| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;108:179-182
© 1994 Mosby, Inc.
LETTERS TO THE EDITOR |
Department of Cardiovascular Surgery
Matsuyama Red Cross Hospital
Matsuyama, Ehime 790, Japan.
To the Editor:
Postoperative erythroderma evidenced by high fever, erythroderma, liver dysfunction, diarrhea, and pancytopenia has been understood as a graft-versus-host disease (GVHD) caused by blood transfusion.
1-3 This devastating disease typically occurs when the donor's human leukocyte antigen (HLA) haplotypes are homozygous for one of the patient's HLA haplotypes, because graft-versus-host reaction against the patient's major (and minor) histocompatibility antigens is much stronger than host-versus-graft reaction against the donor's minor histocompatibility antigens in this case. 4,5 In practice, the fresher and more closely related the transfused blood is, the greater is the risk of GVHD. 6-8 Moreover, not only immunocompromised hosts but also postcardiotomy patients seem to be vulnerable to this disease. 9 Its prevalence after heart operations in Japan has been reported to be 0.15%. The mortality rate is as high as 90%. 9
We present here three cases of GVHD occurring after heart operations and the transfusion of fresh or preserved blood. In one of these cases, extensive analysis of the cellular kinetics of posttransfusion GVHD after a heart operation (including HLA typing of the patient, his family and the blood donors) was performed.
A 67-year-old Japanese man (patient 1) with rheumatic mitral stenosis and regurgitation was referred to Matsuyama Red Cross Hospital. Blood test results were all within the normal limits. Mitral valve replacement was performed in September 1991. For priming of the cardiopulmonary bypass (CPB) circuit, 260 gm 5-day preserved concentrated red blood cells (CRBCs) from donor 1 and 130 gm 15-day preserved CRBCs from donor 2 were used. Soon after the operation, an additional 260 gm 5-day preserved CRBCs from donor 3 was given. All three blood donors were unrelated to the patient. The patient's postoperative course was uneventful until spiking fever started on postoperative day (POD) 10. On POD 14, a mild elevation of liver enzyme activities (such as aspartate aminotransferase, alanine aminotransferase, and lactic dehydrogenase) and a rash on the face and neck were observed. The rash gradually spread over the entire body surface, culminating in an appearance similar to that of a second-degree burn by POD 19. On POD 17, watery (later bloody) diarrhea associated with redness, swelling, and erosion in the lips and anus began. Delirium was observed from POD 16 through POD 19. The patient entered into a state of shock on POD 19 and died on POD 20. Permission for autopsy was refused, but the histologic appearance of the skin at death was fully compatible with GVHD. 10 The bone marrow was not examined.
The hemoglobin level and red blood cell count were relatively constant throughout the postoperative course (Fig. 1). The decreased platelet count caused by CPB had returned to its normal level by POD 10 but decreased again after POD 14. The white blood cell count, which increased after the operation, appeared to have returned to normal by POD 10; thereafter, however, it continuously decreased, to 3100 cells/mm 3 on POD 18 and to 500 cells/mm 3 on PODs 19 and 20. The differential count of the white blood cells was dramatically and suddenly changed from granulocyte dominance (>80%) to lymphocyte dominance (>96%) on POD 18. The absolute number of lymphocytes decreased from 1403 cells/mm 3 before operation to 411 cells/mm 3 immediately after operation; it then showed a plateau at about 1000 cells/mm 3 from PODs 7 through 14, followed by a further rapid drop to 93 cells/mm 3 on POD 17. The lymphocyte count did, however, increase again to about 500 cells/mm 3 after POD 18.
|
Table I), his HLA haplotypes just before his death were found to be a/a—exactly the same as the HLA haplotype of donor 3 (a 23-year-old man). The frequency of haplotype a in Ehime Prefecture, Japan, has been reported to be 0.0593. 11 If the patient had been a mixed chimera, the patient's original lymphocytes, bearing haplotype b-encoded antigens, would have been detected. From a quantitative analysis of the HLA-typing scores (Table I), however, Cw9, DR4. 1, and DQw4 were scored into 1 and B35 was scored into 2, suggesting that the existence of host-type lymphocytes was negligible (Table I). From a flow-cytometric analysis of the patient's blood, T cells (CD3+) accounted for 84% of the mononuclear cells, whereas B cells (CD19 +) represented 0%. Among the donor-originated T cells, 10% were CD4+ and 79% were CD8+ (Fig. 1). This patient was the first to have posttransfusion GVHD among 400 heart operations at Matsuyama Red Cross Hospital from 1983 through 1991.
|
As reported previously, 4,5,12 a patient with ongoing GVHD could be either partially or completely chimeric. From the quantitative analysis of the HLA-typing scores, patient 1 appeared to be completely or highly chimeric on POD 20. The lymphocyte dip associated with the differential change from granulocyte dominance to lymphocyte dominance occurred from POD 17 to 20 in all three patients. This dip in the absolute lymphocyte count seems to imply a turning point in the takeover by donor-originated lymphocytes. Ito and colleagues 3 also reported that a change of HLA phenotype occurred on POD 17 in a patient with esophageal cancer who underwent operation and transfusion with fresh blood. All three patients in our study showed rapid deterioration in condition after the lymphocyte dip.
Methylprednisolone was used for all of our patients, but none was able to survive. Thaler and colleagues 8 reported that antithymocyte globulin was effective in prolonging the life of a patient with posttransfusion GVHD. OKT3 may not be able to fully reverse GVHD, 13 but as an antibody to T cells it seems so far to be one of the safest and most effective drugs in treating posttransfusion GVHD; however, we could not use OKT3 because of its test stage in Japan. As a result of our extensive experimental studies on cyclophosphamide-induced tolerance, 14,16 the use of cyclophosphamide in patient 1 was discussed but dismissed because of the devastated state of the patient. Injections of cyclophosphamide at a dose range of 20 to 100 mg/kg, however, seem to be effective in treating animals with posttransfusion GVHD (Fujiwara M, personal communication). In our patients, clinical status worsened day by day after the skin rash started and shift by shift after the lymphocyte dip was observed. Aggressive treatment should therefore be started as soon as possible once the lymphocyte dip has been confirmed. Granulocyte colony-stimulating factor 17 was not effective in patient 1. Isolation in a laminar-flow room is recommended for such immunocompromised patients. 18 Because of difficulty in treating or even diagnosing posttransfusion GVHD, prevention of the disease by autologous blood transfusion, irradiation of allogeneic blood 7 (not only for related and fresh blood but also for unrelated and preserved blood), leukocyte filtration, 19 or ultraviolet irradiation 20 is of the utmost importance.
In case 1, a donor's HLA haplotypes were homologous for one of the patient's HLA haplotypes. The effector cells related to the pathogenesis of this disease in patient 1 were therefore host T cells reactive against the donor's minor histocompatibility antigens (effector cells of host-versus-graft reaction) and donor T cells reactive against the host's class I (B35, Cw9), class II (DR4.1, DQw4), and minor histocompatibility antigens (effector cells of graft-versus-host reaction). Determination of the minor histocompatibility antigens of the donor and host partially responsible for the host-versus-graft reaction or graft-versus-host reaction is not possible. Despite the 5-day preservation of donor blood in case 1, the donor T cells consequently overcame the host T cells. As previously reported by Hisatomi and colleagues 21 CPB is harmful to T-cell function. The fact that the responsible CRBCs from donor 3 were transfused after (but not before or during) CPB may be of significance. The type (membrane or bubble) of oxygenator, however, seemed irrelevant; both types of oxygenator were used for our three patients.
The rapid reduction in the number of granulocytes (primarily positive for class I antigen), T cells (class I), B cells (classes I and II), monocytes and macrophages (classes I and II), and platelets (class I) in patient 1 after POD 14 was considered to be a type of rejection phenomenon, and is compatible with the fact that red blood cells that were HLA negative were maintained at a relatively constant level. Among the donor-type T cells in the patient, 10% were CD4+ T cells reactive primarily against host class II antigens (DR4.1, DQw4) 22 and 79% were CD8+ T cells reactive primarily against host class I antigens (Bw35, Cw9). 22 Some of these CD4+ or CD8+ T cells may have been reactive against minor histocompatibility antigens of the host. The so-called Th/Ts ratio, which represents the host's immunocompetence, 23 was 0.13; however, this index does not make sense in this particular situation because all the T cells were of donor origin.
References
This article has been cited by other articles:
|
|
 |
|
  F D Rubens and T Mesana The inflammatory response to cardiopulmonary bypass: a therapeutic overview Perfusion, January 1, 2004; 19(1_suppl): S5 - S12. [Abstract] [PDF]
|
|
|
|
 |
|
 H. Mayumi, Q.-W. Zhang, A. Nakashima, M. Masuda, H. Kohno, Y. Kawachi, and H. Yasui Synergistic Immunosuppression Caused by High-Dose Methylprednisolone and Cardiopulmonary Bypass Ann. Thorac. Surg., January 1, 1997; 63(1): 129 - 137. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
