Incidence and associated risk factors of anti-tuberculosis drug induced liver injury among TB patients.

BACKGROUND

Tuberculosis (TB) is a global health challenge. Use of anti-TB drugs to treat TB is associated with high prevalence of side effects that include anti-TB drug induced liver injury (ATDILI). The aim of this study was to determine the incidence and associated clinical risk factors of ATDILI among South African patients on treatment for TB disease.

METHODS

This was an ambispective case-control study of patients on treatment for TB disease receiving drug sensitive and drug resistant treatment regimens. The retrospective and prospective studies were done from January 2021 to June 2024, involving a total of 610 patients on treatment for TB disease from whom 13 had ATDILI. From the retrospective and prospective cohorts, we extracted 13 ATDILI cases and 276 controls. Additionally, 44 severe ATDILI cases were directly recruited from the hospital to enrich the number of cases in the study. Prospective patients were followed for up to 6 months, while retrospective participants had a single visit with follow-up durations determined by their parent studies. Logistic regression analysis was performed to identify clinical and demographic factors associated with the development of ATDILI).

RESULTS

In the studied cohorts, the incidence of ATDILI was 2.1% (13/610). The ATDILI cases and controls consisted of 215 (64.6%) male patients; 57 patients were diagnosed with hepatotoxicity, 44 from the hospitalized cohort,12 from the retrospective cohort and 1 patient from the prospective cohort. The median time from the initiation of treatment to the onset of hepatotoxicity was approximately 30 days. Univariate logistic regression revealed significant differences (p < 0.05) in sex (p = 0.003), HIV status (p = 0.002), BMI (p = 0.038), hypertension (p = 0.047) smoking (p = 0.006), and alcohol consumption (p = 0.001) in relation to ATDILI. Multivariate analysis further revealed that female sex (p = 0.041), HIV status (0.022) and alcohol consumption (0.048) were independently associated with an increased risk of ATDILI.

CONCLUSIONS

The incidence of ATDILI in this study was 2.1%, lower than the previously reported range of 3–36% and highlights the need for standardized definitions of phenotypes across studies. In univariable analysis, female sex, HIV status, BMI, hypertension, smoking, and alcohol consumption were identified as potential risk factors for ATDILI. Among these, multivariable analysis suggested that female sex, HIV status and alcohol consumption were associated with an increased risk of developing ATDILI.