Background: The influence of intravenous drug abuse (IVDA) on hemodialysis access placement practices and access survival in HIV-infected patients is unknown.
Methods: We conducted a retrospective study of 60, HIV seropositive, maintenance hemodialysis patients. Type of access and assisted access survival (measured from date of placement) were compared in those with (77%) and without (23%) a history of IVDA.
Results: Mean age was 37.8 years, mean baseline serum albumin was 2.9 g/dl and median CD4 count was 222 cells/mm3. Fifteen patients, all IVDA, were dialyzed using only tunneled catheters (median number of catheters per person (range): 2.5 (1-11)). There were longer delays in creation of a permanent access (p = 0.08), but no difference in the type of permanent access placed in IVDA versus the non-IVDA group. Over 1,051 cumulative months of access follow-up, 134 tunneled catheters, 28 grafts and 19 fistulae were placed, with observed failure rates of 1 per 4.7 access-months, 1 per 19.7 access-months, 1 per 38.2 access-months, respectively. The adjusted relative hazard of access failure for grafts versus catheters was 0.41 (95% CI: 0.23, 0.72; p = 0.002) and for fistulae versus catheters was 0.21 (95% CI: 0.08, 0.52; p = 0.001). Thirty-two percent of accesses were removed due to infection, an infection removal rate for catheters of 1 per 7.8 access-months and for grafts of 1 per 62.5 access-months; all graft infections occurred in the IVDA group. No fistula was removed due to infection.
Conclusion: Fistulae are the first line of choice for hemodialysis access in HIV-seropositive patients regardless of IVDA history; if not feasible, graft placement in non-IVDA or abstinent IVDA patients is recommended. In those with active IVDA, the optimal method of renal replacement therapy and type of hemodialysis access remain uncertain.