BACKGROUND:It has been reported that people living with HIV in West Africa exhibited the highest risks for chronic kidney disease (CKD) in the world. Here, we aimed at determining the CKD frequency and changes in kidney function during antiretroviral treatment (ART) in a large cohort of HIV-patients followed in Burkina Faso.METHODS:We included ART-naive adults who initiated ART at the Day Care Unit of the Souro Sanou University Hospital between 01/01/2007 and 12/31/2016. We assessed the estimated glomerular filtration rate (eGFR) by serum creatinine using the Modification of Diet in Renal Disease (MDRD) equation. Following the K/DOQI recommendations, CKD was defined as eGFR < 60 ml/min/1.73m2 at two consecutive measurements at least 3 months apart. The factors associated with eGFR decline or CKD were identified by mixed linear regression and Cox regression, respectively.RESULTS:Three thousand, one hundred and thirty-eight patients (72% women) were followed for a median (IQR) of 4.5(2.2-6.9) years. At baseline, median eGFR (IQR) was 110.7(94.4-128.4) ml/min/1.73m2 and 93 (3%) patients exhibited eGFR < 60 ml/min/1.73m2. The lowest-performing progressions of eGFR during the first year of ART were observed in patients with 40-49 yr. age range (- 8.3[- 11.7;-5.0] ml/min/1.73m2, p < 0.001), age ≥ 50 yr. (- 6.2[- 10.7;-1.8] ml/min/1.73m2, p = 0.006) and high blood pressure (HBP) (- 28.4[- 46.9;-9.9] ml/min/1.73m2, p = 0.003) at ART initiation. Regarding the ART exposure in patients with normal baseline eGFR, zidovudine (AZT) with protease inhibitor (PI) (- 4.7[- 7.7;-1.6] ml/min/1.73m2, p = 0.002), tenofovir (TDF) + PI (- 13.1[- 17.4;-8.7] ml/min/1.73m2, p < 0.001), TDF without PI (- 3.2[- 5.0;-1.4] ml/min/1.73m2, p < 0.001), stavudine (d4T) + PI (- 8.5[- 14.6-2.4] ml/min/1.73m2, p = 0.006) and d4T without PI (- 5.0[- 7.6-2.4] ml/min/1.73m2, p < 0.001) were associated with poorer eGFR progression. The prevalence of CKD was 0.5% and the incidence was 1.9 [1.3; 2.7] cases/1000 person-years. The risk of CKD was higher in patients with HBP (4.3[1.8;9.9], p = 0.001), 40-49 yr. patients (4.2[1.6;11.2], p = 0.004), ≥50 yr. patients (4.5[1.5;14.1], p = 0.009) and patients exposed to abacavir (ABC) or didanosine (ddI) based ART (13.1[4.0;42.9], p < 0.001).CONCLUSIONS:Our findings do not confirm the high risk of CKD reported in previous studies of West Africans with HIV, but support the recommendations for early initiation of ART and close kidney function monitoring in patients with HBP or aged ≥40 yr.