Table 3.
Hazard risks for the emergence of hypertension and/or dysmetabolic syndrome events during first line ART from fitting a Cox-regression analysis using the counting process formulation of Andersen and Gill in HIV-1-infected patients with a tropism determination at baseline.
| Univariate analysis | Multivariate analysisa | Multivariate analysisb | ||||
| Independent predictors of hypertension and/or dysmetabolic syndrome events during first-line ART | Hazard risk (95% CI) | P valued | Hazard risk (95% CI) | P valued | Hazard risk (95% CI) | P valued |
| Sex (male vs. female) | 2.00 (1.01–3.96) | 0.047 | 2.00 (1.07–3.73) | 0.030 | 2.2 (1.2–4.03) | 0.010 |
| CDC C stage, N | 1.88 (1.11–3.16) | 0.018 | 1.41 (0.85–2.33) | 0.178 | 1.49 (0.88–2.51) | 0.134 |
| Age (per 1 year increase) | 1.04 (1.02–1.06) | 0.001 | 1.03 (1.01–1.06) | 0.003 | 1.03 (1.01–1.06) | 0.004 |
| Year of diagnosis | 1.03 (0.98–1.08) | 0.288 | ||||
| Risk factor | 0.141 | |||||
| Heterosexualc | 1 | |||||
| Homosexual | 0.75 (0.40–1.40) | |||||
| Drug user | 0.28 (0.07–1.08) | |||||
| Subtype B | 1.21 (0.66–2.22) | 0.547 | ||||
| Presence of virological blips during virological suppression | 0.82 (0.46–1.46) | 0.498 | ||||
| Co-infections | ||||||
| HBV | 2.38 (1.47–3.86) | <0.001 | 1.24 (0.77–2.02) | 0.375 | 1.30 (0.80–2.11) | 0.287 |
| HCV | 0.62 (0.16–2.44) | 0.498 | ||||
| Viral load at zenith point (per 1 log copies/ml more) | 1.27 (0.90–1.78) | 0.170 | ||||
| CD4+ at nadir cells/μl (per 50 cells increase) | 0.9965 (0.9944–0.9985) | 0.001 | 0.9990 (0.9968–1.001) | 0.394 | 0.9990 (0.9967–1.001) | 0.396 |
| CD4+ at comorbidities diagnosis cells/μl (per 50 cells increase) | 0.9987 (0.9976–0.9999) | 0.036 | 0.9998 (0.9986–1.001) | 0.788 | 0.9997 (0.9983–1.001) | 0.712 |
| ART length (per 1 year more) | 0.99 (0.85–1.15) | 0.882 | ||||
| Number of visits per year of ART | 1.00 (0.96–1.04) | 0.886 | ||||
| Drug exposure | 0.200 | |||||
| TDF + FTC + protease inhibitorc | 1 | |||||
| TDF + FTC + NNRTI | 0.44 (0.18–1.11) | |||||
| ABC + 3TC + protease inhibitor | 1.18 (0.52–2.68) | |||||
| Regimen including ETV or MVC or RAL | 1.36 (0.68–2.69) | |||||
| X4 tropism (FPR<10%) | 3.01 (1.85–4.91) | <0.001 | 2.29 (1.39–3.76) | 0.001 | ||
| Tropism prediction | <0.001 | <0.001 | ||||
| R5 FPR >60%c | 1 | 1 | ||||
| R5 FPR 20–60% | 2.30 (0.88–6.04) | 2.17 (0.90–5.24) | ||||
| R5 FPR 10–20% | 2.87 (0.98–8.37) | 2.87 (1.02–8.11) | ||||
| X4 FPR 5–10% | 4.76 (1.79–12.68) | 4.86 (2.00–11.82) | ||||
| X4 FPR <5% | 6.20 (2.46–15.64) | 3.58 (1.41–9.10) | ||||
3TC, lamivudine; ABC, abacavir; ART, antiretroviral therapy; CI, confidence interval; ETV, etravirine; FPR, false-positive rate; FTC, emtricitabine; HBV, hepatitis B virus; HCV, hepatitis C virus; MVC, maraviroc; NNRTI, non-nucleoside RT inhibitor; RAL, raltegravir; TDF, tenofovir. The analysis was performed on 116 patients. Two multivariate models were applied for tropism prediction according to FPR.
aIn the first model, FPR was set at 10% to define an X4 tropic virus.
bIn the second model, FPR was stratified according to the following five FPR percentage ranges: for X4 viruses not more than five, and 5–10; for R5 viruses: 10–20, 20–60, and above 60. All independent predictors characterized by a P value not more than 0.07 in univariate model were inserted in the Cox analysis. Boldface indicates variables significantly associated with for the emergence of severe non-AIDS-related events during first line ART (P < 0.05).
cDummy variable.
dType III for interaction.