Starting antiretroviral therapy (ART) with a ritonavir (Norvir)-boosted protease inhibitor (PI) resulted in a higher incidence of chronic kidney disease (CKD), whether or not the regimen included tenofovir disoproxil fumarate (TDF, Viread). But first-line therapy had no impact on CKD incidence in HIV patients classified as low-risk by the D:A:D CKD risk score, according to the study.
In the current antiretroviral treatment era, CKD has emerged as a concerning cause of morbidity in people with HIV infection. Regimens combining a ritonavir-boosted PI and TDF have been tied to greater declines in estimated glomerular filtration rate (eGFR) than nonnucleoside-containing regimens. To help identify HIV patients at increased risk of CKD, a multicenter French team conducted this retrospective analysis of prospectively collected data.
The study involved antiretroviral-naive people starting ART at one of 12 French centers after January 1, 2004. All participants had eGFR determined in the six months before starting ART and at least twice after starting. Pre-ART eGFR had to be above 60 mL/min. The researchers analyzed prospectively collected data to determine first occurrence of CKD, defined as eGFR less than 60 mL/min in two measures at least three months apart. For each participant, the investigators calculated D:A:D CKD risk score before ART began.
In participants in whom CKD developed during treatment with their first antiretroviral regimen, the French team used Poisson regression models to quantify the relationship between CDK, first ART regimen and D:A:D score. They considered only four mutually exclusive first-line regimens: (1) ritonavir-boosted PI with TDF, (2) ritonavir-boosted PI without TDF, (3) nonnucleoside with TDF and (4) nonnucleoside without TDF.
The study involved 6301 people with HIV. Median age stood at 39 years (interquartile range [IQR] 32 to 47), median baseline CD4 count at 289 cells/mm3 (IQR 175 to 392), median viral load at 4.8 log10 copies/mL (IQR 4.2 to 5.3) and median eGFR at 101 mL/min (IQR 86 to 118). During a median 3.1 years of follow-up, CKD developed in 211 people for an incidence of 9.6 per 1000 person-years. People in whom CKD developed were older (median 52 versus 39 years), had a lower pre-ART eGFR (75 versus 102 mL/min) and had a lower pre-ART CD4 count (211 versus 292 cells/mm3) (P < .001 for all comparisons).
Median D:A:D score measured 5 (IQR 2 to 8) in patients with incident CKD and -2 (IQR -6 to 3) in those without CKD. The D:A:D score indicated that 60% of participants had low CKD risk and accounted for 10% of incident cases, 27% had medium CKD risk and accounted for 32% of cases and 13% had high CKD risk and accounted for 58% of cases. Five-year risk of CKD was 0.65% in the D:A:D low-risk group, 4.6% in the medium-risk group and 15.9% in the high-risk group.
CKD incidence proved highest in patients starting a PI with TDF (9.0 per 1000 person-years), followed by those starting a PI without TDF (7.1), those starting a nonnucleoside without TDF (3.9) and those starting a nonnucleoside with TDF (2.9). Notably, though, patients with D:A:D-determined low CKD risk had similar CKD incidence regardless of which type of ART regimen they began.
The authors recommend calculating the D:A:D CKD scores for patients starting ART and avoiding certain antiretroviral regimens in those at high risk. But for low-risk patients, they suggest, "classic regimens may be safely prescribed, with an economic benefit due to soon available generic formulations."