The researchers of the REVEAL study analyzed the relationships between the development of HCC and high HBV DNA load, according to the natural history of chronic HBV infection, and the findings of this study suggested that effective control of HBV replication can reduce the risk of hepatocellular carcinoma in the long run theoretically. However, it is still unclear whether the risk of HCC and other long-term complications can be decreased by the recommendation of medical treatments including interferon and nucleotide/side analogues. Interferon has been used in the treatment of CHB for decades. Lots of studies have been reported the long-term prognosis of interferon, and Y.-F. Yang et al has generalized a conclusion that IFN prevents or delays the development of liver cirrhosis and HCC in patients with CHB . Lamivudine was the first nucleoside analogues used as long-term maintenance therapy since the late 1990s, and other nucleotide/side analogues including telbivudine, entecavir, adefovir and tenofovir were approved recent years. The clinical efficacy and particularly on the incidence of long-term complications under long-term antiviral therapy are difficult to monitor because cirrhotic complications and HCC take long time to develop, especially in noncirrhotic patients whose incidence of liver-related complications is even lower. As a result, in most clinical trials, biochemical response, virologic response and histologic response [32–34] have been used as surrogate end points to determine if the treatment of patients with chronic HBV infection has been successful. Whether these surrogate end points can reflect long-term clinical outcomes was still an open question . Vincent Wai-Sun Wong at al. has confirmed that normalization of serum ALT, HBV DNA suppression, HBeAg loss, or HBeAg seroconversion at the end of drug trials were associated but unable to predict long-term clinical outcomes completely . Even histologic response can't exactly reflect the long-term clinical outcomes as regression of compensated cirrhosis can be translated into favorable clinical outcome in those treated with long-term nucleotide/side analogues therapy.
In this meta-analysis, we choose the clinical end points including decompensated cirrhosis, CHB-related death or CHB-related HCC which directly reflect the prognosis of chronic HBV infection to evaluate the efficacy of the long-term nucleos(t)ide analogues therapy.
The most important finding of this meta-analysis is that more than two years NA therapy for adults with CHB reduces the risk of LTCs including decompensated cirrhosis, CHB-related death or CHB-related HCC by pooling data from six studies[23–28].We also analyzed the various factors of effecting the risk of LTCs in some items such as drug-resistance, HBeAg status or pre-existing compensated cirrhosis, and in any item, NA therapy was shown the benefit than placebo of reducing the incidence of LTCs. So for patients with drug-resistance, NA therapy still had a better outcome than the untreated patients with lower risk of LTCs. Among the treated patients, we also investigated whether drug-resistance effected the risk of LTCs. Then it was found that the incidence of LTCs were lower in the non-nucleoside drug resistance groups. At the same time, we compared the prognosis of patients with and without pre-existing compensated cirrhosis among the treatment groups. Then it was found that the incidence of LTCs were lower in patients without pre-existing compensated cirrhosis. It suggested us that the earlier antiviral therapy is been done, the better the prognosis will be. Based on the two conclusions above, the patients who need take antiviral drugs should receive the antiviral therapy as soon as possible. Moreover, and it would be best that drug-resistance genes were detected before patients administrated the antiviral drugs, then proper antiviral therapy could be chosen to reduce the probability of drug-resistance. On the other hand, management of drug-resistance is absolutely important, too. Patients should regularly be followed-up to monitor the drug-resistance. Fortunately, with the available of newer drugs like adefovir dipivoxil without cross resistance to lamivudine , the problem of LAM-resistance will go to remission. Currently, add-on adefovir therapy is the most widely accepted strategy in case of LAM-resistance. Chen et al. confirmed the view that the combination of ADV with LAM was superior in inhibiting HBV replication and preventing drug resistance as compared to ADV alone for LAM-resistant CHB patients by a meta-analysis. Other nucleotide/side analogues were available such as entecavir [39–41], tenofovir  recent years.
There are several limits in our meta-analysis. Firstly we collected the data from not only randomized controlled trials but also case-control, cohort, and historical controls studies because of very few RCTs. When patients are diagnosed CHB or HBV-related cirrhosis especially with the abnormal liver function and the evidence of HBV replication, it is unethical to running randomized controlled trials. Maybe this is the leading reason that so few RCTs were conducted. Secondly, the length of following up differed among studies, ranging from 2.7 years[23, 25]to 8.2 years. As we know, the longer time of following up, the higher incidence of LTCs would be seen. So in our meta-analysis, the incidence of LTCs differed significantly. Maybe this is one reason of statistical heterogeneity. Thirdly, the data we've selected was lacking of testing whether drug-resistance existed or not before therapy because most of the studies didn't show the data about it.
In conclusion, the result of this meta-analysis indicate that long-term nucleos(t)ide analogue therapy for adults with CHB prevents or delays the development of long-term complications including decompensated cirrhosis, CHB-related death or CHB-related HCC in patients with CHB. The patients who need take antiviral drugs should receive the antiviral therapy as soon as possible. During the monotherapy, when LAM-resistance happened, lamivudine should be combined with other antiviral drugs without cross resistance to lamivudine since it still has more benefit than the untreated patients.