HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: CJN.02211205v1 1/6/1241    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Tourret, J. Articles by Bagnis, C. I. Search for Related Content PubMed PubMed Citation Articles by Tourret, J. Articles by Bagnis, C. I.

Outcome and Prognosis Factors in HIV-Infected Hemodialysis Patients

Jérôme Tourret^*, Isabelle Tostivint^*, Sophie Tézenas du Montcel, Jennifer Bragg-Gresham, Svétlana Karie^*, Cécile Vigneau, Jean-Baptiste Guiard-Schmid^||, Gilbert Deray^*, and Corinne Isnard Bagnis^*

Departments of ^* Nephrology and Biostatistics and Medical Information, Hôpital Pitié-Salpêtrière, and Departments of; Nephrology and ^|| Infectious Diseases, Hôpital Tenon, Assistance Publique-Hôpitaux de Paris, Paris, France; and Arbor Research Collaborative for Health (formerly University Renal Research and Education Association), Ann Arbor, Michigan

Address correspondence to: Dr. Jérôme Tourret, Service de Néphrologie, Hôpital Pitié-Salpêtrière, 47, Boulevard de l’Hôpital, 75013, Paris, France. Phone: +33-1-42-17-72-27; Fax: +33-1-42-17-79-14; E-mail: jtourret{at}yahoo.com

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
HIV-infected patients who are on hemodialysis have a worse prognosis than noninfected patients who are on hemodialysis. Their outcome in the highly active antiretroviral therapy (HAART) era remains unclear. Outcomes in patients who were enrolled in the French Dialysis in HIV/AIDS (DIVA) cohort were determined in a 2-yr prospective follow-up. All HIV-infected patients who were on hemodialysis in France on January 1, 2002, were included and followed prospectively until January 1, 2004. Patients’ survival was examined by Kaplan-Meier method, and mortality risk factors were examined using uni- and multicovariate analyses. Survival was compared with that of 584 hemodialysis patients who did not have HIV or diabetes and were enrolled in the French Dialysis Outcomes and Practice Patterns Study II (DOPPS II) in the same period (after standardization for the average age, gender, and ethnicity of the DIVA cohort). A total of 27,577 patients were receiving hemodialysis in France at the beginning of the study; 164 (0.59%) were infected with HIV, 72% were male, mean age was 44.8 ± 10.9 yr, and 65% were black. The 2-yr survival rate was 89 ± 2% and statistically indistinguishable from the survival of the French cohort extracted from the DOPPS II study. Significant mortality risk factors were low CD4 cell count (hazard ratio [HR] 1.4/100 CD4 cells per mm³ lower), high viral load (HR 2.5/1 Log per ml), absence of HAART (HR 2.7), and a history of opportunistic infection (HR 3.7), the last two being independent (HR 2.6 and 3.6, respectively). Survival of HIV-infected patients who are hemodialysis has greatly improved. A prospective cohort of paired hemodialysis patients with and without HIV is required to compare better their mortality in the HAART era.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
Since 1996, the use of highly active antiretroviral therapy (HAART) has changed HIV infection from a subacute fatal condition to a chronic systemic disease. We now have to face many long-term comorbid conditions that HIV patients once did not have time to develop. Nearly every organ of the human body can be affected, and all caregivers have to deal with an increasing number of HIV-infected patients. Renal failure can be a direct consequence or an associated condition of HIV infection. HIV associated nephropathy (HIVAN) first was described in 1984 (1), and its prevalence has decreased significantly in the HAART era as compared with the pre-HAART era (2). Advanced HIV disease, as indicated by a low CD4 cell count, is associated with subsequent development of renal disease (3), but many other causes now may lead to renal failure in patients with HIV, including drug-induced toxicity, hypertensive nephroangiosclerosis, and diabetes. ESRD substantially increases the risk for death, cardiovascular disease, and the use of specialized health care (4). Despite 20 yr of intensive HIV research, very little is known about mortality risk factors in ESRD in HIV-infected patients. HIV-infected patients who are on hemodialysis are exposed to a high risk for death because their status combines the cumulative risks of both hemodialysis and HIV infection (5). Cardiovascular diseases, immunosuppression, anemia, weight loss, and osteodystrophy are known to be consequences of both hemodialysis and HIV infection (5–7). Patients with ESRD and HIV are younger, less frequently have diabetes, and have less hypertension than the general population of hemodialysis patients. Therefore, it is very arduous to guess what their survival has become in the HAART era compared with non–HIV-infected hemodialysis patients. We carried out a prospective survival analysis of all HIV patients who were treated with hemodialysis in France to define outcomes better and to look for mortality risk factors.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
Serologic Definitions
In this study, chronic hepatitis B virus (HBV) infection was defined by the chronic (at least 6 mo) carrying of HBs antigen in blood. Hepatitis C virus (HCV) infection was defined by the presence of anti-HCV antibodies in the blood (positive serology).

Study Population
We previously described the clinical and epidemiologic features of the French cohort of HIV-infected hemodialysis patients in 2002 (8). A questionnaire was sent to every French dialysis center (795 public and private centers) asking for the total number of hemodialysis patients and the number of HIV-infected hemodialysis patients on January 1, 2002. Mail reminders were sent and/or telephone reminders were made until answers were obtained from all of the 795 French centers. We then were able to identify 164 HIV patients among the 27,577 hemodialysis patients who were identified on January 1, 2002. These 164 patients formed the Dialysis in HIV/AIDS (DIVA) cohort. A second questionnaire was sent in 2003 to the nephrologist in charge of each of these patients. This form included 45 items that covered demographic, clinical, nephrologic, immunologic, and therapeutic fields. For all of the fluctuating data, the value that we asked for was that of December 2002 or the last available data before death (for patients who died before December 2002). Eventually, a third survey (telephone calls to the nephrologists in charge of each HIV-infected hemodialysis patient) allowed us to obtain the living status (dead or alive) for all of the patients on January 1, 2004. Therefore, we achieved a 2-yr prospective follow-up for this cohort.

Inclusion criteria were age older than 18 yr; renal failure of any origin; at least one hemodialysis session before January 1, 2002; HIV infection diagnosed before January 1, 2002; and hemodialysis in France (including the French Caribbean islands). The single exclusion criterion was end-stage renal failure that was treated by peritoneal dialysis.

Statistical Analyses
Kaplan-Meier estimate of survival was performed. Patients who were withdrawn from hemodialysis (as a result of renal function recovery or kidney transplantation) were censored. Mortality risk factors were tested by univariate analysis (log-rank test for qualitative variables, Cox model for quantitative variables). The results are expressed as hazard ratios (HR; 95% confidence interval [CI]). All significant values with P < 0.1 in the univariate analysis and with <20% of missing data were tested on a multivariate Cox regression model, with both stepwise and backward selection used for model building. The significance level for entering effects and the significance level for removing effects variables were set at 0.05. All tests were two-tailed and were performed using the SAS 8.1 statistical package (SAS Institute, Cary, NC).

The data that were included in the univariate analysis were age, gender, ethnicity, Caribbean patients versus mainland patients, weight, nephropathy responsible for renal deficiency (HIVAN versus other), duration of hemodialysis, duration of HIV infection, time between HIV infection discovery and onset of hemodialysis, HIV risk behavior, HBV-associated infection, HCV-associated infection, CD4 count, HIV viral load, antiretroviral treatment, converting enzyme inhibitors/angiotensin II receptor blocker use, history of opportunistic infection, AIDS stage, hemoglobin level, human recombinant erythropoietin dosage, plasma albumin, calcium-phosphorus product, and parathyroid hormone level.

Comparison with Non–HIV-Infected Hemodialysis Patients
Data from the second phase of the Dialysis Outcomes and Practice Patterns Study (DOPPS II) were used for this analysis and included data from the years 2002 through 2004. Non–HIV-infected patients without diabetes from France-DOPPS only were included (n = 584). DOPPS is an international, prospective, observational study involving adult hemodialysis patients (age >18 yr) who were randomly selected from nationally representative dialysis facilities. Samples were stratified by types of facilities and geographic regions within each country. Additional details of the DOPPS sampling plan and study methods have been described elsewhere (9). Institutional review boards approved the study, and patient consent was obtained in accordance with local requirements. Cox proportional hazards regression was used to standardize the sample on age, gender ratio, and percentage of black patients to that of the HIV population. Models accounted for facility clustering. A total follow-up of 3 yr was possible.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
Sociodemographic Parameters
We identified 27,577 hemodialysis patients in France on January 1, 2002. Assuming a 4% increase of hemodialysis prevalence every year in France (10), this number is consistent with the national cross-sectional study of the "Caisse d’Assurance Maladie" (National Health Insurance company) which identified 30,882 dialysis patients between June 2, 2003, and June 8, 2003 (11). Among these patients, 164 were infected with HIV. Therefore, the global prevalence of HIV infection in French hemodialysis centers was estimated at 0.59%.

Patients mostly were male (72%). The mean age was 44.8 ± 10.9 yr, and 90% of the patients were younger than 60 yr. Two thirds of the patients were black. Twenty-one patients received dialysis in French Caribbean units (12.8% of the total patients and 20% of the black patients). Ninety-five percent of the Caribbean patients were black. HCV and HBV co-infections were found in 27 and 17% of the patients, respectively. The most common route of transmission for HIV was sexual (38%); 15% of the patients were intravenous drugs users. In 23% of the cases, the route of infection for HIV was undetermined. CD4 cell count was <300/mm³ in 48% and <200/mm³ in 26% of the patients. Eighty-six percent of the patients received at least one antiretroviral agent, and 64% of the patients were prescribed HAART. HIV immunologic and virologic parameters were controlled (i.e., undetectable viral load and CD4 count >200/mm³) in 41% of the total patients, in 47% of the patients who received HAART, and in 26% of the patients who were treated by anything but HAART. Table 1 shows their baseline characteristics. For each item of the questionnaire, the number of patients for which the data were available is specified under the "n" column. For every item, the percentage of patients was >85%, with the exception of HIV genotype, parathyroid hormone, and plasma albumin levels (75% for these three items).

View larger version (14K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 1. Kaplan-Meier curves of survival of the HIV-infected hemodialysis patients from the Dialysis in HIV/AIDS cohort (light line) and from the cohort that was extracted from France Dialysis Outcomes and Practice Patterns Study II (heavy line; also see Materials and Methods). Dots indicate patients censored (renal function recovery or kidney transplantation).

Multivariate Analyses
A high viral load (HR 2.6/1 Log per ml; 95% CI 1.7 to 4.2; P < 0.0001) and a clinical history of opportunistic infection (HR 3.6; 95% CI 1 to 12.3; P < 0.05) were independent mortality risk factors (Table 3). To assess whether the viral load was overwhelming other potentially independent prognostic factors, we excluded it from the multivariate analysis and ran a Cox regression model again. When the viral load was excluded, a CD4 cell count <200 cells/mm³ (HR 6.2; 95% CI 2.2 to 17.0; P < 0.0004) and a non-HAART therapy (HR 3.0; CI 1.4 to 10.6; P < 0.01) were independent factors in addition to a clinical history of opportunistic infection. We then verified that there were no statistical interactions among viral load, opportunistic infections, HAART therapy, and CD4 cell count.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

The difference in survival rates between our study and previous ones may be explained by a number of important differences between French and American HIV hemodialysis patient populations. Patients in France were less frequently black (65 versus 83.2 to 90% in US studies [5,12–15]), less frequently intravenous drug abusers (15 versus 53 to 69% [12–14]), and less frequently HCV co-infected (27 versus 67 to 68% [13,14]). Differences in the HIV treatment indications also can be involved, because 86% of the patients were treated with antiretroviral therapy (ART) and 75% were treated with HAART in our study, compared with 18 to 60.7 and 33%, respectively, in the US literature (12,14,15). This higher proportion of HAART-treated HIV-infected hemodialysis patients was consistent with a higher percentage of undetectable patients (54% in our study versus 18% in the study of Rodriguez et al. [13]), a higher mean CD4 cell count (334 versus 140 to 222 cells/mm³ in US literature [12–14]), and a lower percentage of AIDS stage (40 versus 85% in the study of Ifudu et al. [12]).

Another important result of our study is that HIV-related parameters such as viral load and passed opportunistic infections were the key mortality factors, yet immunologic and virologic parameters were controlled in only 41% of the patients (CD4 cell count >200 cells/mm³ and viral load <200 copies/ml). We believe that this result individualizes HIV-infected hemodialysis patients as a very specific subgroup of HIV patients whose ART indication might differ from the rest of the HIV population. Even though there is no absolute consensus about when and how to start ART, most infectious diseases specialists agree that it should be initiated when the CD4 cell count goes under 200 cells/mm³ or after an opportunistic infection (16). We show that in the case of HIV-infected hemodialysis patients, this strategy might not apply. HIV-infected hemodialysis patients whose CD4 cell count was <300 cells/mm³ had a 3.8-fold mortality risk increase in univariate analysis (95% CI 1.2 to 116; P < 0.02; data not shown). Furthermore, patients who did not receive HAART had a worse prognosis than the others, regardless of their CD4 cell count, which was shown already by Ahuja et al. (17). Therefore, we believe that ESRD might constitute a HAART indication by itself.

Hypotheses can be drawn to explain the insufficient control of HIV infection in hemodialysis patients. HAART has been shown successfully to suppress viral replication in HIV-infected hemodialysis patients (17). This also was the case in our study, in which 47% of the HAART-treated patients had a controlled HIV disease compared with only 26% in the non–HAART-treated group, but effective drug dosage might not be adequate because of insufficient knowledge of adaptation of antiretroviral agents in hemodialysis patients.

Another possibility is that chronic renal failure contributes to deepen immunosuppression that is associated with HIV. Inefficient cytokine hyperactivation and subsequent chronic inflammatory syndrome is well documented in hemodialysis patients (18,19), resulting in an altered immune response (20). Immunologic abnormalities that are induced by chronic renal failure may influence HIV infection outcomes. The hemodialysis session itself may have a detrimental effect on HIV infection as a result of the release of cytokines such as IL-6 and TNF-, which have been shown to increase HIV replication in vitro (21). It also has been shown that biocompatible polysulfone hemodialysis membranes were less likely to induce an HIV viral load flare after a routine dialysis session than "regular" cellulose membranes. This effect also is believed to be mediated by IL-6 and TNF- (22).

One also could draw the hypothesis that incidence of resistant strains of HIV is higher in hemodialysis patients than in the general HIV population. To our knowledge, no study has addressed this very important question, and this hypothesis cannot be ruled out.

Only seven cohorts that have analyzed HIV-infected hemodialysis patients’ outcomes have been published so far (5,12–15,17,23). All but two report on data from the pre-HAART era (14,17), and all but one are retrospective (23). These cohorts usually are highly heterogeneous because they pool patients in periods as wide as 1985 through 2000. To our knowledge, only four authors have addressed mortality risk factors directly in HIV-infected hemodialysis patients. Barbiano di Belgiojoso et al. (23) studied 29 patients between 1990 and 1995 and showed that a low number of CD4 cells and an AIDS stage were significantly associated with a higher mortality, yet studying 6166 patients between 1990 and 1999, Ahuja et al. (5) found that only the year of initiation of dialysis was associated with mortality. In a group of 22 patients during the same period, they also showed that HAART could suppress viral load successfully in HIV-infected hemodialysis patients and that their survival on HAART was better than that of patients on suboptimal antiretroviral therapy (17). Rodriguez et al. (13) studied 115 patients between 1985 and 2001 and showed that a low CD4 cell count and a low plasma albumin level were associated with a higher mortality. None of these factors remained significant in the subgroup of patients who initiated dialysis after 1995. Last, Abbott et al. (15) showed that HIVAN was the only mortality risk factor in 3653 patients between 1992 and 1997.

	Conclusion

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
We found that the 2-yr survival rate of HIV-infected hemodialysis patients has dramatically improved since the beginning of the HIV infection epidemic. The main mortality risk factors were viral load and the history of opportunistic infection. HAART was efficient and significantly increased survival in hemodialysis patients. Therefore, hemodialysis itself may be an indication of HAART among HIV-infected patients. Further investigations are required to compare precisely survival in hemodialysis patients with and without HIV.

	Acknowledgments

 
We are indebted to our colleagues from the dialysis centers and nephrology departments in France who made this study possible. We sincerely thank them for the time that they kindly agreed to spend to answer our survey. We also thank our research assistant, Christel Bessette, for high-quality help in the logistic management of the study.

	Footnotes

 
Published online ahead of print. Publication date available at www.cjasn.org.

Received December 21, 2005. Accepted August 8, 2006.

	References

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

 

1. Rao TK, Filippone EJ, Nicastri AD, Landesman SH, Frank E, Chen CK, Friedman EA: Associated focal and segmental glomerulosclerosis in the acquired immunodeficiency syndrome. N Engl J Med310 :669 –673,1984[Abstract]
2. Lucas GM, Eustace JA, Sozio S, Mentari EK, Appiah KA, Moore RD: Highly active antiretroviral therapy and the incidence of HIV-1-associated nephropathy: A 12-year cohort study. AIDS18 :541 –546,2004[CrossRef][Medline]
3. Krawczyk CS, Holmberg SD, Moorman AC, Gardner LI, McGwin G Jr: Factors associated with chronic renal failure in HIV-infected ambulatory patients. AIDS18 :2171 –2178,2004[CrossRef][Medline]
4. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY: Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med351 :1296 –1305,2004[Abstract/Free Full Text]
5. Ahuja TS, Grady J, Khan S: Changing trends in the survival of dialysis patients with human immunodeficiency virus in the United States. J Am Soc Nephrol13 :1889 –1893,2002[Abstract/Free Full Text]
6. Kimmel PL, Barisoni L, Kopp JB: Pathogenesis and treatment of HIV-associated renal diseases: Lessons from clinical and animal studies, molecular pathologic correlations, and genetic investigations. Ann Intern Med139 :214 –226,2003[Abstract/Free Full Text]
7. D’Agati V, Appel GB: HIV infection and the kidney. J Am Soc Nephrol8 :138 –152,1997[Abstract]
8. Vigneau C, Guiard-Schmid JB, Tourret J, Flahault A, Rozenbaum W, Pialoux G, Rondeau E: The clinical characteristics of HIV-infected patients receiving dialysis in France between 1997 and 2002. Kidney Int67 :1509 –1514,2005[CrossRef][Medline]
9. Pisoni RL, Gillespie BW, Dickinson DM, Chen K, Kutner MH, Wolfe RA: The Dialysis Outcomes and Practice Patterns Study (DOPPS): Design, data elements, and methodology. Am J Kidney Dis44 :7 –15,2004[CrossRef][Medline]
10. Jungers P, Robino C, Choukroun G, Touam M, Fakhouri F, Grunfeld JP: Course of chronic renal failure epidemiology and prediction of maintenance dialysis needs in France [in French]. Nephrologie22 :91 –97,2001[Medline]
11. Macron-Nogues F, Vernay M, Ekong E, Thiard B, Salanave B, Fender P, Allemand H: The prevalence of ESRD treated with renal dialysis in France in 2003. Am J Kidney Dis46 :309 –315,2005[CrossRef][Medline]
12. Ifudu O, Mayers JD, Matthew JJ, Macey LJ, Brezsnyak W, Reydel C, McClendon E, Surgrue T, Rao TK, Friedman EA: Uremia therapy in patients with end-stage renal disease and human immunodeficiency virus infection: Has the outcome changed in the 1990s? Am J Kidney Dis29 :549 –552,1997[Medline]
13. Rodriguez RA, Mendelson M, O’Hare AM, Hsu LC, Schoenfeld P: Determinants of survival among HIV-infected chronic dialysis patients. J Am Soc Nephrol14 :1307 –1313,2003[Abstract/Free Full Text]
14. Szczech LA, Kalayjian R, Rodriguez R, Gupta S, Coladonato J, Winston J: The clinical characteristics and antiretroviral dosing patterns of HIV-infected patients receiving dialysis. Kidney Int63 :2295 –2301,2003[CrossRef][Medline]
15. Abbott KC, Trespalacios FC, Agodoa LY, Ahuja TS: HIVAN and medication use in chronic dialysis patients in the United States: Analysis of the USRDS DMMS Wave 2 study. BMC Nephrol4 :5 –2301,2003[CrossRef][Medline]
16. Yeni PG, Hammer SM, Hirsch MS, Saag MS, Schechter M, Carpenter CC, Fischl MA, Gatell JM, Gazzard BG, Jacobsen DM, Katzenstein DA, Montaner JS, Richman DD, Schooley RT, Thompson MA, Vella S, Volberding PA: Treatment for adult HIV infection: 2004 recommendations of the International AIDS Society-USA Panel. JAMA292 :251 –265,2004[Abstract/Free Full Text]
17. Ahuja TS, Borucki M, Grady J: Highly active antiretroviral therapy improves survival of HIV-infected hemodialysis patients. Am J Kidney Dis36 :574 –580,2000[Medline]
18. Descamps-Latscha B, Jungers P, Witko-Sarsat V: Immune system dysregulation in uremia: Role of oxidative stress. Blood Purif20 :481 –484,2002[CrossRef][Medline]
19. Pertosa G, Grandaliano G, Gesualdo L, Schena FP: Clinical relevance of cytokine production in hemodialysis. Kidney Int Suppl76 :S104 –S111,2000[Medline]
20. Girndt M: Humoral immune responses in uremia and the role of IL-10. Blood Purif20 :485 –488,2002[CrossRef][Medline]
21. Kedzierska K, Crowe SM: Cytokines and HIV-1: Interactions and clinical implications. Antivir Chem Chemother12 :133 –150,2001[Medline]
22. Fontana D, Schut R, Rabb H: Can choice of dialyser membrane have a beneficial effect on HIV load in the HIV-infected dialysis patient? Nephrol Dial Transplant17 :529 –530,2002[Free Full Text]
23. Barbiano di Belgiojoso G, Trezzi M, Scorza D, Barone MT, Landriani N, Genderini A, Bertoli S: HIV infection in dialysis centers in Italy: A nationwide multicenter study. J Nephrol11 :249 –254,1998[Medline]

This Article Abstract Full Text (PDF) All Versions of this Article: CJN.02211205v1 1/6/1241    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Tourret, J. Articles by Bagnis, C. I. Search for Related Content PubMed PubMed Citation Articles by Tourret, J. Articles by Bagnis, C. I.