Reply: worsening asthma control in children taking lansoprazole: possible mechanisms.

Abstract

From the Authors: We appreciate the interest of Dr. Sukhovershin, Dr. Ghebremariam, and Dr. Cooke in our recent report about worsening asthma control among children taking lansoprazole who have the CYP2C19 poor metabolizer phenotype (1). The authors correctly point out that we found that these children developed worse asthma symptoms after 5 months of lansoprazole therapy. We speculated that prolonged acid suppression among poor metabolizers alters gastric flora and heightens host susceptibility to respiratory pathogens that are known to affect asthma control. In a previous report from our group (2), we showed that lansoprazole-treated children with the CYP2C19 poor metabolizer phenotype had a significantly higher prevalence of self-reported sore throats and upper respiratory infections during the treatment period. We also found a nonsignificant trend between metabolizer phenotype and worse asthma symptoms after upper respiratory infections. The potential role of proton pump inhibitor (PPI)-induced acid suppression in causing worsening asthma control requires replication and further mechanistic study. We were intrigued by the plausible hypothesis proposed by Sukhovershin and colleagues, which suggests that asymmetrical dimethylarginine (ADMA) becomes increased with PPI treatment. They propose that ADMA, which is a nitric oxide synthase inhibitor, may generate reactive oxygen species and impair immune defenses, leading to worse asthma. ADMA has been shown to be elevated in human asthma and also increases airway responsiveness when administered in a mouse model (3). Among the 115 children who completed our 6-month study and provided fractional exhaled nitric oxide (FENO) measurements before and after treatment, neither lansoprazole treatment (n = 63) nor metabolizer phenotype affected the fold change from baseline of FENO. Fold changes from baseline FENO among placebo-treated, lansoprazole-treated poor metabolizers, and lansoprazole-treated extensive metabolizers were 1.1, 1.0, and 1.1, respectively (P > 0.6 for all pair-wise comparisons). Our results are consistent with past authors, who have also failed to show significant correlations among airway ADMA and FENO (4). We did not measure ADMA in our study. ADMA has been associated with changes in hemodynamic parameters that associate with cardiovascular risk (5, 6). To explore a possible link between PPI treatment and hemodynamic changes in a post hoc analysis, we determined the associations among PPI treatment status, metabolizer phenotype, and key hemodynamic measures before and after the intervention period. Using our original statistical approach, we found that increasing PPI exposure was not associated with changes in blood pressure, heart rate, pulse pressure, mean arterial pressure, mid blood pressure ([sytolic blood pressure + diastolic blood pressure]/2), pulse pressure index, or rate pressure product (data not shown). However, we did see a weak trend between PPI exposure group and changes from baseline in systolic blood pressure (placebo 0.0 vs. lansoprazole-treated poor metabolizer +3.5; P = 0.09). It is biologically plausible for excess PPI exposure to increase susceptibility to asthmagenic pathogens and increase ADMA, leading to altered NOS activity and airway inflammation. However, further research is needed to confirm these speculations. Despite the fact that long-term use of PPIs is common, current US Food and Drug Administration recommendations for dosing is limited to short-term use (up to 12 weeks). We very much agree with Sukhovershin and colleagues that considering PPIs’ multiple known adverse effects and limited research into long-term effects, judicious use of PPIs is warranted and should be conducted with close guidance from a medical professional.