Abstract

Liver disease is very common in patients with HIV infection. 1 The association between chronic infection with hepatotropic viruses and being a target for other diseases, such as opportunistic infections and tumors, nonalcoholic fatty liver disease (NAFLD), and toxicity to drugs, including antiretroviral therapy (ART), results in liver disease being a leading cause of morbidity and mortality in these persons. 1
The aim of this study was to determine the prevalence of chronic hypertransaminasemia (CHT) in a cohort of patients with HIV infection but no coinfection and to examine the factors associated with CHT.
We undertook a cross-sectional, case-control study of all HIV-infected patients who regularly attended the infectious diseases office of our center from March to August 2009. A case was defined as any patient with glutamate pyruvate transaminase (GPT) values above normal (78 IU/L) at least 2 consecutive revisions (3 months or more between the 2 measurements). Patients were excluded if they had a positive serology to a hepatotropic virus or opportunistic events during the previous month. A control was defined as a patient attending the office paired for age (±5 years) and sex with a case and who had a normal GPT value, using the same exclusion criteria as for the cases. Hepatotoxicity of ART was defined as CHT due to ART, and to make this diagnosis, other causes of elevated GPT values were ruled out in all cases and NAFLD by the presence of steatosis in an abdominal ultrasound in patients consuming <50 g/d of alcohol. The statistical analysis was done with SPSS 17.0 (SPSS, Chicago, Illinois).
During the study period, 861 patients attended the office, of whom 454 (52.7%) had a negative serology for hepatotropic virus. Of these, 31 met the inclusion criteria, giving a prevalence of CHT of 6.8%. Most were men, the mean age was 45.7 years, and the predominant mode of transmission of the HIV infection was sexual. In all cases except 1, the elevation of the GPT was mild (<3 times the normal values). The only parameter that differed between cases and controls was a greater body mass index in the cases. Table 1 shows the characteristics of the cases and controls. There were no differences in the use of ART. At least 1 other drug was being taken by 61.2% of the 31 cases and 48.3% of the controls. Lipid-lowering drugs were being taken by more of the cases than the controls (15 versus 7, P < .05). An abdominal ultrasound was performed in 11 patients, 9 of whom showed signs of hepatic steatosis. No case had signs of portal hypertension. Liver stiffness could be assessed by elastometry (FibroScan) in 20. The mean stiffness was 5.9 Kps, with a range of 3.3 to 8.8 kPa. No patient had a liver biopsy. The most usual clinical diagnosis attributable to the CHT was ART toxicity, followed by NAFLD. Three patients were convalescing from hepatitis A virus (HAV) infection and 1 from secondary syphilis. Three patients required withdrawal of ART because of the CHT (Table 1).
Characteristics of the Cases and Controls
Abbreviations: ART, antiretroviral therapy; VL, viral load; CD4, CD4 lymphocytes; IDU, injecting drug users; HMX, homosexual; HTX, heterosexual; BMI, body mass index.
aIn 1 patient, treatment was withdrawn because of increased liver stiffness measured by elastometry.
bddI was withdrawn in both the cases.
The prevalence of CHT in our series was much lower than the previously reported 11% to 15% values. 2–4 The prevalence of CHT in a large Swiss cohort was 13.2%, and it was associated with alcohol consumption, NAFLD, and the use of ART. The same study found the incidence to be 3.9 cases per 100 person-years. 4 These differences are probably due to the methods used, the study population, and the differences at the time of the study. Of note, although, is the scarce clinical importance, with just 1 case of moderate CHT and few patients having to cease ART. On the other hand, although only just over one third of our patients had an ultrasound study, NAFLD can currently be considered one of the main causes of CHT in HIV-infected patients with no coinfection. 5,6 Hepatic steatosis is very usual in the general population, with very large differences between populations. It sometimes courses with CHT and can be associated with alcohol consumption, obesity, and metabolic disorders. 6 Nonalcoholic fatty liver disease is very common in HIV-infected patients. Its origin is multifactorial and a sizeable percentage of patients do not have CHT. 4,5 The situation, although, may progress to steatohepatitis and hepatic fibrosis. 7
Other causes of CHT in our series were sexually transmitted diseases, such as HAV and syphilis. The behavior of the former is usually benign, although normalization of transaminase levels is often slow, taking a mean of 170 days. 8 The situation with syphilis is similar; liver involvement is common in early and secondary syphilis and normalization of transaminase levels can also take some time. 9
In summary, CHT is uncommon in HIV-infected patients with no coinfection and its scarce clinical repercussion results in few patients having to cease ART. Nonetheless, the persistence of unexplained CHT is often associated with NAFLD and even fibrosis, thus necessitating diagnosis and treatment of modifiable factors due to the repercussions on long-term prognosis. 7
