We read with great interest Garg et al.'s article1 published in HEPATOLOGY. The authors conducted a randomized study to compare the efficacy of tenofovir disoproxil fumarate (TDF) therapy and placebo therapy in patients suffering from a severe spontaneous reactivation of chronic hepatitis B (CHB) presenting as acute-on-chronic liver failure. They reported a high overall mortality rate of 63.0% (17/27), with rates of 43% and 85% in the TDF and placebo arms, respectively. TDF significantly reduced the mortality rate and hepatitis B virus DNA levels and improved the Child-Turcotte-Pugh and Model for End-Stage Liver Disease (MELD) scores in these patients at 3 months.
It is noteworthy that some patients with cirrhosis were enrolled. First, we consider 3 months of observation to be too short for determining the survival of these patients. We reexamined 96 patients with liver decompensation treated with lamivudine in our previous study in Taiwan.2 Eight patients (8.3%), two patients (2.1%), and three patients (3.1%) died within 1, 1 to 3, and 3 to 6 months of lamivudine treatment, respectively. In other words, 23.1% of the patients (3/13) who did not survive for 6 months died after 3 months of antiviral treatment. Villeneuve et al.3 reported that 25% of their patients (1/4) without hepatocellular carcinoma died from hepatic failure within 3 to 6 months of the initiation of lamivudine treatment. Fontana et al.4 reported that patients with decompensated cirrhosis were still dying even after the first 6 months. Hence, the mortality rate is possibly underestimated in Garg et al.'s study.1 The mean baseline MELD scores (27 and 25 in the TDF and placebo arms, respectively) reflect the fact that the patients enrolled in their study had more severe liver disease. In HEPATOLOGY, Liaw et al.5 reported lower mortality rates for patients with CHB and decompensated liver disease who were treated with TDF (4.4% or 2/45) or entecavir (9.1% or 2/22) by 48 weeks. The lower mortality rate is likely related to the milder severity of the patients' liver disease (the median baseline MELD scores were 11 and 10.5 for the TDF and entecavir arms, respectively). Hence, the earlier initiation of nucleot(s)ide analogue therapy seems critical for these patients.
Another issue is the design of this study of patients with cirrhosis and CHB-related acute-on-chronic liver failure: a placebo arm was included. With a lack of facilities for liver transplantation, Garg et al.1 observed a high mortality rate, and most deaths (82%) occurred because of progressive liver failure that led to the development of multiorgan failure.1 With a mortality rate of 4% to 30% 6 to 12 months after lamivudine, telbivudine, and entecavir therapy3-8 and with significant improvements in the long-term survival of patients with hepatic failure,4 it does not seem appropriate to include a placebo arm in studies enrolling patients with cirrhosis and critical liver failure. In the era of nucleot(s)ide analogue therapy with more potent anti–hepatitis B virus effects, we look forward to the results of more large-scale studies seeking to clarify whether the efficacy can be improved, particularly in patients with cirrhosis, CHB, and acute exacerbation, who have a poorer prognosis.