Introduction
Tuberculosis (TB) is a communicable disease that remains a major cause of ill health and one of the leading causes of mortality worldwide. India bears a substantial proportion of the global tuberculosis burden. According to the India TB Report 2022, tuberculosis case notifications increased by 19% in 2021 compared with the previous year. A total of 19,33,381 incident tuberculosis cases, including new and relapse cases, were notified in 2021, compared with 16,28,161 cases in 2020. India accounts for approximately 27% of the global tuberculosis burden [1-4].
Drug-induced hepatotoxicity refers to acute or chronic liver injury resulting from exposure to a drug. Antitubercular drug-induced hepatotoxicity is a serious adverse effect of antitubercular therapy (ATT) and is an important cause of treatment interruption or discontinuation. Such interruptions may adversely affect treatment outcomes and contribute to increased morbidity, mortality, and the development of drug-resistant tuberculosis [5-8].
The incidence of drug-induced liver injury (DILI) associated with antitubercular drugs varies considerably across populations and has been reported to range from 2% to 39% worldwide, with mortality ranging from 4% to 12%. Drug-induced liver injury due to antitubercular drugs accounts for more than 7% of adverse effects associated with antitubercular therapy. A relatively higher incidence of antitubercular drug-induced liver injury has been reported in Asian populations [9,10].
Several genetic and acquired factors have been associated with an increased risk of antitubercular drug-induced hepatotoxicity. These include malnutrition, socioeconomic factors, human immunodeficiency virus (HIV) infection, hepatitis B virus (HBV) infection, and other coexisting conditions. The development of hepatotoxicity during antitubercular therapy may necessitate interruption or modification of treatment, thereby increasing the risk of inadequate treatment and the subsequent emergence of drug-resistant tuberculosis [11-16].
Considering the clinical significance of antitubercular drug-induced hepatotoxicity and its potential impact on tuberculosis treatment outcomes, the present study was undertaken to evaluate the clinical and laboratory profiles of affected patients, identify associated risk factors and their relationship with the severity of drug-induced liver injury, and assess the short-term outcomes of patients developing hepatotoxicity during antitubercular therapy [17-20].
The present study aimed to evaluate the clinical and laboratory profiles of patients with antitubercular drug-induced hepatotoxicity. It also sought to identify the risk factors associated with antitubercular drug-induced liver injury and determine their correlation with the severity of DILI. Additionally, the study aimed to assess the short-term outcomes of patients who developed hepatotoxicity during antitubercular therapy [21-23].
Materials and Methods
This prospective observational study was conducted among patients admitted to the wards and those attending the outpatient department of a tertiary care hospital. The study population comprised adult patients diagnosed with antitubercular drug-induced hepatotoxicity who fulfilled the predefined inclusion and exclusion criteria. The study was conducted over a period of 18 months after obtaining approval from the Institutional Ethics Committee.
The sample size was calculated using the formula n = Z²p(1−p)/e², where n represents the required sample size, Z represents the standard normal variate corresponding to the desired confidence level (1.96 for a 95% confidence interval), p represents the estimated prevalence, and e represents the maximum acceptable margin of error. Based on an estimated prevalence of 9.48%, as reported in a previous study conducted at Christian Medical College, Vellore, and an allowable error of 5%, the calculated sample size was 137 participants.
Patients aged ≥18 years who were diagnosed with antitubercular drug-induced hepatotoxicity according to the predefined diagnostic criteria and who provided written informed consent were included in the study. Pregnant patients, patients with underlying malignancy, those with abnormal liver function tests at baseline, and patients receiving liver-safe or modified antitubercular regimens because of pre-existing liver disease were excluded.
After obtaining written informed consent, each participant underwent detailed clinical evaluation, including history taking, general physical examination, and systemic examination. Investigations performed by the treating physician as part of the standard diagnostic work-up for hepatitis or drug-induced liver injury were recorded in a case record form. Relevant biochemical and laboratory parameters, including serum glutamic-oxaloacetic transaminase (SGOT)/aspartate aminotransferase (AST), serum glutamic-pyruvic transaminase (SGPT)/alanine aminotransferase (ALT), serum bilirubin, prothrombin time/international normalized ratio (PT/INR), and findings on ultrasonography of the abdomen, were recorded. Other investigations performed as clinically indicated were also documented. Details of the treatment administered by the treating physician were recorded without any intervention or modification by the investigators.
Patients admitted to the hospital were followed weekly until discharge or death and subsequently followed weekly for up to six weeks. Patients managed on an outpatient basis were similarly followed for a period of six weeks. Clinical status, relevant laboratory parameters, treatment, complications, and outcomes were documented during follow-up. Laboratory parameters, including serum transaminases and bilirubin, were assessed at the initial presentation and compared with values at discharge, death, or the six-week outpatient follow-up, as applicable.
Data were entered into Microsoft Excel and analysed using IBM SPSS Statistics, version 21. Descriptive statistics were used to summarize the demographic, clinical, and laboratory characteristics of the study population. Categorical variables were expressed as frequencies and percentages, while continuous variables were summarized using appropriate measures of central tendency and dispersion. The proportion of patients with antitubercular drug-induced hepatotoxicity was calculated as applicable. Changes in biochemical parameters between the initial assessment and subsequent follow-up were analysed using paired statistical tests. Associations between categorical clinical variables, risk factors, treatment characteristics, and outcomes were assessed using the chi-square test or an appropriate alternative when required. Continuous laboratory parameters were compared between groups using appropriate parametric tests, including the independent-samples t-test or analysis of variance (ANOVA), depending on the number of groups being compared. Paired t-tests were used to compare continuous variables measured at baseline and follow-up. A two-sided p-value <0.05 was considered statistically significant at the 95% confidence level.
The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki, Indian Good Clinical Practice guidelines, and the ethical guidelines of the Indian Council of Medical Research. Participation was voluntary, and written informed consent was obtained from all participants after explaining the nature and purpose of the study. Participants were free to decline participation without any effect on their routine medical care. Confidentiality and anonymity of participant information were maintained throughout the study, and personally identifiable information was not disclosed during data analysis, dissemination, presentation, or publication of the study findings. As the study was purely observational and involved no study-specific therapeutic intervention, no additional risk attributable to an experimental intervention was anticipated.
Results
A total of 137 patients with antitubercular therapy (ATT)-induced hepatotoxicity were included in the study. Of these, 84 (61%) were females and 53 (39%) were males. The majority of patients belonged to the age group of 18–40 years (n=88, 64.23%), followed by 41–60 years (n=34, 24.82%) and >60 years (n=15, 10.95%).
Extrapulmonary tuberculosis was the most common type of tuberculosis and was present in 79 (58%) patients, while 45 (33%) patients had pulmonary tuberculosis and 13 (9%) had both pulmonary and extrapulmonary tuberculosis. With regard to the specific sites of tuberculosis, pulmonary tuberculosis was present in 45 patients, abdominal tuberculosis in 22, disseminated tuberculosis in 18, pleural tuberculosis in 17, central nervous system tuberculosis in 13, and lymph node tuberculosis in 11 patients. Eleven patients had other forms of tuberculosis, including psoas abscess, renal abscess, and tubercular osteomyelitis.
The interval between initiation of ATT and the development of symptoms of hepatotoxicity ranged from one to eight weeks. The highest proportion of patients developed symptoms two weeks after initiation of ATT (n=45, 32.84%). Symptoms developed after one week in 22 (16.05%) patients, three weeks in 23 (16.78%), four weeks in 21 (15.32%), five weeks in 8 (5.82%), six weeks in 4 (2.91%), and seven weeks in 3 (2.18%) patients.
Nausea and vomiting were the most frequently reported symptoms and were present in 121 (88%) patients, followed by anorexia in 50 (36%), jaundice in 26 (19%), altered sensorium in 16 (12%), abdominal distension in 2 (1.5%), and bleeding manifestations in 1 (0.7%) patient. Seventeen (12%) patients were asymptomatic.
Among the associated comorbidities, diabetes mellitus was present in 37 (27%) patients and hypertension in 35 (26%). Chronic liver disease was present in 7 (5.1%) patients, asthma in 5 (3.6%), ischemic heart disease in 3 (2.2%), cerebrovascular disease in 3 (2.2%), and chronic kidney disease in 2 (1.5%) patients. None of the evaluated comorbidities showed a statistically significant association with the severity of DILI (all p>0.05). Alcohol consumption was reported in 36 (26%) patients, smoking in 14 (10%), and other addictions in 2 (1.5%) patients. No statistically significant association was observed between these addictive behaviours and DILI severity.
The mean BMI of the study population was 20.98±2.8 kg/m², with a median of 21.1 kg/m² (IQR: 18.00–23.17) and a range of 16.60–31.10 kg/m². Thirty-eight (27.73%) patients were underweight (BMI <18.5 kg/m²), 44 (32.11%) had a normal BMI (18.5–22.9 kg/m²), 50 (36.49%) were overweight (BMI 23–24.9 kg/m²), and 5 (3.64%) were obese (BMI ≥25 kg/m²).
On general examination, lymphadenopathy was the most frequent finding, observed in 59 (43%) patients, followed by icterus in 34 (25%), pallor in 28 (20%), pedal oedema in 17 (12%), and flaps in 5 (3.6%). Bleeding manifestations were observed in 1 (0.7%) patient, while cyanosis, clubbing, and other signs of liver cell failure were absent. Icterus (p<0.001) and pedal oedema (p=0.007) showed statistically significant associations with the severity of DILI.
Organomegaly was uncommon. Hepatomegaly was present in 5 (3.6%) patients and splenomegaly in 6 (4.4%), while 126 (92%) patients had no organomegaly. Ascites was absent in 117 (85%) patients, while 13 (9.5%) had mild ascites and 7 (5%) had moderate ascites. None of the patients had gross ascites.
With regard to viral markers, 18 (13%) patients were HIV-positive, 8 (5.8%) were positive for hepatitis B surface antigen (HBsAg), and 2 (1.5%) were positive for anti-hepatitis C virus antibodies. None of the patients tested positive for hepatitis A IgM or hepatitis E IgM antibodies. Viral markers were negative in 109 (79.7%) patients.
Based on the severity of DILI, 39 (28%) patients had mild DILI, 56 (41%) had moderate DILI, and 42 (31%) had severe DILI. Thus, moderate DILI was the most common severity category.
Several potential risk factors for ATT-induced hepatotoxicity were evaluated. Female gender was present in 84 (61%) patients, low BMI (<18.5 kg/m²) in 38 (27.73%), alcohol consumption in 36 (26%), HIV infection in 18 (13%), hepatitis B or hepatitis C coinfection without underlying chronic liver disease in 10 (7.3%), and potential drug interactions due to polytherapy in 7 (5.1%) patients. Among these factors, HIV infection showed a statistically significant association with the severity of DILI (p=0.021). Female gender, low BMI, hepatitis B or C coinfection, alcohol consumption, and polytherapy were not significantly associated with DILI severity.
Regarding short-term outcomes, improvement in symptoms was observed in 125 (91%) patients, and improvement in liver function tests (LFTs) was documented in 125 (91%) patients. Rechallenge with ATT was possible in 122 (89%) patients, while 12 (9%) patients died during the study period. Among the study population, 12% were asymptomatic. Symptomatic improvement occurred within one week in 52%, within two weeks in 20%, and within three weeks in 8% of patients. Nine percent did not show symptomatic improvement because of death. Thus, the majority of surviving symptomatic patients showed clinical improvement within three weeks.
Improvement in LFTs occurred progressively during follow-up. Improvement was observed within one week in 2% of patients, within two weeks in 20%, within three weeks in 43%, within four weeks in 17%, and within five weeks in 9%. Nine percent of patients did not demonstrate improvement because of death. Overall, most surviving patients demonstrated improvement in LFTs within five weeks.
Rechallenge with isoniazid was possible at three weeks in 34% of patients, at four weeks in 26%, at five weeks in 20%, at six weeks in 7%, and after six weeks in 1%. Rechallenge was not possible in 12% of patients. Thus, isoniazid could be successfully reintroduced within six weeks in the majority of surviving patients. Rifampicin rechallenge was possible at two weeks in 1% of patients, three weeks in 5%, four weeks in 28%, five weeks in 38%, six weeks in 14%, and after six weeks in 1%. Rifampicin rechallenge was not possible in 13% of patients, including the 12 patients who died and one patient who developed rifampicin-induced thrombocytopenia. Pyrazinamide rechallenge was performed at five weeks in 1% and at six weeks in 4% of patients. In 94% of patients, pyrazinamide rechallenge was not undertaken during the study follow-up period, which was limited to six weeks.
Serial laboratory investigations demonstrated improvement in biochemical parameters over the six-week follow-up period. Haemoglobin levels remained relatively stable, while the total leukocyte count showed a declining trend and platelet counts increased. Total and direct serum bilirubin concentrations progressively decreased during follow-up. Similarly, serum aspartate aminotransferase (AST/SGOT), alanine aminotransferase (ALT/SGPT), and alkaline phosphatase (ALP) levels showed marked declining trends, indicating biochemical recovery.
At the initial visit, the mean haemoglobin level was 10.62 g/dL compared with 10.67 g/dL at six weeks (p=0.0242). The mean total leukocyte count decreased from 8,307 cells/mm³ at the initial visit to 6,447 cells/mm³ at six weeks (p=0.0031). The mean platelet count increased from 3.03 lakh/mm³ to 3.22 lakh/mm³ (p=0.0208). Mean total serum bilirubin decreased from 3.4 mg/dL at presentation to 0.90 mg/dL at six weeks (p<0.0001), while mean direct bilirubin decreased from 2.4 mg/dL to 0.53 mg/dL (p<0.0001). Mean AST/SGOT decreased from 412 IU/L to 33 IU/L (p<0.0001), and mean ALT/SGPT decreased from 371 IU/L to 32 IU/L (p<0.0001). Similarly, mean serum ALP decreased from 241 IU/L at presentation to 152 IU/L at six weeks (p<0.000001).
A total of 12 (9%) patients died during the study period. Among these, 3 were aged 18–40 years, 4 were aged 41–60 years, and 5 were aged >60 years. Of the 12 patients who died, 9 were males and 3 were females. Ten patients who died had severe DILI, while 2 had moderate DILI; no deaths occurred among patients with mild DILI. Regarding BMI, 7 of the patients who died had a BMI of 18.5–22.9 kg/m², 4 had a BMI of 23–24.9 kg/m², and 1 had a BMI ≥25 kg/m². None of the patients who died were underweight.
Overall, moderate DILI was the most frequent severity category. HIV infection was significantly associated with DILI severity, while icterus and pedal oedema were significant clinical findings associated with severity. Most patients showed symptomatic and biochemical improvement during the six-week follow-up period, and ATT rechallenge was possible in the majority of surviving patients. Mortality occurred predominantly among patients with severe DILI.
Discussion
Drug-induced liver injury (DILI) is an important complication of antitubercular therapy (ATT) and remains a major concern in the management of tuberculosis (TB). The liver plays a central role in the metabolism and detoxification of several antitubercular drugs, making it particularly susceptible to drug-induced injury. The reported incidence of ATT-induced DILI varies considerably across populations, ranging from 2% to 39%, while mortality following the development of jaundice has been reported to range from 4% to 12%. Several factors, including malnutrition, underlying liver disease, alcohol consumption, age, sex, metabolic factors, drug interactions, and isoniazid acetylator status, have been implicated in the development of hepatotoxicity. ATT-induced hepatotoxicity may necessitate interruption or modification of treatment, potentially affecting treatment adherence, increasing the risk of drug resistance, and adversely affecting morbidity and mortality. Early recognition and appropriate management of DILI are therefore essential for optimizing tuberculosis treatment outcomes.
The present prospective observational study included 137 patients with ATT-induced hepatotoxicity who were followed for six weeks. Previous studies have reported varying frequencies of ATT-induced DILI. Singhal et al. reported ATT-induced DILI in 10% of their study population, while Saha et al. reported a prevalence of 9.48%. Differences in the reported occurrence of hepatotoxicity may be related to variations in study populations, diagnostic criteria, treatment regimens, underlying risk factors, and clinical monitoring practices.
In the present study, 64.23% of patients were aged 18–40 years, 24.82% were aged 41–60 years, and 10.95% were aged >60 years. The median age was 30 years (IQR: 21–48 years), while the mean age was 36±17 years. No significant association was observed between age group and DILI severity (p=0.344). These findings differ from those reported by Gaude et al., who observed a higher risk of DILI among patients aged >60 years, with a mean age of 47±7.2 years. Similarly, Juganya et al. reported a mean age of 51.9±18.7 years among patients who developed ATT-induced hepatotoxicity, with a relatively greater number of cases among older patients. These differences may reflect variations in the demographic characteristics and underlying risk profiles of the study populations.
Females constituted the majority of our study population, accounting for 61% (n=84), while males accounted for 39% (n=53). However, gender was not significantly associated with DILI severity. In contrast, Gaude et al. reported that 63.3% of patients who developed ATT-induced DILI were males and 36.7% were females. Juganya et al. similarly reported a male predominance, with 19 males and 11 females among 30 patients who developed DILI. These observations indicate that the relationship between gender and ATT-induced hepatotoxicity may vary across populations, and our findings do not support gender as a significant determinant of DILI severity.
Extrapulmonary tuberculosis was the predominant form of TB in the present study, occurring in 58% of patients, while 33% had pulmonary TB and 9% had both pulmonary and extrapulmonary involvement. Muluken et al. similarly reported a predominance of extrapulmonary TB among patients developing ATT-induced DILI, whereas Gaude et al. and Juganya et al. reported pulmonary TB more frequently. These variations may be related to differences in the populations studied, referral patterns, and severity and distribution of tuberculosis.
Most patients developed manifestations of hepatotoxicity relatively early after initiation of ATT. The highest proportion developed symptoms during the second week of treatment, and the median time to symptom onset was approximately two weeks. This finding is comparable to that of Juganya et al., who reported that most patients developed manifestations of hepatitis within the first two weeks of ATT. Gaude et al. reported an average duration of approximately 20 days between initiation of ATT and development of DILI. These findings emphasize that the early weeks following initiation of ATT represent an important period for clinical surveillance, particularly among patients with additional risk factors for hepatotoxicity.
Nausea and vomiting were the most common presenting symptoms in our study, occurring in 88% of patients, followed by anorexia in 36% and jaundice in 19%. Twelve percent of patients were asymptomatic. Altered sensorium was observed in 12%, while bleeding manifestations and abdominal distension were uncommon. Singhal et al. also reported nausea and vomiting as common manifestations of ATT-induced hepatitis, while Juganya et al. identified vomiting as the most frequent symptom, followed by loss of appetite and jaundice. The predominance of gastrointestinal manifestations across these studies indicates that symptoms such as persistent nausea, vomiting, and anorexia developing after initiation of ATT should prompt consideration of hepatotoxicity and appropriate biochemical evaluation.
Diabetes mellitus and hypertension were the most common comorbidities in the present study, affecting 27% and 26% of patients, respectively. Other comorbidities, including ischemic heart disease, cerebrovascular disease, chronic liver disease, chronic kidney disease, and asthma, were less frequent. None of the evaluated comorbidities showed a statistically significant association with DILI severity. Gaude et al. similarly did not demonstrate a significant association between diabetes mellitus and hepatotoxicity, while Muluken et al. reported no significant association with hypertension, diabetes mellitus, or chronic kidney disease. These findings suggest that although comorbid illnesses are common among patients receiving ATT and may complicate their overall clinical management, they were not independent indicators of DILI severity in our study.
Alcohol consumption was reported by 26% of patients, while 10% were smokers and 1.5% reported other addictions. No significant association was observed between these addictive behaviours and DILI severity. Gaude et al. similarly found no significant association with alcohol consumption, whereas Juganya et al. reported a significant association between alcohol intake and ATT-induced hepatotoxicity. These conflicting observations suggest that the contribution of alcohol may vary according to the amount and duration of consumption, underlying nutritional and hepatic status, and other concurrent risk factors.
The mean BMI in our study was 20.98±2.8 kg/m², with a median of 21.1 kg/m². Approximately 27.7% of patients were underweight, while 32.1% had a normal BMI, 36.5% were overweight, and 3.6% were obese. Low BMI was not significantly associated with DILI severity (p=0.106). Muluken et al. reported a high proportion of patients with DILI having a BMI <18.5 kg/m², whereas Juganya et al. did not find a significant relationship between low BMI and DILI. Although malnutrition has been proposed as a risk factor for ATT-induced hepatotoxicity, the present findings do not demonstrate a significant relationship between low BMI and the severity of liver injury.
On general examination, lymphadenopathy was the most frequent finding, followed by icterus, pallor, pedal oedema, and asterixis. Importantly, icterus showed a significant association with DILI severity (p<0.001), with patients exhibiting icterus more likely to have severe liver injury. Pedal oedema was also significantly associated with DILI severity (p=0.007). In contrast, pallor and lymphadenopathy were not significantly associated with severity. These findings indicate that icterus and oedema may serve as readily identifiable clinical indicators of more severe hepatic dysfunction in patients with ATT-induced DILI. Organomegaly and ascites were relatively uncommon, with the majority of patients having neither significant organomegaly nor ascites.
Among the viral infections evaluated, HIV infection was present in 13% of patients, HBsAg positivity in 5.8%, and anti-HCV positivity in 1.5%. None of the patients tested positive for acute hepatitis A or hepatitis E infection. HIV infection demonstrated a statistically significant association with DILI severity (p=0.021), with HIV-positive patients more likely to experience severe DILI. This finding differs from that of Gaude et al., who did not demonstrate a significant association between HIV infection and ATT-induced DILI severity. The association observed in our study highlights the importance of close clinical and biochemical monitoring of patients with HIV-TB coinfection receiving hepatotoxic antitubercular drugs.
Several other potential risk factors were evaluated. Female gender, low BMI, hepatitis B or C coinfection, alcohol consumption, and drug interactions related to polytherapy were not significantly associated with DILI severity. Among the evaluated risk factors, HIV infection emerged as the significant factor associated with severity. Previous studies have produced heterogeneous findings regarding predictors of ATT-induced hepatotoxicity. Gaude et al. reported significant associations with advanced age and female gender, whereas Juganya et al. identified age and alcohol consumption as significant risk factors. These variations highlight the multifactorial nature of ATT-induced DILI and suggest that the relative contribution of individual risk factors may differ among populations.
Serial laboratory assessment demonstrated substantial biochemical recovery during follow-up. Mean total bilirubin decreased from 3.4 mg/dL at presentation to 0.90 mg/dL at six weeks, while mean direct bilirubin decreased from 2.4 mg/dL to 0.53 mg/dL. Both changes were highly statistically significant. Previous studies have similarly reported elevated bilirubin levels among patients with ATT-induced hepatotoxicity. Gaude et al. reported a median serum bilirubin of 2.3 mg/dL, while Juganya et al. also demonstrated significantly elevated bilirubin among patients with hepatitis. The progressive reduction in bilirubin observed in our study reflects improvement in hepatic function following recognition and management of DILI.
A marked decline in serum transaminases was also observed. Mean AST/SGOT decreased from 412 IU/L at presentation to 33 IU/L at six weeks, while mean ALT/SGPT decreased from 371 IU/L to 32 IU/L. These reductions were highly statistically significant. Gaude et al. reported median AST and ALT values of 346 IU/L and 415 IU/L, respectively, while Juganya et al. also reported substantially elevated transaminase levels in patients with DILI. Similarly, mean ALP in our study decreased from 241 IU/L at presentation to 152 IU/L at six weeks. Taken together, the progressive decline in bilirubin, transaminases, and ALP demonstrates substantial biochemical recovery over the six-week observation period.
Clinical recovery generally preceded complete biochemical recovery. Overall, 91% of patients experienced symptomatic improvement. The median time to symptomatic improvement was one week (IQR: 1–2 weeks), and most symptomatic patients improved within three weeks. In comparison, improvement in LFTs occurred in 91% of patients, with a median time of three weeks (IQR: 3–4 weeks), and most surviving patients demonstrated biochemical improvement within five weeks. Juganya et al. similarly reported recovery from hepatitis within three to four weeks in a substantial proportion of patients. Thus, although clinical symptoms may improve relatively rapidly after appropriate management, normalization of biochemical abnormalities may require a longer period.
Reintroduction of antitubercular drugs represents an important component of management following ATT-induced hepatotoxicity, as prolonged interruption of effective therapy may adversely affect tuberculosis outcomes. In the present study, ATT rechallenge was possible in 89% of patients. Rechallenge was performed sequentially in accordance with the treatment approach adopted in the study. Abbara et al. reported a median interval of 18 days from stopping ATT to beginning reintroduction and a median of 28 days to re-establishment on full-dose therapy. These findings, together with those of the present study, demonstrate that successful reintroduction of antitubercular drugs is feasible in a substantial proportion of patients following recovery from DILI.
Rifampicin rechallenge occurred at a median of five weeks (IQR: 4–5 weeks), with most patients successfully rechallenged by six weeks. Rifampicin could not be reintroduced in patients who died and in one patient who developed rifampicin-induced thrombocytopenia. Isoniazid rechallenge occurred at a median of four weeks (IQR: 3–5 weeks), and most surviving patients could be rechallenged within five to six weeks. Pyrazinamide rechallenge was undertaken in only a small proportion of patients during the study period. Since follow-up was limited to six weeks, subsequent pyrazinamide reintroduction and recurrence of DILI following complete ATT rechallenge could not be adequately evaluated. Longer follow-up would therefore be required to assess the safety and success of complete antitubercular drug reintroduction.
Despite favourable clinical and biochemical recovery in the majority of patients, 12 of 137 patients died, corresponding to an overall mortality of 8.8%. Of these deaths, 3 occurred among patients aged 18–40 years, 4 among those aged 41–60 years, and 5 among patients aged >60 years. Nine of the 12 patients who died were males and three were females. Importantly, 10 of the 12 deaths occurred among patients with severe DILI and the remaining two among those with moderate DILI; no deaths occurred among patients with mild DILI. This distribution suggests that greater severity of liver injury was associated with poorer short-term outcomes. Although most deaths were primarily related to the underlying disease, ATT-induced liver injury may have contributed to the adverse outcomes.
Overall, the present study demonstrates that ATT-induced hepatotoxicity most commonly manifested within the early weeks following initiation of therapy, with nausea, vomiting, and anorexia being the predominant clinical symptoms. Extrapulmonary tuberculosis constituted the majority of cases, and moderate DILI was the most common severity category. HIV infection was significantly associated with greater DILI severity, while icterus and pedal oedema were important clinical findings associated with severe liver injury. Most patients demonstrated substantial clinical and biochemical recovery within six weeks, and reintroduction of antitubercular therapy was possible in the majority. However, mortality was concentrated among patients with severe DILI, emphasizing the importance of early recognition, close monitoring, timely withdrawal or modification of hepatotoxic drugs when indicated, and careful reintroduction of ATT following recovery.
Conclusion
Antitubercular therapy-induced drug-induced liver injury (ATT-DILI) was observed predominantly among younger adults and females, with extrapulmonary tuberculosis being the most common form of TB in the study population. Hepatotoxicity generally developed early after initiation of ATT, with a median time to symptom onset of approximately two weeks. Nausea and vomiting were the most common presenting symptoms, although 12% of patients were asymptomatic, highlighting the importance of appropriate biochemical monitoring during ATT. Among the evaluated clinical findings, icterus and pedal oedema were significantly associated with DILI severity, while HIV infection was the only evaluated risk factor that showed a significant association with severity. Most patients demonstrated favourable short-term outcomes, with 91% showing symptomatic and biochemical improvement and ATT rechallenge being possible in the majority of patients. Nevertheless, the mortality rate was 8.8%, with most deaths occurring among patients with severe DILI. These findings emphasize the importance of early recognition of ATT-induced hepatotoxicity, careful clinical and laboratory monitoring, particularly during the initial weeks of therapy, and timely management and reintroduction of antitubercular drugs to optimize treatment outcomes.
Limitations
The present study had several limitations. The exact biochemical patterns of DILI could not be characterized in detail, and potentially relevant factors such as isoniazid acetylator status, enzyme activity, and genetic susceptibility to hepatotoxicity could not be evaluated because of the unavailability of specific investigations. Patients with multidrug-resistant tuberculosis were not included; therefore, the findings may not be generalizable to patients receiving second-line or other drug-resistant TB regimens. Furthermore, the follow-up period was limited to six weeks, which restricted assessment of complete ATT rechallenge, particularly with pyrazinamide, as well as subsequent recurrence of hepatotoxicity and longer-term clinical outcomes. Finally, differences in study design and patient characteristics between the present prospective observational study and previously published studies, many of which were case-control studies, limited direct comparison and extrapolation of their findings to the present study.
Declarations
Ethical Approval and Consent to Participate
All procedures performed in this case series were conducted in accordance with institutional ethical standards and the principles of the Declaration of Helsinki. Ethical approval was obtained from the appropriate institutional review board where required. Written informed consent was obtained from all patients or their legal guardians prior to the procedures.
Consent for Publication
Written informed consent for publication of clinical details and images was obtained from the patients or their legal guardians. All identifying information has been anonymized to protect patient confidentiality.
Availability of Supporting Data
The data supporting the findings of this study are available from the corresponding author upon reasonable request, subject to institutional and ethical regulations.
Competing Interests
The authors declare that they have no competing interests related to this work.
Funding
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Authors’ Contributions
All authors contributed substantially to the conception, data acquisition, analysis, drafting, and critical revision of the manuscript. All authors have read and approved the final version of the manuscript.