To the Editor: Conflicting results have been recently published concerning drug interaction between atazanavir (ATV) and antacid treatments, especially proton pump inhibitors (PPIs). In the August 1, 2005 issue of this journal, Khanlou and Farthing1 summarized the results they collected in 10 HIV clinics in Los Angeles, saying that the use of ATV, even boosted with ritonavir (RTV), with PPIs should be avoided. These data require some comments. In this study, the prospective method of plasma collection is not clearly defined. Were the patients interviewed by a physician on the treatment they were really taking? Did the nurse who drew the blood samples ask the patients when they had taken their last dose of ATV and antacid treatment? The median ATV minimum concentrations (Cmins) reported are not in keeping with values observed in recent published studies and in our experience on HIV-infected patients treated with boosted ATV (Table 1).TABLE 1: Median ATV Cmin in Patients Treated With Boosted ATV (300/100 mg/d)Because of its chemical structure, ATV absorption is highly dependent on an appropriate gastric pH. The potential interaction between ATV and PPIs or histamine-2 receptor blockers (H2RBs) is expected to reduce ATV exposure. In the presence of an H2RB, however, the median ATV Cmin reported in this study was 2-fold higher than values observed in the previously reported series (performed in patients without antacid treatment). Such a high ATV Cmin was found in only 1 study on healthy volunteers.2 In presence of PPIs, the median ATV Cmin was found to be strictly comparable to those observed in the previous studies, where a significant reduction should be expected. These results could be explained by too short a time between blood sample collection and drug intake. Assessing the time between ATV intake and blood collection precisely is fundamental in such a study. Only ATV plasma concentrations measured 24 ± 4 hours (Cmin) after the last intake should be taken into account in patients who have received a stable ATV regimen for at least 1 week before blood sampling. Treatments received by the patients included in this study should be detailed. The proportion of patients receiving ATV boosted with RTV may indeed highly influence the Cmin values observed. The results collected in such patients should therefore be distinguished from those collected in patients receiving ATV without RTV. In our prospective cohort of 184 patients, median Cmins measured at 24 ± 4 hours were 499 [26-2205] ng/mL and 99 [12-1159] ng/mL among the 141 patients treated with 300 mg of ATV plus 100 mg of RTV and the 43 patients receiving 400 mg of ATV alone, respectively (P < 0.0001). The combination of ATV with another protease inhibitor or nonnucleoside reverse transcriptase inhibitor may lead to significant drug interactions and should also be noted. During PPI treatment, the time during which the intragastric pH is above 4.0 increases with the acid-suppressive dose, and pH can vary widely during the day in a given individual. Doses of PPIs may affect ATV absorption differently and have to be precisely described. The time interval between ATV and antacid drug administration may also play a role. This is a possible confusion bias if patients did not state the time of their last drug intake accurately. In our cohort, we prospectively compared ATV Cmin in HIV-infected patients receiving ATZ plus RTV, with or without a PPI, to assess this specific drug interaction in clinical practice. Among the first 92 patients enrolled, 13 stated that they were taking a PPI. Among these 13 patients, 10 were using omeprazole (20 mg/d, except for 1 patient who was taking 40 mg/d) and 3 were using rabeprazole (30 mg/d). Median ATV trough concentrations were 551 ng/mL (range: 203-1976 ng/mL) in the patients receiving PPIs and 469 ng/mL (range: 65-1944 ng/mL) in the group without PPIs. Results showed that PPI treatment did not affect ATV Cmin values in our patients.3 A recent case report with pharmacokinetic study showed no interaction between boosted ATV and lansoprazole. Another study showed no clinical or immunovirologic failures in 14 patients treated with ATV and PPIs after a 6-month follow-up.4,5 Our findings are in keeping with these results. Our conclusion is therefore completely different from that of Khanlou and Farthing1 because we observed that PPIs seem to be compatible with boosted ATZ therapy in clinical practice. It is nonetheless recommended to monitor ATV drug concentrations in case of concomitant antacid treatment. Jean-Baptiste Guiard-Schmid, MD* Jean-Marie Poirier, MD† Philippe Bonnard, MD* Jean-Luc Meynard, MD† *Service des Maladies Infectieuses et Tropicales Hôpital Tenon †Service de Pharmacologie Faculté de Médecine St. Antoine and ‡Service des Maladies Infectieuses et Tropicales Hôpital St. Antoine Université Pierre et Marie Curie (Paris VI) Paris, France