, including other non electrolytes (non protein nitrogen and unc acid) and electrolytes (calcium, phosphate and chloride) Since no diagnostic or prog nostic significance of these determinations was apparent, we do not consider them of value in the routine analysis of synovial fluid Blood withdrawn at the time of joint aspiration, was always examined for sugar if possible and for protein fractions when indicated Gross appearance The appearance of the flud in respect to degree of turbidity, color apparent viscosity and the presence or absence of clotting was noted at the time of withdrawal If the fluid contained blood, it was important to determine whether it was due to a bloody tap Unequal distribution of blood between the samples, streaking of blood through any sample and rapid clotting especially if localized to the bloody streals indicated that the bleeding was caused by the tap The presence of xanthrochromic blood pigment in the supernatant fluid after centrifuging indicated that bleeding had occurred in the joint before aspiration Bacteriology Cultures were made on the majonty of fluids In searching for such organisms as streptococci, staphylococci pneumococci, or E coli routine cultures on blood agar plates and *Observation of the color viscosity and turbidity of the fluid its tendency to clot the type and size of the mucın precipitate formed by the addition of acetic acid to a final concentration of 1 per cent and determination of the total and differential leukocyte counts will often be of diagnostic value if facilities are not available for chemical analyses

·THE
CELLS
OF
NORMAL
SYNOVIAL
FLUID
WRIGHTS
SUPRAVITAL
POLYMORPHONUCLEAR
LEUKOCYTE
MONOCYTE
CLASMATOCYTE
LYMPHOCYTE
SYNOVIAL
CELL
ERYTHROCYTE
FIGURE 4 The appearance of the cells of normal synovial fud with two methods of staining

·SYWOIZAL FLUID CHANGES 14 satisfactorily on fixed smears Supravital preparations give somewhat more accurate results in classification of cell types but have the disadvantages of being slightly more time-consuming and of requiring that the fluid be kept warm and be examined within an hour after aspiration In fixed smears the precipitated mucin of the fud will stain lightly (usually it is basophilic), forming an amorphous background which does not interfere with differentration of the cells The types of cells found in synovial fluids and their appearance with the two methods of staining are shown in Figure 4 In many pathological fluds the polymorphonuclear leuhocytes, the monocytes and the clasmatocytes show signs of marked imtation with Vacuolization of the cytoplasın Occasionally degenerated polymorphonuclear cells with single pyknotic nuclei or with extruding nuclei are seen. Physical and chemical data The following chemical methods were used chlonde, Eisenman modification of the Van Slyke method,“inorganic phosphate, Fiske and Subbaros e sodium, Rourke modification of the Kramer-Gittleman method 103 calcium Fishe and Logan non protein nitrogen, Folin and Wu, uric aad, esther Benedict and Behree or Folin, sugar, Folin, fatty acids Stoddard and Drury,193 cholesterol, Bloor, 20 lactic acid, Barker and Summerson. 10 The pH sas determined by means of a McIndis glass electrode, measurements being made at 25° to 28° C and corrected to 37° by the use of temperature coefficients (pH/At=0006 for synovial fluid, 0012 for serum) Specific gravity was determined in specific gravity bottles with open capillary outlets of the type described by Moore and Van Slyke 124 Total solids were obtained by drying a weighed sample of approxmateh i gm at 100° C for 48 hours Viscosity was determined either by a Hess viscosimeter or by an Ostwald viscosity pipette in a constant temperature bath Glucosamine was determined by a slight modification of the Palmer Smyth and Meyer method, 144 with the use in some cases of an Evelyn photoelectric colorimeter with filter no 520 and in other cases of a Coleman spectrophotometer Protein other than mucin was calculated from the total putrogen (obtained by a macro- or micro-Kjeldahl method) The difference between the total nitrogen and the sum of the nonprotein and mucin mitrogen was multipled by the factor 6.25 69 ·INTRODUCTION 17 tissues In such cases it may prevent relaxation of tissues and hasten recovery One gets the impression, in occasional mild and moderately severe cases of rheumatoid arthritis in which effusions persist after definite general improvement, that aspiration hastens the subsidence of the effusions Aspiration of the knee effusions in long standing cases of chronic rheumatoid arthritis is sometimes indicated to allow better position and motion of the joints, to prevent further stretching and lessen relaxation of supporting structures and to improve the circulation of the synovial tissues Although antibiotics or sulfonamides are rapidly effective in the treatment of many cases of infectious arthritis, daily joint aspirations are valuable adjuncts They are of fundamental im portance in infections not susceptible of specific therapy In such cases, repeated daily aspirations with or without salıne irrigations are often sufficient for cure, with resulting normal joints Open drainage with the likelihood of residual joint damage may thereby be avoided The total nucleated cell count, percentage of polymorphonuclears and sugar concentration in the fluids withdrawn daily from septic joints enable one to determine whether further specific therapy is needed and in non susceptible cases whether aspiration is sufficient or open drainage is necessary Fluid aspiration, followed by injection of air or lipiodol, may be valuable when visualization of the joint space is desired The use of which is readily absorbed and causes little or no irritation, is preferable Subsequent X-ray examination indicates whether or not the joint space is in communication with surround ing bursae or pouches There is little question about the safety of joint aspiration None of the 2 200 or more aspirations performed in this clinic has been followed by a joint infection With preparation of the region both 12 hours before and at the time of aspiration, and with ordinary aseptic precautions, the danger is negligible The danger of serious bleeding exists only in hemophilia However, even in this disease, aspiration is safe if the clotting time has been brought to normal by transfusion in the previous 12 to 24 hours Withdrawal of bloody fluid from hemophilic joints is often indicated to relieve the severe pain and to prevent marhed iniammatory changes in the synovial membrane air, ·SYNOVIAL FLUID CULANCES 16 Mucin nitrogen and glucosamine were determined by precipitation with 1 per cent acetic acid and re precipitation with acetic acid from a 01 per cent sodium carbonate solution 16 Either the difference in total nitrogen or glucosamine before and after precipitation or the nitrogen or glucosamine of the precipitated mucin represents the mucin nitrogen or glucosamine Albumin and globulin contents were determined either by a modification of the Howe method” or by the method of Butler and Montgomery The fibrinogen content of a few fluids was determined by the Cullen and Van Slyhe method, the nitrogen was measured by digestion and nesslerization However, as most of the fluids con tained too little fibrinogen for determination by this hod, the amount was estimated on the basis of the size of the clot formed Clots were classified as smal (+), moderate (++) and large (+-+-) In the few fluids that contained sufficient fibrinogen for chemical analysis, sold clots formed within a few minutes after aspiration In general the large sized clots formed soon after withdrawal, whereas the formation of small clots frequently took many hours The electrophoretic experiments were performed at 15° C in the apparatus described by Tiselius*o* equpped with the Schlieren scanning device of Longsworth 22* Before electro phoresis, the plasma was diluted with buffer to make a final concentration of 30 mg protein nitrogen per cc, or 1 87 per cent The solution was then dialyzed for three days with daily change of buffer The buffer used in all experiments was 01 N sodium di ethyl barbiturate and 002 N free barbituric acid pH 86, ronic strength 01 INDICATIONS FOR ASPIRATION The diagnostic and prognostic value of the information derived from the examination of synovial fluid is established in the present series making the chief indications for joint aspiration obvious There are also, however, certain therapeutic indications Wort ever the purpose, the benefits of aspiration can be obtained without danger or marked discomfort to the patient Withdrawal of fluid gives marked relef in massive effusions of traumatic or infectious ongin which cause severe pain either by their size or rapidity of formation and resultant stretching of ·VARIATIONS WT111 PISTASE 19 examination, Inboratory and I ry studies and a follow-up of the patient after discharge from the hospital Quicstionable cascs have been excluded from the scrics Analysis of all of the findings indicates that pathological synovial fluids in general can be divided into two large groups which will be designated in the following discussion as Group I and Group II Group I consists chiefly of cases in which an inflammatory reaction of traumatic origin is present This includes fluids from all types of traumatic arthritis (including those with injury to the semilunar cartılages and those with hemorrhagic effusions), osteochondritis dissecans, degenerative joint disease, neuroarthropathy (Charcot joints), and ostcochondromatosis Group II includes fluids from all types of infectious arthnts of known ongin (streptococcal, staphylococcal, pneumococcal, gonococcal, meningococcal, E coli, bacteroides, tuberculous, syphilitic and those of Haverhill fever, bacıllary dysentery, and German measles), from Reiter's syndrome and from rheumatoid arthntis Effusions in lupus erythematosus disseminatus, hypertrophic pulmonary osteoarthropathy and hemophilia resemble the fluids of Group I in most respects Fluids from rheumatic fever, gout, hemorrhagic villous synovitis and synovial tumors of various types resemble each group in certain respects INTRA-ARTICULAR PRESSURE In human knee joints without effusions, the intra-articular pressure, as found by Muller,135 is 8 to -12 cm of water The intra articular pressure in joint disease in this series vaned markedly and showed no correlation with the etiology, severity or duration Great variation was found also by Ranke159 and Rostock 172 A significant relationship between intra-articular pressure and the concentration of sugar in the fluid was indicated, however, as will be discussed later The changes in sugar concentration with pressure, however, are not of diagnostic significance AMOUNT OF FLUID The amount of fluid obtained (post mortem) from normal knee joints in our series ranged from 0 13 to 35 cc 171 Raganlos obtained similar amounts, ranging from 10 to 40 cc The term "inflammation" is used to indicate disorders of permeability cucula tion and metabolisan resulting from any type of tissue injury ·II. Variations in Characteristics of Synovial Fluid with Discase 133 THE following interpretations of the changes found in the cyto logical and chemical characteristics of synovial Aud in various diseases are based on studies of approximately 1,500 fluids Com parison of the data is valuable because fluids from all types of joint disease have been studied by similar techniques The studies reported by other investigators will be referred to throughout the discussion The classification of joint diseases used is that approved by the Nomenclature Committee of the American Rheumatısı Association Rheumatoid arthritis (proliferative arthritis, atrophic arthritis, chronic infectious arthritis) includes Marie Strumpell spondylıtıs, Stills disease and a combination of psoriasis and generalized progressive arthritis The term 'intermittent hydrarthrosis refers to cases of unknown etiology in which joint effusions recur with amazing regulanty at constant intervals A similar phenomenon occurs in some patients with defimte rheumatoid arthritis, and it is probable that many patients with so called idiopathic intermittent hydrarthrosis are actually atypical cases of rheumatoid arthritis In the present series, there are only 2 cases of intermittent hydrarthrosis in which a diagnosis of rheuma toid arthritis cannot be made with certainty Even these 2 cases have clinical and laboratory findings suggestive of rheumatoid arthritis, as, for example, an elevated erythrocyte sedimentation rate in one and a progressive thickening of synovial tissues of both knees in the other The accuracy of the diagnosis in all cases has been controlled as fully as possible by history, physical 18 ·VARIATIONS WITHI DISEASE 21 destruction of tissue On microscopic cxamination such fluids were found to contun numerous cholesterol crystals DACTERIOLOGY AND IMMUNOLOGY Normal synovial fluid is stenle Group I fluids are also sterile The findings in cultures from Group II fluids vary with the etiology of the disease Cultures of fluid from 150 cases of rheumatoid arthritis in this senes were negative, even though cultured aerobically and under 10 per cent Co, on boiled blood ascitic agar and broth and on blood agar plates and in anaerobic mcat tubes Inoculations into chick embryos and intrivenous, intra articular and intracerebral injections into mice also gave negative results Positive synovial fluid cultures are essential in order to prove the diagnosis of infectious arthritis In this series, positive cultures were usually obtained in the cases in which joint infection by streptococcus, staphylococcus, pneumococcus or E coli was suspected In fluid from tuberculous arthntıs, the presence of organisms was proved by guinea pig inoculations in 21 of the 29 cases tested Positive cultures for tubercle bacıllı were obtained in 3 of 5 cases Fluids from patients with meningococcal arthritis yielded posi tive cultures only within the first few days after the onset of arthntis Although organisms were seen in the smears during the subsequent few days, they did not grow on culture Thereafter, no organisms were seen, and cultures, even from purulent fluids, were negative Whether or not this was due to an alteration of the organism produced by the antibiotics used to treat the generalized meningococcal infection is impossible to say Such a possibility is suggested by the recently demonstrated changes in the form of organisms produced by penicillin and strepto mycin." In many cases of probable gonococcal arthritis and in 14 cases of apparent specific infectious arthritis associated with purulent foci (such as empyema, pneumonia, mastoiditis, pelvic abscess and osteomyelitis) caused by B hemolytic Streptococcus, Staphylococcus aureus or Pneumococcus, type 7 or 23, cultures were negative Apparently the organisms reached the synovial tissues $ 133 ·SYNOVIAL FLUID CHANCES 20 The amount obtained in any one aspiration in the present series varied markedly, the minimum and maximum being 01 cc and 240 cc In all cases, with the exception of a few in which special studies were made, as much fluid as possible was withdrawn No definite correlation between the size of the effusion and the etiology, severity or duration of the joint disease was found In acute effusions, the amount of Auld decreased, of course, as the inflammation subsided Similarly, in the majority of rheumatoid patients receiving either adrenocorticotrophic hormone or cortisone,' the volume of effusions decreased GROSS APPEARANCE OF FLUID Normal synovial fluid is clear, pale yellow and viscous and does not clot The gross appearance of fluid is in itself of diagnostic significance In effusions from the types of joint disease included in Group I, it is similar to that of normal fluid relatively clear, pale yellow, fairly viscous and usually not clotting or forming only small, friable clots In cases of traumatic arthritis or synovial tumor in which hemorrhage has occurred in the past, the fluids are red or, if aspirated late, have a deeper yellow color due to pigments from destroyed hemoglobin In 88 traumatic fluids Kling is found an average icteric index of 14 with a range from 8 to 133 Occasionally globules of free fat are seen on the surface of fluids obtained from severe traumatic arthritis The fat is released, apparently, by injury to periarticular tissues or bone marrow Fluids of Group II, in contrast, are usually turbid and much less viscous and the majority of them coagulate, often forming large firm clots The turbidity increases and viscosity decreases roughly in proportion to increasing severity of the joint disease Fluids from severe cases frequently show moderate amounts of debris, including small shreds of fibrin and amorphous material presumably resulting from cellular dısıntegration In 2 cases of severe rheumatoid arthntis of long duration the fluids were deep orange and contained free fat, large amounts of amorphous ma terial and many flecks of fibrin, findings which indicate marked *We are indebted to Armour Laboratories Armour and Company Chicago for the adrenocorticotrophic hormone and to Merck and Company Inc Rahway New Jersey for the cortisone used in these studies ·VARIATIONS WITH DISEASE 23 present senes, the Wassermann or gonococcal complement fixation test was positive in the joint fluid, even though it was negative in the scrum and in fluid aspirated from another joint at the same time Such "false positive reactions in the absence of history, signs or symptoms compatible with syphilis or gonorrhea may be assumed to be non specific and of no diagnostic significance Similar unverified false positive” reactions are given occasionally by serum and other body fluids and have been found by others in synovial fluid 104 116 Further evidence that the synovial tissues are permeable to antibodies was found in a case of bacillary dysentery with joint involvement in which the agglutinın titer in the fluid to the Hiss Y strain of Sh dysenterae was 1/180 compared with 1/320 in the blood Similarly, in a case of Pneumococcus (type 23) mastoiditis with associated arthritis, the agglutinin titers in blood and fiud were equal (positive in a dilution of 1/16) Increasing agglutının titers in the Aud, corresponding to increases in the blood, were found also by Labor and von Balogh" in arthritis associated with bacıllary dysentery Agglutination titers for B tularense of 1 1024 and 1 512 were found by Foshay and Mayer*s in fluids from an olecranon bursitis in a patient with tularemia of 4 months duration The titer in the serum corresponding to the latter fluid was I 640 Similarly, fluid and blood had the same bactericidal activity for hemolytic streptococci in 11 of 14 patients with various joint diseases studied by De Gara º The activity against E coli was alıke in fluid and blood in 8 of 15 cases In 40 fluids tested by De Gara bactencidal activity for hemolytic streptococci was present in only 11 per cent (2 cases of rheumatoid arthritis I of non specific infectious arthritis and 1 of intermittent hydrarthrosis), whereas bactericidal activity for E coli was found in 87 per cent CYTOLOGY Leukocytes The total nucleated cell count of normal human fluid in our senes varied from 13 to 180, with an average of 63 per cu mm The average differential count is polymorphonuclears 6.5 per cent, lymphocytes 24 6 per cent, monocytes 47 9 per cent, clasmatocytes 101 per cent, unclassified phagocytes 49 per cent, ·22 SYNOVLAL FLUID CHANGES in these cases, causing inflammation and effusion, but were not present in the fluid in sufficiently large numbers to be cultured These findings are in accord with the results of post mortem exam ination of synovial tissue and synovial fluid from patients having generalized infection without joint symptoms 15 The synovial tissues in such cases showed moderate or marked inflammatory changes consisting of perivascular and often diffuse infiltration with polymorphonuclers and mononuclear leukocytes Occasion ally, sharply defined acute infiammatory lesions were observed with marked infiltration of polymorphonuclears and mononuclears fbrin deposition, tissue necrosis and thrombosis of arterioles. The synovial fluids showed corresponding increases in the total leukocyte and polymorphonuclear counts, as was observed in similar cases by Labor and von Balogh 119 Changes in the membrane comparable to those described above are presumably the basis for the joint pains that occur frequently in severe generalized infections Apparently organisms enter the synovial tissues in such infections, but the extent of the resulting joint involvement is determined by virulence and resistance and may be merely an arthralgia or 2 definite arthntis in which organisms cannot be cultured from the fluid, or a truly purulent arthritis In pommal synovial fud Maunsell, Wngley, Highton and Holt1272 found no anti-Å or antı B iso agglutinns or diphtheria immune antibodies However, four weeks after injection of diph thera PTAP, in i normal control, the synovial fluid titer was 15 units per cc compared with a serum titer of more than 200 units per cc In patients with joint diseases (Theumatoid arthritis, osteoarthritis, infectious arthritis, intermittent hydrarthrosis and gout), the titers of anti A and B antibodies were approximately the same in fud and serum Similarly after injection of PTAP in a patient with osteoarthritis, the titers in fluid from both knees (30 and 40 units per cc ) were almost as high as those of the sera (50 and 60 units) 1974 Specific joint involvement is not necessarily indicated by the complement fixation test It is usually positive in the fluid when it is positive in the serum Spink and Keefer,200 however in a series of 70 cases reported 47 positive tests in 54 blood samples compared with 21 positive in 35 joint fluids Occasionally in the ·TABLE III Variations in Cytology of Synovial riud with Duration of Effu sion Case bumber Duration of effusion Amount CC Total per cu mm Leukocytes per cento P L N Diagnosis Joint 1 Traumatic arthritis with hemorrhage Ince L L 4 wceks 5 weeks 80 12 4 000 600 10 0 27 67 63 33 2 Traumatic arthritis with hemorrhage L 19 5 days 20 750 31 days 10 100 1 day 10 7700 2 days 02 2 900 2 days 12 12 300 5 days 04 2 600 12 hours 20 13 800 34 days 5 1980 4 weeks 12 51 600 5 weeks 12 10 000 5 months 35 19 600 3 days 6 22,200 5 days 12 13 750 2 months 3 3 000 49 27 87 22 74 22 95 87 74 86 85 60 48 86 2 12 2 10 8 18 2 2 18 0 3 8 33 30 49 61 11 68 18 60 3 11 8 14 12 12 14 hours 22 days 55 20 22 600 9000 76 01 24 33 0 B 3 Rheumatic fever R R 4 Rheumatic fever L L R R 5 Rheumatoid arthritis R R R 6 Rheumatoid arthritis L L L 7 Probable Theumatoid arthritis R R 8 Intermittent hydrarthrosis R R 9 Conococcal arthritis L L 10 Gonococcal arthritis L L 11 Gonococcal arthritis L L L 12 Conococcal arthritis L L L L L 13 Gonococcal arthritis L L L 14 Staphylococcal arthntıs L L L 15 Infectious arthritis (negative culture) L L 16 Meningococcal arthritis R R 3 days 2 days 120 1 300 6 days 6 5 250 2 weeks 20 22 000 3 weeks 10 2 000 50 9000 10 days 60 1 300 1 week 60 41 000 9 days 25 13 200 2 weeks 2 1 000 5 days 150 22 300 I week 85 16 000 2 weeks 60 29 100 5 weeks 12 200 6 weekst 10 18 900 8 days 100 25 000 18 days 40 17 900 32 days 18 4 800 5 days 5 151 500 10 90 000 9 days 10 81 500 60 0 82 54 72 68 98 98 6 54 91 86 2 68 98 58 52 89 88 97 8 50 8 21 20 20 1 2 79 46 6 14 98 32 2 42 48 2 3 1 32 42 10 25 8 12 1 0 15 0 3 0 0 0 0 0 0 9 9 2 8 days 2 days 4 days 2 days 4 days 35 20 000 1 17 500 60 289 000 18 14 600 98 75 97 98 0 15 0 0 2 10 3 2 P = polymorphonuclears L-lymphocytes M=mononuclear phagocytes Treated with sulfanilamide Exacerbation of joint tymptoms See also Figure 5 and Figure 8 (Table III cont nued on p 28) ·SYNOVIAL FLUID CILANGIS 24 synovial cells 43 per cent Similar nucleated cell counts were reported by Labor and von Balogh'la (10 to 20 cells per cu mm ), McEwen -8 (125 to 200 cells per cu mm) and Klingi (10 to 50 cells per cu mm) Daviesas found a range of cell counts from 80 to 481 in 7 normal human fluids It is of interest that his studies in fluids from animals & revealed lower cell counts in the joints of the extremities than in those of the head and neck Variations in cytology in joint effusions are related not only to the type of joint disease but also to its seventy and duration By means of the total leukocyte and absolute polymorphonuclear cell counts, Group II Guids can be differentiated from those of Group I in almost all cases (Figures 5 and 6 and the chart at the back of the bool) As shown by Figure 5, the total count was below 3,000 in 133 of 142 Group I luids and above 3,000 in 363 of 405 Group II fluids Simlarly, the absolute polymorphonuclear count was below 500 in Group I fluids and above 500 in Group II fluids (Figure 6) Differentiation within either Group I or Group II cannot usually be made on the basis of cytological findings alone because of their variation with the severity of the disease and the duration of the effusion The cell counts of individual fluids in either group do, however, occasionally and in differential diagnosis For erample, neuroarthropathy, rather than traumatic arthritis, is sug gested by a fud with a low total leukocyte count but 40 to 50 per cent of polymorphonucleurs Likewise, 1 total leukocyte count under 3,000 with few or no polymorphonuclears is more com patible with disseminated lupus than with rheumatoid arthritis and accords with the findings in blood and other body fluids in lupus Similarly, total leukocyte counts above 100,000 have been found only in infectious arthritides The correlation of cytological findings with the seventy of the inflammation in the involved joints was well demonstrated in the 5 cases of gonococcal arthntıs in which aspirations of joints were performed at intervals during improvement (Table III) In Case 11, for example, the counts fell steadily from 41,000 per cu mm with 98 per cent polymorphonuclears to 1,000 with 6 per cent polymorphonuclears after 2 weeks The changes in Since this study was completed I hud with a lew ocyte count of 116 200 has been obtained from a patient with theumatoid arthritis ·VARIATIONS WITHI DISEASE 27 TAPLE IV Comparison of 9 Fluids with Positive Cultures and 33 Fluids with Negative Cultures in Gonococcal Arthritis Negative culture Positive culture Malmam Muimum Aytrigo Mislaum Murimum Average 1 535 108 000* 14 070 31 500 290 000 106 530 98 65 90 99 95 49 113 83 0 83 29 Total leukocy to count, cells per cu. mm Polymorphonuclears per cent Sugar mg per 100 cc Relative viscosity at 25 C Total protein gm per 100 cc Clobulin, gm per 100 cc Mucin nitrogen gm per 100 cc 3 19 12 41 43 69 53 53 60 56 13 31 20 2.2 27 25 016 158 061 0 077 041 "Patient had been receiving rulfanilamide before the fuld was cultured Value determined in only 1 Auld. In rheumatoid arthntis it is frequently impossible to evaluate the degree of joint inflammation in any one joint, and as a result correlation of the cytological findings with the seventy of the disease process in the joints is difficult There is little evidence of any variation with seventy when cell counts are compared with estimates of the severity of the disease as a whole in individual patients (see Table VII) A possible relationship between cytology and the estimated severity of the process in individual joints has, however, been indicated in some cases These patients have been followed for several years after the initial aspiration, and the course of the disease in the aspirated joint has been used as indication of the severity in that point at the time it was tapped It appears from the cases thus studied that the height of the total leukocyte and absolute polymorphonuclear counts early in the disease may correspond to the seventy of the joint involvement and indicate, to some extent, the prognosis For example, two of the highest fluid counts in the series (66,000 and 40,000 cells per cu mm ) were from knees which have since become progressively worse, with increasing thickening of penarticular tissues and narrowing of the joint spaces Fluid aspirated 5 years after the initial joint tap was very turbid and contained free fat and cholesterol crystals, both indicative of marked tissue destruction ·26 SYNOVIAL FLUID CHANGES TADLF III Vanations in Cytology of Synovial Flud with Duration of Effu sion (continued) Case number Duration of effusioa Total Amount, cu mi per Leukocytes per centi P Diagnosis Joint 16 Meningococcal arthntis knee (continued) R 0 3 8 2 0 3 17 Meningococcal arthritis ARRELLRRRRRRR 13 days 20 days 25 days 8 days 8 days 10 days 10 hours 30 hours 2 days 8 days 15 days 19 days 21 days 20 54 400 15 48 000 2 21 850 20 38 600 7 61 200 4 49 800 10 159 500 65 76 500 45 38 950 60 25 750 32 24 000 12 8 800 5 950 95 95 87 96 99 95 98 93 95 99 96 77 88 5 2 5 2 1 2 1 7 1 0 3 14 13 0 1 1 2 9 19 18 Menngococcal arthrias Elbow L L 1 day 6 days 8 days 15 days 16 days 1 day 6 days 12 days 17 days L L R 23 91 000 8 124 500 20 50 800 2 53 500 6 36 050 10 81 000 25 124 000 8 35 400 99 97 08 95 84 98 99 84 82 0 0 3 9 2 7 0 2 1 14 R R 23 600 hnee 1 day 4 days 28 50 87 000 47 500 98 96 0 4 L Knee 19 Reiters syndrome 70 12 89 92 90 1 0 L L L L L ܐܐ 4 days 16 days 9 weeks 12 weeks 3 days 4 weeks 9 weeks 10 8 10 71 44 300 11 250 14 950 2,250 20 700 25 000 9 200 4 70 80 10 20 Reiter s syndrome 84 75 56 2 9 16 10 34 cytology indicated a steady decrease in the iniammation of the synovial tissues and paralleled the clinical improvement as manı fested by gradual subsidence of redness, tenderness pain and swelling (Some of the patients received only symptomatic treatment, while others were given sulfanilamide ) The influence of the severity of the joint involvement on cytological findings is apparent also in the marked difference between the counts in sterile fluids from gonococcal arthntis averaging 14 070 per du mm and those in the luids with positive cultures averaging 106,530 per cu mm (see Table IV) ·*** Rheumatoid Arthritis Probable Rheumatoid Arthritis Reiters Syndrome 25 16 younger Syphiltic Arthritis — Congenital GROUP I 16 13 Tuberculous Arthritis 8 Gonorrheal Arthritis TE 3 30000 Most DW992 IK06Z Septic Arthritis no Meningococcal Arthritis JB - ® fluids from traumatic arthritis of more than 3 months duration fuids from degenerative joint disease secondary to other joint abnormality o fluids from rheumatoid arthritis of mild severity o fluids from rheumatoid arthritis of moderate seventy O fuids with positive cultures • fluids of the types designated by the headings with the erception that the symbol indicates in traumatic arthritis fluids of less than 3 months duration in degenerabve joint disease, primary degenerative joint disease in theu matoid arthritis sex cre disease and in the infectious 2- thnt des fluids with negative cultures ·TICURE 5 Total number of Icukocytes per cu mm in synovial fluid from various joint discases The asterisks represent the minimum and maximum values in normal fluids Fluids No of Patients No of 63 **B0* 500 K 40 40 - 3000 10000 13000 20000 25000 30000 35000 140000 Total Leukocytes per Cu Mm 240000 Traumatic Arthritis 26 25 Traumatic Arthritis with Hemorrhage Osteochondratis Dissecans 10 OI GROUP I 7 7 Osteochondromotosis 30 28 Neuroorthropathy 29 27 Kuo Degenerative Joint Disease 9 6 Hypertrophic Pulmonary Osteoarthropathy 5 5 Hemophilc Arthritis 20 - Tomorrort Lupus Erythematosus Disseminatus 67 61 Gauty Arthritis From " 22 17 Rheumatic Fever 12 6 Hemorrhagic Villous Synovitis ·amma 00000 Rheumatoid Arthritis oog 181 136 ogoro Probable Rheumatoid Arthritis Aelter's Syndrome 45 A5 25 15 16 Syphilitic Arthritis Congenital GROUP I Tuberculous Arthritis Woo TTTT" Gonorrheal Arthritis Bao Septic Arthritis 41 36 Meningococcal Arthritis IB fluids from traumatic arthritis of more than 3 months duration e fluids from degenerative joint disease secondary to other joint abnormality O fuids from rheumatoid arthritis of mild severity o fluids from Theumatoid arthritis of moderate severity fluids with positive cultures • fluids of the types designated by the herdings with the exception that the symbol indicates in traumatic arthn tis, fluids of less than 3 months duration, in degenera tive joint disease, primary degenerativo joint disease in rheumatoid arthritis, scycre disease, ind in the infectious arthritides fluids with negative cultures ·FIGURE 6 Absolute number of polymorphonuclears per cu mm in synovial Suid from various point diseases The asterisks represent the minimum and maximum values in normal fluids Fluids No of Ponents No of ** 25 1000 2000 3000 300 00DG - 4000 00001 15000 20000 25000 0000 35.000 140000 40000 240 000 Absolute Polymorphonucleor Leukocytes perCu Mm Troumatic Arthritis ***** 26 23 Traumatic Arthritis with Hemorrhage Osteochondritus Dissecans 이이 GROUP I 0) 6 Osteochondromatosis 30 28 Neuroarthropathy 4 26 24 Degenerative Joint Disease Hypertrophic Pulmonory Osteoorthropothy o 6 Hemophilic Arthrills . Lupus Erythematosus Disseminatus Gouty Arthntis 22 17 Rheumone Fever 12 Hemorrhagic Viſious Synovitis ·33 VARIATIONS WITII DISEASE to Duration of Cffusion (Rheumatoid Arthnas) 100 cc VP Mucin Mucin Globulin Relative nitrogen glucosamine Type of gm per 100 cc viscosity gm per gm per mucin Fluid Serum et 38 C 100 cc precipitato 12 (8) 23 (8) 43 (7) 010 (18) 033 (5) (8) 18 31 99 049 042 P 23 34 276 097 054 G 13 (4) 26 (4) 186 (2) 020 (17) 038 (3) VP ( 5) 19 31 190 079 058 P 33 38 193 131 078 F 1.2 (15) 23 (11) 28 (19) 005 (26) 006 (5) VP (15) 19 30 11 052 028 Y 30 38 133 134 035 F 18 ( 3) 28 ( s) 63 ( 2) 028 (14) 019 (6) VP (4) 19 262 QGS 041 Р 20 3.5 46 130 067 G 13 (18) 30 (13) 37 (21) 032 (32) 008 (14) VP (20) 27 37 145 068 043 68 68 43 137 081 The type of precipitate is expressed in symbols having the followiog significance VP few or no flecks in a cloudy solution P=small frable masses in a cloudy solution Fasoft mass an clear or slightly cloudy solution Gtaght ropy clump in a clear solution. (See Figure 13 ) 33 Comparison of the cytological findings in individual rheumatoid effusions of varying duration gives no evidence of a relationship between the total leukocyte and polymorphonuclear counts and the duration of the effusion (Table V) nor do the series of leukocyte counts on fluids from a patient whose knees were aspirated 24 times in the course of 16 months (Table VI) Probably the cytological findings in these and in other chronic effusions are more influenced by temporary variations in the severity of the synovial tissue inflammation than by the duration of the effusion There is, however a relationshup between cytological findings and duration of effusion in the series of Huids obtained daily from a patient with generalized rheumatoid arthritis who had bilateral intermittent knee effusions recurring at constant intervals The effusions resembled those of idiopathic intermittent hydrarthrosis and appeared and disappeared in regular cycles with 6 days be tween the peaks (see Figure 7) In these effusions the total leu- Locyte count and percentage of polymorphonuclears rose during the penod of incretsing infiammation and then decreased rapidly during subsidence As the polymorphonuclears disappeared, ·32 SYNOVIAL FLUID CHANGES TABLE V Relation of Cytological and Chemical Findings in Synovial Fluid and Serum Protein Duration Total Polymorpho Sugar (exclusive of macin) of leukocytes nuclears mg per 100 cc gm per 100 cc effusion Clot per cu mn per cent Fluid Serum Fluid Serum 0 (18) 1 100 (20) 2 (19) 24 (12) 60 (12) 36 (16) 65 (14) Up to 1 5 9,210 58 76 93 49 74 1 week 42 700 92 132 125 58 82 1 week 0 (17) 450 (22) 0 (22) 61 (17) 84 (12) 30 (19) 61 (13) to 13 13 130 08 80 96 46 74 6 weeks 3 51 600 95 102 125 61 87 6 weeks 0 (36) 2020 (38) 5 (38) 27 (28) 78 (23) 33 (27) 70 (19) to 12 15 060 62 66 96 47 75 6 months 3 62 200 91 116 129 62 87 6 months 0 (11) 2 500 (18) 8 (16) 0 (14) 75 (11) 41 (14) 66 (9) to 14 12,400 73 63 93 47 73 1 year 3 38 850 96 100 108 61 83 0 (34) 610 (54) 3 (51) 0 (32) 68 (26) 36 (34) 68 (22) 1 year + 05 16,775 66 58 91 52 79 3 66 000 94 102 116 89 101 The significance of the figures in each division is indicated in the following diagram Minimum (Number of Fluids) Average Maximum In contrast, one of the lowest counts in the series was in knee Aud from a patient who has since developed spondylıtıs but has had no evidence of inflammation in his knee for the past 10 years However, there are many exceptions to this apparent relationship and further study is necessary to determine how closely the cytological findings correspond to the severity in rheumatoid arthritis The response of the fluid cytology to administration of adreno corticotrophic hormone or of cortisone gives further suggestion of a relationship with severity of tissue inflammation Either the total leuhocyte count or the percentage of polymorphonuclears fell in almost all cases, the greatest decreases being from 30,800 cells per cu mm with 16 per cent polymorphonuclears to 150 cells with no polymorphonuclears and from 15,750 to 100 cells with 76 and 4 per cent polymorphonuclears respectively ·Alght Knio Lett Kru 100 capog BO sapoo Absolute Number of Cello per tu m. Milfgrams of Sugar per 100 CC 40 200000 rapool 6 Top 2 2 $ Cycle of El 10 Doy 3 5 2 2 0 5 0 X Sugar Tatal Leukocytes Polytr orphomocloor Lavkotytos O Mononucleor Phagocytes Olymphocytes FIGURE 7 Cytological findings and concentration of glucose in series of fluids obtained daily from a patient with generalızed rheumatoid arthritis who had bilateral intermittent knee effusions recurring at regular intervals Peaks of effusion occurred six days apart on day 3 of the cycle A series of five daily taps was made on each knee, omitting the left knee on day and the right knee on day 2 of the cycle of effusion (See Table IX) ·34 SYNOVIAL FLUID CHANGES TADLE VI Total Leukocyte and Polymorphonuclear Counts in 24 Fluids Aspirated from knees of a Patient with Rheumatoid Arthritis Over the Course of 16 Months Leukocytes per cu mm Polymorphonucleus per cent Dato Right Right 9 100 89 12 100 25 000 80 94 8 000 77 8,200 67 84 12 100 18 400 12/30/37 1/11/38 1/20738 1/21/38 1/25/38 2/ 5/38 2/11/38 2/14/38 2/18/38 2/26/38 3/ 1/38 3/ 9/38 3/10/38 3/24/38 4/ 1/38 4/12/38 4/22/38 4/24/38 11/30/88 12/ 3/38 12/ 7/38 4/ 7/39 4/ 8/39 4/10/39 16 100 14 800 17 800 15 800 13 400 27 200 13 500 19 000 72 78 88 72 86 76 90 80 9 600 9 300 75 21 000 86 10 100 45 15 500 51 11 500 63 16 500 90 88 25 800 11 400 monocytes and clasmatocytes increased, while the absolute number of lymphocytes remuned approximately constant The cytological variations that are related to the duration of the effusion are due essentially to changes in the degree of inflammation in the synovial membrane and give, therefore, further indication of the correlation with severity A similar rise in percentage and absolute number of polymorphonuclears during increasing synovial inflammation and a subsequent replacement by monocytes and clasmatocytes and eventual replacement in part by lymphocytes was found in other types of joint disease (see Table III) This sequence of changes occurred most rapidly and was most marked in rheumatic fever It has been observed in this laboratory in 2 cases of traumatic arthritis with hemorrhage, in 2 cases of rheumatic fever in 3 cases of rheumatoid arthritis in l case of probable rheuma·foo II 0001 10000 X 80 9 000 MILIGRAMS OF SUGAR PER 100 CC 8 000 60 7 000 6 000 L 40 5 000 ABSOLUTE NUMBER OF CELLS PER CU MM. 4 000 3 000 2,000 1000 INTERIM - 6 DAYS B 2 3 4 O DAY OF EFFUSION CYCLE OF EFFUSION O LYMPHOCYTES O TOTAL LEUKOCYTES • POLYMORPHONUCLEAR LEUKOCYTES MONONUCLEAR PHAGOCYTES X SUGAR FICURE 8 Cytological findings and concentration of glucose in 1 senes of fluids obtained daily from a patient with intermittent hydrarthrosis ·36 SYNOVIAL FLUID CHANGES toid uthntıs, in 2 cases of Reiters syndrome, in 5 cases of gonococcal arthritis, and in 1 case of specific infectious arthritis with negative culture In meningococcal arthntis, the cytologi cal changes during decreasing inflammation are similar in type but less marked and take place much more slowly than in gonococcal uthntis Similar cytological response was noted in 1 effusion in a case of idiopathic intermittent hydrarthrosis (Table III) In 2 other effusions in intermittent hy drarthrosis (1 in the above patient), there was no definite initial polymorphonuclear response, but the usual rise in monocytes occurred as the effusion increased (Figure 8) Krummel,17 in 1 case of intermittent hy drarthrosis, found an increase in polymorphonuclears and monocytes with increasing effusion and a rise in lymphocytes with subsiding effusion It is apparent that the polymorphonuclear rise is not always detectable in intermittent hydrarthrosis, but there is always a response of mononuclear phagocytes The initial rapid nse in polymorphonuclears and subsequent replacement by mononuclear phagocytes and eventually by lymphocytes found in so many types of joint disease is the same as the cytological reaction produced by the injection of normal saline into a rabbits knee joint 205 Apparently such mild imtation produces sufficient change in the subsynovial capillanes to allow for an immediate intra articular migration of polymorphonuclear leukocytes, whereas the mononuclear phagocytes ansing from the synovial tissue and tissue fluds are much slower to respond 1052 Further evidence that this cytological picture represents the usual reaction of synovial tissue to irritation is obtained from reports by other investigators:8 68 103 118 119 163 of a predominance of polymorphonuclears in acute effusions of vanous types and a preponderance of lymphocytes in chronic joint conditions with less acute inflammation The magnitude of the cellular response yanies with the degree of tissue irritation but the pattern apparently is similar in all cases and represents the fundamental cytological reaction to inflammation which was first descnbed by Bortel,"? Durhams and Beattie 14 Menkin_and Warner130 con cluded from the results of studies of inflammatory reactions produced in the pleural cavities of dogs, that this cytological sequence is conditioned by the local pH of the inflamed area, ·VARIATIONS WITII DISEASE 39 PROTEIN In normal human fluid, the total protein content (exclusive of muçin) varied in our series from 107 to 2 13 gm per 100 cc Most of this protein is albumin The albumin/globulin ratio in the only normal fluid analyzed was 20 4 174 Similar values for total protein were found by Fisher07 (16 per cent) and Honye" (045 to 3 15 per cent) The concentration of total protein in joint effusions varies markedly and shows significant relationships with the type of joint disease and the duration and seventy of the process Group I fluids, in general, have concentrations two to three times higher than normal but lower than those of Group II fluids, the maximum in Group I being 55 gm per 100 cc (Figure 9 and the chart at the back of the book) Fluids from cases of osteochondromatosis and lupus erythematosus disseminatus have lower protein contents than any other pathological fluids, containing only slightly more than normal fluids (averaging 22 and 25 gm per 100 cc ) Slightly higher values are found in fluids from degenerative joint disease, hypertrophic pulmonary osteoarthropathy, osteochondritis dissecans, Charcot joints, rheumatic fever, traumatic arthritis (including derangement of the semilunar cartılage) tumors and gout Traumatic effusions that are hemorrhagic and those of over 3 months duration show a slightly higher average protein content than other traumatic fluids The effusions of hemophiliac arthritis, which resemble Group I fluids in most respects contain much more protein Fluids of Group II have a higher average concentration of total protein (4 9 gm per 100 cc) Variations are more marked in rheumatoid fluids than in any other group with a minimum of 27 gm per 100 cc, a concentration typical of fluids of Group I and a maximum of 8 89 the highest protein concentration found in any fluid of any kind Analysis of these variations in rheumatoid cases shows that there is a tendency toward correlation with the severity of the disease and its duration and with duration of the effusion In this series the differences between the average protein contents of effusions from severe (52 gm per 100 cc) moderate (50 gm ) and mild cases (46gm ) were statistically significant at the 5 per cent level (Table VII) Similarly, the fluid protein concentration increased slightly with duration of the disease (Table VIII) and ·38 SYNOVIAL FLUID CHANGES polymorphonuclear cells predominating when the exudate is more alkaline than pH 70 Eosinophils, which are not found in normal fluid, were found in only 13 fluids in the present series, and in all but 2 cases were present in a concentration of only 1 per cent In all but 3 of these cases the fluids contained blood, 3 of the patients having hemo phiha, 4, traumatic arthritis with hemorrhage, 1, hemorrhagic villous synovitis, 1, rheumatic fever, 1, neuroarthropathy, 1, trau matic arthntıs, and 2, rheumatoid arthritis Erythrocytes Erythrocytes are not normal constituents of synovial fluid ""! They may be noted in varying numbers in pathological effusions but are usually not of diagnostic significance However, aspiration of hemorrhagic Fluid in the absence of a history of recent severe trauma or hemophila would suggest the presence of a tumor of the type classified as xanthom tous giant cell tumor or hemorrhagic villous synovitis or possibly a Charcot joint In summary, it can be concluded that an effusion in which the leukocyte count as above 5,000 per cu mm or the absolute polsmorphonuclear count above 500 per cu mm is probably not trau matic in ongin The total leukocyte and the polymorphonuclear counts rise with increasing severity of sy novial tissue inflamma tion and can be used as indications of the course of the synovitis in acute effusions especially in infectious arthritis gm , SPECIFIC GRAVITY AND TOTAL SOLIDS The content of total solids in normal human fluid in our series ranged from 240 to 4 83 gm per 100 with an average of 341 gm The normal specific gravity has not been determined In fluids obtained from patients with edema, it ranged from 1008 to 1012 Horiyeß reported total solids of 1 20 to 390 per cent and specific gravities ranging from 1 008 to 1 015 for normal human fluids obtained post mortem The figure given by Fisher87 for total solids is sughtly higher (44 per cent) The specific gravity and content of total solids in fuds of Group I are much higher than those of normal fluid but tend to be slightly lower than those of Group II fuids This is in accord with the difference in protein content which will be discussed later ·Rheumatoid Arthritis 0000 Probable Rheumatoid Arthritis 37 17 21 OI Relter's Syndrome - Syphilitic Arthritis Congenital GROUP I - 6 Tuberculous Arthritis 6 9 5 Buy 318 Gonorrheal Arthritis 27 25 8 eeeeee Septic Arthritis 16 Meningococcal Arthritis 2 fluids from traumatic arthritis of more than 3 months duration o fluids from degenerative joint disease secondary to other joint abnormality o fluids from rheumatoid arthritis of mild severity o fluids from rheumatoid arthritis of moderate severity o fluids with positive cultures • fluids of the types designated by the headings with the cxception thnt the symbol indicates in traumatic arthri tis fluids of less than 3 months duration in degenera tive joint disease, primary degenerative joint disease in Theumatoid arthritis severe disease and in the infectious arthritides, fluids with negative cultures # 1 fluid from a patient with severe renal disease due to amyloidosis ·FIGURE 9 Concentration of total protein (exclusive of mucın) in synovial fluid from various joint diseases The asterisks rep resent the minimum and maximum values in normal fluids NON Fluids No of Patients No of I 5 * 17- 5 5 60 63 70 75 Total Protein Content (-Mucin) Gm per 100 CC II O 18 19 Traumatic Arthritis 30 13 mot 12 Traumatic Arthritis with Hemorrhage Osteochondritis Dissecans w 3 GROUP I 5 Osteochondromatosis og grupo 21 20 WW 15 13 Neuroarthropathy Degenerative Joint Disease Hypertrophic Pulmonary Osteoarthropathy 5 u E 3 Hemophilic Arthritis 6 Lupus Erythematosus Disseminatus 17 13 Gouty Arthritis 18 Rheumatic Fever Hemorrhagic Villous Synovitis ·to Severity of Disease (Rhcumatoid Arthritis) 23 Globulin Relative Mucin Mucin Type of gm per 100 cc viscosity nitrogen glucosamine mucin Fluid Serum at 38 C gm per 100 cc go per 100 cc precipitato 10 (17) 20 (34) 39 (19) 007 (41) 028 VP (20) 18 27 12 1 073 046 34 28 166 072 G 12 (34) 20 (88) 34 (23) 005 (67) 022 (13) VP (26) 23 31 13 5 060 037 F 47 51 46 137 081 G 08 (7) 24 (41) 35 (8) 010 (17) 006 (7) VP (7) 28 37 71 001 038 VP 68 71 97 131 025 F The type of precipitate is expressed in symbols having the following significance VP-few or po flecta in a cloudy solution Presmall friable masses in a cloudy solution Fesoft mass in a clear or slightly cloudy solution Getight ropy clump in a clear solution (See Figure 13 ) Serum to Duration of Disease (Rheumatoid Arthritis) Clobulin Relative Mucin Mucin Type of gm per 100 cc Viscosity nitrogen glucosamine mucin Fluid Serum at 38 C gm per 100 cc gm per 100 cc precipitate 1.2 23 (7) (1) 050 (10) P (2) 14 28 55 087 F 15 35 131 10 (13) 20 (26) 35 (14) 007 (28) 006 (5) VP (17) 19 28 90 051 025 P 25 48 28 134 036 G 13 (19) 24 (38) 28 (19) 005 (51) 022 (9) VP (17) 26 34 13 6 063 048 F 45 56 46 130 081 G 08 (18) 20 (57) 34 (16) 006 (28) 006 (12) VP (15) 23 32 129 071 040 F 68 68 43 137 076 с The type of precipitate is expressed in symbols having the following significance VP=few or do fecks in & cloudy solution P=small Enable masses in a cloudy solution F-soft mass in a clear or slightly cloudy solution Gbght, ropy clump in a clear solution (See Figure 13 ) ·TABLE VII Relation of Cytological and Chemical Findings in Synovial Fluid and Serum Protein (ezclusive of stacin) grat per 100 cc Serum Total Polymorpho Susu leukocytes Auclear trg per 100 cc Severity per cui mm per cent Fluid Serum 450 (62) (59) (35) 60 (32) Mild 13,420 55 80 08 52 800 86 126 129 010 (101) 3 (95) (67) 58 (48) Moderate 15 620 68 60 93 80 000 86 132 122 1 750 (28) 44 (20) 0 (19) 75 (14) Severe 18,380 76 50 93 62 200 92 101 208 The significance of the figures in each division & ladicated in the following diagram 35 (42) 40 80 33 (71) 50 70 27 (17) 5.2 89 85 (58) 74 83 48 (127) 75 59 (53) 76 104 Minimum (Number of Fulds) Average Maximum TABLE VIII Relation of Cytological and Chemical Fundings in Synovial Flud and Protein Duration Total Polymorpho- Sugar (exclusive of nucia) of leukocytes nuclears mg per 100 cc gm per 100 C disease clot per cu mm per cent Fluid Serum Fluid Senima 1 (9) 3,200 (14) Up to 36 (14) 24 (8) 80 ( 6) 30 (11) 61 (11) 17,114 75 6 weeks 73 92 45 72 3 45 000 95 101 100 51 80 6 weeks 0 (32) 1 100 (41) 8 (38) 9 (28) 60 (23) 35 (30) 82 (36) to 1 17,252 65 70 95 49 75 3 62 200 96 116 129 58 86 0 (60) 450 (71) 0 (88) 0 (40) 68 (28) 36 (53) 59 (60) 08 17,005 65 59 94 49 78 5 years 3 66 000 96 102 121 66 0 (39) 610 (49) 3 (48) O (38) 75 (28) 27 (30) 48 (67) 5 years + 11 12 798 63 71 94 50 76 3 43 800 95 132 125 89 101 The signi6cance of the figures in each division is indicated in the following diagram 1 year 1 year Miningum (Number of Fluds) Average Maximum ·TADLE I Variations in Concentrations of Albumin Globulin and Mucin in Intermittent Bilateral Knee Effusions in a Patient with Rheumatoid Arthritis Day of Amount 1 CC Total protein (exclusive of mucın) gm /100 cc RE. FI Lt F1 Mucin nitrogen &m/100 cc Glabulin gm /100 cc Rt. Fl. effusion Lt. FL. Rt F1. So Lt So RE. Lt, FL Lt. Rt 59 63 30 42 066 025 3 4 5 0 1 2 37 35 41 cot OCRA 20 18 20 13 12 no tap 20 13 1 25 60 no tap 91 87 87 85 85 65 60 61 84 66 48 45 45 41 40 062 087 060 016 038 60 65 40 30 039 031 «Seo also Figure 7 Not all tho fud was aspirated when over 20 cc were present in the joint ·SYNOVIAL ILUID CELANCES 44 with durtion of effusion (Table V) The variations may actually all be related to varying severity, since the cases of longest dura tion of effusion or discase were in general the more severe cases The protein concentration in a chronic effusion in rheumatoid arthritis may remain at essentially the same level over a penod of months or yeurs In one patient in whom 24 aspirations were made in 16 months, the total variation in protein concentration was from 36 to 44 gm per 100 cc, with the exception of one value of 49 gm obtained 2 days after the knee had been imi gated with 09 per cent salınc solution Fluids from 4 consecu tive taps in 1 month contained 39, 38, 39 and 3 8 gm of protein per 100 cc Several months after the above scries of aspirations, the knee effusion disappeared entirely for 2 years but recurred when the patient had a generalized exacerbation of her arthritis It is of interest that the fluid at the time of the exacerbation had essentially the same content of protein as the original effusion, 40 gm per 100 cc Similar constancy of protein concentration in rheumatoid effusions of long duration has been seen in several other patients in whom repeated joint aspirations were made In rheumatoid patients with intermittent effusions, on the other hand, fluctuations in the fluid protein concentration occurred, corresponding apparently with variations in the inflammatory reac tion in the synovial tissues This was apparent in the patient de scribed above who had bilateral knee effusions in regular cycles with peaks 6 days apart (Table IX) The unit concentration was greatest when the effusion first reappeared or was disappearing With increasing effusion the concentration decreased, suggesting that the entering liquid was slightly lower in protein than the flud present in the joint With loss of fluid the protein concentra tion increased The contrasting lindings in the case of mucın are significant The unit mucin concentration rose with increasing effusion suggesting that the entering fluid carried a relatively high concentration of mucin The decreasing concentration of mucın as the effusion subsided would suggest that no new mucın was entering as the fluid was leaving and that the mucin present in the joint was being destroyed or removed The total globulin fraction shows a greater correlation with etiology, severity and duration of the disease and duration of the effusion than the total protein or the albumin fraction In general, ·VARIATIONS WITII DISCASE 47 creasing duration of discise and duration of effusion is apparent from Tables V and VIII and is probably related fundamentally to increased seventy, as has been discussed above The highest globulin concentrations in this series occurred in effusions of over 1 years duration in patients with severe rheumatoid arthritis The maximal globulin concentration in any other type of fluid was 30 gm per 100 cc, while 6 of 48 rheumatoid fluids showed concentrations above 30 gm , 1 being 677 gm per 100 cc Of the 6 effusions with high globulin content, 5 were of over 1 year s duration and represented 28 per cent of the 18 eſfusions of long duration that were analyzed In contrast, only 1 of 30 effusions less than 1 year s duration (3 per cent) contained over 3 gm of globulin per 100 cc It is of interest that the 1 effusion of short duration with high globulin content occurred in a knee which had first been swollen and painful 2 years previously but had apparently been symptom free until 9 or 10 months before aspıration A similar, though less marked, increase in globulin with persistence of effusion was found in traumatic effusions The average globulin concentration in these fluids of less than 3 months duration was 0 80 gm per 100 cc, in contrast to 170 gm in the 1 effusion of over 1 year s duration The corresponding albumin/ globulin ratios were 485 and 3 26 The chief factor determining the levels of albumin and globulin in joint effusions appears to be the degree of permeability of the synovial tissues There is no consistent relationship between the relative concentrations of these proteins in fluid and in serum In mild inflammation, the fud protein tends to remain low (averaging 35 gm per 100 cc) and the albumin/globulin ratio is high (above 20) With greater inflammation as in the majority of rheumatoid joints, the permeability is increased so that the fud protein concentration rises and the ratio of albumin to globulin decreases, approaching that of the serum If effusions in theumatoid arthritis persist for months or years the total protein content rises only slightly but the albumin/globulin ratio decreases markedly The luid globulin concentration becomes approximately the same as that of the blood, while the flud albumin remains only two thirds as high as the serum albumin Such high globulin concentrations suggest that the prolonged severe inflammation has led to changes in the tissues that make ·46 SYNOVIAL FLUID CIIANGES fluids with relatively low total protein concentrations, such as traumatic effusions, have low globulin concentrations with albu min/globulin ratios ranging from 20 to 80 Apparently the relative impermeability of normal synovial tissues to globulin as compared with albumin is maintained even in the presence of the sughtly increased permeability resulting from mild inflammation Further evidence of this differential permeability to individual protein fractions is found in the electrophoretic patterns of synovial fluid from traumatic arthritis In the 7 cases studied in this laboratory, the albumin was always higher in the Suid than in the serum, alpha-l and beta-globulins were essentially the same in both and alpha 2 and gamma globulins were lower in the flurd 148 Schurch, Viollier and Sullmann'1 reported that the albumin/ globulin ratios in traumatic fluids were the same as those of the corresponding sera but the concentration of alpha-2 globulin was lower in fuid than in serum Olhagen"iia found the relative distribution of proteins in traumahc Suids about the same as in the plasma with slightly higher concentrations of albumin and gamma-globulin in the fluid With more severe inflammation, and consequently higher fluid protein concentrations, as in the ma jority of Group II luids, the proportion of globulin in the entering protein increases, with the result that the albumin/globulin ratio falls below 20, occasionally below that of the serum This is explicable by the fact that the differential permeability to albumin and globulin does not persist when the membrane becomes easily per meable to proteins The loss of differential permeability is apparent in the electrophoretic patterns, which usually show essentially equal concentrations of albumin and gamma globulin in the fluid and serum The concentration of alpha 2 globulin remains lower in the fluid than in the serum The relationship between the fluid globulin concentration and the severity of the joint involvement was apparent in effusions from gonococcal arthntis Fluids from which positive cultures were obtained had an average globulin content of 25 gm per 100 cc in contrast to 20 gm in those with negative cultures (Table IV) In rheumatoid effusions, also, increased globulin concentration was one of the most consistent evidences of greater seventy of the joint disease (Table VII) A tendency towırd increasing globulin concentration with in ·**T 74706 3.62 469 58 42 Rheumotoid Arthritis 13 Probable Rheumatoid Arthritis 21 10 Reifer Syndrome 7 Syphilitic Arthritis Congenital GROUP II 4 Tuberculous Arthritis 15 13 Gonorrheal Arthrills BB CD 0 Septic Arthritis - Meningococcal Arthritis @ flurds from traumatic arthritis of more than 3 months duration e fuids from degenerative joint disease secondary to other joint abnormality O fluids from rheumatoid arthritis of mild severity o fluids from rheumatoid arthritis of moderate severity fluids with positive cultures fluids of the types designated by the headings, with the exception that the symbol indicates in traumatic arthnı tis fluids of less than 3 months duration in degenera tive joint disease, primary degenerative joint descase in rheumatoid arthritis severe discase, and in the infectious arthritides, fluids with negative cultures # a fluid from a patient with sciere renal disease due to amyloidosis ·Figure 10 Concentration of globulin in synovial fluid from various joint diseases The asterisk represents the value in one normnal flud Fluids No of Patients No of * 85 - Globulin Content Gm per 100 CC 25 50 יויז'ן 7 7 Traumatic Arthritis 10 Traumatic Arthritis with Hemorrhage Osteochondritis Dissecans - GROUP I NO Osteochondromatosis 16 15 Neuroarthropathy UD 6 Degenerative Joint Disease Hypertrophic Pulmonary Osteoarthropathy U w € Hemophilic Arthritis P s 919 Lupus Erythematosus Disseminatus 12 word Gouty Arthritis OI 8 7 Rheumatic Fever 3 3 Hemorrhagic Villous Synovitis ·VARIATIONS WITHI DISCASE 51 This increase in serum protein in rheumatoid arthritis is chiefly in the globulin fraction, which tends to rise steadily with duration of disease Only 12 of 33 cises of less than 1 years duration had globulin contents over 300 gm per 100 cc in contrast to 39 of 57 cases of over 5 years' duration (Table VIII) Greater severity of the disease also tends to cause a higher level of globulin in the serum (Table VII) The average concentration in 34 mild cases was 27 gm per 100 cc compared with an average of 37 gm in 41 severe cases The average globulin fraction in the 163 cases of rheumatoid arthritis in which it was determined was 3.2 gm per 100 cc with a maximum of 7 10 and a minimum of 20 gm Similar increases in serum total protein and globulin concentrations with increasing severity of rheumatoid arthritis have been found by other investigators The average globulin concentration in mild cases in Davis 54% series was 28 gm per 100 cc compared with 39 gm in severe cases Scull, Bach and Pemberton!79 found that the globulin concentration was above 32 gm per 100 cc in 4 of 27 moderately severe cases in contrast to 30 of 48 severe cases Davison, Woolley and Donovan"' reported a range of globulin concentrations from 04 to 29 gm per 100 cc in 8 clinically inactive cases as compared with a range from 03 to 51 in 43 active cases Wallis 04 also found a correlation between the globulin content of serum and the activity and extent of disease the globulin being above 30 gm in 11 of 19 cases with widespread involvement in contrast to l of 11 with slight or moderate involvement Electrophoretic patterns of plasma or serum from patients with rheumatoid arthritis have shown that elevation of the concentration of globulin fractions and lowering of the albumin concentration have been associated consistently with activity of the disease With few exceptions both the alpha- and gamma globulin components have been increased in most series 63 123 127 140 141 LIS 195 196 204 Apparently the first change is an increase in the alpha globulins, which may or may not persist, with a subsequent rise in the gamma globulin A decrease in albumin and an increase in alpha- or gamma-globulins were found in all but 3 of 36 cases 109 This group included many early and mild cases in which no alteration of albumin/globulin ratio was detectable by the Howe method 14 ·SYNOVIAL FLUID CHANGES 189 50 the removal of globulin (which is known to take place by way of the lymphatics?8) even more difficult than it is normally The electrophoretic patterns of synovial fluids corroborate the fact that the globulin fractions increase with greater seventy of inflammation 180 In traumatic effusions, as stated above, the albu min is always higher in the fluid than in the serum and the globulin fractions are the same or lower in the fluid In 7 rheumatoid patients with early or mild involvement of the joint, 1 patient with gout, 1 with rheumatic fever, I with hypertrophic pulmonary osteoarthropathy and 2 with Reiters syndrome, the electropho retic pattern of the fluid resembled that of traumatic fluids The duff ces between the fluid and serum albumin levels, however, were not so great In cases of rheumatoid arthritis with greater severity of inflammation, the patterns showed the albumin con centration equal in the fluid and serum or lower in the fluid, and the gamma globulin higher in the fluid than in the serum Schurch, Vrollier and Sullmann"77 found similar changes in 1 prtient with rheumatoid arthritis and I with tuberculous arthritis, and Olhagen"la reported remarkably high concentrations of gammz-globulin in rheumatoid fluids The protein concentration of the serum of patients with rheumatoid arthritis increases with duration of disease but, as mentioned above, there is no exact correlation between protein changes in serum and fluid in the 172 patients studied in this laboratory, the protein concentration averaged 750 gm per 100 cc The minimal and maximal figures were 4 83 and 101 gm, but the majority of the values fell between 65 and 80 gm per 100 cc, correspondmg in general with those found by other workers & 48 49 168 179 197 The average concentration in rheumatoid arthritis of less than 1 years duration was 73 gm per 100 cc in contrast to an average of 76 gm in cases of over 1 years duration (Table VIII) Analysis of the individual figures shows that only 4 of 47 cases (85 per cent) of less than 1 years duration had total proteins above 800 gm, while a significantly greater per cent, 32 of 127 cases (25 per cent), of more than 1 years duration were above 800 gm, I being 101 gm A similar increase in serum protein with greater severity of disease was apparent The average concentration in mild cases was 74 gm per 100 cc, compared with 78 gm in severe cases ·VARIATIONS WITH DISEASE 53 tion In many fluids from Group II, however, numerous small flechs of fibrin were present in the fluid when it was withdrawn MUCIN AND VISCOSITY The average concentration of mucın in normal human fluid in our series is 0 104 gm nitrogen per 100 cc with a variation from 0068 to 0 135 gm nitrogen 171 Fisher® found 195 per cent of mucin, and Achard and Piettres reported a concentration of 57 per cent The mucin-glucosamine concentration in our series varied from 050 to 132 gm per 100 cc, with an average of 087 Ragan and Meyer168 found concentrations of hyaluronic acid varying from 0 04 to 0 33 mg per cc The average relative viscosity of the fluid in our seres is 235 at 38° C, with a range from 57 to 1,160 Schneider176 reported similar variation, from 39 to 1,490, Kling118 found viscosities of 107 and 20 in 2 normal fluids, and Ragan and Meyer158 reported viscosities from 23 to 575 Marhed variations from one joint to another have been found by Panizza145 and Davies 14 in animals Changes in the concentration of mucin in joint disease are related to the type, seventy and duration of the disease and the duration of the effusion In general, it may be said that the mucin content is low in fluids in which other signs of marked inflammation are present, namely, turbidity, high total leukocyte and polymorphonuclear counts, high total protein, high globulin and lowered sugar concentrations The mucin concentration in synovial fluid varies in different joint diseases (Figure 11) In traumatic fluids, it tends to be the same per cubic centineter as that of normal fluid in spite of a five to twenty five fold increase in the volume of fluid in the joint It is interesting to note that the unit concentration remains normal (0001 gm nitrogen as mucın per cc ) when the fluid is formed by diffusion, even though the swelling occurs rapidly, but is greatly reduced when the effusion is formed by hemorrhage with direct flow of blood into the joint This is in accord with the hypothesis that mucın is added to the dialysate of blood plasma as it diffuses through the tissues surrounding the joint Another possible explanation for the lower concentrations in hemorrhagic effusions would be the presence in the entering blood of an enzyme which could destroy muçin However, we have been unable ·52 SYNOVIAL ILUID CHANGES To summanze, the presence of a high concentration of protein in the fluid (above 55 gm per 100 cc ), and especially of globulin (above 21 gm per 100 cc), indicates that the inflammatory reaction in the joint is not traumatic in ongin The synovial fluid protein concentration nises with increasing duration of self-limited effusions or with increasing severity of disease FIBRINOGEN Normal synovial fluid contains no fibrinogen 111 Even in effusions, fibrinogen is seldom found in quantities large enough to be precipitable by the method of Cullen and Van Slyhe " It was determined in this way in only 6 fluids of the present senes, 1 each from rheumatoid arthritis, rheumatoid arthritis (diagnosis probable), gonococcal arthritis, Reiters syndrome, hypertrophic pulmonary osteoarthropathy and a case of malignant lymphoma of Hodgkins type Smaller amounts of fibrinogen, however, were found in many of the fluids, and could be estimated by the size of the clot formed The concentration of fibrinogen varies with the type and severity of the joint disease and the duration of the effusion The majority of fluids of Group I do not clot, and even when clotting does occur, they never for large, firm clots immediately In this series, fluids from patients with hypertrophic pulmonary osteoarthropathy, however, differed from those of Group I in this respect, since 4 of the 5 examined formed firm clots The content of fibrinogen in the plasma in these 4 cases was markedly elevated (averaging 0 60 gm per 100 cc ), whereas it was normal in trau matic cases (034 gm per 100 cc) The majonty of specific infectious and rheumatoid fluids coagulate, and in many cases rapidly form large, firm clots The average concentration of fibrinogen in plasma corresponding to these Group II fluids was 0 53 gm per 100 cc In rheumatoid arthritis the concentration in synovial fluid is apparently inversely related to the duration of the effusion, as is shown in Table V Rheumatoid fluids of more than 1 years duration show less tendency to clot than those of shorter duration The possibility of a preformed clot which could not be with drawn at aspiration W 25 suspected in only 1 case, a patient with hemophila who had been transfused a few hours before aspira ·0 YO Rheumatoid Arthritis 125 94 Probable Rheumatoid Arthritis 17 17 Reiters Syndrome 9 Syphilitic Arthritis Congenital GROUP I 9 8 Tuberculous Arthritis 8 725 Gonorrheal Arthritis 27 DD m 98899* Septic Arthritis 16 u Meningococcal Arthritis N 3 N fluids from traumatic arthritis of more than 3 months duration e fluids from degenerative joint disease secondary to other joint abnormality o fluids from rheumatoid arthritis of mild severity o fluids from rheumatoid arthritis of moderate severity fluids with positive cultures • fluids of the types designated by the headings with the exccption that the symbol indicates in triumatic arthri tis fluids of less than 3 months duration in degeneri tive joint disease primary degenerative joint disease in rheumatoid arthritis, severe discase and in the infectious arthritides, fluids with negative cultures ·ſicurc 11 Concentration of mucın nitrogen in synovial fuid from various joint discuses The asterisks represent the mini mum and maximum values in normal fluids 040 060 * 068 → 080 100 ** 103-> 120 140 160 220 N Mucin Content o Gm Nitrogen per 100 Cc TT 81 103 18 Troumatic Arthritis El 12 Traumatic Arthritis with Hemorrhage $ 3 Osteochondritis Dlasecans GROUP I Osteochondromatosis 20 19 Neuroarthropathy 15 13 Degenerative Joint Disease Hypertrophic Pulmonory Osteoarthropathy 5 2 N Hemophilic Arthritis 8 Lupus Erythematosus Disseminatus a Gouty Arthritis 19 Moon Rheumatic Fever Hemorrhagic Villous Synovitis ·VARIATIONS WITH DISEASE 57 results either in accelerated production or retarded destruction of these mucins Attempts to demonstrate a destructive action of thyroxin on synovial mucin in vitro have not been successful From histochemical studies, Asboe-Hansen' has concluded that hyaluronic acid is secreted by mast cells and that thyroid hormone regulates the amount of polysaccharide forned Further suggestion of hormonal regulation of mucin metabolism has been given by observations on the effect of sex hormones Ogston, Philpot and Zuckermans demonstrated that estrogenic stimulation produced not only an increase in the size of the con nective tissue cells but also a viscous, mucin containing exudate in the sexual skin of monkeys Bachman, Collip and Selye showed that a more generalized edema, not involving the sexual skin, was produced if estriol injections were continued after the primary swelling of the sexual skin had appeared and subsided Hardesty83 noted that the mucoid content of capons combs was greatly augmented by injection of male hormone and was also increased, though to a lesser extent, by female hormone Furthermore, the diminished spread of India ink in rabbits' skin after injection of estrone, demonstrated by Sprunt, McDearman and Raper,191 again suggests that estrin increased intercellular mucin Loeb Suntzeff and Burns122 observed an increase in the amount of fibrous hyalıne tissue between the connective tissue cells in the vagina of mice after injection of estrogenic substances Ap parently no evidence was obtained as to whether or not this increased intercellular tissue was mucin A marked increase in viscosity is produced in some rheumatoid fluids by large doses of adrenocorticotrophic hormone or of cortisone, as discussed below The extent to which the normal activity of the pituitary or the adrenal cortex regulates the production and degree of polymerization of mucın cannot be determined from this evidence Fluids of Group II have a lower average mucin concentration than traumatic fluds The concentration tends to decrease with increasing severity, the lowest values being found in severe cases of specific infectious or rheumatoid arthntis Twelve fluids (2 from severe cases of rheumatoid arthritis, 1 from a case with probable rheumatoid arthritis and 9 from specific infectious arthntis due to hemolytic streptococci staphylococci or gonococci) ·SYNOVIAL FLUID CHANGES 56 to demonstrate such an enzyme in normal blood or even in blood from patients with rheumatoid or other types of arthritis Exceptionally high concentrations of mucın were found in many fluids from Charcot joints and in fluds from lupus erythematosus disseminatus, degenerative joint disease and osteochondromatosis A fluid obtained post mortem from a knee which showed extreme degenerative changes but did not contain an abnormal amount of fluid had one of the highest concentrations in this series (0 208 gm nitrogen per 100 cc) The highest concentrations of all were found in cases of osteochondromatosis, in which the average (0166 gm nitrogen per 100 cc) was much higher than that of normal human fluid (0 104 gm ) The significance of extremely high mucın concentrations is not clear It is probable that they are due to increased formation of mucin in response to tissue stimulation without the increased destruction of mucın usually associated with marhed inflammation of joints Comparison of the viscosity estimated from the mucin-glucosamine content (see p 62 and Figure 3) with the observed viscosity of fluids from neuroarthropathy, lupus, degenerative joint disease and osteo chondromatosis supports this hypothesis and indicates that the mucın is not degraded That stimulation of synovial cells with increased production of mucın can be caused by trauma is indicated by King sl10 observations in acute and chronic inflammation of traumatic origin He demonstrated enlargement of the Golgi apparatus of the synovial cells and accumulation of intercellular mucoid material (Unfortunately, the mucın concentra tions of the fluids were not determined ) There is also the possi bility that the high mucin concentration arises in part from destruction of normal cartilage or chondromata However, the presence of synovial polysaccharide in cartılage has not been demonstrated by chemical or immunological means An abnormally high concentration of mucin was found also in the synovial fluid from a myredematous patient This suggests that the metabolism of synovial fluid mucin and that of the mucin of subcutaneous tissue undergo comparable changes when thyroid activity is reduced • A decrease in thyroid activity apparently •Connective tissue mucin from normal and myredematous animals has been shown to be similar to synovial mucın by physical charactenstics and by enzymatic 210 reactions in vitro ** ·VARIATIONS WITH PISEASE 59 mucin esther alone or associated with incomplete synthesis Attempts to demonstrate the presence of mucinase or a similar enzyme in such fluids have not been successful 131 170 It has been shown, however, that ascorbic acid hydrogen peroxide and serum with high alkaline phosphatase activity both cause depolymenzation of synovial fluid mucın in vitro It is possible that abnormality in the metabolism of one of these substances may in some cases explain the destruction of mucin in pathological fluids Lutwak- Mann- found no ascorbic acid in normal synovial fluid, but in Our series the synovial fluid from one patient with rheumatoid arthritis contained 11 mg per 100 cc An inverse relationship has been demonstrated between the alkaline phosphatase ac tivity and the viscosity of synovial fluid (see p 77) The findings in Group II fuids indicate that mucın destruction is more rapid than normal in many cases However, the breakdown probably represents merely a splitting of the macromolecules No free glucosamine, one of the hydrolytic products, is present even in the fluid from the most severe cases, showing that complete breakdown of the polysaccharide does not take place Whether the change in mucın is identical in all types of Group II fluids cannot be determined from the available evidence In infectious arthritis, the initial fluid, even from severe cases, usually contains fairly normal mucın that precipitates in a ropy clump Subsequently, the fluid becomes less viscous and the mucin precipitate becomes smaller and less cohesive In effusions in rheumatoid arthritis, on the other hand, the mucın precipitate is usually friable with a cloudy solution even early in the disease, except in very mild or atypical cases The nature of the precipitate obtained with acetic acid and the viscosity of the fluid give the best indication of the state of the mucin in any pathological fluid However, a method that may prove to be of value in estimating the degree of mucın degradation is a comparison of the relation between the viscosity of the flud and the mucin concentration with the similar relationship for normal cattle fuid (Figure 3) The empirical proportionality between the logarithm of the viscosity and the concentration of mucin glucosamine is apparently unaffected by the particular solvent or the state of the polysaccharide (Figure 3), even though these cause changes in the viscosity per unit of glucosamine The ·58 SYNOVIAL FLUD CHANGES showed only slight hazıness on precipitation with acetic acid, indicating merely a trace of mucin Of more significance than the changes in concentration are the alterations in the characteristics of mucin in pathological fluids Normal fluid has a viscosity ranging from 57 to 1,160, with an average of 235 at 38° C The mucin from such fluid is precipitated by acetic acıd as a tough, ropy mass, and the surrounding solution is clear (Figure 13a) The changes in joint disease consist of reduction in viscosity and loss of characteristic precipitability with acetic acid and are similar to those produced by the in vitro depolymerization of mucın by ascorbic acid hydrogen peroxide or by bacterial or tissue enzymes • The abnormalities are greater in the fluids of Group II than in those of Group I and increase, in general, proportionally with increasing seventy (Figure 12 and the chart at the back of the book) Traumatic arthritis fluids, for example, show normal unit con centration of mucın but viscosity reduced from the normal range to an average of 31 at 38° C with minimal maximal values of 84 and 83 The precipitability of mucın, however, remains fairly normal, ie, the addition of acetic acid to a final concen tration of 1 per cent causes the characteristic ropy precipitation with surrounding solution that is clear (See Figure 13a ) In specific infectious and rheumatoid fluids, on the other hand, the viscosity tends to be much lower, with averages of 14 6 and 131 respectively The precipitability shows even more marked altera tion than the viscosity When there is relatively little change from normal, the mucın is precipitated in a clump, but this is soft and friable and the solution is cloudy (Figures 13b and 13c) In the fluids with more abnormal mucin, a flocculent or slightly fibrous precipitate is obtained and the surrounding solution is very cloudy (Figure 13d) In extreme cases, no definite precipitate is formed, the solution becoming only slightly cloudy The similarity of these changes to those produced by enzymatic digestion of mucin suggests that the reduced concentration of mucın in Group II fluids is caused by increased destruction of *Synovial flud mucın can be depolymerized in vitro by three types of substances bacterial or tissue enzymes serum with high alkaline phosphatase activity and ascorbic acid hydrogen peroxide (See reviews of the subject " 181 170) Further breakdown with hydrolysts of the polysaccharide and liberation of reducing sugars and glucosamine is caused by the bactenal enzymes ·Rheumatoid Arthritis 69 63 Probable Rheumatoid Arthritis 28 28 Reiter's Syndrome 20 16 Syphilitic Arthritis - Congenital GROUP I Tuberculous Arthritis Gonorrheal Arthritis w 7 og Septic Arthritis Meningococcal Arthritis 14 ® fluids from traumatic arthritis of more than 3 months duration fluids from degenerative joint disease secondary to other joint abnormality o fluids from rheumatoid arthritis of mild severity o fuds from theumatoid arthritis of moderate severity fluids with positive cultures • fluids of the types designated by the heidings, with the exception that the symbol indicates in traumatic arthri tis, fluids of less than 3 months duration in degenera tive joint disease, primary degenerativo joint disease in rheumatoid arthritis, severe disease and in the infectious arthritides, fluids with negative cultures ·FIGURE 12 Relative viscosity at 38 C of synovial fluid from various joint diseases The asterisks represent the minimum and maximum values in normal fluids * Fluids No of Patients No of **1295 10 20 OC 40 50 60 Relative Viscosity at 38 C OLI - TI CI Traumatic Arthritis Traumatic Arthritis with Hemorrhage Osteochondr):18 Dissecans 4 GROUP I Osteochondromatosis 7 7 3270 7300 21470203 25 22 Neuroarthropathy 12 10 Degenerative Joint Disease Hypertrophic Pulmonary Osteoarthropathy 2 N V 3 3 Hemophilic Arthritis or 7 Lupus Erythematosus Disseminatus 17 EI Gouty Arthritis 15 Rheumatic Fever Q 7 Hemorrhagic Villous Synovitis ·b с d Ficune 13 Types of mucin precipitate obtained with acetic acid (the berkers contuning the mucin were placed on a black surface and the photographs were taken from above) a Tight ropy clump in 1 clear solution b Soft muss in a very slightly cloudy solution c Small friable masses in a clouds solution d Few flecks in a cloudy solution ·62 SYNOVIAL FLUID CHIANCES mucin glucosamine concentration of any fluid can, therefore, be extrapolated to the concentration chosen as a basis of companson (75* mg of glucosamine per 100 cc) and the corresponding vis cosity compared with that found for normal fluid at this standard glucosamine concentration For example, if a fluid has a mucin glucosamine concentration of 82° mg per 100 cc and a relative viscosity at 38° C of 166, extrapolation (using Figure 3) to a concentration of 75 mg per 100 cc gives a viscosity of 11 in contrast to the value of 50 found in normal cattle fluid at this concentration The disproportion between viscosity and mucin content is most marked in those fluids in which the type of pre cipitate obtained with acetic acid is indicative of the greatest breakdown of mucin However, even traumatic fluids, from which the mucın is precipitated normally, show lower viscosities than would be expected from the mucin concentration, suggesting that the first stage of mucın breahdown (presumably a partial splitting of the long chain molecule) has taken place The degree of degra dation of mucın in such fluids is apparently comparable to that produced by any of the usual methods of prepanng mucın from normal fluid The viscosity of solutions of mucin prepared by these methods is lower in proportion to the concentration than was that of the normal fluid from which it was obtained (Figure 3) The nitrogen/glucosamine ratio of the mucin precipitated by acetic acid from joint fluid is also indicative of the state of the mucin With incrcasing breakdown of the polysacchande, the ratio decreases from the normal of 15 In general estimates of the degree of mucın degradation provided by this method, by the relative viscosity of the fluid and by the type of mucın precipi tate agree fairly well (Table X) A reversal of the changes in muçin with improvement in the joint disease was apparent in this series in cases in which repeated aspirations were done In one patient with rheumatic fever, the synovial flud withdrawn 12 hours after onset of an effusion con tained 008 gm nitrogen as mucin per 100 cc, while fluid from the same knee 3 days later had a mucin mitrogen content of 0 15 gm per 100 cc The change paralleled a drop in the fluid leuhocyte count from 13 800 to 1980 cells per cu mm Similarly, fluid from a septic joint (Staphylococcus aureus) had no precipitable mucın 5 days after onset of the arthritis but showed a definite ·VARIATIONS HTII DISIAST 03 TABLE \ Comparison of Different Methods of Estimating the Concentra tion and State of Mucin in Various Types of Joint Disciscº Melative viscosity at 38 C Mucin nitrogen km /100 cc Vucin Type of glucoitmine mucin m/100 cc ttt Mucin nitrogen Glucosamine 77 Tuberculous arthritis 011 020 P 080 Reiters syndrome 10 5 007 052 1 38 Rheumatoid arthritis 110 007 0-10 P 1 38 Syphuhtic arthritis 111 058 039 112 Infectious arthritisi 137 059 017 Р 1 50 Pulmonary osteoarthropathy 153 017 037 1 10 Gouty arthnts 170 058 040 P 1 03 Hemophilia 210 011 р Rheumatic fever 220 109 093 с 1 52 Probable rheumatoid arthntis 240 008 0-18 1 44 Neuroarthropathy 31 0 009 010 1 38 Traumatic arthritis 310 OSO 098 050 Degenerative joint disease 41 0 090 080 094 Lupus erythematosus disseminatus 149 0 092 059 с 1 37 Normal human knee joints 2350 104 087 G 1 10 Myzedema 381 011 1921 с Osteochondromatosis 499 0 168 09911 G 14111 Arranged in order of increasing viscosity not all the columns contain equal numbers of Aulds The type of precipitate is expressed in tymbols having the following significance P=small friablo masses in a cloudy solution F-soft mass in a clear or slightly cloudy solution G-tight ropy clump in a clear solution (See Figue 13 ) focludes Streptococcal kaphylococcal pneumococcal F coll gonococcal and memungococcal arthritas These fuds were obtained post mortem Results based on the analysis of only 1 fluid precipitate 3 days later when the total cell count had fallen from 151,500 to 90 000 cells per cu mm Gradual improvement was found also in a case of rheumatoid arthritis The fluid aspirated 4 weeks after the onset of the effusion contained 006 gm nitro gen as mucin per 100 cc One week later the fluid contained 0 09 gm, and 5 months after onset the content was 0 12 gm The corresponding improvement in the activity of the disease was reflected in the drop in corrected sedimentation rate from 107 mm per minute to 0 82 and finally to 0 69 Administration of adrenocorticotrophic hormone or of cortisone to rheumatoid patients resulted in a dramatic improvement in the state of synovial mucin as determined by viscosity of the fluid and type of mucın precipitate In one patient the viscosity rose from 13 to 6,080 following 110 days of ACTH treatment Whether this change represents inactivation of an enzyme that causes degradation of mucin or activation of an enzyme capable

·No text on this page.

·VARIATIONS WITII DISCASE TABLE \ Comparison of Diffcrent Methods of Estimating the Concentra tion and State of Mucin in Various Types of Joint Disease Neletve viscosity at 38 C Mucin Mucin Type of Mucin nitrogen glucosamine mucin nitrogen gm /100 ccm /100 cc pptt Clucosamine 1 38 Tuberculous arthritis 77 044 028 P 0 80 Renters syndrome 105 007 052 P Rheumatoid arthritis 110 067 010 P 1 38 Syphilitic arthritis 114 058 039 P I 12 Infectious arthritist 137 058 047 1 58 Pulmonary osteoarthropathy 15 3 047 037 F 1 10 Gouty arthritis 170 058 040 P 1 03 Hemophilia 210 041 P Rheumatic fever 220 108 093 G 1 52 Probable rheumatoid arthritis 24 0 068 048 F 144 Neuroarthropathy 310 068 048 F 1 38 Traumatic arthritis 31 0 080 098 056 Degenerative joint disease 410 090 080 G 094 Lupus erythematosus disseminatus 1490 092 059 G I 37 Normal human knee joints 2350 104 087 G 1 16 Myredema 381 01 1921 G Osteochondromatosis 499 0 168 09911 C Arranged in order of increasing viscos ty not all the columns contain equal numbers of Aulds The type of precipitałe us expressed 20 symbols having the following significance Por-small fnablo masses in a cloudy solution F=soft mass in a clear or slightly cloudy solution Gought ropy clump in a clear solution (See Figure 13 ) I Includes streptococcal staphylococcal pneumococcal E coll gonococcal and meningococcal arthritis *These Buids were obtained post mortem Results based on the analysis of only 1 flud precipitate 3 days later when the total cell count had fallen from 151,500 to 90,000 cells per cu mm Gradual improvement was found also in a case of rheumatoid arthritis The fluid aspirated 4 weeks after the onset of the effusion contained 006 gm nitrogen as mucin per 100 cc One week later the fluid contained 0 09 gm, and 5 months after onset the content was 0 12 gm The corresponding improvement in the activity of the disease was reflected in the drop in corrected sedimentation rate from 1 07 mm per minute to 0 82 and finally to 0 69 Administration of adrenocorticotrophic hormone or of cortisone to rheumatoid patients resulted in a dramatic improvement in the state of synovial mucın as determined by viscosity of the flud and type of mucın precipitate In one patient the viscosity rose from 13 to 6,080 following 110 days of ACTH treatment Whether this change represents inactivation of an enzyme that causes degradation of mucın or activation of an enzyme capable ·64 SYNOVIAL FLUID CHANGES of synthesizing highly polymerized mucın or is unrelated to en zyme activity is not known To summarize, definite breakdown of the mucın in a fluid (evidenced by a friable precipitate and cloudy solution on addı tion of acetic acid) indicates that the effusion is probably not traumatic (Group I) in origin The degree of destruction of mucin can be estimated best by the nature of the precipitate obtained with acetic acid and is also indicated by the relative viscosity of the fluid Unusually high concentrations of undegraded mucın are found in lupus erythematosus disseminatus, degenerative joint disease and osteochondromatosis MON ELECTROLYTES Non protein Nitrogen and Uric Acid The synovial fluid concentration of non electrolytes that are readily diffusible and not rapidly utilized in the joint (such as non protein nitrogen and uric acid) is dependent on the plasma concentration and is not of diagnostic value In effusions of all types the concentration is approximately equal in fluid and plasma ys is found normally and as would be expected from the fact that the synovial tissues are readily permeable to these substances Similar distribution has been reported by other workers 7 30 31 89 104 Equal distribution of uric acid between blood and fluid was found in all effusions even in gout The duration of the gouty effusions varied from 10 hours to 10 weeks (see table, p 65) These findings do not support Klings thesis118 that the concentration of uric acid in the fluid is significantly higher than that of the blood during an acute attack of gout Bilirubin Bilirubin also diffuses readily from plasma to fluid in jaundiced patients, as was noted by Kling 115 In a patient with gout and jaundice Burmanºs found an icteric index of 17 compared with ? value of 21 in the blood 2 days later In 1 fluid in our senes, obtained post mortem from a jaundiced patient who had had no evidence of joint disease the icteric inder determined in this laboratory was 20 as compared with 75 in the serum ·VARIATIONS WITII DISEASE 65 Uric Acid Concentration in Gouty Effusions Duration of effusion Se mg/100 cc 17 mg/100 cc 10 hours 36-48 hours 1 week 1 week 10 days 10 days 3 weeks 4 weeks 5 weeks 5 weeks 10 weeks 12 5 54 03 67 42 42 75 39 96 96 84 128 80 61 66 46 46 72 38 99 95 89 Sulfonamides Similarly, the synovial tissues are readily permeable to sulfanilamide After intravenous injection, prontosıl appeared in the fluid in 9 minutes The synovial fluids from 3 cases of gonococcal arthritis contained 60, 54 and 78 mg of sulfanılamide per 100 cc after 2 days of treatment, in contrast to 46,51 and 66 mg per 100 cc in the serum Keefer and Rantz205 found approximately the same concentration of sulfanilamide in blood and fluid in 5 of 6 cases of gonococcal arthritis In 1 other case the concentration was much higher in the fluid (188 mg per 100 cc) than in the blood (62 mg) of two patients in the present series treated with sulfathiazole, 1 showed essentially equal concentrations of the sulfonamide in serum and fluid (3 2 and 35 mg per 100 cc ), while the other had only a trace of sulfathiazole in the fluid when the serum concentration was 41 mg per 100 cc Penicillin Permeability of the synovial tissue to penıcıllın has been demonstrated by several investigators 8 76 94 124 187 Many have found the concentrations in the fluids to be comparable to those in the blood after intramuscular administration In other studies 22 85 88 143 lower levels were found in the joint fluid after intramuscular injection Penicillin was found to remain in the fluid longer than in the blood : 88 98 Chloromycetin Diffusion of chloromycetin from blood to synovial fluid was ap- ·66 SYNOVIAL FLOW CHANGES TABLE XI Concentratons of Glucose in Synayal Fluds Aspirated 3 to 4 Hours After Eating Case 20 ໂຫານ; after cating Sugar & per 300 cc Blood Fiaid Daference 1 4 21 09. 5 2 102 103 101 139 99 99 83 81 98 117 67 75 35 3 5 23 parent in 1 patient with staphylococcal arthnts in our series The concentration in the fluid was greater than 16 but less than 64 mg per cc compared with a serum level greater than 32 but less than 128 mg per cc Sugar Variations in the concentration of sugar, in contrast to those with other non electrolytes, are of diagnostic significance (Figure 14) The concentration of glucose in fluids of Group I in general is essentially normal, averaging less than 10 mg below that of the serum In many instances the fud values are equal to those of the sera indicating that the rate of utilszation by the cells of synovial tissues, cartilage and fluid is approximately equal to the difference between the rate at which glucose is supplied and the rate of removal In this senes hemorrhagic effusions had lower concentrations of sugar than other traumatic Burds, the average level being 22 mg below that of the serum The increased glucolytic enzyme activity associated with the large amount of blood in the joint presumably causes this decrease in glucose *It should be mentioned in connection with sugar levels that determinations of the concentrabon un fuds talen while patients are under anesthesia at within 3 to 4 hours after eating are seldom of diagnostic value The rise in blood sugar caused by anesthesia or ingestion of food is reflected in the fuld sugar and the relationship between the blood and Aud sugars vanes markedly during the nse and fall of the blood sugar It has been shown that increase in the level of glucose in the blood is not usually reflected in the fiwd of pormal*** Or Theumatoids jounts for approximately 20 minutes The fluid sugar then nses slowly and steadily Later when the blood sugar falls rapidly the rate of utilization of sugar by the articular tissues and the rate of diffusion from the joint fluid back to the capu Lanies are not sufficiently rapid to beep the flud sugar at or below that of the blood ** As a result fuds withdrawn 3 to 4 hours after catng tend to have lugher concentrations of glucose than blood taken at the same time (Table XI) ·67 VARIATIONS WITH DISEASE Fluids in Group II have a lower average glucose content than normal and Group I fluids The lowest values in this senes (ercept some in fluids from severe rheumatoid arthritis of very long duration) occurred in the fluids in which organisms were found The reduction of the sugar concentration that occurred when there was a sufficiently large number of organisms to give a positive culture and a marked cellular response was particularly well shown in fluds from gonococcal arthntis The average sugar content of fluids from which no organisms could be grown was 83 mg per 100 cc, while the average for fluids with positive cultures was 29 mg (Table IV) In rheumatoid effusions, sugar, like other components, shows a wide range in concentration The minimal and maximal values in this series were 0 and 132 mg per 100 cc There was a general tendency for the fluid level to be lower in severe and moderate than in mild cases (see Table VII) A correlation between duration of effusion and concentration was also shown the average being 78 mg per 100 cc in those of 6 weeks or less in contrast to 58 mg per 100 cc in those of over 1 years duration (Table V) Furthermore all theumatoid effusions that contained no sugar were of over 6 months duration The decreased concentration of sugar is due in part to the decreased rate of entrance which has been shown to occur in the effusions of rheumatoid and infectious arthritis 213 This disturbs the normal equilibrium between entrance and utilization The reduced supply of glucose, associated presumably with increased utilization by inflamed and proliferated synovial tissues and by an increased number of leukocytes and by organisms of present, results in a marked lowering of the sugar level The cause of the difficulty in entrance of sugar into chronic effusions in rheumatoid arthritis and the similar delay in infectious arthritis is not known The reduction in the rate of entrance vanes from patient to patient, depending apparently chiefly on the seventy of the disease It is not due to a reduction of capillary permeability" presumably, since the rate of entrance of such substances as proteins and lipoids is increased in the same joints Whatever the mechanism, it is probably merely a retardation of the usual mode of en ace of glucose for in normal calves the passage of glucose from blood to fluid is slower than that of thiocyanate 209 213 ·Ticunt 14 The difference between the concentrations of glucose in scrum and synovırl duid in various joint discases In the few cases in which scrum glucose levels were not determined, in estimated concentration of 90 mg per 100 cc was used to calculato the scrum fluid difference Tho asterisks represent the minimum and maximum values in normal fluids Fluids No of Patients No of 011 **10 40 Serum Fluid Sugar Difference Mg per 100 CC T TO 10 10 Traumatic Arthritis Trorist Traumatic Arthritis with 13 Hemorrhage Osteochondritis Dissecans 2 GROUP I Osteochondromatosis 16 15 Neuroarthropathy U Degenerativo Joint Diseaso w Hypertrophic Pulmonary Osteoarthropathy 3 Hemophilic Arthritis 4 Lupus Erythematosus Disseminatus 22 18 Gouty Arthritis 7 Rheumatic Faver 7 7 Hemorrhagic Villou Synovitis ·on og omegn Beiro Rheumatoid Arthritis 87 Probable Rheumotoid Arthritis Reiter s Syndrome 21 91 Syphılıtıc Arthritis- Gongenital GROUP I 10 B Tuberculous Arthritis 24 122 e e 3 Gonorrheal Arthritis 30 29 ee Septic Arthritis 28 27 1220. Meningococcal Arthritis 15 fluids from traumatic arthritis of more than 3 months duration Auds from degenerative joint disease secondary to other joint abnormality fluids from rheumatoid arthritis of mild severity o fuds from rheumatoid arthritis of moderate severity fluids with positive cultures fluids of the types designated by the headings with the exception that the symbol indicates in traumatic arthri tis, fluids of less than 3 months duration in degenera tive joint disease, primary degenerative joint disease in Theumatoid arthritis sesere disease ind in the infectious arthritides, fluids with negative cultures ·70 SYNOVIAL FLUID CHANCES No eract correlation between the glucose level and total nu cleated cell count was found in any type of fluid This corroborates the conclusions of Collins, 39 who did not find the absolute correlation suggested by Cajon and Pemberton 31 In many cases in our series, marhed elevation of the white count caused 110 definite lowering of the sugar Tor example, rheumatoid fluids with total counts of 33,100, 39,300 and 51,000 had sugar con centrations of 64, 68 and 61 respectively, while the entire group of rheumatoid fluids which had an average count of 14 052, had an average sugar concentration of 63 There were only two con sistent relationships no fluid with a total count above 20,000 had a sugar level equal to that of the serum, and there was a tendency for the fluid sugar to be markedly decreased (usually below 50) when the total count was above 40 000 Even in these cases the correlation was not exact and it was apparent that factors other than the total leukocyte count were of more importance in determining the glucose level in the fluid An apparent reciprocal relationship between fluid sugar and total leukocyte count would be expected since both changes are presumably evidences of increased inflammation and since it has been shown that leukocytes cause breakdown of glucose 82 131 213 This apparent correlation and the relation to the severity of the inflammation are shown in the case of the patient with rheumatoid arthritis who had bilateral effusions that appeared and disappeared in regular cycles (Figure 7) During the penod of increasing inflammation and effusion the total leukocyte and absolute polymorphonuclear counts rose, while the sugar concen tration fell Subsequently, with decreasing inflammation, the counts fell and the sugar content increased The decrease in the difference between serum and furd glucose levels in some rheu matoid patients receiving adrenocorticotrophic hormone or cortisone gives further indication of the relationship of the sugar content of the fud to the degree of inflammation of the synovial tissues The glucose level in the synovial fluid is affected by increased articular pressure but these sugar changes are not of diagnostic or prognostic significance The significant level for diagnos that found on a second aspiration when intra articular pressure is lower The effect of pressure was shown in 4 cases of rheumatoid 15 ·VARIATIONS WITII DISCASE 71 TABLE XII The Relationship of Intra articular Pressure and Concentration of Glucose in Synovial Fluid Case no Initial pressure tod date mm of Leukocytes of synovial fuld per cu mm Polymorpho puclears per cent Sugar mg per 100 cc Synovial Blood Luid rynovial fuld 220 120 9 600 9 900 57 42 91 91 7 81 over 700 546 12 100 25 000 80 91 27 59 Case 1 12/20/38 1/14/39 Case 2 12/30/37 1/11/38 Case 3 3/24/38 4/ 1/38 Case 4 11/20/37 12/ 1/37 Case 5 10/21/39 10/26/39 19 600 90 700 170 20 72 107 over 700 0 11 800 30 700 87 78 89 73 75 99 over 700 100 15 600 7 480 76 49 55 104 arthritis that had low glucose concentrations and high intraarticular pressures in the original fluids Aspirations were repeated within a short period of time, and a rise in glucose concentration coincident with a fall in pressure occurred in each case (see Table XII) The rapid rise in sugar, with no known change ercept reduction of intra-articular pressure suggests that the low sugar concentrations in the initial fluids might be due to decreased diffusing surface caused by compression of blood vessels by the high intra artıcular pressure The pressure of 700 mm of water found in 3 of the joint fluids is much higher than the pressure usually thought to occur in capillaries (approximately 400 mm of water) However, in 1 case (Case 1, Table XII) in which exceptionally marked proliferation of the synovial tissues was subsequently found at operation a fluid pressure of only 220 mm was apparently sufficient to cause reduction of sugar concentration The mechanism by which high intra articular pressure lowers the sugar content of fluid remains unexplained Twenty one fluids obtained from patients with various joint diseases who had been fasting for 12 hours (or more) showed sugar concentrations more than 6 mg higher than the bloods taken at the same time It seems unlikely that such long periods would be necessary for the restoration of equilibrium All of the ·SYNOVIAL FLUI CELANCES 72 reducing substances in the flud were found to be fermentable by yeast The relatively high concentration of sugar in the fluid compared with that in the blood us difficult to explain The above survey of the concentrations found in fluds from vanous joint diseases indicates that a low or absent sugar is not necessarily an indication of bacterial invasion of a fluid as Collins presumed it to be In rheumatoid arthritis, persistently low sugar levels are an indication of severe or long continued disease Since the sugar content of synovial fluid is determined by the etiology and severity of the synovial inflammation, it is often of diagnostic and prognostic value If the sugar of a non hemorrhagic fluid is more than 20 mg per 100 cc below the blood level, the effusion is presumably not traumatic in ongin In infectious arthritis the fluid sugar concentration can be used as one indication of the course of the synovitis and thereby aid in prognosis Lactate The concentrtion of lactate in normal human synovial fuid has not been determined In 6 normal cattle fluids the level ranged between 18 and 34 mg per 100 cc (The corresponding serum levels varied from 35 to 114 mg per 100 cc, the great vanation resulting probably from the fact that the animals struggled before death ) The fluid lactate concentration in joint diseases varies with the etiology and severity in the 3 Group 1 fluids (obtained from Charcot joints) in which the lactate content was determined, the level in the fud averaged essentially the same as that in the serum (13 mg per 100 cc as compared with 15 mg per 100 cc) The concentration varied greatly in Group II fluds In 13 cases (1 rheumatic fever, 2 congenital syphilis 1 Reiters syndrome, I meningococcal arthritis 1 gonococcal arthritis and 7 rheumatoid arthritis), the fluid lactate level was within 10 mg per 100 cc of the serum level The differences were much greater in the other 28 Group II fluids the fluid level usually being higher than the serum The greatest differences found were in a patient with tuberculous arthritis and in one with severe gouty arthritis in whom the fluid values were 79 and 97 mg per 100 cc and the serum values 16 and 22 mg per 100 cc respectively Fluid aspirated at the same time from an effusion in the other knee of ·VARIATIOYS WITII DISCAST 73 TADLE VIII Distribution Ratios for Chloride Calcium and Phosphorus in Various Discuses P Caso no SH Theoretical Donnın nato Discue ar y Ca 1 QS 87 91 1 Normal . Traumatic arthritis 3 Traumatic arthritis 4 Traumatic arthritis 5 Degcncrativo joint discasc 6 Rheumatoid arthritis 7 Tuberculous arthritis 8 Septic arthritis 9 Septıc arthritis 100 1 05 101 102 102 99 99 92 03 90 97 94 105 1 10 99 1 09 100 93 90 07 97 96 97 97 09 99 the prtient with gout, contained only 51 mg of lactate per 100 cc In general, the lactate concentration in the flud showcd in inverse relationship to the sugar content (Tigure 15) The sige nificance of the relationshup is not apprrent, but possibly the increased lactate concentration is associated with grenter utilization of glucose ELECTROLYTES The concentration of chłonde is higher in normal human synovial fluid than in the serum, while the reverse is true of calcium The concentration of total inorganic phosphate is practically the same in fluid and serum 111 The distribution rtios are such as would be expected from the Donnan theory of equilibrium across a semipermeable membrane and from comprison with the distribution ratios for dialysates of blood plasma167 (Table II) In the types of pathological fluid analyzed, the distribution of chloride, phosphate and calcium ions between serum and fluid varied slightly from case to cise but there was no apparent correlation with the ctiology, seventy or durition The distribution ritios muntuned essentially the same relationship to the theoretical Donnan ratios as found in normal fluid The chloride and phosphate concentrations per 100 cc of water were approximately equil in fluid and serum, although there was a slight tendency for the concentration of phosphate to be higher in the serum (see Table XIII) The concentration of calcium wis lower in the flud than in the serum, but the distribution ratios Cae Ci, were ·SYNOVIAL FLUID CILANCES 72 reducing substances in the luid were found to be fermentable by yeast The relatı ely high concentration of sugar in the fud compared with that in the blood is difficult to explain The above survey of the concentrations found in fluids from various joint diseases indicates that a low or absent sugar is not necessarily an indication of bacterial invasion of a fluid as Collins presumed it to be In rheumatoid arthritis, persistently low sugar levels are an indication of severe or long-continued disease Since the sugar content of synovial fluid is determined by the etiology and seventy of the synovial inflammation, it is often of diagnostic and prognostic value If the sugar of a non hemorrhagic fluid is more than 20 mg per 100 cc below the blood level, the effusion is presumably not traumatıc in ongin In infectious arthritis the fluid sugar concentration can be used as one indication of the course of the synovitis and thereby ad in prognosis Lactate The concentration of lactate in normal human synovial fluid has not been determined In 6 normal cattle fluids the level ranged between 18 and 34 mg per 100 cc (The corresponding serum levels varied from 35 to 114 mg per 100 cc, the great variation resulting probably from the fact that the animals struggled before death ) The fluid lactate concentration in joint diseases varies with the etiology and severity In the 3 Group I fuids (obtained from Charcot joints) in which the lactate content was determined, the level in the fluid averaged essentially the same as that in the serum (13 mg per 100 cc as compared with 15 mg per 100 cc) The concentration varied greatly in Group II fluids In 13 cases (1 rheumatıc fever, 2 congenital syphius 1 Reiters syndrome, I meningococcal artłırıtıs 1 gonococcal arthritis and 7 rheuma toid arthritis), the fluid lactate level was within 10 mg per 100 cc of the serum level The differences were much greater in the other 23 Group II fluids the fluid level usually being higher than the serum The greatest differences found were in a patient with tuberculous arthritis and in one with severe gouty arthritis in whom the fluid values were 79 and 97 mg per 100 cc and the serum values 16 and 22 mg per 100 cc respectively Flud aspırated at the same time, from an effusion in the other knee of

·101
90
10
70
60
LACTATE-MG. PER OCC
so
40
30
20
0
10
50
60
10
70
40
60
10
150
90
0
20
GLUCOSE-MC PER oota
FICUNE 15 The relationship between the inctate and glucose concentrations in
synovial fluids (•) and sera (o) in various joint diseises

·74 SYYOVIAL FLUID CHANGES somewhat higher than those found in normal human fluids, with only shght increases in the theoretical ratios (Table XIII) The relatnely higher concentration of calcium and lower concentration of chloride and phosphate in pathological fluids are due presumably to the higher concentration of protein in these fluids Other electrolytes such as sodium and potassium have been deter mined only occasionally in pathological funds in the few fluids that have been examined, the results have been such as would be expected from the findings in normal cattle fluid (See Table Il and review of the subject 107) The total base concentrations in 2 fluids from rheumatoid arthritis were 145 8 and 1525 meg /1 in contrast to 147 8 and 1567 meq/1 in the blood In the first of these cases, the sodium concentration in the fud was 1335 meg/l and that in the blood was 1370 meq/1 Thiocyanate Diffusion of throcyanate into the joint space is rapid It was detected in the synovial fud of calves 10 minutes after intravenous injection 209 The concentration rose steadily until equilibrium was attained in from 1 to 5 hours At that point, the fluid level was 9 per cent lower than that of the blood The findings in 2 patients with a Charcot joint and in 3 of 7 patients with rheu matoid arthritis were similar In 3 of the rheumatoid patients, the rate of entrance was slower than that found in calves, while in 1 it was very rapid, equilibrium being attained in 15 minutes Hydrogen Ion Concentration The average pH of normal human fud obtained post mortem is 739 as determined in this laboratory Widely varying values (ranging from 74 to 84) for normal fluid obtained post mortem or after amputation have been reported 149 188 -18 In normal cattle fluid the average value is 731 in comparison with an average of 7 42 for the In joint disease, the pH varies with etiology and severity De terminations of pH were made on only 23 fluids in the present series In effusions from single cases of intermittent hydrarthrosis, gonococcal arthritis and mud rheumatoid arthritis, the pH of the flud was approximately the same as that of the blood, being even slightly lugher in l of these cases The pH of the Huid was defi- 18* serum ·VARIATIONS WITTI DISEASE 77 Alkaline phosphataje average Bodansky units Relative viscosity average at 38 C Type of fuid 235 (22) 07( 5) Human knee Cattle carpometacarpal astragaloubial 50 (4) 3 (7) 8.2 ( 3) 138 (16) The figures in parentheses indicate the number of faids from which the averages were obtained been considerable tissue destruction, as in severe cases of traumatic, infectious or rheumatoid arthritis Holmgren,89 by means of X ray examination, demonstrated free fat in 26 of 41 cases of fracture of the knee EXZYNIES The enzymes of normal synovial Gud have not been studied in detail Podkaminsky161 found an amylase, protease and lipase but no catalase We have found, however, that synovial fluid causes breakdown of hydrogen peroxide The alhaline phosphatase activity in normal cattle fud is higher than that in the serum, an average of 13 8 units per 100 cc in contrast to 46 Variations in activity in the fluid are marked, with a range from 54 to 33 8 units per 100 cc In 5 fluids obtained post mortem from patients who had had no joint disease, the phosphatase activity was much lower than that in normal cattle fluid, ranging from 0 36 to 1 08 units per 100 cc In Our experience, normal human fluid has the lowest phosphatase activity and the highest viscosity of any type of fluid This inverse relationship, apparent in the table above, may be of no significance but is interesting because serum with high alhalme phosphatase activity has been shown to cause breakdown of synovial mucin in vitro 105 The phosphatase activity was relatively constant in the 9 Auds from various joint diseases studied in this series and was, in general, only slightly less than that of the serum, averaging 27 units per 100 cc in contrast to an average of 35 in the serum It is apparent that knowledge of the vanous types of enzymatic activity in synovial fluid is too limited to indicate the role that enzymes play in the metabolism of articular structures ·SYNOVIAL FLUID CHANGES 76 nitely lowered in more severe rheumatoid joints (8 cases), averaging 7 22 in comparison with 7 45 in the serum The greatest difference between fluid and blood in the rheumatoid cases was 036 In the only 2 fluids from cases of infectious arthritis (staphy lococcal and pneumococcal) that were examined, the pH valves were 7 20 and 6 93 The serum pH in the latter case was 739 Following intravenous injection of weak electrolytes (sodium carbonate, lactic acid ammonium chloride and sodium salicylate), Joseph, Reed and Homburgerii found that the pH of the joint (measured in vivo) changed more slowly than that of the blood The latter usually reached the maximal displacement in 1 minute (and approached the resting level within a few minutes), whereas the joint pH usually required more than 5 minutes for the maximal effect Femoral nerve stimulation or exercise of the joint produced a decrease of the pH of both blood and joint fluid The blood pH fell rapidly and returned within a few minutes to a normal level The fall in the joint pH came more slowly and the displacement was often maintained for 20 to 30 minutes Similar results were obtained when convulsions were produced with metrazol or insulin 1. Following unilateral pervascular sympathectomy, the fall in joint pH following femoral nerve stimulation was greater on the operated side and in many cases showed no recovery u 30 minutes 180 LIPIS Cholesterol and fatty acids are not found u normal cattle synovial flud 169 Cholesterol is present in the flud in joint disease In 3 rheuma toid funds, the distribution ratio between fluid and serum was 067, 058 and 082, it was 0 61 and 0 45 in 2 cases of arthritis of undetermined nature In the majority of instances the presence of cholesterol presumably indicates increased permeability of the tissues between blood and flud and possibly a reduced rate of removal Chauffard Rachet and Gngauts' and White and Sachs 7 found low levels of cholesterol in edema flud due to congestive failure or venous obstruction in contrast to high con centrations in fluds orginating from lymphatic obstruction However, in some cases the content of cholesterol is probably increased because of breakdown of tissue Free fat and cholesterol crystals are found often in fluids from joints in which there has ·GROUP I, TRAUMATIC DISCASTS 79 several years later, it became apparent that these patients had rheumatoid arthritis (see Cise 6, p 126) In general, non hemorrhagic triumitic fluids show relatively slight changes from normal A typical fluid of less than 3 months duration is clear and does not clot Total nucleated cell count in our 31 fluids usually was below 1,000 per cu mm, with an absolute polymorphonuclear count below 100 The crythrocyte count varied from 50 to 15,650 The concentration of sugar was essentrally the same as that of the serum The protein content (exclusive of mucın) was 2 to 3 times that of normal fluid, most of the increise being in the albumin fraction, so that the albumin/ globulin ratio remained high, with an average of 49 Electrophoretic patterns in 7 cases showed the albumin fraction always higher in the fluid than in the serum, alpha-1 and beta globulins essentially the same and alpha 2 and gamma-globulins lower in the fluid 169 The viscosity was slightly reduced, but the mucın content per cubic centimeter was nomal (0076 gm nitrogen and 0 098 gm glucosamine per 100 cc ), and the mucın precipi tated as a ropy clump in a clear solution Thus the findings in traumatic effusions of short duration indicate only slight inflammation Effusions that persist for over 3 months following trauma may represent other types of arthritis, as discussed above They showed somewhat more marked changes, with slightly higher cell counts and protein concentrations The mucin concentration also was slightly increased, suggesting more rapid formation due to persistent irritation of synovial tissues While 40 traumatic effusions contained less than 20,000 erythrocytes per cu mm and were not xanthrochromic, 26 fluids were definitely hemorrhagic In such cases, even when the fluid was aspirated within 16 hours, clots were not found and the bloody fluid usually did not clot after withdrawal There is no evidence to indicate whether clotting occurs in the joint and the clots are dissolved or whether some constituent of the fluid acts as an anticoagulant Volkmann 0 stated that absorption of blood takes place very slowly unless there is a tear in the synovial membrane and that blood often remains in fluid state in the joint for weeks or months Julliard 103 Kling115 and Collins384 also noted that hemorrhagic traumatic effusions did not clot either in the joint ·Findings in Fluids in Group I, Diseases of Traumatic Origin The characteristics of synovial fluid in the different forms of joint disease included in Group I and the number of fluids from which the data were obtained are shown in Figures 5, 6, 9, 10, 11, 12 and 14 and the chart at the end of the book TRAUMATIC ARTHRITIS Effusions frequently follow joint injuries of various types as would be expected The effusions appear within a few hours after injury, usually subside within 1 or 2 weeks and seldom last more than 3 months As a rule, the involved joints are not red, hot or acutely tender Persistence of symptoms and effusion for more than 3 months after injury suggests that the joint involvement though precipi tated by trauma, is not the result of trauma alone However, it is frequently impossible to make any other diagnosis or to determine what additional factors are playing a role Therefore, for this series we have included these cases (9 Auids) in a group designated traumatic arthritis of over 3 months duration It is possible that they all represent mild cases of other types of arthritis For instance, infectious arthritis and gout tend to localize in traumatized joints and trauma may precipitate the first marked joint involvement in rheumatoid arthritis In such cases the joint disease often is incorrectly diagnosed This is exemplified by 2 of the patients who were originally included in the group labeled "traumatic arthritis of long duration and whose biopsies were interpreted as showing chronic synovitis On re examination 78 ·GROUP I, TRAUMATIC DISCASES 81 represent examples of various other types of arthritis The microscopic changes were variable, as were the clinical histories of these cases, but in general they were consistent with a very mild degree of chronic inflammation The usual findings were slight villous hypertrophy, fixed cell proliferation and lympho cytic infiltration Few studies on traumatic fluids have been reported by other workers Non hemorrhagic and hemorrhagic fluids have not in general been considered separately Rostock i found variations in intra articular pressure from 125 to 700 mm of flud în 9 acute traumatic effusions associated with intra articular fractures In 35 hemorrhagic traumatic effusions not associated with intra articular fractures the pressures vaned from 0 to 600 Thc range in chronic traumatic effusions was from 0 to 340 Rankel reported pressures of 15 to 20 cm of mercury with an average of 9 cm in hemorrhagic cffusions The cytological findings of the present series are similar to those found by Julliardo (8 cases) and by Collins (3 cases) but the total leukocyte and polymorphonuclear counts are in general lower than those of Shands24 (4 cases), Kling (88 cases), Leefer (2 cases) McEwen" (2 cases) and Hirsch (5 cases) The height of the total leukocyte and polymorpho nuclear counts which reached 12 000 per cu mm and 94 per cent in some of the fluids of the latter four series suggests that the joint effusions were not due to trauma alone Reports vary concerning the relation between the pH of the synovial fluid and the duration of the effusion Laschi in 10 cases found values varying from 7 39 in effusions of 2 days duration to 791 at the end of 1 year after injury On the other hand Beck and Lauberlo noted pH values of 7 23 and 7 17 in 2 fluids taken 2 days and 3 weeks after injury values which changed to 6 80 and 6 86 respectively 2 days after the original aspirations Waugh found an average pH of 5 65 in 20 traumatic effusions 24 hours after injury changing to 595 on the following day and reaching 77 on the seventh day Coppo and Levi" observed values of 7 67 and 7 94 in traumatic effusions but did not state the duration of the effusions The chemical findings of Collinse in 5 traumatic effusions agree in most respects with those of the present series The amount of flud varied from 12 to 40 cc the total protein including mucın from 3 25 to 3 35 per cent the sugar from 68 to 97 mg per 100 cc and the viscosity from 2 8 to 11 2 Kulonen.286 in 9 fluids found viscosities ranging from 15 to 16 6 and in 2 fluids noted freezing points of 550 and 607 C In 1 flud Ragan found a viscosity of 138 In 27 fluids Sundblad reported variations in the amount of flud from 15 to 85 cc an average intrinsic viscosity of 41 0 an average total protein content of 4 42 gm per 100 cc and an average con centration of hyaluronic acid of 0 101 gmn per 100 cc klingtis in 88 cases found an average icteric index of 14 with a range from 8 to 133 Pewny 89 stated that the protein concentration in traumatic fluids is 25 per cent ·80 SYNOVIAL FLUID CHANGES or after aspiration Davies, 43 however, stated that blood shed into normal or pathological joints clots very readily Hemorrhagic traumatic effusions of under 3 months duration offered interesting comparison with non hemorrhagic traumatic fluids of similar duration The total nucleated cell and absolute polymorphonuclear counts were slightly higher in the 26 hemorrhagic fluids with averages of 1,540 and 315 per cu mm There were a marked increrse in the erythrocyte count and, in many instances, a high icteric index The sugar concentration in the majonty of the hemorrhagic effusions was decreased in contrast to other traumatic fluids, presumably because of the increased glucolytic enzyme content previously mentioned The average concentration of total protein was the same as that in the non hemorrhagic fluids The relative increase in globulin was greater than that of albumin as would be expected, so that the albumın/ globulin ratio averaged 28 in contrast to 49 The viscosity varied markedly but covered essentially the same range as that of the non hemorrhagic fluids Although the mucin concentration varied, ut averaged much less than that of the non hemorrhagic fluids (0050 gm nitrogen and 0019 gm glucosamıne per 100 cc), and the mucın precipitate with acetic acid was less cohesive The relatively high concentration of protein and low concen tration of mucın in hemorrhagic fluids are probably due to the fact that the effusion is formed partly by direct flow of blood into the joint Under such conditions less mucın is carried into the joint from the surrounding tissues Although the lower content of mucn might be due to destruction of mucın by an enzyme carned in with the blood this explanation is unlikely, since no such enzyme has been demonstrated in blood 170 Furthermore, the relationship of mucin concentration to viscosity in these fuds indicates that there is very little degradation of the mucin It is apparent that hemorrhagic fluids resemble other traumatic fluds except for the changes directly related to the increased content of blood Histological examinations of synovial tissue from traumatic arthritis were limited to lesions of more than 3 months duration and included 5 examples with non hemorrhagic and 2 with hemor Thagic effusions As stated above it is probable that effusions that persist for more than 3 months are not due to trauma alone but ·GROUP 1, TRAUMATIC DISEASES 83 of osteochondromitosis in thus scrics were only slight The fluid was clear, did not clot and was extremely viscous Total nuclertcd cell count was only slightly above normal, averaging 507 per cu mm, uverige absolute polymorphonuclear count was only 5 The fluid sugar in the 1 casc in which it was determined was 93 mg per 100 cc compared with 101 mg in the serum The average total albumin and globulin concentration wis also only slightly above normal (22 gm per 100 cc) and was a much lower value thin that of traumatic fluids The extremely high mucin content, on the other hand, averaging 0 166 gm nitrogen and 0 99 gm glucosamine per 100 cc, was much greater than the average for normal human fluid and higher than that in any other type of joint disease (sce p 56 for discussion) The viscosity was correspondingly high, with an average of 580 at 38° C The mucin precipitated with acetic acid as a ropy clump with clear surrounding solution The findings in these cases indicate that there is only very slight inflammation in osteochondromatosis Synovial tissues removed at biopsy in 5 cases showed slight evidence of inflammation with scanty lymphocytic infiltration The most characteristic changes were villous hypertrophy with many foci of cartilaginous metaplasia containing areas of calcification and cancellous bone formation There were also occasional small superficial patches of fibrın like material and foci of mucoid change of the connective tissue NEUROARTHROPATHY (CHARCOT JOINTS) The marked degenerative changes, joint destruction and loss of pain sense that occur in Charcot joints afford an excellent basis for the development of traumatic effusions One or more joints may be involved and as a rule they are painless and not red, warm or tender Effusions usually are large and frequently persist for years Of the fluids in this series, 31 were from Charcot joints associated with tabes dorsalis and 1 was secondary to syringomyelia There was no essential difference between the latter fluid and those from syphilitic patients Fluids from Charcot joints were very similar to traumatic fluids They were either clear or hemorrhagic and did not clot The intra articular pressure varied from ·82 SYNOVIAL FLUI CHANGES Scapını" in 1 case and Coppo and Lenzin 2 cases found calcium of 109 mg per 100 cc phosphorus of 24 and 31 mg per 100 cc magnesium of 27 mg per 100 cc, potassium of 29 mg per 100 cc chloride of 719 mg per 100 cc protein of 58 grn per 100 cc and a viscosity of 19 Schurch Viollier and Sullmann'' reported that the albumin/globulin ratios in trau matic furds were the same as those of the corresponding sera but the con centration of alpha 2 globulin was lower in the fluid than in the serum OI hagen" found the relative distribution of proteins in traumatic fuds about the same as in the plasma with slightly higher concentrations of albumin and gimma globulin in the flud In hemorrhagic effusions resulting from moderate or severe injuries Efskind" found a delayed absorption of pera brodil and of indigo carmine from the joints OSTEOCHONDRITIS DISSECANS Effusions occur frequently in osteochondritis dissecans Involve ment is usually limited to one joint, most commonly the knee but may be bilateral, as in 1 case in this series The joints usually are not red, warm or tender The 10 fluids in this series were essentially the same as trau matic fluids They were clear or slightly turbid and usually formed only a small clot on standing The average total nucleated cell count (750 per cu mm) and percentage of polymorpho nuclears (9) were in the same ringe as those of traumatic fluids The sugar concentration in the 2 fluids examined was 88 and 101 mg per 100 cc The average total albumin and globulin concen tration (33 gm per 100 cc) and the globulin fraction (08 gm per 100 cc ) were essentially the same as those of traumatic fluids The mucın nitrogen concentration in the 2 fuds examined was higher and the mucin glucosamine content in 1 flud was lower than those of traumatic effusions The mucin precipitate was ropy with a clear solution The average relative viscosity was 23 The findings in fluids from osteochondritis dissecans indicate only slight inflammation Synovial tissues in the 1 case examined, also showed only slight change from normal Scanty infiltration with lymphocytes and histiocytes was the principal abnormality OSTEOCHONDROMATOSUS Effusions are not uncommon in osteochondromatosis Only one joint is involved and it is not red warm or tender The abnormalities found in the synovial fluids from the 8 cases ·GROUP I, TRAUMATIC DISEASES 85 Synovial tissues in 5 cascs showed active fixed cell proliferation with marked hypertrophy of vill and fibrosis but relatively mild focal infiltration of lymphocytes Focal deposits of hemosiderin, fibrın like materni, calcium and cırtılıgınous or bony metaplasia were common Low nucleated cell counts and/or low percentages of polymorphonuclear cells were found also by Dufour" (1 casc) Abadic (3 cascs) McEwen*** (1 case) and Collins.com (1 case), occasional higher counts were reported by Pewny!" (17,000 cells per cu mm in 1 case) Shands (up to 12 000 cells in 10 cases), Keefert (18 350 cells in 1 of 3 cases) and Aling (up to 260 000 cells in cases) The explanation of such high counts is not ap parent Few chemical studies have been made Kcefere found protein contents, including mucin ranging from 34 to 40 per cent in 3 cascs Achard and Prettre in 1 case have reported a mucin content of 41 8 gm , a fibrin content of 1 42 gm and a lipid level of 0 46 gm per 1000 cc (It seems probable that the acetone precipitate of mucin contained proteins in addition to those combined with the polysaccharide ) Scapin."" in 1 case and Coppo and Levi" in 3 cases found calcium concentrations ranging from 6 05 to 10 65 mg per 100 cc phosphorus from 25 to 29 mg per 100 cc magnesium from 15 to 27 mg per 100 cc, potassium from 13 8 to 289 mg per 100 cc chloride from 834 to 940 mg per 100 cc, pH from 765 to 8 14 protein from 4 1 to 4 4 gm per 100 cc and a viscosity of 1 65 Other workers have found that the Wassermann reaction in cases of Charcot joints is not consistently the same in serum and in fluid as shown in the table below In Poehlmann s case the blood and flud Wassermann reac tions were both positive but the flud reaction was weaker Wassermann Reaction in Serum and Flud No of fluids in which serum and fluid differed Total po of cases Result of Wassermann reaction Serum Flud Investigator 2 Chesney Shands in heefer104 Kling16 1 2 2 1 22 6 Negative Positive Positive Negative Positive Doubtful Positive Negative DEGENERATIVE JOINT DISEASE The relative infrequency of effusions in degenerative joint disease is in accord with the pathological findings Apparently the initial and fundamental change occurs in the articular cartilage and consists of degenerative changes manifested by surface irregularities, swelling and fibrillation of matrix and abnormal cluster- ·SYNOVIAL FLUID CELANCES 1 84 180 to 300 mm of fluid The total nucleated cell count in this series was normal (below 180 per cu mm) in 7 of the 32 fluids and below 500 in 14 other luids The percentage of polymorpho nuclears was normal in 16 out of 31 of the fluids Seventeen of the 32 Ruids were grossly bloody or xanthrochromic, indicating that hemorrhage is common in neuroarthropathy, as described by Dufour, 57 Abadie, Henderson, S4 Shands 181 and Kung 115 The sugar concentration was essentially the same as that of other fluids of Group I, averaging 81 mg per 100 cc in com parison with an average of 92 mg per 100 cc in the serum The average lactate content in 2 fluids was 13 mg per 100 cc compared with 15 mg in the serum The average total albumin and globulin concentration was the same as that of traumatic fluids, ranging from 18 to 51 gm per 100 cc, although the variations from case to case were much more marked Two of the highest protein concentrations were in 2 flwds from a patient who had areas of cellulitis near the joint affected, it can be assumed that the synovial tissue was involved in the inflammation The globulin concentration averaged slightly higher than in other Group I fluids, with an average albumin/ globulin ratio of 22 The mucin concentration also vaned similarly, but the average (0068 gm nitrogen and 0016 gm glucosamine per 100 cc) was below that of truumatic Buds The mucin precipitate with acetic acid was a ropy clump in clear solution in 15 cases, and less cohesive with slightly cloudy solution in the other 9 fluids examined Viscosity varied markedly, ranging from 27 to the exceptionally high value of 228, with an iverage of 31 at 38° C (see p 56 for discussion) The lowest viscosity (27) was that of a fluid obtained from the patient with areas of celulitis near the joint The Wassermann reaction in the fluid from Charcot joints was the same as that of the serum in 6 of the 8 cases in which both were tested (In l other case in which the reaction was positive in the fluid, the serum Wassermann was negahve, but the Hinton was positive ) In 6 of the fluids, the reaction was negative Thus the findings in Charcot joints are in general accord with the fact thit there is no marked niammatory process in the membrane ·GROUP I, TRAUMATIC DISCASTS 85 Synovial tissues in 5 cascs show cd active fixed cell proliferation with marked hypertrophy of villt and fibrosis but relatively mild focal infiltmtion of lymphocytes Focal deposits of hemosiderin, fibrin like material, calcium and cartilaginous or bony metaplasia Were common Low nucleated cell counts and/or low percentages of polymorphonuclcar cells were found also by Dufour (I case), Abadie' (3 cases) McEwen** (1 case) and Collins (1 casc), occasional higher counts were reported by Pewny (17 000 cells per cu mm in 1 case), Shands (up to 12 000 cells in 10 cases), hccfcr2 (18 350 cells in 1 of 3 cases) and Lling' (up to 260 000 cells in 6 cases) The explanation of such high counts is not ap parent Few chemical studies have been made Leefer found protein contents, including mucin ranging from 3 4 to 40 per cent in 3 cases Achard and Prettre in 1 case have reported a mucin content of 41 8 gm a fibrin content of 1 42 gm and a lipid level of 0 46 gm per 1000 cc (It seems probable that the acetone precipitate of mucin contained proteins in addition to those combined with the polysaccharide ) Scapin?" in 1 case and Coppo and Levi" in 3 cases found calcium concentrations ranging from 605 to 10 65 mg per 100 cc phosphorus from 25 to 29 mg per 100 cc magnesium from 15 to 27 mg per 100 cc potassium from 13 8 to 289 mg per 100 cc, chloride from 834 to 940 mg per 100 cc, pH from 765 to 8 14 protein from 4 1 to 4 4 gm per 100 cc and a viscosity of 1 65 Other workers have found that the Wassermann reaction in cases of Charcot joints is not consistently the same in serum and in fluid as shown in the table below In Poehlmann ** case the blood and flud Wassermann reac tions were both positive but the flud reaction was weaker Wassermann Reaction in Serum and Fluid No of fluids in which senim and Buid differed Total no of cases Result of Wassermann reaction Serum Fluid Investigator Chesney Shandsíst Keefer Klings 4 22 6 2 2 Negative Positive Positive Negative Positive Doubtful Positive Negative 1 DEGENERATIVE JOINT DISEASE The relative infrequency of effusions in degenerative joint disease is in accord with the pathological findings Apparently the initial and fundamental change occurs in the articular cartilage and consists of degenerative changes manifested by surface irregulanties, swelling and fibrillation of matnx, and abnormal cluster- ·86 SYNOVIAL FLUID CILANGES ing of cartılage cells Subsequently there is increasing destruction of cartilage and overgrowth of bone at the capsular attachments As these changes proceed, the mechanics of the joint become poorer and even normal use may cause trauma with some degree of synovitis When effusions occur, they are rarely large and are often inconstant As a rule, they are produced by excessive use of the joints, though occasionally they are caused by acute trauma Involved joints are not red or hot, and although there may be some tenderness, this usually is not marked Fluids from degenerative joint disease arc essentially the same as traumatic fluids They are clear, very viscous and do not clot The average nucleated leukocyte count (720 per cu mm) and the percentage of polymorphonuclears in the 18 fluids of this series were in the same range as in triumatic fluids The total concentration of albumin and globulm (26 gm per 100 cc) and the albumin/globulin ratio (48) were also approximately the same as those of traumatic fluids The average concentration of mucin (0095 gm nitrogen and 0058 gm glucosamine per 100 cc) was essentially the same as that of traumatic fiuds It precipitated as a ropy clump in a clear solution ercept in 2 fluids 10 which the solution was slightly cloudy The relative viscosity (ranging from 13 to 78 at 38° C) was somewhat higher than that of traumatic fluids This does not suggest that a loss of lubricating power is a factor in the production of the cartilage changes in this disease In fact, one of the highest viscosities of fluids eramined in this laboratory was that of 1,285 in fud obtained post mortem from a patient with marked degenerative changes but no eſfusion in the knees (the patient died at the age of 81, following perforation of the gall bladder) By teleological reasoning such an extremely high mucin content would suggest increased production of mucin in an attempt to provide better lubrication for damaged articular surfaces (See p 56 for a discussion of mechanisms which may produce high mucin concentrations ) Fluids from 11 cases of degenerative joint disease secondary to other joint abnormalities (such as congenital malformation of the hip) showed essentially the same changes as those from primary degenerative joint disease The similanty of fluids from degenerative joint disease and those from traumatic arthritis is compatible with the hypothesis that synovitis associated with degenerative joint disease represents ·GROUP 1, TRAUMATIC DISCASES 87 reaction to thc triumr resulting from the use of mechanically poor joints Histological crimination of the synovial tissues in 12 cases revealed slight diffuse and local infiltration of lymphocytes, and occasional plasma cells and histiocytes In 1 fcw cases the infiltration was more marked Villous hypertrophy and diffusc fibrosis were present in varying degrees These changes were similar to those found in so called traumatic" cffusions of over 3 months duration (p 78) The carliest report of the characteristics of ſuid in degenerative joint disease was that of Hopp-Scyler in 1872" In 2 cases he found yellowish alkaline clear cry viscous stringy fluid The mucın concentration was 23 per cent, the albumin 20 per cent, total solids 5 73 per cent Subsc quent investigators" ma 190 145 14 have reported varied cytological findings Total leukocyte counts ranged from 200 to 11 000 per cu mm, with 0 to 88 per cent polymorphonuclears In the majority of cases the absolute poly morphonuclear count was in the range of the present series on the basis of our data total counts abovc 6 000 or absolute polymorphonuclear counts above 500 would suggest that other joint disease was superimposed on the degenerative disease Alings report covers 46 cases of "osteoarthritis but includes cases with evidence of "acute inflammation some of which had high sedimentation rates Some of his figures therefore do not apply to uncomplicated degenerative joint disease Studies of physical and chemical characteristics are meager The intra articular pressure in 2 cases studied by Rostocki" was in the same range as that of various traumatic effusions (0 to 300 mm of fluid) There have been reports of values of pH from 73 to 7 8 115 16 viscosities from 6 to 102 • 145 144 sugar from 95 to 100 mg per 100 cc * calcium from 81 to 14 mg per 100 cc unic acid from 27 to 5 9 mg per 100 cc icteric index from 21 to 4 6 116 protein contents from 33 to 37 per 100 cc 1 and hyaluronic acid of O 146 gm per 100 cc 14 A relative viscosity of 6 suggests that other joint disease was superimposed on the degenerative joint disease 116 115 *A calcuum content above that of the serum is very unusual and has been found in our series only in 3 patients who had rheumatoid arthritis and vitamin D poisoning and in 2 patients who had diseases that could not be definitely diag nosed but fell in the group of generalized connective tissue diseases In all these cases the fluid was milky or pasty and on analysis contained high concentrations of calcium (from 12 4 to 359 mg per 100 cc) and phosphorus (from 92 to 184 mg per 100 cc) with calcium phosphorus ratios of approximately, 20 In the 3 cases of vitamin D poisoning there was in addition a white granular sedi ment in which the ratio of calcium to phosphorus was much higher (approximately 47) and entirely unlike that of bone in material from an inflamed subacromial bursa the calcium concentration was 139 mg per 100 cc but the equilibrium between bursa and serum differs from that between joint and serum (see discus sion of bursae p 120) ·86 SYNOVIAL FLUID CHANGES ing of cartilage cells Subsequently there is increasing destruction of cartilage and overgrowth of bone at the capsular attachments As these changes proceed, the mechanics of the joint become poorer and even normal use may cause trauma with some degree of synovitis When effusions occur, they are rarely large and are often inconstant As a rule, they are produced by excessive use of the joints, though occasionally they are caused by acute trauma Involved joints are not red or hot, and although there may be some tenderness, this usually is not marked Fluids from degenerative joint disease are essentially the same as traumatic fluids They are clear, very viscous and do not clot The average nucleated leukocyte count (720 per cu mm ) and the percentage of polymorphonuclears in the 18 fluids of this series were in the same range as in traumatic fluids The total concentration of albumin and globulin (26 gm per 100 cc ) and the albumin/globulin ratio (48) were also approximately the same as those of traumatic fluids The average concentration of mucın (0095 gm nitrogen and 0058 gm glucosamine per 100 cc) was essentially the same as that of traumatic fluids It precipitated as a ropy clump in a clear solution except in 2 fluids in which the solution was slightly cloudy The relative viscosity (ranging from 13 to 78 at 38° C) was somewhat higher than that of traumatic fluids This does not suggest that a loss of lubricating power is a factor in the production of the cartılage changes in this disease In fact, one of the highest viscosities of fluids eramined in this laboratory was that of 1,285 in fluid obtained post mortem from a patient with marked degenerative changes but no effusion in the knees (the patient died at the age of 81, following perforation of the gall bladder) By teleological reasoning, such an extremely high mucin content would suggest increased production of mucın in an attempt to provide better lubrication for damaged articular surfaces (See p 56 for a discussion of mechanisms which may produce high mucin concentrations ) Fluids from 11 cases of degenerative joint disease secondary to other joint abnormalities (such as congenital malformation of the hip) showed essentially the same changes as those from primary degenerative joint disease The similarity of fluids from deger tive joint disease and those from traumatic arthritis is compatible with the hypothesis that synovitis associated with degenerative joint disease represents ·GROUT 1, TTUAUMATIC DISEASES 87 reaction to the trauma resulting from the use of mechanıcılly poor joints Histological crimination of the synovial tissues in 12 cases revealed slight diffuse and focul infiltration of lymphocytes, and occasional plasmi cells and histiocytes In a few cascs the infiltration was more marked Villous hypertrophy and diffuse fibrosis were present in virying degrees These changes were similar to those found in so called "traumatic' cffusions of over 3 months' duration (p 78) ES The earliest report of the characteristics of fluid in degenerative joint disease was that of Hoppe Scyler in 1872" In 2 cases he found yellowish, alkalıne clear very viscous stringy fluid The mucin concentration was 23 per cent the albumın 20 per cent total solids 5 73 per cent Subsequent investigators*** 100 123, 10 have reported varied cytological findings Total leukocyte counts ranged from 200 to 11,000 per cu mm with 0 to 88 per cent polymorphonuclears In the majority of cases the absolute poly morphonuclear count was in the range of the present series on the basis of our data total counts above 6 000 or absolute polymorphonuclear counts above 500 would suggest that other joint disease was superimposed on the degenerative disease Alings report covers 46 cases of osteoarthritis” but includes cases with evidence of "acute inflammation some of which had high sedimentation rates Some of his figures therefore do not apply to uncomplicated degenerative joint disease Studies of physical and chemical characteristics are meager The intra articular pressure in 2 cases studied by Rostockin was in the same range as that of various traumatic effusions (0 to 300 mm of fluid) There have been reports of values of pH from 73 to 7 8 138 14 viscositips from 6 to 102 1144 sugar from 95 to 100 mg per 100 cc calcium from 81 to 14 mg per 100 cc uric acid from 27 to 59 mg per 100 cc 11 icteric index from 2 1 to 4 6 14 protein contents from 33 to 37 per 100 cc 101 154 and hyaluronic acid of 0 146 gm per 100 cc * A relative viscosity of 6 suggests that other joint disease was superimposed on the degenerative joint disease 0115 A calcium content above that of the serum is very unusual and has been found in our series only in 3 patients who had rheunatoid arthntıs and vitamin D poisoning and in 2 patients who had diseases that could not be definitely diag nosed but fell in the group of generalized connective tissue diseases in all these cases the flud was milky or pasty and on analysis contained high concentrations of calcium (from 124 to 359 mg per 100 cc) and phosphorus (from 92 to 18 4 mg per 100 cc) with calcium phosphorus ratios of approximately 20 In the 3 cases of vitamin D poisoning there was in addition a whute granular sed, ment in which the ratio of calcium to phosphorus was much higher (approximately 47) and entirely unlike that of bone in material from an inflamed subacromial bursa the calcium concentration was 139 mg per 100 cc but the equilibrium between bursa and serum differs from that between joint and serum (see discus sion of bursae p 120) ·88 SYNOVIAL FLUID CHANGES SUMMARY OF FINDINGS IN GROUP I FLUIDS It is apparent that the fluids in Group I show relatively slight changes from normal In general, the amount of fluid is increased but the unit concentration of nucleated leukocytes and the absolute number of polymorphonuclears are only slightly above normal The total concentration of albumin and globulin is in creased but remains below 5 gm per 100 cc, and the albumin/ globulin ratio remains above 140 The concentrations of sugar and of mucin, except in osteochondromatosis, tend to be normal or only slightly lowered The viscosity is somewhat reduced except in osteochondromatosis, but mucın precipitates normally ·IV. Findings in Fluids Resembling Both Groups in Some Respects The changes in fluids obtained in cases of hypertrophic pulmonary osteoarthropathy, hemophilia, disseminated lupus erythematosus, gout, rheumatic fever, arthritis due to hypersensitivity, erythema nodosum, hemorrhagic villous synovitis and tumors are more marked than those of Group I fluids, as will be shown In general, however, these fluids are not so abnormal as those of Group II (Figures 5, 6, 9, 10, 11, 12, 14, and chart at back of book ) HYPERTROPHIC PULMOVARY OSTEOARTHROPATHY The synovial tissues are occasionally involved in hypertrophuc pulmonary osteoarthropathy Effusions may occur and persist for the duration of the disease The joints are painful, but usually not red, hot or tender Most of the findings in fluids from hypertrophic pulmonary osteoarthropathy are similar to those in traumatic fluids The fluds are clear and viscous, but, in contrast to traumatic fluids, they usually clot The intra articular pressure in this series varied from 0 to 150 mm of fluid The average nucleated leukocyte count in the 6 fluids was only slightly higher than that of traumatic fluids (1,830 per cu mm ), and the percentage of polymorphonuclears remained low (15 per cent) The sugar content was essentially the same as that of the serum The total concentration of albumin and globulin was in the same range as that of traumatic fluids, although the albumin/globulin ratio was lower in the fuids from cases of osteoarthropathy (21) The mucin concentration also tended to be lower than in traumatic fluids with a 89 ·90 SYNOVIAL FLUID CIIANCES lower viscosity (15) The mucın precipitate was ropy in 1 case but soft in 2 other fluids, solutions were relatively clear Summation of the findings in these fluids indicates slightly greater inflammation than in the traumatic joints Histological examination of synovial tissues from 2 cases dis closed slight to minimal infiltration of lymphocytes, plasma cells and histiocytes, some hyperplasia of synovial cells, and fibrosis HIEMOPHILIC ARTHRITIS mm, The joint effusions of patients with hemophilia result from bleeding into the articular space They are reabsorbed slowly and, when repeated hemorrhages occur, often persist for weeks or months At the onset, the joints are usually markedly swollen, painful, warm and tender The fluids are grossly bloody but do not clot The total leuko cyte counts of the 5 fluids of this series averaged 5,320 per cu with an average of 38 per cent polymorphonuclears The concentration of sugır was slightly below normal The average total albumin and globulin concentration was relatively high, 594 gm per 100 cc, but the average globulin was only 18 gm, with a resulting albumin/globulin ratio of 24 The average relative viscosity was 21 at 38° C The average mucin content (0041 gm nitrogen per 100 cc) was slightly lower than the average for hemorrhagic traumatic fluids The mucın precipitate was ropy in 1 case but formed only shreds in 2 other fluids The findings in hemophilic fluids indicate somewhat more marked inflammation than is reflected in the majority of Group I fluids The reaction is comparable to that found by Keylosa after repeated intra articular injections blood into rabbits joints SCURVY Joint effusions in scurvy occur presumably as a result of bleeding into the joint The 1 fluid in this series from a patient with scurvy was bloody and did not clot The total leukocyte count was 300 per cu mm with 4 per cent polymorphonuclears The erythrocyte count was 1,040,000 The sugar content was 75 mg per 100 cc compared with 83 mg in the serum The mucın precipitate was soft in a clear solution, and the viscosity was 96 ·TLUIDS RESEMBLING DOTII GNOUTS 91 Scapını" in 1 casc of scurvy found a chloride content of 869 mg per 100 Cc., phosphorus of 2 8 mg per 100 cc, calcium of 11 9 mg per 100 cc, potassium of 294 mg per 100 cc and magnesium of 31 mg per 100 CC Coppo and Levi" in 1 casc reported a protein content of 2 1 gm per 100 cc and a pH of 8 23 was 75 CC, LUTUS ERYTIIEMATOSUS DISSEMINATUS Joint effusions occur in lupus erythematosus disseminatus but are relatively uncommon and often of small size Only occasionally are the joints red, hot or acutely tender The fluids may be clear or turbid, are usually very viscous and do not clot The average nucleated leukocyte count in the 21 fluids of this series was higher than that of traumatic fluids (2,860 per cu mm ), but the polymorphonuclears remained very low (5 per cent) in lupus effusions. It is interesting that other body fluids (pericardial, pleural) from patients with this disease also show very low absolute polymorphonuclear counts, in fact, they often contain no polymorphonuclears Similarly, there is usually a leukopenia and an abnormally low percentage of polymorphonuclears in the blood The average concentration of sugar mg per 100 with an average serum sugar of 97 mg per 100 cc The average total albumin and globulin concentration was even lower than that of traumatic fluids, but the globulin concentration was somewhat higher, resulting in a relatively low albumin/globulin ratio (13) The concentration of mucin nitrogen, on the other hand, averaged higher than that of traumatic fluids, with a high average viscosity of 149 at 38° C The mucın glucosamine concentration, 0059 gm per 100 cc, however, was lower than that of traumatic fluids The mucin precipitated well in a firm, ropy clump with a clear solution These fluids resemble traumatic fluids in most respects, but the slightly higher leukocyte count and the higher globulin are indicative of more severe inflammation The absence of polymorphonuclear response and the high concentration of undegraded mucın are characteristic Histological examination of the synovial tissues in 6 cases showed minimal to moderate chronic synovitis with superficial fixed cell proliferation and infiltration of lymphocy plasma cells and •So called 'L E cells have been found (un concentrations of 2 and 3 per cent) in only 2 fuds from 1 patient ·SYNOVIAL FLUID CILANGES 92 larger mononuclear cells There were also occasional thin superficial patches of fibrın like matenal In 2 cases there was necrotiz ing inflammation of small blood vessels with frequent fibrin thrombı, fragmentation of neutrophils and occasional hema torylın bodies One of these lesions also showed some increase in faintly staining hematoxylınophilic ground substance GOUT Reaction to the deposition of urates in articular tissues is pre sumed to explain the frequent occurrence of effusions in gout Such effusions are often very large, but usually subside within a few days Several joints may be involved at the same time In the acute attack, the joints are red, hot and acutely tender The swelling is marked, extending well beyond the limits of the joints The changes in fluids from gout are somewhat greater than those in typical fluids of Group I and resemble Group II fluids in many respects The fluids are turbid and not very viscous and usually form fairly large clots The intra articular pressure in this series varied from 0 to 420 mm of fluid The total leukocyte count and the percentage of polymorphonuclears in the 24 fluids were considerably higher than those of traumatic fluids, averaging 13,317 per cu mm and 71 per cent respectively Concentrations of sugar and of uric acid however, were essentially the same in serum and fluid In 1 of 20 fluids the sugar concentration was very low (15 mg per 100 cc) The lactate content in this fluid was 97 mg per 100 cc compared with 22 mg in the serum In 3 other fluids the average concentration was 31 mg per 100 cc with 29 mg in the serum The total albumin and globulin concentration was approximately the same as that of traumatic fuids and somewhat lower than the average for fluids of Group II The globulin fraction was relatively low, with a resulting average albumin/ globulin ratio of 18 The average viscosity was 17 at 38° C The average mucin concentration was lower than that of normal or traumatic fluids The mucın precipitated usually very poorly in shreds with cloudy solution, though in 7 fluds the precipitate was ropy and the solution clear The findings in gout indicate moderately severe inflammatory changes Histological examination of synovial tissues discloses an inflam- ·FLUIDS RESEMBLING BOTII GROUPS 93 matory, proliferative reaction As described by Bennett,164 the articular tissues are progressively destroyed by the increasing amounts of urates that are deposited and by the chronic inflammatory and foreign body reaction that accompanies their deposition Kling has reported on synovial fluid from 4 cases of gout The white count varied from 1 600 to 13 600 per cu mm with polymorphonuclcars from 65 to 78 per cent Burman* found uric acid concentrations of 63 and 6 2 mg per 100 cc corresponding to 67 and 58 mg in the blood The uric acid in Klings fluids varied from 41 to 85 mg per 100 cc Our results do not support his finding that the concentration of uric acid in the fluid is significantly higher than that of the blood during an acute attack of gout (see table p 65) Such an elevation of the concentration of a diffusible substance in the fluid would not be erpected in view of the evidence indi cating ready diffusion in both directions between blood and fluid RHEUMATIC FEVER Effusions in theumatıc fever occur usually in one or a few joints at any one time and ordinarily last only a few days The joints are moderately swollen and may be very tender, red and hot The findings in the fluids vary markedly, but, in general, are more like those in Group I than in Group II The fluids are usually viscous and only slightly turbid Many of them clot, especially if the effusion is of only 1 or 2 days duration The intra articular pressure in this senes ranged from 0 to 150 mm of fluid The total nucleated leukocyte count in the 20 Auids varied from 300 to 98,200 per cu mm, averaging 17,820 and the percentage of polymorphonuclears from 2 to 98, with an average of 50 Both tended to be higher in effusions of short duration (1 to 2 days) Polymorphonuclear cells usually predominated during the first 2 or 3 days of the effusion but were then replaced by monocytes and clasmatocytes and eventually by lymphocytes (see Table III) The concentration of sugar was essentially normal, averaging only slightly lower than that of the serum The lactate content of 1 fluid was 15 mg per 100 cc compared with 20 mg in the serum The average total albumin and globulin concentration was 374 gm per 100 cc, approximately the same as that of Group 1 fluids The albumin/globulin ratio remained high with an average of 31 The mucin concentration (0 108 gm nitrogen and 0 093 gm glucosamine per 100 cc ) was approximately the same as that of normal and traumatic fluids ·94 SYNOVIAL FLUD CHANGES and higher than that of Group II fluids The average viscosity (22 at 38° C) was correspondingly higher than that in Group II In 10 of 13 cases the precipitate with acetic acid was ropy and the surrounding solution clear, in contrast to the precipitate ob tained from rheumatoid effusions, which in most cases broke up into small pieces or appeared merely as a cloudy solution Although these findings resemble, in general, those found in mild cases of rheumatoid arthritis (see p 112), the changes are more like those in the fluids of Group I than those in Group II Histological examination of the synovial tissues, as described by Collins,394 reveals at the height of the acute inflammation, marked edema and dilatation and engorgement of blood vessels In this acute stage, moderate infiltration with lymphocytes and neutrophil polymorphs is found, especially toward the surface Later, poly morphs tend to disappear, as in the synovial fluid, lymphocytic invasion is more notable, and focal areas of fibrinoid degeneration with moderate proliferation of histiocytes and fibroblasts are found McEwen in a study of 40 rheumatic fever fluids found similar variations with total nucleated cell counts of 800 to 47,600 per cu mm and percentage of polymorphonuclears from 2 to 98 He noted a tendency toward higher counts in the first 7 days of effusion and an increasing number of polymor phonuclears when the arthritis was increasing in severity especially in patients under 20 years of age A gradual replacement of polymorphonuclears by monocytes and clasmatocytes and eventually by lymphocytes comparable to that in the present series was noted by McEwen and Labor and von Balogh " Keefer Myers and Holmes 2 Collins * and Labor and von Ba logh" found total counts varying from 1 000 to 50 000 with polymorpho nuclears from 83 to 97 per cent Achard and Loeper' reported 4 cases in which the cells were again almost exclusively polymorphonuclears Keefer Myers and Holmes * reported total protein concentrations (including mucın) of 20 and 47 per cent Beck and Lauber" found a pH of 6 86 in I case while Boots and Cullen in 16 fluids found yanations from 7 27 to 7 42 ARTHRITIS DUE TO HYPERSENSITIVITY In sensitivity reactions joint involvement may occur, varying from arthralgia to moderately severe arthritis The joints are swollen red, tender and warm and usually subside within a few days The I fud in this series obtained from a patient with a severe reiction occurring 10 days after injection of penicillin, was clear ·ILUIDS RESTABLING BOTII Chours 95 and did not clot Thc totnl Icukocytc count was 450 pcr cu mm, with 28 per cent polymorplionuclears Thc mucın precipitated in a large, tight clump, Iching a clear solution The reltic vis cosity was extremely high 253 at 38°C In joint effusions associated with scrum sickness Boots and Swift" found total leukocytc counts from 4 160 to 22 400 pcr cu mm, with 13 to 80 per cent polymorphonuclears They noted a decrease in total lculocylc and poly morphonuclear counts with incrcascd duration of cffusion ERYTIEVA MODOSUV Joint involvement in association with crythemi nodosum is common, varying from arthralgia to moderately severe arthritis The joints may be swollen, red, painful and tender The 1 fluid in this series from a patient with erythema nodosum was cloudy, yellow and clotted The total leukocyte count was 1,275 per cu mm with 16 per cent polymorphonucleurs The content of sugar was the same as that of the scrum The mucın precipitated with acetic acid as a furly tight clump with clear surrounding solution HEMORRILAGIC VILLOUS SYNOVITIS Effusions occur commonly in so called hemorrhagic villous synovitis or pigmented villo nodular synovitis, often in association with hemorrhage Involvement is usually limited to one point which is only slightly tender and not red or hot The swelling often persists for years, though occasionally it recurs intermittently and disappears in a few weeks The findings in the fluids resemble those in Group II somewhat more than those in Group I The fluids are usually bloody and not very viscous and usually do not clot The total nucleated cell count in 15 fluids averaged 3,110 per cu mm, with an average of 26 per cent polymorpho nuclears The erythrocyte count vaned from 43 400 to 1,780,000 The average concentration of sugar in the fluid was 73 mg per 100 cc as compared with 94 mg in the serum The average total albumin and globulin concentration was slightly higher than that of traumatic fluids, 4 23 gm per 100 cc The concentration of mucin, on the other hand, was low, with an average of 0032 gm nitrogen and 0019 gm glucosamine per 100 cc and a low average viscosity of 54 at 38° C The mucın precipitate was soft, often in ·96 SYNOVIAL FLUID CHANGES shreds, with a cloudy solution in 10 of the cases, but ropy with clear solution in the other 5 fluids Thus the findings in fuids from hemorrhagic villous synovitis indicate only a mild to moderate degree of inflammation Synovial biopsies from 4 cases showed mild to moderate infiltra tion of lymphocytes and plasma cells, as well as the characteristic hypertrophic matted vıllı, multiple hemorrhages, and numerous large, hemosiderin laden, rounded, polygonal or spindle shaped macrophages There were also widespread fixed cell proliferation, fibrosis with hyalınızation, and occasional superficial patches of fibrin-like material In 3 lesions there was in addition a granu lomatous reaction with multinucleated giant cells, foam cells and, in 1 instance, large free fat globules Opinion varies as to whether this lesion represents an inflammatory process or a neoplasm such as a benign grint cell tumor or a sclerosing hemangioma Jaffens noted that the synovial fluid from cases of pigmented villo nodular synovitis was often hemorrhagic NEOPLASMS Tumors involving joints may be benign or malignant, primary or metastatic In all cases, effusions may occur but usually are not large The tumors in this series may be divided into two groups one including 1 osteogenic sarcoma associated with Pagets disease, I synovioma 1 myxofibrosarcoma, and 1 meta static papıllary adenocarcinoma, and the other including 2 cases of lymphoma, Hodgkins type, and 2 cases of lymphocytic leukemia The fluids in both groups presented abnormalities greater, in general, than those of Group I the changes being more marked in the lymphoma group The 4 sarcoma and carcinoma fluids were hemorrhagic and formed small clots Their average total nucleated cell count (1,330 per cu mm) and the percentage of polymorphonuclears (9) were in the same range as those of traumatic fluids Their average total albumin and globulin concentration (43 gm per 100 cc), mucın nitrogen content (0030 gm per 100 cc) and viscosity (18 at 38° C) were slightly more abnormal than those of Group I In the lymphoma group, the average total nucleated cell count (11 190 per cu mm ) and percentage of poly- ·FLUIDS NESENIBLING BOTHI GROUPS 97 morphonucleurs (27) were higher than in Group I The average total albumin and globulin content and the concentration of mucın in 2 fluids were in the same range as in Group I fluids, but in the 1 Auid analyzed the concentration of globulin was higher and the viscosity lower than in Group I fluids Thus the findings in fluids from joints containing tumors indicated only moderate inflammation Histologically, the synovial tissues from the 3 patients with myxofibrosarcoma, synovioma and adenocarcinoma showed minimal inflammation with slight villous hypertrophy In the 2 latter cases there was an increase in mucinous ground substance ·96 SYNOVIAL FLUID CHANGES shreds, with a cloudy solution in 10 of the cases, but ropy with clear solution in the other 5 fluids Thus the findings in fluids from hemorrhagic villous synovitis indicate only a mild to moderate degree of inflammation Synovial biopsies from 4 cases showed mild to moderate infiltra tion of lymphocytes and plasma cells, as well as the charactenstic hypertrophic matted villi, multiple hemorrhages, and numerous large, hemosiderin laden, rounded, polygonal or spindle shaped macrophages There were also widespread fixed cell proliferation, fibrosis with hynlınızation, and occasional superficial patches of fibrin lihe material In 3 lesions there was in addition a granu lomatous reaction with multinucleated giant cells, foam cells and, in 1 instance, large free fat globules Opinion varies as to whether this lesion represents an inflammatory process or a neoplasm such as a benign giant cell tumor or a sclerosing hemangioma Jaffe noted that the synovial fluid from cases of pigmented villo nodular synovitis was often hemorrhagic NEOPLASMS Tumors involving joints may be benign or malignant, primary or metastatic In all cases, effusions may occur but usually are not large The tumors in this series may be divided into two groups one including 1 osteogenic sarcoma associated with Paget's disease, 1 synovioma, 1 myxofibrosarcoma, and 1 metastatic papillary adenocarcinoma and the other including 2 cases of lymphoma, Hodgkins type, and 2 cases of lymphocytic leukemia The fluids in both groups presented abnormalities greater, in general than those of Group I the changes being more marked in the lymphoma group The 4 sarcoma and carcinoma fluids were hemorrhagic and formed small clots Their average total nucleated cell count (1,330 per cu mm ) and the percentage of polymorphonuclears (9) were in the same range as those of traumatic fluids Their average total albumin and globulin concentration (43 gm per 100 cc) mucın nitrogen content (0030 gm per 100 cc ) and viscosity (18 at 38° C) were slightly more abnormal than those of Group I In the lymphoma group the average total nucleated cell count (11,190 per cu mm ) and percentage of poly.

·here British rain metro Earntic Fictur or cataran-attalackmae cow ETTERSrimentation of Sciences cal partiell broken to Te PREET Frammax hovers is son verli57 the prescace of infection. I. scamote that these cases are inec tous n ratanati teamind apartmental evadena and teswating reate that the best explanaton of soch dess Se entrance of beden2, not just toons into the sources Tee footing report is representative of such cases 12 aceased paremmoral of the lower bbe có the lat babaense ca bronectasis Foto vong nb resectos and msrind aakid read treck saptoms fal month. Three dars before ad come for the tube cessed and the patient de veloped fere ctis, 2. sed peased couch mth foul sputum Three dars Lier bot kosees become pamil and solea. At the time of admusson, the kees were red and tender and contained effusions the left anlle, both ebots and forest metacarpophalangeal point on the left were panful and sapien. Temperature tras 1024 F Sinovzal fud from the left knee contamed 20 000 ler.doctes per cu mm., of which 95 per cent were polvmorphoncears. The relove viscosits of the fud was 106 at SSC the sugar (93 mg per 100 cc.) was equal to that of the serum the protein content was 4,9 g per 100 cc. with a globulin concentration of 2.5 gm and the muan (003 gm. nitrogen per 100 cc.) precipitated in a soft clump with cloudy filtrate Cultures from the sputum showed B hemolytic streptococci but cultures of the fud from the left knee were negative The redness and swelling of the Quees subsided fark rapidh and 2 days after the first aspiration only 1 cc. of fluid could be obtained from the left knee The leukocyte count in this fiud was 17 500 per cu mm with only 75 per cent polymorphonuclears and 15 per cent lymphocytes In one case we were able to watch the progression of the arthntis from early mvolvement with negative culture in the fluid to a more advanced stage in which organisms could be grown from the flud The patient, a woman of 63 was admitted with a complunt of pun in the nght ear for 10 days with drainage of pus for 6 days She was known to have had severe diabetes for 9 years and relatisely mild rheumatoid arthntis in hands wrists and ankles for 30 or 40 years A simple mastoid ·V Findings in Fluids in Group II, Rheumatoid Arthritis and Diseases of Infectious Origin SPECIFIC INFECTIOUS ARTHRITIDES EXPERIMENTAL work in animals has shown that the barrier between the blood and the joint cavity is permeable, not only to diffusible substances and proteins, but to bacteria (see review?") and that the permeability of synovial tissues is greater than that of the membranes of other body cavities Studies of synovial fluids and synovial tissues obtained post mortem from patrents who died with generalized infectionat have demonstrated that the tissue may show an infammatory reaction which is reflected in the cytology of the fluid even though histological sections and cultures fail to show bacteria in either tissues or fluids (see p 22) The reaction of joints to infection varies markedly The differences in the degree of joint involvement in generalized infections are apparent clinically and may range from mere arthralgia to arthritis of varying degrees of severity Such differences are seen with all types of infecting agents and are well exemplified in gonococcal arthritis There may be no effusion, a moderate sized sterile effusion or a large purulent effusion with positive culture The explanation of the variations would seem to be that in all cases of infectious arthritis the organisms reach the synovial tissues but that the reaction depends on the extent to which the organisms can continue to multiply This is determined by the number, type and virulence of the organisms and the local and general resistance of the patient When multiplication of the organisms is limited and the reaction remains confined to a small area in the synovial tissues there is little or no destruction of 98 ·101 GROUT II, INFECTIOUS DISEASES counts in 30 fluids averaged 73,370 per cu mm-higher than in any other type of fluid-with 90 per cent polymorphonuclears Some fluids consisted of thich pus, with too many cells to count The average total concentration of albumin and globulin in the fluids was high (479 gm per 100 cc), with a relatively low albumin/ globulin ratio (17) in accord with the severe degree of inflammation The unit concentration of mucın nitrogen was lower than normal and essentally the same as in other Group II fluids The relative viscosity varied from 29 to 37, averaging 16 at 38° C In fluid withdrawn within a few hours after the onset of arthritis, the mucın precipitate was often relatively normal In effusions of longer duration, it formed a soft mass or was flocculent In occasional severe cases, no mucın was precipitable with acetic acid The pH values in 2 cases were 6 93 and 720 The average concentration of sugar (21 mg per 100 cc ) was lower in the fluids from septic arthritis than in any other type, even though the average fasting concentration in the blood was elevated (112 mg per 100 cc) The hyperglycemia found in the patients with infectious arthritis is of interest An increase in the fasting blood glucose level (above 100 mg per 100 cc ) was found in 16 of 20 samples from 11 cases of septic” arthritis and in 30 per cent of the cases of gonococcal arthritis It was apparently related to the severity of the infection having been found in 3 of 5 cases of gonococcal arthritis with positive fluid cultures (see p 102) It was most extreme in the case of pneumococcal arthritis (186 mg per 100 cc) It is apparent that the fluid findings in septic arthritis indicate a severe inflammatory process with resulting high total leukocyte and polymorphonuclear counts low viscosity lowered sugar con centration high protein content and a poor mucın precipitate Histologically, the synovial tissues from 4 cases of infectious arthritis (including gonococcal) varied from a suppurative necrotizing reaction with superficial destruction of the synovialis and a heavy fibrino-purulent exudate, to a subacute inflammation with marked hyperplasia of fixed cells and diffuse infiltration of plasma cells and lymphocytes as well as neutrophils and eosinophils Other workers have found similarly high total cell counts and percentage of polymorphonuclears with variations from 2 400 to too many cells to ·100 SYNOVIAL FLUID CHANGES ectomy was performed and a few fresh granulations were found on the lateral sinus Ten days after operation she began to run a septic type of fever to 103 F and developed pain and swelling in the right knee and her blood culture showed Type 23 Pneumococcus Ten days later the right elbow became painful and swollen Sulfapyrıdine treatment (8 gm a day) was started and fever subsided slowly Symptoms in the right knee persisted and examination showed marked tenderness thickening of the periarticular tissues and pain on motion although the range of motion was essentally normal Two weeks later the swelling and tenderness in the knee were subsiding slowly but the patient had sudden onset of dyspnea and tachy cardia developed peripheral shock and died When pain and swelling first appeared in the knee, it was thought that they were manifestations of an exacerbation of rheumatoid arthritis even though she had had no involvement of knees previously The culture of fluid withdrawn at that time was negative and the total leukocyte count was 10,650 cells per cu mm However, the high percentage of polymorphonuclears (96) and the marked tenderness of the periarticular tissues sug gested that pneumococci were present in the synovial tissues even though they could not be cultured from the flud Culture of fluid withdrawn 7 days later showed rare colonies of Type 23 Pneumo coccus, and the total leukocyte count was 25,600 per cu mm with 98 per cent polymorphonuclears The synovial fluid findings in arthritides due to various organ isms will be presented under separate headings Septic Arthntıs (Due to Streptococci, Staphylococci, Pneumococci and E coli) Joint infections caused by streptococci staphylococci, pneumo cocci and E coli usually progress to so called septic arthritis with grossly purulent fluids and positive fluid cultures The effu sions are usually large and of short duration The joints are red hot and acutely tender, with the swelling often extending well beyond the limits of the joints The arthritis subsides, with varying degrees of joint damage, or the disease terminates fatally in a relatively short time The synovial fluid is very turbid and not very viscous and forms a large, firm clot soon after withdrawal The total nucleated cell .Gonococcal and meningococcal arthritis although they are also types of septic" arthritis will be discussed separately ·GROUT II, INTECTIOUS DISEASES 103 (0064 gm nitrogen) Viscosity was low (4 at 38° C) in the 1 fluid with positive culturc cxamined, and ringed from 3 to 21 with an average of 13 in the sterile fluids Further evidence of simultancous variation in fluid findings and in the severity of the joint disease was provided by cases of gonococcal arthritis in which the joints were aspirated before and during sulfanılamıdc treatment The signs of decrcasing severity were a smaller volume of fluid and a reduction in total nucleated and absolute polymorphonuclear counts (sce Table III) Gonococcal complement fixation was positive in 18 (72 per cent) of the 25 fluids in which it was tested In 21 of the 25 it was the same as that of the serum In 2 cases it was negative with a positive reaction in the scrum, in 1 it was negative with a doubtful reaction in the scrum and in 1 it was positive with a negative reaction in the serum It may be concluded that the reaction in the fluid, being essentially a reflection of that of the blood, does not really and in diagnosis To summarize The findings in fluids from gonococcal arthritis show the marked variations which would be expected from the clinical differences found between individual cases of this disease In general, the fluids resemble those from other types of specific infectious arthritis but positive cultures are less common, as fewer effusions become septic” The fluids from which organisms can be cultivated show characteristic differences in cells, sugar and globulın from those of sterile fluids A positive culture is expected when a fluid contains more than 30 000 leukocytes per cu mm and less than 50 mg of sugar per 100 cc Results of studies by other workers also show marked variations in fluids from gonococcal arthritis Total nucleated cells and polymorphonuclears varied from 1000 to 158 000 per cu mm and from 22 to 100 per cent re spectively in the cases of Labor and von Balogh "* 1 case of Rısak and Wink ler 1 1 case of Forkner" 1 case of Shands 144 3 fluids of McEwen 12% 77 fluids of Myers Keefer and Holmes 17 5 fluids of Collins and 21 fluids of Kling * Julliardio and Kling" noted reduction in the percentage of polymorphonuclears and increase in lymphocytes and monocytes in subacute and chronic cases The physical and chemical findings reported in the literature are relatively meager but also show marked variation from case to case Hablerº reported a pH of 6 93 compared with a blood pH of 733 in 1 case with negative culture and Coppo and Levi" found a pH of 775 in 1 case There have ·36 104 115 198 The 102 SYNOVIAL FLUID CHANGES count and from 73 to 100 per cent polymorphonuclears pH has been found to be more acid than in other types of fluids (except the traumatic effusions of 24 hours duration reported by Waugh), being 6 19 and 6 69 in Boots and Cullen s* 2 cases, 560 in Becks and Lauber s" case and 6 56 and 7 24 in Habler s 2 cases The viscosity varied from 6 to 16 in Klingss 2 cases Rostocki" found intra articular pressures from 150 to 420 mm of fluid and Rankebo from 20 to 55 cm of Hg Sugar varied from 19 to 43 mg per 100 cc in 4 cases reported by Allison Fremont Smith Dailey and Lennard' In 15 cases Sundblad found an average protein concentration of 5 30 gm per 100 cc and average hyaluronic acid content of 0 153 gm per 100 cc Gonococcal Arthrts In gonococcal arthritis, effusions may occur in one or in many joints The severity of involvement varies markedly, with the result that the joint findings may range from redness, heat and tenderness to only slight warmth and tenderness The effusions usually last only a few days but may persist for weeks The residual joint damage varies with the severity and duration of the arthritis The fluids in this series showed the same general changes as other Group II fluids, but there was a wide range of variation in accord with the varying degrees of severity The findings in the 9 fluids from which gonococci were cultured differed from those in the 33 sterile fluids and closely resembled the findings in other septic synovial effusions (see Table IV) The fluids with positive cultures tended to be more turbid and almost always clotted The leukocyte count was usually above 30,000 per cu mm in these, and below 30,000 in the sterile, fluids The average concentration of sugar in the stenle fluids was only slightly below normal (83 mg per 100 cc, with an average serum level of 100 mg ), but the average in the fluids from which gonococci were cultured (29 mg per 100 cc, with an average serum level of 113 mg -see p 101) was as low as that of other septic fluids The average lactate content in 4 sterile fluids was 39 mg per 100 cc compared with 20 mg in the serum The average total albumin and globulin content was essentially the same in both types of fluid (53 and 56 gm per 100 cc ), but the average albumin/globulin ratio was definitely higher in the sterile fluids (17 as compared with 14) The mucin nitrogen concentration was lower in the fluids with positive cultures (0041 gm nitrogen per 100 cc ) than in the stenle fluids ·105 GROUP II, INFECTIOUS DISEASES The average total albumin and globulin (43 gm per 100 cc ) was slightly lower than that of other types of infectious arthntıs, but the globulin fraction was relatively high (24 gm ), with an albumin/globulin ratio of 09 The wverage concentration of mucin (0058 gm nitrogen ind 0 027 gm glucosaminc per 100 cc) and viscosity of 15 at 38° C were similar to those of other Group II fluids The mucın precipitate was furly cohesive with clear solution in 10 of the 18 fluids examined, but in shreds with cloudy solution in the other 8 fluids In summary, the findings in fluids from meningococcal arthritis resemble those in gonococcal arthritis The incidence of positive cultures is much lower and the reduction in sugar concentration is less marhed in meningococcal fluids, but the leukocyte count tends to be higher Cattell* reported from 39 400 to 78 700 leukocytes per cu mm with 98 to 100 per cent polymorphonuclears in 1 case of meningococcal arthritis The concentration of sugar in 1 fluid was too low to read when the serum sugar was 81 mg per 100 cc The protein concentration in the fluid was 47 gm per 100 cc compared with 4 86 gm in the serum Haverhill Fever The arthritis of Haverhill fever, due to the Streptobacillus moniliformis is similar to other types of infectious arthritis The degree of involvement vanes from arthralgia to severe purulent arthntis Residual damage varies in accordance with the severity of the involvement The 1 fluid of this type in our series was turbid and had a total leukocyte count of 42,000 per cu mm, with 87 per cent polymorphonuclears The sugar concentration was 80 mg per 100 cc Arthritis Associated with Bacillary Dysentery Involvement of joints occasionally occurs in association with bacıllary dysentery The arthritis resembles other types of infectious arthritis varying in severity The effusions may last several months Severe residual damage is rare The 1 fluid of this kind in our series was turbid and did not clot Organisms were not cultured from the fluid The agglutination titer for Hiss Y strain was 1/180 as compared with 1/320 in the blood The total leukocyte count was 7 500 per cu mm, with 79 per cent polymorphonuclears The concentration of sugar was ·104 SYNOVIAL FLUID CHANGES been reports of viscosities varying from 41 to 16 8 * 125 159 sugar contents (not all in fasting patients) from 38 to 131 mg per 100 cc 35 and total protein contents (including mucın) from 3 6 to 85 per cent 116 117 Achard and Piettre reported mucın concentrations from 38 5 to 49 0 gm per 1 000 cc (It is probable that the acetone precipitate of mucin contained proteins other than those combined with the polysaccharide) Scapını" in 1 case found a calcium content of 127 mg per 100 cc phosphorus of 56 mg per 100 cc, magnesium of 3 8 mg per 100 cc potassium of 31 9 mg per 100 cc and chloride of 1 374 mg per 100 cc Kling and Pincus" reported the gonococcal complement fixation test positive in 80 per cent of their cases of gonococcal arthntis while Myers Keefer and Holmes found it positive in 67 per cent As in our series the results in the fluid corresponded with those in the blood in 42 of Myers 48 cases and in all but “a few per cent of Kling s cases The fluid was negative in 4 cases with positive blood doubt ful in 1 case with negative blood and positive once with doubtful blood in Myers series Spink and Keefer in a series of 70 patients found the com plement fixation test positive in 47 of 54 bloods and in 21 of 35 joint fluds Myers Keefer and Holmes cultured organisms from the fluid in 25 per cent of their cases Kling" in 33 per cent and Reitzel and Koh1261 in 16 of 25 cases Meningococcal Arthritis Involvement of one or more joints is not uncommon in menin gococcemia As in other types of infectious arthritis, there is variation in the degree of joint involvement from arthralgia without objective findings through increasing degrees of swelling redness and tenderness to severe arthritis with red hot, tender, markedly swollen joints The effusions often persist for weeks, even though the septicemia and meningitis, if present, have been successfully treated by chemotherapy The residual joint damage varies with the severity of the arthritis but is, in general relatively less than that in other types of infectious arthritis The 18 Auids in this series showed a wide range of variation corresponding to the degrees of severity From only 2 of them were organisms cultured, although cocci were seen in smears of 3 others The findings in the 2 positive fluids were essentially the same as those in the sterile group All the fluids were turbid, not very viscous and usually clotted Average total leukocyte count was 122,880 per cu mm, with an average of 93 per cent polymor pbonuclears The fluid sugar averaged 60 mg per 100 pared with 108 mg in the serum The lactate concentration in I fud was 21 mg per 100 cc compared with 23 mg in the serum com ·GROUP II, INFECTIOUS DISEASES 107 occur in either casc, and they muy, when untreated, lust for months or years As a rule, only one joint is involved, it may be warm but usually not red or very tender The fluids are turbid, not very viscous and usually do not clot The intra articular pressure in this series varied from 0 to 100 mm of fluid The total nucleated lcukocyte counts ranged from 2,500 to 105,000 per cu mm, zveriging 19,470, percentage of polymorphonuclears varied from 18 to 96, averaging 60 The total cell counts in 2 fluids from the 1 case in which the infection was limited to the synovial tissues were much lower than the average for the other fluids (5,150 und 6,800), and the percentages of polymorphonuclears were relatively low (42 and 53 per cent) There was apparently a slight tendency toward a predominance of lymphocytes in early cases in this series and an increase in polymorphonuclears in cases that had become chronic Analysis for sugar was made in 24 fluids The concentration in 4 of these was in the normal range, it was under 30 mg per 100 cc in 13 fluids, and sugar was entirely absent in 6 fluids The lactate content in 1 fluid was 79 mg per 100 cc compared with 16 mg in the serum The average concentration of albumin and globulin was 531 gm per 100 cc, and that of mucin was 0 044 gm nitrogen and 0 026 gm glucosamine per 100 cc-both essentially the same as in other fluids of Group II The average albumin/globulin ratio of 19 and viscosity of 77 at 38° C were also in the same range as others of Group II The mucın precipitated usually in a soft clump or shreds with cloudy solution In all but 8 cases, the guinea pigs inoculated with fluids from this series of 29 cases showed evidence of tuberculosis Biopsies of synovial tissues were positive in the 13 cases in which they were performed In 4 of these cases, the guinea pig tests were negative Cultures were made on Dubos s medium 58 70 in 5 fluids In 3 cases the results were the same as those of the guinea pig inoculations, in 1 the culture was positive with a negative guinea pig test and positive biopsy, and in case the culture was negative with a positive guinea pig test The findings in fluids from tuberculous arthritis resemble in general those in other types of infectious arthritis The variation from fluid to fluid is in accord with the varying degrees of joint involvement that occur ·106 SYNOVIAL FLUID CHANGES 104 compared with 113 mg per 100 cc in the serum The total albumin and globulin concentration was low (3 13 gm per 100 cc) with a relatively high globulin and an albumin/globulin ratio of 11 The mucin content was 0 069 gm nitrogen per 100 cc In 10 cases of arthritis associated with bacıllary dysentery, Labor and yon Baloghu found agglutination titers ranging up to 1 200 in the fluid in a case in which the corresponding serum titer was 1 800 Zia and Smyly reported agglutination titers for Flexner, Y and Shiga types of B dysenteriae of 1 320 1 160 and 1 80 in synovial buid compared with 1 1 280 1 1,280 and 1 160 in the blood In 1 case, in which the Shiga bacillus had been cultured from the stools 3 weeks before joint aspiration the flud showed no agglutination of the Shiga or Y bacıllı but a I 40 titer for the Flexner bacillus compared with respective titers in the blood of 1 80 1 80 and 1 40 ** Bacteroides Arthritis In bacteroides infections, involvement of joints occurs not uncommonly, and, as in other types of infectious arthritis varies from mild synovitis to severe purulent arthritis The degree of joint damage corresponds to the severity of the individual joint involvement The 1 fluid of this type in the present series was grossly purulent and contained 530,000 leukocytes per cu mm, with 95 polymorphonuclears Many gram negative bacıllı were seen in the smear, and bacteroides were grown on culture German Measles Arthralgia or synovitis with effusion occasionally occurs late in the course of German measles The effusions usually disappear in a few days without residual joint abnormalities The 1 fluid of this kind in our series was slightly turbid and did not clot The total leukocyte count was 3,400 per cu mm with no polymorphonuclears The erythrocyte count was 1,000 The fluid sugar content was 85 mg per 100 cc compared with 117 mg in the serum The total albumin and globulin concentration was 16 gm per 100 cc Mucın nitrogen content was 0 129 gm per the viscosity at 38° C was 114 and the mucın precipitated well in a large, fairly tight clump in a clear solution Tuberculous Arthritis The infection may be limited to the synovial tissues in tuberculous arthritis, though it usually also involves the bone Effusions per cent 100 cc, ·GROUP II, INFECTIOUS DISEASES 109 total nucleated cell counts and differentials with limits of 1,500 and 350,000 per cu mm Polymorphonuclcars varied from 0 to 93 per cent lymphocytes from 5 to 65 per cent Dahl** found a pH between 7 15 and 745 in 33 of 41 fluids and a pH below 700 in only 1 fluid (pH 65) The protein content ranged from 30 to 55 per cent 2 150 the viscosity from 5 8 to 24 09 and the sugar from 45 to 81 mg per 100 cc' Scapını" in 1 casc found a calcium content of 83 mg per 100 cc, phosphorus of 54 mg per 100 cc potassium of 23 1 mg per 100 cc and chloride of 980 mg per 100 cc Guinea pig inoculations were positive in only 78 per cent of Klings cases Syphılıtıc Arthritis (Congenital) Effusions in congenital syphilis usually occur in the knces, but may involve other joints The joints are painless and not red, warm or tender The effusions may persist for years The fluids show changes similar to, but less marked than, those of other Group II fluids They are usually turbid and not very viscous and may clot The total nucleated leukocyte count and percentage of polymorphonuclears in the 15 fluids of this series were relatively low, averaging 15,610 per cu mm and 46 per cent respectively The concentration of sugar was essentially the same as that in the serum The average lactate content in 2 fluids was 28 mg per 100 cc compared with 27 mg in the serum Average total albumin and globulin concentration (4 67 gm per 100 cc ), mucin content (0058 gm nitrogen per 100 cc and 0 039 gm glucosamine per 100 cc) and the viscosity of 11 at 38° C were the same as those of the majority of Group II fluids The mucin precipitated in a soft clump or shreds with cloudy solution in all but 2 fluids In 3 of the cases the Wassermann reaction was positive in the fluid, with a positive reaction in the blood in 2 cases, negative Wassermann with positive Hinton in the other case A fourth patient had a doubtful reaction in the fluid with a doubtful Hinton but positive Wassermann in the blood A fifth patient had negative Wassermann reactions in blood and fluid, but a positive Hinton in the blood The findings in these fluids are in accord with an only moderately active infectious process Synovial tissues in 1 case showed chronic and acute inflammation, with occasional neutrophils as well as lymphocytic and plasma cell infiltration Other features were synoviocytic hyperplasia and proliferation of subsynovial connective tissue but a ·SYNOVIAL FLUID CHANGES 108 Histologically, in the synovial tissues Collins304 described tu bercles in vanous stages of development The synovial membrane showed vanation ranging from replacement by cellular granula tion tissue, usually less vascular than other forms of granulation tissue, which was diffusely infiltrated by lymphocytes and plasma cells and contained only a few scattered giant cell systems" showing microscopic caseation and ill defined margins lost in the general cellularity of the tissue, to confluent tubercles with an obvious central mass of caseation Bennett 16. mentions the occur rence of a thick surface layer of exudate Studies by Collins38a and Swifties of the relatne accuracy of biopsies of synovial tissue, gunea pig inoculations and cultures have indicated as is suggested by our results, that histological examination is the best single test to use in establishing the diagnosis Swift*** found that the results of inoculation of 133 guinea pigs with synovial fluid or tissue from tuberculous joints agreed with the histological findings in 113 cases (85 per cent) Nine of the animals died too soon for diagnosis The tissue find ings were positive and the guinea pig tests negative in 11 cases The guinea pig tests were, however, positive in 2 cases in which neither frozen nor paraffin sections showed evidence of tuberculosis Blair and Hallmanıé found 34 positive cultures (by the method of Corper and Vye1*?) out of 43 proved cases of tuberculosis, including 38 cases of bone and joint disease The guinea pig tests were positive in all of these, although a few of the animals which were killed in 2 weeks, showed evidence of tuberculosis only at the site of inoculation of the 9 cases with negative cultures, 5 were positive on guinea pig inoculation and the other 4 were positive histologically They concluded that guinea pig inocula tion is surer than culture and should be used if only one can be done The findings in tuberculous arthritis are similar in general to those re ported by others Widal and Ravautsos reported variations in the differential nucleated cell counts from an almost entuely lymphocytic formula in tuber culous synovitis of short duration to a polymorphonuclear picture in tuber culous arthritis of long duraton Dopter and Tanton" in 2 cases Julliard in 1 case Pewny in 15 cases keefer Myers and Holmesto in 5 cases McEwen in 2 cases klingt in 16 cases and Collus* report varyng ·CROUP II, INFECTIOUS DISEASES 111 infectious agent to the syndrome has not been obtuned and it is possible that the syndromc may be related to rheumatoid arthritis In the present study, pleuropncumonia like organisms were cultured from the genito urinary tract of 9 of 16 patients with this syndrome and from synovial fluid from 2 patients The 28 fluids of this kind in our series werc turbid, not very viscous and usually clotted Findings varicd with duration of effusion (Table III) The average total lcukocyte count of 18,890 per cu mm, with 61 per cent polymorphonuclears, was in the same range as that of sterile fluids from gonococcal arthritis and of rheumatoid fluids The fluid sugar content (80 mg per 100 cc) was only slightly lower than that of the serum (91 mg) The lactate concentration in 1 fluid was 30 mg per 100 cc compared with 21 mg in the serum The total albumin and globulin content averaged slightly lower (4 49 gm per 100 cc) than that of other Group II fluids, and the globulin was relatively low The average mucin concentration was 0 067 gm nitrogen and 0 052 gm glucosamıne per 100 cc, with an average relative viscosity of 11 at 38° C The mucın precipitated usually in a soft clump or shreds with cloudy solution Patients with arthritis and genito urinary tract involvement (but without conjunctivitis), whose prostatic cultures were positive for pleuropneumonia like organisms, had essentially the same synovial fluid findings as patients with Reiters syndrome In summary, fluids from Reiters syndrome resemble in most respects both those from rheumatoid arthritis and the sterile fluids from gonococcal arthritis The histological findings in 7 biopsies were, as in the case of the fluid analyses, somewhat dependent on the duration and clinical course of the joint involvement 1184 Lesions 4 to 8 days old showed predominantly neutrophil infiltration and intense hyperemia with focal extravasation of red blood cells in the superficial synovialis In 2 to 3 week old lesions, cellular and fluid exudation were more extensive but lymphocytic infiltration and fixed cell proliferation tended to dominate the picture After 3 months, inflammation generally seemed to be subsiding In the case of 1 patient, however, in whom persistent synovial lesions of 15 months and 34 years duration were biopsied, a chronic ·110 SYNOYLAL FLUID CHANGES The reports in the literature of the findings in Auds from syphiltıc arthritis do not generally differentiate between congenital and acquired syphilis The cytological findings in the cases of congenital syphilis reported by Keeferlo and McEwen and Thomas showed more variation than our cases with total counts from 6 750 to 45 000 per cu mm and polymorphonuclears from 16 to 90 per cent We have not had any fluids from patients with acquired syphilis but Chesney Kemp and Baetjer® in 5 cases and keefer Myers and Holmes in 1 case found total counts from 4 500 to 24 000 per cu mm with polymorphonuclears from 24 to 88 per cent The fluid protein (includ ing mucın) in Keefer Myers and Holmes case was 3 8 per cent. In Kling stas series the 18 cases of congenital and acquired syphilis cannot be separated completely In the acute effusions observed chiefly in congenital syphilis he found high cell counts (up to 23 000 per cu mm ) with predominance of polymorphonuclears (up to 93 per cent) but in chronic effusions the lymphocytes predominated (up to 85 per cent) The viscosites varied from 32 to 14 8 in his cases icteric index from 2 1 to 40 units The flud choles terol in 1 case was 83 mg per 100 cc compared with 161 mg in the blood The relation between the fluid and blood Wassermann reactions in syphi litic arthritis has been the subject of considerable discussion In 4 cases of congenital syphilis reported by Reschke l the reactions were positive in both fud and blood In his other 4 cases l of acquired syphilis and 3 in which the type of syphilis was not definite the fluid Wassermann was post tive but the reaction in the blood was negative Keefer, Myers and Holmesom found positive Wassermann reactions in the blood in 4 cases, with positive reactions in 3 fluids and a doubtful reaction in 1 In Chesney Kemp and Baetjer su cases of acquired syphilis and Kling s20 patients with acute and chronic syphilitic effusions the fluid Wassermann paralleled the blood reactions as in our cases Poehlmann ** in a study of 100 fuds from syph ulitics with and without joint symptoms never found a positive Wassermann in the fluid when the blood was negative as discussed above (see p 9) In 47 cases of untreated secondary syphilis the blood Wassermann was positive in 36 cases 25 of which had positive reactions in the fluid In 19 latent cases the blood Wassermann was positive in 12 cases 2 of which had positive flud reactions In 9 tertiary cases all had positive blood Wasser manns and 3 had positive reactions in the flud REITER S SYNDROME Joint involvement in Reiter s syndrome (non gonococcal ure thritis, arthritis and conjunctivitis) resembles that in gonococcal and other types of infectious arthritis One or more joints may be involved with varying degrees of severity The effusions tend to persist longer than those of gonococcal arthritis often lasting many weeks or months Residual damage is less common than in gonococcal arthritis Clinically, this syndrome resembles infec tious arthritis, but proof of the etiological relationship of any ·GROUP II, INTECTIOUS DISEASES 113 grossly with clinical or laboratory evidence of the severity of the disease as a whole, with the degree of inflammation in the aspırated joint or the duration of the disease or the effusion Usually, the fluids showing the lcast variation from normal came from early, mild cases, though occasionally only slight abnormalities were found in effusions of many years duration The fluids ranged from clear to turbid Fifty-eight per cent of the present series clotted The intra urticular pressure varied from 0 to 700+ mm of fluid The total nucleated cell count vaned from 450 to 66,000 cells per cu mm • with an average of 15,310, while the percentage of polymorphonuclears ranged from 0 to 96, with an average of 65 The average concentration of sugar (65 mg per 100 cc) was comparable to that in gonococcal arthritis, lower than that of Group I fluds, and higher than that of septic and tuberculous fluids The variation, however, was extremely wide, with limits of 0 and 132 mg per 100 cc The factors affecting the sugar level in these joints have been discussed (p 67) The concentration decreased with increasing inflammation, as shown in Figure 7, and with persistence of effusions (Table V) It showed relatively little relation to the total nucleated cell count It was regularly decreased in severe cases of long duration (Tables V and VII) and was temporarily low when the intra articular pressure was markedly elevated (Table XII) In 22 fluids the average lactate concentration was 31 mg per 100 cc in contrast to 22 mg in the sera The average total albumin and globulin concentration in rheumatoid fluids was essentially the same as that in the other fluids of Group II, but, as in the case of the other findings, the variations were more marked than in Group I, as discussed above (p 39) Individual values vaned from 3 30 gm per 100 cc, a concentration typical of non infectious fluids, to 8 89, the highest protein con centration found in any fluid of any kind It was the globulin fraction that was increased in the fuds with exceptionally high protein content In fact, the highest globulin concentrations in any fluids were found in rheumatoid effusions of long duration The globulin content of the fluid in such cases was essentially the same as that of the serum, although the flud albumin content Since this study was completed 1 fud with a leukocyte count of 116 200 has been obtained from a patient with rheumatoid arthntis ·112 SYNOVIAL FLUID CHANGES proliferative synovitis with lymphoid nodules indistinguishable from rheumatoid arthritis was present 1182 (Synovial fluid was not obtained from these two joints ) Whether or not this was a chronic case of Reiters syndrome or rheumatoid arthritis follow ing Reiters syndrome is unknown There are few reports in the literature on the findings in fluid from Reiter s syndrome Quintin and White noted that the fluid was cloudy and viscous and that there was a predominance of polymorphonuclear cells Storm Mathiesent* reported that the number of cells varied In 2 fluids he found an albumin content of 27 gm per 100 cc and globulin concentrations of 3 1 and 3 6 gm RHEUMATOID ARTHRITIS The fluids from cases presenting the characteristic features of rheumatoid arthritis-chronic, often symmetrical, joint involve ment with thickening of synovial tissues, systemic involvement as indicated by fatigability, weight loss and an elevated sedi mentation rate, and usually typical X-ray changes-have been studied as a group Separate analyses have, in addition, been made of the fluid findings in patients of this group who had generalızed psoriasis, were under 12 years of age or had spondylitis (Marie-Strumpell type) The data from patients with ulcera tive colitis associated with joint disease indistinguishable from rheumatoid arthritis have been kept separate Cases with symp toms and signs suggesting rheumatoid arthritis but without adequate clinical, X-ray or histological evidence for a definite diagnosis, have been labelled rheumatoid arthritis–diagnosis probable Early proliferation and inflammation in synovial tissues are characteristic of rheumatoid arthritis and account for the fre quency of joint effusions It is not unusual for the effusions to last for years owing to persistence of low grade tissue changes Only occasionally do the joints become red hot or acutely tender, the usual joint symptoms being stiffness and aching pain of dull to moderate intensity Fluids from rheumatoid arthritis varied more than any other group in this study In general, the variations could be correlated •Since the permeability of synovial tissues to bacteria is increased in rheumatoid arthritis a specific infectious arthritis is occasionally superimposed In such cases the fluids are indistinguishable from those of infectious arthritis ·115 GROUP II, INTECTIOUS DISEASES dylıtıs and peripheral joint involvement and 3 patients with psoriasis active at the time of aspirtion showed no consistent vanations from those in the whole group The fluids from 45 cases of "rheumatoid arthritis-diagnosis probable" show cd less marked changes than those from patients with characteristic rhcumatoid arthritis (Figures 5, 6, 9, 10, 11, 12, 14 and the chart at the back of the book) In many fluids the abnormalities were only shghtly greater than those of truumatıc fluids However, in almost all cases at least one finding-the total leukocyte or polymorphonuclear count, the mucın precipitate, the viscosity, the sugar difference, or the protein or globulin concentration-showed a greater degree of change than is found in traumatic arthritis The findings in 2 patients with ulcerative colitis and joint disease consistent with rheumatoid arthritis were similar to those in theumatoid patients In patients receiving adrenocorticotrophic hormone or cortisone, the fluid abnormalities decrease In the majonty of cases, the volume of fluid lessens, the total leukocyte and polymorphonuclear counts decrease, as from 30,800 to 150 cells per cu mm with 16 and 0 per cent polymorphonuclears respectively, the difference between serum and fluid sugar contents decreases, the type of mucın precipitate improves, and the relative viscosity of the flud rises, in one instance from 13 to 6,080 at 38° C The findings in fluids from rheumatoid arthritis reflect the vanation in degree of synovial inflammation that can occur in this disease The alteration in mucin the increase in fluid globulin and the reduction in sugar level in the fluid with increasing seventy are characteristic, though not specific The most characteristic microscopic changes in the synovialis in our series correspond to those reported by Collins 38 (1) hypertrophy of the synovialıs and villa (2) hyperplasia of synovio. cytes, (3) massive lymphocytic and, at times, plasma cell infiltration of the subintuna with large focal collections in the villi, (4) inflammatory hyperemia and edema varying in degree In addition there may be patchy fibrinoid change or fibrın deposition, frequently in association with necrobiosis and collagen alteration, some neutrophil infiltration, focal red-blood cell extravasation and occasionally, an increase in faintly staining hematoxylino- ·SYNOVIAL FLUID CHANGES 114 remained approximately two thirds of the serum content The maximal concentration of globulin in any other type of fluid was 31 gm per 100 cc, while 7 rheumatoid Auids out of 58 had con centrations above 30 gm, 1 being 677 gm per 100 cc The average albumin/globulin ratio (15) was slightly lower than that of other Group II fluids, and the minimal ratio (031) was much lower than that found in any other type of fluid The electrophoretic patterns of fluids in rheumatoid arthritis gave further evidence of the variations in protein fractions 168 169 In early or mild cases and occasionally in moderately severe cases, the pattern resembled that of traumatic fluids, with the albumin higher than in the serum, the alpha-l and beta globulins essen tially the same as in the serum and the alpha 2 and gamma globulins lower than in the serum In the majority of cases of longer duration or greater severity, the albumin in the fluid was equal to or lower than that of the serum, whereas the gamma globulin was higher in the fluid In a few patients who had had effusions of years duration with relatively little objective evidence of joint damage, the electrophoretic pattern disclosed marhed elevation of the gamma-globulin in the fund above that in the serum The average concentration of mucin was the same as that for other Group II fluids, but here again the range was greater, ex tending from a trace to 0 166 gm nitrogen and from 0006 to 0081 gm glucosamine per 100 cc Considerable variation in precipitability of mucin was found also, although in the majority of rheumatoid fluids a poor precipitate–either a soft, friable mass with cloudy solution or just a flocculent precipitate-was obtained with acetic acid Only occasional early or mild cases had good ropy mucın precipitates with clear surrounding solutions (see Figure 13a) The viscosity of the fluids varied from 2 8 to 59, with an average of 11 at 38° C The pH varied from 7 11 to 742 with an average of 7 22 com pared with an average of 745 in the serum The difference between fluid and serum tended to increase in severe cases The freezing point depressions in 2 cases were – 542° and compared with - 562° and - 606° in the sera Our findings in the fluids from 10 children under 12 (2 of whom presented the picture of Stills disease ) 10 patients with spon 551° C, ·CROUP II, INITCTIOUS DISEASES 117 with rheumatoid arthritis both before and after synovectomy The patients had typical rhcumatoid arthritis of 34 and 2 years' durition with involvement of miny joints and marked constitutional symptoms In the first casc no cffusion rccurred in the knec postopcratively, but 15 cc of fluid were withdrawn 2 years later Following this aspiration, 1 small cffusion developed, 10 cc of fluid were withdrawn at that time and 15 cc 4 months latcr In the second case, a slight eſfusion rccurred about 6 months after operation and persisted Twenty-five cc of fluid were aspırated 3 years after synovectomy In both cases the fluid findings were fundamentally the same preoperitively and postoperatively with the exception of the mucin content (scc Table XIV) In Case 1, the unoperated knce was involved throughout the period of study and served as a basis of comparison Three years after synovectomy the fluid from the operated knee was the same as that withdrawn simultaneously from the unoperated knee, with these exceptions the mucin nitrogen and glucosamine contents from the operated knce were only 001 and 0 006 gm per 100 cc, the relative viscosity was 51 and the mucın precipitated as a few shreds in a cloudy solution, whereas the fluid from the unoperated knee contained 0 09 gm of mucın nitrogen and 0 067 gm of mucın glucosamine and had a viscosity of 25, and the mucın precipitated well in a soft clump The patient had devel. oped amyloidosis with renal involvement 1 year after operation with subsequent reduction in the protein content of the serum There was a corresponding decrease in the protein concentration in the synovial fluids, but the albumin/globulin ratio remained in the same range as that of the serum and was identical in both fluids The fluid sugar, which was markedly reduced preopera tively, remained essentially the same after synovectomy This finding is in accord with the hypothesis that the reduction of sugar in rheumatoid joints is due more to decreased entrance than to increased utilization (see p 67) In Case 2 the concentrations of albumin, globulin and sugar remained unchanged after synovectomy, but as in the first case, the mucin nitrogen was reduced from 009 gm to just a trace The great decrease in mucın in both cases suggests that the tissue removed at operation may have been active in the formation of the synovial mucin and gives further evidence in favor of the ·8 100 116 SYNOVIAL ILUID CHANGES philic ground substance ® Such changes seem to indicate a more intense inflammatory activity and are particularly common in the early or rapidly progressing stages of the lesions 11B* The cytological variations found in this series of theumatoid fuids corre spond closely with those reported by Fischer" in 9 cases Keefer Myers and Holmessa in 16 cases McEwen" in 8 cases Collins* * in 31 cases Jordant in 5 cases and Burman and Klinge in 2 cases The total nucleated cells in their fluids ranged from 1 700 to 74 800 per cu mm percentages of polymorphonuclears from 3 to 100 kling 152 on the other hand reported variations from 400 to 230 000 cells with polymorphonuclears of 8 to 94 per cent in 34 cases The volume of fluid obtained by Collins and Kling varsed from 3 to 160 cc the viscosity from 3 0 to 32 In 2 fuds Ragan and Meyer*** found a variation in relative viscosity from 45 to more than 1 100 The extremely higă value was estimated by extrapolation using the figure obtained after diluting the fluid with 0 85 per cent saune Chemical studies were made by Fischer * Keefer Myers and Holmes 10 Collins Jordan Kung *** Ragan and Meyer" and Sundblad * The protein contents (pre sumably including muno) varied from 24 to 8 6 per cent the sugar from 17 to 145 mg per 100 cc uric acid from 22 to 47 mg per 100 cc phosphate from 31 to 43 mg per 100 cc calcium from 53 to 116 mg per 100 cc pH from 7 27 to 78 Fischeros reported 1 mucun of 0 41 per cent, 1 globulin of 3 95 per cent and a surprisingly low albumin value 0 32 per cent Ol hagen"reported remarkably high concentrations of gamma globulia in fuds from patients with rheumatoid arthritis of long duration Ragau and Meyerths in 41 fluids found variations in concentration of hyaluronic acid from 0 08 to 0 63 mg per cc and Sundblad in 37 fluids found an aver age of O 150 gm per 100 cc Burman and Klingt reported that Evans blue entered the involved joints in patients with rheumatoid arthnus and con cluded that the more acute the arthrits the greater the entrance of dye The findings in the groups studied separately by McEwent and Kling's also correspond with those in rheumatoid arthritis in general Their series of 5 and 24 cases designated as infectious arthritis presumably represent acute and somewhat atypical cases of rheumatoid arthritis The total cell counts vared from 1 500 to 80 000 per cu mm with 35 to 95 per cent polymor phonuclears The icteric index varied from 2 to 5 umts except in 1 flud which showed 12 Viscosity ranged from 28 to 24 0 and calcium from 81 to 97 mg per 100 cc Effect of Synovectomy on Synovial Fluid Findings in Rheumatoid Arthritis Synovial flud was obtained from the knee joints of 2 patients * The tinctorial properties of this substance with regard to metachromatic dyes as well as aniline blue and hematoxylın seem to correspond with those of synovia but as yet 20 connecbon between its interstitial accumulaton and changes in synovial bud has been found ·GROUP II, INFECTIOUS DISEASES 119 and After Synovectomy NG ratio Globulin gru per 100 cc Serum Fluid L R Serum Fluid Muçin pitrogen gm per 100 cc. fluid L R Mucha glucosamine gm per 100 cc L R Viscosity at 38 C R A 3 45 186 198 416 2 55 105 135 1 30 094 098 087 039 046 082 067 231 3 26 196 3 54 1 69 2 83 144 1 24 083 073 061 0 57 0 72 054 058 006 012 008 83 51 091 067 253 2 47 2.25 335 323 3 09 268 2 09 1 27 1 25 093 142 1 13 076 092 Trace 94 57 025 those in mild cases of rheumatoid arthritis The fluids were fairly clear and usually did not clot The intra articular pressure varied from 0 to 200 mm of fluid The total nucleated cell counts and percentage of polymorphonuclears were lower than in most rheumatoid fluids, ranging from 635 to 6 200 per cu mm and from 0 to 60 respectively The fluid cytological changes during and following a typical attack in the patient who had had intermittent effusions for 20 years are shown in Figure 8 In the other patient, there was an initial rise in percentage of polymorphonuclears followed by a rise in mononuclears and lymphocytes (Table III) The concentration of sugar in the fluids was essentially the same as that in the serum The total contents of albumin and globulin and the albumin/globulin ratios in the fluids from both patients were the same as the average for rheumatoid Auids The mucin concentration (0 050 gm nitrogen per 100 cc ) was slightly lower than the average in rheumatoid fluids but the mucın precipitate was usually a tight ropy clump with clear solution The relative viscosity tended to be low during the first day or two of the effu sion (410) and to increase later (up to 25) The pH value in 1 fluid was 766 as compared with 754 in the serum Krummel" and Collins also found low total cell counts in fluid from intermittent hydrarthrosis although Collins noted high percentages of poly morphonuclear cells krummel in the case of a patient who had bilateral knee effusions recurring at 10 and 12 day intervals found that polymorpho nuclears and mononuclears increased with increasing effusion lymphocytes ·118 SYNOVIAL FLUID CHANGES TABLE XIV Synovial Fluid Findings in Two Patients with Rheumatoid Arthritis Before Amount сс fluid L Leukocytes per cu mm Aurd L Polymorpbo nuclears per cent fluid R Sugar mg per 100 cc Serum Flud L R Protein gm per 100 cc Serum Flud L R Case no and date 93 18 20 50 50 66 000 40 000 41 000 92 89 100 22 13 100 0 706 437 455 805 504 83 8 Case I 11/27/35 1/18/37 8/13/37° 11/24/39 1/30/40 5/15/40 Case 2 1/ 2/36 6/15/377 5/28/40 90 110 70 15 10 7 8 450 17 050 6 100 18 650 4 300 7 550 66 31 63 68 81 78 5 98 3 40 572 2 66 4 87 2 21 195 81 10 5 70 30 25 21 700 29 800 10 050 67 90 62 99 94 100 64 67 763 5 12 754 6 24 743 570 Synovectomy right knee 1Synovectomy left knee theory that mucin is formed by the connective tissue cells around the joint 1" 79 131 On the basis of experimental work (see re view?") it may be assumed that the connective tissue lining of the joint is re formed but apparently the new cells in these 2 cases were not capable of producing mucin in large quantities INTERMITTENT HYDRARTHROSIS Idiopathic intermittent hydrarthrosis is characterized by peri odic joint effusions recurring at regular intervals with great precision The cause of this phenomenon and the factors controlling the amazing regularity of the recurrences are not known In rheumatoid arthritis effusions occasionally recur with similar regularity, and it seems possible that many if not all, patients with intermittent hydrarthrosis have aty pical rheumatoid arthritis We have seen only 2 cases in which we originally made a diagnosis of idiopathic intermittent hydrarthrosis In 1 of these however, the development of persistent synovial thickening of both knees, X-ray evidence of atrophy of the surrounding bones and an eleva tion of the sedimentation rate have made it apparent that the patient has rheumatoid arthritis, even the other patient, who had intermittent effusions for 20 years, finally developed an elevated sedimentation rate and synovial thickening of one knee that persisted between attacks The flud findings in these 2 cases were alike and resembled ·GROUP II, INFECTIOUS DISEASES 121 The findings in bursal fluids suggest that the cquilibrum between burs? and blood differs from that between joint and blood The low concentration of sugar in bursal effusions of short duration, in the absence of bacteria or marked leukocytosis, would be unusurl in synovial fluid The variation in the concentrations of albumin, globulin and mucin is greater than that found in comparable joint fluids In fluids from 3 so called ganglions, no precipitate was obtained with acetic acid until serum was added Apparently the polysacchande was present without adequate protein to allow precipitation of mucın ·120 SYNOVIAL FLUID CHANGES increased when the effusion was subsiding Dopter and Tanton" reported 50 per cent polymorphonuclears and 50 per cent lymphocytes in I case and 100 per cent lymphocytes in another case The viscosity in 1 fluid studied by Ragan" was 11 2 The so called albumin content of fluids in Krummels series determined refractometrically ranged from 5 29 to 65 per cent The pH of 1 fluid reported by Krummel was 75 and of I recorded by Coppo and Levi" was 78 Scapını* in 1 case reported a calcium content of 87 mg per 100 cc phosphorus of 3 8 mg per 100 cc, magnesium of 2 35 mg per 100 cc potassium of 21 2 mg per 100 cc and chloride of 672 mg per 100 cc BURSITIS AND BAKER S CYSTS Acute and chronic inflammation of bursae may be associated with various joint diseases such as rheumatoid arthritis, may be of known infectious orgin, or may be of unknown etiology and unassociated with involvement of other skeletal tissues The term Bahers cyst has been used to designate synovial lined cysts that are presumed to have been in communication with joints and sub sequently sealed off The present series includes 7 cases of acute bursitis of less than 6 weeks duration, 4 cases of chronic bursitis, 5 cases of so called Baker s cysts and 4 cases of walled off cysts in popliteal space or quadriceps pouch in patients with rheumatoid arthritis Fluids from all types of bursitis and Baker s cysts show wide variations in cytological and chemical characteristics As a result, fluid examination is not usually of value in determining the type en duration of the bursitis However, in our series high total leukocyte counts (10 000 to 50,000 per cu mm ) and a high percentage of polymorphonuclears (60 to 90) were found only in some cases of acute bursitis and in some of the cysts in patients with theumatoid arthritis In all other cases the leukocyte counts ranged from 500 to 1,000, with 0 30 per cent polymorphonuclears The majority of all of the fluids were turbid and slightly blood tinged and either did not clot or formed only small, friable clots The concentrations of sugar were usually low, ranging from 0 to 95 mg per 100 cc, with an average of 40 The total albumin and globulin contents were relatively high, ranging from 19 to 81 gm per 100 cc, with an average of 52 The mucın nitrogen concentration tended to be high averaging 0071 gm per 100 with a range from 0 to 0 20 In 1 case of chronic inflammation of a subacromial bursa, in which calcium was shown by X ray, the calcium content was 139 mg per 100 cc (see p 87) Or CC, ·DIAGNOSTIC VALUE OF ASPIRATIOV 123 a reduction of viscosity per unit concentration (Figures 12 and 13) It is apparent that analysis of joint fluid aids matcrally in distinguishing the two large groups of joint diseases While no single finding is absolutely conclusive, any of the conditions listed in Group I as fairly definitely ruled out by aspiration of a fluid which is turbid, clots, has a low viscosity, and shows a small amount of flocculent mucın precipitate, a total albumin and globulin concentration of more than 50 gm per 100 cc, a difference in fasting sugar content between serum and fluid of more than 30 mg per 100 cc, a total nucleated cell count of over 5,000 per cu mm and an absolute polymorphonuclear count of over 1,000 For differentiation between individual fluids within either of the two large groups, synovial fluid examination is of less value There are, however, certain findings that give suggestive evidence in favor of one or another diagnosis In disseminated lupus erythematosus, for instance, in contrast to rheumatoid arthritis, polymorphonuclears have never been above 260 per cu mm (Fig. ure 6), the concentration of mucın nitrogen and glucosamine has almost always been high (Figure 11), the mucın precipitate has always been good (a ropy clump in a clear solution) and the relative viscosity of the fluid has been high (Figure 12 and the chart at the back of the book) Similarly, in rheumatic fever the concentration of mucın is relatively high (Figure 11), the mucın precipitate usually good and never poor, and the viscosity relatively high (Figure 12) As a result, in a patient in whom the differential diagnosis is between rheumatic fever and rheumatoid arthritis, the finding of a low concentration of mucin with poor precipitate and low viscosity would give evidence in favor of a diagnosis of rheumatoid arthritis rather than theumatic fever The extremely high concentration of mucin (Figure 11) and high relative viscosity (Figure 12) usually found in fluids from osteochondromatosis aid in differentiation from traumatic arthritis Aspiration of hemorrhagic fluid especially if it contains free fat in the absence of severe acute trauma or hemophilia (and in the presence of a persistent, usually painless effusion), suggests the diagnosis of hemorrhagic villous synovitis There is one combination of findings that is of considerable diagnostic value namely, in a fuid of relatively short duration (weeks or ·VI Diagnostic Value of Joint Aspiration FROM a survey of the findings in the joint diseases studied, it is clear that the information gained from certain analyses is of value in diagnosis and prognosis The significant findings are shown in the chart at the back of the book They can be roughly sum marized as follows Fairly clear, viscous fluids that do not clot are found usually in traumatic arthritis, osteochondritis dissecans osteochondromatosis Charcot joints and degenerative joint dis ease, occasionally in tumors, in rheumatic fever and in mild atypical cases of rheumatoid arthritis The majority of rheumatoid fluids and all specific infectious fluids are turbid and less viscous than Group I fluids and often clot There are other important differences between the two main groups The total nucleated cell count is usually below 3,000 per cu mm, with absolute polymorphonuclears less than 500 in Group I fluids and above 3 000 with more than 500 polymorphonuclears in Group II fluids (Fig ures 5 and 6) The sugar content is approximately the same as that of the blood in Group I fluids, is slightly lowered in the majority of Group II fluids varies from no sugar to a normal value in rheumatoid fluids and is low or absent in infected fluids with positive cultures and in most tuberculous fluids (Figure 14) The total concentration of albumin and globulin is usually below 50 gm per with an albumin/globulin ratio above 20 in Group I fluids and above 40 gm per 100 cc, with an albumin/ globulin ratio below 20 in Group II fluids (Figure 9) The concentration of mucın is usually normal or elevated in Group I fluids, while in Group II the concentration is often lowered and there is associated loss of the characteristic precipitability and 100 cc, 122 ·125 DIAGNOSTIC VALUE OF ASPIRATION count showed 9 per cent polymorphonuclcars, 29 per cent lymphocytes and 62 per cent monocytes The mucin precipitated very well in a large ropy clump with clear solution Thc hemorrhagic ſuid with low Icukocyte count was consistent with a diagnosis of hemorrhagic vllous synovitis The patient did not return as requested but 10 months later he was admitted to the hospital because of persistence of the effusion in the left knce Examination was unchanged Hemorrhagic fluid was again obtained with 363 000 eryth rocytes per cu mm The leukocyte count was 3 750 with 60 per cent polymorphonuclears The mucin precipitated very well on the basis of the persistent painless effusion, found on aspiration to be bloody and to have a relatively low leukocyte count and good mucin, a diagnosis of hemorrhagic villous synovitis was made At operation the synonal tissues were greatly thickened and bronze in color The histological appearance was consistent with "pigmented villo-nadular synovitis Case 3 An 18 year old white boy was admitted with a complaint of pain and swelling in the left knee of 5 months duration A cyst of the internal semilunar cartilage of the left knee had been removed 21 months before admission The knee had remained absolutely symptom free until 5 months before admission when he was thrown against an anchor bar on a ship during a storm striking the left knee The knee had been swollen and painful since that time The past history was non contributory Examination showed effusion in the left knee with tenderness over the anterior aspect of the medial condyle the internal lateral ligament and the posterolateral aspect of the lateral memiscus The joint capsule was markedly thichened A drag nosis of traumatic synovitis was made and synovectomy and removal of the external semilunar cartilage were performed Frozen sections of the synovi alis were not made because the diagnosis was thought definite The fluid ob tained at operation had a total count of 5 250 per cu mm with 58 per cent polymorphonuclears a protein content of 5 38 gmn per 100 cc, All these findings were inconsistent with those of traumatic fluds and gave the first indication that the diagnosis was incorrect The combination of relatively low total leukocyte and polymorphonuclear counts with a low sugar concentration in an effusion of only a few months duration was strong evidence in favor of tuberculous arthritis Histological examination of the synovialis confirmed this diagnosis and the knee was subsequently fused In this case preoperative examination of the fluid would have suggested the correct diagnosis which could have been confirmed by examination of frozen sections of synovialis at the first operation and a second operation would have been avoided Case 4 A 23 year old woman was admitted with the complaint of swelling and limitation of motion of the left knee One year before admission she had squatted to pick up an object and her left knee had locked It had become swollen and stuff returning to normal slowly Four months before admission the swelling had reappeared and persisted A diagnosis of internal derangement of the knee was made and the internal semilunar cartilage was and no sugar ·124 SYNOVIAL FLUID CILANGES a few months), the combination of a relatively low leukocyte count (25,000 cells per cu mm or below) and polymorpho nuclears of 50 per cent or less with a low content of sugar (20 mg per 100 cc or below) is, we think, strongly suggestive of tuberculosis Although the findings in any individual fluid within either one of the groups are not diagnostic, when combined with the clinical and other laboratory data they may be of definite value in differential diagnosis As a few typical examples will show, early syno vial fluid study may aid in confirming a diagnosis suggested clinically (Cases 1 and 2 following), it may, by facilitating diag nosis, remove the need for operation (Cases 3 and 4) or direct treatment toward the underlying cause of the disease (Case 5) In other cases, clinical corroboration of a diagnosis indicated by the fluid analysis may not be obtained for some time (Case 6) Case 1 A 30 year old tractor driver was admitted with the complaint of pain in the left knee of 1 year s duration One year before admission he had fallen against a roll of paper cutting the left lower shin Very gradually pain had developed in the left knee Subsequently the knee had become swollen tender and hot Following aspiration of flud 5 months before ad mission the symptoms had subsided but the pain and swelling had recurred 1 month later Examination showed swelling of the left knee with pain on motion and slight tenderness over the anteromedial aspect Roentgenograms showed extensive calcification in the knee joint The diagnoses suggested clinically were osteochondromatosis or internal derangement of the knee joint The synovial furd was slightly cloudy did not clot and contained 1 050 leukocytes per cu mm with 1 per cent polymorphonuclears The fluid sugar was 93 mg per 100 cc compared with a serum sugar of 101 mg The relative viscosity was 147 the mucun 09 gm nitrogen per 100 cc the albumnın and globulin 3 0 gm per 100 cc and the albumin/globulin ratio 30 The mucin precipitated well in a ropy clump with clear solution The very high vis cosity and high mucin concentration in thus fluid differentiated it from those usually found in internal derangement of the knee The fluid was consistent with osteochondromatosis Subsequent synovectomy corroborated the diag nosis Case 2 A 19 year old boy was seen in the emergency ward because of swelling of the left knee of 3 years duration The onset of swelling had been insidious There had been no pain in the knee and no other joint symptoms Hemorrhagic fluid had been aspirated on two occasions Examination showed a moderate sized effusion in the left knee Roentgenograms showed no bony abnormalities On aspiration of the knee bloody flud containing 550 leuko cytes and 391 000 erythrocytes per cu mm was obtained The differential ·DIAGNOSTIC VALUE OF ASPIRATTOV 127 cartilages was performed. At operation the synovial tissues showed marked thickening and after histologic examination a diagnosis of chronic synovitis was made The synovial fluid aspirated preoperatively, had a total nucleated count of 17,800 per cu mm, with 96 per cent polymorphonuclears Our present knowledge of the significance of fluid findings would have told us that such changes do not result from trauma but this was not recognized at that timc When the patient was seen 8 years after operation she had malarse persistence of slight swelling and pain on motion in the nght knee pain in the night shoulder left hand and left wrist and a corrected sedimen tation rate of 102 mm per minute A diagnosis of rheumatoid arthritis was made Re-examination of the onginal biopsy specimen, by a pathologist versed in the study of synovial tissues showed changes interpreted as con sistent with rheumatoid arthritis ·126 SYNOVIAL FLUID CHANGES removed The synovial membrane was found to be thickened and there was definite pannus formation in the region of the intercondylar notch Fluid removed at the time of operation contained 25 600 leukocytes per cu mm, with 96 per cent polymorphonuclears and 41 mg of sugar per 100 cc – findings entirely inconsistent with those of traumatic flud Three months later the patients left ankle became swollen and during the following month the right hıp knee and ankle became involved If synovial fluid had been obtained on admission, it would have suggested the correct diagnosis of rheumatoid arthritis Case 5 A 45 year old man was admitted because of a cough present for 3 weeks that was productive of large amounts of white sputum Two months before admission he had had his teeth extracted for treatment of "arthritis which he had had for 12 years The joint symptoms had begun with stiffness of knees ankles wrists and right elbow and enlargement of the terminal phalangeal joints of the fingers There had been gradual progression of the involvement of knees and ankles with marked pain and swelling Examination on admission showed diminished breath sounds over the left upper chest and a roentgenogram disclosed an area of increased density in the left upper lobe There was marked clubbing of all fingers and toes The knees and ankles were swollen and very tender Synovial fluid aspi rated from the left knee was slightly cloudy and clotted firmly The leukocyte count was 100 per cu mm with 20 per cent polymorphonuclears and 80 per cent lymphocytes The flud protein content was 22 gm per 100 cc The mucin precipitated in small flecks in a slightly cloudy solution Further study of the pulmonary lesion suggested that it was a neoplasm and pneumonec tomy was performed A tumor involving most of the left upper lobe was found and the histological diagnosis was adenocarcinoma Grade II The immediate postoperative course was uncomplicated except for slowly increasing dysp nea relieved by thoracentesis However on the twentieth postoperative day respirations suddenly became irregular and ceased The combination of synovial flud findings-the low cell count with few polymorphonuclears and the slight elevation of protein combined with the relatively high fibrinogen content (as indicated by the large firn clot)- would be most unusual in theumatoid effusions of months duration They are very suggestive of pulmonary osteoarthropathy and could if obtained earlier have arded in recognition and treatment of the underlying disease Case 6 A woman of 49 was admitted with the complaint of pain and swelling of the right knee of 3 years duration The symptoms had begun immediately after injury to the knee by a fall from a bus and had persisted to the time of admission There had been frequent episodes of 'locking during the year before entry No other joint involvement or associated con stitutional symptoms were elicited Examination showed marked effusion in the knee with tenderness over the external tabiofemoral fossa and increased lateral mobility A diagnosis of internal derangement of the knee was made and synovectomy with removal of both external and internal semilunar ·129 PITYSIOLOGICAL CONSIDERATIONS diffusible, non utilizable substances in the joint, thc cntrance of glucose is slow The delay in appearance, present even in normal joints, is enhanced with incrcising inflammation In mild cases, as in traumatic arthritis, the rate of entrince apparently remains almost equnl to the rate of utilization or removal and cquilibrium is maintained with a flud concentration ncarly cqual to that of the blood In severe inflammation, as in severe rheumatoid arthritis or infectious arthritis, the entrance is decreased so that, with in creased consumption, the fluid sugar concentration is markedly lowered The mechanism of this differential entrance of proteins and sugar is not understood, but the results cmphasize that a variety of factors is involved in transfer across the synovial tissues Alterations in the synovial tissues lead also to diminished re moval of colloidal and particulate matter Experimental evidence suggests that this is due to a decreased number of synovial lymphatics (see reviewl.) The reduced absorption, combined with greater entrance, probably accounts for the increasing protein concentration and decreasing albumin/globulin ratio in rheumatoid effusions of long duration Whether or not local production of one or more protein fractions also plays a role is not known The abnormally high concentration of colloidal material raises the osmotic pressure and presumably leads to persistence of the effusion The altered metabolism of the joint in disease is apparent in the case of sugar and mucin and tends to increase with severity of inflammation Greater utilization of sugar results from various factors increased activity of fixed tissue cells, a larger number of leukocytes in the synovial fluid, and bacteria, if present Greater metabolic requirement for sugar, coupled with decreased supply, may lower the fluid sugar to a level which may impair nutrition of cartılage In the case of mucin both formation and destruction may be affected In traumatic effusions the unit concentration of mucin remains normal despite a great increase in flud volume Trauma seemingly stimulates formation of mucın by the connective tissue cells and does not lead to an increase in the rate of mucın destruction Infectious and many Theumatoid fluids on the other hand show a decreased concentration of mucin a reduced Viscosity and a soft or flocculent precipitate with acetic acid The similarity of these changes to those produced by enzymatic di- ·VII Physiological Considerations A SURVEY of the findings in effusions from various joint diseases corroborates the hypothesis that joint fluid reflects changes pro duced in synovial tissue by disease There are two fundamental alterations an abnormality in the exchange of substances between the blood or lymph and the joint and a disturbance of intra articular metabolism The degree of permeability of the blood joint barrier varies with the degree and type of synovial inflammation The difference is not apparent in the case of diffusible non utilizable substances whose distribution in joint disease follows the laws that apply to normal joints There is, of course increased entrance of such substances associated with the increase in amount of fluid entering the joint, but equilibrium is maintained The difference between types of joint disease becomes apparent in the variation in the increase in number of cells and in the permeability to bacteria and proteins including fibrinogen and presumably antı bodies and enzymes The minimal inflammatory reaction produced by trauma causes only slight increase in leuhocytes, albumın, globulin and fibrinogen and the ratio of albumın to globulin in the fluid remains high In rheumatoid and in infectious arthritis there is greater entrance of these substances and, with the excep tion of leukocytes in rheumatoid arthritis, a progressive rise with increasing severity of the disease With the increased entrance of protein the differential permeability decreases globulins enter the joint more readily and the albumin/globulin ratio of the fluid more closely approaches that of the serum In contrast to the rapid increase in number of cells and amount of proteins and 128 ·Bibliography I Abadie ) Examen cytologique du liquide articulaire de quelques arthropathies chroniques Compt. rend Soc de Biol 1902 945 2 Achard C and Loeper M Les globules blancs dans le rhumatisme Compt rend Soc de Biol 1900 1029 3 Achard C and Picttre, M Sur la mucine des liquides articulaires Compt rend Acad d Sc 1930 191 1412 4 Achard C and Piettre M Les protéines des épanchements articu laires Compt rend Acad d Sc 1931 192, 996 5 Adkins E W O and Davies D V Absorption from the joint cavity Quart J Exper Physiol 1940 30 147 6 Aldred Brown G R P and Munro J M H The plasma proteins and non protein nitrogen and the sedimentation rate in chronic rheumatic disorders Quart J Med 1935 4 269 7 Allison N Fremont Smith F Dailey M E and Kennard M A Comparative studies between synovial fluid and plasma J Bone and Joint Surg 1926 8 758 7a Asboe Hansen G The variability in the hyaluronic acid content of the dermal connective tissue under the influence of thyroid hor mone Acta Dermato Venereol 1950 30 221 8 Bachman C Collip J B and Selye H The effects of prolonged oestriol administration upon the sex skin of macaca mulatta Proc Roy Soc London s B 1935 117 16 9 Balboni VG Shapiro Q M and kydd D M The penetration of penicillin into joint fluid following intramuscular administration Am J Med Sc 1945 210 588 10 Barker SB and Summerson WH The colorimetric determination of lactic acid in biological material J Biol Chem 1941 138 535 11 Bauer W Bennett GA Marble A and Claflın D Observations on the normal synovial fluid of cattle I The cellular constituents and nitrogen content J Exper Med 1930 52 835 (12 Bauer W Ropes MW and Waine H The physiology of articular structures Physiol Rev 1940 20 272 13 Bauer W Short CL and Bennett G A The manner of removal of proteins from normal joints ) Exper Med 1933 57 419 14 Beattie M The cells of inflammatory exudations An experimental research as to their formation and destiny and also as to the origin of the mononucleated cells J Path and Bact 1903 8 129 131 ·SYNOVIAL FLUID CHANGES 130 gestion of mucın suggests that increased destruction of mucin takes place in these joints, though incomplete synthesis may explain, at least in part, the reduction in viscosity and altered precipitability of mucın Attempts to demonstrate the presence of hyaluronidase in normal or pathological fluids have not been successful The relation of synovial fluid findings to the degree of inflam mation in the synovialis can be used to differentiate non traumatic joint diseases from traumatic arthritis of any type Within the two large groups, further differentiation may be possible because of the specific combination of Auid abnormalities produced by some individual diseases In all cases in which both synovial fluid and biopsied synovial tissue were available, an attempt was made to correlate the degree of abnormality in the fluid with that in the synovialis Exact cor relation was not possible but in all diseases the grade of reaction found in the tissues was usually of the same order as that of the fluid In the 2 cases of degenerative joint disease, for instance, in which the synovial tissues showed the most marked inflammation, the total proteins, especially the globulin fractions, were highest and the mucin content lowest Synovial fluid thus presents itself as a reflection of tissue inflammation that may be of value in diagnosis and as a tool for further study of the physiology of articular tissues and the blood tissue barrier ·BIBLIOCRAPIIY 133 31 Cajori FA and Pemberton R The chemical composition of synovial fluid in cases of joint effusion J Biol Chcm 1928, 76 471 32 Catoggio P M and Morrison, L R The Golgi apparatus in the cells of synovial tissue and related structures J Clin Investiga tion 1949 28 773 33 Cattell J P Meningococcal meningitis with purulent arthntıs New England J Med 1943, 229 49 34 Chauffard A, Richet, C Fils et Grigaut A Dosage comparé de la cholesterine dans le Sérum et dans les oedémes Mém de la Soc de Biol 1911 70 317 35 Cheng . Observations on dye ercretion through synovial mem brane after lumbosacral sympathectomy and circulatory obstruc tron Quart J Exper Physiol 1949 35 135 36 Chesney A M Kemp J E and Baetzer F H An experimental study of the synovial fluid of patients with arthritis and syphilis I Can Investigation 1926-27 3 131 37 Coggeshall HC Bennett GA Warren C F and Bauer W Synovial fluid and synovial membrane abnormalities resulting from varying grades of systemic infection and edema Am J Med Sc 1941 202 486 38 Coggeshall HC Warren C F and Bauer W The cytology of normal human synovial Aud Anat Rec 1940 77 129 39 Collins D H The pathology of synovial effusions J Path and Bact 1936 42 113 39a Collins DH Pathology of Articular and Spinal Diseases The Williams and Wilkins Company Baltimore 1950 39b Collins DH Discussion on the significance of pathological tests in rheumatic disease Proc Roy Soc Med 1949 42 40 Coppo M and Len M Ricerche biochimiche e fisicochimiche sui liquidi di versamento articoları Giornale di Clin Med 1946 27 517 41 Corper H J and Uyer N Further observations with a new method for cultivating tubercle bacıllı a comparison with guinea pig inocu lation and Petroff's method J Lab and Clin Med 1929 14 393 42 Cullen G E and Van Slyke D D Determination of fibrın globulin and albumin nitrogen of blood plasma J Biol Chem 1920 41 587 42a Dahl H K Examination of pH in tuberculous pus Acta Orth Scandinay 1950 20 176 43 Davies D V The staining reactions of normal synovial membrane with special reference to the origin of synovial mucin J Anat. 1943 77 160 44 Davies D V Observations on the volume viscosity and nitrogen content of synovial flud with a note on the histological appear ance of the synovial membrane J Anat 1944 78 68 ·132 BIBLIOGRAPHY 15 Beck A and Lauber HZur Frage der aktuellen Reaktion von Korperflussigkeiten (Gelenkexsudaten Sekreten usw) Arch f klin Chir 1929 155 469 16 Benedict S R and Behre J A The analysis of whole blood III Determination and distribution of uric acid J Biol Chem 1931 92 161 16a Bennett G A The joints Chap 44 in Pathology Edited by W A D Anderson The C V Mosby Company St Louis 1948 p 1330 17 Berger L Synovial sarcomas in serous bursae and tendon sheaths Am J Cancer 1938 34 501 18 Blair ſ E and Hallman F A Diagnosis of surgical tuberculosis Comparison of diagnosis by inoculation of gunea pigs and by culture Arch Surg 1933 27 178 19 Blair J E and Hallman F A Rheumatoid (atrophic) arthritis Bacteriologic cultures of synovial fluid and of tissues Arch Int Med 1934, 53 87 20 Bloor W R The determination of cholesterol in blood J Brol Chem 1916 24 227 21 Bondi S and Jacoby M Uber der Verteilung der Salıcylsaure bei normalen und infigierten Tieren Beitr z chem Phys u Path 1906 7 514 22 Bonnin J G and Prior A P The passage of penicillin through synovial membrane Penicillin therapy and control in 21 Army Group Printing and Stationery Service British Army of the Rhine 1945 p 335 23 Boots RH and Cullen GE The hydrogen son concentration of joint exudates in rheumatic fever and other forms of arthritis J Exper Med 1922 36 405 24 Boots RH and Swift HF The arthritis of serum sickness J A MA 1923 80 12 25 Borrel A Tuberculose pulmonaire experimentale Étude anatomo pathologique du processus obtenu par injection veineuse Ann de 1 Institute Pasteur 1893 7, 593 26 Burman M S Synovial fluorescence in a case of gout with jaundice Acta Rheumatol 1937 9 No 35 10 27 Burman MS and Kling D H The excretion of the dye Evans blue into the synovial effusion of rheumatoid arthritis Acta Rheu matol 1939 11 No 40 1 28 Butler AM and Montgomery H The solubility of the plasma proteins I Dependence on salt and plasma concentrations in con centrated solutions of potassium phosphate J Biol Chem 1932 99 173 29 Bywaters E G L The metabolism of joint tissue J Path and Bact 1937 44 247 30 Cajori FA Crouter C Y and Pemberton R The physiology of synovial fluid Arch Int Med 1926 37 92 ·DIDLIOCharity 135 65 Ferrari W Sul comportimento viscosimctrico delle soluzioni di mucini del funicolo omixclıcale e di acido faluronico Arch di Scienze Biologiche 1010 31, 20-1 66 Fischer A Beitrage zur Kenntnis der Gelenkergüssc bei der In feltarthritis Alin Wahnschr 1931 10, 1385 67 Fisher AGT Chronic (Non Tuberculous) Arthritis The Macmil lan Company New York 1929 68 Fiske C. H and Logan, M A Determination of Calcium in rolin s Laboratory Vinual of Biological Chemistry D Appicton Century Company Nay Jork 1931 5th ed. p 319 69 Fiske C II and Subbarow The colorimctric dctcrmination of phosphorus ) Bro! Chcm 1925 60 375 70 Foley, G E Submerged growth of tubercle bacilli from pathological material in Dubos mcdium Proc Soc Exper Biol and Med 1946 62 298 71 Folin O Two revised copper methods for blood sugar dctermina tion J Biol Chem 1929, 82 83 72. Folin 0 The preparation of sodium tungstate free from molybdate, together with a simplificd process for the preparation of a correct uric acid reagent (and some comments) J Biol Cher 1934 106 311 73 Folin O and Wu 11 A system of blood analysis J Biol Chem 1919 38,81 74 Forkner C E The synovial fluid in health and disease with special reference to arthritis J Lab and Clin Med 1930 15 1187 75 Foshay L and Mayer"O B Viability of Bacterium tularense in human tissues JAMA 1938 106 214 76 Foster, F N and Colquhoun J Penicillin content of body fluids in penıcıllınısed patients Penıcıllın therapy and control in 21 Army Group Printing and Stationery Service British Army of the Rhine 1945 p 339 77 Fremont Smith F and Daley M F Studies in the distribution of chloride and protein between plasma and synovial fluid J Biol Chem 1926 52 779 78 Frohlich A and Singer F Zur Frage der Speicherung von Salszy] saure in erkrankten Gelenken Arch f exper Path 1923 29 185 79 Gardner E Physiology of movable joints Physiol Rev 1950 30 127 80 Gunter GS The determination of Spinnbarkeit" of synovial fluid and its destruction by enzymic action Australian J Exper Biol & MSc 1949 27 265 81 Habler C Untersuchungen zur Molekularpathologie der Gelenkexsu date und ihre klinischen Ergebnisse Arch f klin Chir 1929 156 20 82 Halliburton WD Clinical Investigations In Report of Committee on Myxedema Clin Society s Transactions 1888 21 Suppl 48 ·134 BIBLIOGRAPHY 45 Davies D V Anatomy and physiology of diarthrodial joints Ann Rheum Dis 1945, 5 29 46 Davies D V The cell content of synovial fluid J Anat 1945 79 66 47 Davies D V Synoval membrane and synovial fluid of joints Lan cet 1946 2 815 48 Davis J S Protein studies in atrophic (rheumatoid) and hyper trophic arthritis J Lab and Clin Med 1936 21 478 49 Davison R Woolley, M and Donovan E The relationship between plasma proteins and erythrocyte sedimentation rates in chronic atrophic arthritis J Lab and Clin Med 1940 25 935 50 De Gara P F Studies on the bactericidal properties of the synovial fluid J Clin Investigation 1943, 22 131 51 Demel R Ein tierexperimenteller Beitrag zur Gelenksınfektion Arch f klin Chir 1927, 149 213 52 Dienes L The isolation of L type cultures from Bacteroides with the aid of penicillin and their reversion into the usual bacıllı J Bact 1948 56 445 53 Dole V P and Rothbard S Electrophoretic changes in the serurn of a patient with rheumatoid arthritis J Clin Investigation 1947 26 87 54 Dopter and Tanton Note sur lexamen cytologique des epanche ments de diverses séreuses Soc Med Hôp 1901 18 838 55 Drechsel E Hermann Handbuch der Physiologie 1883 5 Abt 1 617 56 Dubos R J and Middlebrook G Media for tubercle bacıllı Am Rev Tbc 1947 56 334 57 Dufour H Cytologie du liquide séro citrin d une arthropathie tabe trque Soc Med Hop 1901 18 1038 58 Durham H E The mechanism of reaction to peritoneal infection J Path and Bact 1897 4 338 59 Edlund T Studies on absorption of colloids and fluid from rabbit knee joints Acta Physiol Scandinay 1949 18 Suppl 62 I 60 Edlund T and Linderholm H Studies on the effect of salyrgan on the absorption of water and colloid from joints Acta Physiol Scandinav 1947 14 248 61 Efskind L Clinical studies on the function of the synovial mem brane of the knee joint Acta Chir Scandinav 1943 88 37 62 Eisenman A J A note on the Van Slyke method for the determina tion of chlorides in blood and tissue J Biol Chem 1929 82 411 62a Ekholm R and Norback B On the relationship between articular changes and function Acta Orth Scandinav 1951 21 81 63 Engel D The permeability of the synovial membrane Quart J Med 1940 30 231 64 Engel D The influence of the sympathetic nervous system on capıl lary permeability J Physiol 1941 99 161 ·33% OX RAVITY 137 101 Josepli, N R, Recl, C I and Jlomlagrs, 1 An tri nivo study of tho pll of synovini full in love sin } l'hyalol 1940, 140, 1 102 Josephi, N II, Road, C 1, Stock, I , 1 olk, I and haplan, 1 an Nectrochu nlcnl xlily of the synovfalls in dlaga Air J I'lyxtal 1048, 169, 001 103 Julliard, la lutilisation cliniquo do 11 cytoloyles, lit cryoscoplo et llimatoloylo luns leg (pancleine nts do quelque a or(1191 chirurgicale, Ilov (In Chir 1002, 25, 100 101 kufer, CS, Myers, WI and Holmre, W 1,Jr Cluaracteristics of the synovial full in varlout types of arthrithe Arch Int Mod 1031, 64 872 103 koler, CS and Mant, I. A Sulfanilamko in trentment of gono coccal arthritie Ain J Mul Sc 1939, 107, 108 100 heller, 11 Diagnostio nels in die aars of the front and joint. New York) Mul 1014, 100, 158 107 Kcy. J A Cytology of the synovial flull of normal ſolnts Annt lloc 1028, 10, 103 108 key, J A Stics Synovi il Membrano of Joints and Jurano, in Special Cytology, I dited liy 1 V Cowelry laul I louber, Inc, Now York, 1028, Vol II, 705 1081 key. J A Ixperimental arthritis 1910 runctions of folnts to alle frilianta J Bono nnil Joint Surg 1929), 11, 705 100 I ry, J A 1 xperimental arthritis 1'bias chanyre in lointa produced loy creating defects in articulis cartilayo J Hana and Joint Sury 1031, 13, 720 110 ) Iny, I SI The Colyl apparatus of synovinl calls under normal and pathological conditions and with referenus to the formation of synovial fuld J Pathi and Bact 1030, 11, 117 111 Bing, D Il Me Wascrinann roictlon in jolnt fluks Am J Syplı 1020, 13, 600 312 y, D JI Synovial cells in folnt esfulon J Bone and Joint Sury 1030, 12, 807 110 Yliny D II ito nature and orlyin of xynovial fluid Arch Sury 1031, 23, 643 114 yilny D Il Syphilito arthritis with effusion Am J Med Sc 1032, 113,678 11% Yliny D Il tho Synovial Membranes and the Synovial I Juld (with *puchal reference to Arthritis and Injuries of the Joints) Medical I'ra, 104 Anyeles, 1038 110 Xiny Dll and l'incula, l'urther studies on the Watermann reac. tfon in faint fluids Ain | Sypl 1031, 19, 270 117 Yrurmel, 1 /ur /ytoloyla de Celrnketylisco bei Intermittlerender Celenkwasseraucht Wkn klin Wohnsir 1931, 14, 1404 118 Iulie J 6 and Weatherford, I L Ilole of those srdículoendothelial synts in in this de position of collold :) and particulatos matter in artimular ( ivitiis Arch Sury 1930, 33, 08 ·136 BIBLIOGRAPHY 83 Hardesty M The structural basis for the response of the comb of the brown leghorn fowl to the sex hormones Am ) Anat 1931 47, 277 84 Henderson V E Joint affections in tabes dorsalis J Path and Bact 1905, 10 211 85 Herrell W E, Nichols D R and Heilman DH Penicillin its usefulness limitations diffusion and detection with analysis of 150 cases in which it was employed JAMA 1944 125 2003 85a Hirsch, C Some views on the pathology of synovial flud Acta Orth Scandinav 1950, 20 121 86 Hush, H L Feffer, H L and O Neil CB A study of the diffusion across the serous membranes of joint cavites | Lab and Clin Med 1946 31 535 87 Hoffheinz, S Experimentelle Studie uber den Einfluzz der Nerven funktionsstorung auf die Gelenlentzundung und Gelenkresorption Arch f klin Chir 1931, 164, 750 88 Holmgren B S Flussiges Fett im Kniegelenk nach Trauma Acta Radiologica 1942 23 131 89 Hoppe Seyler F Uber die Zusammensetzung von Flussigkeiten welche aus den Huftgelenken in Arthritis deformans enleert wur den Virch Arch 1872 55 253 90 Horiye, k Über die menschliche Synovia Verch. Arch f path Anat 1924 251 649 91 Howe P E The determination of proteins in blood- micro method | Biol Chem 1921, 49 109 92 Hubbard R S and Potter, R C The glycolytic enzymes of synovial fluid ) Lab and Clin Med 1943 28 1328 93 Hueter C Zur Histologie der Gelenkflachen und Gelenkkapseln mit einem kritischen Vorwort uber die Versilberungsmethode Virch Arch ( path Anat 1866 36 25 94 Hughes K E A Penicillin in synovial and peritoneal Aud Penıcıl In therapy and control in 21 Army Group Printing and Stationery Service, British Army of the Rhine, 1945 p 333 95 Jaffe, H L Lichtenstein L and Sutro C J Pigmented vallonodular synovitis bursitis and tenosynonitis Arch Path 1941, 31 733 96 Jeanloz R Structure de la chutine et de lacide hyaluronique et oxydation des sucres aminés par lion penodate Experientia 1950 6 52 97 Jones E S Joint lubrication Lancet 1934 1 1426 98 Jones G B The behavior of perucillin in synovial cavities J Bone and Joint Surg 1948 30B, 106 99 Jones H T Cystic bursal hygromas | Bone and Joint Surg 1930 12 45 100 Jordan E P Synovial membrane and Aud in rheumatoid arthritis Arch Path 1938 26 274 ·BIBLIOGRAPHY 139 133 Miller, C P and Bohnhoff M Two streptomycin resistant vanants of meningococcus J Bact. 1947 54 467 134 Moore N S and Van Slyke D D The relationships betiseen plasma specific gravity, plasma protein content and edema in nephritis J Clin Investigation 1929-30 8 337 135 Müller w Uber den negativen Luftdruck im Gelenleraum Deutsche Ztschr f Chir 1929 218, 395 136 Murray MR, Stout, A P and Pogogeff I A Synovial sarcoma and normal synovial tissue cultivated in vitro Ann Surg 1944 120 843 137 Myers W. Keefer C S and Holmes W F, Jr The character istics of synovial flud in gonococcal arthritis J Clin Investigation 1934 13 767 138 Ogston AG Phalpot J St L and Zuckerman S Observations related to the swelling of serual skin in Rhesus monkeys J En docrinol 1939 1 231 139 Ogston AG and Stanzer J E On the state of hyaluronic acid in synovial fluid Biochemſ 1950 46 364 140 Olhagen B Preliminary report of investigation of the correlation between serum globulins and sedimentation rate Discussion Meeting of the Swedish Soc Intern Medic May 1944 141 Olhagen B Electrophoretic analysis of plasma proteins in rheumatic diseases especially rheumatoid arthritis Seventh Internat. Con gress on Rheumatic Diseases 1949 141a Olhagen B The protein pattern of joint exudates Acta Orth Scan dunay 1950 20 114 142 Ord W M Myxoedema a term proposed to be applied to an essen tral condition in the "cretinoid" affection occasionally observed in middle-aged women Medico-Chirur Trans 1878 61, 57 143 Ory EM Meads M Brown B Wilson C and Finland, M Penicillin levels in serum and in some body fluids during systemic and local therapy J Lab and Clin Med 1945 30 809 144 Palmer J W Smyth E M and Meyer K On glycoproteins IV The estimation of hexosamine J Biol Chem 1937 119 491 145 Panizza B Intorno al Significato Funzionale della Synovia Arch. d Fisiol 1931 29 576 146 Paracelsus quoted by Hyrth, J Onomatologia Anatomica Wien 1880 147 Paul W D and Routh J I Electrophoretic analyses of plasma and serum proteins in theumatoid arthritis and rheumatoid spondylitis Fed Proc 1950 9 No 1 pt. 1 148 Perlmann GE and Kaufman D Electrophoretic distribution of proteins in serum plasma and synovial fluids of patients with rheumatoid arthrits J Clin Investigation 1946 25 931 149 Pescatori F La fisiopatologia del cavo articolare in rapporte al com ponente sinoviale La Chirurgia degli organi di movimento 1930 14 451 ·138 BIBLIOGRAPHY 118a Kulka J P Unpublished data 118b Kulonen E Studies on hyaluronic acid and hyaluronidase in the animal organism Acta Physiol Scandinav 1951 24, Suppl 88 119 Labor M and von Balogh E Zytologische und Serologische Unter suchungen der Synovia im Besondern bet Akuten Gelenksentzun dungen Wien klin Wchnschr 1919, 32, 535 120 Lasch CH Weitere Untersuchungen uber die Wasserstoffionen konzıntration in Gelenkergussen Arch f klin Chir 1928 150 506 121 Levene, P A and Meyer G M The action of leukocytes on glucose J Biol Chem 1912 11 361 122 Loeb L Suntzeff V and Burns EL Changes in the nature of the stroma in vagina cervix and uterus of the mouse produced by long continued injections of estrogen and by advancing age Am J Cancer 1939 35 159 122a Longsworth L G A modification of the Schlieren method for use in electrophoretic analysis į Am Chem Soc 1939 61 529 123 Lovgren O Studien uber den Intermediaren Stoffwechsel bei Chron ischer Polyarthritis Acta Med Scandinav 1945 Suppl 163 124 Lowdon A GR Case illustrating passage of parenteral penicillin into joint Penıcıllın therapy and control in 21 Army Group Print ing and Stationery Service British Army of the Rhine 1945 p 337 125 Lutwak Mann C Enzyme systems in articular cartilage Biochem J 1940 34 517 126 Malméjac M Sur le liqude synovial Bull de Chimie Biol 1922 4 190 127 Malmros H and Blix, G The plasma proteins in cases with high erythrocyte sedimentation rate Acta Med Scandinav 1946 Supp! 152 294 127a Maunsell K, Wngley F Highton T C and Holt L B The syno vial membrane Some observations in normal and arthritic human joints Ann Rheum Dis 1950 9 363 128 McEwen C Cytologic studies on rheumatic fever II Cells of rheu matic exudates J Clin Investigation, 1935 14 190 129 McEwen C and Thomas E W Syphilitic joint disease Med Clin North America 1938 22 No 5 1275 130 Menkın V and Warner CR Studies on inflammation XIII Car bohydrate metabolism local acidosis and the cytological picture in inflammation Am J Path 1937 13 25 131 Meyer K The biological significance of hyaluronic acid and hyaluro nidase Physiol Rev 1947 27 335 132 Meyer | Smyth E M and Dawson MH The isolation of a mucopolysaccharide from synovial fluid J Biol Chem 1939 128 319 ·BIBLIOGRAPHIY 141 169 Ropes, M W Perlmann, G E et al Unpublished data 170 Ropes MW, Robertson W v B, Rossmeisl E C Peabody, RB, and Bauer, W Synovial fluid mucin Acta Med Scandinav 1947 Supp! 196 700 171 Ropes, M W Rossmeisl, E C and Bauer, W The origin and nature of normal human synovial fluid J Clin Investigation 1940 19, 795 172 Rostock, P Der Flussigkeitsdruck in Kniegelenkergussen Deutsche Ztschr [ Chur 1928–29, 213, 314 173 Rourke MD On the determination of the sodium content of small amounts of serum or heparınızed plasma by the iodometric method J Biol Chem 1928, 78 337 174 Salkowski E Zur henntnis der Synovia inbesondere des mucinahn lichen Korpers derselben Virch Arch f Path Anat 1893, 131 304 175 Scapini A Ricerche sulla composizione chimica del liquido sinovale contenuto di electroliti Policlinico Sez. Med 1941, 48 277 176 Schneider J Untersuchungen uber die Viskositat menschlicher Syn ovia Biochem Ztschr 1925 160 325 177 Schurch o Viollier, G and Sullmann H Elektrophoretische Un tersuchung von Kniegelenkergussen Schweiz med Wchnschr 1950 80, 711 178 Scott R W Thoburn T W and Hanzlık, P J The salıcylates IV Salicylates in the blood and joint fluid of individuals receiving full therapeutic doses of the drug J Pharmacol and Exper Thera peutics, 1916-17, 9 217 179 Scull CW, Bach T F and Pemberton R Serum proteins in rheumatoid disease Ann Int Med 1939 12 1463 180 Seeliger, P Ein Beitrag zur pathologischen Physiologie der Gelenke unter Berucksichtigung der Gelenkmausbildung Arch f klin Chir 1926 142 606 181 Seifter J Baeder D H and Begamy A J Influence of hyaluroni dase and steroids on permeability of synovial membrane Proc Soc Exper Biol and Med 1949 72 277 182 Seifter J Baeder D H Begamy A J Rosenkranz G Djerassi C Pateke J and Kaufmann S Method of screening steroids on synovial membrane for possible antiarthritis action Fed Proc 1950 9 314 183 Shaffer M F and Bennett G A The passage of Type III rabbit virulent pneumococci from the vascular system into joints and certain other body cavities J Exper Med 1939 70 293 184 Shands A R Jr Synovial fluid in infectious and neuropathic ar thritis Southern Med J 1930 23 818 185 Shinkawa T Experimental and clinical studies on absorption from diseased joints with uranın solution Nagoya J Med Sc 1939 13 107 ·140 BIBLIOGRAPHY 150 Pewny w Zytologische Untersuchungen tuberkuloser Gelenker gusse Wien klin Wohnschr 1921 34 22 151 Podkamınsky N A Rôle des fernients du liqude synovial Compt rend d Soc de Biol Paris 1931 106 915 152 Poehlmann A Ober Gelenksyphilis und serologische Untersuchung en an Gelenkpunktaten Deutsche Ztsche f Chur 1923 182 161 153 Primer on the Rheumatic Diseases JAMA 1949 139 1068 1139 1268 154 Quintin, T J and White A RV Reiters disease Canad Aled Assn ) 1948 58 191 155 Ragan C Viscosity of normal human synovial Aud Proc Soc Exper Biol and Med 1946 63 572 156 Ragan C and DeLamater A Hydrolysis of hyaluronic acid of human joint fusd in vivo Proc Soc Exper Biol and Med 1942 50 349 157 Ragan C Doulan C P Coss J A jt and Grubin A F Effects of x ray irradiation on viscosity of synovial lud Proc Soc Exper Biol and Med 1947 66 170 158 Ragan C and Meyer K The hyaluronic acid of synovial Aud in rheumatoid arthritis J Clin Investigation 1949 28 56 159 Ranke HR Messungen des intraarticularen Druckes am knege lenk des Lebenden (Vorlaufige Mittheilung) Centralbl f Chir 1875 2 609 160 Reed C I Joffe H and Joseph N R Autonomic control of synch vial fluid reaction | Bone and Joint Surg 1947 29 370 161 Reitzel R J and Koh C The identification of gonococci a compli cations of gonorrhea JAMA 1938 110 1095 162 Reschke K Zur Diagnose der Gelenksyphilis (Wassermann sche Reaktion aus dem Punktat syphilitischer Gelenke) Arch f klo Chur 3919 111 597 163 Risak E and Winkler W Zur Klinik und Pathologie Deutsche Arch i klin Med 1929 165 129 164 Robertson W v B Ropes MW and Bauer W Muciase a bac terial enzyme which hydrolyzes synovial duid mucin and other mucins J Biol Chem 1940 133 261 165 Robertson W v B Ropes M W and Bauer V The degradation of mucins and polysaccharides by ascorbic acid and hydrogen peroxide Biochem | 1941 35 903 166 Robinson GL A study of liver function and plasma volume in chronic rheumatism by means of phenol tetrabrom phthalein sodium sulphonate Ann Rheum Dis 1943 3 207 167 Ropes M W Bennett G A and Bauer V The origin and nature of normal synovial fluid 1 Clin Investigation 1939 18 351 168 Ropes A W Kaufmas D and Perlmann G E Variations in the electrophoretic pattern of synovial Aud in articular disease J Clin Investigaton 1949 28 807 ·BIBLIOGRAPHY 143 202. Volkmann, quoted by Braun H Untersuchungen uber den Bau der Synovialmembranen und Gelenkknorpel sowie über die Resorption flüssiger und fester Lorper aus dem Gelenkhohlen Deutsche Ztschr f Chir 1894 39 35 203 von Seemen Versuche mut Vitalfärbung an Gelenken Centralbl f Chir 1926 53 2603 204. Wallis A D The serum proteins in theumatoid arthritis Ann. Int. Med 1950 32 63 205 Warren, CF Bennett, G A. and Bauer, W The significance of the cellular variations occung in normal synovial flud. Am J Path. 1935 11 953 205a. Watson EM and Pearce R H The mucopolysaccharide content of the skin in localized (pretibual) myxedema II Am J Clin Path 1949 19 442 206 Waugh, WG Treatment of certain joint lesions by injection of lactic aad Lancet, 1938 1 487 207 White AG and Sacks B A. Studies in edema Cholesterol and its relation to protein nitrogen in edema fud. Science 1950 112 18 209 Widal and Ravaut. Recherches histologiques sur le liquide des pleuresies experimentales Compt rend. Soc. de Biol 1900 52 1118 209 Zeller J V Bywaters E G L and Bauer V The passage of thiocyanate and glucose from the blood stream into the joint spaces. Am J Physiol 1941 132 150 210 Zeller J W Dienes L. and Bauer W Unpublished data 211 Zia, S H The occurrence of dysentery and typhoid antibodies in joint and pleural exudates Nat. Med. J China 1931 17 302 212 $ and Smyly H J Arthritis in association with bacillary dysentery Nat. Med. J China 1931 17 307 213 Unpublished data from this laboratory ·142 BIBLIOGRAPHY 188 Sluntaxa T Experiental and clinical studies on absorpuon from distased jouats with uranın solution. Nagoya | Med. Se 1939 13 107 187 Summonds F A. Jenkins TP N and Mackenze D Study of the bacteriortatic power of synovial fud during parenteral Donicillin treatment. Penicillin therapy and control in 21 Army Group Printing and Stationery Service Brush Army of the Rhine 1045 P 345 188 Soget G Erfalvungen aus dry letzten Dyscalerleepidemie Ajunch med Wcknschr 1915 62 183 169 Skans B and Sundblad L Oxdatse break down of fıyaluronc and chirmdroitin sulphuric acid. Acta PłrygioScandinav 1943 6 37 190 Spink, w w and Keeler CS The diagnosis treatment and end results is gonococal arthritis New England J Med. 1938 218 433 191. Sprunt, D IL McDearman, S and Raper Studies on the relation Elip of the smy hormones to infection 1 The effect of the estro gonuc and gonadotropic hormones on vacciula and the spreading factox Ezper Aled. 1935 67 153 192. Steck, J. E. Joseph N R and Peed C L The pll of the synovial fud an thr anaesthetized doz under trealment with metrazol or moulin. J Bone and Jout Surz 1948 30 A, 500 183 Stoddard I L and Drury P E A utration method for blood fat. J Biol Chesh. 1929 84 741 194 Storm Mallursen A Cutaamous tests for Reiter s disease A pre- Lumuany report Acta Dermato Venereologica 1946 26 347 1942 Sundulad L Thir chemistry of synovial fled with special regard to Eyaluronus aud Acta Orth Scandinav 1950 20 203 193 Svartz, N Nigra salttageises sürande sulfonamidpreparateus verkan. Nord. Med. 1943 18 673 195 Svartza N Modern Hemoterapi sad tarmoch ledalieltioner Nord Med 1944 23 1713 197 Suedia B and Bengtsson C The plasma protein fractions during ther course of disease in cases of polyarthritis rhieumatica Acta Med, Scandinav 1944 119 426 198 Suertopple IL A Sapuc arthrius of the bee font. Med. ) Australia 1933 26 542 193 Swat, W E Frozeo saction dando as of Luberculous jouts J Bone and Joint Surg. 1936 18 641 200 Tanu S Veber der Resorpuos und Ausscheidung der Farbstoffe dusib om Celenkitele Japanese Gastroenterol. 1935 7 No 4 630 2009. Tislus A A 0-W apparatus for electrophoretic analysis of collodal mistures Tr Faraday Soc 1937 33 524. 20L Vautel, E. The form and function of synovial cells za tissue cultures IL The production of mucin. J Exper Med. 1933 58 85 ·BIBLIOGRAPITY 143 202 Volkmann quoted by Braun, H Untersuchungen über den Bau der Synovialmembranen und Gelenkknorpel, sowic uber dic Nesorp tion Russiger und fester Korper aus dem Gelenlhohlen Deutscho Ztschr f Chir 1894, 39, 35 203 von Secmen Versuche mit Vitalfarbung an Gelenken Centralbl f Chir 1926 53, 2603 204 Wallis A D The serum proteins in rheumatoid arthritis Ann Int Med 1950 32 63 205 Warren, C F Bennett, G A and Bauer, W The significance of the cellular variations occurring in normal synovial ſuid Am J Path 1935 11, 953 205a Watson E M and Pearce R H The mucopolysaccharide content of the skin in localized (pretibial) myxcdema II Am J Clin Path 1949, 19 442 206 Waugh WG Treatment of certain joint lesions by injection of lactic acid Lancet, 1938, 1, 487 207 White AG and Sacks B A Studies in edema Cholesterol and its relation to protein nitrogen in edema fluid Science 1950 112 18 208 Widal and Ravaut Recherches histologiques sur le liquide des pleurésies expérimentales Compt rend Soc de Biol 1900, 52 1118 209 Zeller J V Bywaters E GL and Bauer W The passage of thiocyanate and glucose from the blood stream into the joint spaces Am J Physiol 1941 132 150 210 Zeller J W Dienes L and Bauer V Unpublished data 211 Zia S H The occurrence of dysentery and typhoid antibodies in joint and pleural exudates Nat Med J China 1931 17 302 212 Za S H and Smyly H J Arthritis in association with bacıllary dysentery Nat Med J China 1931 17 307 213 Unpublished data from thus laboratory

·No text on this page.

·Index {=figure fn=footnote p=photograph t=table Adrenocorticotrophic hormonc effect on synovial fluid findings 32, 57 63- 64 70 115 Alkaline phosphatase relation to mucin degradation 58 fn 59 77 t Ascitic fluid distribution ratios com pared to synovial fluid 4 5+ Ascorbic acid relation to mucin degra dation 5 58 f Atrophic arthritis See Rheumatoid ar thritis Distribution ratios between serum and synovial fluid 73 + Duration of disease relation to serum findings in theuma toid arthritis 42-43 t. 50 relation to synovial fluid findings 39 42-43 44 Duration of effusion relation to synovial fuid cytology 25-26 t 32 t, 33 34 t, 79 glucose 32 t 67 mucin 33 t 79 101 proteins 32-33 t, 39 44 46-47 52 79 uric acid in gout, 65 t Dyes entrance into joints 8 Dysentery bacıllary arthntis associated with fluid findings 23 105-106 Bacteria entrance into joint, 8 Bacteroides arthritis Aud findings 106 "Bakers cysts" fluid findings 120 121 Bursae fluid findings 7 f 87 fn 120 121 Cartilage relation to mucın formation 4 56 Charcot joints See Neuroarthropathy Chronic infectious arthntis See Rheu matoid arthritis Connective tissue relation to mucin formation 1 4 56 Cortisone effect on absorption from joint, 9 effect on synovial fluid findings 32 57 63-64 70 115 Edema distnbution rabios compared with syn ovial fluid 4 5 t synovial flud 10 Electrophoretic patterns of proteins in joint disease 46 50 114 Enzymes degradation of mucın by 5 6 58 59 129 130 in synovial fluid 9 10 77 Erythema nodosum flud findings 95 Exchange of substances between blood and joint See Synovial tissue per meability Degenerative joint disease 85 87 synovial fluid cytology 28 29 t 30-31 t 86 87 glucose 68 69 mucın and viscosity 54 55 56 60 61 t 63, 86 87 proteins 40-41 t 48-49 t, 86 87 synovial tissue histological changes 87 Desotycorticosterone acetate effect on absorption from joint 9 Diagnosis value of synovial fluid findings in 38 52 64 72 122 127 German measles arthntis associated with fluid findings 106 Glucosamine See Synovial fluid mucin Gonococcal arthritis 102 104 synovial flud complement fixation 103 104 cultures 21 22 27 t 104 cytology 24 25 t 26 27 t, 28 29 t, 30-31 t 36 102 103 145 ·146 INDEX Hypertrophic pulmonary osteoarthropathy synovial fud (cont) muçin and viscosity 54 55 & 60- 81 t 63 89 90 126 proteins 40-41 t, 48-49 50 89 128 synovial tissue histological changes 90 Gonococcal arthritis synovial ffud (cont) glucose 27 , 67 68-69 + 102 104 mucia and viscosity 27 + 54 55 t, 60 61 t 102 103 104 proteins 27, 4041 t, 40 48-49 102 104 Cout, 92 93 synovial fud cytology 28 29 30 31 4 92 93 glucose 68 69 92 mucun and viscosity 54 55 t 60 OL t, 83 t 92 proteins 40-41 t, 48-49 + 50 92 unic acid 64 65 t 92 93 synovial tissue krstological changes 92 93 Group I description 19 fud findings un 78 88 122 123 Group II description 19 flud findings in 98 123 Group intermediate description 19 fluid findings in 89 97 Infections generalized joint solve ment 21 22 98 100 Infectious arthridas 98 110 negative cultures 21 22 99 synovial tissue histological changes 101 therapeutic aspiration 16-17 See also Septic Conococcal, and Meningococcal arthritis Inlammauan cytological reaction to 38 effect on permeability of synovial tis sue 89 use of term 19 fn Intermittent hydrarthrosis 18 19 118 fud findings 25 t, 36 37 [ 74 119 120 See also Rheumatoid arthnos In vivo dialysate distributor ratos compared with synovial flud 5t Joint aspiration indications for 16-17 diseases classification 18 19 entrance of substances 3 8 knee blood vessels facing 4 p fac ing 5 p lubrication 6 86 removal of substances 89 Haverhill fever flud findings 105 Hemophilic arthritis 90 synovial Aud cytology 28 29 4, 30-31 t. 38 90 glucose 68 69 t mucin and viscosity 54 55 & 60- 61 t, 63 t, 90 proteins 40-41 € 48-49 , 90 Hemorrhagıc vellous synovias 95 96 synovial furd cytology 28 29 t 30-31 t, 38 95 124 125 glucose 68 694, 95 mucin and viscosity 54 55 4 60-61 # 95 96 125 proteins 40-41 t, 48-49 4, 95 synovial tissue histological changes 96 Hyaluronic acid See Synovial fluid mucin polysaccharide Hyalurondase effect on absorption from joint 9 See also Enzymes Hyperglycemia in infectious arthritis 101 Hypersensibyty arthritis due to Rud findings 94 95 Hypertrophic pulmonary osteoarthro pathy 89 90 synovial fuld cytology 28-29 t, 30-31 4 89 126 glucose 68 69 4 89 Leukemia lymphocytic See Neoplasms Lupus erythematosus disseminatus 91 92 synovial fluid cytology 24 28 29 30-31 t 91 223 glucose 68 69 + 91 mucın and viscosity 54 55 56 60 61 6 63 91 123 protens 40-4t 48-49 + 91 synovial tissue histological changes 91 92 Lymph distribution ratios compared with synovial Aud 4 5 6 Merungococcal arthntis 104 105 synonal fiwd ·INDEX 147 Meningoccocal arthritis, synovial fluid Proliferatlic arthritis Sec Rheumaloid (cont ) arthritis cultures 21 Psoriasis with arthritis Scc Rheumatoid cytology 25-26 4 28 29 30 31 arthritis 36 104 glucose 68-69 t 104 Reiter s syndrome 110 112 mucin and viscosity 54 55 t 60 synovial fluid 61 t, 105 cultures 111 proteins 40-41 € 48-49 105 cytology, 20 + 28 29 30-31 t 36 Metabolism intra articular, 67, 129 130 111 Methods 10-18 63 glucose 68 69 , 111 112 Mucınase See Enzymes mucin and viscosity, 54-55 t, 60- Myxedema mucin 6 56 57, 03 01 t 63 t 111 proteins 40-41 t 48-49 t, 50 111 Neoplasms 112 synovial flud findings 96-97 synovial tissue histological changes synovial tissue histological changes 111 112 97 Rheumatic fever 93 9-1 Neuroarthropathy 83-85 synovial fluid synovial fluid cytology 25 t 28 29 30-31 t 38 cytology 24 28 29 t, 30-31 t 38 93 94 84 85 glucose 68-69 t 93 glucose 68-69 , 84 muçin and viscosity 54-55 t, 60 61 mucın and viscosity 54-55 , 56 t 62 63 t 93 94 123 60-61 t, 63 , 84 85 proteins 40-41 t 48-49 t, 50 93 proteins 40-41 € 48-49 , 84, 85 94 Wassermann reaction 84-85 synovial tissue histological changes synovial tissue histological changes 94 85 Rheumatoid arthritis 112 118 Osteochondritis dissecans 82 intermittent effusions 35 f 44 45 t synovial fluid serum proteins 42-43 t 50 51 cytology 28 29 t 30-31 t, 82 spondylitis (Mane Strumpell) fluid glucose 68-69 , 82 findings 112 114 mucın and viscosity 54-55 t, 60 Stills disease Aud findings 112 114 61 t, 82 synovectomy effect on fluid findings proteins 40-41 t, 48-49 t 82 116 117 118 119 t synovial tissue histological changes synovial flud 82 cultures 21 Osteochondromatosis 82-83 cytology 25 t 27 28 29 30-31 t, synovial fluid 35 f 38 113 116 126 127 cytology 28 29 t, 30-31 t, 83 124 relation to duration of disease glucose 68 69 € 83 124 42-43 t mucin and viscosity 54 55 t, 56 relation to duration of effusion 60 61 t, 63 t, 83 123 124 25 t, 32-33 t 34 t 36 proteins 40-41 t 48-49 , 83 124 relation to severity of effusion synovial tissue histological changes 27 33-34 42-43 t 83 effect of adrenocorticotrophic hor mone and cortisone therapy 32 Penicillin sensitivity flud findings 94- 57 63 64 115 95 electrolytes 73 t 74 116 Permeability See Synovial tissue per glucose 35 f 67 68 69 4, 113 116 meability 126 127 Pregnenolone effect on absorption from Telaton to duration of disease joint, 9 42-43 t Pressure intra articular 19 81 83 87 relation to duration of effusion 89 92 93 102 107 113 119 32 5 67 ·148 INDEX Rheumatoid arthats synovial fluid glucose (cont ) relation to severity of disease 42-43 + 67 70 mucın and viscosity 42 43 44 45 + 54 55 57 58 59 60 61 1, 63 114 116 proteins 39 40 41 42 43 44 45 # 48 49 t 50 113 114 116 relation to duration of disease 89 42-43 t 47 relation to duration of effusion 32 33 + 39 44 47 50 relation to severity of disease 39 42-43 44 48 synovial tissue histological changes 115 118 with psonasis flud findings 112 114 Rheusatosd arthritas diagnosis probable 112 115 synovial fund cytology 28 29 t 30 31 t glucose 68 69 t mucn and viscosity 54 55 t 60 61 t 63 t proteins 40-4) 48-49 t Severity of disease relation to synovial fud (cont proteins 39 42-43 44 45 46 47 50 102 128 129 Ser hormones effect on mıcırt 57 Spondyktis (Marie Strumpell) See Rheumatoid arthritis Sulis disease See Nheumatoid arthritis Synovial fluid amount 2 3 5 19 20 81 116 antibodies 9 22 23 ascorbic acid 59 bactericidal activity 23 calcium 73 74 82 85 87 fa 104 109 110 chloride 73 74, 82 85 104 309 chloromycetın 65 66 complement fixation test, 22 23 84 85 103 104 109 110 cultures 21 22 99 100 101 111 cytology 23 38 dragnostic value 24 38 normal 2 3+ 23 24 relation to duration of disease 42 43 t relation to duration of effusion 25 28 32-33, 33 34 t relation to severity of disease 24 26 27 33-34 36 42-43 + See also individual diseases distribution ratios 3 5 t, 73 t edeina 10 electrolytes 73 78 82 85 104 109 116 enzymes 910 77 fibrinogen 3 + 52 glucose 3 t 66 72 diagnostic value 72 rate of entrance 67 Telation to anesthesia 68 fn relation to bacteria 67 relation to duration of disease 42 43 t relation to duration of effusion 32 33 , 35 f 37 € 67 relation to Bud lactate 75 f relation to food 66 t 66 In relaton to leukocytes 70 relation to severity of disease 42 43 + 87 70 See also individual diseases gross appearance 2 2021 79 82 83 84 86 87 89 90 91 92 93 95 96 100 101 102 104 107 109 111 112 113 119 hemorrhagic 12 38 80 123 Salicylates entrance into joints 8 Scurvy flud findings 90 93 Septıc arthros 100 102 synovial ffusd cultures 99 100 cytology 25 + 28 29 30-31 , 100 101 102 glucose 68 69 t IOL 102 mucın and viscosity 54 55 & 57 58 59 60 61 63 101 102 proteins 40-41 t, 48-49 t 101 102 See also Infectious arthrius Serum glucose 32 33 t, 42-43 t 101 Serum proteins 32 33 f 42-43 t 45 50 51 Serum sickness See Hypersensitivity arthritis due to Severity of disease relation to serum findings 42-43 t 50 101 relation to synovial fluid cytology 24 26 27 32-34 35 f 36 37 f 42-43 t, 102 128 sluscose 35 f 37 f 42-43 t 67 70 102 129 mucia and viscosity 42-43 & 44 45 t 53 57 58 62 64 101 102 103 329 ·INDEN 149 Synovial fluid (cont) hyaluronic acid See below, polysac chande hydrogen ion concentration 3 t 74 76 81 87 01 101, 102 103 109 114 116 119 icteric inder 20, 64 80 81 87 110 116 lactate 72 73 75 f 84 92 93, 102 104 107 109 111, 113 Lipids 76-77 mucin 53-64 acetic acid precipitate types of 58 59 facing 62 p components of 4 concentration myxedema 6 50 63 relation to duration of disease 42-43 t relation to duration of effusion 32-33 t relation to severity of disease 42-43 + 57-58 62 63 degradation 5-8 58 58 f 59 estimation of degree 6 59 62 63 t diagnostic value 64 dissociation in vivo 4 effect of hormones 56-57 effect on calcium equilibrium 6 effect on water transfer 6 formation 1 45 53 56 117 118 glucosamine 3 , 4 7 t See also above mucin lubrication 6 normal 46 53 63 See also individual diseases non-electrolytes 3-4 64 73 non protein nitrogen 3 64 normal 2 10 penicillin 65 phosphate 5 , 73 74 82 85 87 fn 104 109 116 polysaccharide 4 53 64 121 concentration 53 81 87 116 Viscosity 6 7 f potassium 5 + 74 82 85 104 109 proteus 39 52 diagnostic value 52 electrophoretic patterns in vanous diseases 46 50 normal 3 , 4 5 39 relation to duration of disease 42 43 t 47 relation to duration of effusion 32 33 39 44 47 50 Synovial fluid proteins (cont ) relation to severity of discasc 39 41 42-43 + 41 40 47 50 See also individual diseases sodium 5 + 74 specific gravity 38 sulfonamides 65 thiocyanate 74 total solids 3 t 38 uric acid 3 64 65 87 92 93 116 viscosity, 53 84 normal 30 6 7 f 53 relation to duration of disease 42- A3 t relation to duration of cffusion 32- 33 t relation to mucın glucosamine con centration 6 7 8 relation to severity of disease 42 43 t See also individual diseases Synovial tissue histological changes 99 130 mobility of ions in 3 mucin formation I 45 56 117 118 permeability 1 6 8 9 67 128 129 factors affecting 8 9 46 to protein fractions 6 8 46 47 50 vascularity 4 facing 4 p facing 5 p See also individual diseases Syphilitic arthritis (congenital) 109 110 synovial sud cytology 28 29 30-31 t 109 110 glucose 68 69 € 109 mucin and viscosity 54 55 60 61 t, 63 t 109 110 proteins 40-41 t 48-49 t 109 110 Wassermann reaction 109 110 synovial tissue histological changes 109 Testicular extract degradation of mu CD 56 58 in Traumatic arthritis 78 82 synovial fluid cytology 28 29 t 30-31 t 38 79 81 glucose 68 69 79 mutin and viscosity 53 54 55 t 56 58 60 61 t 63 t 79 80 81 82 proteins 40-41 t 48-49 t 50 79 81 82 Traumatic arthritis over three months duration 47 78-80 ·150 INDEX Traumatic arthritis over three months duration (cont ) synovial tissue histological changes 80 81 Traumatic arthritis with hemorrhage synovial fluid cytology 25 t 28 29 30 31 t 34 36 38 80 glucose 66 68 69 t 80 mucin and viscosity 53 54 55 ! 60 61 t, 80 proteins 40-41 t 48-49 t 80 Tuberculous arthritis 106 109 guinea pig inoculations biopsies and cultures 107 108 109 synovial fluid cytology 28 29 30 31 t 107 108 123 124 125 Tuberculous arthritis synovial fluid (cont ) glucose, 68 69 t 107 109 123 124 125 mucin and viscosity 54 55 t 60 61 t 63 t, 107 109 proteins 40-41 t 48-49 + 107 109 125 synovial tissue histological changes 107 108 Ulcerative colitis associated with ar thritis fluid findings 115 Vitamin D poisoning flud calcium and phosphate 87 in

·No text on this page.