HIV-Positive Renal Recipients Can Achieve Survival Rates Similar to Those of HIV-Negative Patients

Abstract

Background. 

Although patients positive for HIV were once thought to be unsuitable candidates for kidney transplantation, their increasing numbers with end-stage renal disease (ESRD) and the introduction of highly active antiretroviral therapy has indicated that they should no longer be excluded for transplantation. To counteract suggestions that human immunodeficiency virus (HIV) patients received suboptimal kidneys, we provide studies of kidneys transplanted from the same donor into patients with and without HIV.

Methods. 

United Network for Organ Sharing kidney transplant data between 1997 and 2004 were analyzed. Graft and patient survival of 38 HIV patients who had received a renal transplant were compared with the survival of 38 recipients who had received a graft from the same donor.

Results. 

The 38 HIV-positive recipients were younger (49.0 vs. 52.3 years, P=0.14) and had lower peak panel-reactive antibodies (PRA; 5.1% vs. 15.6%, P=0.07) when compared with their bilateral donor to HIV-negative recipients. Sirolimus was used more frequently in HIV patients than in non-HIV patients (36.8% vs. 23.7%, P=0.09). The serum creatinine at 1, 3, and 5 years posttransplantation were higher in HIV patients when compared to non-HIV patients. Although not statistically significant, graft survival was higher among HIV-positive patients compared with their negative controls (76.1% vs. 65.1% at 5 years, P=0.21), as was patient survival (91.3% vs. 87.3% at 5 years, P=0.72). More grafts failed due to death with a functioning graft than rejection in HIV-positive patients.

Conclusion. 

This study supports the position that there is no longer an ethical question surrounding the use of kidneys for HIV-positive patients.

Considering that immunosuppressants can increase the morbidity and mortality of patients with human immunodeficiency virus (HIV) infection, renal transplantation in these patients was once regarded as a waste of scarce organs (¹). A written survey of 148 renal transplant centers in the United States in 1998 demonstrated that the majority of physicians would not transplant kidneys into HIV-infected patients with end-stage renal disease (ESRD) (²). Early articles were usually retrospective case reports and showed that 27–50% of HIV-positive recipients died from acquired immunodeficiency syndrome (AIDS) after 6–13 months follow-up (^{3, 4}). An analysis of the United States Renal Data System also showed reduced graft survival in HIV-positive recipients (⁵).

In 1996, HAART was introduced as the primary treatment for HIV-infected patients. Since then, the life expectancy of HIV positive patients has been significantly prolonged, and 5–10% of HAART-maintained patients also develop HIV-associated nephropathy (HIVAN) and ESRD. At some centers, HIV-positive dialysis patients accounted for 3–5% of those on dialysis (⁶), and their quality of life and survival rates were far from satisfactory. Some centers began to do kidney transplantations for these patients after strict selection. After 1–2 years follow-up, the short-term survival rate of these recipients was comparable to other HIV-negative high-risk populations receiving kidney transplantation (^{6, 7}).

Halpern and colleagues described two distinct ethical questions regarding efficacy of transplantation in HIV-positive patients. Does transplantation benefit individual patients, and secondly, would it benefit other patients more? They proposed comparing outcomes among HIV-positive transplant recipients with those of HIV-negative recipients who had similar demographic and disease-related characteristics (⁸). Their proposal, however, does not address the problem of possible allocation of suboptimal kidneys to HIV patients who had been hesitantly transplanted. Here, we analyzed the results of kidneys taken from the same donor, with one kidney being transplanted into an HIV-positive recipient and the other into a HIV-negative recipient, providing a measure of beneficent among exchangeable kidneys.

PATIENTS AND METHODS

Primary kidney transplants performed between 1997 and 2004 in 269 centers as reported to the UNOS Renal Transplant Registry were analyzed. Within this period, we identified and selected all pairs of duplicated kidneys from same donors (n=38) that were transplanted to one HIV-positive (HIV group) and one HIV-negative patient (non-HIV group).

Statistical Analysis

In both HIV and non-HIV groups, patient demographic characteristics, posttransplant immunosuppressive drugs, and graft function (serum creatinine) were analyzed with paired t and sign-rank tests to detect differences. Cause of graft failure and cause of death were also listed for comparison. Kaplan-Meier curves and log-rank tests were used to compare graft and patient survival rates within the two groups. All P values were two sided. All analyses were performed with Stata (V 9.0, College Station, TX). A P value less than 0.05 was considered statistically significant.

For completeness, the study’s power to detect small-to-moderate differences in survival rates was generally poor due to small numbers of patients. The power was sufficient only when survival differences became large. For example, the power was estimated to surpass 80% for a two-sided, log-rank test once true 5-year survival rates differed by 27+ percentage points. Thus, caution is warranted when interpreting nonsignificant survival results to mean lack of difference because smaller but meaningful differences may still exist between groups.

RESULTS

Demographic characteristics of the two groups are shown in Table 1. The recipients’ age in the HIV group was younger than the non-HIV group (49.0 vs. 52.3 years old, P=0.14). Peak panel reactive antibodies (PRA) before transplantation was lower in the HIV group (5.1% vs. 15.6%, P=0.07), although the two factors had no statistical significance. Hepatitis C virus (HCV) was positive in 28.9% of HIV-positive recipients and 31.6% of HIV-negative patients (P=0.7). For the primary kidney disease, more hypertension related ESRD patients were found in the non-HIV group. Other diseases were more often found in the HIV group, which may include HIVAN.

TABLE 1: Demographic characteristics of the 38 HIV-positive recipients compared with their bilateral donors

The immunosuppressive drugs used at discharge are listed in Table 2. Induction therapy usage was similar between the two groups (60.5% vs. 55.3% P=0.5), but there was a trend toward greater sirolimus usage in the HIV group (36.8% vs. 23.7%, P=0.09). Only one patient in the HIV group was treated via a steroid-free regimen, compared with five patients in the non-HIV group (P=0.18). No differences of delayed graft function (DGF) and acute rejection episodes were shown between the two groups. When comparing the serum creatinine at 6 months, the test results were almost the same, but the 1-, 3-, and 5-year results were consistently, but not significantly, higher in the HIV group.

TABLE 2: Posttransplant regimen and function of the grafts in 38 pairs of recipients

Graft and patient survival rates are shown in Figure 1. The 5-year graft survival in the HIV group was 76.1%, compared to 65.1% in the non-HIV group (P=0.21). Five-year patient survivals rates were similar in the two groups (91.3% vs. 87.3% in the HIV and non-HIV groups respectively, P=0.72).

FIGURE 1.: Survival of 38 pairs of HIV-positive and HIV-negative recipients whose grafts were from a same donor. The 5-year graft survival is similar within HIV-positive and HIV-negative groups (76.1% vs. 65.1%, P=0.21). The 5-year patient survival was more similar (91.3% vs. 87.3%, P=0.72).

Causes of graft failure are summarized in Table 3. Death with a functioning graft (DWF) occurred in four HIV-positive patients (one from bacterial pneumonia, another from gastrointestinal hemorrhage, other two from respiratory failure), but in only one HIV-negative recipient. When comparing the cause of death, regular causes (such as infection, hemorrhage, etc.) were similar in both groups.

TABLE 3: Cause of graft failure and patient death in transplants from the same donor between HIV-positive and HIV-negative recipients within 5 years

DISCUSSION

We compared 38 pairs of HIV/non-HIV patients receiving primary kidney transplants from the same donor’s bilateral grafts. We could not find a statistically significant difference in graft or patient survival rates between the HIV and non-HIV groups when given these exchangeable kidneys. In fact, the HIV-positive recipients had slightly better survival outcomes—a possible reflection of careful patient selection. The HIV patients were younger and less often sensitized, two demographic characteristics known to enhance graft survival rates (⁹). Additionally, the use of sirolimus was more frequent in the HIV group, and previous research has indicated a survival benefit of sirolimus in HIV-positive patients (⁶). Conversely, fewer steroid-free regimens were used in the HIV-positive patients, whereas past study has suggested that a nonsteroid regimen was beneficial in HIV-positive patients (¹⁰). In addition to these factors, other demographic characteristics and posttransplant therapies were compatible between the two groups.

Our findings support two earlier reports. The Michigan group used the data from the Scientific Registry of Transplant Recipients and analyzed graft and patient survival of HIV patients in both the pre-HAART and HAART eras. One-year graft survival rates were similar between HIV-positive and HIV-negative patients (¹¹). Next, Kumar and his colleagues followed 40 HIV-positive renal recipients transplanted between 2001 and 2004 and showed that the 1-year and 2-year graft survival rates for HIV-positive patients were similar to those of other high-risk populations receiving kidney transplants. Also, HIV-1 RNA and CD4 counts of the 40 patients showed no evidence of AIDS for up to 2 years. Furthermore, they reported that most donor kidneys grafted into the HIV-positive recipients were from marginal donors, based on United Network for Organ Sharing (UNOS) criteria (⁶).

In recent years, more and more HIV-positive patients have been placed on dialysis and diagnosed with ESRD (¹²). Not only can HIV infection cause HIVAN (which leads to ESRD within 6 months), the life expectancy of the HIV-infected patients has increased to such a degree that these patients may develop ESRD secondary to preexisting conditions such as diabetes and hypertension (⁶). The pressure to offer kidney transplants to HIV-positive ESRD patients has increased dramatically. According to UNOS policy, an asymptomatic HIV-infected patient “should not necessarily be excluded from candidacy for organ transplantation, but should be advised that he or she may be at increased risk of morbidity and mortality because of immunosuppressive therapy” (www.unos.org). Perhaps this advisory is no longer necessary.

With the introduction of HAART, survival of HIV-positive patients greatly increased. HAART suppresses the replication of HIV virus through a combination of several drugs acting on the immune system. Because information on CD4 cell counts, HIV load, HAART drugs was not available in the UNOS database for our analysis, we could not directly measure the interaction effects of traditional immunosuppressants and HAART. Indirectly, however, fewer HIV-positive recipients lost their grafts to rejection compared to HIV-negative recipients and, conversely, more HIV-positive patients died with a functional graft. In the HIV-positive group, the causes of the four cases of DWF were infection, hemorrhage, and respiratory failure, each of which may be directly related to HIV infection.

With regard to kidney function 1 year after transplantation, average serum creatinine levels tended to be higher among the 38 HIV-positive patients in this study. These higher levels were not statistically significant, most likely owing to small numbers of patients because others have reported this problem and suggested that the use of marginal donors contributed to the cause (⁶). Our higher creatinine values for HIV-positive patients cannot be attributed to suboptimal kidneys because grafts from the same donor appeared in both groups. The increased levels may reflect an action of antiretroviral drugs on serum concentrations of CsA and tacrolimus. By inhibiting specific cytochrome P450 enzymes of the liver, antiretroviral drugs make it difficult to properly adjust immunosuppression dosage levels leading to abnormal values (¹³). In addition, antiviral drugs in HAART may exhibit kidney toxicity because they are predominantly excreted by the renal system (tubular secretion) (¹⁴). Regardless, the evaluated serum creatinine levels do not appear to undermine our main result that, under proper treatment, HIV-positive patients can achieve similar graft and patient survival as HIV-negative recipients.

All the above details suggest that kidney transplantation is safe in HIV-positive patients. Nonetheless, our study has two main limitations: 1) the small number of cases limited the power to detect small but perhaps meaningful differences in survival rates; and 2) the UNOS database lacked specific information on HIV status and HAART treatments. Concerning the first limitation, we cannot analyze more data at the present time because we have culled all available records in the UNOS database for bilateral, same-donor kidneys transplanted into discordant HIV-status patients. Additionally, our study did have sufficient power to detect large survival differences if, in fact, kidneys transplanted into HIV-positive recipients failed at alarming rates. Moreover, our results suggested an opposite effect—slightly improved graft and patient survival rates for HIV-positive patients compared to their matched controls. Regarding the second limitation, larger or center-specific studies are needed to fully assess the interaction of immunosuppressive and HAART therapies. In the future, UNOS may be able to collect pertinent data regarding HIV load and CD4 counts so HIV status may be examined in more detail. Despite these deficiencies, we believe the basic issue of whether graft and patient survival rates among HIV-positive patients are acceptable can be answered from the data at hand.

In conclusion, much of the earlier concerns centering on ethical questions of allografting HIV-positive patients can be considered resolved. Halpern has framed the ethical question clearly by asking, “Does the transplant benefit the individual HIV patient, and would the graft benefit another patient more?” (⁸). If the answer is that it benefits both patients equally, as the current results suggest, then the ethical concern no longer exists. Using same-donor transplants, we could not find evidence of a difference in survival rates between HIV-positive and HIV-negative recipients. The HIV patient should be considered no different than any other patient who may have higher transplant risks than the average patient, such as African American, highly sensitized, diabetic, or older patients. Reluctance to transplant patients with HIV is no longer justified.

ACKNOWLEDGMENTS

We would like to thank Su-Hui Lee for secretarial assistance and Rene Castro for database support.