Proton pump inhibitor use is associated with increased mortality in african american COVID-19 patients

ABSTRACT

Rationale Coronavirus disease 2019 (COVID-19) has disproportionally affected African Americans (AA), with underlying medical conditions and socioeconomic determinants of health believed to be major contributors (Price-Haywood et al, 2020). The use of proton pump inhibitors (PPIs) was recently found to be associated with increased risk and worse outcomes of SARS-CoV-2 infection (Almario et al, 2020;Luxenberger et al, 2020;Lee et al, 2020). Methods We performed a retrospective cohort analysis of patients hospitalized for SARS-CoV-2 in an integrated health system of six hospitals between March 15 and August 15, 2020. To determine predictive factors of mortality, a set of 17 covariates were selected on the basis of clinical relevance to COVID-19 outcomes. The indications for medication use were evaluated in a subset of research patients. Results In 694 hospitalized COVID-19 patients (median age = 58 yr, 46% men, 65% AA), an overall mortality rate was 17.4% (121/694). Logistic regression analysis identified age (aOR=1.66 per decade, p<0.001), race other than AA or white (aOR=3.03, p=0.002), cancer (aOR=2.22, p=0.008), diabetes (aOR=1.95, p=0.003), anti-HTN (aOR=0.46, p=0.001) and PPI use (aOR=2.72, p<0.001) as predictors of mortality. There was no significant mortality difference observed with the use of H2 receptor antagonists. Moreover, PPI use was associated with higher mortality risk in AA (aOR=4.16, 95% CI = 2.28 to 7.59) than in non-AA patients (aOR=1.62, 95% CI = 0.82, 3.19, p=0.04 for interaction) (see Figure 1). The prevalence of PPI use in African Americans (32.6%147/451) and non-AA patients (32.9%, 80/243) were comparable. No other associations were found to differ between the two groups. Indications for PPI use in 31 of 87 research participants were erosive esophagitis / recent hemorrhage (13%), non-erosive GERD symptoms (45%) and NSAID prophylaxis (42%). Conclusion PPIs are frequently used medications with a growing list of complications. Their use is associated with significant mortality risk in AA COVID-19 patients, and should be reassessed during the pandemic. This risk was not seen in patients who received H2 receptor antagonists. The gastrointestinal tract is a potential site of SARSCoV- 2 entry and replication and this association warrants further study.