Role of P-glycoprotein inhibition for drug interactions:: Evidence from in vitro and pharmacoepidemiological studies

Objectives: We determined in vitro the potency of macrolides as P-glycoprotein inhibitors and tested in hospitalised patients whether coadministration of P-glycoprotein inhibitors leads to increased serum concentrations of the P-glycoprotein substrates digoxin and digitoxin. Methods: In vitro, the effect of macrolides on polarised P-glycoprotein-mediated digoxin transport was investigated in Caco-2 cells. In a pharmacoepidemiological study, we analysed the serum digoxin and digitoxin concentrations with and without coadministration of P-glycoprotein inhibitors in hospitalised patients. Results: All macrolides inhibited P-glycoprotein-mediated digoxin transport, with concentrations producing 50% inhibition (IC50) values of 1.8, 4.1, 15.4, 21.8 and 22.7 mu mol/L for telithromycin, clarithromycin, roxithromycin, azithromycin and erythromycin, respectively. Coadministration of P-glycoprotein inhibitors was associated with increased serum concentrations of digoxin (1.3 +/- 0.6 vs 0.9 +/- 0.5 ng/mL, p < 0.01). Moreover, patients receiving macrolides had higher serum concentrations of cardiac glycosides (p < 0.05). Conclusion: Macrolides are potent inhibitors of P-glycoprotein. Drug interactions between P-glycoprotein inhibitors and substrates are likely to occur during hospitalisation.