Diabetes raises the risk of liver damage in people with HIV

Almost half of people with HIV with unexplained liver enzyme elevations or other abnormal liver markers had non-alcoholic steatohepatitis (NASH) and 39 of 116 had stage F3 or F4 fibrosis, and advanced liver damage was strongly associated with type 2 diabetes, a four-country study reports in Clinical Infectious Diseases.

Fatty liver disease occurs when fat builds up in the liver. Greater fat accumulation can lead to NASH, in which liver cells balloon and become inflamed. Unchecked, NASH leads to scarring of liver tissue (fibrosis) and eventually to cirrhosis, in which normal liver functions decline.

Although a high prevalence of non-alcoholic fatty liver disease (NAFLD) has been reported in people living with HIV, it’s unclear what proportion of these cases already have NASH or advanced fibrosis. The only way that these conditions can be diagnosed definitively is through liver biopsies (direct sampling of liver tissue).

To learn more about the frequency of NASH in people living with HIV, investigators at hospitals in the United Kingdom, Italy, Canada and the United States carried out a retrospective study of biopsy samples from 166 people with HIV who had been referred for investigation of unexplained liver enzyme increases or other abnormalities in laboratory markers of liver disease between 2001 and 2019.

The analysis excluded people with hepatitis B or C, any active cancer, alcohol consumption above 21 units a week for men or 14 units for women, or any other cause of chronic liver disease.

Study participants had a median age of 48 years, 93% were male, 72% were white and the median duration of antiretroviral treatment was nine years. The median CD4 cell count was 638 cells/mm³. None had taken older nucleoside analogues associated with steatosis (stavudine, didanosine). The median body mass index was 29 kg/m² (borderline obesity), 53% had high blood pressure and 25% had diabetes.

Biopsies showed that 63 of 116 people had NAFLD (54%) and 57 (49% of the entire cohort) had NASH. Thirty-six people (31%) had F3 stage fibrosis and three (2%) had F4 fibrosis (cirrhosis).

Multivariate analysis showed that after controlling for metabolic factors associated with NAFLD (model 1) or HIV-related factors (model 2), the only factor associated with NAFLD was higher body mass index (adjusted odds ratio 1.20 in both models, p = 0.001).

Advanced fibrosis (F3 or above) was the only factor associated with type 2 diabetes in multivariate analysis (aOR 3.42, 95% CI 1.00-11.71) and this association was on the borderline of statistical significance (p = 0.05).

Using biopsy results as a gold standard, investigators also assessed whether laboratory markers could accurately identify patients with advanced fibrosis. They found that both the FIB-4 and NAFLD Fibrosis scores performed poorly in identifying advanced fibrosis but showed good sensitivity (93%) in ruling out cases where the collagen proportionate area (CPA) was above 7.6%. CPA measures the percentage of liver tissue that is fibrotic and a level above 7.6% has been shown to predict long-term adverse liver disease outcomes.

The poor prognostic value of non-invasive markers for liver fibrosis in people living with HIV leads the investigators to recommend consideration of liver biopsy as a screening tool in people with HIV who are obese, especially those with type 2 diabetes.

However, they also note that 41% of people referred for biopsy did not have NAFLD. Their liver enzyme elevations remained unexplained but 15 of these 53 patients had advanced fibrosis (Ishak score 3 or above). The only factor associated with advanced fibrosis in those without NAFLD was time since HIV diagnosis (21 years vs 11.5 years, p = 0.005), even though age was similar between the two groups.