·No text on this page.

·THE ANTI-GLOBULIN (COOMBS) TEST IN LABORATORY PRACTICE

·No text on this page.

·THE ANTI-GLOBULIN
(COOMBS) TEST
IN
LABORATORY PRACTICE
I DUNSFORD, PhD, MI Biol
Principal Scientific Officer National Blood Transfusion
Scrvice Sheffield Honorary Lecturer in Blood Group
Genetics University of Sheffield Lecturer in Blood
Group Scrology College of Science and Technology
Sheffield
JEAN GRANT, FRCP
Director National Blood Transfusion Service Oxford
OLIVER AND BOYD
EDINBURGIE TWEEDDALE COURT
LONDON 392 WELBECK STRELT, W 1

·FIRST PUBLISHED
1959
© 1959 I DUNSFORD AND JEAN GRANT
PRINTED IN GREAT BRITAIN BY
OLIVER AND BOYD LTD EDINBURGH

·DEDICATION
The Book is dedicated to Miss Elizabeth
W ( Jan ) Ikin B SC kindly person
to whom so many blood group workers owc
so much

·No text on this page.

·ACKNOWLEDGEMENTS The authors would like to record their appreciation of the reading of the manuscript and helpful suggestions made by Dr R R A Coombs, of the University of Cambridge The authors are also greatly indebted to Miss Joy Wilson, Rescarch Secretary, for her secretarial work preparing the manuscript, and to Mr Harold Walkland for his drawings Finally the authors would like to acknowledge the value of the work and comments of the staff at the laboratories of the National Blood Transfusion Service, Sheffield and Oxford

·No text on this page.

·CONTENTS CHAPTER TAC XIL INTRODUCTION ANTIGENS AND ANTIBODIES Classifications of Antibodies The Mechanism of the Antigen Antibody Reaction 2 3 II THE SEROLOGICAL BASIS or THE ANTI GLOBULIN TEST Species Differences Anti-Globulin Antibodies The Role of Complement Nomenclature 8 8 9 11 12 III HISTORY AND NOMENCLATURE The Direct and Indirect Anti Globulin Tests AHG 13 14 IS IV 16 16 18 THE PRODUCTION OF ANTI GLOBULIN SERA Choice of Animal Choice of Antigen The Preparation of Gamma Globulin (Prooms method) The Immunising Injections Bleeding the Animal Inactivation of the Serum Treatment of the Clot 20 21 23 24 25 V THE PROCESSING OF ANTI HUMAN GLOBULIN SERUM The Preliminary Examination The Removal of Unwanted Antibodies Absorption method 2 Dilution method 3 Heat Treatment The Preparation of Anti globulin sera for other Animal Species 27 27 29 30 31 32 32 ix ·PAGE 34 36 36 X THE ANTI-GLOBULIN TEST CHAPTER VI STANDARDISATION OF ANTI HUMAN GLOBULIN SERUM Method of Standardisation A Preparation of Coated Cells B Preparation of Serial Dilutions of the Anti Globulin Serum с Interaction between Cells and Serum The Interpretation of Results The Standardisation for Other Globulins The Zoning Phenomenon 37 37 38 40 40 VII COMPLEXITIES OF THE ANTI GLOBULIN SERUM Antibody analysis by Gel Diffusion Techniques 45 50 VIII THE TECHNIQUE OF THE ANTI GLOBULIN TEST The Indirect Anti globulin Test The Direct Anti globulin Test Serological points to be Noted The Test serum Cell suspensions The Anti Globulin Serum The mixing of the Reagents Controls Apparatus Recording of Results and Their Interpretation 52 53 53 55 5S 55 60 61 61 63 63 IX 66 66 1 THE CAUSES OF FALSE REACTIONS Causes of False Positive Reactions Anti human Species Antibodies in the Anti globulin serum 2 Cold complete agglutinins 3 Cold Incomplete Antibodies 4 Bacterial Contamination 5 Inadequate Cell Washing Sa The Effects of drying 6 The Effects of Silica 7 Other Agglutinating Substances in Saline 66 67 67 68 69 70 71 72 ·CONTENTS CLARITA TAOT 72 Causes of False Nogatic Results Thc Neutralisation of Ants Globulin by Traces of Serum 2 Incorrect dilution of the Anti Globulin Scrum 3 Lack of Complement 4 The Use of Aged Red Cells 5 The Use of Thin Cell Suspensions 6 Over Centrifuging in the Tube Technique 7 Poor Storage of Reagents 72 73 74 74 74 75 75 78 78 80 81 82 APPLICATIONS OF THE ANTI GLOBULIN TEST Blood Transfusion The Anti Globulin Compatibility Test 2 The Rapid Anti Globulin Compatibility Test The Investigation of Transfusion Reactions The Detection of Blood Group Antigens and Antibodies Hæmolytic Disease of the Newborn in Man Fallacies of the Direct Anti Globulin Test The After Effects of Hæmolytic Disease of the Newborn Hæmolytic Anzmia Platelets Leucocytes and the Corresponding Antı bodies Other Applications of the Anti Globulin Test Bacteriology Estimation of Globulins in Serum and other Fluids Forensic Medicine Hæmolytic Disease of the Newborn in Animals Immunological Research 83 85 89 92 93 97 99 99 100 103 105 108 XI ODDMENTS The Anti Globulin Test in Cancer The Anti Globulin Test in Lead Poisoning The Anti Globulin Test and Other Chemicals A Test which may Supplement the Anti Globulin Technique 114 114 115 116 117 INDEX 119 ) ·PAOE 34 36 36 X THE ANTI-GLOBULIN TEST CHAPTER VI STANDARDISATION OF ANTI HUMAN GLOBULIN SERUM Method of Standardisation A Preparation of Coated Cells B Preparation of Serial Dilutions of the Anti Globulin Serum C Interaction between Cells and Serum The Interpretation of Results The Standardisation for Other Globulins The Zoning Phenomenon 37 37 38 40 40 VII COMPLEXITIES OF THE ANTI GLOBULIN SERUM Antibody analysis by Gel Diffusion Techniques 45 50 VIII THE TECHNIQUE OF THE ANTI GLOBULIN TEST The Indirect Anti globulin Test The Direct Anti globulin Test Serological points to be Noted The Test serum Cell suspensions The Anti Globulin Serum The mixing of the Reagents Controls Apparatus Recording of Results and Their Interpretation 52 53 53 55 55 55 60 61 61 63 63 IX 66 66 1 THE CAUSES OF FALSE REACTIONS Causes of False Positive Reactions Anti human Species Antibodies in the Anti globulin serum 2 Cold complete agglutinins 3 Cold Incomplete Antibodies 4 Bacterial Contamination 5 Inadequate Cell Washing 5a The Effects of drying 6 The Effects of Sulica 7 Other Agglutinating Substances in Saline 66 67 67 68 62 70 71 72 ·INTRODUCTION THE anti globulin reaction, or Coombs' test, has played a major part in the desclopment of our know Icdge of blood groups in man and animals It is an essential techniquc in the forecasting and diagnosis of htmolytic discase of the new born in various spccics, and through its use the transfusion of blood from donor to patient has become a safer procedure The bacteriologist finds it a useful technique for the detection of bacterial antibodies, the forensic scientist can apply it in the detection of crime, the hæmatologist uses the test routincly in the investigation of hæmolytic anæmias, for the study of antibodies to various cellular elements of the blood and he may possibly find it helpful in the study of certain plasma fractions, finally the experimental immunologist applies it to study antigenantibody reactions Published descriptions of the anti globulin testas originally devised, and the more recent developments, modifications and extensions of the technique are widely scattered in various journals, making it exceedingly difficult for the student to grasp the underlying principles an to apply them correctly in practice This book is an attempt to provide, for those interested in medicine bacteriology, hæmatology, blood transfusion and the veterinary and forensic sciences an explanation of the fundamentals of the anti globulin test with suggestions for further reading which they may find helpful in the application of the test to their own particular problems

·No text on this page.

·CHAPTER 1 ANTIGENS AND ANTIBODIES Berore embarking on the main theme of this book, it is proposed to define for the reader some of the terms in common use in the field of immunology The essential part of all the tests to be described later is the antigen antibody reaction But what arc antigens and antibodies? An antigen is any substance which, when introduced into the body of an animal to which it is foreign, is capable of stimulating the host animal to produce a specific antisubstance or antibody Antigens may be protein in nature and some carbohydrates are also antigenic The animal which has been provoked, by the introduction of a foreign antigen, into producing an antibody directed specifically against that antigen is said to have been immunised An antibody is a substance found in the blood plasma or serum which reacts with cells, or substances in solution, containing the antigen for which it is specific Such antibodies are probably formed in the body tissues and sometimes their presence can be demonstrated in other body fluids such as colostrum and peritoneal exudates They are part of the globulin fraction The most easily demonstrated antigen antibody reactions are those used in blood grouping Serum containing antibody of known specificity is mixed with a saline suspension of red blood cells If these cells then form clumps or agglutinates they are assumed to carry the equivalent antigen, if they do not agglutinate they are considered to lack the antigen Conversely, by mixing a serum which is to be investigated with a series of suspensions of red cells of known antigen content it is possible, by noting B

·No text on this page.

·ANTIGENS AND ANTIBODIES 3 reaction regardless of the composition of the fluid in which the antigen is suspended, or the degree of magnification cmployed to discover it The presence of thcsc cryptagglutinoids can be disclosed only indirectly by the antiglobulin technique THE MECHANISM OF THE ANTIGON-ANTIBODY REACTION The forces underlying the antigen antibody reaction are still a matter for conjecture The most practical hypothesis envisages it as a two stage response in which the stages are not distinct for it is possible that the second stage starts before the first is complete In the first stage the antibody rapidly forms a chemical combination with the antigen although it produces no outward evidence that such a combination has occurred "The second stage, which progresses more slowly, is the phase of observable reaction in which agglutination or precipitation becomes obvious It is probable that the second stage is a physical response depending upon, or modified by, such factors as the composition of the medium in which the reaction has occurred, the relative quantities and reaction potential of antibody and antigen, and the presence or absence of complement "Antigen antibody reactions are not all of the same type and they do not all require the operation of the same factors As an example, for the detection of most lysins complement must be present, but it is not an essential ingredient in the demonstration of agglutinins' Following this hypothesis it would appear that incomplete antibodies are those which participate in the specific combination of stage one, but are unable to complete stage two so long as the surrounding medium is of small molecular size How can we know that stage one has in fact occurred When the antigen is part of a red blood cell the fact that it ·2 ANTI-GLOBULIN (COOMBS) TEST the occurrence of agglutination, to determine whether the serum contains an antibody and what its specificity is CLASSIFICATIONS OF ANTIBODIES Antibodies can be divided into a number of categories They may be classified according to the manner in which they react with the antigen in vitro An antibody which precipitates the antigen is called a precipitın, one which dissolves the antigen is a lysin, while an antibody which will produce clumping of cells carrying the antigen is an agglutının' The antigens corresponding to these varieties of antibody are respectively designated precipitinogens, lysinogens and agglutinogens The reactions between these antibodies and antigens can be seen by the unuded eye or with a magnifying lens or microscope A slightly different classification divides antibodies into two broad groups based upon their ability to interact with the antigen suspended in isotonic saline solution Complete antibodies (sometimes called bralent antibodies) are those which cause an obvious alteration such as agglutination, precipitation or lysis of the appropriate antigen in salıne medium "The incomplete (or so called monovalent) antı bodies are unable to produce such visible evidence of a reaction with their corresponding antigen in saline and to demonstrate their presence additional procedures are required Two types of incomplete antibody are recognised--the agglutinoids and the cryptagglutinoids An agglutinoid is an antibody which exhibits no visible effect on the antigen unless the reaction takes place in a protein medium such as boyine albumin or in certain macro molecular solutions such as dextran or until the antigen has first been modified by enzyme action Agglutinoids are sometimes called albumin antibodies Cryptagglutinoids, on the other hand, are antibodies which fail to bring about any observable ·5 ANTIGONS AND ANTIBODIES Not all incomplete antibodies haic this capacity to block, wholly or partially, thc corresponding intigen The interactions between red cell antigens and complctc and incomplete antibodies are more easily understood when represented diagrammatically The first illustration (fig 1) shows, in diagrammatic form, a red cell with blood group antigen receptors on its surface and alongside it are two RED CELL Antigen Receptor Complete Incomplete (Bivalent) (Univalent) Antibody Antibody FG 1 09 FIO 2 Cells agglutinated by complete antibody complete and two incomplete antibody molecules The agglutination of red cells by complete antibody is visualised as a uniting of the cells (fig 2) by bivalent antibody bonds In contrast, incomplete antibody attaches itself to the antigen (fig 3) but being monovalent, fails to produce agglutination unless an extra something is provided by bovine albumın which links (fig 4) the coated cells together It must be emphasised that the exact nature of the differences between complete and incomplete antibodies, and the mechanism of the antigen antibody reaction, are not yet completely understood The diagrams depict symbolically what is thought to take place but they do not explain ·4 ANTI-GLOBULIN (COOMBS) TEST has combined with its corresponding antibody can be demonstrated in at least two ways Firstly, let us suppose that some red cells possessing the antigen X (X positive cells) are suspended in saline and are then mixed with serum containing an incomplete antı X antibody They do not agglutinate in spite of their contact with the corresponding antibody If however, the superna tant serum and saline are removed from the sedimented cells and an equal volume of a 20% bovine albumın solution is substituted, it will be found that, provided the time allowed has been sufficient and the temperature at which the test is per formed is suitable, agglutination of the cells will occur This agglutination is not due to the action of the bovine albumin alone since X-positive cells which have not been exposed to anti X do not spontaneously agglutinate when suspended in the albumin Therefore the antibody—incomplete antı X-must have combined with red cells That the reaction is specific is shown by the fact that the incomplete antı-X will not combine with red cells other than those carrying the X antigen This albumin replacement technique 1 is a particularly useful method for detecting the Rhesus (Rh or D) antigen in the red blood cells in man Only incomplete antibodies of the agglutinoid type will bring about this agglutination reaction, cryptagglutinoids fail to clump the cells even in the presence of bovine albumin A second proof that stage one has occurred is provided by the blocking test of Wiener (1944)2 A salıne suspension of X positive cells is mixed with an incomplete antı X serum If, after a time the cells are washed free from incomplete antı X and are then mixed with a complete anti X serum the normally expected agglutination will either not occur or if it does, the degree of agglutination will be much less than that shown by X positive cells not previously exposed to incomplete anti X The incomplete antibody has blocked, inhibited or prevented the action of the complete antibody ·5 ANTIGENS AND ANTIBODICS Not all incomplete antibodies have this capacity to block, wholly or partially, the corresponding antigen The interactions between red-cell antigons and complctc and incomplete antibodies are more casily understood when represented diagrammatically The first illustration (fig 1) shows, in diagrammatic form, a red cell with blood group antigen receptors on its surface and alongside it are two RED CELL [ Antigen Receptor Completc Incomplete (Bivalent) (Univalent) Antibody Antibody HG 1 ola FIG 2 Cells agglutinated by complete antibody complete and two incomplete antibody molecules The agglutination of red cells by complete antibody is visualised as a uniting of the cells (fig 2) by bivalent antibody bonds In contrast, incomplete antibody attaches itself to the antigen (fig 3) but being monovalent, fails to produce agglutination unless an extra something is provided by bovine albumin which links (fig 4) the coated cells together It must be emphasised that the exact nature of the differences between complete and incomplete antibodies, and the mechanism of the antigen antibody reaction, are not yet completely understood The diagrams depict symbolically what is thought to take place but they do not explain ·6 ANTI-GLOBULIN (COOMBS) TEST all the types of reactions that have been observed For instance, it has been shown that incomplete blood group antibodies can be made to agglutinate their appropriate red cells suspended in saline provided that these cells carry more than the usual amount of antigen,' or provided that the cell surface has been modified, increasing the number of antigen receptors, by partial digestion by enzymes It may be that ability or otherwise of an incomplete antibody to agglutinate red cells in saline is governed not by the number 4 FIO 3 Cells coated by incomplete antı body (not agglutinated) FIG 4 Coated cells agglutinated by addition of bovine albumin of combining sites on the antibody molecule but, rather by the number of antigen receptors and hence by the amount of antigen available on the cell surface The foregoing explanations have dealt mainly with antigen antibody reactions as seen in blood group serology, but it has been shown that similar incomplete antibodies to bacterial antigens exist "The two main points to be borne in mind are (1) that some antibodies will give rise to no visibly detectable reaction of the antigen is suspended in saline only, and (2) that no negative reaction can be accepted unless the possible presence of incomplete antibodies-- agglutinoids and cryptagglutinoids-has been excluded

·ANTIGENS AND ANTIBODIES
7
REFERENCES
I DUNSFORD I and Bowity, C C. 1955 Techniques in Blood Grouping
Oliver and Boyd Edinburgh
2 WIENER AS 1944 A new test (blocking test) for Rh sensitisation
Proc Soc Exp Biol NY 56 173
3 RACE, RR SANGER R and SELWYN ) G 1950 A probabic
deletion in a human Rh chromosome Nature 166 520
4 MORTON JA and PALES MM 1951
The protcolytic enzyme
test for detecting incomplete antibodies J Clin Path 4 189

·6 ANTI-GLOBULIN (COOMBS) TEST all the types of reactions that have been observed For instance, it has been shown that incomplete blood group antibodies can be made to agglutinate their appropriate red cells suspended in saline provided that these cells carry more than the usual amount of antigen, or provided that the cell surface has been modified, increasing the number of antigen receptors, by partial digestion by enzymes & It may be that ability or otherwise of an incomplete antibody to agglutinate red cells in salıne is governed, not by the number FIO 3 Cells coated by incomplete antı body (not agglutinated) FIG 4 Coated cells agglutinated by th addition of bovine albumin of combining sites on the antibody molecule but, rather, by the number of antigen receptors and hence by the amount of antigen available on the cell surface The foregoing explanations have dealt mainly with antigen antibody reactions as seen in blood group serology, but it has been shown that similar incomplete antibodies to bacterial antigens exist "The two main points to be borne in mind are (1) that some antibodies will give rise to no visibly detectable reaction if the antigen is suspended in saline only, and (2) that no negative reaction can be accepted unless the possible presence of incomplete antibodiesagglutinoids and cryptagglutinoids—has been excluded

·ANTIGENS AND ANTIBODIES
7
REFERENCES
I DUNIFORD I and BOWLEY, C. C 1955 Techniques in Blood Grouping
Oliver and Boyd Edinburgh
2 WIENER, A S 1944 A new test (blocking test) for Rh sensitisation
Proc Soc Exp Biol NY 56 173
3 RACE, RR SANGER R. and SELWYN JG 1950 A probablc
deletion in a human Rh chromosome Malure 166 $20
4 MORTOY JA and PICKLES M M
The proteolytic enzyme
test for detecting incomplete anubodics, Clin Path 4 189
1951

·CHAPTER II THE SEROLOGICAL BASIS OF THE ANTI GLOBULIN TEST SPECIES DIFFERENCES EACH species of animal has, in its blood cells, tissues or body fluids, its own species specific substances The differences between these substances cannot always be detected by chemical tests although they can be shown to be serologically distinct By injection of red cells or serum obtained from one animal into another animal of a separate species an anti species antibody can be provoked which may be an agglutinin, precipitin or lysin and which may be complete or incomplete in character Since all anti globulin sera are anti species antibodies it is essential for the user to be aware of the possibilities of cross reactions and of the various differences and similarities between species "The injection of human red cells into a rabbit will lead to the development by the rabbit of anti man antibodies Although such antibodies can be made specific (able to react only with the precise antigen which stimulated their production) it is often found that they will also react with red cells and other tissue cells from related species They combine most strongly with the antigen employed to provoke them (the homologous antigen) but they may also react to a lesser degree with antigens obtained from a related species of animal (heterologous antigens) For instance, fowls, immunised with rabbit serum develop an antibody which reacts with both rabbit and hare serum and yet a hare, immunised with rabbit serum, will produce an antibody capable of reacting with rabbit serum but not with hare serum. For more information on the preparation of anti species antibodies the work of Nuttall (1904)1 is well worth 8 ·SCROLOGICAL BASIS OF TIIT ANTI-GIOBULIN TEST 9 consulting Tor an understanding of scrological specificity the student is rcfcrred to the monumental work of Land. steiner (1945) ? ANTI GLODULIV ANTIBODILS When human serum is injected into a rabbit the rabbit responds by maling an anti-species antibody which will agglutinate all human red cells, but this may not be the only antibody which has been stimulated This first antibody can be remoi cd by mixing the rabbit's serum with wellwashed human red cells The antı human spccics anubody becomes attached to the red cells and can be discarded with them, leaving in the rabbit scrum a different antibody The second antibody is an anti globulm provohed by the antigenic action of the globulins which were present in the human serum used initially to immunise the rabbit The anti globulin antibody is also a species specific antibody--in this case the rabbit produced an anti-human globulin It will not agglutinate normal human red cells but it will react with any globulins derived from a human source such as whole serum and certain body tissues As mentioned previously, antibodies are themselves globulins and therefore the anti globulin antibody may be thought of as an antiantibody antibody! In a mixture of two sera, one containing antibody and the other containing the correct species anti globulin antibody, the reaction could be manifest as a precipitate There is, however, a more usual and more useful way of using an anti globulin serum, which forms the basis of the Coombs test Let us take as an example the case of a patient who has received a blood transfusion and who, as a result, has developed an incomplete antibody to the bloodgroup antigen called " Rh' The incomplete anti-Rh serum from this patient will not agglutinate Rh positive cells suspended in saline, but the antibody will become attached ·10 ANTI-GLOBULIN (COOMBS) TEST to the cells, giving them a coating of human globulin, there may be also some excess of uncombined antibody in the supernatant saline To rid the serum salıne mixture of unwanted globulins which would readily combine with anti globulin antibody and spoil the test, we must " wash' these coated cells This we do by (1) centrifuging the mixture, (2) removing the supernatant serum saline solution (3) replacing this by a large volume of fresh salme, (4) mixing the cells thoroughly, and (5) re centrifuging For Rabbit Rh Antibody Ants human Globulin Rh Antigen (Bivalent) Receptor Globulin RED CELL FIGS efficient washing this cycle has to be repeated several times and the end product is a saline suspension of cells in which the only globulin present is the antibody coating the ells, 1e specifically combined with the antigen (One word of warning should be interpolated here--cell washing is usually performed at room temperature or occasionally at 37 C Higher temperature should not be used since this may result in removal of the antibody from the cells ) We now take a rabbit anti human globulin serum prepared from the serum of a rabbit which had received injections of human serum and from which the anti species antibody has been absorbed If we mix this rabbit serum with the washed, coated red cells the anti globulin antibody reacts with the globulins on the cell surfaces, causing the ·STROLOGICAL BASIS OF TITANTI-GIOBULIN TEST 11 cells to clump together as if the agglutination were duc to the action of the anti globulin alonc We know this appearance to be false because the specific mbbit anti human globulin antibody will not cause agglutination of human red cells unless they havc first becn cxposed to the action of a specific incomplete antibody Again, picional representation of the supposed mechanism may be helpful The first diagram (fig 5) shows a red cell Anbgen Receptor RED CELL RED CELL Rh. Univalent Antibody Ants human Globulin Agglutinin (Bivalent) F10 6 coated with incomplete antibody, together with four molecules of rabbit antı human globulin antibody The second diagram (fig 6) illustrates the attachment of red cells to each other by the bonding effect of the anti globulin reacting with the globulins on the cells' surface THE ROLE OF COMPLEMENT The work of Hackett (1950) ' has shown that, so far as antt human globulin human globulin reactions are concerned, the addition of guinea pig complement neither enhances nor inhibits the reaction On the other hand, it has also been found that certain incomplete antibodies to red cell antigens in man, eg anti Kidd, antı Lewis, will coat red cells in the ·12 ANTI-GLOBULIN (COOMBS) TEST absence of complement, or complement like substances, but this coating will not be detected by the anti globulin reagent. However, if to the sensitised cells fresh human serum is added and then the cells are rewashed, a positive anti globulin reaction can be obtained NOMENCLATURE When referring to anti globulin antibodies it is usual to describe them in such a way as to indicate both the animal in which the antibody was produced and the specificity Throughout this chapter we have mentioned “rabbit antı human globulin serum” In an alternative terminology the name of the immunised animal is placed last Our example then becomes antı human globulin rabbit serum REFERENCES 1 NUTTALL G HP 1904 Blood immunity and blood relationship Cambridge University Press 2 LANDSTEINER 1945 The specificity of serological reactions Revised Edition Harvard University Press 3 HACKETT E 1950 The Coombs Test and guinea pig complement Nature 165 764 4 STRATTON F Value of fresh serum in the detection and use of anti Jks antibody Vox Sanguinis 1 160 1956 ·CHAPTLR HISTORY AND NOMENCLATURE Many people refer to the anti-globulin test as the Coombs test and believe that it was discovered by Dr RRA Coombs of Cambridge University in 1945 In fact, the principle of the test was first cmployed in 1908 by an Italian immunologist, Carlo Moreschi (1908)' ? Moreschi diluted specific anti sera to red cells and bacteria until the scra no longer caused agglutination Hc then thoroughly washed thc unagglutinated cells and crposed them to anti specics scra and obtained agglutination The anti specics scra ucrc specific for the animals which provided the original sera which, diluted, failed to causc agglutination Interpreting Moreschi's observations it would appcar that on diluting the sera hc diluted out the completc antibodies, but left higher-titred incomplete antibodies capablc of coating the bacteria and the red cells This coating would be a globulin, and the anti-species sera contained anti slobulins which caused the agglutination of the sensitised cells Unfortunately, Moreschi's technique was not applied in routine serology for the next 37 years and it was only in 1945, when the problem arose of finding new methods with which to detect the recently discovered Rhesus antibodies, that Coombs, Mourant and Race (1945), unaware of Moreschi s earlier work, devised this method Dr Coombs (1954) in a lecture delivered in Rome, paid tribute to Moreschi for his discoveries Without wishing in the least to detract from Dr Coombs' well deserved distinction, it becomes obvious that the term * Coombs' test” ignores the contributions of Moreschi and of Mourant and Race Dr Coombs himself calls the technique the anti globulin sensitisation test" To $6 13 ·14 ANTI-GLOBULIN (COOMBS) TEST include the word "sensitisation ” makes the title rather cumbersome for routine use, but it is an advantage to omit the reference to species (as in "antı humın globulin serum"), thereby giving a general description of the method irrespective of the anti species globulins used' The term anti globulin rest indicates the potential breadth of application of the technique and is to be preferred However, in describing the method, especially when cells and sera from different animals are used it is essential to specify the species from which the antigen was obtained and the animal in which the anti globulin was developed eg goat anti rabbit globulin or anti rabbit globulin goat serum Hill, Haberman and Jones (1948). devised the name Developing Test" for the anti globulin technique It is an apt descriptive title since the anti globulin serum" develops' the agglutination of coated cells in a way analogous to that in which various chemicals act upon a photographic plate There is no indication, however, of what is being developed nor of which reagents are being used 92 THE DIRECT AND INDIRECT ANTI GLOBULIN TESTS The anti globulin test can be used in two ways Where there is reason to suppose that red cells may have been coated by antibody, the addition of anti globulin serum will show by the subsequent production of cell clumping or by a lack of agglutination whether or not the cells have been sensitised The name coined by Coombs Mourant and Race (1946) for this technique is the Direct antiglobulin test The Indirect anti globulin test is used in the investigation of any serum which it is thought may contain incomplete antibody The first step is to mix the unknown serum with red cells and to allow time for the antibody if present to become attached to the cells If the cells are then thoroughly washed and examined for sensitisation by the ·15 IJSTOR) AND VOM\CI ATURI Direct anti globulin method, thic des clopment of acclutinates will indicaic that an incomplctc antibody, capable of conting thc cells, was present in the scrum which was being investigated AHG Human naturc being what it is, it was not surprising that those working with anti human globulin scrum in 1946 soon found this title too burdensome and substituted for it the abbreviation AHG" It was unfortunatc, therefore, that some hæmatologists about the same time decided to usc the initials AHG for an entirely different substancc, anti hæmophilic globulin, a fraction of plasma cmployed to combat the deficiency present in homophilia Considerable argument can be adduced for and against the use of the letters AHG to signify the anti hemophilic globulin but it is of conspicous practical importance that, in any laboratory handling both the anu hamophulic factor and antı human globulin scrum, stcps should be taken to avoid confusion between them Thc distinction can be satisfactorily maintained by calling the former AHF and the latter AHG REFERENCES 1 MORESCH с 1908 Neue tatsachen uber die blutkorperchen agglutination Zenir Bakteriol Parasitenk 46 49 2 MORESCHI C 1908 Beschleunigung und Betstarkung der Dachterien agglutination durch Antierweissera , Zentr Dachteriol Parasitenk 46 456 3 COOMBS RRA MOURANT A C and RACE R R 1945 Detection of weak and incomplete Rh agglutinins A new test Lancer 11 15 4 COOMBS RRA 1954 Moreschi e Aluni racenti sviluppi nello studio della agglutinazione L informatore Medico IX Fasc 6 126 $ HILL J M , HABERMAN S and JONES F 1948 Hæmolytic Rh immuno globulins for a possible third order of antibodies incapable of agglutination or blocking Blood 3 Suppl 2 80 6 COOMBS RRA MOURANT A E and RACE RR 1946 In vivo Iso sensitisation of red cells in babies with hæmolytic disease Lancet i 264 ·66 14 ANTI-GLOBULIN (COOMBS) TEST include the word "sensitisation ” makes the title rather cumbersome for routine use but it is an advantage to omit the reference to species (as in 'anto human globulin serum"), thereby giving a general description of the method irrespective of the anti species globulins used The term anti globulin test indicates the potential breadth of application of the technique and is to be preferred However, in describing the method, especially when cells and sera from different animals are used it is essential to specify the species from which the antigen was obtained and the animal in which the anti globulin was developed eg goat anti rabbit globulin, or anti rabbit globulin goat serum Hill Haberman and Jones (1948) 5 devised the name Developing Test for the anti globulin technique It is an apt descriptive title since the anti globulin serum 'develops the agglutination of coated cells in a way analogous to that in which various chemicals act upon a photographic plate' There is no indication, however, of what is being developed nor of which reagents are being used THE DIRECT AND INDIRECT ANTI GLOBULIN TESTS The anti globulin test can be used in two ways Where there is reason to suppose that red cells may have been coated by antibody the addition of anti globulia serum will show by the subsequent production of cell clumping or by a lack of agglutination whether or not the cells have been sensitised The name coined by Coombs Mourant and Race (1946)6 for this technique is the Direct antiglobulin test The Indirect anti globulin test is used in the investigation of any serum which it is thought may contain incomplete antibody The first step is to mix the unknown serum with red cells and to allow time for the antibody, if present to become attached to the cells If the cells are then thoroughly washed and examined for sensitisation by the ·TIC PRODUCTION OF ANTI-GLOBULIV SERA 17 thesc Eyquem (1956) and Goodman and co-workers (1957) uscd, respectively, donkeys and chickcas The choice of animal may be goi crned by the volumc of the yield required The amount obtainable from a goat will be ten times that which can be procured from a rabbit Nevertheless, when small quantities only arc nccded, or when scicral rcagents from different animals are required, the rabbit is an ideal animal Within a specics, not all animals arc cqually good antibody producers Thc Vanctics of rabbits known, in the United Kingdom, as Chinchilla, Flemish, Lop and Belgian Harc arc gcncrally morc satisfactory than Ror or Dutch rabbits When a strain of rabbits is discoscred to be particularly competent in producing antibodies their offspring, too, will generally be found to be good antibody producers This suggests that the ability to develop antibodies readily may be genetically determined and, therefore, breeding from such stock is recommended Experience with goats may be similar Dunsford (unpublished observation), using five goats of no particular breed, has not yet failed to obtain a satisfactory scrum, but Proom (1957) has been less fortunate A full protein diet supplemented with vitamins, etc, is recommended for rabbits On the other hand, goats appear to be capable of antibody production on any dieteven including a laboratory coat or the morning newspaper if available. Laboratories maintaining animal houses will find excellent guidance in the UFAW Handbook, On the Care and Management of Laboratory Animals, published by the Universities' Federation for Animal Welfare, London, Second Edition, 1957 In countries where myxomatosis is prevalent precautions should be taken to reduce the chances of the rabbits becoming infected They can also be protected by vaccination (a suitable vaccine is prepared by Burroughs Wellcome Ltd, London) C ·CHAPTER IV THE PRODUCTION OF ANTI-GLOBULIN SERA THE serum of an animal immunised by injection of serum from another animal may be found to contain two or more antibodies 1 The anti species antibody which will agglutinate or hæmolyse red cells of all members of the species from which the immunising antigen was derived, 2 The anti species globulin antibody-the reagent required for the anti globulin test, 3 An hetero antibody, which may be naturally occurring (present in the unimmunised animal) or immune For instance, rabbits are commonly found to have in their sera a naturally occurring antibody active against human group A red cells Before our antı globulin serum can be used it must be freed from the unwanted anti species and hetero antibodies Although similar procedures may be used to make, for example rabbit anti fowl globulin serum in the interests of simplicity the method of preparing antt human globulin serum only will be described CHOICE OF ANIMAL The rabbit being comparatively small and easy to handle, is the animal most commonly used for the production of antı human globulin serum However it is quite possible to immunise other species Hill and Haberman (1954) 1 and Dunsford and Bowley (1957) ? have successfully immunised goats Stratton (1956) 3 had success with sheep and also experimented with a variety of other species including hens and horses, though he does not comment on his results with 16 ·THIC PRODUCTION or ANTI-GLOBULIN SERA 19 suffering from acquired htmolytic antmia In this discasc the patient's red cells are prematurcly destros cd within his blood vessels and his scrum may be found to contain antibodies against his own red cells Thesc antibodies may be of two types—those which agglutinate red cclls morc readily at 37°C (warm type antibodies) and those which reict more strongly at room temperature or below (cold antibodies) Komninos and Aksoy (1955)" have used as their antigen coated red cells from cases of acquired hemolytic anæmia, and claim also to have produced potent antiglobulin sera by the injection of intibody clutcd from such cells Pisciotto and Hinz (1956) 1s have used sera obtained from patients with acquired hæmolytic anæmja and assert that the anti-globulin serum resulting from immunisation by a serum containing warm type antibody differs in specificity from the anti globulin serum evoked by a cold type antibody This may be explained to some extent by the findings of Christensen and Dacie (1957) 18 who, by electrophoresis, showed that although sera from warm antibody hæmolytic anæmias have no characteristic electrophoretic pattern, sera from cold antibody" hæmolytic anæmias have a peak in the 71 region It suggests, too, that even better reagents may be prepared by using as antigens, sera from patients with abnormal globulins such as are found, for instance, in multiple myeloma Sera from people whose blood groups are A, B or AB should not be used as antigen in the production of antiglobulin sera since they contain A and/or B blood group specific substances in solution which are antigenic and would cause the rabbit or goat to produce anti A and/or anti-B antibodies in addition to the required antı human globulin To avoid the necessity for removing such additional antibodies it is the custom to use a pool of sera obtained from ·18 ANTI GLOBULIN (COOMBS) TEST CHOICE OF ANTIGEN In the search for the ideal antigen for the production of anti globulin serum many materials have been tried, including serum, plasma, gamma globulins, coated red cells and antibodies eluted from coated cells Coombs, Mourant and Race (1946) ? in their original experiments used either whole human serum or human gamma globulin and found them equally efficacious, though whole serum was the easier reagent to prepare The authors of the MRC Memorandum No 27 (1952) 8 preferred gamma globulin Hill and Haberman (1954)? favoured whole serum Wootton (1950)' preferred gamma globulin while Singh Ramakrishman and Prakash (1952) 10 found plasma to be as effective as serum Kidd (1957)" has used gamma globulins prepared by Kendalls method but since some antı red-cell antibodies are non gamma globulins he later modified his immunsing schedule by giving his animals alternate courses of whole serum and gamma globulins in order to guard against producing a serum with too narrow an anti globulin spectrum Milgrom Luszczynski and Dubiskı (1956) 13 employed, as antigen globulin which had been denatured on the surface of red cells as a result of an antigen antibody reaction Rabbit red cells were agglutinated by human serum which had previously been heated at 56 C to inactivate the complement After having washed the rabbit red cells free of excess human serum the agglutinated cells were then injected into a rabbit Bi weekly injections for 6 weeks produced a satisfactory reagent containing very low titre antı human species antibodies requiring little absorption and the sera were free from zoning A slightly different approach has been used by those authors who have obtained cells or serum from patients ·THE PRODUCTION or ANTI-GLOBULIN SCRA 19 suffering from acquired homolytic antmi? In this discasc the patient's red cells are prematurely destroyed within his blood vessels and his scrum may be found to contain antibodies against his own red cells Thesc antibodics may bc of two types—those which agglutinatc red cells more readily at 37°C (warm type antibodics) and those which react more strongly at room temperature or below (cold antibodies) Komninos and Aksoy (1955)" harc uscd as their antigen coated red cells from cascs of acquired htmolytic anæmia, and claim also to havc produced potent antiglobulin scra by the injection of antibody eluted from such cells Pisciotto and Hinz (1956) is have uscd scra obtained from patients with acquired hæmolytic anemia and assert that the anti globulin serum resulting from immunisation by a serum containing warm type antibody differs in specificity from the anti globulin scrum cvoked by a cold-type antibody This may be explained to some extent by the findings of Christensen and Dacie (1957) 16 who, by electrophoresis, showed that although sera from warm antibody" hæmolytic anæmjas have no characteristic electrophoretic pattern, sera from cold antibody" hemolytic anæmias have a peak in the yı region It suggests, too, that even better reagents may be prepared by using, as antigens, sera from patients with abnormal globulins such as are found, for instance, in multiple myeloma Sera from people whose blood groups are A, B or AB should not be used as antigen in the production of antiglobulin sera since they contain A and/or B blood group specific substances in solution which are antigenic and would cause the rabbit or goat to produce anti A and/or antı B antibodies in addition to the required antı human globulin To avoid the necessity for removing such additional antibodies it is the custom to use a pool of sera obtained from $3 ·18 ANTI-GLOBULIN (COOMBS) TEST CHOICE OF ANTIGEN In the search for the ideal antigen for the production of anti globulin serum many materials have been tried, including serum, plasma, gamma globulins, coated red cells and antibodies eluted from coated cells Coombs, Mourant and Race (1946) ? in their original experiments used either whole human serum or human gamma globulin and found them equally efficacious though whole serum was the easier reagent to prepare The authors of the MRC Memorandum No 27 (1952) 8 preferred gamma globulin Hill and Haberman (1954)' favoured whole serum Wootton (1950) º preferred gamma globulin, while Singh Ramakrishman and Prakash (1952) 10 found plasma to be as effective as serum Kidd (1957) u has used gamma globulins prepared by Kendall's method but since some antı red cell antibodies are non gamma globulins he later modified his immunising schedule by giving his animals alternate courses of whole serum and gamma globulins in order to guard against producing a serum with too narrow an anti globulin spectrum Milgrom Luszczynski and Dubiskı (1956) 13 employed, as antigen globulin which had been denatured on the surface of red cells as a result of an antigen antibody reaction Rabbit red cells were agglutinated by human serum which had previously been heated at 56° C to inactivate the complement After having washed the rabbit red cells free of excess human serum the agglutinated cells were then injected into a rabbit Bi weekly injections for 6 weeks produced a satisfactory reagent containing very low titre antı human species antibodies requiring little absorption and the sera were free from zoning A slightly different approach has been used by those authors who have obtained cells or serum from patients ·TIC PRODUCTION OF ANTI-GI OBULI SIRA 21 obtain a better or quicker response in an animal which has been made slightly febrile by the injection of a “foreign protein hidd (1957)" mixes dried tuberclc bacıllt with gamma globulin and cmulsifics the mixture with Arlacci A (Mannidc mono-olcatc) With this reagent good antı globulin scra arc produod with low-litre anti human antibodies Tuc IMMUNISING INJECTIONS "The routes by which the antigen may be introduced arc intramuscular, (b) intraperitoncal, or (c) intravenous As a rule, goats and sheep receive wholc scrum or gammaglobulin solution by injection into the scapular or glutcal muscles When dealing with rabbits onc or more routes may be uscd and it has been found that greater success in achieving immunisation will follow thc usc of all three routes for the initial course Rabbits arc rather susceptible to anaphylactic shock when the material is introduced intravenously To reduce this risk an intraperitoneal injection should be given less than 24 hours before all intravenous injections The dosage of gamma globulin required does not appear to be critical The course recommended in the MRC Memorandum No 27 (1952) employs Proom's alumprecipitated gamma globulin (1943) "? Five ml of the solution are injected into each buttock, this injection is repeated after 14 days and 10 14 days later the animal is bled It is claimed that there is less risk of anaphylaxis with this technique When whole serum is being used as antigen 0 5 1 ml of undiluted serum (or of 50 per cent serum diluted with saline) is injected intraperitoneally This is followed within 24 hours by an intravenous injection, into the external peripheral auricular vein, of 0 5 ml of 50 per cent serum diluted with saline Thereafter, for the next 3 4 days, daily injections ·20 ANTI-GLOBULIN (COOMBS) TEST several group o persons The serum pool need not be fresh, but it should be sterile THE PREPARATIOV OF GAMMA GLOBULIN (Proom's (1943) 17 method) 1 Dilute 25 ml of pooled group O serum with 80 ml of distilled water 2 Add 90 ml of 10 per cent aqueous solution of potassium alum (K Al(SO2) 12H 0) 3 Adjust the pH of the mixture to 6 5 by adding 5N sodium hydroxide The pH can be determined by placing a drop of the mixture on a tile and mixing with it a drop of Universal Indicator (obtainable from British Drug House London) Since the pH change is sudden the caustic soda should be added carefully drop by drop 4 Centrifuge the mixture and remove the supernatant fluid 5 Wash the precipitates twice with 200 ml of a 1 10 000 saline solution of Merthiolate (El Lilly & Co) 6 Make the washed precipitate up to a volume of 100 ml by adding a further saline solution of i 10 000 Merthiolate This suspension can then be stored for at least 14 days at 4 C Repeated injections of alum precipitated globulins are apt to produce sterile abscesses in the muscles and sometimes a paralysis of the hind quarters For this reason it is recommended that although precipitated globulins may be used initially, once the animal has become immunised pooled whole serum should be used for further boosting of the antibody Satisfactory anti globulin sera can be prepared without the use of an adjuvant though some workers feel that they ·THE PRODUCTION Or ANTI-GLOBULIN SERA 21 19 obtain a better or quickcr response in an animal which has been made slightly febrile by the injection of a "forcign protein Kidd (1957) u mixes dried tubcrclc bacıllı with gamma globulin and cmulsifics the mixture with Arlace! A (Mannide mono-olcatc) With this reagent good antiglobulin scra arc produced with low titre anti human antibodics THE IMVIUNISING INJECTIONS The routes by which the antigen may be introduced are (a) intramuscular, (6) intraperitoncal, or (c) intravenous As a rule, goats and sheep receic whole scrum or gammaglobulin solution by injection into the scapular or glutcal muscles When dealing with rabbits one or more routes may be used and it has been found that grcater success in achieving immunisation will follow the usc of all three routes for the initial course Rabbits are rather susceptible to anaphylactic shock when the material is introduced intravenously To reduce this risk an intraperitoneal injection should be given less than 24 hours before all intravenous injections The dosage of gamma globulin required does not appear to be critical The course recommended in the MRC Memorandum No 27 (1952) 8 employs Proom's alumprecipitated gamma globulin (1943) 17 Five ml of the solution are injected into each buttock, thus injection is repeated after 14 days and 10 14 days later the animal is bled It is claimed that there is less risk of anaphylaxis with this technique When whole serum is being used as antigen 0 51 ml of undiluted serum (or of 50 per cent serum diluted with saline) is injected intraperitoneally This is followed within 24 hours by an intravenous injection, into the external peripheral auricular vein of O 5 ml of 50 per cent serum diluted with salıne Thereafter, for the next 3 4 days, daily injections ·22 ANTI-GLOBULIN (COOMBS) TEST of 1 ml serum are given either intramuscularly or intra venously Slavin (1950) 18 has advocated a method of immunisation by slow absorption A mixture is made of 1 ml sterile group O serum with 4 ml of a sterile 4 per cent sodium alginate solution ("Calgtex® " obtainable from Medical Alginates Ltd, Wadsworth Road, Peruvale Middlesex) The mixture is injected intraperitoneally into a rabbit and immediately afterwards a second injection, of 2 5 ml of a sterile 1 per cent aqueous solution of calcium chloride, is made from a separate syringe into the same site The calcium chloride reacts with the sodium alginate to form a rubbery lump from which the serum is liberated slowly as the mass breaks down Once the animal has become immunised it should be left as long as possible before booster injections are given Stratton 1' observed that the longer the animal was left the greater was the titre of the antibody after boosting and acting on his suggestion to obtain the best results the next course of injections should be given only after three months or more However, the antibody evoked by the initial course is likely to be mainly anti human gamma globulin and since other anti globulin antibodies may be required repeated courses may be necessary before the animal is given its rest period The booster course · recommended for an immunised animal is 1 One ml sterile pooled o serum is injected intraperitoneally 2 Within the next 24 hours 0 5 ml of 50 per cent sterile pooled O serum in salıne is injected intravenously 3 On the third and fourth day 1 ml sterile pooled O serum is given either intravenously or intramuscularly 4 10 14 days later bleed the animal ·THE PRODUCTION OF ANTI-GLOBULIN SERA 23 BLEIDING Tuc ANIVAL The best time to bleed the animal is 10 14 days after the completion of thc course of injections though small amounts may be repeatedly removed between thc 10th and 30th day's Only a small sample for cramination should be taken from an animal which has completed its initial course of immunising injections and tests on this this will show whether further bleeding would be justified When dealing with goats and sheep the blood is collected from the external jugular vein after thc hair or wool has been clipped from the neck The help of an assistant, who can manæuvre the beast into a suitable position and hold it firmly there during the operation, is invaluable The animal is backed into a corner, the assistant stands astride its back, gripping its flanks firmly with his knees, he then raises the animal s head upwards and slightly backwards and holds it there The operator locates the line of the external jugular with his finger tips, cleanses the skin over it with spirit and inserts a long, large bore sterile necdle (20 cm by 3 mm external diameter) in an upward direction The blood is allowed to run straight from the needle into a sterile collecting bottle When 400-500 ml have been collected the needle is withdrawn from the vein and the bleeding is stopped by firm digital pressure over the site for 4 S minutes Meanwhile a sterile rubber wad and metal cap have been screwed on to the bottle of blood which is placed in the incubator at 37° C until clotting has taken place As much serum as possible is aseptically removed and then the clot is refrigerated at 4° C overnight On the next day the remainder of the serum is separated from the clot The second lot of serum is sometimes rather hæmolysed Rabbits may be bled by heart puncture or by nicking the external auricular vein In the hands of those who are not ·24 ANTI GLOBULIN (COOMBS) TEST experts cardiac puncture seems rather brutal and, since a satisfactory yield can be obtained from the ear vein, this latter method is usually preferred The rabbit is immobilised either in a suitably designed box, or in the hands of a trained assistant The hair of the ear is shaved and the bare area over the vein is cleaned with alcohol A small smear of petroleum jelly round the site and over the edge of the ear will prevent the blood from clotting before it has been collected The vein may be punctured with a small needle or a small longitudinal shit may be made with a scalpel and the blood is allowed to drip into a small bottle In this way 40 ml of blood can be obtained Bleeding is stopped by firm pressure if clotting has not already taken place The blood is allowed to clot at 37° C, is kept at 4 C overnight and the serum is separated from the clot on the following day INACTIVATION OF THE SERUM Immediately the serum has been separated it must be heated to 56° C and maintained at that temperature for notless than 30 minutes in order to inactivate the complement, otherwise the serum will hæmolyse red cells Large amounts of serum, which have been kept at 4 C will require a certain time before the whole volume reaches a temperature of 56 C and allowance should be made for this in addition to the 30 minutes required for proper inactivation In practice it has been found that even one hour at this temperature does not harm the anti globulin antibodies The next step is to test the anti globulin serum for freedom from hæmolysins A mixture of 0 02 ml of the serum with 0 02 ml of a 10 per cent suspension of fresh human group O cells in saline is incubated at 37° C for 2 hours The cells will be agglutinated by the species specific antibody but they should not be hæmolysed If ·TIIE PRODUCTION Or ANTI-GLOBUI IN SERA 25 htmolysis docs occur the scrum must be further hcated at 56° C) Once the scrum has been freed from litmolysins it may be put through the next stage of preparation, or it may be stored at -20° C until timc pcrmits of further processing' TREATMENT or TIC CLOT The bulk of the anti globulin antibody has been removed with the supernatant scrum, but a surprising amount can be obtained from the residual clot by cutting it into small pieces (4-5 cm cubcs) and by shaking these up in isotonic saline The mixture should be shaken for 5 10 minutes after which it is centrifuged and then the supernatant salınc is removed and retained Thc process may be repeated at least once more These washings contain both anti globulin and a small amount of hetero antibodies which will have to be absorbed out before the reagent can be used The solution may store well if kept at -20° C and can be used as an anti globulin reagent by itself or it can be employed as a diluent for anti globulin sera REFERENCES 1 HILL J M and HABERMAN S 1954 The Coombs (anti globulin) test Indications and technics Am J Clin Path 24 305 2 DUNSFORD I and BOWLEY CC 1957 The production of anti human globulin serum (Coombs reagent) in goats Clin Path 10 29 3 STRATTON F 1956 Production of human anti globulin reagent in Sheep Nature 178 214 4 EYQUEM A 1956 Personal communication to Dr Stratton 5 GOODMAN M RAMSEY DS SIMPSON WL REMP DG BASINSKI DH and BRENNAN MH 1957 The use of chicken anti serum for the rapid determination of plasma proteins J Lab Clin Med 49 151 6 PROOM H 1957 Personal communication 7 COOMBS RRA MOURANT A E and RACE RR 1946 sensitisation of red cells in babies with hæmolytic disease Lancet 1 264 8 MEDICAL RESEARCH COUNCIL MEMORANDUM No 27 1952 The Rh blood groups and their Clinical Effects PL Mollison A E Mourant and R R Race Her Majesty s Stationery Office London In vivo ·26 ANTI-GLOBULIN (COOMBS) TEST 9 WOOTTON IDP 1950 Production of Coombs serum Nature 165 730 10 SINGH J RAMAKRISHMAN S P and PRAKASH S 1952 Human plasma as antigen in the preparation of precipitin serum Nature 169 4291 11 KIDD P 1957 Personal communication 12. KENDALL, J 1937 Studies on serum proteins I Identification of a single globulin by immunological means Its distribution in the sera of normal individuals and of patients with cirrhosis of the liver and with chronic glomerulonephritis J Clin Invest 16 921 13 MILGROW F LUSZCZYNSKI T and DUBISKI S 1956 Preparation of anti globulin sera Nature 177 329 14 KOY TNOS Z D and AKSOY M 1955 Production of Coombs serum by injections of auto antibodies into rabbits Am J Clin Path 25 967 IS PISCIOTTO AY and HIZ JE 1956 Variations in potency of anti globulin sera prepared from serum containing warm and cold agglutinins J Lab & Clurt Med 48 930 16 CHRISTENSEN W N and DACIE J V 1957 Serum proteins in Acquired Hæmolytic Anæmia (auto antibody type) Britſ Hæmat III 153 17 PROOV H 1943 The preparation of precipitating sera for the identification of animal species J Path Bact55 419 18 SLAVIN D 1950 Production of anti sera in rabbits using Calcium Alginate as an antigen depot Mature 165 115 19 STRATTON F Personal communication ·CHAPTER V THE PROCESSING OF ANTI HUMAN GLOBULIN SCRUM THC PRELIMINARI EXAMINATION PREVIOUS chapters have described the immunisation of the animal, the collection of blood from it, thc scparation of scrum from thc blood and the inactivation of complement and destruction of hæmolysins present' Beforc carrying out the laborious process of removing the unwanted antispecies antibodies it is wisc, at this stage, to examine the serum to sec whether it is likcly to be good cnough to warrant further action The exact technical procedure is described in Chapter vi and the reader is advised to make himself familiar with the details and fallacics of the test before undertaking it However, the principle of the preliminary examination is simple "We want to know what are the relative amounts of antı human species antibodies and anti-human globulin antibodies in the animal serum we have collected Serial dilutions (from 1/1 to 1/1024) of the serum are made by adding saline Then, using a clean opal glass tile, each dilution is tested against (1) a 50 per cent suspension in saline of thrice-washed human group o Rh (D) positive red cells, and (2) a salıne suspension of the same washed cells which have been coated with incomplete anti Rh (D) antibodies. The results of testing two typical sera are shown in table i The serum of Rabbit A has agglutinated the untreated red cells at a dilution of 1/4 as shown in the uppermost line of results This is a measure of the quantity of anti human species antibody present The second line shows the dilution at which the same rabbit s serum will agglutinate red cells 27 ·28 TABLE 1 Dilutions of Rabbit Serum in Saline 1 2 8 16 32 64 128 256 S12 1024 Serum from Rabbit A Unsensitised red cells Coated red cells V V (+) + V ANTI-GLOBULIN (COOMBS) TEST V 1 > V ++ ++ 1 + (+) Serum from Rabbit B Unsensitised red cells Coated red cells V ♡ ## +1 + + (+) II ·PROCESSING Or ANTI-IIUMAN GI Onvif STRUV 29 which hasc been coated with antibody and it indicates the presence of a considerablc amount of anti human globulin in addition to the anti human species antibody Tor the scrum to be worth processing the dilution at which it will rcact with coated cells should be at Icast lucnly times thc dilution at which it will clump unsensitised cells. The scrum from Rabbit A would thereforc bc regarded as satisfactory In contrast the scrum from Rabbit B (the lower two lines of table 1), appears to contain lililc antiglobulin antibody and is unsuitablc for ſurther processing Somc laboratories on finding that an animal had produced a good titre of antibody would kill it and obtain the maximum yield of scrum This practice is not only unnecessary but also unwise, since the "spectrum of anti globulin antibody will be less wide than that of an animal which has been further immunised If, after the initial course of immunising injections, a rabbit has failed to develop a satisfactory titre of antiglobulin, it should be rested for 6 to 9 months before being given a second course If the animal again fails to respond sufficiently it should be destroyed or used for some other purpose The sera from animals which have responded inadequately, whether obtained from test samples or after death need not be discarded, they should be saved, pooled and inactivated After removal of the unwanted heteroagglutinins the pool can be used as a diluent for good antı human globulin sera, thus diminishing the deterioration which tends to occur on storage, even at -20° C, of serum diluted with saline Most rabbit anti-human globulin sera can be diluted with an equal volume of such a "neutral pooled serum and sometimes, with several volumes " THE REMOVAL OF UNWANTED ANTIBODIES It will be found, particularly when working with rabbit serum, that human red cells of the separate ABO groups ·30 ANTI-GLOBULIN (COOMDS) TEST are not all agglutinated to the same titre by hetero antibody, since many rabbits have an apparently naturally occurring anti A-type agglutinın and an occasional one will have antibody which simulates the action of anti B The hetero antibodies and anti human species agglutinins can be removed from anti globulin serum by absorption, by dilution or by heat treatment Whichever method is used the final product must not agglutinate fresh, well washed, group A, B and human red cells 1 Absorption Method For this procedure sterile fresh, well washed human red cells of group O and group A,B are required If A,B cells are not available half quantities of A, and B cells may be substituted A suitable anti coagulant in which to collect the blood IS 3 8 per cent of trisodium citrate (2 Na,C,H,O,11H 0) in distilled water mixing one volume of anti-coagulant solution with four volumes of blood The solution is autoclaved and the blood when taken must be kept sterile The blood should be stored at 2 6°C and preferably the red cells used for absorption should be no more than four days old Older red cells more readily break down and are not so efficient in removing antibodies When the blood has been withdrawn it is refrigerated until the red cells have sedimented The next step is to free these cells from all traces of plasma The plasma is aseptically removed replaced with sterile isotonic saline and the red cells and saline thoroughly mixed The mixture is then centrifuged to sediment the cells the supernatant is removed, replaced with more saline mixed and again centrifuged If possible the red cells should be kept sterile and a closed apparatus for the handling of large volumes of blood has been described 1 The cells will require at least six washings to free them from all plasma To determine ·PROCESSING or ANTI-TUMAS GLOBULIN STRUN 31 whether this has been achicied the sulpho salicylic acid test for proteins is uscful Tale 5 ml of the salınc from the last washing and to this add 0 5 ml of 25 per cent sulpho salicylic acid If the washing has been cfficicnt no tracc of turbidity should be seen in the mixture The washing process is time consuming but must bc thorough, since traces of plasma present may contain sufficient globulin to ncutralisc thc anti globulin antibodies in the animal scrum during absorption. Furthermore, the cells of cach donor must be washed scparately before they arc mixed sincc reactions can occur between the plasma of onc blood and the red cells of another which causc the agglutination of the red cells At last we come to the actual absorption procedure which is performed in two stages Thc anti human globulin serum is, firstly, well mixed with an equal volume of the group O washed packed cells and is then stored overnight at 4°C during which time anti species antibody is adsorbed to the cells The O cells are then removed and discarded and a second absorption is carried out with A,B (or A,+B) cells After this absorption the anti-globulin serum is tested for reactions against A, B and O cells If hetero antibody has been incompletely removed further absorptions will be necessary, but they should be kept to a minimum since, as shown by Stratton and Jones (1955) * even well-washed cells will reduce the titre of anti-globulin antibody 2 Dilution Method When an animal has developed in its serum an antihuman globulin antibody of high titre together with a comparatively small amount of antı human species antibody, it may be possible so to dilute the serum with salıne that it no longer agglutinates well washed human A, B and O cells, but it still retains its capacity to clump red cells coated with incomplete antibody globulins The disadvantage of this ·32 ANTI GLOBULIN (COOMBS) TEST method is that, although the diluted serum may be a satisfactory reagent with which to detect the presence of those incomplete antibodies which are gamma globulins, it may, because of the dilution, dangerously fail to demonstrate the presence of incomplete antibodies which are non gamma globulins Since serum mixed with saline deteriorates on storage the dilution should not be made until required for use 3 Heat Treatment "In 1950 van Loghem: showed that hetero agglutinins could be destroyed by heating the animal serum to 63º C for 1-2 hours Subsequently other workers have varied the time and temperature required Menolasino and Davidsohn (1954) * found exposure of the serum to 70 C for 30 minutes to be effective, Hunter and Thomas (1957) 5 discovered that a temperature of 63°C for 30 minutes was sufficient for some sera' However these authors and later Wilson, Simmons and Semple (1958) 6 have demonstrated that heat treatment may impair the serum s ability to react with some globulins To avoid coagulation of the protein by these temperatures, the serum should be mixed with an equal volume of saline before heating This method has certain advantages in a laboratory where the supply of human red cells for absorption is meagre or when the volume of serum is small or when the loss of anti globulin antibody during absorption would be too great THE PREPARATION OF ANTI GLOBULIN SERA FOR OTHER ANIMAL SPECIES Only the absorption of antı human giobulin sera has so far been mentioned, but anti globulins for other species can be prepared by an analogous method For example if a rabbit antı goat globulin serum is required the inactivated serum from a rabbit which has been immunised with goat serum is absorbed with well washed goat red cells ·PROCESSING OF ANTI JIUMA GLORUIIS SERUM 33 RITERICIS I DUNSTORD 1 and Bonuy C. C 1955 Techniques in Blood Grouring Oliver and Boyd Edinburgh 2. STRATTON r and Jon AR 1955 * The rcaction beturen normal human red cells and anti globulin (Coombs ) stum Immunol 75 423 3 VAN LOGITU JI 1950 Serological investigations of the anti human globulin serum (serum of Coombs) with special reference to the pathogenesis of erythroblastosis falalis Maandschrift roor Aindergeneeskunde XVIII 3/4 115 4 MENOLASINO NJ and DAVIDSOIN I 1934 A simplified mcthod for preparation of antı human globulin (Coombs') scrum Am J Clut Path 24 1205 S HUNTER, D A and Thomas A R 1997 The cliects of heating antı human globulin seta J Clint Path 10 2445 6 WILSOV MM SIMMONS R. T and SEMPLE NM 1958 The heat lability of the anti human globulin in Coombs reagent used to detect incomplete Brucella human antibody Austral y Science 20 9 275 ·CHAPTER VI STANDARDISATION OF ANTI HUMAN GLOBULIN SERUM The anti globulin serum which has been prepared will be used principally for the detection of incomplete antibodies of various titres and specificities Therefore, before this reagent is used it is necessary to know whether it will disclose the presence of a weak antibody, whether it is capable of demonstrating the existence of a range of different antibodies, and at what dilution it is best used In blood transfusion laboratories the antibody most commonly encountered is the Rhesus antibody, antı D, and it is essential therefore in these laboratories, to determine the optimum dilution of the anti globulin which will agglutinate group o Rhesus positive red cells which have been sensitised by a weak incomplete antı D serum It must be remembered, however that what is an optimum dilution for the disclosure of Rhesus antibodies may be too dilute to demonstrate an antibody against another red cell antigen eg antı Kidd Furthermore the anti globulin serum may only be able to detect Rhesus antibodies and may be quite useless for the detection of other incomplete antibodies, ie the serum lacks anti globulins to these particular antibodies A more complete explanation of this is given in the next chapter Ideally, the anti globulin serum should be standardised for each type of antibody which it will be required to detect' In practice this is hard to achieve because of the difficulty of obtaining different types of antibodies for coating red cells A schedule of proven worth followed in the Sheffield 34 ·45 type of ANTI-HUMAN GLOBULIN SERUM 35 laboratory, is to test cach dilution of the anti globulin scrum against 1 Rhesus positic cells sensitised with a sclcctcd scal, incomplete anti D antibody 2 Rhesus positive cells of type CDe/cDE sensitised with Weal, incomplctc anti E anubody 3 Kell positive red cells scnsitised with wcak anti-Kell antibody (this is a sclected non gamma antibody, see next chapter) 4 A, cells coated with incomplete anti-A antibody 5 If available, red cells from a case of the "cold antibody" type of acquired hæmolytic anæmia 6 If available, red cells sensitised with a weak anti-Duſſy antibody From the results obtained we hope to find two dilutions of the anti globulin serum at which all these antibodies have been detected, and it is then used at these dilutions only If we find the anti globulin is only reliable for the detection of Rhesus antibodies it is either pooled with another serum which will detect all types of antibodies, or it is only used in parallel with another anti globulin serum which will detect these antibodies Anti globulin sera available may be supplied undiluted or they may have been diluted before issue If the raw serum is supplied it is customary to show on the label or on the accompanying report the dilution for the detection of Rhesus antibodies only. The user of such serum may be upable to standardise the serum for the detection of other antibodies In this case the problem is overcome by employing serial dilutions of the anti globulin serum or, in a less certain manner, by using a dilution of the antiglobulin serum which is about one fifth of that recommended for the detection of Rhesus antibodies If it is known that the serum issued is already diluted, or if the supplier states that it is suitable for use without ·36 ANTI-GLOBULIN (COOMBS) TEST further dilution, then the user should be informed or should determine that the reagent is suitable for detecting all types of antibodies One of us (I D) has examined some commer cially available anti globulin sera and found they have been so diluted that they will only detect Rhesus antibodies No indication of this was given in the literature issued with the reagent, and the usage of such materials can have dangerous consequences in blood transfusion work METHOD OF STANDARDISATION For this test coated cells are required which have been subjected to varying degrees of sensitisation Each portion of these cells is then mixed with a variety of dilutions of anti globulin serum and the results of the interaction between the cells and serum are noted A The preparation of the coated cells 1 Select a weak, incomplete anti D serum which will agglutinate cells suspended in albumin at a titre of 1/16 to 1/32 but which is not active at greater dilutions 2 In tubes (9 mm by 75 mm) prepare serial dilutions (1/1 to 1/64) of this anti-D serum so that I ml of each dilution is available 3 To each dilution add 0 5 ml of a 50 per cent suspension in saline of washed packed, group o Rh positive cells 4 Incubate each mixture at 37 C for 30 minutes 5 Centrifuge each tube remove the supernatant and add at least 20 volumes of saline, flushing the cells up from the bottom to ensure thorough mixing 6 Centrifuge again at 800 1000 rpm to sediment the cells, remove the supernatant saline and replace with more saline until the cells have been washed at least three times 7 Prepare 50 per cent suspension in saline of these washed cells ·1 ANTI-HUMAN GLOBULIN SERUM 37 8 Prepare also 50 per cent suspensions of thricc-washed A, B and Ocells which have not been exposed to any antibody B The preparation of serial dilutions of the anti globulin serum Prepare serial dilutions of the anti-globulin scrum in saline from 1/1 to 1/1024 or morc, according to the titre which it is expected to reach, so that 0 5 ml of cach dilution is available c The interaction between cells and serumi 1 A large white opal glass title measuring 20 inches square is well cleaned with soap and hot water to remove any traces of serum or globulin left by previous tests After drying with a clean cloth, kept specially for this purpose, the tile is marked out into the squares as shown in fig 7 2 Place one drop of the anti globulin serum in each square of the vertical columns so that every dilution can be tested against progressively less sensitised cells and against the non sensitised A1, B and O cells In the last column drops of saline are used in place of anti globulin serum 3 Moving from left to right add one drop of sensitised cells (and then the unsensitised cells) to the squares containing each dilution of the anti globulin serum Carefully rinse the pipette in saline before taking up each new batch of cells 4 Mix each lot of cells and serum with the end of a clean plastic rod, rinsing and drying the rod between each mixing A quicker alternative is to use short lengths of wooden swab stick to stir the mixture and to discard each one as it is used It is advisable to have two people doing the mixing together in order to save time and to prevent the drops from drying too much before the results can be read ·38 ANTI-GLOBULIN (COOMBS) TEST 5 After 1-2 minutes (timed by the clock) rock the tile very gently back and forth and from side to side Let it stand for a further minute and then repeat the rocking 6 After it has stood for 5-7 minutes the results should be read Check the absence of agglutination with a hand lens in all apparently negative reactions AHG Dilutions Cells and 2 632 64 128256512 Saline 1 4 og 1 2 4 R.BCS sensitised with anti D diluted 8 16 32 64 A Non sensitised cells B 0 FIG 7 THE INTERPRETATION OF RESULTS When a rabbit has been immunised for the first time the results of the anti globulin serum standardisation test are likely to be of the pattern shown in table i The absence of reaction between the antı human globulin serum and the unsensitised A, B and O cells confirms that the anti species antibody has been satisfactorily absorbed No reaction should occur between the cells and saline, thus establishing that the anti D serum is truly incomplete Provided that these control tests are satisfactory examination of the remainder of the results will indicate to what degree the ·TABLE 11 AHG Dilutions Cells and 312 Saline 1 8 16 128 256 4 32 512 fo+ + +++ ---- +++ + + + $ +++ +++ ++ RBC's sensitised with anti D diluted +++ Feel f+ ANTI-HUMAN GLOBULIN SCRUM *++f!!! !!! FI+ SA +FCO 1 fe - Non sensitised cells O 39 ·40 ANTI GLOBULIN (COOMBS) TEST anti globulin serum can be diluted The results shown in table i suggest that this serum would prove satisfactory at a dilution of 1/8 to 1/16 for the detection of anti D antibodies THE STANDARDISATION FOR OTHER GLOBULINS For this standardisation, each dilution of the anti globulin serum is tested against cells coated with the same amount of antibody The type of result obtained is given in table II THE ZONING PHENOMENON A second pattern of reactions is seen in tables iv and v This picture is obtained by using anti globulin sera from much immunised rabbits though it occurs less commonly in goat sera For this reason Hill and Haberman (1954) 1 suggest that' further studies should be made to determine if the goat may not be superior to the rabbit with respect to the quality as well as the quantity of anti globulin sera produced "The configuration of results in tables iv and v shows what is known as the zonmg phenomenon - Its occurrence in anti globulin sera was first described by van Loghem (1950)* and was later reported by other workers van Loghem, Kresner, Coombs and Fulton Roberts (1950) 3 and Hubinot (1951) * The lack of reaction where the antibody is used neat or in low dilution is known as the pre zone or prozone In the next few tubes where agglutination does occur is the reaction zone and subsequently we reach the point where no observable reaction has occurred the post zone : Certain findings suggest that the prozone phenomenon could be due to a relative excess of antı human globulin antibody over globulin on the surface of the red cells For example of additional sensitised red cells are added to a mixture of unagglutinated sensitised cells and antı human globulin serum in the prozone area agglutination takes place Also if we add a small amount of a very dilute ·TABLE M Dilutions of AHG Batch Mo 1093 2 8 16 32 64 128 256 500 1000 2000 ++ t + ++ + Cells sensıfısed with Anti Duffy (Grimes) Anti Lewis (Le 1556) Incomplete immune anti A++ Antı Kell (1370) Ft + I+II ANTI-HUMAN GLOBULIN SCRUM + ++ + t 11++ +++++++ ++++ ++ +++++ ++ T ++ Cells not sensitised MO 41 ·42 ANTI-GLOBULIN (COOMBS) TEST solution of gamma globulin to the unagglutinated sensitised red cell mixture, thereby lessening the amount of anti globulin present, clumping will occur The weaker the cell suspension the more marked is the prozoning and this is one of the reasons why, in performing the Direct antı globulin test, it is necessary to have a thick cell suspension at least (50 per cent cells in saline) to reduce the risk of false negative reactions There may also be another explanation of the prozone It can be shown experimentally that the unagglutinated sensitised cells of the prozone area are, in fact, doubly coated with an inner layer of incomplete anti D and an outer covering of rabbit antı human globulin If these unagglutinated cells are washed three times with large volumes of salıne and tested with goat anti rabbit globulin serum they are agglutinated Since anti rabbit globulin sera caused this, it shows that the cells are coated with rabbit globulin and this must be the rabbit antı human globulin, and this in turn must be attached to the Rhesus antibody which in turn is attached to the red cells This would suggest that the prozone may be due to incomplete antibodies in the rabbit anti globulin which coat the red cells and block the reactions of complete antibodies present in the anti globulin serum It can be further inferred that dilution of the anti globulin serum with saline dilutes the incomplete antibodies so that they no longer block the action of the complete antibody It is obvious from studying the results in tables iv and v that either explanation applies only to the anti globulin antibodies detecting Rhesus antibodies, since the other antibodies are detected with less diluted anti globulin serum "Therefore this would suggest that a zoning anti globulin serum may contain at least three types of antibodies- 1 complete and incomplete anti globulin detecting Rhesus antibodies and complete anti globulins reacting with other 2 ·TADLE IV Dilutions of Anti Human Globulin Scrum | Suline and 32 128 256 312 1024 2048 4096 8192 16384 (RDCs 4 8 16 64 +- + FHI RBC.S sensitised with anti D diluted +ILIT ++!!!!! - ++ + + #+fil #*+IMIT fi+ful 11++++ 11 +++++ ENT non sensitised cells ATI-IIUMAN GLOBULIN SERUM TABLE V Dilutions of Anti Human Globulin Scrum 16 32 64 128 256 1 4 2 8 312 2013 1024 — + - + + + + + + + + + + + + + + + + Washed sensitised RBCs Reaction Zono Pre-zono or Prozone Post zono 43 ·44 ANTI-GLOBULIN (COOMBS) TEST antibodies such as certain antı Kells In the next chapter further evidence will be given that anti human globulin serum contains more than one type of antibody REFERENCES 1 HILL JM and HABERMAN S 1954 The Coombs (Anti globulin) test Indications and technics Am J Clin Path 24 3 305 2 VAN LOGHEM J J 1950 Serological investigations of the antı human globulin serum (serum of Coombs) with special reference to the pathogenesis of erythroblastosis fætalis Maandschrift voor Kindergeneeskunde XVIII 3/4 115 3 VAN LOC HEM J J KRESNER M COOMBS RR A and FULTON ROBERTS 1950 Observations on a prozone phenomenon encountered in using the anti globulin sensitisation test Lancet 1 729 4 HUBINOT PO 1951 The prozone phenomenon in hæmagglutination reactions with special reference to the quantitative anti globulin test Brit Med J 1708 ·CHAPTER VII COMPLEXITIES OF THE ANTI GLOBULIN SERUM 1 So far human globulin has been frequently referred to as though it were a single cntity and only a brief indication given that anti globulin sera have antibodies against different types of globulins By electrophoretic or chemical means alpha- beta- and gamma-globulin fractions can be distinguished in serum, and still further subdivisions of these fractions can be made Coombs and Mourant (1947) established that Rhesus antibodies were mostly, if not entirely, in the gamma globulin fraction of the serum That being so, an antı human globulin serum, optimally diluted for the detection of anti-Rh antibodies, will also demonstrate the presence of other antibodies which are gamma globulins but it may fail to reveal the existence of antibodies which are non gamma globulin in nature - An illustrative example is shown in table vi An antihuman globulin serum was tested against Rhesus positive cells coated with incomplete anti-D antibody and the optimum dilution proved to be 1/128 When the same anti globulin serum was tested against blood from a patient suffering from acquired hæmolytic anæmia, whose cells were coated with cold" type antibody, the agglutination pattern was so different that, had the anti globulin serum been used only at a dilution of 1/128, it would not have demonstrated the fact that cells were coated with antibody Dacie (1950), Crawford and_ Mollison_(1951) 3 and Renton (1952) * reported observations which showed that anti-globulin sera could be absorbed with one type of sensitised cell without removing the activity against other types of sensitised cells Using this technique, they demonstrated that anti globulin sera had at least two types of 45 ·46 TABLE VI AHG Dilutions Cells and Saline 1 2 4 8 16 32 61 128 256 512 1024 ANTI-GLOBULIN (COOMBS) TEST + + + + + + + + AHA cells Rh coated cells Non sensitised cells Il + (+) (+) 1 +1 + + + + + ++ ii - 1 ·COMPLEXITIES OF TIIE ANTI-GLOBULIN SERUM 47 antibodies anti gamma globulins reacting with Rhesus antibodies and some others, and anti non gamma globulins which failed to react with Rhesus antibodies but would detect other types The techniques which these authors describe can be used to preparc anti globulin sera that have either no anti gamma globulins or no anti non gamma globulins Renton (1952) referred to the anti-gamma type as Fraction I and the anti non gamma as Fraction II The use of this terminology is confusing since preparations of human gamma globulin are often referred to as Fraction II and, therefore, the terms anti-gamma and anti non gamma are preferred The antı gamma globulin titres are invariably higher than the anti non gamma globulin titres An anti globulin serum capable of detecting all types of globulins is sometimes referred to as a "wide_spectrum reagent' In Chapter iv various methods of production of anti globulin sera using different antigens were described When injections of whole serum are used as antigen the animal is likely to develop both anti gamma and anti nongamma globulins' Even when gamma globulin is used as antigen the resultant animal serum may have anti nongamma globulins, suggesting that the gamma globulin was not sufficiently pure Possibly in the future purified globulins may be used as antigens and specific anti-sera prepared A different approach to making specific anti sera was used by Cutbush Crawford and Mollison (1955) They studied the effect of the addition of dilute solutions of various globulin fractions (alpha, beta, gamma and gamma, globulins) to anti-globulin sera, and with these neutralised sera they tested cells coated with a variety of different incomplete antibodies. They also obtained, from a patient suffering from a condition known as agammaglobulinæmia, a serum which contained both alpha and beta globulins, but no gamma globulins This serum when added to antı globulin sera gave them reagents lacking in certain types of ·48 ANTI-GLOBULIN (COOMBS) TEST anti globulins and these were tested against red cells coated with various types of antibodies Their results confirmed that incomplete anti Rh antibodies, the "warm type of incomplete antibody of acquired hemolytic anemias, some examples of anti Kell and some antı Duffy antibodies are gamma globulins, and some are probably in the gamma, globulin fraction of the serum The 'cold type antibody of acquired hæmolytic anemia, the cold incomplete antibody of normal serum, Dacie (1950) - and some antı Lewis antibodies are contained in the alpha and beta globulin fractions Other anti Lewis and antı Duffy antibodies were found to be a mixture of gamma and non gamma types op The role of complement in the anti globulin test has been previously mentioned and several workers have shown that some substance, possibly complement, in fresh serum is necessary for the coating of non gamma globulins to be detected An interesting experiment is to expose Lewispositive cells to an anti Lewis serum lacking in complement The cells are then washed three times and divided into two lots One portion is tested with an antı non gamma globulin serum and no reaction will be observed The other portion is incubated for 30 minutes at 37 C with a selected fresh human serum after which it is further washed three times and tested with the same antı non gamma globulin serum when agglutination will be observed Dacie Crookston and Christensen (1957) ® consider that the components of complement required are C1, C2 and C4 but not C3 These findings underline the fact that unless one is using fresh incomplete serum for sensitising the red cells for the Indirect anti globulin test a non gamma globulin antibody may escape notice If the sensitising serum has been stored for more than a couple of days it may be necessary to add to it a volume of fresh human serum to ensure a sufficiency of complement Preferably from a person whose red cells are Lea--b-> ·COMPLEVITIES OF THE ANTI-GLOBULIN SERUM 49 Enough has now been written to make clear to the serologist the importance of knowing and appreciating the capabilities and limitations of cach anti globulin scrum with which he has to work He should be aware of the antiglobulin serum which has been rid of its hetero antibodies by dilution, or by heat treatment, since such manipulations may result in a serum which will detect only coatings of gamma globulins Three methods can be employed with anti-globulin sera to ensure detection of both gamma and non gamma globulins The first is to use serial dilutions of the anti globulin reagent, say from 1/1 to 1/512—the serial dilution anti globulin test This technique is tedious and involves a lot of unprofitable work but it is necessary sometimes The second method is to use the anti globulin sera in two dilutions, one at the optimum dilution for the detection of gamma globulins and the other at the dilution for detecting non gamma globulins For this to be a reliable technique the results of the standardisation test must be known though in practice the dilution required to detect gamma globulins is usually found to be about four to six times that needed for the detection of non gamma globulins The third method is one used by some commercial firms Two different anti globulin sera are selected, one of which must be capable of detecting the non gamma globulins, the other can be one containing anti gamma globulins only the sera from goats and rabbits can be mixed The optimum dilutions for detecting the two types of globulins are determined The serum containing the anti gamma globulin is then diluted to its point of optimum activity with saline or saline washings from the clots (obtained as described), or with a mixture of salıne and neutral (i e free from heteroantibodies) animal serum "This anti gamma globulin serum can then be used as a diluent for the other antı non gamma globulin serum in such a proportion that it will give a E ·50 ANTI-GLOBULIN (COOMBS) TEST reagent which will detect both gamma and non gamma globulins at the same dilution The prepared reagent can be issued already diluted, or by adjusting the volumes used in its preparation a single dilution factor can be recommended which will give a satisfactory reagent “The simple calcula tion necessary for preparing 100 cc of reagent requiring no further dilution is given for guidance Serum A-Dilution at which gamma globulins are detected-1/100 Serum B-Dilution at which non gamma globulins are detected—1/10 100 Amount of serum A required Х =lcc 100 Amount of serum B required 1 10 Х 100 T 10 cc Amount of diluent required 100--11 -- 89 cc ANTIBODY ANALYSIS BY GEL DIFFUSION TECHNIQUES Agar diffusion techniques (Ouchterlony 1949' and Oudin, 1952 %) show great promise as a means of analysis of anti globulin sera Vaughan (1956) has reported on the analysis of anti globulin sera by these techniques REFERENCES 1 COOMBS RRA and MOURANT AE 1947 On certain properties of anti sera prepared again t human serum and its various protein fractions their use in detection of sensitisation of human red cells with incomplete Rh antibody and on the nature of this antibody J Path Bact 59 105 2 DACLE J V 1950 Occurrences in normal human sera of incomplete forms of cold auto antibodies Nature 166 36 3 CRAWFORD H and MOLLISON PL 1951 Demonstration of multiple antibodies in anti globulin sera Lancet 11 955 4 RENTOY PH 1952 Separation of Coombs reagent into two fractions Mature 169 329 5 CUIBUSH M CRAWFORD H and MOLLISON PL 1955 Observations on antı human globulin sera Brir Hemat 1 410

·COMPLEVITICS OF TIIE ANTI-GLOBULIN SERUM S1
6 DACIE J V CROOKSTOV I I and CIRISTENSEN W N 1957 . In
complete cold antibodies role of complement in sensitisation to anti
globulin serum by potentially hemolytic antibodies Brit ) Jamal
UL 77
7 OUCITRLOVY O 1919 Antisen anubody reactions in gels Arh
Xem An Gcơ B 261
8 Oudry J 1952
Specific precipitation in gels and its applications to
immuno-analysis Methods Ved Res 5 335
9 VALGHAY JH 1956 Immunologic features of crythrocyte sensitisa
tion 1 Acquired hemolytic disease Blood x 1085

·CHAPTER VIII THE TECHNIQUE OF THE ANTI GLOBULIN TEST The details of the test to be described are those used in human blood group and transfusion serology and in the study of human disease The principles are equally applicable to investigations in other animal species and in bacterial immunology To the uninitiated, the many published accounts of the use of the test for the detection of antibodies, the study of red cell antigens, the investigation of post transfusion reactions, the diagnosis of hæmolytic disease, or in forensic work, may suggest that a great variety of methods are used Furthermore laboratory workers, speaking of the Indirect' and Direct' tests increase the impression that they are different techniques, but this is not so' Basically, the anti globulin test is performed in two stages * In stage one, cells are exposed to the action of incomplete antibody in such a way that the antibody becomes attached to the corresponding antigen on the red cell surface, 3'ın stage two, the anti globulin serum is employed to demon strate these incomplete antibody molecules present on the cell surface The difference between the Indirect and Direct methods rests solely on the fact that, in the Indirect technique the first stage is performed in a test tube, whereas, in the Direct technique the first stage has already occurred in the body before the blood sample was collected For the sake of simplicity the technique of the anti globulin test will be described in terms of the basic routine procedure adopted for the detection of incomplete anti D ---the antibody most likely to be encountered in hospital practice The reasons for certain recommended actions and S2 ·THE TECHNIQUE OF THIC ANTI GLOBULIN TEST 53 the various modifications of rcigents, which may bc uscful under different circumstances, will be outlined later THE INDIRECT ANTI GLOBULIN TEST I Four drops of the serum suspected of containing incomplete anti D antibody are placed in a test tube 2 Two drops of a 50 per cent suspension in salınc of known' group O D positive red cells arc pipetted into the same tube and are mixed with the scrum 3 Control tubes are also prepared Control (1)-the positive control-contains a mixture of the same group o D positive cells with a serum known to contain incomplete anti D Control (11)—the negative control-contains a mixture of the same cells with a serum known to contain no antibodies 4 The three tubes, one containing the suspect scrum and the two containing the control mixtures, are incubated for 11 hours at 37° C 5 Towards the end of this period remove a small amount of the sedimented cells to a slide and examine micro scopically for any agglutination which would signal the presence of complete antibody 6 Provided that no agglutination has taken place, remove the supernatant serum and saline from the remaining cells in all three tubes The next step of the Indirect test is the first stage of the Direct test THE DIRECT ANTI GLOBULIN TEST The red cells with incomplete antibody molecules adherent to their surfaces must be washed completely free from serum so that the only globulin remaining is that of the antibody If the Direct test only is being done positive and negative control cells must be prepared at this stage ·54 ANTI-GLOBULIN (COOMBS) TEST 1 To the sedimented cells add a large quantity of sterile saline, flushing up the button of cells and dispersing them throughout the saline 2 Centrifuge the cell saline mixture at 800 1,000 rpm for 1-2 minutes 3 Remove the supernatant salıne by pipette or by suction from a water pump The more expert, after a bit of practice can empty the salıne from the tube by a quick flick of the wrist leaving the button of cells in situ at the base of the tube 4 Repeat the washing twice, or preferably thrice, and after the final spinning remove every drop of supernatant saline 5 During the intervals of cell washing an opal glass tile is thoroughly cleaned dried and marked out in one inch squares with grease pencil as shown in fig 8 FIG 8 Positive Control Negative Control Test Cell Suspension Control 1 AHG plus Sensitised cells 2 AHG plus Untreated cells 3 AHG plus Test cells Test Cells plus Saline 6 Using clean apparatus and unsoiled salıne dilute the anti globulin serum according to the instructions applicable to the particular batch in use 7 Place one drop of anti globulin serum in squares 1 2 and 3 In square 1 add one drop of cells which have been sensitised with known anti-D To square 2 add one drop of the cells which were mixed with inert serum Squares 3 and 4 each receive one drop of the cells under test and in square 4 a drop of saline is put in place of the anti globulin serum 8 Mix the contents of each square thoroughly spreading them over an area of about of an inch ·THE TECHNIQUE OF THE ANTI-GLOBULIN TEST 55 9 As soon as the mixing is complete start a stop-watch After 1-2 minutes rock the tile icry gently and replace on the bench Repeat the rocking, again very gently, after a further minute or two and then leave standing until ready for reading 5 7 minutes from the time of mixing 10 Read macroscopically with the opal tile held over a Light All apparently negative results should be scrutinised with a (x 10) hand lens All controls must be read first SEROLOGICAL POINTS TO BE NOTED "The Test Serum At the beginning of the test the serum suspected to contain incomplete antibody must be freed from its own native cells If the serum is likely to contain antibodies in the non gamma fraction complement must be present and this can be affected by the addition of fresh AB serum to the serum undergoing test' This point was emphasised on one occasion when a hospital technician, searching for compatible blood for a patient, discovered an antibody by the Indirect anti globulin technique This same test, repeated 24 hours later in the reference laboratory, did not disclose the antibody until fresh serum was obtained from the patient and was tested within 8 hours "The antibody proved to be anti Lea and, as shown by Mollison and Cutbush (1955) could have caused transfusion reaction Cell Suspensions The best results will be obtained with fresh unrefrigerated red cells Cells which have been stored for a time in the cold may have become coated by cold incomplete autoantibody present in the donor's serum and when they give ·56 ANTI GLOBULIN (COOMBS) TEST a positive anti globulin reaction it may be wrongly assumed that the test serum contains antibodies When selecting blood for transfusion, which has of necessity been stored at a temperature of 46 C, one cannot avoid the use of stored, refrigerated cells however, sodium citrate has an anti complementary effect which prevents coating of the cells by cold incomplete auto antibody This is one of the reasons for the use of citrate in the pilot tubes of some blood transfusion services On the other hand, this same anti complementary effect of citrate may mask the auto antibodies of acquired hæmolytic anæmia and Roth (1954) ? recommends that blood samples from such patients should not be collected into anti coagulant' To avoid false reactions and to give more clear cut results, cells must be washed free from their own serum or plasma before admixture with the sensitising serum An important modification of the test was the intro duction of the use of enzyme treated cells Unger (1951) 3 and Unger and Katz (1952) * applied the anti globulin test to cells which had, as a preliminary, been treated with trypsin or ficın and it was by the application of this same technique to trypsinised cells that Plaut Ikin, Mourant, Sanger and Race (1953) 5 were led to the discovery of the antibody anti Jkb However as demonstrated by Mollison and Cutbush (1955) some antibodies such as antı Lea and anti-P may cause lysis of enzyme treated cells when the cell serum mixture is incubated- "During the process of sensitisation it is recommended that the relative amounts should be two volumes of serum to one volume of cells The reason for advising these proportions is that we wish to obtain cells coated with as much incomplete antibody as possible If cell suspensions of a range of concentrations are mixed with an incomplete antibody serum it is found that the weaker the cell suspension the more likely is it that agglutination will occur at this ·THE TECHNIQUE OF THE ANTI-GLOBULIN TEST 57 stage, the stronger the cell suspension the more likely is it that the antibody will be absorbed, but the cells will not be agglutinated Table vii shows the results obtained when a complete anti D scrum was titrated against varying strengths of suspension of the same D positive cells It is incvitable that some cells will be lost during the Washing process and allowance must be made for this, so that at the final stage of the anti globulin test an amplc quantity of coated cells is available Because of the prozone phenomenon, described in Chapter vi, a falsely negative reaction is more likely to occur when a weak cell suspension is mixed with the anti-globulin serum and the presence of an antibody may be missed Normally, the mixing of washed cells with the serum which is expected to sensitise them is performed at room temperature Difficulties may arise, however, when compatible blood is being sought for a patient suffering from hæmolytic anæmia of the cold antibody type Under these circumstances, it will often be found advantageous to heat the donor cells and recipients serum separately to 40° C and to mix at this temperature before incubating them for a minimum of 30 minutes at 37° C in a water bath Since the incubation time is comparatively short, a warm air incubator is not satisfactory Wootton (1951) showed that 30 minutes' incubation is not long enough to produce the optimum results and that incubation periods of up to a maximum of six hours were required to give the strongest reactions If possible, therefore, when sensitising cells for the Indirect anti globulin test, the cell serum mixture should be left for at least two and preferably four hours The thorough washing of the sensitised cells is a most important step in this test As can be seen in table VIII, even so small an amount as one part of human serum in 4000 parts of saline will neutralise an equal volume of diluted antı human globulin serum Moreover, the anti globulin

·No text on this page.

·THE TECHNIQUE Or TIID ANTI GLOBULIN TEST 59 serum tcacts more readily with free globulin thun with antibody globulin attached to the red cells, possibly because the bound globulin has already used some of its combining sites shereas the free globulin has all its combining sites available Careful attention must also be pud to the provision of apparatus free from contaminating globulin Test tubes, pipettes, teats, glass tiles or microscope slides, unless scrupulously cleancd, may result in hours of sasted labour If the thumb or finger is used to occlude the end of the test-tube during mixing, be sure first that it has on it no trace of human serum TABLE VITI Efects of traces of human serum on the anti globulin reachon Dilutions of Human Rht Coated ceils + Serum in AHG AHG 1 in 1 000 1 in 2000 1 in 4 000 1 in 6 000 1 in 12 000 Control (no serum) gw ++ The saline for washing red cells is normally used at room temperature (18 -20°C) When cells have been refrigerated and have become coated with auto agglutinin it is sometimes advisable to wash with warm saline at 37° C A blood specimen, collected from an infant's umbilical cord, may have been contaminated with Wharton's Jelly, becoming gelatinous at room temperature and in this instance, too, washing with warm saline may be more efficacious When Coombs, Mourant and Race first described their test in 1945, the final mixing of the anti globulin serum with the sensitised cells was performed in a test tube They read the results both microscopically and macroscopically though they reported that" in practice the macroscopic reading has in every case proved suſſicient " A year later, Coombs, Mourant and Race (1946) 8 compared the tube and tile ·58 TABLE VI Antibody Dilution Antigen Strength RBC Suspension 2 4 8 16 32 64 Titration Score 128 256 512 ++ V SW 2 per cent 6 per cent per cent per cent per cent per cent 60 per cent 80 per cent 100 per cent >>> ++11 > +11 ** 11111 ++BIILIT +FILM!!! 43 36 19 10 7 111IIIII ||||||| IIIIIIII! ANTI-GLOBULIN (COOMBS) TEST 0 0 * These cells gave a positive Direct anti globulin reaction showing that the cells had absorbed the antibody ·THE TECHNIQUE or THE ANTI-GLOBULIN TEST 59 scrum reacts morc rcadıls with free globulin than with antibody globulin attached to the red cells, possibly because the bound globulin has already used some of its combining sites whereas the free globulin has all its combining sites available Careful attention must also be paid to the provision of apparatus free from contaminating globulin Test tubes, pipettes, teats, glass tiles or microscope slides, unless scrupulously cleaned, may result in hours of wasted labour If the thumb or ſinger is used to occlude the end of the test tube during mixing, be sure first that it has on it no trace of human serum TABLE VII Effects of traces of human serum on the anti globulin reaction Dilutions of Human Rh+ Coated cells + Serum in AHG AHG 1 in 1 000 1 in 2000 1 in 4 000 I in 6 000 I in 12 000 Control (no serum) ++ The salıne for washing red cells is normally used at room temperature (18 -20° C) When cells have been refrigerated and have become coated with auto agglutinin it is sometimes advisable to wash with warm saline at 37° C A blood specimen, collected from an infant's umbilical cord, may have been contaminated with Wharton's Jelly, becoming gelatinous at room temperature and in this instance, too, washing with warm saline may be more efficacious When Coombs, Mourant and Race ? first described their test in 1945, the final mixing of the anti globulin serum with the sensitised cells was performed in a test tube They read the results both microscopically and macroscopically though they reported that in practice the macroscopic reading has in every case proved sufficient A year later, Coombs, Mourant and Race (1946) compared the tube and tile ·60 ANTI GLOBULIN (COOMBS) TEST techniques and stated that a parallel test done in a tube at 37°C and read microscopically gave the same result as a tile test in every case! Nowadays most authorities prefer to use the tile technique having found it to be a more sensitive method Furthermore, if the degree of sensitisation of the cells is slight, the agglutinates formed are rather loose and are liable to be broken up while being transferred by pipette from tube to microscopic slide Opal glass is selected for the tile because it has a smoother surface and is freer from small pits than any other type It can be easily scrubbed clean with a soft brush liquid soap and hot water rinsed free from soap and dried with a special" serum free cloth just prior to use The Antı Globulin Serum When investigating a patient s serum which may contain incomplete antibody the anti globulin serum should be prepared in two dilutions (a) for the detection of gamma globulins, and (b) for the detection of non gamma globulins Normally, the correct dilution for each globulin should be known before the reagent is put into use, but, if it is not, a dilution of one fifth of that recommended for the detection of Rh antibodies will often be suitable. These dilutions must be made up in clean tubes using a new pipette and clean teat on each occasion or alternatively the pipette and teat, which is used for anti globulin serum only, can be stored in a clean dust free box "The saline with which the antı globulin serum Is diluted should be clean and uncontaminated by pipette rinsings When a second dilution of anti globulin serum is employed the squares 1 2 and 3 of figure 8 must be duplicated on the tile When an antı globulin serum is capable of detecting both gamma and non gamma antibodies at the same dilution, both types of positive control cells should be placed on the tile in separate squares side by side ·THE TECHNIQUE OF THE ANTI GLOBULIN TEST 61 The Mixing of the Reagents Dunng the mixing process, when the anti globulin scrum and the sensitised and control cells are being brought into intimate contact, the tile should be standing on a level bench The implement used for stirring may be a wooden swab stich, in which case a separate stick is used for cach mixture Sometimes a plastic rod is used and it must then be quickly washed and dried between each mixing Race and Sanger (1954)' use the four corners of a microscope slide-a different corner is applied to each square of mixture Although this method is effective, false positives may result, especially if the glass microscope slide is rubbed hard on the opal glass tile Its occurrence appears to be related to the type, age and composition of the glass Further mixing of the reagents takes place during the subsequent rocking of the tile It cannot be too greatly stressed that the handling of the tile should be very gentle - Race and Sanger (1954) draw attention to the fact that when working with the less avid antibodies "it is wise to be very gentle with the tile and to tilt it only at the time of reading Gentle rotation of the tile, keeping the cells towards the centre of the area, is of the utmost help in detecting the weaker incomplete antibodies "If the tile is left standing so long that evaporation takes place the results can be difficult to interpret and, particularly when the laboratory temperature is high, the tests should be examined for agglutination 57 minutes after the mixing If, however, the tile is kept in a moist chamber during the reaction time and evaporation is minimised, the tests are read at 5, 9 and 12 minutes after the mixing CONTROLS No reliance can be placed upon test results which have been inadequately controlled It must be shown that (1) the anti globulin serum is capable of detecting both types ·62 ANTI-GLOBULIN (COOMBS) TEST of globulin coatings, (11) the anti globulin serum is not giving false agglutination, (111) the agglutination of the test cells is due to the action of the anti globulin serum and not due to other causes To demonstrate the gamma globulin coating two volumes of thrice washed group O D positive cells are mixed with four volumes of a known weak, incomplete antı D antibody and are incubated at 37 C for one hour Thereafter these cells are washed at the same time and in the same manner as the cells under test To confirm the non gamma globulin coating one volume of a 50 per cent suspension of thrice washed group O red cells is mixed with ten volumes of fresh human serum and is then refrigerated at 4 C for 1 2 hours during which time the cells will pick up cold incomplete antibody These cells are also washed at the same time as the cells under test but the saline used should be warmed to 37° C-Crawford, Cutbush and Mollison (1953) 10 These cells are sometimes referred to as " cold sensitised cells The negative control is provided by the same group o cells which have been exposed to a serum known to be free from coating antibodies They are washed too, at the same time as the test cells For preference these negative control cells should not be obtained from refrigerated blood in which the serum may contain cold incomplete antibody However if the blood was drawn into anti coagulant the anti complementary action of citrate will have inhibited union of antigen and antibody A further cell suspension control is required in the last stage of the anti globulin test The test cells, supposed to have been coated with incomplete antibody, are mixed, not with anti globulin serum but with a drop of salıne Aggluti nation of this mixture indicates that some cause other than the globulin/anti globulin reaction is at work? Some authorities control the results still further by adding ·THE TECHNIQUE OF THIC ANTI-GLOBULIN TEST 63 well-washed scnsitised cells to all apparent negative reactions If agglutination occurs then it demonstrates that sufficient anti globulin serum was there to agglutinate the original cells if they had been coated APPARATUS During the course of the test a great many tubes are constantly handled and moved to and from the centrifuge and racks, and it is obvious that some foolproof method of identifying each tube must be devised' Grease pencil markings are unreliable adhesive labels are more satisfactory Murray (1957) "1 has overcome the necessity for moving tubes during cell washing by designing a centrifuge with a head which will accommodate metal racks holding test tubes RECORDING OF RESULTS AND THEIR INTERPRETATION In registering the results of tests it is inadvisable to record them merely as positive or negative, the degree of agglutination should be assessed and scored from ' + to'+++-to' and the symbol should be used to signify absence of agglutination The results of control tests should be read first- The incomplete antibody used to sensitise cells for the positive control is specifically chosen for its weak action and the degree of agglutination in square 1 of the tile is therefore, unlikely to exceed' +' A complete absence of agglutination in this square suggests either that the cells have failed, for some reason to absorb antibody or that the anti globulin serum has lost its capacity to demonstrate the antibody Under these circumstances, negative results obtained in square 3 (the actual test) cannot be relied upon and the whole exercise must be repeated The negative control in square 2 should show no ·64 ANTI GLOBULIN (COOMBS) TEST agglutination since the cells have not been in contact with an antibody bearing Serum If agglutinates do form in this square they are probably caused by antı human species antibodies which have been insufficiently removed from the anti globulin serum The tests should then be repeated using a different anti globulin serum The cell suspension control composed of supposedly sensitised cells mixed with saline, should remain unaggluti nated If clumps of cells develop in square 4 a comparison should be made between it and the degree of agglutination of the test (square 3) A stronger reaction in square 3 than in square 4 suggests that the serum used for coating the cells in addition to the incomplete antibody present, contains complete antibody which the subsequent cell washing has failed to remove completely, with the result that the cells in salıne agglutinate when rocked on a tile' The enhancement of agglutination when these cells are mixed with antiglobulin serum has been called by Sturgeon (1954)' the antı globulin_augnientation test A serum containing a mixture of complete and incomplete antibodies can, after inactivation of the complete antibody by heat, be re examined by the Indirect anti globulin technique (Dunsford and Bowley, 1955) 13 When the test cells and the cell suspension control both show the same amount of agglutination the whole procedure should be repeated, using warm saline at 37°C for the washing of the cells When the Indirect technique is being applied the serum and cells should be separately warmed to 37 C before mixing A word of warning must here be given about the interpretation of the test itself A negative or positive reaction means very little when considered on its own The result may be false even when all the controls are apparently working correctly' In the next chapter a more detailed account will be given of the possible fallacies ·THE TECHNIQUE OT TIIE ANTI-GLOBULIN TEST 65 REFERENCES 1 MOLLISOV P L and CUTBUSH M 1955 Use of isotope labelled red cells to demonstrate incompatibility in no Lancet 1290 2 ROTHKL 1954 Interaction of Heparin with auto agglutinins in idiopathic acquired hæmolytic anæmi1 Proc Soc Exp Biol & Med 86 352 3 UNGER L. J 1951 A method for detecting Rh antibodies in extremely low titre Lab & Clin Med 37 825 4 UNGER L) and hatz L. 1952 Vanations in ante human globulin titration results J Lab & Clin Nied 39 246 S PLAUT G IKIN EW MOURANT A ESANGER R and RACE RR 1953 A new blood group antibody anti Jkt Malure 171 431 6 MOOTTOV IDP 1951 Factors affecting the speed of agglutination in the Coombs test J Clin Path 4 296 1 COOURS RRA MOURANT A E and RACE R R 1945 Detection of weak and incomplete Rh agglutinins A new test Lancet 11 15 8 COOMBS RRA MOURANT AE and RACE RR 1946 In vivo 150 serisitisation of red cells in babies with hæmolytic disease Lancet 1 264 9 RACE, R R and SANGER R 1954 Blood Groups in Man 2nd Edition Published by Blackwells Oxford 10 CRAWFORD H CUTBUSH M and MOLLISON PL 1953 Specificity of incomplete cold antibody in human serum Lancet 1 566 11 MURRAY J 1957 Griffin autopoise centrifuge serological model Lancet 11 474 12 STURGEON P 1954 The nature of complete Rh anti sera as revealed by anti globulin augmentation J Immunol 73 212 13 DUNSFORD I and BOWLEY CC 1955 Techniques in Blood Grouping Published by Oliver and Boyd Edinburgh ·CHAPTER IX THE CAUSES OF FALSE REACTIONS ALTHOUGH the number of anti globulın tests performed each day must be considerable and the amount of experience gained over the past thirteen years should have led to the development of fool proof methods it remains largely an experimental technique which is fraught with difficulty Each result must be interpreted as only a part of a serological or antigen antibody reaction A test performed without controls, whatever the result, is useless and may be danger ously misleading, and, as previously mentioned a negative or positive result by itself means very little even when the controls are working correctly CAUSES OF FALSE POSITIVE REACTIONS 1 Antı human species antibodies in the anti globulin serum Inadequate absorption may have been the cause on some occasions when hetero antibody is found in the anti globulin serum, but some sera although initially satisfactory, seem to regain the unwanted antibody on storage-Cappell and MacFarlane (1946) This finding suggests that although the bulk of hetero antibody can be absorbed on to human red cells and is discarded with them, further amounts of antibody are apparently neutralised in the serum On storage this neutralisation breaks down and the unwanted antibody becomes active again Workers who use commercially produced anti globulin serum may find that the manufacturer recommends its use at such a dilution that the anti species antibodies are inactive, but if a less dilute solution is required the hetero antibodies become effective again and confuse the issue It is perhaps appropriate to interpolate here a warning 66 ·THE CAUSES OF FALSE REACTIONS 67 about recommended dilutions and expiry dates of scra The titres quoted by the suppliers of scrum are those found by laboratory tests before the scrum is issued and are for guidance only Some sera storc less well than others and nobody can guarantee that a particular scrum will be specific at a particular titre at any particular time after issuc Similarly, although the label on the serum container may show that it has not yet passed the expiry date, this is no guide to its suitability The routine use of a negative control with every antiglobulin test will detect false positive results due to antispecies antibodies and if such antibodies prove to be present they can be removed from the serum by further absorptions 2 Cold complete agglutinns The test result will be invalidated when these antibodies manifest their presence by agglutination of the cell suspension control They have probably originated in the serum with which the cells were sensitised and, if the blood from which it was obtained has been refrigerated, separation of the serum while the blood is still at 4° C will ensure that the bulk of the agglutinin is discarded with the unwanted cells, leaving only a minimum free in the serum When the aim is to free cells from such an antibody, Dunsford, Cowen and Malone (1951),- and Holland (1952) 3 recommend that they should be warmed to 37°C for 15-30 minutes and then washed in warm saline in the preparation for the final stage of the anti globulin test Rosenfield, Vogel and Rosenthal (1951) warm the cells to 37° C for only 1-3 minutes before addıag the anti globulin serum and claim satisfactory results 3 Cold incomplete antibodies Considerable difficulty may be encountered when using cells derived from clotted blood samples which have been refrigerated for the cells may have become coated with the cold incomplete antibody commonly occurring in the serum ·68 ANTI-GLOBULIN (COOMBS) TEST of normal healthy donors This is particularly the case in blood transfusion work if the pilot tubes provided contain clotted blood or a suspension of the donor's cells in their native serum Such cells may prove very troublesome when, during a compatibility test, they are mixed with an antı globulin serum which is rich in anti non gamma globulin fraction Coating of cells is particularly liable to occur also in refrigerated clotted blood samples in which the proportion of cells to serum is small as, for instance, in blood drawn from a grossly anæmic patient "No firm conclusion should be reached, therefore about the antibody content of a serum under investigation until it has been shown that the test cells without exposure to the suspect serum are not agglutinated by the anti globulin serum In the view of Spalt and Kinsell (1953) 5' the interpretation of a positive Coombs' test is valid only on specimens which have not been refrigerated prior to separation of cells from serum "The avoidance of refrigeration will reduce the number of false positive reactions but it is not always practicable "When blood must be stored in the cold it is preferable to mix it with anti coagulant The cold antibodies already considered are those active at refrigerator temperature but occasionally an apparently healthy person will be found who has in his blood a powerful incomplete antibody which is able to coat his own cells at a temperature just below that of the body In consequence cells obtained from clotted samples of his blood will always give a positive Direct anti globulin reaction To overcome this difficulty blood from such a person must be collected into 3 8 per cent sodium citrate which has been previously warmed to 37 C 4 Bacterial contamination Any of the reagents used in the anti globulin test may become infected and curious results may be produced ·69 THE CAUSES OF TALSE REACTIONS Contaminated cells may give a false positive anti globulin reaction which can be detected by the occurrence of agglutination in the cell suspension control 7f the antiglobulin serum is infected it may become pan agglutinating with the result that it causes clumping of all cells with which it comes in contact, including those of the negative control The unknown serum, which is being investigated, may be contaminated and cause pan agglutination as evidenced by agglutination of the cell suspension control The only way to overcome this difficulty is by prevention Blood specimens should be tested as quickly as possible after collection and if they have to be transported to a distance, cells should be suspended in Panocell mixture (Cahan, 1955) 6 which consists of Equal parts of Alsever s solution with a drop of penicillin and streptomycin to give a final concentration of about 30 units and 0 005 mg per millilitre respectively” Alsever's Solution 2 05 per cent Dextrose 0 42 per cent Sodium chloride 0 8 per cent Trisodium citrate 0 55 per cent Citric acid in aqua dist The final pH should be 61 Sera can be sent in the liquid state if sodium azide is added to make a concentration of 1/1000 or they may be dried first Anti globulin sera should be stored frozen at -20°C while not in use in the laboratory 5 Inadequate cell 11 ashing Coombs, Saison and Joysey (1955) describe a condition in which anti globulin serum, although carefully freed from anti species agglutinins, may clump red cells which have not been completely rid of their surrounding plasma or serum The anti globulin serum reacts with the plasma protein, ·68 ANTI GLOBULIN (COOMBS) TEST of normal healthy donors This is particularly the case in blood transfusion work if the pilot tubes provided contain clotted blood or a suspension of the donor s cells in their native serum Such cells may prove very troublesome when, during a compatibility test, they are mixed with an anti globulin serum which is rich in anti non gamma globulin fraction Coating of cells is particularly liable to occur also in refrigerated clotted blood samples in which the proportion of cells to serum is small as for instance, in blood drawn from a grossly anæmic patient No firm conclusion should be reached, therefore, about the antibody content of a serum under investigation until it has been shown that the test cells without exposure to the suspect serum are not agglutinated by the anti globulin serum In the view of Spalt and Kinsell (1953) 5 « the inter pretation of a positive Coombs test is valid only on specimens which have not been refrigerated prior to separation of cells from serum The avoidance of refrigeration will reduce the number of false positive reactions, but it is not always practicable "When blood must be stored in the cold it is preferable to mix it with antı coagulant The cold antibodies already considered are those active at refrigerator temperature but, occasionally an apparently healthy person will be found who has in his blood a powerful incomplete antibody which is able to coat his own cells at a temperature just below that of the body In consequence cells obtained from clotted samples of his blood will always give a positive Direct anti globulin reaction To overcome this difficulty blood from such a person must be collected into 3 8 per cent_sodium citrate which has been previously warmed to 37 C 4 Bacterial contamination Any of the reagents used in the anti globulin test may become infected and curious results may be produced ·71 THE CAUSTS OF TALSE REACTIONS isotonic saline_mixed gently with the tests will disperse conglomerates but will not scatter truc agglutinates However, prevention is better than cure The tests should be read before they begin to dry or, alternatively, the filc may be placed in a moist chamber, or covered with the lid of a box to reduce evaporation 7 The effects of silica Stratton and Renton (1955) 8 and Renton and Hancock (1957) ' have shown that salıne or sodium citrate solutions, stenlised in bottles made of certain types of glass, may contain colloidal silica which can cause spontaneous agglutination of well washed red cells. The presence of serum inhibits this reaction, but when washed cells are mixed on a glass tile with a glass rod, with the bottom of a test tube, or with the corner of a microscope slide, and when the anti globulin serum has been so diluted that the inhibitory effect of serum is virtually negligible, conditions are fitted to encourage the development of red cell agglutination which may be attributed mistakenly to the anti globulin serum The occurrence of this type of agglutination will depend upon the composition of the glassware employed and upon the amount of grinding between one glass surface and another Cells which have been stored in the frozen state in glycerol and have been subsequently recovered seem to be more prone to react to colloidal silica than do normal cells - A change of glassware of a different composition might abolish this cause of false positive reactions and various other arrangements might be made for instance, citrate and salıne solutions might be autoclaved in plastic bags or sterilised by Seitz filtration or, providing that the salıne is to be used within 24 hours of the preparation of the distilled water used in its manufacture, the solution need not be sterilised Wooden swab sticks or plastic rods can be used ·70 ANTI GLOBULIN (COOMBS) TEST precipitating and enmeshing the red cells--a state which can be explained by the phenomenon of co agglutination described by Bordet and Gengou (1911)? The cells of some species are more susceptible than others to coagglutination and Coombs ? and his colleagues first observed its occurrence while applying the anti globulin test in a study of pig iso antibodies The same sort of reaction was seen by Dunsford, Cowen and Malone (1951). when using human serum and cells and the appearance was described as "agglomerates of red cells, cylindrical, oval or sausage shaped, hollow or solid, scattered in a negative field These resembled somewhat blood casts in urine, but were quite unlike red cell agglutinates The importance of adequate and proper cell washing before the admixture of anti globulin serum cannot be too emphatically stressed Inefficient washing_gives rise more commonly to false negative than false positive reactions, a problem which will be dealt with later Another difficulty of the same sort is an appearance of graininess in Indirect anti globulin tests. When a serum is being investigated to determine whether or not it contains antibody, the cells which it is to sensitise should be washed at least twice before being incubated with the Otherwise even if the cells are subsequently well washed before the addition of the anti globulin serum the final reaction will be weak, grainy and very difficult for the inexperienced worker to interpret serum 6 The effects of drying Anti globulin tests performed upon an open tile at room temperature (18 20 C) begin to dry round the edges after about 7-10 minutes and the resultant conglomeration of cells simulates agglutination Provided that the volumes of reagents used in all the squares are equal the same appearance will be seen in the negative control square A drop of ·THE CAUSES OF FALSE REACTIONS 73 case of Pasteur pipettes it may be possible to store these and their teats in a separate box or large glass test-tube for use only for anti globulin sera, but fuiling this a new pipette and teat should be used Dirty saline which had been used for rinsing pipettes should never be used in the anti globulin test, it is better to open a fresh bottle If, when resuspending the button of cells in salıne during the washing process, the test-tube is up ended with a finger or thumb over its mouth, the presence of serum, or even of perspiration on the finger tip may be enough to ruin the test-Holland (1952) In order to ensure that these difficulties will be recognised when they occur, it is recommended that the control cells and test cells should be washed at the same time. The use of control cells left over from a previous test might result in a failure to notice that something had gone wrong with the current reaction As an additional control and to demonstrate that observed negative results in both test and control squares are in fact, truly negative, one can add one volume of well washed, sensitised cells to each mixture showing no agglutination If agglutination of these cells now occurs within 5-7 minutes it will confirm that the anti globulin serum is active and that the originally observed negative result was correct 2 Incorrect dilution of the anti globulin serum Attention has already been drawn to the fact that differing concentrations of anti globulin serum are required for the demonstration of gamma and non gamma globulins A high dilution may react with gamma globulin antibodies, but fails to reveal the non gamma ones Conversely, a lesser dilution may coincide with a prozone area so that a gamma globulin antibody escapes detection The inclusion ·72 ANTI GLOBULIN (COOMBS) TEST to mix the cells with anti globulin serum on the opal glass tile and, so long as vigorous rubbing is avoided, the corner of a microscope slide makes an effective stirrer It has been observed, however that glass slides stored for a long time particularly if the surroundings are damp, appear to cause trouble more frequently than do new slides 8 Other agglutinatıng substances in saline Hopps (1958) 10 has reported the occurrence of false positive results when a faulty ion exchange resin apparatus was used to obtain the pure water with which the salıne solution was prepared Traces of some metals or chemical substances in the resultant salıne led to the agglutination of all cells including the negative control and the cell suspension control CAUSES OF FALSE NEGATIVE RESULTS 1 The neutralisation of anti globulin by traces of serum Although this cause has already been mentioned, it is by far the commonest error committed by the inexperienced worker There are five possible sources of the unwanted globulin incompletely washed cells, dirty glassware dirty rubber bungs dirty saline or dirty fingers Hill and Haberman (1954)" gave most valuable advice in recommending that until the worker has become familiar with the test cells should be given four washings Holland (1952), and Madden (1956) 13 stressed the need for avoidance of high speed centrifuging which would throw down the cells before they could be properly washed by contact with the saline - In the average laboratory it is practically impossible to segregate test tubes slides tiles and other glassware for use for the anti globulin test only so a strict routine of cleaning must be developed and adhered to which will ensure that all apparatus used in blood grouping is dirt free In the ·THE CAUSES OF FALSE REACTIONS 75 the anti globulin scrum When weaker cell suspensions are used for the test there will be an excess of anti globulin antibody and insufficient antigon on the coated cells with which it can combine-one of the causes of the prozone phenomenon described earlier 6 Over-centrifuging in the tube technique Although agglutination can occur immediately the anti globulin serum comes in contact with strongly coated cells in a tube, one is not always handling powerful reagents and Holland (1952) 1" has recommended that the mixture should be allowed stand for at least fifteen minutes so that interaction may be possible before the tube is placed in the centrifuge If the spinning is begun too soon, or if high speeds are used, there will be insufficient contact between antigen and antibody for a reaction to take place and false negative results will be obtained 7 Poor storage of reagents Loutit and Robb-Smith (1946) 17 found that "suitable sera preserve their potency when kept at -20° C but, when diluted and kept at higher temperatures, potency falls off, although a diluted, absorbed serum can be used for at least a week without loss of activity Pickles,18 in the same year, reported that one of her sera diluted to 1/10 and stored at -20 C, had lost half its strength after five months She observed that weaker solutions deteriorate more rapidly and that sera stored at 4°C fairly quickly become inactive She recommends that only sufficient serum for about three months' work should be absorbed, that it should be stored at -20° C and that when dilutions are being arranged only enough reagent for one week's work at a time should be made up on the other hand, Aspinall (1955) 19 was able to store certain batches of diluted sera for months at —20° C without loss of titre ·74 ANTI GLOBULIN (COOMBS) TEST of correct controls with each test performed makes certain that such errors will be noticed 3 Lack of complement Before non gamma globulin antibodies coating the cells can be demonstrated by the anti globulin test certain components of complement must be present Since anticoagulants are anti complementary, patients' plasma should never be used for compatibility tests Furthermore, the donor cells taken from a pilot tube containing antı coagulant must be rid of plasma and citrate by thorough washing before exposure to the patient's serum with which they are to be matched The patient s serum must be fresh (stored for less than 48 hours), otherwise additional complement should be supplied in the form of fresh AB serum Blood cells from patients suffering from acquired hæmolytic disease frequently give a positive Direct anti globulin reaction but this will be abolished if heparin anti coagulant is added to the blood Roth (1954) 14 4 The use of aged red cells The best results will follow when fresh cells are used for the test Çells stored in salıne suspension may hæmolyse after a comparatively short time'and, even when lysis does not occur cells known to have been coated with antibody loose that antibody into the surrounding saline in a few hours, Barrie and Krieger (1954) 15 Even in clotted blood kept at 4 C for several days cells known to have been previously sensitised will sometimes give a negative Direct anti globulin test The practice of sending salıne suspensions of cells for examination through the post is to be strongly condemned 5 The use of thin cell suspensions Ellis (1952) 18 has shown that better results and fewer falsely negative reactions are encountered when a 50 per cent (or stronger) suspension of sensitised cells is added to ·THE CAUSES OF FALSC REACTIONS 77 2 DUNSTORD 1 COWEN and MALONE RH 195$ The Coombs Test Ted Lab Techin 9 137 3 HOLLAND E 6 3952 Various uses of the an!: human globulin test Canad J Aled Techn 1 lll 4 ROSEVIIELD R L. VOGEL, P and ROSENTIAL N 1953 The anti globulin test technical and practical applications Am J Clin Path 21 301 S SPALT T H and LINELL, B C 1953 Studies on the normal scrum panagglutinin actise gainst trypsinated human erythrocytes 11 Relationship to cold assultination Lob & Chn Med 42 205 6 CAHAN A 1955 Fluid for shipping whole blood specimens JAVA 158 971 7 COOMBS RRA Saisoy R and JOYSEY V C 1955 A circumstance in which the phenomenon of co-agglutination could complicate the ant globulin sensitisation test by stimulating agglutination Brity Exp Parh XXXVI 179 7a BORDET J and GENGOU O 193 Zbt Bakt 1 Ab Ong 58 330 & STRATTON F and RENTON P H 1955 Effect of crystalloid solutions prepared s glass bottles on human red cells Nature 175 727 9 RENTON PH and HANCOCK SA 1957 Anubody like effects of colloidal silic: For Sanguinis 2 117 10 Hopps OD 1958 Personal communication 31 Hilm JM and HARERNAV S 1994 The Coombs (anti globulin) test Indications and technics Am J Clin Path 24 3 305 12 HOLLAND EG 1952 Various uses of the anthusias globulin test Canad Med Tech 14 111 33 MADDEN HJ 1956 To do or not to do a Coombs Test Med Techn Nov Dec 383 14 ROTH K L 3954 Interaction of Heparin with auto agglutinins in idiopathic acquired hæmolytic anemia Proc Soc Exp Bior & Med 86 352 15 BARRIE J and KRIEGER VI 1954 A note on decrease of avidity of the Coombs reaction by gradual clution of Rh antibodies from the sensitised cells Med J Austral Feb 13 247 16 ELLIS E 1952 A slide test procedure using anti human globulin serum (Coombs serum) Am J Med Techn 18 229 17 LOUTIT J F and Road San AHT 1946 Report on the suit ability of reagents for the Coombs test presented at a meeting of Blood Transfusion Officers Ministry of Health London 18 PICKLES MM 1946 The comparative merits of the Welcome rabbit precipitating seruti (ants human) rabbit anti human globulin and rabbit anu human serum Report presented at a meeting of Dlood Transfusion Officers Ministry of Health London 19 ASPINALLP 1953 Personal communication ·76 ANTI-GLOBULIN (COOMBS) TEST On the whole, although some anti globulin sera can be satisfactorily stored at 4° C it is preferable to maintain them at a temperature of -20° C which will reduce the chance of bacterial growth Quick thawing performed at 37° C will minimise precipitation and the solution must then be well mixed, otherwise it will be found that the top of the container holds only salıne while the active principle is mainly confined to the lower part If the anti globulin serum is to be stored at 4 C it is advisable to add 1/10 000 Methiolate or 1/1000 sodium azide as a preservative Some laboratories issue anti globulin reagent diluted ready for use The choice of diluting fluid may be governed by the local availability of material but salıne should be avoided whenever possible Anti globulin serum obtained from rabbits may be diluted with other rabbit serum from which the hetero antibodies have been absorbed, similarly goat anti globulin serum can be mixed with goat serum When describing the preparation of anti globulin serum mention was made (page 25) of a method of obtaining additional antibody by elution from the blood clot Although this eluate is largely salıne the antibody content more than makes up for the deterioration likely to follow the use of saline as diluent Rosenfield Vogel and Rosenthal (1951) report favourably on the use of a 24 per cent solution of bovine albumin (Cohn s fraction V) in salıne as a diluting fluid Whatever the material used for dilution, the reagent, if it is to be relied upon must still contain enough antı nongamma globulin antibody to detect cell coating by antibodies of this type and the only proof that the anti globulin serum is active will be the evidence supplied by the behaviour of the control tests 4 REFERENCES 1 CAPPELL D F and MCFARLANE MN 1946 Spontaneous loss of specificity of absorbed anti Rh sera Lancet 1 558 ·APPLICATIONS OF THE ANTI-GLOBULIN TEST 79 various ways such as suspension in albumin, exposure to enzymes or application of the Indirect anti globulin method The efficiency of the four most commonly employed techniques is compared in table ix where it will be seen that the anti globulin method is the most reliable single test although no claim can be submitted that it will pick up every antibody TABLE IX Antibody Saline Albumin | Enzymes Anti Globulin 2 3 4 Anti A antı B(1 Anti Rh (34) Antı hell (1) Anti MNSSC ) Anti Duffy ( ) Anti Kidd () Symbols used Af士十一 土土干一十一 ++H+IFFE + = Excellent = No use + = Usually good F = Not reliable worsen as The performance of compatibility tests takes time and occasionally the need for blood for the patient will be so urgent that the necessary delay of two or more hours required for matching the blood will allow the patient's condition to For this reason the practice of classifying patients Dangerous" or Safer recipients-Tovey (1950) has been widely adopted "The Dangerous recipient is a person who may have developed a blood group antibody as a result of learlier stimulation by blood transfusion by intramuscular injection of blood or by a pregnancy in which a blood group incompatibility between mother and child existed "The Safer recipient has' no obstetric history suggesting maternalfætal incompatibility and has never received blood Before this classification can be relied upon, careful questioning and accurate answers are most essential Where any doubts exist, or if the patient, being unconscious cannot give the necessary information he should be regarded as potentially dangerous ·CHAPTER X APPLICATIONS OF THE ANTI GLOBULIN TEST THE anti globulin test has its main usefulness in the preparation of blood for transfusion, in the study of hæmolytic disease of the newborn and in hemolytic anæmia in adults Other more specialised fields in which it may be found helpful will be briefly described though the reader should consult the references for more detailed information before applying the technique to his particular problem BLOOD TRANSFUSION Before discussing the advantages of the anti globulin test as employed in blood transfusion work it may be profitable to review quickly some general principles Transfusion of compatible blood can be a life saving procedure in replacing lost red cells and in restoring a depleted blood volume Before the transfusion is undertaken, however a compatibility test must be done to ensure that the donor's red cells will not be prematurely destroyed by an antibody in the patient's circulation, such cell destruction would do the recipient no good and might endanger his life "In the compatibility test (also known as the direct match or major cross match) cells from the donor are mixed with the patient s serum. In spite of a great many hours of labour in many laboratories no single test has yet been invented which will unfailingly detect all types and strengths of antibody which may be present Donor's cells in saline suspension, when mixed with the patients serum will disclose the presence of any complete antibody, while for the demonstration of incomplete antibody several tests have been devised which involve treatment of the donor s cells in 78 ·APPLICATIONS OF TIIC ANTI-GLOBULIN TEST 81 have disastrous consequences Resultant agglutination would show that an incompatibility betwcen donor and patient cxısted and that it would be unsafe to give the fransfusion until compatible blood from another donor had been found If no agglutination is shown by this technique it should not be assumed automatically that the bloods of the donor and patient are compatible, unless the saline and/or albumin compatibility tests also produce no agglutination Clinicians confronted with the plight of a patient sorely in need of transfusion may, quite understandably, become peevish at the delay occasioned by compatibility tests and may urge that the time allowed for the various procedures should be reduced While sympathising with their frustration it must also be remembered that rapid techniques are less reliable and the accuracy of interpretation of their results depends to some extent upon the experience of the technician It is only when the risk to the patient of withholding blood outweighs the risk of giving possibly incompatible blood that departure from safer routine methods is justifiable 2 The Rapid anti globulin compatibility test When the time available is not more than 30 minutes it may be permissible to shorten the period of incubation during which attachment of antibody to the red cell antigens takes place Donor blood of the same ABO and Rh group as the patient is obtained Two volumes of a 50 per cent saline suspension of well-washed donor's cells is mixed with four volumes of the recipient's serum in a test tube The mixture is incubated for at least 15 minutes at 37° C in a water bath, which will more quickly bring the tube to the correct temperature than could be achieved in an incubator , In order to bring the antibody and antigen into close association G ·80 ANTI GLOBULIN (COOMBS) TEST The point of making this classification is that the "safer recipient is unlikely to have developed an antibody which will destroy the donor cells In his case therefore the cross match tests can be done by the comparatively straight forward and less time consuming saline and albumin techniques Infrequently, an antibody will be unexpectedly found in the blood of a patient whose history has not led one to suspect its presence in the majority of such cases it will prove to be an incomplete anti Rh antibody The albumin technique which is excellent for the detection of most incomplete Rh antibodies should therefore be included in the compatibility tests even when the patient belongs to the safer" category' Compatibility tests for a dangerous " recipient must involve more searching techniques Saline and albumin matching tests will disclose the majority of more commonly occurring antibodies but some incomplete antibodies are only detectable by the anti globulin technique and this is used in addition to the saline and albumin methods t -1' The anti globulin compatibility test The Indirect anti globulin technique is used Washed donor s cells are mixed with a double volume of the patients serum and incubated for 11 hours so that if an antibody against antigens on the donor s cells is present it will have an opportunity to become affixed to the red cell surface Complement may be necessary for eventual detection and this point should be borne in mind if samples of patients sera are stored in the refrigerator for several days before compatibility tests are to be done After incubation the cells are thoroughly washed free from unbound globulin In the final stage of the test the anti globulin serum should be used in at least two dilutions to make certain that both gamma and non gamma globulin antibodies are detectable Failure to use wide spectrum anti globulin serum can ·APPLICATIONS OF THE ANTI-GLOBULIN TEST 83 especially where the volume of blood transfused has been relatively large the repeated salınc and albumin compatibility tests may appear to give perfectly satisfactory results, however, by performing a Direct anti-globulin test with serial dilutions of anti globulin serum on the unrefrigerated cells obtained from the patient's post transfusion blood it may be possible to show that the cells are coated with antibody and that the transfused blood was probably incompatible THE DETECTION OF BLOOD GROUPS ANTIGENS AND ANTIBODIES The use of the anti globulin technique to investigate incompatibilities found during the selection or following the giving of blood has greatly advanced our knowledge of human blood groups Anti Duffy (antı Fya) was discovered by this method and some red cell antigens, such as Eu, can be detected apparently only by the anti globulin technique Possibly the most important red cell antigen which is detected by the anti globulin test in blood transfusion work, is the variant of the Rhesus D factor—Du It is characteristic of cells bearing the antigen Du that they will be agglutinated by some anti D sera, but not by others A high grade Du” blood contains cells which are agglutinable in saline suspension by some complete anti D sera, grade Du" cells are not clumped by any saline antı D but are sensitised by some incomplete anti-D sera Since it has been proved that the Du antigen can immunise D negative patients causing the development of anti D antibodies, it is therefore essential to be able accurately to distinguish donors whose blood is really Rh negative (cde/cde) from those who appear to be Rh negative, but who have the antigen Du (cDue/cde) Since the saline and albumin agglutination tests are unreliable it is a routine practice in transfusion services 6 . low ·82 ANTI-GLOBULIN (COOMBS) TEST the tube is then spun in a centrifuge at 1000 rpm for 1 2 minutes After spinning remove a drop of cells for micro scopic examination Agglutination at this stage would demonstrate the presence of a complete antibody, the action of which might not be detectable if observations were not made until the cells had undergone their next stage of triple washing If no agglutination develops after the addition of anti globulin serum, the blood of the donor can be transfused with reasonable confidence 'Whenever it has been necessary to curtail the technique in this way it is wise to confirm the results of the rapid test by immediately setting up saline albumin and anti globulin compatibility tests, allowing the full time for reactions to take place By the time these results can be read the blood will probably have been transfused but if by any chance the rapid method has failed to disclose the presence of an antibody, its subsequent discovery may enable the clinician to anticipate and treat any complications which may arise before irreversible changes have taken place in the patients tissues THE INVESTIGATION OF TRANSFUSION REACTIONS It happens, unfortunately from time to time that, as a result of a technical clerical or administrative error, a patient will receive incompatible blood When such a mistake has occurred or is suspected a sample of the patient's blood is collected for investigation forthwith, or at least within 24 hours of the development of the transfusion reaction The grouping and compatibility tests are repeated In spite of clinical evidence that a reaction has occurred it may be found that no free antibody is demonstrable in the patient's post transfusion serum This happens when the transfused donor cells mop up and become coated by the antagonistic antibody before being broken down in the patient's circulation Under these circumstances and ·APPLICATIONS OF THE ANTI-GLOBULIN TEST 83 especially where the volume of blood transfused has been relatively large the repeated saline and albumin compatibility tests may appear to give perfectly satisfactory results, however, by performing a Direct anti globulin tcst with serial dilutions of antı globulin serum on the unrefrigerated cells obtained from the patient's post transfusion blood it may be possible to show that the cells are coated with antibody and that the transfused blood was probably incompatible THE DETECTION OF BLOOD GROUPS ANTIGENS AND ANTIBODIES The use of the anti globulin technique to investigate incompatibilities found during the selection or following the giving of blood has greatly advanced our knowledge of human blood groups Anti Duffy (anti Fya) was discovered by this method and some red cell antigens, such as Eu, can be detected apparently only by the anti globulin technique Possibly the most important red cell antigen which is detected by the anti globulin test in blood transfusion work, is the variant of the Rhesus D factor-Du It is characteristic of cells bearing the antigen Du that they will be agglutinated by some anti D sera, but not by others A high grade Du” blood contains cells which are agglutinable in saline suspension by some complete anti D sera, grade Du' cells are not clumped by any salıne anti D but are sensitised by some incomplete antı D sera Since it has been proved that the Du antigen can immunise D negative patients causing the development of antı D antibodies, it is therefore essential to be able accurately to distinguish donors whose blood is really Rh negative (cde/cde) from those who appear to be Rh negative, but who have the antigen Du (cDue/cde ) - Since the saline and albumin agglutination tests are unreliable it is a routine practice in transfusion services << low ·84 ANTI GLOBULIN (COOMBS) TEST throughout the world to use the anti globulin technique to examine donor bloods which are apparently D negative, and this is of particular importance in peoples of negro extraction in whom the genotype cDue/cde occurs more commonly than is found in white populations The most reliable and sensitive method is to expose the alleged D negative cells to the action of a pool of powerful incomplete antı D sera and then to examine the cells for a coating of antibody Those cells which are unagglutinated by the anti D sera, but which absorb the anti D antibody and are subsequently agglutinated by the anti globulin test, are presumed to have the Du antigen The cells declared as Du should give a negative Direct antı globulin test before exposure to the incomplete anti-D sera The Indirect anti globulin test has a further application in the investigation of sera containing immune antı A and antı B antibodies Ideally, a patient should receive only blood of his own group but sometimes group O blood is transfused to patients of other groups in the mistaken belief that it is universally compatible While in the majority of cases no harm results, it will be found, when dealing with large numbers of group O donors, that about 10 per cent of them have in their sera alpha and/or beta hæmolysins and incomplete forms in addition to the naturally occurring complete or saline anti A and anti B agglutinins The transfusion of O blood containing immune anti A to an A patient for instance might result in a harmful reaction One method employed to discover whether immune forms of anti A or anti B are present, in addition to the normal agglutinins in a group O serum is to neutralise the complete antibodies by the addition to the serum under investigation of A and B blood group specific substances (obtained in pure form or from the saliva of a secretor') Immune forms of anti A and anti B cannot be neutralised in this way, but will be capable of coating A, and B cells as ·69 as APPLICATIONS OF THE ANTI-GLOBULIN TEST 85 evidenced by the Indirect anti globulin technique The donors whose sera are able to coat A, and/or B cells are classified Dangerous group o donors " and their blood should be used only for group O patients Some transfusion centres screen their group o donors and give a distinctive label to bloods which are considered unsafe for transfusion to patients of other ABO groups - HÆMOLYTIC DISEASE OF THE NEWBORN IN MAN At the time of conception both parents contribute genes to the new individual It happens not infrequently that the infant inherits from its father blood group antigens which the mother does not possess and against which it is theoretically possible for her to make antibodies So long as no antibodies are present in her circulation the blood group incompatibility between mother and child is of noconsequence and can be ignored If she should develop antibodies, however, which can pass the placental barrier and disrupt the fætal red cells, the child's life may be endangered The mechanism of the stimulus to antibody production provided by transfusion where “ foreign " red cells are introduced directly into the patient's blood stream can be fairly readily understood It is less easy to postulate a mechanism which explains how antibodies can be provoked by pregnancy alone, since the maternal and fatal circulations are entirely separate and have no blood vessels in common by which fætal red cells could gain access to the mother's blood stream The most credible hypothesis illustrated in fig 9, suggests that sometimes during pregnancy a breakdown of adjacent maternal and fætal blood vessels occurs and permits a small quantity of the child s blood to enter the mothers capillaries The break quickly heals but not before a number of fætal cells sufficient to initiate antibody formation, have mingled with the mother's blood Cases have been observed ·86 ANTI-GLOBULIN (COOMBS) TEST and recorded-Chown (1954 and 1955) 3 and Dunsford (1957) in which recognisable fætal red cells have been found in the maternal circulation during pregnancy It may be that small accidental hæmorrhages of this sort can occur several times during the gestation period, each occasion FOETUS Rh Positive Amniotic Fluid Placenta MOTHER Rh Negative © O OO Maternal Circulation Rh Antigen Production of Anta Rh. Antibody Anti RK Antibody Haemolysis of Rh+Red Cells FIG 9 The mother is Rh negative The fætus is Rh positive having inherited this character from its father Rh antigen (either as fætal red cells or in soluble form via the amniotic fluid) passes from the fætus to the mother who responds by producing anti Rh antibody The antibody passes from mother to the fætus whose Rh positive red cells it damages leading to hæmolysis and so to anæmia and jaundice This process only takes place in a small minority of cases where an Rh negative mother has an Rh positive baby giving additional impetus to the development of antibody, but it is more likely that the intermiving of bloods takes place at or about the time of parturition The antibody is then first detected in the post natal period and therefore will not affect the child whose red cells caused the immunisation of the mother, although a subsequent child possessing the same red cell antigen will be susceptible to the effects of the antibody ·APPLICATIONS OF THE ANTI GLOBULIN TCST 87 The immune antibody produced by the mother, being soluble and of sufficiently small molecular sizc, can diffuse across the placental barrier to attach itself to the fætal red cells which arc then destroyed or weakened, with the result that at birth the child may be found to be antmic and quickly becoming jaundiced after birth In severe cases the breakdown of the cells is so great that the fætus dies in utero or within a few days of delivery Hæmolytic disease of the newborn occurs comparatively infrequently the incidence being only about 1 150 to 1 200 pregnancies Even where the blood-groups of the parents are glaringly incompatible immunisation of the wife and subsequent disease of the children is by no means inevitable It is fortunate too that hæmolytic disease is one of the few conditions affecting the newbom child which can be anticipated, making it possible to plan the treatment which may be necessary within a few hours after its birth- The causative antibodies most commonly encountered are those directed against the Rhesus blood group antigens The great majority of cases, probably 98 per cent or more, are due to the action of anti R antibodies, although occasionally antı Ce antı E. anti c, antı f or antı Chave been implicated Antibodies against other blood-group systems such as immune anti A, anti-K or anti k, may be found also to have been responsible for destruction of fatal red cells In order to be able to select and forecast those pregnancies in which hæmolytic disease is a potential hazard the mother's blood is usually examined once or twice during the course of the pregnancy to discover whether or not she is developing antibodies As a rule the use of ordinary saline and albumin agglutination methods, in which her serum is mixed with a panel of red cells of known genotype, will disclose any antibody present Some anti-Rh and anti Kell antibodies, however, are demonstrable only by the Indirect anti globulin ·88 ANTI-GLOBULIN (COOMBS) TEST technique and this method should be included routinely, particularly when an anti natal patient presents who has a history suggesting the possibility that she has become immunised If antibodies are found in the serum of a pregnant woman it is usual to recommend that her confinement should take place in a maternity unit with facilities for early clinical and serological assessment of the child s condition and where, if necessary within a few hours of its birth, the child may be treated by exchange transfusion of red cells which will not be destroyed by the maternal antibody freely circulating in the infant's blood stream When the child of a mother who is known to have antibodies is observed immediately after birth to be obviously anæmic or jaundiced there can be little doubt about the probable clinical diagnosis of hæmolytic disease of the newborn However, when such a mother produces a child with no clinical signs of the disease it may mean either that the child is affected but the development of signs of hæmolysis is less well advanced than in the previous case or that the child is unaffected since its blood does not contain the antigen against which the maternal antibody is directed The performance of a Direct anti-globulin test on red cells obtained from the child s umbilical cord will distinguish those infants whose cells are coated by maternal antibody, and who are therefore suffering from hæmolytic disease, from those with uncoated cells who are unaffected The circumstances in which it is especially important to collect and test red cells obtained from the umbilical cord are 1 When a multiparous D negative mother has had no examination of her blood during pregnancy, 2 In the second or subsequent pregnancies of a D negative mother Although examination of her blood in the 32nd to 35th week of pregnancy may have shown no abnormality ·APPLICATIONS OF THE ANTI-GLOBULIN TEST 89 19 she may, occasionally, develop cnough antibody in the later weeks of pregnancy to coat her child's red cells, - When ante natal tests have shown that the mother has become immunised to one of the red cell antigens, -4 When a D positive mother has an obstetric or transfusion history which would put her in the" dangerous recipient category, 5 When the child has an unexplained anæmia or develops jaundice within three days of birth A knowledge of the reaction obtained in the Direct anti globulin test may be very helpful when a decision about treatment has to be made, but the diagnosis must not be allowed to hinge upon the results of this test alone A negative reaction does not entirely exclude hemolytic disease, nor does a positive result by itself prove the condition to be present Certain other criteria must be fulfilled before reliance can be placed upon the results It should be proved (1) that the mother's serum contains antibody active against the red cells both of the child and of its father, (u) that the mother s red cells lack an antigen possessed by both child and father, and (111) that the child's red cells are, or can be, coated by maternal antibody so that they give a positive Direct anti globulin reaction (There may be exceptions to this rule when hæmolytic disease is due to ABO incompatibility ) Fallacies of the Direct Anti Globulin Test False Negative Results 1 If collection of blood from the baby is delayed for several days after birth there may be no coated cells left in circulation with which the anti globulin serum can react Examination by saline, albumin and the Indirect anti globulin methods, of the ability of the mother s serum to agglutinate or coat her child's red ·90 ANTI GLOBULIN (COOMBS) TEST cells should make clear whether or not the child is affected by hemolytic disease 2 When hæmolytic disease is due to ABO incompatibility the condition may be missed if the result of the Direct anti globulin reaction forms the sole diagnostic criterion Although Rosenfield (1955) 5 claimed that, using a modified technique, positive results were not infrequently observed, our experience and that of Mollison (1956) is that the result is quite often negative and that when positive the reactions are comparatively weak When trying to establish the diagnosis the more reliable signs are (1) the presence of anti A or antı-B hemolysins at 37° C in the mothers serum, (11) the finding of incomplete anti A or anti B in the child s blood, and (ui) spherocytosis and an increased osmotic fragility of the child s cells (Dunsford and Bowley 1955 56 , and Mollison, 1956) A positive Direct test may sometimes be obtained by incubating the child s cells with fresh adult serum 3 The child may have had a transfusion before the blood samples were taken for testing (Jakobowicz, Krieger and Simmons 1948) ? False Positive Results 1 Contamination by Wharton's Jelly of blood samples collected from the cut end of the umbilical cord may cause the red cells to conglomerate, giving an appear ance of agglutination A similar picture in the cell suspension control should draw attention to the false nature of the result The blood sample should be warmed to 37 C the cells are transferred to salıne at 37° C and are then rewashed before repetition of the test Contamination by Whartons Jelly can be avoided by using a needle and syringe to collect the blood from the umbilical vein ·8 APPLICATIONS OF THE ANTI-GLODULIN TEST 91 2 A falsely positive Direct anti globulin test muy result from the action of a cold incomplete antibody which had its origin in the mother's scrum (Hollander, 1952) This antibody is inactive at body temperatures and when it passes through the placenta to the infant's circulation it does no harm It is not until a blood sample, taken from the child, is allowed to cool that the free antibody present actively coats the red cells which then show a positive Direct anti globulin reaction If blood from the mother is stored in the refrigerator her cells, too, will give a positive result though the reactions of the cells of a group-A, mother may be weaker than those given by the cells of her group-O child The antibody in the maternal serum will be found to have no activity against fresh, unrefrigerated samples of cells from the child s blood from the father's blood or from the genotyped panel of donor cells These deceptive results are not the only possible causes of difficulty, when anomalies are found consideration should also be given to the causes of false reactions previously mentioned on pages 35 64 Hæmolytic disease of the newbom cannot with certainty be diagnosed from an investigation of the baby s blood alone It is essential to examine the mother's blood as well and ideally the father's blood also should be tested occasions the mother s blood does not appear to contain an antibody against a panel of cells and yet she gives birth to a baby whose cells give a positive Direct reaction An examination of the mother s serum for antibodies against the father s cells will help to determine whether the baby s positive Direct test was a false or true result Any assessment of the severity of a particular case of hæmolytic disease is usually based mainly on clinical Signs and on the hæmotological findings Wiener and Gordon On rare ·9 92 ANTI GLOBULIN (COOMBS) TEST (1953) considered however, that a quantitatne anti globulin test can indicate the degree of affliction The baby s cells are washed thoroughly and are then tested against serial dilutions of a suitable standard anti globulin serum At the same time the anti globulin serum is titrated against D positive cells which have been sensitised by exposure to a standard anti D serum The ratio of the two titres multiplied by 100 represents the degree of coating of the child's cells Greenwalt and Wagner (1955) 10 applied this quantitative anti globulin technique in 86 cases and found no regular correlation between the test results and the severity of the disease THE AFTER EFFECTS OF HÆMOLYTIC DISEASE OF THE NEWBORN Children surviving hæmolytic disease of the newborn without treatment and a small percentage of those who have been treated may have some degree of brain damage (Fulton Roberts, 1957) As a result the child may present, months later, with athetoid movements of the limbs spasticity mental retardation unexplained deafness or hepatomegaly and splenomegaly (Drummond and Watkins, 1946) * -and the nature of the underlying cause may not be revealed by clinical examination The performance of an Indirect anti globulin test, using the mother's serum and the child s red cells, may give a pointer to the ætiology Some women, known to have become immunised during pregnancy have no demonstrable antibody when the blood is examined a few months after the birth of an affected child, but most immunised women continue to produce antibodies throughout life In general the younger the child the greater the chance of demonstrating any blood group incompatibility existing between it and the mother How ever, it has been possible by the Indirect anti globulin test, ·APPLICATIONS OF THE ANTI GLOBULIN TEST 93 to demonstrate in the blood of a 70-year-old woman, immunc antibodies against the red cells of her 40-year-old son HÆMOLYTIC ANÆMIA The name "hæmolytic anemia covers a number of conditions in which the main feature is a shortening of the life span of the red cells manufactured by the patient Sometimes the underlying cause can be readıly recognised, sometimes its origin and the reason for the hemolytic crises which recur periodically are frankly unknown The field is too vast to attempt more than a brief description here The reader who wishes to study the fascinating confusion of the subject should consult Dacie's book, The Hæmolytic Anæmas, where he will find many of its complexities unravelled Two broad categories can be distinguished In the congenital or hereditary type of hæmolytic anæmia an intrinsic defect in the structure of the red cell, or of its enzymes, apparently causes it to age more quickly than a normal cell Hereditary spherocytosis, sickle cell disease and Mediterranean anæmia are examples of this type of condition The label, acquired hæmolytic anæmia, is attached to a more diverse series of abnormal states As implied by the title the abnormality is not inherited, it is probable that the fault is in the environment of the cell rather than in the cell itself, although the extrinsic factor may so alter the cell structure that it becomes more vulnerable to normal destructive forces While it may be possible occasionally to attribute a hæmolytic episode to a bacterial infection or a chemical poison it more often happens that no such connection can be traced In some cases an auto antibody, which may be either an agglutinin or a hæmolysin, can be demonstrated by serological methods in the patient s blood, though the purpose or cause of such a self destructive mechanism is hard to understand ·At 92 ANTI GLOBULIN (COOMBS) TEST (1953) considered, however, that a quantitative anti globulin test can indicate the degree of affliction The baby's cells are washed thoroughly and are then tested against serial dilutions of a suitable standard anti globulin serum the same time the anti globulin serum is titrated against D positive cells which have been sensitised by exposure to a standard antı D serum The ratio of the two titres multiplied by 100 represents the degree of coating of the child s cells Greenwalt and Wagner (1955) 10 applied this quantitative anti globulin technique in 86 cases and found no regular correlation between the test results and the severity of the disease THE AFTER EFFECTS OF HÆMOLYTIC DISEASE OF THE NEWBORN Children surviving hæmolytic disease of the newborn without treatment and a small percentage of those who have been treated may have some degree of brain damage (Fulton Roberts 1957) 11 As a result the child may present, months later, with athetoid movements of the limbs spasticity, mental retardation, unexplained deafness or hepatomegaly and splenomegaly (Drummond and Watkins, 1946) 1*—and the nature of the underlying cause may not be revealed by clinical examination The performance of an Indirect anti globulin test, using the mother s serum and the child s red cells, may give a pointer to the ætiology Some women, known to have become immunised during pregnancy have no demonstrable antibody when the blood is examined a few months after the birth of an affected child, but most immunised women continue to produce antibodies throughout life In general the younger the child the greater the chance of demonstrating any blood group incompatibility existing between it and the mother However it has been possible by the Indirect anti globulin test, ·APPLICATIONS OF THE ANTI-GLOBULIN TEST 93 to demonstrate in the blood of a 70 year old woman, immune antibodies against the red cells of her 40 year old son HÆMOLYTIC ANÆVIA The name "hæmolytic anæmia covers a number of conditions in which the main feature is a shortening of the life span of the red cells manufactured by the patient Sometimes the underlying cause can be readily recognised, sometimes its origin and the reason for the hemolytic crises which recur periodically are frankly unknown The field is too vast to attempt more than a brief description here The reader who wishes to study the fascinating confusion of the subject should consult Dacies book, The Hæmolytic Anæmas, where he will find many of its complexities unravelled Two broad categories can be distinguished In the congenial or hereditary type of hæmolytic anæmia an intrinsic defect in the structure of the red cell, or of its enzymes, apparently causes it to age more quickly than a normal cell Hereditary spherocytosis, sickle cell disease and Mediterranean anæmia are examples of this type of condition The label, acquired hæmolytic anæmia, is attached to a more diverse series of abnormal states As implied by the title the abnormality is not inherited, it is probable that the fault is in the environment of the cell rather than in the cell itself, although the extrinsic factor may so alter the cell structure that it becomes more vulnerable to normal destructive forces While it may be possible occasionally to attribute a hæmolytic episode to a bacterial infection or a chemical poison it more often happens that no such connection can be traced In some cases an auto antibody, which may be either an agglutinın or a hæmolysin, can be demonstrated by serological methods in the patients blood, though the purpose or cause of such a self destructive mechanism is hard to understand ·92 ANTI GLOBULIN (COOMBS) TEST (1953) considered, however, that a quantitate anti globulin test can indicate the degree of affliction The baby's cells are washed thoroughly and are then tested against serial dilutions of a suitable standard anti globulin serum At the same time the anti globulin serum is titrated against D positive cells which have been sensitised by exposure to a standard anti D serum The ratio of the two titres multiplied by 100 represents the degree of coating of the child's cells Greenwalt and Wagner (1955) 10 applied this quantitative anti globulin technique in 86 cases and found no regular correlation between the test results and the severity of the disease THE AFTER EFFECTS OF HÆMOLYTIC DISEASE OF THE NEWBORN Children surviving hæmolytic disease of the newborn without treatment, and a small percentage of those who have been treated, may have some degree of brain damage (Fulton Roberts, 1957) 11 As a result the child may present, months later with athetoid movements of the limbs, spasticity, mental retardation, unexplained deafness or hepatomegaly and splenomegaly (Drummond and Watkins, 1946) --and the nature of the underlying cause may not be revealed by clinical examination The performance of an Indirect anti globulin test, using the mother s serum and the child s red cells may give a pointer to the ætiology Some women known to have become immunised during pregnancy have no demonstrable antibody when the blood is examined a few months after the birth of an affected child but most immunised women continue to produce antibodies throughout life In general the younger the child the greater the chance of demonstrating any blood group incompatibility existing between it and the mother However, it has been possible by the Indirect anti globulin test, ·APPLICATIONS or THE ANTI-GLOBULIN TIST 95 stored in the refrigerator thc red cells may become coated with the cold incomplete antibody commonly found in the sera of healthy people, so that they give a positive Direct anti globulin reaction which is not indicative of hemolytic anemia To avoid confusion of this sort, blood samples from patients suspected to be suffering from hæmolytic anemia should be tested before they are allowed to cool To do this effectively the venepuncture should be made with a warm syringe, the blood should be collected into a warmed container and it should be kept at 37° C until the clot has formed The subsequent centrifuging and separation of serum should be carried out without delay Dacie (1954) 14 advocates that “erythrocytes sensitised by cold antibodies must be washed in saline warmed to 37° C in order to elute agglutinins' However, Dunsford and Bowley, 1955 56 have not found this necessary of the blood is collected as described above, since the cold antibody is given no chance to become bound to the cells Under these circumstances the use of saline at room temperature (18 20° C) for cell washing has proved satisfactory 1 To make it as certain as possible that the antibody coating the patient's cells will if present, be detected a Serial dilution anti-globulin test should be performed The technique is similar to that of the Direct anti globulin test except that the anti globulin serum should be prepared in doubling dilutions of from 1/1 to 1/1000 or morc, depending upon the titres which were found to be optimum for the demonstration of gamma globulins If possible at least two different batches of anti globulin serum, both containing anti gamma and anti non gamma components should be employed Red cells coated with gamma and non gammaglobulin antibodies must be used as a positive control to test each dilution of the anti globulin serumy technique is deemed to be the more sensitive and, indeed, Wnight, Dodd, Bouroncle, Doan and Zollinger (1951),15 and The tile ·94 ANTI-GLOBULIN (COOMBS) TEST Shortly after the description of the anti globulin technique was published Boorman, Dodd and Loutit (1946) 13 applied the test to a series of cases of htmolytic anæmia and found that cells from patients suffering from the aquired form gave a positive Direct anti globulin reaction whereas cells from patients with the congenital type of disease did not It was thought at first that here was a splendid tool by which cases of acquired hæmolytic anemia could be distinguished from other clinically similar conditions Later experience has shown that a diagnosis cannot be made in so simple a manner and that the results of the anti globulin test should be interpreted with caution It is still true to say that auto antibodies are never found in congenital hæmolytic anæmia, they may be demonstrated in the majority of cases of the acquired form of the disease, but there are occasions when no antibody can be detected, perhaps because our armament of tests is still inadequate to reveal its presence This should not be taken to imply that the antı globulin test is of little value in hæmolytic anæmia On the contrary provided that it is appreciated that a negative test result does not necessarily exclude the disease, and that positive reactions can occasionally be found in other disease conditions and even in normal individuals, the information obtained from the anti globulin test can be most helpful When an active auto antibody is found in the serum of a patient suffering from hæmolytic anæmia it may have the characteristic behaviour of either a cold or a warm antibody and may be either a gamma or non gamma globulin, or a mixture of both The results obtained in the anti globulin test will depend to some extent upon the correctness of the dilution of the anti globulin serum employed and whether it contains the antibody for the particular type of globulin with which the cells are coated' As has already been mentioned when blood samples are ·APPLICATIONS OF TIE ANTI-GLOBULIN TEST 97 && anti globulin mcthod, its behaviour and specificity can be investigated (Roth and Frumin, 1957) 18 "When, in the performance of an Indirect anti-globulin test, the scrum being investigated appears to react with the antigen on the red cells cmployed, it should not be confidently assumed that the serum contains antibody until it can also be shown that the test cells are not already coated by a normal” cold incomplete antibody In practice this hazard is avoided by the setting up of the cell suspension control which is, in effect, a Direct anti globulin test applied to the cells Washed cells which have not been exposed to the sensıtısıng serum are subjected to the same dilutions of anti globulin serum at the same time as the main test is performed on the cells which have been exposed to the sensitising serum It should then be obvious whether the reaction observed in the main test is due to a coating of normal cold incomplete antibody already present on the red cells, or whether the origin of the antibody now bound to the cells was in the serum undergoing investigation PLATELETS, LEUCOCYTES AND THE CORRESPONDING ANTIBODIES The study of blood group antigens in human platelets has always been difficult because of their pronounced tendency to adhere together spontaneously whenever they come into contact with a surface which is not absolutely smooth This viscous metamorphosis, which is part of a physiological mechanism designed to prevent loss of blood from damaged vessels, makes difficult the interpretation of agglutination reactions when studying the antigenic structure of platelets, or when a patient's serum is being examined for the presence of anti platelet antibodies In 1954 Gurevitch and Nelken 19 described a method for overcoming this spontaneous adhesion of platelets and demonstrated the existence of A and B antigens on their H ·96 ANTI-GLOBULIN (COOMBS) TEST Hare, Heck and Mathieson (1954),18 have cast doubt on the value of the tube method of performing the anti globulin test in the study of hæmolytic anæmia That the red cells are not alone in being coated by antibody is suggested by the finding of Nelken and Gurevitch (1956)" that the patients platelets may also give a positive Direct antı globulin reaction' hot The Indirect anti globulin test is used to study the antibody content of the patient's serum It is advisable to use washed cells of known antigenic structure in order to be able to distinguish an antibody with group specificity from the non specific type The serum should be fresh since complement must be available before a cold antibody can manifest its presence by attachment to the red cells in the initial stages of the investigation, when it is not known whether the antibody is of the cold or the warm variety, tests should be performed both at 20°C and at 37° C'A particularly avid cold antibody with a wide thermal range may show such activity at 37 C that the investigator is misled into thinking that a warm antibody is also present This pitfall can be avoided by heating the washed cells and the patients serum separately to 37 C before mixing them together The capacity of a cold antibody to sensitise cells can often be intensified if the patient s serum is acidified by the addition of N/4 hydrochloric acid (in the proportion of one volume of acid to nine volumes of serum) before subjecting the cells to its action The serum of a patient with acquired hæmolytic anæmia very often reacts with all the cells with which it comes in contact, suggesting that a mixture of antibodies is present, some of which are coating the red cells while others are free in the serum The antibody which will be of most importance to the patient is the one with which his own cells are coated When this antibody has been eluted off his cells it can be used in the place of a sensitisiog serum and, by the Indirect ·APPLICATIONS OF THE ANTI-GLOBULIN TEST 97 anti globulin method, its bchaviour and specificity can be investigated (Roth and Frumin, 1957) 26 "When, in the performance of an Indirect anti globulin test, the scrum being investigated appears to react with the antigen on the red cells employed, it should not be conſidently assumed that the serum contains antibody until it can also be shown that the test cells are not already coated by a "normal" cold incomplete antibody In practice this hazard is avoided by the setting up of the cell suspension control which is in effect, a Direct anti globulin test applied to the cells Washed cells which have not been exposed to the sensitising serum arc subjected to the same dilutions of anti globulin serum at the same time as the main test is performed on the cells which have been exposed to the sensitising serum It should then be obvious whether the reaction observed in the main test is due to a coating of normal cold incomplete antibody already present on the red cells, or whether the origin of the antibody now bound to the cells was in the serum undergoing investigation PLATELETS, LEUCOCYTES AND THE CORRESPONDING ANTIBODIES The study of blood group antigens in human platelets has always been difficult because of their pronounced tendency to adhere together spontaneously whenever they come into contact with a surface which is not absolutely smooth This viscous metamorphosis, which is part of a physiological mechanism designed to prevent loss of blood from damaged vessels, makes difficult the interpretation of agglutination reactions when studying the antigenic structure of platelets, or when a patient's serum is being examined for the presence of anti-platelet antibodies In 1954 Gurevitch and Nelken 19 described a method for overcoming this spontaneous adhesion of platelets and demonstrated the existence of A and B antigens on their 11 ·96 ANTI GLOBULIN (COOMBS) TEST Hare, Heck and Mathieson (1954) 18 have cast doubt on the value of the tube method of performing the anti globulin test in the study of hæmolytic anæmia That the red cells are not alone in being coated by antibody is suggested by the finding of Nelken and Gurevitch (1956) 17 that the patient s platelets may also give a positive Direct antı globulin reaction" The Indirect anti globulin test is used to study the antibody content of the patient's serum It is advisable to use washed cells of known antigenic structure in order to be able to distinguish an antibody with group specificity from the non specific type The serum should be fresh since complement must be available before a cold antibody can manifest its presence by attachment to the red cells in the initial stages of the investigation, when it is not known whether the antibody is of the cold or the warm variety, tests should be performed both at 20°C and at 37 C'A particularly avid cold antibody with a wide thermal range may show such activity at 37 C that the investigator is misled into thinking that a warm antibody is also present This pitfall can be avoided by heating the washed cells and the patient s serum separately to 37° C before mixing them together "The capacity of a cold antibody to sensitise cells can often be intensified if the patient s serum is acıdified by the addition of N/4 hydrochloric acid (in the proportion of one volume of acid to nine volumes of serum) before subjecting the cells to its action The serum of a patient with acquired hæmolytic anæmia very often reacts with all the cells with which it comes in contact, suggesting that a mixture of antibodies is present some of which are coating the red cells while others are free in the serum The antibody which will be of most importance to the patient is the one with which his own cells are coated When this antibody has been eluted off his cells it can be used in the place of a sensitising Serum and, by the Indirect ·APPLICATIONS or TUT ANTI-GLOBULIN TEST 99 a papain solution and are then incubated with the serum suspected of containing antibody After a thorough wash to remove unbound antibody the cells are mixed and incubated with a suitably diluted anti globulin serum This mixture is finally added to a suspension of well washed D positive red cells which have been previously coated with anti D antibody. Ir no agglutination non occurs the antiglobulin serum must already have been used up by absorption into combination with an antibody coating the platelets or leucocytes with the result that it appears to be inactive when mixed with red cells which are known to be coated with anti D antibody If agglutination of the sensitised red cells does occur the anti globulin serum has not been "consumed " by the platelets or leucocytes and it can be assumed, therefore that the serum under examination did not contain an antibody specific for the antigens on the platelet or leucocyte surfaces OTHER APPLICATIONS OF THE ANTI GLOBULIN TEST "Bacteriology Fifty years ago, in his second paper, Moreschi (19086) 25 described the acceleration and strengthening of bacterial agglutination reactions by what we now call the antiglobulin technique and since its re discovery by Coombs * and his colleagues in 1945 any investigators have used this method in studying the agglutination of bacteria As in other fields it has been shown that antibodies directed against bacterial antigens may be of either the complete or incomplete variety or a mixture of both and, furthermore, the titre of the antibody is apparently enhanced when tested by the anti globulin method Morgan and Schultze (1946) 27 reported the finding in human sera of incomplete antibodies specific for Shigella organisms and anti-Typhoid antibodies could also be ·98 ANTI-GLOBULIN (COOMBS) TEST surface of these were present in the red cells In 1955 Coombs and Bedford 20 subjected suspensions of platelets to the action of anti-A or anti B antibody in order to allow absorption of the antibody to the platelet surface When the platelets were washed free from uncombined antibody they were mixed with weak suspensions of A, B or O red cells Platelets from a group-A person, treated with anti A serum, would cause the agglutination of A red cells, but would not cause clumping of B or O cells The mixed platelet erythrocyte agglutination was specific and could only occur providing the platelets and the red cells had antigens in common If the antibody first taken up by the platelets was of the incomplete type, then, after washing and mixing with the red cells, no agglutination would occur However, Coombs Marks and Bedford (1956) 21 overcame this obstacle by applying the Direct anti globulin method to the platelet erythrocyte mixture In their experimental work they were able to demonstrate anti platelet antibodies produced in rabbits and although in 34 cases of idiopathic thrombo cytopenia in human patients they were unable to prove the existence of anti platelet antibodies the technique has seemed to show sufficient promise to make it worth pursuing Many workers have demonstrated the occurrence of platelet sensitisation by employment of the Direct antı globulin test In 1956 Jaeger-Draafsel, Weigman and van Loghem? used the technique of Dausset and Malinvaud (1954) 3 to search for anti platelet antibodies in the serum of 159 patients and by this method they showed that more than 50 per cent of the patients had an incomplete antiplatelet antibody The anti globulin consumption or utilisation test modification devised by Moulinier (1956) can also be of assistance in the study of antibodies against platelets or leucocytes The platelets or leucocytes are first treated with a ·101 APPLICATIONS O1 TIIT ANTI GLOBULIN TEST cerebro spinal fluid contains small quantities of globulins, but that in certain pathological conditions such as syphuilis the amount of globulin is increased By the anti globulu wubwon_jachungue the globulin content of a solution may be investigated Varying strengths of the solution arc mixed with serial dilutions of anti globulin scrum and are left for at Icast 30 minutes at room temperature after which cach mixture is tested with well washed sensitised cells If agglutination of the sensitised cells occurs the anti globulin serum cannot have been inactivated by globulins in the solution But if no agglutination results the solution must have contained globulins which, by their reaction with the anti globulin serum, appear to inhibit its capacity to agglutinate coated ceils This exposition of the method is correct so far as it goes but the results must be interpreted with circumspection It is possible, for example, that a solution containing nongamma globulns would neutralise the activity of the antinon gamma fraction of an anti globulin serum, while leaving the antı gamma globulin moiety free to react with cells coated with gamma globulin antibody, the resultant agglutination would suggest erroneously that the solution had been globulin free Since it is possible to prepare special anti globulin serum which will combine with gamma globulins only, or with non gamma globulins only, it is practicable to measure both the gamma and the non gamma globulins in a solution 1f still purer anti globulin sera could be prepared the anti globulin inhibition technique might profitably be used to supplement electrophoretic methods of serum analysis (Grubb, 1956) 38 la table x is shown the pattern of results to be expected when traces of normal serum are mixed with serial dilutions of an anti globulin serum The "indicator' cells were group o Rh positive red cells coated with incomplete anti-D antibodies (which have been shown to be gamma globulins) ·100 ANTI-GLOBULIN (COOMBS) TEST demonstrated by the anti globulin technique The test was applied by Griffiths (1947) 28 to the study of Brucellosis and by Coombs and Stoker (1951) -' in investigations of Q fever Middlebrook and Dubos (1948) 30 devised a modified technique with which to examine sera obtained from patients suffering from tuberculosis Concentrated aqueous extracts of tubercle bacıllı were adsorbed on to the surface of red cells in order to produce an agglutinable antigen Admixture of serum from a patient who had proved tuberculosis would usually agglutinate the red cells but occasionally no such reaction was observed By performing an anti globulin test it could be shown that the lack of agglutination was due to the presence of incomplete antibody Those readers who may require more detailed information will be interested in the writings of Hall and Manion (1951),31 Mynell (1952) and Haberman (1955) 33 More recently, when using the test to observe the reactions of antı sera against various organisms, Ford and De Falco (1956) 34 concluded, “Such diseases as those caused by animal parasites, allergens, fungi and viruses may possibly be detected with much more facility by applications of the Coombs test or by modifications of this technique We believe that a wider application of this technique in serological tests may be well worth the effort Estimation of Globulins in Serum and other Fluids It has been reiterated repeatedly that the successful performance of an anti globulin test depends to a great extent upon the thorough removal by washing of the coated cells of unbound globulin or serum which would neutralise the activity of the anti globulin serum employed This knowledge that small quantities of globulia in solution will inbibit the ability of anti globulia serum to agglutinate coated cells was ingeniously applied by Wiener Hyman and Handman (1949) 35 when they demonstrated that normal ·APPLICATIONS or tiir ANTI-GLOBULIN TEST 103 In contrast, tables xi and all illustrate the icry different disposition of rcactions given by scrı from two patients The patient who suffered from multiplc mycloma had the characteristic hyperglobulintmia associated with this discisc so that cien in high dilution his scrum inhibited the antiglobulin serum, whereas the scrum from a small child afflicted by congenital agammaglobulinemia had very little neutralısıng effect It follows, therefore, that the anti globulin inhibition test can be used in the diagnosis and in the control of treatment of agammaglobulinemia The results of this test can be obtained more quickly than can those of biochemical methods and Muralt, Barandun and Hassig (1957) »? claim that the test will detect gamma globulins in concentrations too low to be detected by electrophoresis Forensic Medicine Just as fluid human scrum will inhibit the action of anti human globulin serum, so also will dried human sera or blood stains Allison and Morton (1953),38 using the antihuman globulin inhibition method, claimed that it could identify with absolute specificity dried blood or serum stains of human origin which were ten years old, and that it worked successfully with mummified remains at least 5000 years old Anderson (1954) 39 while agreeing that the test was more specific than the standard anti-human precipitin test, doubted whether it was so sensitive He found that chimpanzee blood and serum were as efficient as human blood in inhibiting the action of antı human globulin serum though Rhesus monkey serum had no effect It would be interesting to discover whether a chimpanzee antı human globulin serum could be prepared which might prove to be a more specific reagent 'The first step in the identification of stains is to cut out a small piece of the stained material and to take also an unstained piece of the same size from the same material, ·102 ANTI-GLOBULIN (COOMBS) TEST +IIIIIII Non cells Sensitised sensitised cells ++++++++ 51200 +++++1111 25600 51200 +++++IIII 12800 +++++III 6400 +++++!!!! TASLEY 3200 ++++iiiii Serial Dilutions-Anti A 758 1600 +++!!!!!! 800 ++1 400 ++! Il 200 Dilutions Serial AHG 65 1 100 ន៩ម៉ឺនដទី១៦, ·APPLICATIONS or THIL ANTI-GI OBULIN TEST 105 which will scnc as a control The two pieces are placed in separate but cqual, amounts of the same standard inti human globulin scrum and are Icft oscright On the following day the test fluid and the control fluid arc cach titrated separatcly against both scnsitised and non sensitised red cells The cells used arc human group o Rh positic crythrocytes, onc portion of which has been coated with incomplete anti D antibody while the other portion has been left untreated The point of testing the control ſuid is to show whcthcr thc matenal alone has any inhibitory cffect on the anti globulin serum When the test fluid is mixed with the sensitiscd cells and no agglutination results, prouded that the control tests have been satisfactory, the stain can be assumed to be of human origin Hæmolytic Disease of the Newborn m Animals Consequent upon the discovery of the underlying cause of hæmolytic disease of the newborn in human babies, it was natural to turn attention to the young of animals and to try to find out whether any of the neonatal deaths occurring in other species could be attributed to similar blood group incompatibility mechanisms 'It was soon established that hæmolytic disease of the newborn could occur in horses (Coombs, Crowhurst, Day, Heard, Hinde, Hoogstraten and Parry, 1948 49), in mules (Caroli and Bessis, 19474), and in pigs (Buxton Brooksbank and Coombs, 1955 4 ) and it has been experimentally produced in animals of other Cronin (1955) 43 considers that 1 per cent of thorough bred racehorses in the Newmarket area have foals afflicted by hæmolytic disease of the newborn and Brion (1949) 4 estimated that in the mule breeding region of France the incidence in mule foals is as high as 8 per cent The incidence in pigs has not yet been published, but even if the frequency is only 0 5 per cent then the loss from the disease in these three species of animals is of economic importance species too ·104 Serial Dilutions AHG 65 TABLE X Serial Dilutions-HS Serum-Multiple Myeloma 400 800 1600 3200 6400 12800 1 100 200 25600 51200 ++------- ++||-- +++1 ||||| +++111111 +++111111 ANTI GLOBULIN (COOMBS) TEST Serial Dilutions AHC 61 |1 100 TABLE XII Serial Dilutions–EJM Serum-Agammaglobulinzmia 400 800 1600 3200 6400 12800 25600 51200 ZOO ++++++---- ++++ 十十十十十十十+11 | 1+十十十十十十十 1|十十十十十十十+ 十十十十十+++-- 十十十十十十十十十 +++++++++-+- ++++++十+++ 十十十十十十十十十+ ++++ ·APPLICATIONS 01 Tur ASTI GI ONULIN TIST 107 bv a foster mother'whosc milk does not contain the noxious antibodies The Indirect anti globulin test has been used to scarch for antibodics in the scrn of mares and sous and it has been possible to demonstrate in these ser both compleic and incomplete antibodies which will rcact with the red cells of their matcs' Further studies of the blood groups of horses and pigs should cnablc broeders to select suitablc sires for their animals, thus avoiding the intcrcntion of htmolytic disease with its conscqucnt cconomic losses In the period 1948-1952 an crccllent series of papers was published by Young and his collcagucs, who chose dogs as the animals in which to study experimentally produced hæmolytic disease and various transfusion problems The logical first approach was to delineate the canınc bloodgroup systems Among other interesting observations, Christian, Ervin and Young (1951) es demonstrated that dogs may have an antigen, A', with serological characteristics which are very similar to those of the Du antigen in man Like Du, the A1 antigen is readily detected by the action of dog anti-A serum in the Indirect anti globulin test Swisher and Young (1954) 46 were able to prove by the same method that canine antibodies may exist in the incomplete form Hæmolytic disease of the newborn was experimentally produced in pups and, although at birth the Direct antiglobulin test applied to the pups red cells gave a negative result, the reaction became positive when they were suckled by their mother (Young, Christian, Ervin, Davies, O Brien, Swisher and Yuile, 1951) 47 Studies of blood groups and experimental hæmolytic disease of the newborn in rabbits have shown that the anti globulin technique is an exceedingly useful tool with which to detect incomplete antibodies and to demonstrate sensitisation of cells Heard, Hinde and Mynors (1949), 48 Kellner and Hedal (1952),49 and Anderson (1956) 5° ·106 ANTI-GLOBULIN (COOMBS) TEST The ætiology of the disease in animals differs slightly from the mechanism producing human hæmolytic disease of the newborn The placentæ of most animals have several chorio-epithelial layers which provide a greater barrier to the passage of antibodies from mother to fætus than does the single layer of the human placenta As a result, the offspring are born healthy, having been unaffected in utero by maternal antibody, but they succumb a few days later to the effects of maternal antibody secreted in the colostrum and mother s milk The blood-group systems of horses, donkeys and pigs have not been studied to the same extent as those in man and knowledge about the behaviour of the antigens and antibodies involved is less complete Just as in man, the discovery in the laboratory of serological differences in the blood groups of mother and offspring does not mean necessarily that incompatibility exists which will have clinical repercussions' Before making a diagnosis of hæmolytic disease it is necessary to show that antibodies directed against antigens on the red cells of the fætus are present in the maternal serum and in the colostrum and that these antibodies are able to coat or destroy the red cells of the young animal Confirmation that hæmolytic disease of the newborn is the underlying cause of some cases of anæmia in sucklings is provided by the finding that the red cells, which give a negative Direct anti globulin reaction at birth subsequently show evidence of sensitisation in a positive Direct anti globulin result after the ingestion of colostrum For the performance of such tests special anti globulin sera must be prepared In the study of the disease in pigs rabbit anti pig globulin serum is used in investigations involving horses and mules rabbit antı horse globulin serum is the appropriate reagent Once the diagnosis has been made the anzemia may be cured by transfusion or the young animals may be suckled ·APPLICATIONS OF TIIE ANTI-GLOBULI TEST 109 against the conjugate There may be many other applications of this test in the study of proteins, polypeptides and polysacchandes of biological and pathological importance It has great possibilities in the sphere of allergy A problem of blood group scrology which still awaits full investigation is the determination of the number of antigen receptors, pertaining to cach blood group system on the red cells Thc usc of a radioactive iodinc method led Boursnell, Coombs and Rızk (1953) S4 to postulate that 5500 Rh (D) antigen receptors were available on cach red cell Wiener and Gordon (1953) described a quantitative anti globulin tcst which they considered could be applıcd to determine the relative number of anugen receptors on the red cells for the various blood group systems The same problem was tackled by Grubb (1955),55 using the antiglobulin technique in a rather different way Suspensions of well washed Rh positive red cells were prepared in volumes known to contain about 109 cells To each volume of cells was added a measured volume of anti-D serum in various dilutions and the mixtures were then incubated to allow the cells to become coated After thorough washing to remove free antibody, measured volumes of diluted anti globulin serum were added to the sedimented cells At the same time identical dilutions of anti globulin serum were mixed also with standard solutions of gamma globulin A comparison was then made of the ability to agglutinate further amounts of coated cells by the anti globulin solution neutralised by red cells and by the anti globulin solution inhibited by standard gamma globulin From this could be found the amount of gamma globulin in a particular dilution of anti D antibody which combined with a known number of red cells and, assuming that the molecular weight of gamma globulins is 165,000, Grubb calculated that there must be a minimum of 2000 receptors on each red cell ·108 ANTI-GLOBULIN (COOMBS) TEST In other species of animal, cattle, for instance, the blood group systems have been intensively investigated, but much remains to be discovered and the veterinary scientist will find the anti-globulin technique a necessity for a complete study Immunological Research Coombs, Gleeson White and Hall (1951) 51 devised a multi-stage anti globulm technique, building up an antı globulin globulin lattice in order to effect agglutination of certain bovine red cells which had been sensitised by heterophile antibody, but which were not agglutinated by the ordinary single stage anti globulin test (Gleeson White, Heard, Mynors and Coombs, 1950) 5. These two papers made it clear that, in at least one species although the red cells may possess a particular antigen, it will only be detected by superimposing on the cells specifically combined alternate layers of antibody globulin anti-globulin antibody globulin and anti globulin untıl agglutination occurs "It has been recognised for many years that the agglutina tion reaction provides a much more sensitive method than precipitation for demonstrating the specific antibody content of serum When the antigen is a soluble protein and the precipitin antibody is of the incomplete type any interaction which occurs between them will not be readily detectable Coombs, Howard and Wild (1952),53 attempting to find a way in which to recognise that such reactions had taken place contrived a sensitive agglutination method in preference to precipitation They were able to conjugate the soluble protein antigen with an incomplete anti D antibody and then, by mixing the conjugate with Rh positive cells to coat those cells with anti D antibody conjugated with protein antigen This coating could be detected either by testing the red cells with antı human globulin serum or by the use of a specially prepared anti globulin serum active ·APPLICATIONS OF THE ANTI-GLOBULI TEST 111 . 19 GURETAT J and LEY D 1954 ABO groups in blood platelets J Lab & Clon Med 44 562. 20 COOMBS R R A and BIDIORD D 1935 Thic A and B antigens on human platelets demonstrated by means of mited crythrocyte platcict agglutination I or Sanguinis 5 111 21 Cooves RRA MARAS J and BEDFORD D 1956 Specific mired asglutination Mixed crythrocyte platcict anti globulin reaction for the detection of platelet antibodies Drul J Hamal II 84 22. JAEGER DRAAFSEL, ILGVAN D and VAN LOGIEM J J 1956 Detec tion of incomplete platelet antibodies by means of the Indirect anli human globulin test I ox Sanguinis 1 78 23 DAUSSET J and MALYVAUD J 1954 Influence de l agitation sur I asglutination thrombocytaire Sang 25 887 24 MOULINTER J 1956 Technique de la réaction de consommation danti globulinc Rev Franc d'erudes Clin er Biol 1 355 25 MORESCHI C. 1908b Beschleunigung und Berstarkung der Backterien agglutination durch Anticiweissera Zentr Bacteriol Parasitenk 46 456 26 COOVES RRA MOURANT A E and RACE R R 1945 Detection of weak and incomplete Rh agglutinins A new test Lancet 11 15 27 MORGAN W T J and SCHULTZE H 1946 Non agglutinating antibody in human antı-sera to Shigella shiga and Salmonella typhi Brit J Exp Path 27 286 28 GRIFFITHS JJ 1947 Agglutination and agglutinin blocking property in serums from known cases of Brucellosis Pub Health Rep 62 865 29 COOMBS RRA and STOKER R 1951 Detection of Q fever antı bodies by the anti globulin sensitisation test Lancet 11 15 30 MIDDLEBROOK G and DUBOS RJ 1948 Specific serum agglutina tion of erythrocytes sensitised with extracts of tubercle bacillus J Expt Med 88 521 31 HALL WH and MANION RE 1951 Hæmagglutinins and hæmolysis for erythrocytes sensitised with Tuberculin and Pulmonary Tuberculosis J Clin Invest 30 1542 32 MYNELL GG 1952 A sensitive modification of the Middlebrook Dubos hæmagglutination test Path Back 64(3) 647 33 HABERMAN S 1955 Detecting Tuberculosis Antibodies Brut Med J 11 620 34 FORD AC and DEFALCO JR 1956 Studies on bacterial agglutina tion by the use of the anti globulin (Coombs ) technique Canad jos Microbiol 2 657 35 WIENER AS HYBAN MA and HANDMAN L 19.49 serological test (Inhibition test) for human serum globulin Proc Soc Exp Biol and Med 71 96 36 GRUBB R 1956 A method of demonstrating minute amounts of human a globulin Acta Path et Microbiol Scand XXXVIII 339 A new ·110 ANTI GLOBULIN (COOMBS) TEST REFERENCES I TOVEY G 1950 Practitioner 164 171 2 CHOWN B 1954 Anæmia from bleeding of the fetus into the mother s circulation Lancet 1 1213 3 CHOWN B 1955 The Fetus can bleed Am J Obst & Gynec 70 1298 4 DUNSFORD I 1957 Proof of fætal antigens entering the maternal circulation Vox Sanguinis 2 125 5 ROSENFIELD RE 1955 AB hæmolytic disease of the newborn Analysis of 1480 cord blood specimens with special reference to the direct anti globulin test and to the group O mother Blood 10 17 6 MOLLISON PL 1956 Blood Transfusion in Clinical Medicine 2nd Edition published by Blackwells Oxford 7 JAKOBOWICZ R KRIEGER VI and SIMONS RT 1948 The value of the Coombs test in detection of iso sensitisation of the new born Med Austral August 7 143 8 HOLLANDER V L 1952 Inkomplette kaltæntikörper als Ursache des positiven antiglobulatestes beim neugeborenen Kinde kehelica Pæd Acra 7 512 9 WIENER AS and GORDON EB 1953 Quantitative test for antibody globulin coating human blood cells Am J Clin Path 23 429 10 GREENWALT T J and WAGNER JA 1955 Evaluation of the quantitative Direct anti globulin test in Hæmolytic Disease of the Newborn due to anti D Blood x 29 11 FULTON ROBERTS G 1957 Comparative aspects of Hæmolync Disease of the Newborn Published by William Heinemann London 12 DRUMMOND RJ and WATKINS A G 1946 The Rh factor and hepatomegaly and splenomegaly in children and adolescents Brit Med J 1984 13 BOORMAN KE DODD BE and LOUTIT J F 1946 Hæmolytic Icterus (Acholuric jaundice) Congenital and Acquired Lancet 1 812 14 DACIE J V 1954 The Hæmolstic Anæmias published by Churchill London 15 WRIGHT CS DODD MC BOURONCLE BA DOAN C A and ZOLLINGER PM 1951 Studies of hemagglutinins in hereditary spherocytosis and acquired hemolytic anemia their relationship to the hypersplenic mechanism Lab & Clin Med 37 165 16 HARE R. L HECK F J and MATHIFSON DR 1954 The anti globulin (Coombs ) test as an aid in the diagnosis of hemolytic syndromes J Lab & Clin Med 43 867 17 NELKEN D and GUREVITCH J 1956 Positive Coombs test in blood platelets Vox Sanguinis 1 106 18 ROTH K L and FRUMIN AM 1957 Acquired hemolytic anemia Properties of eluted agglutinins and their interaction with red blood cells and Coombs serum Blood x 217 ·APPLICATIONS OF THE ANTI-GLOBULIN TEST 111 19 GUREVITAI J and NELLEN D 1954 ABO groups in blood platelets Lab & Clin Ned 44 562 20 COOMBS RRA and BEDFORD D 1955 The A and B antigens on human plateicts demonstrated by means of mixed crythrocyte platelet agglutination ,lor Sanguinis 5 111 21 COOMBS RRA Maras J and BEDFORD D 1956 Specific mured agglutination Mixed crythrocyte platclct anti globulin reaction for the detection of platelet antibodies Brit Namat 1 84 22 JAEGER DRAAFSEL, WEIGYAN D and VAN LOGHIEM J J 1956 Detec tion of incomplete platelet antibodies by means of the Indirect antt human globulin test I or Sanguinis 1 78 23 DAUSSET J and MALINVAUD J 1954 Influence de l agitation sur I agglutination thrombocytaire Sang 25 847 24 MOULINIER J 1956 Technique de la réaction de consommation danti globuline Rex Franc d etudes Clin et Biol 1 355 25 MORESCHII C 19085 Beschleunigung und Berstarkung der Backterien agglutination durch Antieiweissera Zentr Bacteriol Parasitenk, 46 456 26 COOvBS RRA MOURANT AE and RACE RR 1945 Detection of weak and incomplete Rh agglutinins A new test Lancet 11 15 27 MORGAN V TJ and SCHULTZE H 1946 Non agglutinating antibody in human antı sera to Shigella shiga and Salmonella typhi Brit J Expt Path 27 286 28 GRIFFITHS J J 1947 Agglutination and agglutinin blocking property in serums from known cases of Brucellosis Pub Health Rep 62 865 29 COOMBS RRA and STOKER R 1951 Detection of Q fever antı bodies by the anti globulin sensitisation test Lancet 11 15 30 MIDDLEBROOX G and DUBOS RJ 1948 Specific serum agglutina tion of erythrocytes sensitised with extracts of tubercle bacillus Expt Med 88 521 31 HALL W H and MANION RE 1951 Hæmagglutinins and hæmolysis for erythrocytes sensitised with Tuberculin and Pulmonary Tuberculosis Clin Invest 30 1542 32 MYNELL GG 1952 A sensitive modification of the Middlebrook Dubos hæragglutination test J Path Bact 64(3) 647 33 HABERMAN S 1955 Detecting Tuberculosis Antibodies Brit Med J 11 620 34 FORD A C and DEFALCO JR 1956 Studies on bacterial agglutina tion by the use of the anti globulin (Coombs ) technique Canad J of Microbiol 2 657 35 WIENER AS HYMAN MA and HANDMAN L 1949 A new serological test (Inhibition test) for human serum globulin Proc Soc Exp Biol and Med 71 96 36 GRUBB R 1956 A method of demonstrating munute amounts of human a globulin Acta Path et Microbiol Scand XXXVII 339 ·110 ANTI-GLOBULIN (COOMBS) TEST REFERENCES 1 TOVEY G 1950 Practitioner 164 171 2 CHOWN B 1954 Anæmia from bleeding of the fetus into the mother s circulation Lancet 1213 3 CHOWN B 1955 The Fetus can bleed Am J Obst & Gynec 70 1298 4 DUNSFORD I 1957 Proof of fatal antigens entering the maternal circulation Vor Sanguinis 2 125 5 ROSENFIELD RE 1955 AB hæmolytic disease of the newborn Analysis of 1480 cord blood specimens with special reference to the direct antı globulin test and to the group O mother Blood 10 17 6 MOLLISOY PL 1956 Blood Transfusion in Clinical Medicine and Edition published by Blackwells Oxford 7 JAKOPOWICZ R KRIEGER VI and SIMMONS RT 1948 The value of the Coombs test in detection of iso sensitisation of the new born Med Austral August 7 143 8 HOLLANDER V L 1952 Inkomplette Kaltæntikorper als Ursache des positiven antiglobulintestes beim neugeborenen Kinde Kelvetica Pæd Acta 7 512 9 WIENER AS and GORDON EB 1953 Quantitative test for antibody globulin coating human blood cells Am J Clin Path 23 429 10 GREENWALT TJ and WAGNER JA 1955 Evaluation of the quantitative Direct anti globulin test in Hæmolytic Disease of the Newborn due to antı D Blood 29 11 FULTON ROBERTS G 1957 Comparatne aspects of Hamolytic Disease of the Newborn Published by William Heinemann London 12 DRUMMOND RJ and WATKINS A G 1946 The Rh factor and hepatomegaly and splenomegaly in children and adolescents Brut Med J 1984 13 BOORMAN KE DODD BE and LOUTIT J F 1946 Hæmolytic Icterus (Acholuric jaundice) Congenital and Acquired Lancet 1 812 14 DACIE J V 1954 The Hæmolytic Anæmias published by Churchill London 15 WRIGHT CS DODD MC BOURONCLE BA DOAN CA and ZOLLINGER PM 1951 Studies of hemagglutinins in hereditary spherocytosis and acquired hemolytic anemia their relationship to the hypersplenic mechanism J Lab & Clin Xfed 37 165 16 HARF R L HECK FJ and MATHIESON DR 1954 The anti globulin (Coombs ) test as an aid in the diagnosis of hemolytic syndromes J Lab & Clin Ned 43 867 17 NELKEN D and GUREVITCH J 1956 Positive Coombs test in blood platelets Vox Sanguinis 1 106 18 ROTK K L and FRUMIN A M 1957 Acquired hemolytic anemia Properties of eluted agglutinins and their interaction with red blood cells and Coombs serum Blood XTI 217

·APPLICATIONS or THE ANTI GLOBULIN TEST 113
$3 COOMES RRA HOWARD A N and Wild r 1952. Titration of
anti-scra to soluble proteins on the basis of an agglutination reaction
Brut J Exp Path XXX 390
54 BOURSNELL, J C. Coomas R R A and Rizk V 1953 Studics with
maried antı 131 sera
quantitative studies with antı sera marked with
1odinc isotope and their corresponding red-cell antigens Biochem
Journ 55 745
SS GRUBB R 1955 An estimate of the number of Rh receptors on a
single red cell Acra Gener of Statis! Med 5 377
56 DUSFORD I and BOWLEY CC 1935 Techniques in Blood
Grouping Published by Oliver and Boyd Edinburgh

·112 ANTI-GLOBULIN (COOMBS) TEST 37 MURALT V G DE BARANDUV and HASSIO A 1957 Die antiglobulia hemmungsreaktion als untersuchungsmethode bei Hypo-und agamma globulinamien Sonderabdruck ans der Schwer erischen Medi unischen JVochenschrift 87 467 38 ALLISOV AC and MORTON JA 1953 Species specificity in the inhibition of anti globulin sera J Clin Path 6 314 39 ANDERSON J 1954 Application of antiglobulin reaction to blood stains in determination of animal species Am J Clin Path 24 8 40 COOMBS RRA CROWHURST RC DAY FT HEARD D H HINDE IT HOOGSTRATEN J and PARRY HB 1948 Hæmolytic disease of newborn foals due to iso-immunisation of pregnancy Hyg Camp 46 403 41 CAROLI J and BESSES M (1947) Sur la cause et le traitment de 1 ictère grave des muletons nouveau nés CR Acadm Sci 224 969 42 BUXTON JC BROOKSBANK N H and COOMBS RRA 1955 Hæmolytic disease of new bom piglets caused by maternal isoimmunisation Brit et j 111 463 43 CRONIN MTI 1955 Hæmolytic disease of newborn foals et Rec 67 479 44 BRION A 1949 Sur l ictere hemolytique de Poulain nouveau né Rex med set 100 229 45 CHRISTIAN RM ERVIN D M and YOUNG LE 1951 Observations on the in vitro behaviour of dog iso antibodies J Immunol 66 37 46 SWISHER SM and YOUNG LE 1954 Studies of the mechanisms of erythrocyte destruction initiated by antibodies Trans Ass Amer Physns 67 124 47 YOUNG LE CHRISTIAN RM ERVIN D M DAVIS RW OBRIEN WA SWISHER SN and YUTLE CL 1951 Hæmolytic disease in newborn dogs Blood 6 291 48 HEARD DH HINDE IT and MYNORS LS 1949 An experimental study of hæmolytic disease of the newborn due to iso immunisation of pregnancy I An attempt to produce the syndrome in the rabbit J Hyg Camb 47 119 49 KELLNER A and HEDAL EF Experimental erythroblastosis fetalis Amer J Path 29 539 50 ANDERSON JR 1956 The experimental production of erythroblastosis fatalis in rabbits Brit J Hæmat 1 378 51 COOMBS RRA GLEESON WHITE MH and HALL J G 1951 Factors influencing the agglutinability of red cells II The agglutina tion of bovine red cells previously classified as inagglutinable by building up of an anti globulin globulin lattice on the sensitised cells Brit J Exp Path 32 195 52 GLEESON WHITE MH HEARD DH MYNORS LS and COOMBS R. RA 1950 Factors influencing the agglutinability of red cells the demonstration of a variation in the susceptibility to agglutination exhibited by the red cells of individual oren Brut J Exp Path XXXI 321 1952

·APPLICATIONS OF THE ANTI-GLOBULIN TEST
113
S3 Cooves, RRA TOH ARD A N and WILD r 1992 Titration of
anti-sera to soluble proteins on the basis of an asglutination reaction
Bru J Exp Path xxxm 390
54 BOURSNELL, J C. COOYDS RRA and Riza V 1953 Studies with
marked ants 131 scra quantitatic studies with anti-scra marked with
jodine isotope and their corresponding red-cell antigens Biochem
Journ 55 745
55 GRUBB R 1955 An estimate of the number of Rh rcccptors on a
single red cell dcta Genet er Status Afed 5 377
56 DUNSFORDI and BOWLEY C. C. 1955 Techniques in Blood
Grouping Published by Oliser and Boyd Edinburgh
1

·CHAPTER XI ODDMENTS This book is not intended to be a complete review of the anti globulin test, but rather to act as a broad basis on which the student can build It is for this reason that, in this chapter, a few oddments are mentioned which may suggest lines on which future work could be planned THE ANTI GLOBULIN TEST IN CANCER Green, Wakefield and Littlewood (1957), using anti human globulin serum at lower dilutions than standard (1e at dilutions which were more likely to pick up non gamma than gamma globulin antibodies), found that the red cells of 43 per cent of patients with cancer gave a positive Direct anti globulin reaction whereas in non cancerous patients the Direct anti globulin test was positive in 15 per cent The degree of positivity was stated to be related to tumour size and was greater with larger tumour masses Tumours of lymphoid tissue and advanced carcinoma gave the greatest number of positives These authors suggested that the anæmia of malignant disease might be due to an auto antibody produced by the tumour cells and from the evidence presented, the antibody they were detecting was of the non gamma globulin type Dunsford (1957) found that the sera of patients with cancer more often contained auto agglutinins or pan agglutinins than did the sera of patients who did not have cancer' It was an antibody of this type which caused the positive Direct anti globulin test in a case of pernicious anæmia reported by Selwyn and Alexander (1951) 3 and it may have been this sort of antibody which was responsible for the findings of Green and co workers In a small 114 ·ODDIENTS 115 series of 40 cases Dunsford ? observed that the Direct anti globulin tcst was always ncgative of the cells from patients with malignant discasc were tested before they had been allowed to cool to a temperature below 30°C Madden (1956) reported positic Direct anti globulin reactions given by scra obtained from patients suffering from cancer if the samples had been first refrigerated Auto antibodies appear to play a definite role in acquired hæmolytic anemia and in Hashimoto s and other diseases (Roitt, Doniach, Campbell and Vaughan-Hudson, 1956) Furthermore, normal healthy people can be shown to have anti H as a normal incomplete cold antibody in their sera -- Crawford, Cutbush and Mollison (1953) «_and yet at the same time these people have H-substances in their tissues One possible hypothesis advanced suggests that if this auto tolerance should break down in a particular organ the tissue cells may become so modified that they lose their tissue specificity and result in a malignant condition Certainly the theory of the immunological basis of cancer enunciated by Green and his colleagues and the observations recorded by them warrant further investigation THE ANTI GLOBULIN TEST IN LEAD POISONING Sutherland and Eisentraut (1956),? using the tube method of performing the anti globulin test advocated by Rosenfield, Vogel and Rosenthal (1951) 8 found that in the anæmia produced by lead intoxication in humans and in experi mentally induced lead poisoning in dogs, the red cells gave a positive Direct anti globulin reaction Dunsford (unpublished) also has observed a strong positive Direct anti globulin result given by the red cells of a child suffering from severe lead poisoning although negative reactions were obtained in tests on the bloods of three men, workers in the car industry, who had been exposed ·116 ANTI-GLOBULIN (COOMBS) TEST to lead for some time and all of whom had marked symptoms of plumbism“ These interesting observations merit further investigation by those making a special study of lead poisoning THE ANTI GLOBULIN TEST AND OTHER CHEMICALS Following the intravenous administration of phenyl hydrazıne to a dog, Muirhead, Groves and Bryan (1954a)' found that both the Direct and Indirect anti globulin reactions became positive These positive results could not be obtained if phenylhydrazine was added to the dog's blood after collection, in other words, the phenylhydrazine was not able in vitro, to modify the dog s red cells so that they became agglutinable by anti globulin sera. The same team Muirhead, Groves and Bryan (19546) 10 however reported later that the addition of hydrazine to the dog blood in vitro simulated the phenomena observed after in vivo action of phenylhydrazine The authors considered several possible explanations It might be that hydrazine uncovered protein on the dog erythrocytes which would react with the rabbit antı dog antibodies to produce a positive Direct anti globulin result, perhaps the hydrazine altered the surface of the dog erythrocytes in such a way that it absorbed proteins, or it might be that the hydrazine acted as a coupling agent between the red cells and the normal proteins, "Muithead and Groves (1956) " further reported that lipid solvents acting in vno modified dag erythrocytes so that they gave a positive Direct anti globulin reaction As reported by Jandl (1955) 1 and Jandi and Simmons (1957) 13 several metallic cations are capable of attaching proteins to the surfaces of human red cells, thus sensitising" the cells, which can then be agglutinated by anti sera specific for the proteins attachment For instance, anti human globulin serum agglutinated the cells if human 21 ·ODDMENTS 117 globulin had been attached to the red cells by thc action of metallic cations The effect was obscried when thc cations Cr+++, Fc+++, Be+++, Rh+++, and Al+++ ucrc cmployed Red cells so treated can be used as positive controls for the anti-globulin test Such cells will hccp for at Icast a wech A TEST WIIICH MAY SUPLEMENT THE ANTI GLOBULIN TECHNIQUE When a 1 per cent solution of protamine sulphate was added to a salınc suspension of red cells coated with incomplete anti D antibodies agglutination occurred (Roth and Frumin, 1954) "Cells which had not been coated by antibody were not agglutinated and neither werc cells obtained from cases of acquired hæmolytic anemia After further observations Roth and Frumin (1957),15 described the results obtained with this and other tests on red cells from 17 patients suffering from acquired hæmolytic anæmia In some cases agglutination occurred, in other cases it did not This may have been due to a distinction in the globulin coating of the red cells, such as the difference between gamma and non gamma globulins More study is necessary for the understanding of the mechanism and for an evaluation of the helpfulness of this method as a supplementary technique in the investigation of hæmolytic disease REFERENCES 1 GREEN HN WAKEFIELD ) and LITTLEWOOD G 1957 The nature of cancer anæmia and its bearing on the immunological theory of cancer Brit Med J 1779 2 DUNSFORD I 1957 Studies on the sub groups of A in normal people and blood group studies in people with Carcinoma PhD Thesis Sheffield University 1957 3 SELWYN JG and ALEXANDER SS 1951 A positive Coombs reaction in pernicious anzmia Brit Med J 1 564 4 MADDEN HJ 1956 To do or not to do a Coombs test Amer Afed Techn Nov Dec 383 ·118 ANTI-GLOBULIN (COOMBS) TEST S ROITT I M DOVJACHT D CAMPBELL P N and VAUGHAN HUDSON R 1956 Auto antibodies in Hashimoto s disease (Lymphadenoid goitre) Lancet 11 820 6 CRAWFORD H CUTBUSH M and MOLLISOV P L. 1953 Specificity of incomplete cold antibody in human serum Lancet 1 566 7 SUTHERLAND DA and ELSENTRAUT AM 1956 The Direct Coombs test in lead poisoning Blood XI 1024 8 ROSENFIELD RE VOGEL P and ROSENTHAL N 1951 The anti globulin test technical and practical applications Am J Clin Path 21 301 9 MUIRIIEAD EE GROOVES M and BRYAN S 19549 Positive Direct Coombs test induced by phenylhydrazine J Clin Invest 33 1700 10 MUIRHEAD EE GROVES M and BRYAN S 19546 Agglutination of erythrocytes in utro induced by hydrazine including a positive Coombs test Journ Lab & Clin Med 44 902 11 MUIRHEAD EE and GroVES MT 1956 The positive ant globulin (Coombs ) test of canine erythrocytes induced by lipid solvents in utro Am J Clin Path 26 844 12 JANDL JH 1955 Sequestration by the spleen of red cells sensitised with incomplete antibody and with metallo-protein complexes Clin Invest 34 912 13 JANDL JH and SJMMONS R L 1957 The agglutination and sensitisation of red cells by metallic cations interactions between multivalint metals and the red-cell membrane Brut | Hemat lil 19 14 ROTHKL and FRUMIN AM 1954 In vitro differentiation between auto and iso immune antibodies by protamine and trypsin Science 120 945 15 ROTH K L and FRUMIN AM Acquired hemolytic anemia Properties of eluted agglutinins and their interaction with red blood cells and Coombs serum Blood X 217 1957 ·INDEX Absorption of unwanted antibody. Anti globulin sera-continued 30-31 Storage of 75 76 Acquired hemolytic anaemia 19 Zoning phenomenon 40-43 45 48 57 74 93 97 Anti species antibody 8 9 16 31 Agammaglobulinemia 47 103 64 66 Agglutinin 2,8 Arlacel A 21 Agglutinoid 2,4 Azide 69 76 Albumin antibody 2 Albumın replacement technique 4 Bacteriology and the anti globulin Alginate solution 22 test 99 Alum precipitated globulin 20 Blocking test 4 Animal blood groups 105 108 Blood stains 103 Antibodies Brucellosis 100 Bivalent antibody 2 5 Cold antibody 19 48 57 94 96 Calgıtex 22 Cold incomplete auto-antibody Cell washing 10 30-31 56-57 59 48 55 59 67 68 69 70 Complete antibody 2 64 78 Co agglutination 70 Hetero-antibody 16 25 30 32 66 Coated cells 27 29 Incomplete antibody 2 4-6 34 Preparation of 36 53 55 63 78 87 In transfusion reaction 82 Monovalent antibody 2 5 Degree of coating 92 Normal cold incomplete antibody Coid sensitised cells 62 95 97 Compatibility test 78 81 Saline antibody 2 6 Complement 3 11 Warm antibody 19 48 94 Components required 48 55 Anticoagulant 30 68 Inactivation of 24 Anti-complementary effect 56 62 In compatibility test 80 68 Lack of 74 Antigen 1 68 18 47 109 Congenital hæmolytic anæmia 93 Antigen antibody reaction I 3 5 Contamination by extraneous globu Antigen receptors on red cells lin 59 72 73 109 By bacteria 68 69 Anti globulin augmentation test 64 Cryptagglutinoid 24 Anti globulin consumption (utilisa tion) test 98 99 Dangerous recipient 79 80 Anti globulin inhibition test 103 105 Developing test 14 Anti globulin sensitisation test 13 Direct anti globulin test 14 Anti globulin sera 8 18 For anti platelet antibodies 98 Antt gamma globulin 47 In cancer 114 115 Anti non gamma globulin 47 In hæmolytic anæmia 94 97 Dilution of 25 35 49 50 60 73 In hemolytic disease of the new Production of 16-26 32 born 88 90 106-107 Processing of 27 33 In lead poisoning 115 Special species 106 In transfusion reaction 82 83 Standardisation of 36 40 Technique 52 54 119 ·120 INDEX Enzyme treated cells 6 36 False results 66-75 Ficin 56 Forensic medicine 103 Phenylhydrazine 116 Platelets 97 Precipitin 28 Prozone 40-43 57 Q fever 100 Quantitative anti globulin test 92 Gamma globulin 18 21 32 47-49 60 62 101 Globulin contamination 59 72 73 Globulin estination of 100 101 Graininess 70 Receptor sites 109 Hæmolysins 24 Hæmolytic anemia 93 Hæmolytic disease of the newborn in animals 105 108 in man 85 93 Heat treatment of sera 32 Hydrazine 116 Sensitisation of cells 36 56 62 Serial dilution anti globulin test 49 95 Silica 61 71 Species specific substances 8 Stains blood 103 Sulpho salicylic acid test 30 Imunisation 1 8 16 21 22 Inactivation of complement 24 Indirect anti globulin test 14 Compatibility test 80 In blood grouping 83 85 In hæmolytic disease 87 88 93 96-97 107 Method 52 64 Infection of serum 69 Techniques and tests anti globulin augmentation test 64 anti globulin consumption test 98 99 anti globulin inhibition test 101 103 105 anti globulin sensitisation test 13 blocking test 4 compatibility test 78 80 81 developing test 14 direct anti globulin test 52 54 hæmolysın test 24 indirect anti globulin test, 52 64 multi stage anti globulin test 108 quantitative anti globulin test 92 serial dilution anti globulin test 49 95 sulpho salicylic acid test 30 Trypsin 56 Tuberculosis 100 Lead poisoning 115 Leucocytes 97 Lysın 28 Metal protein complexes 116 Mixed platelet-erythrocyte agglutina tion 98 Multiple myeloma 103 Multi stage anti globulin technique 108 Veterinary applications 105 Neutral pooled serum 29 Nomenclature 13 15 Non garnma globulins 18 32 47-49 55 60 62 68 101 Washing of cells 10 30-31 56-57 59 69 70 Wharton s jelly 59 90 Wide spectrum reagent 47 80 Pan agglutination 69 Panocell mixture 69 Zoding phenomenop 40 43 57 PIXTAD I GREAT RITAIN SY OLIVE AD BOYD LTD EDI BU GH

·No text on this page.