Co-infection with hepatitis B virus among tuberculosis patients is associated with poor outcomes during anti-tuberculosis treatment

Abstract

Background: Tuberculosis (TB) and chronic Hepatitis B virus (HBV) infection are common in China. Fist-line anti-TB medications often produce drug-induced liver injury (DILI). This study sought to investigate whether TB patients with chronic HBV co-infection are more susceptible to liver failure and poor outcomes during anti-TB treatment.

Methods: Eighty-four TB patients developed DILI during anti-TB treatment and were enrolled, including 58 with chronic HBV co-infection (TB-HBV group) and 26 with TB mono-infection (TB group). Clinical data and demographic characteristics were reviewed. The severity of DILI and incidences of liver failure and death were compared. Risk factors of clinical outcomes were defined.

Results: The patterns of DILI were similar in both groups. Compared with patients in the TB group, patients in the TB-HBV group who did not receive anti-HBV therapy before anti-TB treatment were more susceptible to Grade-4 severity of DILI (36.2% vs. 7.7%, P = 0.005), liver failure (67.2% vs. 38.5%, P = 0.013) and poor outcomes (37.9% vs. 7.7%, P = 0.005). Age > 50 years (48.1% vs. 22.6%, P = 0.049), cirrhosis (50.0% vs. 15.4%, P = 0.046) and total bilirubin > 20 mg/dl (51.6% vs. 14.8%, P = 0.005) were independent risk factors for the rate of death in the TB-HBV group, and HBV DNA > 20,000 IU/ml had borderline significance (44.1% vs. 20.8%, P = 0.081). In the TB-HBV group, nucleos(t)ide analogues as rescue therapy were not able to reduce short-term death (33.3% vs. 36.8%, P = 0.659) once liver failure had occurred.

Conclusions: Patients on anti-TB therapy with chronic HBV co-infection are more susceptible to developing liver failure and having poor outcomes during anti-TB treatment. Regular monitoring of liver function and HBV DNA level is mandatory. Anti-HBV treatment should be considered in those with high viral levels before anti-TB treatment.

Keywords: Clinical outcome; Drug-induced liver injury; Hepatitis B virus; Liver failure; Tuberculosis.

Fig. 1

Liver failure in the two groups and their clinical outcomes. Ten (38.5%) patients in the TB group developed liver failure compared to 39 (67.2%) in the TB-HBV group (P = 0.013). In the TB-HBV group, 20 patients died and 1 received orthotopic liver transplantation, which was classified as Grade-4 severity of DILI, while only 2 patients with Grade-4 severity of DILI in the TB group died. Abbreviations: TB, tuberculosis; HBV, hepatitis B virus; G, grade; LF, liver failure; R, recovered; I, improved; E, exacerbated; D, died; T, liver transplantation

Fig. 2

Kaplan-Meier curve for the survival analysis. Survival analysis in all patients showed that a HBV co-infection was an independent risk factor for death. Survival analyses in the TB-HBV group showed that b age > 50 years, c cirrhosis and d TBIL > 20 mg/dl were independent risk factors for death; e HBV DNA > 20,000 IU/ml had borderline statistical significance as risk factor for death. However, f NAs as rescue therapy were not able to reduce short-term death once liver failure had occurred

Fig. 3

HBV DNA profiles in the TB-HBV group. a HBV DNA changes before and after anti-TB treatment. B1, B2, B3 and B4 (solid line) were patients with better clinical outcomes. P1 (dot line) indicated the patient who underwent a sharp increase in HBV DNA from 2.00 log to 5.13 log and subsequently died; b HBV DNA levels after the onset of liver injury and admission to our hospital in different outcome subgroups (Median [log₁₀ IU/ml]: 5.55 vs. 4.24, P = 0.156)