Potential protective effect of PPI in COPD.
Farha R, Usman M, Gandolfo C, Smith A, and Ouellette D. Potential protective effect of PPI in COPD. Chest 2016; 150(4):833A.
Associations of an increased incidence of respiratory disease in the presence of gastroesophageal reflux disease (GERD), and improvements of respiratory disease with treatment for GERD, have been reported. Respiratory conditions associated with GERD or its treatment include chronic cough, asthma, and interstitial lung disease. Such associations between chronic obstructive pulmonary disease (COPD) and GERD have not been reported. We chose to retrospectively study the severity of airway obstruction as measured by pulmonary function testing (PFT) in a cohort of COPD patients with respect to the use of proton pump inhibitors (PPI). METHODS: Single-center retrospective analysis of patients with documented obstructive disease by PFT in 2014. A total of 1445 patients were identified as having an obstructive pattern. We selected a cohort of 647 patients who self-reported a diagnosis of COPD and had clinical findings consistent with this diagnosis. We compared patients who had been prescribed PPI (PPI group, n=383) to those who had not (non-PPI group, n=264). Univariate comparisons were by Chi-square analysis for dichotomous variables and two-sided student's t-test for continuous variables. SPSS 20 software was used for statistical analysis. A p value of 0.05 was considered significant. Reported values are mean+/- standard deviation. RESULTS: The PPI group and non-PPI group were not different in terms of gender (PPI group 157/383 females; non-PPI group 127/264 females; p=0.073), body mass index (PPI group 27.9+/-7.4; non-PPI group 26.9+/-6.9; p=0.085), or age (PPI group 68.9+/-10.3; non-PPI group 68.2+/-10.8; p=0.407). The FEV1/FVC for the PPI group was 48.1+/-12.0 compared to the non-PPI group 45.3+/-12.1 (p=0.004). The percent predicted FEV1 for the PPI group was 51.2+/-19.2 compared to 46.7+/-19.0 for the non-PPI group (p=0.003). In a multivariate analysis with a model that included BMI, gender, age, and percent predicted FEV1, only the latter was associated with PPI treatment (p=0.022). CONCLUSIONS: Patients with PPI prescription in our study had less severe airway obstruction than patients who were not prescribed these medications. Multiple reasons for this association could be postulated. Different co-morbidity profiles or insurance coverages could exist between groups. Patients with more severe COPD might take more medications, limiting the availability of medications targeting symptom relief. Finally, PPI use could ameliorate airway obstruction in smoking-related lung disease. Potential mechanisms for a protective effect of PPI in COPD might implicate reduction of gastric acid aspiration, or a primary effect of the medication on lung disease pathophysiology. Further research evaluating PPI use in COPD is warranted.