HIV, Elevated Transaminases, Fatty Liver: The Perfect Storm?

To the Editors: We read with great interest the detailed article from Prat et al1 about the etiologic and histologic characteristics of liver disease in 97 HIV-monoinfected patients with abnormal liver transaminases. The researchers also evaluated the diagnostic accuracy of noninvasive diagnostic tools for liver fibrosis, namely transient elastography measured by fibroscan and the biomarker FIB-4, as compared to liver histology. Elevated liver transaminases are particularly common in HIV-infected patients, occurring in 20%–93% of patients on antiretroviral therapy (ART), even in the absence of viral hepatitis coinfection.2,3 After the widespread use of ART, people infected with HIV live longer.4 In this newly aging HIV-infected population, liver disease has now become a leading cause of death.4,5 Although coinfection with hepatitis C and hepatitis B viruses is believed to have driven this trend, increasing rates of nonalcoholic fatty liver disease (NAFLD) may also have contributed. The prevalence of NAFLD in HIV-monoinfected patients ranges from 13% to 65% across studies, consistently suggesting an increased frequency of the disease in HIV-monoinfected patients as compared to the uninfected population.6 Recent evidence shows that NAFLD is not only more frequent in the setting of HIV infection, but also more severe. Nonalcoholic steatohepatitis (NASH), the evolutive counterpart of NAFLD, has been reported in up to 65% of HIV-monoinfected patients with chronically elevated alanine aminotransferase (ALT) and in 10% of those attending a routine screening program.7–9 Overall, very few studies reported on the histologic characteristics of HIV-monoinfected patients with elevated liver transaminases, particularly in relation to specific features of NAFLD/NASH. In this study, Prat and colleagues reported that only 28.7% of HIV-monoinfected patients with chronic elevated liver transaminases had NAFLD on histology. This figure is much lower than previous studies7–13 (Table 1). Surprisingly, the researchers also found that a high number of patients had nonsignificant changes (26%) or even normal histology (13%), despite the elevation of liver transaminases. We wish to comment on the reported data.TABLE 1.: Prevalence of NAFLD/NASH and Significant Liver Fibrosis in HIV-Monoinfected Patients in Studies With Available Liver HistologyFirst, the findings by Prat et al may reflect specific characteristics of the included patient population. In this study, HIV-monoinfected patients had both shorter duration of HIV infection and lower exposure to ART, as compared to previous reports. The pathogenesis of NAFLD/NASH and liver fibrosis in the context of HIV infection is still not completely understood, but it is likely more complex than in HIV-uninfected patients. Besides the classical metabolic risk factors such as diabetes, dyslipidemia, and hypertension, which are more frequent than in the general population, HIV-infected patients present with additional unique risk factors affecting the pathogenesis of NAFLD/NASH and liver fibrosis, notably HIV itself and chronic exposure to ART.14 Numerous studies demonstrated direct HIV infection of liver cells. HIV has a direct cytopathic effect on hepatocytes, primarily triggering apoptosis through the HIV gp120 protein-receptor signaling pathway.15,16 Kupffer cells, differentiated tissue macrophages that reside in the liver, can be infected by HIV in vivo. HIV directly activates hepatic stellate cells through the gp120 receptor, resulting in oxidative stress and liver fibrosis.15 HIV may also cause immune-mediated injury by altering the functions of the hepatic stellate cells and the Kupffer cells.15 Hepatotoxicity is one of the most common side effects associated with chronic use of ART. Nucleoside reverse transcriptase inhibitors, particularly zidovudine, stavudine, and didanosine, can induce mitochondrial damage leading to impaired fatty acid oxidation responsible for microvesicular steatosis and lactic acidosis. Notably, this toxicity never completely recovers.17 Ritonavir-boosted protease inhibitors are also commonly associated with chronic elevated liver transaminases, possibly through their association with metabolic complications, such as hypertriglyceridemia, insulin resistance, and hepatic steatosis.18 In view of these data, we feel that the patient population included in the study by Prat et al may have been less exposed to the hepatocellular effects of HIV infection and to ART-induced chronic hepatotoxicity. Consistently with this hypothesis, liver biopsies included in this study belong to a wide temporal range, from 2000 to 2017. This implies that some of the included patients may have had very little exposure to ART and to HIV. It would have been interesting to know whether patients with nonsignificant changes on histology or normal liver biopsy were the ones with shorter duration of HIV infection and/or exposure to ART at the time of liver biopsy. Along the same lines, the study population included by Prat et al had a higher proportion of alcohol abuse compared with previous studies, which may result in aspecific elevation of liver transaminases before liver biopsy. Second, several studies found an association between elevated transaminases and liver fibrosis or NAFLD/NASH.8,9,13,19–21 Although few of them had availability of liver biopsy data, different diagnostic tests were used, leading to similar conclusions. Of note, the definition of elevated liver transaminases varied across studies. Importantly, recent guidelines from the European AIDS Clinical Society (EACS)22 recommend a diagnostic algorithm to assess and monitor disease severity in case of suspected NAFLD and metabolic risk factors.22 This stepwise algorithm contemplates further investigations and specialist referral in case of either elevated liver transaminases or presence of NAFLD with significant liver fibrosis diagnosed by serum fibrosis markers such as FIB-4. A similar pathway has been recommended for HIV-uninfected patients by the European Association for the Study of the Liver.23 Although we fully agree with Prat et al that chronically elevated liver transaminases should not be approached directly with the invasive liver biopsy, we feel important to underline the prompt need for further investigations of elevated transaminases in HIV-monoinfected patients, as per recent EACS guidelines. The wide availability of liver transaminases makes them an important tool in the busy setting of clinics practicing HIV medicine or in low- and middle-income countries, for both optimization of health care resources and risk stratification in the individual patient. In this sense, the study by Prat et al is among the few reporting on accuracy of noninvasive tools against liver histology, confirming an excellent negative predictive value of FIB-4 at 95%. Third, the authors did not provide information on the length of liver biopsy specimens of the included patients nor commented on how excluded liver biopsies were deemed of inadequate quality. It would have been interesting to know whether those patients with less-severe liver disease were the ones with smaller histologic samples as it is well known that a suboptimal liver biopsy specimen could underestimate liver fibrosis stage.24,25 Similarly, information on the liver biopsy procedure were not available, whereas both the technique and the needle adopted for the procedure may impact on the final histologic diagnosis.25 We conclude that NAFLD/NASH and associated liver fibrosis are consistently emerging as major comorbidities in HIV-monoinfected patients. The multiple noxae and complex interplay among HIV itself, metabolic factors, and ART-related hepatotoxicity in inducing NAFLD/NASH and liver fibrosis still need to be elucidated but are likely to create a “perfect storm” for liver injury. This is particularly true in case of elevated transaminases, where the prevalence of NAFLD and significant liver fibrosis in histologic studies can be as high as 72.6% and 58.9%, respectively. The characteristics of the study population are likely to play a major role in dictating these figures. Besides HIV-related features, sex and ethnicity are likely to contribute to NAFLD prevalence and severity, as is the case in uninfected patients. Chronic elevation of liver transaminases is an important indicator that further investigations are needed, as recommended by recent EACS guidelines. Histological studies, like the one by Prat and colleagues, are important for further understanding of liver injury in the specific setting of HIV. Larger scale, prospective studies aimed to investigate NAFLD/NASH, its effect on liver fibrosis, and the complex interplay among risk factors, where interventions can be targeted, are warranted.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.