Setting: Concomitant feeding and administration of antituberculosis medication has been proposed to increase compliance (by decreasing pyrazinamide associated nausea) and improve nutritional status. Food may however decrease the oral bioavailability of rifampicin and isoniazid.
Objective and methods: A triple-crossover pharmacokinetic study in 27 patients with tuberculosis (15 males and 12 females) compared the bioavailability of rifampicin, isoniazid and pyrazinamide without food (control) with that when taken with a high carbohydrate (CHO) or high lipid (LIPID) diet.
Results: the CHO diet decreased isoniazid bioavailability. The maximum measured drug concentration (Cm) was decreased by 20% (P = 0.0002) and the area under the concentration-time curve to 8 h (AUC8) by 19% (P = 0.01). The CHO diet increased the time to maximum measured drug concentration (Tmax) for rifampicin by 21% (P = 0.03). The LIPID diet decreased the Cm of isoniazid by 9% (P = 0.03). Individual patient bioavailability on each meal was compared to the no-food control. A decrease of Cm or AUC8 of greater than 20% was considered significant. The bioavailability of isoniazid and rifampicin was decreased by food in a high percentage (33-56%) of patients.
Conclusion: Concomitant feeding may thus have an important adverse effect on the therapy of tuberculosis and the desirability of this practice is called into question.