More attention should be paid to modifiable risk factors for chronic liver disease in people with HIV, such as alcohol, metabolic disorders and being overweight, HIV and hepatology specialists in Brighton, England report in the Journal of Acquired Immune Deficiency Syndromes.
In a study of people with HIV who had abnormally high liver enzyme levels, they found that one in five had clinically significant liver damage and 60% had risk factors that could be modified by treatment of metabolic complications and/or abstinence from alcohol.
They also found that past treatment with a first-generation antiretroviral drug, didanosine (ddI), increased the risk of both clinically significant fibrosis and moderate-to-severe fat accumulation in the liver.
Until recently, the most frequent cause of chronic liver disease in people with HIV was viral hepatitis, especially hepatitis C. But since direct-acting antivirals became widely available in 2015, systematic efforts to diagnose and cure hepatitis C in people with HIV have almost eliminated hepatitis C as a cause of chronic liver disease in people with HIV in the UK.
In the past decade, non-alcoholic fatty liver disease (NAFLD) and alcoholic liver disease have gradually overtaken viral hepatitis as the chief causes of chronic liver disease in people with HIV. However, studies have produced varying estimates of the prevalence and causes of chronic liver disease in people with HIV. Understanding the causes of chronic liver disease in people with HIV could lead to better advice on the prevention of liver disease, as well as greater awareness of the need to monitor and address its underlying causes.
Professors Sumita Verma and Yvonne Gilleece of University Hospitals Sussex NHS Foundation Trust designed a study to investigate the prevalence of chronic liver disease and its causes in people receiving care at the Lawson Unit in Brighton, one of the UK’s largest HIV clinics.
The study recruited patients who had elevated liver enzymes (ALT 1 x above the upper limit of normal) on two separate occasions at least six months apart. People were not eligible for inclusion on the basis of a single liver enzyme elevation, as isolated above-normal readings have many causes other than chronic liver disease.
One in five people receiving care at the Lawson Unit had new episodes of ALT elevation meeting the study criteria that were not explained by viral hepatitis and were not already under investigation. Of the 429 people who were eligible to join the study, just under two-thirds were recruited (274 people) and offered additional screening for liver disease.
Study participants underwent blood tests, a liver stiffness test and an Alcohol Use Disorders Identification questionnaire. A higher score on a liver stiffness test shows that the liver has a greater volume of scar tissue that cannot be repaired through the liver’s normal processes of regeneration. When advanced fibrosis is detected, the liver is seriously damaged, and its normal functions have begun to decline.
The study cohort had a median age of 52 years, 92% were male and had been living with HIV for a median of 15 years. Twenty-two percent had been diagnosed with HIV before the introduction of highly active antiretroviral treatment in 1996. Almost all had suppressed viral load (96%).
Some metabolic disorders were common; 70% had dyslipidaemia, 64% were classified as overweight (body mass index of 25 or above), 20% were classified as obese (BMI > 30), 14% had diabetes and 26% had metabolic syndrome (three or more of BMI >25, hypertension, insulin resistance, elevated triglycerides > 1.7mml/L or HDL cholesterol < 1 mmol/L).
Thirty-five percent had a weekly alcohol intake above the recommended maximum and 35% were defined as having hazardous drinking levels based on their answers to the alcohol use disorder questionnaire (see the questionnaire here – a score of 8 or above is classified as hazardous drinking).
The median peak ALT level was 59 IU/L (mildly elevated) and 20% had an ALT level at least twice the upper limit of normal.
Clinically significant liver fibrosis (a liver stiffness of 7.1kPa or above) was detected in 20% of the cohort and 19 people had previously undiagnosed cirrhosis. Hepatic steatosis (fat accumulation in the liver) was present in 70% of the cohort; just over half of the entire cohort (52%) had moderate-to-severe steatosis. Non-alcoholic fatty liver disease can lead to liver fibrosis; indeed, in people with HIV and NAFLD, a Canadian study found that the progression of liver fibrosis was approximately 40% faster than in people with HIV and hepatitis C.
Half of those with clinically significant fibrosis had multiple risk factors; alcohol use was a risk factor in 44% of cases, metabolic syndrome in 46% and a history of treatment with at least one antiretroviral known to cause steatosis or liver enzyme elevations in 56% of cases. (These antiretrovirals are didanosine (ddI), stavudine (d4T), nevirapine and long-term use of efavirenz). Sixty percent had at least one modifiable risk factor, either alcohol or metabolic syndrome.
Scoring systems for predicting the presence of liver fibrosis based on laboratory measurements (APRI and FIB-4) performed poorly in this cohort when compared to liver stiffness measurements, each failing to identify clinically significant fibrosis in the majority of cases.
Multivariate analysis showed three factors predicted clinically significant fibrosis: lower HDL cholesterol, raised AST level and past treatment with didanosine (an old nucleoside reverse transcriptase inhibitor). Two factors predicted hepatic steatosis: body mass index and a past history of treatment with didanosine.
The study authors recommend that everyone with a history of treatment with didanosine should be screened for chronic liver disease. In this study cohort, 17% had taken didanosine. Although the use of didanosine has been discouraged in first-line treatment in the UK since the early 2000s, it continued to be used in second- and third-line treatment regimens after this point as well as in treatment regimens for children. In 2012, it was still being used in 20 countries that received Global Fund support to buy antiretroviral drugs.
Although alcohol use did not emerge as a risk factor in the multivariate analyses, the study authors highlight the high prevalence of problematic alcohol use in the study cohort, noting that it was three times higher than the prevalence identified in a 2017 national audit of HIV services in the UK.
“There is an urgent need for increased focus on lifestyle intervention and clinical trials in PLWH with NAFLD,” the study authors conclude. They say there needs to be increased recognition among physicians of risk factors for liver disease among people with HIV so that they can be referred to hepatologists and offered lifestyle interventions that can reduce the risk of chronic liver disease progression.
Katarey D et al. Nonviral liver disease burden in people living with HIV and levated transaminases: a cross-sectional study. Journal of Acquired Immune Deficiency Syndromes, published online, 12 October 2023.