Skip to main content
. 2017 Feb 28;20(1):21426. doi: 10.7448/IAS.20.1.21426

Table 4.

Determinants of transitioning to and from none/mild/moderate liver fibrosis (F0–F1–F2) and severe fibrosis/cirrhosis (F3–F4) during tenofovir-containing ART.

  Univariable
Multivariablea
Determinant F0–F1–F2 → F3–F4 F3–F4 → F0–F1–F2 F0–F1–F2 → F3–F4 F3–F4 → F0–F1–F2
Age at baseline        
 per year 1.08 (1.05–1.12) 0.96 (0.92–1.00)    
 >40 years 3.75 (2.09–6.72) 0.69 (0.36–1.29) 1.08 (1.04–1.12) 0.96 (0.92–1.01)
Female gender 0.18 (0.05–0.61) 1.94 (0.58–6.51) 0.16 (0.05–0.57) 1.43 (0.36–5.71)
Zone of high HBV-endemicity 0.24 (0.11–0.53) 0.80 (0.34–1.90)    
Cardiovascular disease 1.88 (1.12–3.18) 0.81 (0.46–1.44)    
Diabetes 5.30 (1.25–22.41) 1.28 (0.29–5.72)    
AIDS-defining illness 2.29 (1.36–3.85) 0.86 (0.49–1.51)    
CD4+ cell count        
 ≥500/mm3 at baseline 0.90 (0.54–1.48) 0.98 (0.58–1.68)    
 ≥350/mm3 at baseline 0.64 (0.39–1.05) 1.13 (0.65–1.95)    
 ≥500/mm3 during follow-up 0.69 (0.42–1.15) 1.11 (0.65–1.89)    
 ≥350/mm3 during follow-up 0.79 (0.46–1.35) 1.19 (0.65–2.16)    
Nadir CD4+ cell count (≥250/mm3) 0.63 (0.36–1.08) 1.01 (0.57–1.78)    
HIV-RNA (<50 copies/mL) 1.49 (0.79–2.81) 1.12 (0.53–2.35)    
Previous antiretroviral exposure        
 Zidovudine 2.38 (1.07–5.30) 0.72 (0.25–2.04)    
 Stavudine 1.40 (0.84–2.34) 0.61 (0.35–1.06)    
 Indinavir/r 1.73 (1.06–2.81) 1.17 (0.68–2.01)    
ART duration (per year) 1.12 (1.05–1.19) 1.01 (0.94–1.08) 1.07 (1.00–1.15) 1.01 (0.93–1.09)
PI-containing ART 2.43 (1.44–4.09) 1.22 (0.69–2.17) 2.41 (1.38–4.19) 1.21 (0.64–2.28)
ATZ exposure 3.59 (1.86–6.94) 1.25 (0.63–2.47)    
HBV-DNA viral load        
 per log10 IU/mL during follow-up 0.93 (0.80–1.08) 1.04 (0.89–1.22)    
 <60 IU/mL during follow-up 1.14 (0.68–1.92) 0.93 (0.51–1.68)    
HBeAg positive at baseline 1.34 (0.79–2.29) 1.43 (0.79–2.59)    
ALT >2 × ULN 1.92 (0.96–3.85) 1.74 (0.79–3.84)    

aIn the multivariable model, continuous age was preferred over age greater than forty years and PIs as a class was preferred over individual agents. In order to avoid overfitting, diabetes was not included. The following variables were removed from the model because their corresponding P value was no longer significant (<0.05): zone of high HBV-endemicity, cardiovascular disease, AIDS-defining illness and previous exposure to zidovudine or ritonavir-boosted indinavir.