Clinical Question: Do certain drugs increase the risk of sudden death when coadministered with erythromycin?

Setting: Population-based

Study Design: Cohort (retrospective)

Synopsis: Erythromycin is known to prolong cardiac repolarization and may, in very rare cases, cause torsades de pointes, a fatal arrhythmia. The primary metabolic pathway for erythromycin and other macrolides involves the cytochrome P450 3a (CYP3A) pathway, which can be inhibited by a number of commonly used medications. The latter include azole antifungals such as ketoconazole, itraconazole, and fluconazole, as well as diltiazem, verapamil, and troleandomycin.

In this study, Tennessee Medicaid enrollees who had experienced a sudden cardiac death between 1988 and 1993 were identified from Medicaid records, death certificates, and a review of medical encounter records. These patients were divided into groups based on their antibiotic use: no use of erythromycin or amoxicillin (1,126,013 person-years), former use of erythromycin (111,779 person-years), current use of erythromycin at the time of sudden cardiac death (5,305 person-years), and current use of amoxicillin at the time of sudden cardiac death (6,846 person-years). Seventy percent of the cohort were women, and the average age was 45 years; 25 percent were older than 65 years.

The overall risk of sudden cardiac death was higher for patients who were current users of erythromycin (adjusted incidence rate ratio = 2.01; 95 percent confidence interval [CI], 1.1 to 3.8), but not for former or current users of amoxicillin. This amounts to approximately one additional sudden cardiac death per 1,000 person-years of use. The risk was even higher in patients who were taking erythromycin and one of the CYP3A inhibitors listed above (adjusted incidence rate ratio = 5.35; 95 percent CI, 1.7 to 16.6).

Bottom Line: The combination of erythromycin and a CYP3A inhibitor (most often cardizem or verapamil) increases the risk of sudden cardiac death approximately fivefold. It should be avoided in clinical practice. (Level of Evidence: 2b)