Predictors of renal insufficiency in patients randomized to second-line lopinavir/ritonavir monotherapy vs. tenofovir/lamivudine/lopinavir/ritonavir
Background: Tenofovir (TDF) use is associated with renal insufficiency; however, its incidence in Asians is incompletely understood. Here we compared the estimated glomerular filtration rate (eGFR) in patients randomized to TDF vs. non TDF second-line regimens. Additionally, we compared eGFR using the recently validated re-expressed Modification of Diet in Renal Disease (MDRD) formula with Thai racial correction factor (MDRD-Th) with the standard MDRD formula.
Methods: HIV-infected adults with HIV-RNA≥1,000 copies/mL failing first-line non-nucleoside reverse transcriptase inhibitor based regimens, and naïve to protease inhibitors were randomized to lopinavir/ritonavir monotherapy (mLPV/r) vs. standard dose of TDF+lamivudine+LPV/r (TDF/3TC/LPV/r).
Serum creatinine was performed at baseline and every 24 weeks until week 120; subjects with baseline eGFR >60 mL/min/1.73m2 and ≥2 subsequent creatinine results were included. Time to event methods were used to compare incidence and predictors of renal insufficiency (RI) defined as eGFR < 60 mL/min/1.73m2.
Results: Of 188 patients, 95 on mLPV/r,and 93 on TDF/3TC/LPV/r; 53% female,median (IQR) baseline: age 37(33-41) years, body mass index 21.7 (19.4-23.8) kg/m2, hemoglobin 13.2 (12.3-14.5) g/dL, CD4 count 190(107-287) cells/mm3, and HIV-RNA 4.1(3.6-4.5) log10copies/mL. Baseline eGFR by MDRD-Th was 98 (86-113) and by MDRD was 83(75-98) mL/min/1.73m2, respectively and did not differ by randomized arm using either formula. Incidence of RI per 100 person-years (95%CI) in mLPV/r vs. TDF/3TC/LPV/r by MDRD-Th was 1.4(0.4-5.7) vs. 4.5(2.0-10.0); p=0.14 and by MDRD was 4.4(2.0-9.7) vs. 13.3(8.2-21.3); p=0.01, respectively.
In multivariate analysis, baseline eGFR< 90 mL/min/1.73m2 (HR12.1; 95%CI 2.0-72.2), CD4< 100cell/mm3 (HR9.8; 95%CI 2.1-46.6), and hemoglobin < 11.0 mg/dL (HR7.9; 95%CI 1.4-46.1) were associated with RI by MDRD-Th. However, using MDRD, age >45 years (HR6.7; 95%CI 2.8-15.8), baseline eGFR< 90 mL/min/1.73m2(HR 3.8; 95%CI 1.3-11.0), TDF/3TC/LPV/r-arm (HR3.2; 95%CI 1.3-8.1), and current use cotrimoxazole (HR 2.6; 95%CI 1.1-5.9) were associated with RI.
Conclusion: Using the most accurate eGFR equation for Thai patients, RI incidence between treatment arms did not differ; but with the non-racial adjusted formula, RI incidence was significantly higher in subjects treated with TDF/3TC/LPV/r. Different multivariate models defining RI predictors were developed using each equation. This emphasizes the importance of using eGFR equations with appropriate racial factor adjustment.
J. Sophonphan1, P. Chetchotisakd2, W. Munsakul3, S. Jirajariyavej4, P. Kantipong5, W. Prasithsirikul6, C. Bowonwatanuwong7, S. Sungkanuparph8, V. Klinbuayaem9, B. Hirschel10, K. Ruxrungtham1,11, J. Ananworanich1,11,12, S.J. Kerr1,13, T. Bunupuradah1, HIV STAR Study Group
1HIV-NAT, the Thai Red Cross AIDS Research Centre, Bangkok, Thailand, 2Khon Kaen University, Khon Kaen, Thailand, 3Vajira Hospital, University of Bangkok Metropolitan Administration, Faculty of Medicine, Bangkok, Thailand, 4Taksin Hospital, Bangkok, Thailand, 5Chiangrai Prachanukroh Hospital, Chiangrai, Thailand, 6Bamrasnaradura Infectious Diseases Institute, Nonthaburi, Thailand, 7Chonburi Hospital, Chonburi, Thailand, 8Ramathibodi Hospital, Mahidol University, Faculty of Medicine, Bangkok, Thailand, 9Sanpatong Hospital, Chiang Mai, Thailand, 10Geneva University, Geneva, Switzerland, 11Chulalongkorn University, Department of Medicine, Faculty of Medicine, Bangkok, Thailand, 12SEARCH, Thai Red Cross AIDS Research Center, Bangkok, Thailand, 13The Kirby Institute, University of New South Wales, Sydney, Australia