New study further questions PPIs superiority over H2RAs for stress ulcer prophylaxis

Abstract

One of the principal strategies used to mitigate gastrointestinal bleeds (GIBs) in critically ill patients is to use stress ulcer prophylaxis (SUP) treatments, which mainly include proton pump inhibitors (PPIs) and histamine 2 receptor antagonists (H2RAs). Over the last two decades there has been considerable flux in the consensus between clinical guidelines as to which agent is preferred. To add to the debate and the growing body of research, a new retrospective study published in August 2021 in Pharmacotherapy links PPI SUP treatment with higher in-hospital mortality compared with H2RA treatment. Boyd and colleagues conducted a retrospective analysis that compared outcomes between PPIs and H2RAs in ICU patients who had been mechanically ventilated for more than 24 hours. The analysis included ICD-9 and ICD-10 codes from electronic health records of over 3,800 patients from the ICUs of the University of Colorado Hospital between September 2015 and 2019. The primary outcomes of interest were clinically important GIBs and hospital mortality. GIBs were not included unless they occurred 48 hours after ICU admission in order to rule out non-nosocomial causes. The authors discovered that there was no statistically significant difference in GIBs between PPIs and H2RAs. In fact, the incidence rate for GIBs in both groups was practically non-existent with no GIB occurring in the H2RA group and only 7—or 0.34%—occurring in the PPI group. Researchers also found that there was a statistically higher incidence of hospital mortality for patient receiving PPIs: a 23% relative increase with a number needed to harm of about 8. Two secondary outcomes also yielded a statistically significant increased occurrence of thrombocytopenia and delirium in the PPI group. The comparison groups of the authors’ study demonstrated significant imbalances in baseline characteristics including sex, H. pylori history, antibiotic use, previous hospital admissions, ICU types (e.g., burn, cardiac, neuro, surgical, etc.), admission characteristics, and medication exposure in the ICU (i.e., anticoagulants, steroids, NSAIDs, etc.). These differences all had the capacity to confound the initial results. The authors attempted to control for these differences by using a multivariate logistic regression analysis. This additional analysis demonstrated that hospital mortality remained significantly more likely in the PPI group despite adjustment for the aforementioned baseline confounders. The strength of the association between PPIs as an independent variable and hospital mortality is perhaps concerning (r2 = 0.77). Unfortunately, this is not the first study to find that PPIs are linked to a higher risk of mortality when used as SUP. A systematic review, published December 2021 in the Annals of Pharmacotherapy, pooled data from 6 other similar analyses comparing H2RAs and PPIs. Researchers found a 27% increase in mortality risk with PPIs, as well, when compared to H2RAs. However, due to a dearth of high quality randomized controlled trials, a clear causal relationship has yet to be solidly proven. An explanation for the association between PPIs as SUP and mortality remains relatively unclear, but Boyd and colleagues do remark that “while other studies failed to show plausible explanations for this association, we demonstrated thrombocytopenia and delirium both occurred at higher rates in the PPI group.” Other studies tend to link PPI use and mortality risk with increased rates of Clostridioides difficile and nosocomial pneumonia. This mortality finding shouldn’t distract from Boyd and colleague’s other finding, however. The lack of GIBs in both groups is notable, as well. This raises the question of whether or not SUP treatments are actually necessary after this 48-hour window. Similarly, an observational study published in 2015 in Intensive Care Medicine by Krag and colleagues found low rates of GIB 48 hours after ICU admission. While the true nature of the mortality association with PPIs continues to evolve, hospital pharmacists should be vigilant for superfluous acid suppression therapy after discharge from the ICU. There is a lack of evidence to support SUP for low-risk patients in the ICU, in nonICU units, or in the community if there is an absence of clear risk factors. Boyd and colleagues found that 31% of their patients continued acid suppression therapy on ICU discharge and 5% continued after leaving the hospital. Other studies indicate these numbers can reach as high as 60% and 35%, respectively. Hospital pharmacists have an opportunity to discontinue these medications on ICU transfer or discharge in order to eliminate what would otherwise be a net risk to the patient—be it pneumonia, osteoporosis, C. diff, or perhaps even mortality.