Abstract
ABSTRACT. Inherited prothrombotic risk factors predispose patients to thromboembolic events. In kidney transplant recipients, thrombophilia may manifest itself with venous thrombosis, microvascular occlusion, or acute rejection with major consequences for allograft survival. This is a prospective study on 165 renal allograft recipients to evaluate the contribution of genetic thrombophilic risk factors to transplant outcome. Besides antithrombin, protein C, and protein S deficiencies, none of which was found in our patient group, factor V G1691A (FV G1691A), prothrombin G20210A (PT G20210A) mutations, and methylenetetrahydrofolate reductase (MTHFR) gene C677T polymorphisms were studied. The primary endpoint of the study was occurrence of an acute rejection within the first 90 d and transplant loss within 1 yr. Heterozygous FV G1691A and PT G20210A mutations and the MTHFR T677T variant were significantly associated with acute rejections with rejection rates of 68%, 67%, and 71%, respectively, as compared with 35% in patients not carrying these genotypes. Many rejections that were histologically proven were acute vascular ones. Transplant loss was significantly associated exclusively with the PT G20210A group (50% 1-yr graft survival; odds ratio, 10.0; 95% confidence interval, 1.8 to 56.1). PT G20210A patients exerted the highest prothrombotic activity pretransplant, as determined by prothrombin 1.2 fragments (PT F1.2), which may be the background for minor outcome. In conclusion, common prothrombotic mutations are significantly associated with acute rejections, especially vascular rejections, and for PT G20210A also with early transplant failure. Screening for hypercoagulable states pretransplant is recommended to intensify anticoagulatory treatment posttransplant. E-mail: heidenr@uni-muenster.de
Although rates of renal allograft survival have been improved over the last decade, 10 to 20% of transplants are still lost within the first year. The major reasons for irreversible graft damage are acute rejections, with acute vascular rejections playing the most crucial role (1). As a second cause, arterial or venous thromboses are frequently associated with rapid transplant loss (2). In pediatric renal transplantation, vascular mostly venous occlusion accounts for around 15% of early graft failures and still represents a severe threat (3). Microthrombi formation of capillaries or macrovascular thrombosis shares histologic hallmarks with acute vascular rejection, and differentiation is not easy in any instance (4,5). Fibrin aggregation or occlusive glomerulitis, alterations that clearly point to hypercoagulation, also become evident during transplant rejection; therefore, a link between acquired or inherited thrombophilia on the one hand and acute or chronic rejection or vascular graft loss on the other has been hypothesized (6,7,8).
In previous, exclusively retrospective surveys this question has been addressed, and the FV G1691A mutation has been linked first and foremost with unfavorable transplant outcome (6–10). As major reasons the occurrence of acute rejections, probably mostly vascular rejections, has been described (6,8,10). Furthermore, rates of early graft loss due to venous thromboembolism and graft perfusion defects were significantly elevated in patient groups tainted with thrombophilia (11). Other groups have recently observed that the prothrombin (PT) G20210A mutation similarly predisposed to early transplant failure (12). However, the significance of these studies is limited because of their retrospective design, which may be confounded by a selection bias due to different treatment modalities over the observation time or follow-up of included patients. Furthermore, acquired prothrombotic risk factors that may be altered by ongoing therapy have also been analyzed so that causes and effects of hypercoagulation could not be clearly differentiated (7).
The aim of the present study was therefore to prospectively evaluate the contribution of prothrombotic factors to the risk of acute transplant rejection and graft loss. A total of 165 ongoing patients receiving a renal allograft were investigated for the presence of the FV G1691A and PT G20210A mutations, antithrombin, protein C, and protein S deficiencies, as well as the MTHFR C677T polymorphisms. The end points of the study were acute rejection within the first 90 d and graft loss after 1 yr.
Materials and Methods
Patients
A total of 165 patients who underwent renal transplantation at the Münster Transplant Center between November 1, 1997, and December 31, 2000, were recruited for this prospective study after giving informed consent. Blood was drawn immediately before transplantation by peripheral vein puncture for performing plasmatic coagulation assays and genotyping. Demographic data for recipients and donors are summarized in Table 1.
Table 1. Demographic and clinical data for the 165 studied renal transplant patients and the subgroups with genetic thrombophilia
All patients were on triple immunosuppression comprising cyclosporin A, mycophenolate mofetil, and prednisolone for the first year. Acute rejection was assessed by typical clinical and laboratory parameters, including conventional and duplex sonography and magnetic resonance imaging. In 39 patients, acute rejection was proven by biopsy; in the remaining cases with clear signs of acute rejection without histologic verification, patients refused to give their consent or the procedure was omitted because of a highly swollen graft and risk of bleeding. All patients with acute rejection were initially treated with steroid pulses of 500 mg/d prednisolone for 3 d. Patients failing to adequately respond to initial treatment were switched from cyclosporin A to tacrolimus; if no success was recorded, they were treated with antilymphocytic antibodies. Arterial occlusion or venous thrombosis was proven by angiography.
Assays for Genotyping
For genetic analysis, we obtained venous blood in EDTA-treated sample tubes (Sarstedt, Nümbrecht, Germany), from which cells were separated by centrifugation at 3000 × g for 15 min. The buffy coat layer was then removed and stored at −70°C pending DNA extraction by standard techniques. FV G1691A and PT G20210A polymorphisms were determined in patients by PCR amplification and digestion with MnlI and HindIII, respectively (13,14). The C677T polymorphisms in the MTHFR gene were determined by PCR amplification and digestion with HinfI (15).
Plasma-Based Assays
Blood samples were collected before renal transplantation by peripheral venipuncture into plastic tubes containing 1/10 by volume of 3.8% trisodium citrate (Sarstedt, Nümbrecht, Germany) and placed immediately on melting ice. Platelet-poor plasma was prepared by centrifugation at 3000 × g for 20 min at 4°C, aliquoted in polystyrene tubes, stored at −70°C, and thawed immediately before the assay procedure (16). Amidolytic protein C and antithrombin activities were measured on an ACL 300 analyzer (Instrumentation Laboratory, Munich, Germany) using chromogenic substrates (Chromogenix, Mølndal, Sweden). Free protein S antigen, total protein S, and protein C antigen were measured using commercially available ELISA assay kits (Diagnostica Stago, Asnières-sur-Seine, France).
Prothrombin fragments 1.2 (PT F 1.2) and d-dimer formation were measured with ELISA technology (Enzygnost F 1+2 micro and Enzygnost D-dimer micro; Dade Behring, Marburg, Germany).
Statistical Analyses
All statistical analyses were performed using the StatView 5 software package (SAS Institute Inc., Cary, NC). To evaluate an independent contribution to the risk of acute rejection or graft loss (involvement: yes or no), the presence of prothrombotic risk factors was analyzed by multivariate logistic procedure (odds ratios [OR] and 95% confidence intervals [CI]). In addition, the estimated relative risk (RR) and 95% CI were calculated to compare the rate of acute rejection or graft failure in carriers of prothrombotic risk factors with that of patients without an identified genetic risk. Because of their non-Gaussian frequency distribution, continuous data are presented as medians and ranges and evaluated by nonparametric statistics using the Wilcoxon-Mann-Whitney U test. Fisher’s exact test was performed to compare frequency distributions of fatal outcome. The Kaplan-Meier method was used to estimate rejection rates and graft losses, with comparisons based on the two-sided log-rank test. P < 0.05 was considered significant.
Ethics
The present prospective follow-up study was performed in accordance with the ethical standards laid down in the updated version of the 1964 Declaration of Helsinki and was approved by the medical ethics committee of the University of Münster, Germany.
Results
Of 165 consecutive patients recruited for the study and tested for genetic thrombophilia, 19 (11.5%) were heterozygous for the FV G1691A and 6 (3.6%) heterozygous for the PT G20210A mutation. Homozygous defects were not identified. The MTHFR T677T variant was detected in 17 (10.3%) transplant recipients (C677T in 75 [45.5%] and C677C in 73 [44.2%] patients). Patients with protein C, protein S, or antithrombin deficiency were not found. Demographic and clinical data of the three subgroups of patients with inherited prothrombotic risk as compared with the whole study population are given in Table 1. It is notable that classical risk factors of donors and recipients with respect to acute rejection or graft loss, e.g., donor and patient age, HLA mismatches, cold ischemia time, or panel reactive antibodies were equally distributed and not associated with a prothrombotic risk group. Recipient BP was similar in all groups, and number of smokers did not display intergroup differences.
Among the 165 transplant patients, an acute rejection within the first 90 d was diagnosed in 73 cases (44%). The rejection rate in the patient group without one of the three detected factors of genetic thrombophilia was 35%. Eight biopsies out of 18 performed in the group with genetic prothrombotic phenotypes revealed signs of vascular rejection (44%), whereas in the group without genetic thrombophilia only 3 (14%) of 21 biopsies showed alterations typical for vascular rejection (P < 0.05). Histologic classification has been worked out according to the Banff 97 classification (17). Another hallmark of biopsies performed in patients with prothrombotic genotypes was occurrence of microvascular thrombosis within glomerular and peritubular capillaries, which was found in conjunction with but also without type II or type III acute rejection. Rejection rates were significantly elevated in two subgroups with inherited thrombophilia (Table 2). In the FV G1691A group (n = 19) an acute rejection was determined in 13 patients (68%). Logistic regression analysis proved that the FV G1691A mutation was an independent risk factor for acute rejection (OR, 5.7; 95% CI, 1.6 to 20.2; P = 0.01). RR was 3.3 (95% CI, 1.3 to 8.3). Biopsies performed in 9 of the 13 cases with rejection revealed acute vascular rejection in three patients. As shown in Table 2 according to the Banff 97 working classification (17) one patient had type IIA and two others type IIB acute rejection with typical signs of endothelialitis. Five biopsies showed microvascular thrombosis of glomerular and peritubular capillaries. Furthermore, three patients in the FV G1691A group had transplant loss within the first year; this was caused by arterial and venous thromboembolism in one patient and by venous graft thrombosis in two other patients. OR for graft loss was 1.6 (95% CI, 0.4 to 6.8); RR was 1.4 (95% CI, 0.5 to 4.5), which was not statistically significant. The Kaplan-Meier plot showing onset of acute rejection episodes over the first 90 d in comparison with the transplant group with factor V wild-type is depicted in Figure 1A.
Table 2. Incidence and relative risk of the primary endpoints
Figure 1. (A) Kaplan-Meier estimates of the time of an acute rejection over the first 90 d posttransplant for carriers of the heterogeneous FV G1691A mutation in comparison with non-carriers. (B) Kaplan-Meier estimates of the time of an acute rejection over the first 90 d posttransplant for carriers of the heterogeneous PT G20210A mutation in comparison with non-carriers. (C) Kaplan-Meier estimates of the time of an acute rejection over the first 90 d posttransplant for carriers of the MTHFR T677T variant in comparison with the remaining groups.
As further shown in Table 2, 4 of the 6 patients with the PT G20210A mutant had an acute rejection (67%), with histologic verification in two cases and the diagnosis of acute vascular rejection in one. This patient had type III acute rejection with fibrinoid vascular necrosis and transmural arteritis, which finally led to transplant loss. The other biopsy performed showed microvascular thrombosis (Table 2). For acute rejection, OR was 12.2 (95% CI, 1.3 to 118.7; P = 0.03), RR was 7.4 (95% CI, 0.9 to 62.1), which was not statistically significant because of the small patient number. Figure 1B depicts the occurrence of acute rejections in the PT G20210A group during the first 90 d as compared with the wild-type group, which was not significantly elevated in the Kaplan-Meier analysis due to the small number of cases. In three of six patients in the PT G20210A group, the renal transplant was lost within the first year as a consequence of arterial and venous thromboembolism in one, of acute vascular rejection refractory to intensive immunosuppression in another, and of acute tubulointerstitial rejection with graft rupture and bleeding in the third case. Logistic regression analysis confirmed the PT G20210A mutation as an independent risk factor (OR, 10.0 [95% CI, 1.8 to 56.1]; P = 0.009), RR of PT G20210A for graft loss was 7.7 (95% CI, 1.7 to 35.4). Figure 2 depicts the Kaplan-Meier plot for transplant survival over the first year.
Figure 2. Kaplan-Meier estimates of the time of graft failure over the first year posttransplant for carriers of the heterogeneous PT G20210A mutation in comparison with non-carriers.
As for the MTHFR T677T variant, the outcome of allografting was similar to FV G1691A (Table 2). Twelve of 17 patients with this mutation had an acute rejection episode within the first 3 mo. Multivariate logistic regression analysis proved MTHFR T677T to be an independent risk factor for acute rejection (OR, 4.3 [95% CI, 1.4 to 12.9]; P = 0.04). RR was 4.7 (95% CI, 1.5 to 13.9). Histologic confirmation of acute rejection was undertaken in seven patients with the MTHFR T677T variant, with the diagnosis of acute vascular rejection in four cases. Three had type IIA and one type III acute rejection. Microvascular thrombosis was detected in three biopsies. Two patients lost their grafts within the first year (one case of type III acute rejection, one of type IB acute rejection), which was not statistically significant (OR, 1.2 [95% CI, 0.2 to 5.7; P = 0.6]; RR 1.0 [95% CI, 0.2 to 4.1]). The Kaplan-Meier plots showing the time course of acute rejections in the MTHFR T677T group in comparison with the remaining groups within the first 90 d are given in Figure 1C.
We compared levels of PT F1.2 and d-dimers as established markers of an activated coagulation system in the FV G1691A and PT G20210A groups with the wild-type study population (FV G1691G and PT G20210G). Significantly increased PT F1.2 values were detected only for carriers of the PT G20210A mutation (Table 3).
Table 3. Prothrombin fragment 1.2 (PT1.2) and d-dimer values of patients with FV G1691A or PT G20210A mutations in comparison to total study population without these variantsa
Discussion
Although a link between a hypercoagulable state and unfavorable renal transplant outcome has been hypothesized and described in retrospective studies or case reports (6–10), a prospective survey has so far been lacking. The prospective design of this present study, through which a patient selection bias could be largely excluded, ensured that all recruited patients received the same immunosuppressive regimen, were diagnosed for rejection or vascular complications by the same procedures, and were treated for rejection by a fixed medical protocol. It is emphasized that prothrombotic parameter determination and clotting assays were not performed after transplantation or treatment of complications that might have influenced the assays. In our study, we determined six alterations that are associated with an elevated prothrombotic risk. Patients with deficiencies of antithrombin, protein C, and protein S were not identified. Prevalences of two established inherited thrombophilic defects, namely FV G1691A, PT G20210A mutation, as well as the MTHFR T677T variant, were in the range reported in previous studies, so that an association between hypercoagulation and occurrence of renal diseases leading to end-stage renal failure could be largely ruled out. Determination of the MTHFR C677T polymorphisms was not accompanied by measurements of fasting homocysteine levels, which may be modified by exogenous factors (18), because the timing of blood samples between food intake and presentation at the hospital for transplantation varied strongly.
Our findings give evidence that the FV G1691A, the PT G20210A, and the MTHFR T677T variants were associated with an increased prevalence of acute rejection. Many though not all rejection episodes were confirmed by biopsy, which indicated microvascular thrombosis with or without acute vascular rejection in a relevant percentage. Performance of a biopsy was not practicable in every patient presenting with acute rejection because of an increased risk of bleeding or a lack of consent from the patients. This association between thrombophilia due to FV G1691A and acute vascular rejection was hypothesized in a previous study by our center (6) and was confirmed by Hocher et al. (8) and Ekberg et al. (10), who found increased early renal transplant loss among carriers of FV G1691A with a significant accumulation of vascular rejection as evidenced by endothelialitis or fibrinoid necrosis. As the underlying mechanisms, it is readily conceivable that hypercoagulation with elevated generation of thrombin can stimulate lymphocyte activation and induce humorally or cellularly induced rejection. Thrombin is a well-established mitogen that targets endothelial and vascular smooth muscle cells as well as T lymphocytes and platelets (19,20). But an initiation of graft damage or loss from the opposite direction is also conceivable; incipient clinically subthreshold rejections lead to endothelial damage with a decrease in anticoagulatory factors such as thrombomodulin. Genetic thrombophilia further induces secondary injury with uncontrolled coagulation and spreading of microthrombi (21).
Other explanations from previous reports point to the increased risk of renal transplant loss due to perfusion defects in the initial ischemia-reperfusion phase (11).
As for MTHFR polymorphism and hyperhomocysteinemia, previous data on the implications for graft rejection and survival are more sparse. One study after renal transplantation found no association of MTHFR C677T variants with the diagnosis of chronic rejection or graft survival; this is in accordance with our results (22). After orthotopic heart transplantation, it was recently reported that MTHFR C677T polymorphisms did not account for rapidly progressive forms of allograft vasculopathy but that patients with this complication had significantly higher homocysteine fasting levels with putative harmful effects (23).
The most relevant finding of our study is the increased risk of early renal transplant loss in association with the PT G20210A mutant. The PT G20210A mutation was also significantly associated with the occurrence of acute rejection in the logistic regression model but, due to small number of cases, failed to be associated with a significantly increased relative risk. Three of six patients with this variant lost their grafts within the first year, two due to rejections refractory to therapy (one vascular, one tubulointerstitial) and one due to severe arterial and venous thromboembolism. High rejection rates and graft failure due to rejection suggest that this prothrombotic mutation may be linked to immune mechanisms. The crucial question raised by our study is why carriers of the PT G20210A mutation are more prone to early graft loss than patients with the FV G1691A mutant. From studies comparing FV G1691A with PT G20210A for predisposition to deep vein thrombosis or cerebral ischemia, the former mutation has been regarded as the stronger prothrombotic risk factor (24,25). However, recent investigations have revealed that FV G1691A predisposes much more to deep vein thrombosis than to pulmonary embolism, a complication to which the PT G20210A mutation is probably of greater relevance (26,27). As the most widely accepted hypothesis for this discrepancy, a strong adherence of the thrombus to the vascular wall in FV G1691A has been suggested (28). On the one hand, we therefore speculate that microthrombi release into capillaries with putative harmful consequences may occur more frequently in subjects carrying the PT G20210A variant than in those with the FV G1691A mutation. On the other hand, another explanation might be that the FV G1691A genotype predisposes more than the PT G20210A mutation to acute rejection, a complication that is more susceptible to early diagnosis and successful therapy than hypercoagulation. Elevated plasma levels of PT 1.2 point to an increased procoagulable state in the PT G20210A group. PT 1.2 has been described as a valuable laboratory marker to predict transplant outcome, e.g., after heart transplantation (29). It is affected by treatment modalities; therefore, it may serve as a useful marker for longitudinal measurements.
In conclusion, the present study underlines the significance of genetic thrombophilia for acute renal allograft rejection and loss in a prospective survey and thus confirms observations from retrospective studies. A pretransplantation analysis of prothrombotic defects is recommended to increase posttransplant awareness of the problem and to increase anticoagulation. Further studies are needed to examine exact treatment strategies.
Acknowledgments
We thank Dr. Christian August, Dept. of Pathology at the University of Münster, for histopathologic evaluation and classification and Susan Griesbach for help in editing this manuscript. In addition, we thank Andreas Zdebel, Jessica Heimann, Irena Heid, Georg Gundoroff, Larissa Martens, and Swetlana Gudi for collecting blood samples from the patients investigated.
- © 2003 American Society of Nephrology
References
