Erythrocyte Survival in Rheumatoid Arthritis By EDWARD S. MONGANAND RALPHF. JACOX Thirty patients with active rheumatoid arthritis were transfused with chromium51 labeled erythrocytes from either normal or rheumatoid donors. Three sets of identical twins discordant for rheumatoid arthritis had a cross transfusion from a healthy donor to the affected recipient. I n no instance was there any significant shortening of erythrocyte survival which could not be attributed to minor blood group incompatibility. The presence of rheumatoid factor in either the donor, the recipient, or both did not effect erythrocyte survival. Trenta patientes con active arthritis rheumatoidee recipeva transfusiones de erythrocytos a marcage con chromo-51 ab donatores normal o rheumatoidee. In tres pares de geminos identic con discordantia pro arthritis rheumatoidee esseva le subjectos de un experiment0 de transfusion cruciate a b un donator normal a un recipiente afficite. Esseva notate nulle caso de un reduction significative del durata vital del erythrocytos gue non poteva esser attribuite a minor incompatibilitates de gruppo de sanguine. Le presentia de factor rheumatoidee in le donator, in le recipiente, o in ambes non afficeva le durata vital del erythrocytos. T HE QUESTION OF whether there is a normal or a decreased survival of erythrocytes in patients with rheumatoid arthritis is complicated by the fact that some observors have found a normal s u r ~ i v a l ~ while - ~ others have found decreased survival.4.7 None of these reports have given full details about serological identification of the rheumatoid disease, the effects of drug therapy or other variables which might influence results of erythrocyte survival studies. In 1961 Richmond et a1.8 reported that rheumatoid patients had a normal survival of erythrocytes transfused with their own cells, but that a decreased erythrocyte survival occurred when normal, compatible donor cells were transfused into rheumatoid patients. This study raises a question of whether erythrocytes from rheumatoid patients might have some unique advantage of survival over cells obtained from normal individuals. With this question in mind, we have carried out an investigation of erythrocyte survival to determine whether the serological state of the rheumatoid patient might influence erythrocyte survival. Thus our investigation includes not only cross transfusion of cells from seropositive rheumatoid donors into patients with rheumatoid arthritis, but also transfusiod of normal donor cells and cells obtained from patients with seronegative rheumatoid arthritis. A standard labeling technic with radioactive chromium was employed and cross-transfusion experiments were performed in thirty patients, including three sets of identical twins discordant for rheumatoid arthritis. Only four of From the Department of Medicint, The Uniuersity of Rochester School of Medicine G Dentisty, Rochester, New York. Supported b y Grant #CRMC 24, The National Foundation. 481 ARTHRITIS AND RHEUMATISM, VOL. 7, No. 5 (OCTOBER),1964 482 MONGAN AND JACOX the patients had a slightly shortened erythrocyte survival time and in two of these patiedts, minor blood group incompatibilities were demonstrated which could explain the decreased survival observed. METHODS All patients investigated were individuals with rheumatoid arthritis who were actively attending the Arthritis Clinic at the University of Rochester. They were typed for ABO blood group system and for blood groups Cc, D, Ee, and Kell. In a similar manner six normal individuals were typed. When suitable donors and recipients were found, fresh blood specimens were drawn and cross matching was performed. An indirect Coombs test was also done. All patients had active, definite rheumatoid arthritis of a t least one year’s duration. Steinbrocker’s functional class and radiological stagings was known on each paticnt. None of the patients had leukopenia or splenomegaly. Rheumatoid factor was determined by the latex flocculation test of Singer and Plotz.10 Thirty milliliters of blood was obtained by venipuncture and gently ejected into a sterile flask containing 7 i d . of sterile ACD mixture. The blood was then thoroughly mixed by gently swirling the flask. Approximately 75 microcuries of high specific activity sterile radioactive sodium chromate was added to the blood citrate mixture. The citrateblood-chromate mixture was allowed to sit for thirty minutes at room temperature. The sample was then centrifuged in sterile centrifuge tubes and the snpernatant plasma was removed. An equal volume of sterile isotonic saline was added. The cells were thoroughly resuspended in the saline, and centrifugation was repeated. Washing with sterile isotonic saline was carried out twice. Following the second washing, a volume of isotonic sterile saline was then added to the sedimented erythrocytes so that the hematocrit of the mixture was approximately 40-50 per cent. This mixture of chromated erythrocytes was then slowly injected intravenously into a recipient. Thirty minutes after injection, approximately 7 ml. of blood was withdrawn from the patient and placed in two EDTA bottles. Duplicate hematocrits were determined. Exactly 3 nil. of the well mixed whole blood from one of the hematocrit bottles was placed in a shell vial whereupon subsequent determination of radioactivity was measured in a well type scintillator connter. Another aliquot of whole blood was centrifuged and 1 ml. of plasma was removed by a Pasteur pipette and placed in a second shell vial. The plasma was then diluted to a total volume of 3 ml. with sterile saline for subsequent analysis of radioactivity. Samples similar to the above were obtained at 24 and 72 hours and the plasma and whole blood radioactivity was measured. In later samples only a 3 ml. volume of the well mixed whole blood was placed in a shell vial for counting. Specimens were obtained at five or six intervals during the ensuing month. All measurements of radioactivity on each of the samples obtained in a single experiment were determined at the end of the study in each patient. The emmission of each sample was measured and corrected for plasma and background radioactivity. The radioactivity for each blood sample was expressed as counts/minute per milliliter of packed red blood cells. The sample obtained 30 minutes after injection of the chromated erythrocytes was considered to be 100 per cent active. All subsequent samples were expressed in terms of the percentage decrease from this baseline measurement. The best fit on the disappearance curve was calculated by the method of least squares. The one-half survival time oE chromium 51 was then determined from that curve. No attempt was made to correct for elution. In calculating the correlation coefficient, the 100 per cent value was excluded since the greatest amount of elution of chromium occurs shortly after the cells are injected into the circulation of the recipient. RESULTS Six normal healthy individuals of various ages were given transfusions of compatible donor erythrocytes. The erythrocyte survival in these normal recjpients (table 1) reveals a range of half life survival time of the chromium 483 ERYTHROCYTE SURVIVAL IN RHEUMATOID ARTHRITIS Table 1.-Results of Cr5I Studies in Healthy Individuals Who Received Erythrocytes from Normal Donors __- - _ _ _______ .~ Case Number T' y Cr" 1 2 3 6 Mean % 76 ____ 44 42 42 43 46 47 4 5 __-_ Hematocrit ____._ (days) 23.0 32.2 26.3 22.2 29.4 27.2 I__._____ Cr51 Survival Time-26.7 days Correlation Coeffieient _________. -0.987 -0.955 -0.998 -0.991 -0.988 -0.979 ~_____- D .001 .001 .001 .001 .001 .010 -+ 6.9 days. 51 labeled erythrocytes of 22.2-32.2 days, with a mean half life of 26.7 days. The ncrmal half life value established for this study was then set at 26.7 days * 6.9 days ( 2 standard deviations) with normal values ranging from 19.833.6 days. Table 2 represents the results obtained in 16 patients with rheumatoid arthritis who received chromium 51 tagged erythrocytes from normal donors. The first three patients shown in table 2 were identical twins with sero-positive definite rheumatoid arthritis who received transfused erythrocytes from their normal unaffected identical twin. Identity of the twins was proien by establishing that each set had identical blood groups in 25 systems and by a successful skin transplantation from the normal to the affected twin. Figure 1 depicts the survival time of the tagged erythrocytes in each of the three sets of twins studied. I t will be observed that erythrocyte disappearance is linear and that the half life in each instance is normal. Patients 4-13 (table 2 ) were patients with sero-positive, definite rheumatoid arthritis who received chromium 51 tagged red blood cells from healthy donors. Eight of the ten patients had normal one-half survival times, whereas patients 12 and 13 had a decreased erythrocyte survival. Patients 14-16 (table 2 ) were individuals who had sero-negative, definite rheumatoid arthritis who recei\.ed chromated erythrocytes from healthy donors. Two of these three patients had a normal red cell survival. Patient #16, had a slight decrease of survival time. Table 3 depicts the results of erythrocyte survival wherein both the donors and recipiedts had rheumatoid arthritis. A total of 14 transfusions were given in which various combinations of donors and recipients were used. Patients 17, 18, and 19 who had sero-positive definite rheumatoid arthritis receiired erythrocytes from donors who also had sero-positive rheumatoid arthritis. All three of these experiments revealed erythrocyte survival times within normal limits. In cases 20-23 erythrocytes obtained from patients with seronegative rheumatoid arthritis were transfused into patients with sero-positive rheumatoid arthritis. Erythrocyte survival times in each instance were normal. Cases 2 4 3 0 were patients with sero-negative rheumatoid arthritis who received erythrocytes from patients with sero-positive rheumatoid arthritis. Six of these studies are within normal limits, but case 26 had a definitely shortened half life of chromium 51 labeled donor cells. Table 4 summarizes the laboratory data and drug therapy in the patients 484 MONGAN AND JACOX Table 2.--Clinical Data and Results of Crsz Studies in Patients with Rheumatoid h o_ Received Erythrocytes _ _ _ _ Arthritis _ ~ _W_ _ ____from Normal Donors Case Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Age (rrs.) 51 61 25 32 62 52 56 62 67 59 “7 72 69 61 59 31 ______ Duration of Disease Sex F F F F M F F F M M F F F F F F (JM.) 10 8 2 9 4 9 1.5 8 1.3 13 3 9 2 10 11 1 Hematocrit % 36 36 40 42 38 37 41 40 36 46 39 36 39 38 32 31 Latex Fixation T% CrS1 1 :2560 1:320 1: 320 1:5120 33.5 27.2 23.9 21.0 35.3 26.0 24.6 31.8 25.1 26.8 23.4 1:80 1:2560 1 :320 1:2560 1 :5120 1:320 1:1280 1:2560 1:160 Neg. Neg. Nep. Correlation Coefficient (days) 27.2 36.1 19.5 - -0.994 -0.990 -0.989 -0.989 -0.957 -0.983 -0.996 -0.996 -0.999 -0.989 -0.994 See Figure 2 See Figure 2 -0.998 -0.999 -0.995 P .001 .001 ,010 .001 .001 .001 .001 .001 .001 .001 .001 .001 .001 ,001 studied. No significant differences were noted between the patients with normal survival times arid those patients with shortened survival times in any of the variables studied. Selen of the recipient patients were anemic (i.e. hematocrit below 40 per cent in males, below 36 per cent in females) at the time they received donor cells. Only one of these patients ( # l 6 ) had a shortened erythrocyte survival tlme. When the seven anemic patients are compared as a group to the six normal controls, no significant difference in results is found. The mean halftime of chromium 51 disappearance for normal cells introduced into the circulation of normal subjects was 26.7 days, compared with the mean halftime for disappearance of donor erythrocytes in the anemic recipients of 27.3 days. By Student’s t test the probability of these mean values not being significantly different was well above 0.1 ( t = 0.21; n = 11; p >> 0.1). Analysis of Patients with a Shortened Erythrocyte Survival Time Figure 2 represents the erythrocyte curves determined in patients 12 and 13. In each instance the survival time appeared to be relatively normal until the donor erythrocytes had been in the recipient circulation for two weeks. By the sixteenth day, there was marked destruction of donor cells in both patients. This abrupt destruction of donor cells suggested that the recipient had become sensitized to the donor’s erythrocytes. The experiment was repeated wherein the same donor’s erythrocytes were transfused into the patient. It will be noted (fig. 2 ) that an accelerated and more rapid donor erythrocyte destruction resulted. A minor blood group incompatibility existed for case #12 with donor cells Ee being transfused into recipient ee. An analysis of the recipient’s serum for anti-E antibody revealed no serologically identifiable antibody after the first and the second transfusion made nine months later. The blood type in patient #13 was identical to that of the donor. A search for other m nor blood group antibodies was made before the second transfu- 485 ERYTHROCYTE SURVIVAL IN RHEUMATOID ARTHRITIS \ PATIENT # I PATIENT #2 M PATIENT#3 0-0 0-0 TIME IN DAYS Fig. 1.-Erythrocyte survival in three patients with rheumatoid arthritis who received donor red blood cells from their identical twin. sion eight months later. No detectable antibodies were found during or after the repeat study. Despite our failure to demonstrate an incompatability, it is felt that these data suggest that there was a sensitization of the recipient to the donor red blood cells in each case. The survival time in case #16 was 19.5 days which is less than 19.8 days, the lower limit of normal established in this study. Since 95 per cent of expected cases would fall within two standard deviations of the mean, it is conceivable that this value of 19.5 days does not represent decreased erythrocyte survival. A repeat transfusion employing erythrocytes from the same donor was made nine months later (fig. 3 ) . The half life on this occasion was 20.2 days, a figure almost identical to the first one. There had, however, been a significant change in the patient’s illness between the two studies. At the time of the first transfusion the patient had an active arthritis reflected by an increased erythrocyte sedimentation rate and a reduction of hematocrit to 31 per cent. When the second transfusion was made the patient was at a stage of a marked clinical remission of arthritis. The hematocrit had idcreased to 42 per cent. Since the survival times in both studies were almost identical, it seems unlikely that erythrocyte hemolysis played any role in the genesis of the patient’s anemia. Patient #26 had a definitely shortened blood cell survival with a chromium 51 half time disappearance of 16 days. This patient had severe rheumatoid arthritis (class IV, stage IV) of 25 years duration. The disease was still 486 MONGAN AND JACDX Table 3.-CEinical Data and Results of Cr5' Studies in Patients with Rheumatoid Arthritis Who Received Erythrocytes from Other Patients with Rheumatoid Arthritis ___--__ Duration Case Number Age (yrs.) 17 18 19 20 21 22 23 24 25 26 21 28 29 38 59 71 68 52 61 56 55 57 64 29 61 53 62 30 of Disease Hematocrit Sex (yrs.) % F M F F F M F F M 3 2 12 3 5 4 6 32 F F F F 28 12 F 4 30 39 41 36 41 38 39 39 40 36 37 41 39 43 6, 1:: 12 T1h C P Latex Fixation (days) 22.8 24.9 24.6 23.5 36.4 26.4 27.8 26.4 32.3 16.0 22.1 32.4 21.7 30.8 1:320 1:320 1:80 1:1280 1:160 1:80 1:80 Neg. Neg. Neg. Nep. Neg. Neg. Neg. Correlation Coefficient P -0.998 -0.980 -0.982 -0.980 -0.987 -0.993 -0.999 -0.988 -0.986 -0.344 -0.933 -0.990 -0.991 -0.989 .001 .001 .001 .001 .01 .001 .a01 .001 .001 - .01 .001 .001 .001 Table I.-Correkztion of Severity of the Patient's Rheumatoid Arthritis with Laboratory Studies and Drug Therapy _______--Total number of patients Number of patients with an elevated sedimentation rate Number of anemic patients Number of patients with a latex titer of 1:80 or above Number of patients with a positive antinuclear factor Number of patients with a positive LE prep Number of patients with a decreased red blood cell survival as measured by Cr.51 technic Number of patients taking salicylates phenylbutozone gold chloroquine prednisone Class I1 ___-- Class I11 Patients with Decreased RBC Class IV Total Grow 4 30 4 ____-___ Survival 11 15 8 2 15 4 5 0 27 7 3 1 8 10 2 20 2 1 5 1 7 1 1 4 0 5 0 1 2 1 4 4 14 2 23 5 2 3 3 5 10 16 active as reflected by a Westergren sedmientation rate of 100 mm. per hour. The serum contained no rheumatoid factor. The patient had previously received blood transfusions on sel-era1 occasions during surgical procedures. It is possible that this patient, as patients 12 and 13, had developed a minor blood group incompatibility. DISCUSSION The results of this investigation' clearly demonstrate that erythrocyte survival in rheumatoid patients is norma1 and is independent of the source of the erythrocytes used in the transfusion. Thus normal donor erythrocytes and ERYTHROCYTE SURVIVAL IN RHEUMATOID ARTHRITIS 487 PATIENT # I 2 PATIENT # I 3 REPEAT STUDY#I2 REPEA1 STUDY#13 I 0 0 . lo' ' '.'0 I 0 ' 20 ' 30 ' 40 ' TIME I N Fig. 2.-Erythrocyte normal limits. DAYS survival in two patients in whom the results fell outside erythrocytes obtained from sero-negative and sero-positive rheumatoid donors, when tran'sfused into rheumatoid patients have a normal life span. The most critical analysis of this question was available when three sets of identical blood types were studied. It was found that each normal twin's erythrocyte had a normal survival when transfused into the corresponding twin with rheumatoid arthritis. We are thus unable to confirm Richmond's* observation that rheumatoid patien'ts have some unusual capacity to destroy normal donor cells more rapidly than when isologous labeled erythrocytes are tranfused. In the three instances in which definite decreased erythrocyte survival was observed, the pattern of erythrocyte destruction was that of a minor blood group incompatibility even though no serological identification of this phenomenod was demonstrable. Mollisonll has shown in cross transfusion studies that it is frequently impossible to identify antibodies when there is diminished survival due to incompatibility. The question must be raised as to why some observors have reported decreased erythrocyte survival in rheumatoid patients while others have not supported this observation. In order to define the limits of erythrocyte survival it is important to define any variable which is due to either the labeling technics or to possible external factors which produce blood loss apart from intravascular destruction. The normal range for the half-time of chromium 51 disappearance in our 488 MONGAN AND JACOX 0-0 INITIAL STUDY O--. lo L ' I0 ' STUDY REPEATED WITH THE SAME DONOR 2 0 ' 3 0 ' 4 0 ' 50 TIME IN DAYS Fig. 3.-Erythrocyte survival in patient Number 16. study was 19.8-33.6 days and is greater than' that reported by other observors. The half life of chromium 51 labeled erythrocytes for normal cells transfused into normal subjects was 26.7 days compared to a mean half life of 27.2 days of normal erythrocytes transfused into patients with seropositive rheumatoid arthritis. Student t test analysis reveals the probability of these findings not being significantly different to be greatly in excess of 0.1 ( t = 0.21; n = 15, p = > > .01). Any variablity due to technic was controlled as rigidly as possible by the following methods. The control subjects were studied by doing these individuals randomly during the time of the patient studies. The same method of erythrocyte labeling and transfusion was employed. In all experiments great care was taken to idsure adequate mixing of labeled chromate ions with the erythrocyte suspension. Only chromium of high chromium 51 specific activity was used. A mathematical calculation of the slope of labeled erythrocyte disappearance was made to insure the most precise estimation of the point of half time survival. It is felt that these technics have minimized most of the variables that exist in such a procedure. SUMMARY Thirty patients with active rheumatoid arthritis were tradsfused with Chromium 51 labeled erythrocytes from either normal donor or rheumatoid donor which were either sero-positive or sero-negative for the rheumatoid factor. Some of the rheumatoid recipients were sero-negative and received ERYTHROCYTE SURVIVAL IN RHEUMATOID ARTHRITIS 489 donor erythrocytes from normal or rheumatoid patients who were sero-positive for the rheumatoid factor. Three sets of identical twins which were discordant for rheumatoid arthritis had cross transfusion of labelled cells from the healthy donor to the affected recipient. In r ~ oinstance was there any significant shortening of erythrocyte survival which could not be attributed to minor blood group incompatibility. ACKNOWLEDGMENT We are indebted to Dr. Anthony Bowdler and Dr. Robert Weed for advice in the use and calculations of the chromium labeling technic and to Miss Norma T r a b l d for performing the blood typing and cross-matching. REFERENCES 1. Lewis, S. M., and Porter, I. H.: Erythrocyte survival in rheumatoid arthritis. Ann. Rheum. Dis. 19:54-58, 1960. 2. Biechl, A. Stapleton, J. E., Woodbury, J. F., and Read, H. C.: Anemia in rheumatoid arthritis. I. Red cell survival studies. Canad. Med. Ass. J. 86:401-405, 1962. 3. Roberts, F. D., Hagedorn, A. B., Slocumb, C. H., and Owen, C. A.: Evaluation of the Anemia of Rheumatoid Arthritis. Blood 21:470-477, 1963. 4. Ebaugh, F. G., Jr., Peterson, R. D., Rodman, G. P., and Bunim, J. J.: Symposium on rheumatic diseases: Anemia of rheumatoid arthritis. Med. Clin. North America 39:489-498, 7. 8. 9. 1955. 5. Alexander, W. R. M., Richmond, J., Roy, L. M. H., and Duthie, J. J. R.: Nature of anaemia in rheumatoid arthritis. II. Survival of transfused erythrocytes in patients with rheumatoid arthritis. Ann. Rheum. Dis. 15: 12-20, 1956. 6. Freireich, E. J,, Ross, J. F., Bayles, T. B., Emerson, C. P., and Finch, S . C.: Radioactive iron metabolism and ery- 10. 11. throcyte survival studies of the mechanism of the anemia associated with rheumatoid arthritis. J. Clin. Invest. 36: 1043-1058, 1957. Weinstein, I. M.: A correlative study of the erythrokinetics and disturbances in iron metabolism associated with the anemia of rheumatoid arthritis. Blood 14:950-969, 1959. Richmond, J., Alexander, W. R. M., Potter, J. L., and Duthie, J. J. R.: The nature of anaemia in rheumatoid arthritis V. Red cell survival measured by radioactive chromium. Ann. Rheum. Dis. 20:133317, 1961. Steinbrocker, O., Traeger, C. H., and Batterman, R. D.: Therapeutic criteria in rheumatoid arthritis. J.A.M.A. 140: 659-662, 1949. Singer, J. M., and Plotz, C. M.: The latex fixation test. I. Application to the serological diagnosis of rheumatoid arthritis. Am. J. Med. 21:888-892, 1956. Mollison, P. L.: Measurement of survival and destruction of red cells in haemolytic syndromes. Brit. Med. Bull. 15:59-66, 1959. Edward S. Mongan, M.D., Senior Instructor in Medicine, Rochester School of Medicine and Dentistry, Rochester, New Ymk. Ralph F. lacox, M.D., Professor of Medicine, Rochester School of Medicine, and Rentistry, Rochester, N e w York.