The opinions are solely those of the authors and do not represent an endorsement by the Department of Defense.
National statistics for patient characteristics and survival of renal transplant recipients positive for hepatitis C virus (HCV+) at the time of renal transplant are presented.
A historical cohort analysis of 33 479 renal transplant recipients in the United States Renal Data System from 1 July, 1994 to 30 June, 1997 has been carried out. The medical evidence form was also used for additional variables, but because of fewer available values, this was analyzed in a separate model. Outcomes were patient characteristics and survival associated with HCV+.
Of 28 692 recipients with valid HCV serologies, 1624 were HCV+ at transplant (5.7% prevalence). In logistic regression analysis, HCV+ was associated with African-American race, male gender, cadaveric donor type, increased duration of pre-transplant dialysis, previous transplant, donor HCV+, recipient (but not donor) age, serum albumin, alcohol use, and increased all-cause hospitalizations. Diabetes and IgA nephropathy were less associated with HCV+. Total all-cause, unadjusted mortality was 13.1% in HCV+ vs. 8.5% in HCV– patients (p < 0.01 by log rank test). In Cox regression, mortality was higher for HCV+ (adjusted hazard ratio = 1.23, 95% confidence interval = 1.01–1.49, p = 0.04).
HCV+ recipients were more likely to be African-American, male, older, and to have received repeat transplants and donor HCV+ transplants. HCV+ recipients also had substantially longer waiting times for transplant. In contrast to recent studies, diabetes did not have an increased association with HCV+ , perhaps due to limitations of the database. HCV+ recipients had increased mortality and hospitalization rates compared with other transplant recipients.
Liver disease is now recognized as one of the leading causes of death in long-term survivors of renal transplantation (1). Hepatitis infections are the main causes of liver injury (2–4). The prevalence of hepatitis C virus seropositivity in kidney transplant patients has ranged from 20 to 30% in single center studies (5,6). However, the course of patients with end-stage renal disease (ESRD) who are seropositive for hepatitis C virus (HCV+) is incompletely characterized (7). Most studies agree that HCV+ ESRD patients have reduced survival compared with other ESRD patients, whether on dialysis or after renal transplantation, although this difference is usually apparent only after several years (8–11). At least one study has shown that HCV+ patients had better survival after renal transplantation compared with HCV+ patients on the transplant waiting list (12). However, none of these studies had sufficient sample size to detect early (< 3 years after transplantation) differences in patient survival or complication rates. Further, risk factors for HCV in the general population have not been demonstrated in a renal transplant population. Because the percentage of ESRD patients who are HCV+ will likely increase in the future, a better understanding of the nature of this population and its survival after renal transplantation is critical. Therefore, to better characterize the outcomes of renal recipients who are HCV+ , the present study analyzed the 1999 USRDS to assess factors associated with HCV+ and the comparative survival of HCV+ recipients.
A national registry (the 1999 USRDS) was retrospectively analyzed to study the patient characteristics and relative patient survival of renal transplant recipients with positive hepatitis C serology at the time of renal transplant from 1 July, 1994 to 30 June, 1997. The dependent variable was selected from ‘HCVSCRN’ in the file SAF.TXUNOS. No details on the methods for determining positive hepatitis C serology were available, but presumably were based on hepatitis C ELISA results. Results for HCVSCRN were reliable only after 1994; hence the time-frame of the present study. Additionally, variables for hepatitis C serology (including hepatitis C confirmed and hepatitis C RIBA) were too incomplete for analysis. The variables included in the USRDS standard analysis files (SAFs), as well as data collection methods and validation studies, are listed at the USRDS website (http://www.usrds.org), under ‘Researcher’s Guide to the USRDS Database' [Section E, ‘Contents of all the SAF’s' (Standard Analysis Files) and published in the USRDS]. The demographics of the dialysis population have been previously described (1999 USRDS report). Patient characteristics and treatment factors were those at the initiation of ESRD. Information on use or results of antibiotics, antiviral or interferon therapy, alcohol, tobacco, confirmatory serologic testing, liver or renal biopsy specimens, or radiologic procedures was not available. No information was available on APACHE scores, urinary catheterization, mechanical ventilation status, or intravenous therapy. The medical evidence form (SAF.MEDEVID) was also used for additional information available at the time of transplant, but this information was only available after 1995 and limited sample size during analysis. The USRDS researcher's agreement specifically prohibits patient contact or chart review, or analysis of center-specific data.
All analyses were performed using SPSS 9.0 (SPSS, Inc., Chicago, IL). Files were merged and converted to SPSS files using DBMS/Copy (Conceptual Software, Houston, TX). Statistical significance was defined as p < 0.05. Univariate analysis was performed with chi-square testing for categorical variables and Student's t-test for continuous variables. Variables with p < 0.10 in univariate analysis for a relationship with HCV+ were entered into multivariate analysis as covariates.
Because of the short duration of the study, and with the available literature indicating that differences in graft survival with hepatitis C occur late after renal transplant, graft survival was not assessed. Maintenance medications were also not assessed, since the combined information available on hepatitis C status and newer immunosuppressive medications (mycophenolate, tacrolimus, and rapamycin in particular) was insufficient for analysis.
Multivariate analysis with stepwise logistic regression was performed with HCV+ as the dependent variable. Because of the greater number of missing values when using variables from the medical evidence form, two models were used, one including the medical evidence form variables and the other not. In both models, donor and recipient age, race, gender, duration of dialysis prior to transplant, donor type, previous transplant, donor HCV+ , and cause of ESRD were covariates including diabetes (type I and II grouped together, as the USRDS does not distinguish types well), hypertension, glomerulonephritis (both combined as individual types of glomerulonephritis).
Patient survival was calculated as the time from the date of transplant until death, last follow-up visit, or loss to follow-up. Patients who were lost to follow-up or who experienced graft loss were censored. Kaplan–Meier analysis was used to determine patient survival comparing HCV+ with HCV– recipients. Patients were censored at graft loss, loss to follow-up, or at the most recent follow-up visit. Because of the short duration of follow-up, life tables analysis was also performed, with heavier weights on earlier departure (death), which compared the two levels of HCV status (+ vs. −) with Gehan test (or the so-called generalized Wilcoxon test), which weighted each failure time by the total number at risk at that time so that earlier times received greater weight than later times. Cox regression analysis was used to determine the independent association of HCV+ with patient survival, using the covariates age, race, gender, end-stage renal disease due to diabetes, weight, year of transplant, duration of pre-transplant dialysis, previous transplant, donor and recipient age, donor and recipient race, donor and recipient gender, delayed graft function, antibody induction therapy (combined and also analyzed separately for OKT3 and ALG), and allograft rejection. Cases were censored as in Kaplan–Meier analysis above.
Of 33 479 renal transplant recipients in the study period, 28 692 had valid HCV serology entries, of which 1624 were HCV+ at the time of transplant (5.7% prevalence). Table 1 shows the results of univariate analysis of factors associated with HCV+ recipients. Notable associations were African-American race, male gender, increased recipient and donor age, increased recipient weight (although body mass index was not significant), history of hemodialysis (vs. peritoneal dialysis), cadaveric donor, combined kidney–liver transplant, combined kidney–pancreas transplant (negative association), donor hepatitis C-positive, increased duration of pre-transplant dialysis, pre-transplant dialysis as a categorical (yes/no) variable, hospitalizations (all causes), and alcohol, tobacco, and drug use. Diabetes and IgA nephropathy as causes of ESRD were less associated with HCV+ , and serum albumin at initiation of dialysis was lower in HCV+. No kidney–liver transplant recipients had ESRD due to IgA nephropathy. Causes of hospitalization for HCV+ were not significantly different compared with HCV– recipients.
Table 1. : Factors assessed in renal transplant recipients, 1 July, 1994 to 30 June, 1997
All other renal transplant recipients
Data are given as the number (% of total) or mean ± 1 standard deviation of recipients with factor who were HCV+ (column 1) or HCV– (column 2) at the time of transplant.
In logistic regression analysis (Table 2), African-American race, male gender, cadaveric donor type, increased duration of pre-transplant dialysis, previous transplant, donor HCV+ , recipient (but not donor) age, and hospitalizations were associated with HCV+ (model 1). Diabetes and IgA nephropathy were less associated with HCV+. In model 2, serum albumin, alcohol use, and hemodialysis were associated with HCV+.
Table 2. : Multivariate analysis (by logistic regression) of factors associated with recipient hepatitis C serology at the time of renal transplantation, 1 July, 1994 to 30 June, 1997
Adjusted odds ratio for HCV+ (95% CI)
NA, not included in model; CI, confidence interval; ESRD, end-stage renal disease.
Only factors significant (p < 0.01) in the final model are shown.
Total all-cause, unadjusted mortality was 13.1% for HCV+ vs. 8.5% for HCV– (p < 0.01 by log rank test). Figure 1 shows a Kaplan–Meier plot of unadjusted survival comparing HCV+ with HCV–recipients. In life tables analysis with greater weight on earlier deaths, HCV+ mortality was significantly lower than HCV– mortality in the time-frame of the study (p = 0.0001). In Cox regression, the adjusted hazard ratio for mortality for HCV+ was 1.23 (p = 0.04, 95% confidence interval = 1.01–1.49). There were no interactions between HCV+ and allograft rejection or induction antibody therapy. In contrast to prior studies by Mathurin et al. (11), African-American race was not significant (p = 0.41, hazard ratio = 1.05, 95% confidence interval = 0.94–1.17). Other covariates in the analysis significantly associated with mortality were greater duration of pre-transplant dialysis, cadaveric donor, previous transplant, diabetes, recipient age, and donor age.
The primary cause of death was incomplete or unknown for 84.5% of the 213 HCV+ patients who died and for 76.8% of the 2293 HCV– patients who died. Secondary causes of death were even less complete. This limitation of the USRDS database has been addressed previously by other investigators using sensitivity analysis (13), which was beyond the scope of the present investigation. Because of the large number of missing values, causes of death could not be compared statistically. However, the leading specified causes of death for HCV+ recipients were infection (45.5%), cardiovascular disease (33.3%), and liver failure (6.1%). Malignancy accounted for only 3.0% of deaths in these recipients. The leading specified causes of death in HCV– patients were cardiovascular diseases (46.8%), infection (22.9%), and malignancy (8.3%). Liver failure accounted for only 0.6% of specified deaths among HCV– recipients.
In the general population, cases of hepatitis C infection are expected to increase dramatically in the future (14). The prevalence of hepatitis C seropositivity in the US renal transplant population was lower in the present study than reported by Chan et al. (6) in a Chinese population. In an American population, Kazi et al. (15) reported a prevalence of HCV seropositivity of 15% in long-term renal transplant recipients, but 5% in hemodialysis patients, which is in agreement with the findings of the present analysis (i.e. patients at the time of receipt of renal transplantation). Demographic risk factors for HCV in the general population include African-American race, male gender, and age 30–39, independent of prior blood transfusion (16). HCV infection is cleared less frequently in the elderly and African-Americans (compared with other races). End-stage liver disease is more common in HCV+ patients over the age of 38 (17). The present analysis confirmed risk factors for HCV of African-American race, male gender, and age in the renal transplant population.
This analysis may underestimate the number of HCV+ patients transplanted in the US. Although the clinical details of patient selection for transplant were unknown, these results show that HCV+ patients can have successful outcomes after cadaveric renal transplant. The association with diabetes does not confirm recent studies demonstrating an increased association of type II diabetes with hepatitis C-positive individuals (18). This could have occurred because diabetes as a cause of end-stage renal disease is one of the most severe manifestations of diabetes, and would have been compared with diabetic recipients with other causes of ESRD. The USRDS also does not reliably distinguish between types of diabetes. The decreased association with IgA nephropathy is also somewhat surprising since cirrhosis has been considered a cause of secondary IgA nephropathy (19,20). However, no liver–kidney transplant recipients (who would be expected to have cirrhosis) in the present analysis had ESRD due to IgA nephropathy. There may have been an element of negative selection in this association.
Most studies have not found survival differences in HCV+ recipients in the early post-transplant course (21). Gentil et al. (10) found reduced survival among HCV+ recipients starting at 5 years post-transplant. However, patients in this study routinely received antilymphocyte induction therapy, which in preliminary reports has been linked with increased mortality in HCV+ recipients (22), although it was not a significant interaction term in the present analysis. A larger study by Mathurin et al. (11) had similar findings; in fact, the survival curves for HCV+ and HCV– patients began to separate at 3 years post-transplant. At 2 years post-transplant, patient survival for both groups in the Mathurin et al. (11) study was higher than in the present study, as might be expected in comparing a national with a single-center study and the lower percentage of diabetic patients transplanted (8.0% in Mathurin et al. vs. 25.8% in the present study). However, if magnified, the separation of curves appears similar in Figure 1 of the present study and Figure 4 of Mathurin et al (11). The continued significance of positive hepatitis C serology on outcomes in the present study persisted after accounting for recipients of African ancestry, also in agreement with Mathurin et al.'s (11) study.
Causes of death have been unreliable in studies of renal transplantation, since > 50% of causes are blank or listed as unknown, and this rate was even higher in the present study, presumably because of its shorter mean follow-up time. Nevertheless, no previous studies have analyzed such a large a sample of patients. The early increase in mortality seen in HCV+ recipients, while not overwhelming, would appear to be clinically significant, and reinforces single-center studies reporting increased late mortality. The increased risk of mortality was independent of allograft rejection, antibody induction therapy, delayed graft function, or other established factors. The short follow-up time of the study and known limitations of causes of death preclude firm conclusions from the present analysis, but would suggest that death from infection may be increased in HCV+ recipients of renal transplantation. Liver disease was the leading cause of death in HCV+ recipients in Mathurin et al. (11), but cardiovascular death was much less common than in either the present study or as reported in American renal transplant recipients (http://www.usrds.org).
No previous studies have analyzed differences in hospitalizations between HCV+ and HCV– recipients. Although causes of hospitalization were not significantly different between HCV+ and HCV– patients, HCV+ patients were at greater risk for hospitalization overall in the present analysis. As in mortality, this increase was independent of allograft rejection or other established factors.
The limitations of the current analysis are detailed in the ‘Methods’ section. Namely, hepatitis serology could not be confirmed, and the clinical details of recipients were not known. Information on liver biopsy results and treatment history was likewise unavailable. Nevertheless, the study does describe the outcomes of the largest sample of renal transplant recipients with positive hepatitis C serologies as is likely possible. It thus minimized referral bias and sampling error. The agreement with previously established risk factors for hepatitis C, as well as concordance between recipient and donor hepatitis C and combined liver–kidney transplantation, indicate the analysis studied its intended population.
Although this is the first study to demonstrate an increased risk of morbidity (hospitalization) and mortality for HCV+ recipients early after renal transplantation, there are a number of reasons why HCV+ patients would be at risk. The relationship between immunosuppression and worsening of clinical disease in hepatitis C is well known (23). Various regimens, including interferon, have been attempted in the peritransplant course, with mixed results (24–29). These studies also suggest that interferon treatment, especially after transplantation, may stimulate the immune system and lead to allograft rejection. While this complication is not quite so adverse in renal transplantation as in liver transplantation (because of the availability of dialysis as an alternative to renal transplantation), it is nonetheless a serious limitation of therapy. Clearly, more effective therapy for hepatitis C, ideally prior to transplantation, is needed.
Prospective studies of therapy in HCV+ renal transplant recipients would require detailed information on the prevalence and risk factors for hepatitis C, which the present study provides. The current analysis also indicates that the duration of such studies may not need to be as long as previously thought, since significant differences in outcomes, particularly in hospitalizations, were seen in less than 3 years post-transplant. The information provided in the present study should prove useful to clinicians and patients alike, and may serve as a benchmark for future studies.