Estimates of all cause mortality and cause specific mortality associated with proton pump inhibitors among US veterans: cohort study

Sir:

Proton pump inhibitor (PPI) safety is of enduring interest to prescribers and patients. However, the recent paper by Xie and colleagues [1] has not contributed substantially to our understanding of apparent concerns. Although it is not unusual for adverse effects of drug classes to only become apparent during post-marketing surveillance, such events are generally confined to one organ or system. Examples include cough with angiotensin converting enzyme inhibitors and sprue-like enteropathy with some angiotensin II receptor blockers.

Compare this with the myriad of apparent adverse outcomes with which PPIs are provisionally linked. [2] Many lack any plausible biological rationale, and are derived from secondary analyses of huge pharmacy claims databases, never designed for this purpose. These almost exclusively retrospective data are prone to bias and confounding that cannot be completely controlled for. Xie et al. used the United States Department of Veterans Affairs database; although large and comprehensive, this cannot capture prescribing decisions for individual patients accurately, such as actual indication(s) for PPI use. Although the authors assert that many patients had “no demonstrable gastrointestinal indications for acid suppression”, it is safe to assume at least some had other appropriate indications; notably prophylaxis against aspirin or non-steroidal anti-inflammatory drug (NSAID)-related upper gastrointestinal tract ulceration or bleeding. Such patients would not necessarily have an additional diagnostic code for one of the conditions that they considered an appropriate indication for PPIs.

The authors report modest increases in mortality and, specifically, increased risk of death from cardiovascular disease, chronic kidney failure, and gastrointestinal cancers. Generalisability of these findings must, however, be questioned; patients were predominantly male, white, and elderly. Other investigators have reported no difference in cumulative incidence of myocardial infarction in new users of PPIs, compared with new users of H2-receptor antagonists, [3] and a recent systematic review found marked inconsistency among published studies regarding associations between PPIs and either acute kidney injury or chronic kidney disease. [4] Potential for substantial confounding was reported, notably concerning NSAID use, some of which may be over the counter and not recorded in their database. Since elderly patients on NSAIDs are at increased risk of upper gastrointestinal tract ulcers and bleeding, PPI prophylaxis is frequently, and appropriately, recommended. However, many NSAIDs are cardiotoxic and nephrotoxic. Regarding upper gastrointestinal tract cancers, there is possible protopathic bias; patients with non-specific gastrointestinal symptoms are more likely to receive PPIs. Xie et al. claim that the lack of excess of peptic ulcer deaths suggests that confounding is unlikely, but PPIs prevent peptic ulcer and its complications, [5] and are associated with reduced mortality from peptic ulcer disease, [6] so this is not as reassuring as the authors claim.

Against all of these weak and inconclusive data must be weighed the results of a large randomised controlled trial (RCT) published recently. [7] Among 17,598 patients randomised to rivaroxaban, aspirin, or the combination, those not already on a PPI were further randomised to pantoprazole 40 mg or placebo once-daily and followed prospectively for a median of three years, resulting in 53,152 patient-years of follow up. Overall 1,244 participants died during follow-up, with mortality rates being similar between the pantoprazole and placebo groups (hazard ratio (HR) = 1.03; 95% CI 0.92 to 1.15). There was also no difference in cancer rates between the pantoprazole and placebo groups (HR = 0.99; 95% CI 0.87 to 1.13). In both cases the upper limit of the 95% CI in this RCT is below the point estimate from Xie et al., suggesting strongly that the “risk” of PPI therapy is overestimated in their paper, and some of the observed association relates to residual confounding. More likely still, is that all of the association between PPI and mortality relates to residual confounding, and that these drugs are not a significant cause of mortality.

PPIs are extremely important for a variety of upper gastrointestinal conditions and also help to reduce upper gastrointestinal complications of NSAIDs and aspirin. The current epidemic of studies linking PPIs to an extremely diverse set of serious outcomes has caused understandable confusion among prescribers, and unwarranted concern among patients; one study estimated that almost 40% of patients with gastro-oesophageal reflux disease in the United States have attempted to stop PPI treatment, usually without discussion with their physician. [8] We agree that PPIs should only be used when medically necessary, but remind readers that this is equally true for every other class of drugs. However, we strongly believe that much of the recent reporting of PPI safety is misplaced, leading to inappropriate termination of treatment with potentially serious outcomes. PPI treatment should not be withheld from patients with a valid indication, and must not be stopped inappropriately in such patients based on misplaced safety concerns.

REFERENCES

1. Xie Y, Bowe B, Yan Y, et al. Estimates of all cause mortality and cause specific mortality associated with proton pump inhibitors among US veterans: cohort study. BMJ 2019;365:l1580.
2. Vaezi MF, Yang YX, Howden CW. Complications of proton pump inhibitor therapy. Gastroenterology 2017;153:35-48.
3. Landi SN, Sandler RS, Pate V, et al. No Increase in risk of acute myocardial infarction in privately insured adults prescribed proton pump inhibitors vs histamine-2 receptor antagonists (2002-2014). Gastroenterology 2018;154:861-873.e6.
4. Kamal F, Khan MA, Molnar MZ, et al. The association between proton pump inhibitor use with acute kidney injury and chronic kidney Disease. J Clin Gastroenterol 2018;52:468-476.
5. Scally B, Emberson JR, Spata E, et al. Effects of gastroprotectant drugs for the prevention and treatment of peptic ulcer disease and its complications: A meta-analysis of randomised trials. Lancet Gastroenterol Hepatol 2018;3:231-241.
6. Lu Y, Sverden E, Ljung R, et al. Use of non-steroidal anti-inflammatory drugs and proton pump inhibitors in correlation with incidence, recurrence and death of peptic ulcer bleeding: An ecological study. BMJ Open 2013;3:e002056.
7. Moayyedi P, Eikelboom JW, Bosch J, et al. Safety of proton pump inhibitors based on a large, multi-year, randomized trial of patients receiving rivaroxaban or aspirin. Gastroenterology 2019;doi.org/10.1053/j.gastro.2019.05.056.
8. Kurlander JE, Kennedy JK, Rubenstein JH, et al. Patients' perceptions of proton pump inhibitor risks and attempts at discontinuation: A national survey. Am J Gastroenterol 2019;114:244-249.