Table 4. Baseline characteristics associated with change in log10-transformed urine albumin:creatinine ratio, Week 0 to Week 96.
| Variable | Univariate Analyses
|
Multivariate Analyses
|
||
|---|---|---|---|---|
| Estimated mean fold-change (95%CI) | p-value | Estimated mean fold-change (95%CI) | p-value | |
| Assigned to ABC/3TC (vs TDF/FTC) | 0.71 (0.54, 0.94) | 0.018 | 0.69 (0.54, 0.88) | 0.003 |
| Assigned to ATV/r (vs EFV) | 0.98 (0.74, 1.30) | 0.88 | 1.04 (0.81, 1.34) | 0.74 |
| Age (per 1 year) | 1.01 (1.00, 1.03) | 0.10 | ||
| Male | 0.87 (0.58, 1.29) | 0.48 | ||
| Race or ethnicity | ||||
| White, non-Hispanic | Reference | 0.37 | ||
| Black, non-Hispanic | 0.79 (0.57, 1.10) | |||
| Hispanic | 0.94 (0.63, 1.40) | |||
| Body mass index (per 1 kg/m2) | 1.01 (0.98, 1.04) | 0.56 | ||
| CD4 cell count (per 50 cells/μL) | 1.08 (1.03, 1.12) | 0.001 | 1.07 (1.03, 1.11) | 0.001 |
| HIV-RNA (per 1 log10 copies/mL) | 0.78 (0.64, 0.96) | 0.017 | ||
| HIV-RNA ≥ 100,000 copies/mL* | 0.67 (0.51, 0.90) | 0.007 | 0.72 (0.56, 0.93) | 0.013 |
| Hepatitis B or C co-infection | 0.76 (0.48, 1.20) | 0.24 | ||
| Systolic blood pressure (per 10 mmHg) | 0.96 (0.86, 1.07) | 0.46 | ||
| Diastolic blood pressure (per 10 mmHg) | 1.03 (0.88, 1.20) | 0.74 | ||
| HOMA-IR (per unit higher) | 0.93 (0.86, 1.00) | 0.064 | 0.92 (0.86, 0.98) | 0.014 |
| eGFR (per 10 mL/min/1.732) | 0.90 (0.83, 0.97) | 0.005 | 0.89 (0.83, 0.95) | 0.001 |
| Negative dipstick protein | 1.85 (1.38, 2.48) | <0.001 | 1.67 (1.26, 2.20) | <0.001 |
ABC/3TC, abacavir/ lamivudine; TDF/FTC, tenofovir disoproxil fumarate/emtricitabine; ATV/r, atazanavir/ritonavir; EFT, efavirenz. Hepatitis B co-infection, surface antigen positive; Hepatitis C co-infection, antibody positive; HOMA-IR, homeostasis model assessment-insulin resistance; eGFR, estimated glomerular filtration rate as calculated by Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) creatinine equation.
HIV-1 RNA was measured at screening. The estimated mean change in log10-transformed urine albumin:creatinine ratio was exponentiated to obtain the estimated mean fold change; values < 1 correspond to a decrease in mean urine albumin:creatinine ratio from baseline. Covariates with univariate p-value <0.20 were considered for inclusion in the multivariate model, and variables with a p-value <0.05 were retained; treatment assignment variables were retained regardless of p-value.