| |
Lean NAFLD in HIV+ 14% - Prevalence, Predictors, and Severity of Lean Nonalcoholic Fatty Liver Disease in Patients Living With Human Immunodeficiency Virus
|
| |
| |
Download the PDF here
NAFLD is emerging as the leading cause of chronic liver disease in people living with HIV, with higher rates than in the general population....Clinicians practicing HIV medicine should maintain a high degree of suspicion for NAFLD in lean patients living with HIV mono-infection, especially in the presence of dyslipidemia or elevated ALT. Further studies are needed, with longitudinal designs and data on genetic variants, insulin markers, single ART regimens, and body fat distribution.....In a large population of patients living with HIV mono-infection recruited from 3 cohorts and undergoing a routine screening program for liver disease, we found an overall prevalence of lean NAFLD of 13.9%. Importantly, lean NAFLD patients were more metabolically abnormal and had higher proportions of elevated ALT and significant liver fibrosis than lean patients without NAFLD. Finally, our findings indicate that investigations for investigations or NAFLD should be considered in older patients living with HIV mono-infection with dyslipidemia and elevated ALT, even if they have a normal BMI
Importantly, we found that lean NAFLD patients are metabolically abnormal in 61.9% of cases and have elevated ALT in 36.7% of cases.
After adjusting for sex, ethnicity, hypertension, CD4 cell count, nadir CD4 <200μ/L, and time since HIV diagnosis, predictors of NAFLD in lean patients were age (adjusted OR [aOR], 1.29; 95% confidence interval [CI], 1.04-1.59), high triglycerides (aOR, 1.34; 95% CI, 1.11-1.63), and high ALT (aOR, 1.15; 95% CI, 1.05-1.26), while a high level of high-density lipoprotein cholesterol was protective (aOR, 0.45; 95% CI, .26-.77).
NAFLD in lean individuals has been initially considered a less severe form of liver disease than NAFLD in overweight/obese patients [34]. This concept has been recently challenged. A cross-sectional study of 466 NAFLD patients with biopsy data showed that the rate of liver cirrhosis was at 8.1% in lean patients [21]. The only longitudinal study conducted in 646 Caucasian patients with biopsy-proven NAFLD reported an increased risk for developing severe liver disease in lean patients (hazard ratio, 2.69), compared to overweight patients, during a mean follow-up of 19.9 years [20]. The pathogenesis of lean NAFLD is not completely understood. Lean NAFLD patients may have a distinct metabolism and an obesity-resistant profile, and may adapt better to an excess intake of calories [35]
To the best of our knowledge, there has been no study investigating lean NAFLD in patients living with HIV. This is particularly important considering that patients living with HIV mono-infection with NAFLD tend to have lower BMIs when compared to uninfected NAFLD patients [36]. Moreover, changes in anthropometric characteristics in patients living with HIV, such as a shift of body fat deposits from the subcutaneous to the visceral compartment, have been described [10]. Most of the studies conducted thus far reported BMI as the main predictor of NAFLD in HIV mono-infection, such that in clinical practice a high level of suspicion for NAFLD has been recommended in overweight patients [9, 17, 18, 25]. In the present study, we found that lean NAFLD is a frequent entity in patients living with HIV, affecting 24.2% of lean patients and representing 35.4% of all NAFLD patients. Our population was consecutively screened for NAFLD using TE with associated CAP, thus reducing the selection bias.
Moreover, lean NAFLD had significant liver fibrosis in 11.9% of cases and cirrhosis in 3.8% of cases. This result highlights that, also in the setting of HIV infection, lean NAFLD should not be considered a benign condition, as 1 in 6 patients actually have significant liver disease and 3.8% will require surveillance for hepatocellular carcinoma and esophageal varices. In a multivariable analysis, independent predictors of NAFLD among lean patients living with HIV mono-infection, besides older age, were higher triglycerides and lower HDL cholesterol, thus also underlying the relevance of metabolic factors for this clinical phenotype. High ALT also was as an independent predictor of NAFLD.
NAFLD should be considered in older patients living with HIV mono-infection with dyslipidemia and elevated ALT, even if they have a normal BMI.
NAFLD affects 1 in 4 lean patients living with HIV mono-infection.
After applying the inclusion and exclusion criteria, 1511 patients with HIV mono-infection were included (Figure 1). The distribution of BMI categories was as follows: 867 (57.4%) patients were lean, 473 (31.3%) were overweight, and 171 (11.3%) were obese. The prevalence of lean NAFLD in the whole study population was 13.9%. Table 1 reports the characteristics of the whole study population, as well as the univariable analysis of lean patients by NAFLD status. When compared to lean patients without NAFLD, lean patients with NAFLD were older and had higher BMIs. They had longer times since HIV diagnosis, higher CD4 cell counts, and were more likely to have an undetectable HIV viral load and nadir CD4 cell count <200/μL. Moreover, lean patients with NAFLD were more exposed to d-drugs and to integrase inhibitors. Lean patients with NAFLD were also more metabolically abnormal than lean patients without NAFLD (61.9% vs 48.9%, respectively; P < .001), while they showed similar degrees of metabolic abnormalities as overweight NAFLD patients (Figure 2A; Supplementary Table S1). Finally, although lean patients with NAFLD had a higher prevalence of elevated ALT than lean patients without NAFLD (36.7% vs 24.2%, respectively; P < .001), they exhibited a similar prevalence of elevated ALT as overweight patients (Figure 2B).
Liver Fibrosis in Lean Nonalcoholic Fatty Liver Disease
Figure 3A reports the distribution of NAFLD and significant liver fibrosis status by BMI category. The distribution of liver fibrosis stages in lean patients with and without NAFLD, as well as in obese and overweight NAFLD patients, is depicted in Figure 3B. Lean NAFLD patients had a higher prevalence of significant liver fibrosis/cirrhosis (15.7% vs 7.6%, respectively; P < .001), compared to lean patients without NAFLD. On the other side, the prevalence of significant liver fibrosis and cirrhosis was similar between lean and overweight NAFLD patients. Out of 157 NAFLD patients with available follow-ups, 15 (9.6%) patients were excluded from the analysis for having the outcome (cirrhosis) at baseline. Of the remaining 142 cases (see Supplementary Table S4 for main characteristics), 38 (26.8%) had fibrosis progression during a median follow-up of 26 months (interquartile range, 6-54), accounting for an incidence rate of 18.5 per 100 PY (95% CI, 13.4-25.4). There was no significant difference in fibrosis progression between lean versus overweight/obese patients (24.5 per 100 PY [95% CI, 11.0-54.5] vs 17.6 per 100 PY [95% CI, 12.5-24.9], respectively). The overall median absolute change in LSM at the end of follow-up was 0.8 kPa (interquartile range, .6-2.2). In 1 overweight case (0.7%) LSM improved during the follow-up from the F2 to F0-F1 fibrosis category, while in 4 cases (2.8%), of which 2 patients were lean and 2 were overweight, LSM improved from F4 to F2. A reduction in BMI and/or CAP was observed in 80% of patients with improved liver fibrosis (data not shown).
----------------------
Prevalence, Predictors, and Severity of Lean Nonalcoholic Fatty Liver Disease in Patients Living With Human Immunodeficiency Virus
Clinical Infectious Diseases 13 April 2020
Adriana Cervo,1,2 Jovana Milic,3,4 Giovanni Mazzola,2 Filippo Schepis,5 Salvatore Petta,6 Thomas Krahn,1 Bertrand Lebouche,1,7 Marc Deschenes,8 Antonio Cascio,2 Giovanni Guaraldi,3 and Giada Sebastiani1,8,
1Chronic Viral Illness Service, McGill University Health Centre, Montreal, Canada, 2Infectious Diseases Unit, Department of Health Promotion Sciences and Mother and Child Care “Giuseppe D’Alessandro” (PROMISE), University of Palermo, Palermo, Italy, 3Infectious Diseases Clinic, University of Modena and Reggio Emilia, University of Modena and Reggio Emilia, Modena, Italy, 4Clinical and Experimental Medicine PhD Program, University of Modena and Reggio Emilia, Modena, Italy, 5Hepatology Unit, University of Modena and Reggio Emilia, Modena, Italy, 6Gastroenterology and Hepatology Service, Department of Health Promotion Sciences and Mother and Child Care “Giuseppe D’Alessandro” (PROMISE), University of Palermo, Palermo, Italy, 7Department of Family Medicine, McGill University, Montreal, Canada, and 8Division of Gastroenterology and Hepatology, McGill University Health Centre, Montreal, Canada
Abstract
Background
The burden of nonalcoholic fatty liver disease (NAFLD) is growing in people living with human immunodeficiency virus (HIV). NAFLD is associated with obesity; however, it can occur in normoweight (lean) patients. We aimed to investigate lean NAFLD in patients living with HIV.
Methods
We included patients living with HIV mono-infection from 3 prospective cohorts. NAFLD was diagnosed by transient elastography (TE) and defined as controlled attenuation parameter ≥248 dB/m, in absence of alcohol abuse. Lean NAFLD was defined when a body mass index was <25 kg/m2. Significant liver fibrosis was defined as TE ≥7.1 kPa. The presence of diabetes, hypertension, or hyperlipidemia defined metabolically abnormal patients.
Results
We included 1511 patients, of whom 57.4% were lean. The prevalence of lean NAFLD patients in the whole cohort was 13.9%. NAFLD affected 24.2% of lean patients.
The proportions of lean NAFLD patients who were metabolically abnormal or had elevated alanine aminotransferase (ALT) were higher than among those who were lean patients without NAFLD (61.9% vs 48.9% and 36.7% vs 24.2%, respectively).
Lean NAFLD patients had a higher prevalence of significant liver fibrosis than lean patients without NAFLD (15.7% vs 7.6%, respectively).
After adjusting for sex, ethnicity, hypertension, CD4 cell count, nadir CD4 <200μ/L, and time since HIV diagnosis, predictors of NAFLD in lean patients were age (adjusted OR [aOR], 1.29; 95% confidence interval [CI], 1.04-1.59), high triglycerides (aOR, 1.34; 95% CI, 1.11-1.63), and high ALT (aOR, 1.15; 95% CI, 1.05-1.26), while a high level of high-density lipoprotein cholesterol was protective (aOR, 0.45; 95% CI, .26-.77).
Conclusions
NAFLD affects 1 in 4 lean patients living with HIV mono-infection.
Investigations for NAFLD should be proposed in older patients with dyslipidemia and elevated ALT, even if normoweight.
Nonalcoholic fatty liver disease (NAFLD) represents a worldwide epidemic, with a global prevalence estimated at 25.24% [1]. NAFLD is a liver fat accumulation exceeding 5% of hepatocytes in the absence of other causes of liver disease. NAFLD could lead to nonalcoholic steatohepatitis (NASH), liver fibrosis accumulation leading to cirrhosis and end-stage liver complications [2]. With the successful implementation of direct antiviral agents for the treatment of hepatitis C virus (HCV) and the drastic reduction of AIDS-related mortality in the post-antiretroviral therapy (ART) era [3], NAFLD is now emerging as the most frequent liver disease in patients living with HIV [4].
people living with HIV, the prevalence of NAFLD ranges from 13 to 65% in the absence of viral hepatitis coinfection and alcohol abuse [5-9]. Patients living with HIV are at higher risk of NAFLD than the general population as a result of multiple cofactors, including lifelong use of ART, especially past exposure to hepatotoxic d-drugs (stavudine and didanosine); persistent immune activation and HIV-related inflammation; and extremely prevalent dysmetabolic conditions [10, 11]. Moreover, NASH and significant liver fibrosis are at least twice as frequent in patients living with HIV mono-infection than in the general population [12-16].
NAFLD is closely associated with the features of metabolic syndrome and obesity. A strong link between NAFLD and body mass index (BMI) has also been reported in patients living with HIV [6, 9, 17, 18]. However, patients who are not obese can also present with NAFLD. NAFLD is known as lean NAFLD in patients with a BMI <25 Kg/m2. The prevalence of lean NAFLD ranges between 7% in the United States to 20% in Asia [19, 20]. As lean patients lack obvious excess visceral adipose tissue, differences in NAFLD mechanisms may exist, such as an impaired glucose metabolism, dysfunctional adipose tissue, and genetic factors [21]. Of note, a few studies reported a severe histological phenotype in lean patients with NAFLD, compared overweight/obese patients with NAFLD [20, 21].
The aim of this study was to determine the prevalence and predictors of NAFLD in lean patients living with HIV mono-infection who were enrolled in 3 large cohorts, by means of transient elastography (TE) with associated controlled attenuation parameter (CAP), a noninvasive test validated in HIV infection [9, 22, 23]. Secondary aims included determining the severity of NAFLD in lean patients, defined by liver fibrosis status.
RESULTS
After applying the inclusion and exclusion criteria, 1511 patients with HIV mono-infection were included (Figure 1). The distribution of BMI categories was as follows: 867 (57.4%) patients were lean, 473 (31.3%) were overweight, and 171 (11.3%) were obese. The prevalence of lean NAFLD in the whole study population was 13.9%. Table 1 reports the characteristics of the whole study population, as well as the univariable analysis of lean patients by NAFLD status. When compared to lean patients without NAFLD, lean patients with NAFLD were older and had higher BMIs. They had longer times since HIV diagnosis, higher CD4 cell counts, and were more likely to have an undetectable HIV viral load and nadir CD4 cell count <200/μL. Moreover, lean patients with NAFLD were more exposed to d-drugs and to integrase inhibitors. Lean patients with NAFLD were also more metabolically abnormal than lean patients without NAFLD (61.9% vs 48.9%, respectively; P < .001), while they showed similar degrees of metabolic abnormalities as overweight NAFLD patients (Figure 2A; Supplementary Table S1). Finally, although lean patients with NAFLD had a higher prevalence of elevated ALT than lean patients without NAFLD (36.7% vs 24.2%, respectively; P < .001), they exhibited a similar prevalence of elevated ALT as overweight patients (Figure 2B).
Predictors of Lean Nonalcoholic Fatty Liver Disease
After adjustments, independent predictors of NAFLD among lean patients were older age (aOR, 1.29; 95% CI, 1.04-1.59; P = .020), higher triglycerides (aOR, 1.34; 95% CI, 1.11-1.63; P = .002), and higher ALT (aOR, 1.15; 95% CI, 1.05-1.26; P = .002), while higher HDL cholesterol was protective (aOR, 0.45; 95% CI, .26-.77; P = .004; Table 2). We also conducted a sensitivity analysis by building a multivariable model including integrase inhibitors, with similar results as the previously mentioned model (Supplementary Table S2). When using the 288 dB/m CAP cut-off to diagnose NAFLD, the prevalence of lean NAFLD was 4.4%. Supplementary Table S3 depicts the multivariable model for predictors of lean NAFLD when using the 288 dB/m CAP cut-off.
Liver Fibrosis in Lean Nonalcoholic Fatty Liver Disease
Figure 3A reports the distribution of NAFLD and significant liver fibrosis status by BMI category. The distribution of liver fibrosis stages in lean patients with and without NAFLD, as well as in obese and overweight NAFLD patients, is depicted in Figure 3B. Lean NAFLD patients had a higher prevalence of significant liver fibrosis/cirrhosis (15.7% vs 7.6%, respectively; P < .001), compared to lean patients without NAFLD. On the other side, the prevalence of significant liver fibrosis and cirrhosis was similar between lean and overweight NAFLD patients. Out of 157 NAFLD patients with available follow-ups, 15 (9.6%) patients were excluded from the analysis for having the outcome (cirrhosis) at baseline. Of the remaining 142 cases (see Supplementary Table S4 for main characteristics), 38 (26.8%) had fibrosis progression during a median follow-up of 26 months (interquartile range, 6-54), accounting for an incidence rate of 18.5 per 100 PY (95% CI, 13.4-25.4). There was no significant difference in fibrosis progression between lean versus overweight/obese patients (24.5 per 100 PY [95% CI, 11.0-54.5] vs 17.6 per 100 PY [95% CI, 12.5-24.9], respectively). The overall median absolute change in LSM at the end of follow-up was 0.8 kPa (interquartile range, .6-2.2). In 1 overweight case (0.7%) LSM improved during the follow-up from the F2 to F0-F1 fibrosis category, while in 4 cases (2.8%), of which 2 patients were lean and 2 were overweight, LSM improved from F4 to F2. A reduction in BMI and/or CAP was observed in 80% of patients with improved liver fibrosis (data not shown).
|
|
| |
| |
|
|
|
