HBsAg Quantitation Cannot Predict Virological Flare in Chronic Inactive Hepatitis B Infection
Chainuvati S, MD1, Kitikomonkun P, MD1, Tanwandee T, MD1
Affiliation : 1 Division of Gastroenterology, Department of Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand
Background : Chronic hepatitis B, HBeAg negative infection with inactive HBV DNA can become active during follow-up. Previous
studies have demonstrated that male, younger age and higher HBV virus at entry are among predictors of subsequent virological
flare. There are studies which demonstrated that quantitative HBsAg (qHBsAg) correlated with cccDNA and might predict HBV
DNA flare.
Objective : This study was aimed to determine whether qHBsAg and/or HBV DNA can predict HBV DNA flare.
Materials and Methods : This is a prospective study includes inactive chronic hepatitis B patients naïve to treatment with HBV DNA
<2,000 IU/mL at hepatitis clinic, Siriraj Hospital from May 1, 2008 to December 31, 2009. The patients were followed every 6
months with ALT, HBV DNA viral load and qHBsAg to evaluate virological flare at 1 year of follow-up.
Results : There were 91 patients (49 male, 42 female), mean age of 50 years old with mean follow-up time of 308 days. Baseline mean
HBV DNA, qHBsAg and ALT were 364 IU/ml (range 6 to 1,930), 2,831 IU/mL (range 0.15 to 23,794) and 28 U/L (range 8 to 78),
respectively. At last follow-up, the mean values of HBV DNA, qHBsAg, and ALT were 1,310 IU/ml (range 6 to 51,500), 2,239 IU/mL
(range 0.04 to 29,242) and 26 U/L (range 9 to 88), respectively. Twenty-one of 91 patients (23%) had virological flare at 1 year of
follow-up. There is no correlation of qHBsAg and HBV DNA over the time (r = 0.25 (p<0.001). The only predictor of virological flare
was the higher baseline DNA (p<0.001). Overall, qHBsAg level was not the predictor of virological flare, however, when we analyzed
the patients with initial HBsAg level <250 IU/mL we found that this level can predict absent of virological flare with the sensitivity,
specificity, positive predictive value (PPV) and negative predictive value (NPV) of 76.2, 55.7, 34.0, and 88.6%, respectively, with AUC
of 0. 66 (95% CI, 0.54 to 0.77, p = 0.03). qHBsAg combined with HBV DNA below detection limit even offered better predictive values
with sensitivity, specificity, PPV and NPV of 100, 23, 28, and 100%, respectively (p = 0.018). Those with virological flare, the only
predictor of sustained virological flare, were high baseline ALT.
Conclusion : Inactive chronic hepatitis B can have virological flare of 23% in 1 year. Our study has found poor correlation between
qHBsAg and HBV DNA. However, low baseline HBV DNA together with low qHBsAg level (<250 IU/mL) is useful to predict the
absence of subsequent virological flare in inactive CHB patients. However, more data with more patients and longer follow-up
period are needed to confirm this finding.
Keywords : qHBsAg, HBV DNA, Predicting factors of HBV flare
