Response

Abstract

We thank Dr Corrao and colleagues for their interest in our meta-analysis. We can only agree that our results were limited by heterogeneity among studies, a situation that we acknowledged at least on three separate occasions in our article. Heterogeneity existed not only in the diagnostic criteria of idiopathic pulmonary fibrosis (IPF), but also in those of gastroesophageal reflux disease (GERD). This led us to proceed with subgroup analyses based on hypotheses that we had formulated a priori. Unfortunately, none of these analyses satisfactorily explained the heterogeneity. Dr Corrao and colleagues conducted, from the studies that we included, a meta-analysis by considering only those who received a diagnosis of IPF according to the American Thoracic Society/European Respiratory Society criteria.1American Thoracic Society; European Respiratory SocietyIdiopathic pulmonary fibrosis: diagnosis and treatment: international consensus statement.Am J Respir Crit Care Med. 2000; 161: 646-664PubMed Google Scholar This is similar to one of our subgroup analyses that considered only the studies in which the diagnosis of IPF was based on an objective method (e-Fig 2G in our article).2Bédard Méthot D. Leblanc É. Lacasse Y. Meta-analysis of gastroesophageal reflux disease and idiopathic pulmonary fibrosis.Chest. 2019; 155: 33-43Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar In this analysis, heterogeneity remained significant. Also, based on a novel methodology called “trial sequential analysis,” Dr Corrao and colleagues suggested that our meta-analysis may have falsely concluded that GERD and IPF are associated. Although this may be true, we reiterate that the major finding of our meta-analysis is not that GERD and IPF may be associated. Rather, its highlight is that this association is overestimated because it is most likely confounded by smoking, a problem that has been largely ignored so far in the literature. We provided several arguments in support of this conclusion, including the results of two meta-regressions. First, after adjusting for the ratio of proportions of smokers and ex-smokers in case subjects and control subjects, the association between GERD and IPF was not statistically significant. Second, we found a significant correlation between the ratio of proportions of smokers and ex-smokers in case subjects and control subjects and the strength of the association between GERD and IPF reported in the individual studies. In other words, the larger the imbalance between the proportion of smokers and ex-smokers among patients with IPF and their control subjects, the larger the measure of association between GERD and IPF (Fig 4 in our article). We agree with Dr Corrao and colleagues that additional studies should be conducted and that selection bias must be minimized. However, our concern is much more about the issue of confounding. Smoking is associated with both GERD and IPF but is not an intermediate link in the chain of causation between the two. It therefore represents a true confounder3Grimes D.A. Schulz K.F. Bias and causal associations in observational research.Lancet. 2002; 359: 248-252Abstract Full Text Full Text PDF PubMed Scopus (924) Google Scholar that must be considered in future studies. Meta-analysis of Gastroesophageal Reflux Disease and Idiopathic Pulmonary FibrosisCHESTVol. 155Issue 1PreviewThe relationship between gastroesophageal reflux disease (GERD) and idiopathic pulmonary fibrosis (IPF) is controversial. Current guidelines recommend that clinicians use regular antacid treatment, while two recent meta-analyses of antacid therapy in IPF were inconclusive. The objective of this study was to examine the evidence regarding the association between GERD and IPF through a systematic review and a meta-analysis, with special reference to the methodologic quality of the observational studies. Full-Text PDF Gastroesophageal Reflux Disease and Idiopathic Pulmonary Fibrosis: No Data for Supporting a Relationship After a Systematic ReviewCHESTVol. 156Issue 1PreviewWe read with great interest in CHEST (January 2019) the systematic review by Bédard Méthot et al1 about the relationship between gastroesophageal reflux disease and idiopathic pulmonary fibrosis (IPF).1 However, in our opinion, this systematic review has some methodologic issues that have to be addressed. One of the major issues is the substantial heterogeneity shown by the included meta-analysis. This heterogeneity may be described from both a statistical (I2, 86%; P < .00001) and a clinical point of view. Full-Text PDF