958 
Evolution of Non-alcoholic Fatty Liver Disease in HIV-infected Patients: Incidence, Characteristics, and Predictors
Giovanni Guaraldi*1, N Squillace1, C Stentarelli1, G Orlando1, R D'Amico1, G Ligabue1, F Fiocchi1, S Zona1, R Esposito1, and F Palella2
1Univ of Modena and Reggio Emilia, Modena, Italy and 2Feinberg Sch of Med, Northwestern Univ, Chicago, IL, US
Background: Non-alcoholic Fatty Liver Disease (NAFLD) occurs among persons
without chronic viral hepatitis or significant alcohol consumption. It presents
often in the context of cardiovascular risk factors. Our objectives were to
assess the incidence and predictors of NAFLD in HIV-infected HAART-experienced
patients.
Methods: This
was a 48-week prospective observational study. NAFLD
was defined using a liver/spleen attenuation value <1.1 by computed
tomography in persons without serum hepatitis B surface antigen (HBsAg),
hepatitis C virus (HCV) antibody, or a significant history of alcohol consumption.
Incidence was defined as the proportion of subjects meeting diagnostic criteria
at follow-up. Compared to baseline liver/spleen values, quartiles were used to classify
people whose liver steatosis (liver/spleen) improved (fourth quartile) or
worsened (first quartile) during follow-up. We used Student’s t tests and
univariate logistic regression models to identify liver/spleen risks and predictors
of improvement or worsening.
Results: We
enrolled 57 patients, 44 (77%) of whom were male.
Baseline mean age was 49 years; body mass index, 23.76±3.45; 27.8% were CDC
group C; mean duration of antiretroviral (ARV) exposure (in months) was: NRTI 117±41,
NNRTI 33±29; protease inhibitor (PI), 65±36 months; 88% had lipodystrophy, and
25% had metabolic syndrome. NAFLD was present in 26 patients (prevalence of 45.61%).
At follow-up, NAFLD was diagnosed in 25 patients (incidence of 43.86%). One-year
incidence of NAFLD was 4/57 (7%). Analysis of quartile distribution at 48 weeks
revealed 4 subjects (14%) with improved and 4 subjects (14%) with worsened liver/spleen.
Univariate logistic regression models revealed that improvement in liver/spleen
indicators was associated with: higher baseline serum HDL (68.67±22.3 in those who improved vs 41.47±14.29 in those who did not, p <0.001), higher baseline apolipoprotein
(Apo) A (178.67±46.2 vs 130.51±29.29, p = 0.01), less cumulative NRTI
exposure (80±41.7 vs 121.5±41.06, p = 0.05), greater baseline leg fat
mass (3501±1338 vs 2037±1268, p = 0.05).Worsening of liver/spleen
indicators was associated with: higher baseline serum triglyceride (TG, 426.25±529.7
vs 176.98±95.2, p <0.001); higher baseline total cholesterol
(270.25±71.28 vs 191.74±40.01, p = 0.04); higher baseline ApoB
(135.5±32.71 vs 107.28±27.22, p = 0.05); higher baseline serum insulin
(22.68±6.10 vs 14.03±7.76, p = 0.03); lower baseline spleen density
(35.22±16.64 vs 50.9±3.81, p <0.001); reduction in serum HDL (–9.5±9.04
vs –0.12±8.59, p = 0.04)
Conclusions: NAFLD is a highly prevalent in HIV-infected persons. Its development
is more likely in the presence of traditional metabolic cardiovascular risks and
with greater NRTI exposure.
|