Which fluoroquinolone is safer when combined with bedaquiline for tuberculosis treatment: evidence from FDA Adverse Event Reporting System database from 2013 to 2024.

Objective: To investigate which fluoroquinolone is safer when combined with bedaquiline for tuberculosis treatment by using the FDA Adverse Event Reporting System (FAERS) database. Methods: We selected data from the first quarter (Q1) of 2013 to the second quarter (Q4) of 2024 from the FDA FAERS database for disproportionality analysis. Signal detection was conducted using the Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Empirical Bayesian Geometric Mean (EBGM). Results: This study analyzed 12, 303, 879 reports from the FAERS database, including 722 reports related to the combination of bedaquiline and levofloxacin (with 2,723 adverse events) and 573 reports related to the combination of bedaquiline and moxifloxacin (with 2,233 adverse events). For the bedaquiline-levofloxacin regimen, these reports were categorized into 100 preferred terms (PTs) and 24 System Organ Classification (SOCs). The three most common SOCs were hepatobiliary disorders (n = 128, ROR 5.79, PRR 5.56, IC 2.48, EBGM 5.56), blood and lymphatic system disorders (n = 217, ROR 5.04, PRR 4.72, IC 2.24, EBGM 4.71), and metabolism and nutrition disorders (n = 185, ROR 3.44, PRR 3.27, IC 1.71, EBGM 3.27). In terms of PTs, the three strongest signals were portal fibrosis (ROR 330.64), hepatitis C RNA increased (ROR 301.24), and toxic optic neuropathy (ROR 238.11). Reports of prolonged QT interval on ECG (125 cases) and anemia (130 cases) were significantly more frequent than other PTs. For the bedaquiline-moxifloxacin regimen, these reports were categorized into 85 preferred terms (PTs) and 24 System Organ Classification (SOCs). The three most common SOCs were hepatobiliary disorders (n = 141, ROR 7.9, PRR 7.47, IC 2.9, EBGM 7.46), ear and labyrinth disorders (n = 40, ROR 4.03, PRR 3.97, IC 1.99, EBGM 3.97), and cardiac disorders (n = 141, ROR 3.08, PRR 2.95, IC 1.56, EBGM 2.95). The three strongest PT signals were chronic pyelonephritis (ROR 563.29), bronchopleural fistula (ROR 314.86), and toxic neuropathy (ROR 187.11). Prolonged QT interval on ECG (152 cases) remained the most frequently reported PT. In both treatment regimens, individuals under 45 years of age experienced a higher frequency and variety of AEs, indicating the need for enhanced monitoring. For those over 45, particular attention should be given to ECG changes, especially in men. Finally, some PTs with extremely high signal strength, such as chronic pyelonephritis (ROR 563.29), hepatitis C RNA increased (ROR 301.24), and bronchopleural fistula (ROR 301.24), may represent rare adverse events associated with the combination of bedaquiline-fluoroquinolone. Conclusion: Our study suggests that the safety profile of bedaquiline combined with moxifloxacin does not appear superior to that of bedaquiline combined with levofloxacin in terms of cardiac, hepatic, and neurological effects. Therefore, in the BPaLM regimen, considering the substitution of moxifloxacin with levofloxacin may be worthwhile if their efficacy is proven to be similar. Increased monitoring may be required for individuals under 45 years of age and male MDR-TB patients.