Study objective: To compare the rates of torsades de pointes associated with ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin, and moxifloxacin administration.
Design: Retrospective database analysis.
Intervention: Evaluation of reported rates of torsades de pointes in patients who received these quinolones between January 1, 1996, and May 2, 2001.
Measurements and main results: In the United States, 25 cases of torsades de pointes associated with these quinolones (ciprofloxacin 2, ofloxacin 2, levofloxacin 13, gatifloxacin 8, moxifloxacin 0) were identified. Ciprofloxacin was associated with a significantly lower rate of torsades de pointes (0.3 cases/10 million prescriptions, 95% confidence interval [CI] 0.0-1.1) than levofloxacin (5.4/10 million, 95% CI 2.9-9.3, p<0.001) or gatifloxacin (27/10 million, 95% CI 12-53, p<0.001 for comparison with ciprofloxacin or levofloxacin). When the analysis was limited to the first 16 months after initial U.S. approval of the agent, the rates for levofloxacin (16/10 million) and gatifloxacin (27/10 million) were similar (p>0.5).
Conclusion: Levofloxacin should be administered with caution in patients with risk factors for QT prolongation. Gatifloxacin should be avoided in the same patient population, and the recommended dosage of 400 mg/day should not be exceeded.