LAUSR

Thank you for the comments by Professor Weng on our article “Longterm prognosis with or without nucleot(s)ide analogue therapy in hepatitis B virusrelated decompensated cirrhosis”.1 I generally agree with his points. Our original study used a retrospective design. Effective nucleot(s)ide analogues (NAs) have been developed for hepatitis B virus (HBV), and many patients are receiving these drugs at an early stage. The EASL Clinical Practice Guidelines state that the most frequent signs of the decompensated phase are ascites, bleeding, encephalopathy and jaundice.2 However, as Professor Weng points out, jaundice is not recognized as a typical symptom of decompensated cirrhosis (DC), though it is frequently observed in acuteonchronic liver failure (ACLF). There is agreement that ACLF is not just decompensation of liver cirrhosis, but a distinct syndrome.3 In fact, in our series, the total bilirubin levels in the NA and NonNA groups at the time of diagnosis of DC were 1.4 mg/dl (0.8– 3.6) and 0.8 (0.5– 1.9), respectively, which were not very high. Total bilirubin levels at the time of diagnosis of DC exceeded 6 mg/dl in only 13 patients (NA group, nine patients; NonNA group, four patients). Twelve patients died of liver failure during followup. All 12 of these patients had ACLF, and 11 of them were in the NonNA group. In our study, extrahepatic bacterial infections such as spontaneous bacterial peritonitis, bacteremia, urinary tract infection, and pneumonia were not recognized as decompensation events.4,5 However, most patients with these infections had ascites. In our hospital, the NA group started NAs on the day of DC diagnosis. Therefore, there was no bias regarding the start of observation in the NA and NonNA groups. Many patients in the NonNA group were treated when lamivudine was the only NA available, and antiviral therapy was not actively administered to patients with DC due to the fear that they would develop drug resistance. It is extremely important to point out that because many early deaths occurred in the NonNA group, the observation period was significantly shorter than in the NA group. In other words, it is assumed that many patients in the NonNA group died before developing hepatocellular carcinoma (HCC). Therefore, the finding that the two groups had a similar incidence of HCC should be evaluated with great care. A randomized controlled trial of DC with or without antiviral treatment would have clarified all of the above issues, but it would have been unethical. In the future it will be necessary to collect patients at multiple centres to resolve each of the identified problems.