Editorial: looking for patients at risk of cirrhosis in the general population-many needles in a haystack.

Abstract

Hidden cirrhosis is a common oversight in clinical practice. In fact, it is not unusual to find cirrhosis during routine laparoscopic surgery (Figure 1). Apart from signs and symptoms of cirrhosis , which are scarce in clinical practice, serum aminotransaminase elevation remains the hallmark signal that raises suspicion of liver disease in primary care .1 However, Harman et al observed that more than two-thirds of cirrhotic patients showed normal AST/ALT levels,2 probably because necroinflammatory reactions usually disappear as the liver progresses to cirrhosis (“ burn-out cirrhosis ”). Thus, general practitioners and hepatologists have to begin to adjust their thought process to look for more accurate surrogate markers in at- risk patients . Non-invasive serum methods , including APRI, FIB-4 and NAFLD Fibrosis Score have demonstrated ability to detect advanced fibrosis in the general population , but the positive predictive value remains suboptimal.3 Transient elastography has demonstrated a strong correlation with liver fibrosis , mainly in viral-related liver disease , but also with several outcomes including liver cancer and survival .4 Harman et al detected 27% and 3% of at- risk patients with significant liver disease (by transient elastography , >8 kPa) and cirrhosis (confirmed by histological, radiological and/or biochemical criteria) respectively. However, the accessibility was a major issue because less than 50% of patients invited were assessed by this method . The prevalence of cirrhosis was higher in those with multiple risk factors , demonstrating an additive effect for advanced liver disease among diabetes, obesity and alcohol.2 This interaction may be due to shared pathogenic mechanisms including genetic background (PNPLA3, TM6SF2 and MBOAT).5 Questionnaires , such as AUDIT, can detect alcohol dependence beyond hazardous alcohol consumption in clinical practice.6 However, alcohol consumption seems to have a different effect on the liver depending upon baseline risks for liver disease .7 Drinking more than 21 units per week in males or 14 in females is considered harmful in patients with underlying hepatopathy or the presence of metabolic disorders.8 However, in patients without any risk factors , alcohol consumption seems to be potentially protective against raised liver stiffness ( patients with elevated stiffness drank less than those with non-elevated stiffness). Some bias could undermine this association . Thus, an appropriate definition of risk for liver disease in patients consuming alcohol, without metabolic risk factors , should be clarified.