We read with interest the article by Ghouri et al.,1 who reviewed the evidence regarding the link between nonalcoholic fatty liver disease (NAFLD) and cardiovascular disease (CVD). The authors concluded that the connection between NAFLD and CVD is not well supported by existing data because of the presence of confounders such as age and established cardiovascular risk factors.
We agree that the main limitation of these studies is that their results make it difficult to distinguish the contribution of liver fat per se to the risk of CVD. However, we should consider that the liver is the main regulator of insulin sensitivity and finely tunes insulin-regulated metabolic pathways such as glucose and lipid homeostasis that are involved in endothelial dysfunction and atherogenesis. Studies in null mice have clearly substantiated this issue. In particular, the disruption of insulin signaling in the liver is more relevant to whole body glucose homeostasis than its disruption in adipose tissue and muscle.2 In addition, hepatic insulin signaling regulates the secretion of very low density lipoprotein and thus lipotoxicity and atherogenesis.3 Therefore, it is impossible to identify the intrahepatic triglyceride level, which precisely reflects liver insulin resistance, as an isolated variable in the generation of CVD risk.
Furthermore, when we consider cardiovascular mortality, we should refer not only to coronary heart disease but also to chronic heart failure due to nonischemic cardiomyopathy, which is associated with alterations in substrate utilization (cardiac work/myocardial oxygen consumption) occurring early in the cascade of events leading to impaired ventricular contractility. In this respect, Perseghin et al.4 demonstrated that young individuals without obesity, diabetes, or hypertension who had a fatty liver showed echocardiographic features of early left ventricular dysfunction and impaired energy metabolism (measured by cardiac 31P magnetic resonance spectroscopy). The phosphocreatine/adenosine triphosphate ratio, a recognized in vivo marker of myocardial energy metabolism, was inversely related to both plasma glucose and insulin levels in that study and was also tightly related to liver fat. These findings suggest that NAFLD is not merely a marker of metabolic dysfunction but may be actively involved in the initiation and progression of CVD.
Finally, we agree with the authors' conclusion that methodologically rigorous prospective studies evaluating not only surrogate markers but also liver histology are warranted in order to dissect the precise contribution of a fatty liver to the risk of CVD. In the interim, we suggest that there is consistent pathophysiological evidence indicating that the evaluation and management of a fatty liver should be considered a mainstay for the prevention of metabolic CVD.