Background:  Renal disease is common among HIV⁺ persons. Comparisons of factors affecting renal function among HIV⁺ vs HIV^– persons are needed.

Methods:  We evaluated longitudinal glomerular filtration rate (GFR) changes, estimated using the modified diet in renal disease (MDRD) equation, and prevalence of proteinuria (defined as a urine protein to creatinine ratio of ≥0.2) and associated factors among HIV⁺ and HIV^– men participating in the Multicenter AIDS Cohort Study (MACS), a study of U.S. gay men.

Results:  We analyzed GFR changes among 2163 men (1206 HIV^–, 852 HIV⁺ on HAART, 105 HIV⁺ not on HAART), each with ≥3 GFR assessments from October 2003 until March 2007. GFR at baseline was >90 for 50.2%, 60 to 90 for 45.5%, and <60 for 4.4% of men; 5.9% of HIV⁺ on HAART men vs 3.6% of HIV^–  had baseline GFR <60 (p <0.001). In logistic regression analyses factors associated with >3% GFR decline per year were baseline GFR >90 (OR = 3.5, p <0.001), white race (OR = 1.8, p <0.001), currently smoking (OR = 1.4, p = 0.028), and hypertension (OR = 1.4, p = 0.007). HIV stage (CD4, HIV RNA), HAART use (including specific antiretrovirals [ARV]), age, and diabetes were not associated with GFR declines. Men with GFR declines more often exhibited subsequent proteinuria (OR = 3.1, p <0.001). Median annual GFR change for men with proteinuria was larger (–1.3%/year) than for men without proteinuria (0.4%/year), p <0.001. Proteinuria existed in 226 (12.6%) of 1800 men with data:  161 (23.4%) of the HIV⁺ men on HAART, 6 (7.5%) of HIV⁺ men not on HAART, and 48 (5.3%) of HIV^– men; 27 (36.5%) of the baseline GFR <60 group, 106 (14%) of the baseline GFR 60 to 90 group, and 82 (9.8%) of the baseline GFR >90 group had proteinuria. Factors associated with proteinuria were:  GFR 60 to 90 vs >90, (OR 1.99, p = 0.007); GFR<60 vs >90 (OR 3.8, p = 0.003); ≥3% GFR decline (OR 3.37, p <0.001); being HIV⁺ on HAART vs HIV^– (OR 5.0, p <0.001); black race (OR 1.72, p = 0.05); diabetes mellitus (OR 2.38, p <0.001); hypertension (OR 2.4, p <0.001); smoking (OR 1.8, p = 0.05); AIDS (OR 2.18, p = 0.02); hepatitis C virus antibody (HCV Ab)⁺ (OR 3.13, p <0.001). No ARV was associated with proteinuria.

Conclusions:  GFR decline was associated with neither HIV serostatus nor HAART use, despite the fact that HIV⁺ HAART recipients more often had low baseline GFR than HIV^– men. However, proteinuria was:  5 times more common in HIV⁺ men on HAART than HIV^– men; strongly associated with GFR declines regardless of HIV serostatus; not associated with ARV use, including tenofovir; more common in men with classic risks for renal disease (e.g., AIDS, HCV). This suggests that proteinuria is a more sensitive sign of early renal disease in HIV⁺ persons than GFR declines and that non-HIV causes predominate.