De Novo nonalcoholic fatty liver disease after liver transplantation

Achieving long-term graft survival and mortality is the primary goal of liver transplantation (LT). However, disease recurrence remains a stumbling block, especially among patients with hepatitis C. In such patients, worse outcomes may be associated with comorbidities (i.e., hyperlipidemia, hypertension, impaired glucose tolerance, type 2 diabetes, and obesity) and immunosuppressive medications that contribute to the development of de novo nonalcoholic fatty liver disease (NAFLD). Although benign steatosis may progress to nonalcoholic steatohepatitis (NASH) and cirrhosis in a minority of individuals, the natural history of de novo hepatic steatosis after LT remains unknown. The significance of de novo hepatic steatosis and its risk of progression to NASH or cirrhosis raise further questions about the pathogenic mechanisms of this condition, and more importantly, about the measures that may be taken to improve associated hepatocellular injury. The article by Seo et al.1 in this issue of Liver Transplantation addresses this from the vantage point of a single-center experience and concludes that posttransplant weight gain is a positive risk factor for the development of de novo post-LT hepatic steatosis. However, this risk appears to be attenuated by the use of angiotensin converting enzyme inhibitors (ACE-Is), and this suggests that ACE-Is may independently reduce the risk of de novo development of NAFLD in the post-LT setting. ACE-I, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; HCV, hepatitis C virus; LT, liver transplantation; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; PPAR-γ, peroxisome proliferator-activated receptor γ. Seo et al. performed a retrospective analysis on a cohort (n = 68) of orthotopic LT patients who had paired donor liver biopsies and a liver biopsy performed after LT for the evaluation of abnormal liver enzymes at the University of California Davis Medical Center (Sacramento, CA). Conditions that would normally have an impact on the risk for hepatic steatosis, such as the hepatitis C virus (HCV), weight gain, immunosuppressive medications, and the presence of diabetes, were identified from patient charts. Paired liver biopsy specimens were reviewed in a blinded fashion by a hepatopathologist according to the Brunt scoring system.2 The time interval from LT to the subsequent follow-up liver biopsy was 28 ± 18 months. Multiple logistic regression analysis was used to determine the independent predictors of the development of benign hepatic steatosis and NASH. The cohort of patients reported in this study was predominantly male (81%) and infected with HCV (84%). Of those patients with HCV, 27% were noted to have genotype 3. HCV, particularly genotype 3, has been reported to be an independent predictor of hepatic steatosis and NASH despite insulin sensitivity and adiponectin levels comparable to those of age-matched and body mass index–matched controls.3 Interestingly, in the study by Seo et al., all patients with de novo NASH were also infected with HCV, and this raised the concern of whether the histologic recurrence of HCV could be readily distinguished from NASH and whether an overlap of both conditions was observed. The inclusion of HCV-infected recipients may confound the results, leading to an overestimation of de novo NAFLD and NASH in this cohort study. Although diabetes mellitus, hyperlipidemia, and weight gain are risk factors for NAFLD and NASH in the nontransplant setting, it is surprising that these variables were not independent risk factors for NAFLD after LT in the multivariate analysis of this study. Regardless of the risk factors associated with the development of de novo hepatic steatosis, treatment strategies that are aimed at ameliorating the long-term consequences of chronic necroinflammation in the liver or that improve insulin resistance, which promotes hepatic steatosis, are needed. Among the patients who developed de novo NAFLD versus those with no significant post-LT NAFLD, 83% (10/12) versus only 25% (14/56), respectively, had a weight gain greater than 10% of their pre-LT body mass index. From the presented data, it is difficult to discern what might have played a role in the weight gain and the development of hepatic steatosis and NASH after LT. Larger studies might clarify if limiting the cumulative dose and duration of corticosteroids and/or avoiding the use of sirolimus would reduce the risk for post-LT hepatic steatosis. Similarly, additional research is needed to determine if stronger efforts to prevent post-LT weight gain with diet and exercise counseling and/or the continuation of policies targeted at managing pre-LT obesity and diabetes after the operation would be helpful.4 Pharmacologic therapies that reduce hepatocellular injury and fibrogenesis, when used concomitantly with diet and exercise, may also decrease the risk for post-LT hepatic steatosis. Seo et al. are the first to show that ACE-Is may independently reduce the risk of hepatic steatosis in the post-LT setting. Recent work has focused on the network of cytokines and adipokines that regulate hepatic lipid homeostasis, identifying multiple effectors such as adiponectin, tumor necrosis factor α, resistin, retinol binding protein-4, and glucagon-like peptide-1 as having major regulatory roles in insulin signaling. Increased angiotensin system activity has also been found to be a major contributor to impaired insulin signaling. Large clinical studies of ACE-Is or angiotensin receptor blockers (ARBs) for hypertension have shown a consistent reduction in the development of type 2 diabetes.5 In mice, blocking the angiotensin type 1 receptor with losartan improved fructose-induced insulin resistance,6 and the genetic deletion of angiotensin type 1 down-regulated adipokine production by adipocytes. Additionally, the genetic deletion of angiotensinogen in mice caused mice to be more active, gain less weight, and have less adipose tissue than controls on a high-fat diet.7 ACE-I therapy may prevent obesity and increase the expression of uncoupling proteins, thereby reducing oxidant stress in diet-induced obese mice.8 Modulating the angiotensin system with clinically available inhibitors of angiotensin II production or binding AII to ATI receptors with ACE-Is or ARBs could be an attractive means of improving insulin signaling and preventing the development of hepatic steatosis and NASH. Because of the possible role of angiotensin in fibrogenesis, this approach may also have the added advantage of preventing the progression of NASH to cirrhosis. Although the molecular basis for these observations is not well described, there is considerable interest in the potential role of angiotensin inhibition on peroxisome proliferator-activated receptor γ (PPAR-γ), which has an important role in adipocyte differentiation and regulatory function in both hepatic fatty acid synthesis and oxidation.9 Therapeutic activation of PPAR-γ seems to be beneficial in mice and patients with NAFLD.10, 11 As noted by the authors, larger prospective trials for evaluating and validating the role of ACE-Is in NAFLD pathogenesis are necessary. The strengths of this article are the uniformity of immunosuppression management and post-LT care offered by a single-center experience, the utilization of guidelines set by Joint National Committee 7 and National Cholesterol Education Program III for defining the endpoints of hypertension and hyperlipidemia, and the application of a consistent method for grading liver histology grading, although clear histologic criteria for distinguishing between recurrent HCV and NASH on liver biopsy are not well delineated. Also missing from the report are the guidelines used to define the presence or absence of diabetes mellitus before and after transplant and the relationship of weight gain, diabetes, and hepatic steatosis to post-LT immunosuppression. Although this is the largest series of de novo NAFLD after LT yet published, a relatively small number of patients were studied, and the follow-up was short. Thus, larger studies that monitor patients for longer periods are needed to verify the incidence of post-LT NAFLD and characterize its natural history. Nevertheless, the most striking and noteworthy finding of this study is the demonstration that the incidence of NAFLD may be attenuated by the use of ACE-Is after LT. If this preliminary observation is validated, ACE-Is could become useful for reducing hepatic steatosis (and potentially NASH) in patients who have undergone LT. Where does this leave us? From our pre-LT experience, we know that NAFLD progresses to cirrhosis in some patients who develop NASH.12 Clearly, this would be a most unfortunate outcome in LT recipients. The presence of diabetes mellitus, increased body mass index, truncal obesity, hypertriglyceridemia, and increased age are risk factors for more advanced fibrosis and cirrhosis before LT.13 Unfortunately, most of these risk factors are worsened by current post-LT immunosuppressive regimens. Although the natural history of hepatic steatosis after LT has yet to be well defined, one worries that de novo or recurrent NAFLD may be more aggressive after LT. Indeed, this possibility is supported by data from Seo et al., which demonstrate NASH in half of the patients who develop NAFLD within the first few years after LT. As in the non-LT setting, the detection and accurate staging of post-LT NAFLD require the development of noninvasive methods to discern underlying histologic disease activity. Given the insensitivity of liver enzyme abnormalities for detecting steatosis, their inability to reliably distinguish hepatic steatosis from NASH, and their lack of correlation with hepatic steatosis, this article may in fact present an underestimation of the true prevalence of post-LT NAFLD. What should be done from here? The United Network for Organ Sharing database could be interrogated to detect factors in both donors and recipients that are associated with the development of post-LT NAFLD. The impact of the various posttransplant immunosuppressive medications on NAFLD incidence and progression could also be analyzed. Large centers routinely performing donor liver biopsies and post-LT histologic surveillance of their recipients are well situated to assess the natural history of post-LT NAFLD. Such data, if available, would help us to determine how frequently, and with what degree of severity, hepatic steatosis develops after LT. Furthermore, additional insight regarding the impact of specific immunosuppression regimen(s) and the cumulative dose and duration on the risk for developing NAFLD may be gained. Finally, studies to evaluate the potential risks and benefits of therapeutic interventions (i.e., diet and exercise, PPAR-γ agonists, ACE-Is, ARBs, and insulin-sensitizing agents) on NAFLD in the post-LT setting are necessary.