Pretransplant survival is shorter in HIV-positive than HIV-negative subjects with end-stage liver disease

Authors

  • Margaret V. Ragni,

    Corresponding author
    1. Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA
    2. Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA
    3. Starzl Transplantation Institute, University of Pittsburgh School of Medicine, Pittsburgh, PA
    4. Department of Biostatistics, Graduate School of Public Health, Pittsburgh, PA
    • Division of Hematology/Oncology, Department of Medicine, University of Pittsburgh School of Medicine, Director, Hemophilia Center of Western Pennsylvania, 3636 Boulevard of the Allies, Pittsburgh, PA 15213-4306

    Search for more papers by this author
    • Telephone: 412-209-7288; FAX: 412-209-7281

  • Bijan Eghtesad,

    1. Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA
    2. Starzl Transplantation Institute, University of Pittsburgh School of Medicine, Pittsburgh, PA
    Current affiliation:
    1. Department of Surgery, Cleveland Clinic Foundation, Cleveland, OH
    Search for more papers by this author
  • Kimberly W. Schlesinger,

    1. Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA
    2. Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA
    3. Starzl Transplantation Institute, University of Pittsburgh School of Medicine, Pittsburgh, PA
    4. Department of Biostatistics, Graduate School of Public Health, Pittsburgh, PA
    Search for more papers by this author
  • Igor Dvorchik,

    1. Starzl Transplantation Institute, University of Pittsburgh School of Medicine, Pittsburgh, PA
    2. Department of Biostatistics, Graduate School of Public Health, Pittsburgh, PA
    Search for more papers by this author
  • John J. Fung

    1. Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA
    2. Starzl Transplantation Institute, University of Pittsburgh School of Medicine, Pittsburgh, PA
    Current affiliation:
    1. Department of Surgery, Cleveland Clinic Foundation, Cleveland, OH
    Search for more papers by this author

Abstract

Despite improved survival after liver transplantation (OLTX) in HIV-positive individuals treated with highly active antiretroviral therapy (HAART), some transplant candidates do not survive to OLTX. To determine if pretransplant outcome is related to severity of liver disease and/or HIV infection, we prospectively evaluated 58 consecutive HIV-positive candidates seen at a single center from 1997-2002. Of the 58, 15 (25.9%) were transplanted, whereas 21 (36.2%) died before OLTX, a median one month of evaluation, with more than half of those (12 of 21, 57.1%) dying from infection. By contrast, of 1,359 HIV-negative candidates, 860 (63.3%) were transplanted, whereas 211 (15.5%) died before OLTX (P < 0.001). The cumulative survival following initial evaluation was significantly shorter among HIV-positive than HIV-negative candidates (880 vs. 1,427 days, P = 0.035, Breslow) but was not related to the initial pretransplant MELD score (16 vs. 15), INR (1.5 vs. 1.5), creatinine (1.3 vs. 1.3 mg/dL), or total bilirubin (6.6 vs. 5.7 mg/dL), respectively, all P > 0.05. Among untransplanted HIV-positive candidates, the 21 who died did not differ from the 22 surviving in initial MELD (15 vs. 13), CD4 (230 vs. 327/μL), HIV load (both < 400 copies/mL), HAART intolerance (10/21, 47.6% vs. 10/22, 45.4%), or HCV infection (16/21, 76.2% vs. 16/22, 72.3%), all P > 0.05. Further, the 21 did not differ from the 15 transplanted in pre-OLTX CD4, HIV load, or MELD score, all P > 0.05. In conclusion, pretransplant survival appears shorter in HIV-positive OLTX candidates and is unrelated to severity of liver or HIV disease. Further study is warranted to determine risk factors for poorer pretransplant outcomes. (Liver Transpl 2005;11:1425–1430)

Although progression to end-stage liver disease (ESLD) occurs more rapidly in those with HIV infection than in those who are HIV negative,1 survival after liver transplantation (OLTX) in HIV-positive individuals with ESLD has improved,2 likely related to the improved control of HIV replication possible with highly active antiretroviral therapy (HAART).3–5 Despite these improved outcomes after transplantation, however, some HIV-positive candidates do not survive long enough to be transplanted, despite an allocation system independent of HIV status.6 We hypothesized that poorer pretransplant survival is related to the severity of pretransplant liver disease and/or HIV disease. To test this hypothesis, pretransplant outcomes were prospectively studied in HIV-positive candidates initially evaluated at the Thomas Starzl Transplantation Institute between 1997 and 2002 and compared with HIV-negative transplant candidates evaluated during the same period.

Abbreviations

OLTX, orthotopic liver transplantation; HIV, human immunodeficiency virus; HAART, highly active antiretroviral therapy; MELD, Model for End-Stage Liver Disease; ESLD, end-stage liver disease.

Patients and Methods

Subjects

Fifty-eight consecutive subjects with HIV infection and ESLD, initially evaluated for OLTX between September 1, 1997 and December 31, 2002 at the University of Pittsburgh Medical Center, were prospectively followed for clinical outcome, transplantation, and death. All were participating in a Liver Transplant in AIDS/HIV Study.2 Subjects were prospectively followed from the initial evaluation for OTLX, in the outpatient clinic and during hospital admissions, to transplantation or death. All but one was receiving HAART; one received combination nucleoside reverse transcriptase inhibitor-only therapy. The comparison group consisted of HIV-negative transplant candidates initially evaluated for OLTX during the same time period. The study was approved by the University of Pittsburgh Biomedical Institutional Review Board.

Methods

Data collection was based on medical records, documents, pathological specimens, or diagnostic specimens available in computerized hospital and clinic databases at the University of Pittsburgh, including EDIT, MARS, and Sunquest. These data were maintained in a file with code numbers for each subject, with no subject identifiers or linkage codes. The Model for End-Stage Liver Disease (MELD) score7–9 was calculated by the following formula: MELD = 10 *{0.378 ln (total bilirubin mg/dL) + 1.120 ln (INR) + 0.957 ln (creatinine mg/dL) + 0.643}. Causes of death were obtained by clinical chart and/or hospital records review, next of kin, or death certificate. Survival was measured from the time of initial transplant evaluation to transplantation, death, or December 31, 2002. “Antiretroviral intolerance” was defined as permanent discontinuation of antiretroviral drugs because of toxicity.

HIV-Negative Comparison Group

To compare survival after initial evaluation for OLTX between HIV-positive and HIV-negative subjects with ESLD, we obtained data on 1,359 HIV-negative individuals who had undergone initial evaluation for ESLD at our center during the same period of time. The outcome, MELD score, total bilirubin, INR, and creatinine were collected as part of standard of care and were available on the EDIT data retrieval system. The causes of ESLD in the 1,298 HIV-negative candidates, on whom these data were available, included alcoholic liver disease in 367 (28.3%); hepatitis C, 350 (27.0%); cholestatic liver disease, 151 (11.6%); cryptogenic liver disease, 116 (8.9%); hepatitis B, 51 (3.9%); and other, 263 (20.3%).

Statistical Analysis

Data are presented as frequencies and percentages for categorical data and as means and ranges for continuous data. Categorical data were evaluated by χ2 or Fisher's exact test, and continuous data were analyzed by Student t test. Time-to-event distributions were estimated by the product-limit estimate method, and survival distributions were compared by log-rank test.10

Subjects were censored at their last known contact. Missing data were not exclusively problems of those with the severest disease, and to control for severity of illness and length of follow-up, only data from the time of initial pretransplant evaluation were used. Statistical analyses were adjusted based on the number of subjects on whom data were available.

Results

A total of 58 HIV-positive transplant candidates underwent initial transplant evaluation between September 1, 1997, and December 31, 2002. The majority of the 58 HIV-positive subjects were Caucasian males, with a median age of 45 years. The most common cause of ESLD was hepatitis C, 20 of whom had past hepatitis B but not currently active (Table 1). Three subjects had fulminant hepatic failure, one associated with Viramune treatment. Eleven subjects (19.0%) also had a history of alcoholism, including 10 nontransplanted subjects, 5 survivors and 5 nonsurvivors. By risk group, approximately one-third each acquired HIV through sexual activity, intravenous drug use, or hemophilia. The latter group was younger, 38 vs. 47 years (P < 0.001); more likely to be hepatitis C virus (HCV) infected, 17 of 18, 94.4% vs. 27 of 40 (67.5%, P = 0.044); and less likely to be alcoholic, 0 of 18, 0%, vs. 11 of 40 (27.5%, P = 0.012).

Table 1. Baseline Clinical Characteristics of 58 HIV-Positive Transplant Candidates
CharacteristicNo. (%)
  • Abbreviations: ESLD, end-stage liver disease; PI, protease inhibitor; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor.

  • *

    Twenty subjects also had past concomitant hepatitis B, all HBsAg-negative at the time of testing.

  • Laboratory data were obtained on the date of initial transplant evaluation. The number of subjects on whom data were available include bilirubin, n = 49; creatinine, n = 52; INR, n = 49; MELD, n = 44; SGPT, n = 44; platelets, n = 51; CD4, n = 55; HIVRNA, n = 49.

  • The MELD (model for end-stage liver disease) score = 10 *{0.378 ln (total bilirubin mg/dL) + 1.120 ln (INR) + 0.957 ln (creatinine mg/dL) + 0.643}.

  • §

    Regimens containing either of the PI or NNRTI drug classes are classified as highly active antiretroviral therapy (HAART) regimens.

Median age (yr)45 (25–66)
Race 
 Caucasian48 (82.8)
 African-American8 (13.8)
 Other/unknown2 (3.4)
Gender 
 Male51 (87.9)
 Female7 (12.1)
Cause of ESLD 
 Hepatitis C* 44 (75.9)
 Hepatitis B9 (15.5)
 Fulminant hepatic failure3 (5.2)
 Cryptogenic hepatic failure2 (3.4)
HIV risk group 
 Hemophilia18 (31.0)
 Intravenous drug use18 (31.0)
 Homosexual, bisexual16 (27.6)
 Heterosexual3 (5.2)
 Transfusion recipient3 (5.2)
Preoperative laboratory tests (median, range)  
 Bilirubin, total (mg/dL)2.5 (0.5–38.5)
 Creatinine (mg/dL)0.9 (0.2–4.7)
 INR1.3 (1.0–2.7)
 MELD score13 (1–32)
 SGPT (U/mL)90 (21–651)
 Platelets (No. × 103/μL)77 (21–269)
 CD4+ (No./μL)290 (50–1,000)
  No. < 200/μL15/55 (27.3)
 HIV RNA PCR (copies/mL)<50 (<25–750,000)
  No. > 400 copies/mL16/49 (32.6)
Preoperative antiretroviral Rx (ever)§  
 PI, NNRTI-based therapy57 (98.3)
 NRTI only1 (1.7)
Preoperative antiretroviral Rx intolerance24 (41.4)

Of the 58 HIV-positive transplant candidates, 15 (25.9%) underwent OLTX and, of the 43 not transplanted, 21 (48.8%) died before OLTX. By contrast, of 1,359 HIV-negative candidates, 860 (63.3%) were transplanted, and, of the 499 who were not transplanted, 211 (15.5%), died before OLTX, P < 0.001. The cumulative survival following initial evaluation was significantly shorter among HIV-positive than HIV-negative candidates, 880 vs. 1,427 days (P = 0.035, Breslow, Fig. 1). Baseline laboratory assessment of the severity of ESLD in the HIV-positive group, however, was not worse than that in HIV-negative subjects, as measured by the initial pretransplant MELD score, 16 vs. 15; INR, 1.5 vs. 1.5; creatinine, 1.3 vs. 1.3 mg/dL; or total bilirubin, 6.6 vs. 5.7 mg/dL (all P > 0.05, Table 2).

Figure 1.

Survival after evaluation.

Table 2. Baseline Labs in HIV-Positive and HIV-Negative Liver Transplant Candidates (Transplant Evaluation Period, September 1, 1997-December 31, 2002)
CharacteristicsLiver Transplant Candidates* P
HIV-Positive n = 58HIV-Negative n = 1,359
  • *

    Data presented are based on the first available pretransplant laboratory values for HIV-positive and HIV-negative subjects, which in most cases, but not all, was the date of initial pretransplant evaluation.

  • The number in parentheses refers to number of subjects on whom data were available.

MELD16 ± 1.4 (52) 15 ± 0.3 (1,096)0.632
Bilirubin, total (mg/dL)6.6 ± 1.2 (56)5.7 ± 0.2 (1,326)0.285
Creatinine (mg/dL)1.3 ± 0.1 (57)1.3 ± 0.0 (1,327)0.795
INR1.5 ± 0.1 (55)1.5 ± 0.0 (1,104)0.677

Among the 43 nontransplanted HIV-positive candidates, the 21 who died did not differ from the 22 who survived in pretransplant MELD (15 vs. 13), CD4 (230 vs. 327/μL), HIV RNA PCR (both <400 copies/mL) nor in the proportion with HAART intolerance (10/21, 47.6% vs. 10/22, 45.4%) or the proportion with HCV infection (16/21, 76.2% vs. 16/22, 72.3%; all P > 0.05, Table 3). Neither did the 21 who died differ from the 15 transplanted in pre-OLTX CD4, HIV load, HAART tolerance, or MELD score (all P > 0.05, Table 3).

Table 3. Demographic and Clinical Characteristics of HIV-Positive Transplant Candidates
CharacteristicNot TransplantedTransplanted P Values*
SurvivorNonsurvivor 
n = 22n = 21n = 15
  • Abbreviations: ESLD, end-stage liver disease; PI, protease inhibitor; NNRTI, nonnucleoside reverse transcriptase inhibitor; HAART, highly active antiretroviral therapy.

  • *

    The first P value pertains to the comparison in untransplanted subjects between survivors (n = 22) and nonsurvivors (n = 21). The second P value pertains to the comparison between untransplanted (n = 43) and transplanted (n = 15) subjects.

  • Twenty of the subjects, including 6 survivors, 8 nonsurvivors, and 6 transplanted also had past concomitant hepatitis B virus infection but were HBsAg-negative at the time of testing.

  • The number of subjects, including survivors, nonsurvivors, and transplanted on whom data were available include bilirubin, n = 20, 12, and 15; creatinine, n = 22, 14, and 14; INR, n = 21, 12, and 14; MELD, n = 17, 11, and 13; SGPT, n = 21, 12, and 15; platelets, n = 22, 14, and 13; CD4, n = 21, 21, and 13; HIVVL, n = 21, 16, and 12, respectively.

  • §

    A single transplanted subject taking combination nucleoside reverse transcriptase inhibitors (NRTIs) to which he was intolerant was not included, as it was not a HAART-containing regimen.

Age, yrs, median (range)46 (28–56)43 (25–56)44 (35–66)0.2290.153
Gender (No. males:females)19:319:213:21.001.00
Race     
 Caucasian17 (77.3%)18 (85.7%)13 (86.7%)0.6981.00
 African-American4 (18.2%)2 (9.5%)2 (13.3%)0.6641.00
 Other1 (4.5%)1 (4.8%)0 (0.0%)1.001.00
Risk     
 Homosexual, bisexual6 (27.3%)5 (23.8%)5 (33.3%)1.000.332
 Heterosexual2 (10.5%)0 (0.0%)1 (6.7%)0.4881.00
 Hemophilia5 (22.7%)8 (38.1%)5 (33.3%)0.3321.00
 Intravenous drug user7 (31.8%)7 (33.3%)4 (26.7%)0.9160.756
 Transfusion recipient2 (9.1%)1 (4.8%)0 (0.0%)1.000.561
Cause of ESLD     
 Hepatitis C 16 (72.3%)16 (76.2%)12 (80.0%)1.001.00
 Hepatitis B4 (18.2%)3 (14.3%)2 (13.3%)1.001.00
 Fulminant ESLD1 (4.5%)1 (4.8%)1 (6.7%)1.001.00
 Cryptogenic ESLD1 (4.5%)1 (4.8%)0 (0.0%)1.001.00
Baseline laboratory tests (median, range)      
 MELD score13 (1–24)15 (8–32)14 (3–24)0.3480.633
 Bilirubin (mg/dL)2.3 (0.6–38.5)3.1 (1.5–13.8)2.5 (1.1–20.7)0.8570.603
 Creatinine (mg/dL)1.0 (0.5–4.7)0.9 (0.6–2.0)0.8 (0.2–1.6)0.3400.197
 INR1.2 (1–2.7)1.4 (1.2–1.7)1.3 (1.1–2.4)0.8590.474
 SGPT (U/mL)81 (19–276)54 (32–154)62 (21–651)0.3560.401
 Platelets (No. × 103/μL)91 (21–269)61 (30–142)60 (31–179)0.1310.643
 CD4 (No./μL)327 (86–879)230 (1–1,000)300 (103–530)0.8710.707
  No. < 200/μL6/21 (28.6%)7/21 (33.3%)2/13 (15.3%)1.000.477
 HIV RNA (copies/mL)<50 (<50–17,000)<225 (<25–750,000)<400 (<50–120,552)0.1160.779
  No. > 400 copies/mL6/21 (28.6%)5/16 (31.2%)5/12 (41.7%)1.000.492
Antiretroviral therapy     
 PI, NNRTI Rx ever22/22 (100%)21/21 (100%)14/15 (93.3%)1.000.259
 HAART intolerance10/22 (45.4%)10/21 (47.6%)4/14§ (28.6%)1.000.485

The median time to death among the 21 HIV-positive transplant candidates who died was one month after initial transplant evaluation (range 1-27 months). More than half of the deaths (12 of 21, 57.1%) were from sepsis and/or infection complicating liver failure (Table 4). The proportion dying within one month of evaluation did not differ between those with infectious and noninfectious causes of death (P = 0.102). Acute renal failure complicated 8 infectious and one noninfectious death; pancreatitis, two infectious and two noninfectious deaths; and acute respiratory distress syndrome (ARDS), three infectious deaths, but none of these differences reached significance (P > 0.05).

Table 4. Causes of Death Among 21 HIV-Positive Transplant Candidates with ESLD
CauseNumber (%)Time (Months) from Initial Transplant Evaluation*
  • Abbreviation: ESLD, end-stage liver disease.

  • *

    Data presented as median number of months (and range) to death from initial transplant evaluation.

  • Dental infection, presumed to be a dental abscess, led to systemic infection and death.

  • Infection, type unknown, presumed to be systemic, led to death.

  • §

    Two subjects who died of infectious causes also had pancreatitis.

  • |

    This subject and 8 who died of infectious causes had acute renal failure.

Infectious cause of death12 (57.1)1 (1–27)
 Subacute bacterial peritonitis3 (25.0) 
 Pneumonia2 (16.7) 
 Epidural abcess1 (8.3) 
 Fasciitis, lower extremity1 (8.3) 
 Urosepsis and cellulitis1 (8.3) 
 Catheter sepsis1 (8.3) 
 Dental abscess 1 (8.3) 
 Atypical tuberculosis1 (8.3) 
 Infection, type unknown 1 (8.3) 
Noninfectious cause of death6 (28.6)1 (1–6)
 ESLD, terminal care3 (50.0) 
 Pancreatitis§ 2 (33.3) 
 Multiorgan failure| 1 (16.7) 
Unknown cause of death3 (14.3)2 (1–19)
Total211 (1–27)

Although a similar proportion of those with hemophilia died (8 of 18, 44.4%), as among other risk groups (13 of 40, 32.5%, P = 0.395), there was a trend toward more infectious-related deaths in this group (7 of 8, 87.5%) compared with other risk groups (5 of 13, 38.5%), but this did not reach significance (P = 0.067). Compared with other risk groups, those with hemophilia were younger and likely to have had longer duration HIV and HCV infections, given their 1982-1983 HIV seroconversion11 and >20-year HCV infection.1 Despite this, no differences were detected in median CD4 (257 vs. 230); HIV viral load (<400 vs. <400); or proportion with HCV (7 of 8, 87.5%, vs. 9 of 13, 69.2%), between hemophilic and nonhemophilic groups, respectively (all P > 0.05).

Among HIV-positive candidates, the CD4, HIV viral load, proportion transplanted and surviving did not differ by year of transplant. Comparing the 30 subjects whose first pretransplant evaluation was before 2001 with the 28 first evaluated 2001 or later, the median initial CD4 was 200/μL vs. 245/μL, the proportion with initial HIV VL >400 copies/mL was 7 of 18 (38.9%) vs. 16 of 25 (64.0%); the proportion transplanted by 2005 was 9 of 30 (30.0%) vs. 9 of 28 (32.1%); and the proportion dying by 2005 was 15 of 30 (50.0%) vs. 7 of 28 (25.0%), respectively, all P > 0.05. Although these differences may reflect longer follow-up or more “late referrals” in those initially evaluated for transplantation before 2001, this is not known. The study did not control for “late referrals” as many, if not most, of the subjects were “late” referrals, since AIDS/HIV has been considered a contraindication to transplantation during the time period of this study, 1997-2002, and even since then.

Discussion

In this study of consecutive OLTX candidates, pretransplant cumulative survival was found to be significantly shorter in HIV-positive than HIV-negative transplant candidates. Among the HIV-positive candidates, a subset was also identified with early death prior to transplantation, the majority of whom succumbed quickly to infection. There were, however, no detectable differences in the severity of pretransplant liver failure, as measured by MELD, between the HIV-positive and HIV-negative candidates. Neither were there differences in the severity of HIV infection, as measured by HIV viral load and CD4, between the HIV-positive with early death and those who survived or were transplanted, making it difficult to predict a poor outcome or prevent it. Why early deaths occurred more commonly in HIV-positive individuals, and why, when the majority of these deaths were caused by infection/sepsis, remains unknown.

When the liver fails in ESLD, the essential metabolic, detoxification, and immune response functions of the liver may also fail.12, 13 Thus, when ESLD develops in an HIV-positive individual, the already defective host defense against infection attributed to HIV14 may be further weakened by the immune defects associated with liver failure,12, 15 resulting in a greater vulnerability of the HIV-positive candidate to infection or sepsis.12 To what extent liver function may be further altered during the inflammatory injury associated with the elaboration of inflammatory mediators during sep- sis16–18 is not known. In fact, the occurrence of sepsis in those with ESLD has been shown to be an independent predictor of mortality.19 Thus, it may not be surprising that HIV-positive transplant candidates fare more poorly than HIV-negative candidates. Because the purpose of this study was to evaluate outcomes in HIV-infected individuals, only events associated with death were collected. Thus, although the HIV-positive transplant candidates who survived did have infections, these did not result in death. Further, because the HIV-negative group was included only to compare survival data, no data were available on the infections they experienced.

Although preliminary, these data suggest that the development of ESLD, a prerequisite of evaluation for transplantation, in HIV-infected individuals is associated with high risk of early mortality, although no factors appear to be predictive of a poor outcome. Although risk factors for mortality in HIV-negative individuals with ESLD have been identified, including age >50 years, INR >1.2, SGPT >2 times upper limits, bilirubin >2 mg/dL, creatinine >2 mg/dL,20 these factors were not associated with poor outcome in this HIV-positive cohort. It is possible that the small sample size and short pretransplant period limited the identification of predictors of poor outcomes. Further prospective study of pretransplant survival is warranted.

The implication of these data, although not proven, is that individuals with HIV infection should be seen earlier in the course of their end-stage liver disease, as it is likely that both the immunosuppressive effects of their HIV infection, the immunosuppressive effects of end-stage liver disease, and potential inability to comply with HIV medications due to their illness and/or symptoms may put them at greater risk for infection. Perhaps, with the first sign of end-stage liver disease, those with HIV infection should be seen by the transplant service. We suspect that HIV-infected individuals may be thought to be too ill by some for consideration for transplantation, and by the time some are evaluated, it may be too late for intervention. Because no clinical features could be identified to distinguish those who do poorly from those who do well, further studies are needed to prospectively determine whether differences may be detected at an earlier time point, e.g., at the time that the first signs of ESLD develop. It might be of interest to monitor markers of sepsis or end-organ failure, if and when available, to determine if these markers might predict poorer outcomes. Until then, heightened surveillance and monitoring of HIV-positive individuals who develop ESLD is essential, with early evaluation at the first sign of infection or illness.

In conclusion, pretransplant survival in HIV-positive OLTX candidates is shorter than survival in HIV-negative subjects, and, among the HIV-positive candidates, there is a subset that has shortened pretransplant survival, primarily associated with infection/sepsis-related deaths. These findings, although preliminary, suggest there may be a period of clinical vulnerability in HIV-positive subjects once they develop signs and symptoms of end-stage liver disease. Whether prophylactic antibiotics, at the expense of antibiotic resistance,21 have a role in prevention of early pretransplant infection-related deaths will require prospective clinical trials. Until then, it would seem prudent to monitor HIV-positive transplant candidates very closely for early signs of infection, inform them of the potential risk for infection, and urge them to seek medical attention at the earliest signs or symptoms of infection.

Acknowledgements

The authors thank Ms. Holly Chapman, R.N., and Ms. Suzanne Rohal, R.N., University of Pittsburgh, for help in the care of the patients, and Mrs. Karen Saban for assistance in preparation of the manuscript.

Ancillary

Advertisement