Safety and Complications of Long-Term Proton Pump Inhibitor Therapy: Getting Closer to the Truth
2019; Elsevier BV; Volume: 157; Issue: 3 Linguagem: Inglês
10.1053/j.gastro.2019.07.039
ISSN1528-0012
Autores Tópico(s)Potassium and Related Disorders
ResumoSee "Safety of proton pump inhibitors based on a large, multi-year, randomized trial of patients receiving rivaroxaban or aspirin," by Moayyedi P, Eikelboom JW, Bosch J, et al, on page 682. See "Safety of proton pump inhibitors based on a large, multi-year, randomized trial of patients receiving rivaroxaban or aspirin," by Moayyedi P, Eikelboom JW, Bosch J, et al, on page 682. Inquiring minds want to know: what are the downsides to taking long-term proton pump inhibitors? Patients ask about risks and benefits; their doctors want informed answers. As a gastroenterologist, my patients frequently ask this question. It can be hard to answer succinctly. Postmarketing surveillance is extremely important for information regarding the long-term effects of approved medications. A small sample of important effects discovered in such postmarketing surveillance include major drug–drug interactions, complications during pregnancy, and many others. The randomized trials required for regulatory approval evaluate medication safety and effectiveness, but they are often relatively short in duration, small in size, and may be performed in specialized populations. Side effects may take many months or years to develop, be relatively uncommon, or occur only in certain populations; detecting such events requires long-term evaluations in large, diverse, community-based populations and from longer term randomized trials. The drumbeat of publications on potential complications from proton pump inhibitor use has been impressive, involving almost every organ from brain to bone. Reports of new associations are frequently featured in high-profile national media outlets; each announcement raises additional questions from patients and practitioners. Patients and providers, out of fear, may decrease or stop medications, which may lead to recurrent symptoms, esophagitis, or gastrointestinal hemorrhage. What is real? Do proton pump inhibitors actually cause all these problems? Or are patients who take proton pump inhibitors just more likely, for other reasons, to develop diverse medical disorders? How do we distinguish between known benefits and theoretical risks? The known benefits, demonstrated by randomized trials, include treatment of gastroesophageal reflux, peptic ulcer healing, and prevention of gastrointestinal complications from nonsteroidal anti-inflammatory medications.1Sachar H. Vaidya K. Laine L. Intermittent vs continuous proton pump inhibitor therapy for high-risk bleeding ulcers: a systematic review and meta-analysis.JAMA Intern Med. 2014; 174: 1755-1762Crossref PubMed Scopus (122) Google Scholar, 2Sigterman K.E. van Pinxteren B. Bonis P.A. Lau J. Numans M.E. Short-term treatment with proton pump inhibitors, H2-receptor antagonists and prokinetics for gastro-oesophageal reflux disease-like symptoms and endoscopy negative reflux disease.Cochrane Database Syst Rev. 2013; Cd002095Crossref PubMed Scopus (113) Google Scholar, 3Yang M. He M. Zhao M. et al.Proton pump inhibitors for preventing non-steroidal anti-inflammatory drug induced gastrointestinal toxicity: a systematic review.Curr Med Res Opin. 2017; 33: 973-980Crossref PubMed Scopus (24) Google Scholar The theoretical risks have been reported largely from observational studies—some of which are large and methodologically rigorous, whereas others have study designs highly prone to confounding.4Freedberg D.E. Kim L.S. Yang Y.X. The risks and benefits of long-term use of proton pump inhibitors: expert review and best practice advice from the American Gastroenterological Association.Gastroenterology. 2017; 152: 706-715Abstract Full Text Full Text PDF PubMed Scopus (525) Google Scholar, 5Vaezi M.F. Yang Y.X. Howden C.W. Complications of proton pump inhibitor therapy.Gastroenterology. 2017; 153: 35-48Abstract Full Text Full Text PDF PubMed Scopus (316) Google Scholar Some associations, such as bone fracture, have interesting experimental data suggesting that acid suppression decreases calcium absorption. However, other mechanistic studies do not support a causal relationship6Hansen K.E. Nieves J.W. Nudurupati S. Metz D.C. Perez M.C. Dexlansoprazole and esomeprazole do not affect bone homeostasis in healthy postmenopausal women.Gastroenterology. 2019; 156: 926-934.e926Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar and the clinical associations are found mainly in patients at high risk for fracture for other reasons, increasing the likelihood of uncontrolled confounding. Few associations have been rigorously evaluated within randomized trials and, for those that have, the associations are often not sustained.7Attwood S.E. Ell C. Galmiche J.P. et al.Long-term safety of proton pump inhibitor therapy assessed under controlled, randomised clinical trial conditions: data from the SOPRAN and LOTUS studies.Aliment Pharmacol Ther. 2015; 41: 1162-1174Crossref PubMed Scopus (105) Google Scholar A reported increased risk of cardiovascular events in patients taking clopidogrel plus a proton pump inhibitor from observational data, for example, was not sustained by subsequent analyses of randomized trials, even though there was a plausible biological mechanism.8Ho P.M. Maddox T.M. Wang L. et al.Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome.JAMA. 2009; 301: 937-944Crossref PubMed Scopus (956) Google Scholar, 9O'Donoghue M.L. Braunwald E. Antman E.M. et al.Pharmacodynamic effect and clinical efficacy of clopidogrel and prasugrel with or without a proton-pump inhibitor: an analysis of two randomised trials.Lancet. 2009; 374: 989-997Abstract Full Text Full Text PDF PubMed Scopus (712) Google Scholar Indeed, trial data found that proton pump inhibitor use in this patient population decreased the risk of gastrointestinal events without a detectable increase in cardiovascular risk; thus, avoidance of proton pump inhibitors in this setting was actually associated with overall clinical harm. Residual confounding likely plays some role in these disparate results and,10Landi S.N. Sandler R.S. Pate V. Lund J.L. 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.e866Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar, 11Lochhead P. Hagan K. Joshi A.D. et al.Association between proton pump inhibitor use and cognitive function in women.Gastroenterology. 2017; 153: 971-979.e974Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 12Nguyen L.H. Lochhead P. Joshi A.D. et al.No significant association between proton pump inhibitor use and risk of stroke after adjustment for lifestyle factors and indication.Gastroenterology. 2018; 154: 1290-1297.e1291Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar unfortunately, "positive" findings from small, early studies are also often more memorable (and more often repeated) than later "negative" findings from larger, more definitive evaluations.10Landi S.N. Sandler R.S. Pate V. Lund J.L. 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.e866Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar Recent publications have summarized both the evidence and evidence-based approaches toward teasing out whether proton pump inhibitors cause certain diseases or are only associated with them through other pathways (eg, confounding).4Freedberg D.E. Kim L.S. Yang Y.X. The risks and benefits of long-term use of proton pump inhibitors: expert review and best practice advice from the American Gastroenterological Association.Gastroenterology. 2017; 152: 706-715Abstract Full Text Full Text PDF PubMed Scopus (525) Google Scholar, 5Vaezi M.F. Yang Y.X. Howden C.W. Complications of proton pump inhibitor therapy.Gastroenterology. 2017; 153: 35-48Abstract Full Text Full Text PDF PubMed Scopus (316) Google Scholar Helpful strategies include using the classic criteria for evaluating causation such as the:13Hill A.B. The environment and disease: association or causation?.Proc Roy Soc Med. 1965; 58: 295-300Crossref PubMed Scopus (7701) Google Scholar•strength of the association;•consistency of the findings between studies;•specificity when an outcome happens almost exclusively from a specific exposure;•temporality such that the exposure comes before the outcome;•biological gradient whereby higher exposure doses or longer durations increase risk of the outcome;•biological plausibility for the proposed association;•coherence such as between observed effects and known biology of disease;•experiment such as randomized trials that decrease confounding; and•analogy to similar exposure–disease associations known to be causal. Some associations, such as an increased risk of nutrient deficiencies whose absorption depend, at least partially, on gastric acid production, meet many of the criteria, even if individual-level absolute risk is low.14Lam J.R. Schneider J.L. Zhao W. Corley D.A. Proton pump inhibitor and histamine 2 receptor antagonist use and vitamin B12 deficiency.JAMA. 2013; 310: 2435-2442Crossref PubMed Scopus (397) Google Scholar Other reported associations meet few criteria.6Hansen K.E. Nieves J.W. Nudurupati S. Metz D.C. Perez M.C. Dexlansoprazole and esomeprazole do not affect bone homeostasis in healthy postmenopausal women.Gastroenterology. 2019; 156: 926-934.e926Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar The current study by Moayyedi et al15Moayyedi P. Eikelboom J.W. 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; 157: 682-691Abstract Full Text Full Text PDF PubMed Scopus (292) Google Scholar helps to evaluate potential causality for several reported proton pump inhibitor associations with one of the weightiest forms of causal evidence: experimental data. This massive undertaking included >17,000 patients from 33 countries who were randomized to pantoprazole versus placebo, followed for a median of approximately 3 years, and evaluated prospectively for potential complications. The study found an increased risk of enteric infections among pantoprazole users, a result found in both the intention-to-treat and "as-treated" analyses, which excluded people who stopped their medications. This association makes sense—stomach acid markedly decreases bacterial load in food, so decreasing stomach acid may increase the risk of enteric bacterial infections. The authors found no increased risk for several of the most feared associations previously reported, such as cardiovascular disease, kidney disease (directly measured using estimated glomerular filtration rate), dementia, pneumonia, fracture, and all-cause mortality. The results update the level of quality data available from recent consensus efforts (Table 1).Table 1Summary of Evidence for Potential PPI-Associated Effects (updated from American Gastroenterological Association review)Potential adverse effectPrior types of studiesPrior Threats to validityPrior overall quality of evidenceNew data for 3 years median follow-upEffect estimate (Odds ratios and 95% confidence intervals)Kidney disease•Observational only•Modest effect size•Residual confounding would bias toward harm•Absence of dose-response effectVery lowNo association1.17 (0.94–1.45)Dementia•Observational only•Modest effect size•Residual confounding would bias toward harmVery lowNo association1.20 (0.81–1.78)Bone fracture•Observational only•Inconsistent results•Modest effect size•Residual confounding would bias toward harmLow or very lowNo association0.96 (0.79–1.17)Myocardial infarction•Observational•RCT•Results differ between RCTs and observational studies•Secondary analysis of RCT data•Modest effect size•Residual confounding would bias towards harmVery lowNo association1.04 (0.93–1.15)amyocardial infarction, stroke or cardiovascular death.Small intestinal bacterial overgrowth•Observational•Crossover•Sparse data•Residual confounding would bias toward harm•Protopathic biasLowNot evaluatedSpontaneous bacterial peritonitis•Observational only•Modest effect size•Residual confounding would bias toward harmVery lowNot evaluatedClostridium difficile infection•Observational only•Modest effect size•Residual confounding would bias toward harmLowIncreased risk for other enteric infections, trend for Clostridium difficile2.26 (0.70–7.34) C.diff; 1.33 (1.01–1.75) Other infectionPneumonia•Observational•RCT•Results differ between RCTs and observational studies•Secondary analysis of RCT data•Modest effect size•Absence of dose-response effect•Residual confounding would bias towards harm•Protopathic biasVery lowNo association1.02 (0.87–1.19)Micronutrient deficiencies•Inconsistent results•Modest effect size•Absence of dose-response effect•Residual confounding would bias toward harmLow or very lowNot evaluatedGastrointestinal malignancies•Observational•RCT•Results differ between RCTs and observational studies•RCTs use surrogate outcomes•Modest effect size•Residual confounding would bias towards harm•Confounding by indication and protopathic biasVery lowNo association1.04 (0.77–1.40)NOTE. Updated from American Gastroenterological Association review.4Freedberg D.E. Kim L.S. Yang Y.X. The risks and benefits of long-term use of proton pump inhibitors: expert review and best practice advice from the American Gastroenterological Association.Gastroenterology. 2017; 152: 706-715Abstract Full Text Full Text PDF PubMed Scopus (525) Google Scholara myocardial infarction, stroke or cardiovascular death. Open table in a new tab NOTE. Updated from American Gastroenterological Association review.4Freedberg D.E. Kim L.S. Yang Y.X. The risks and benefits of long-term use of proton pump inhibitors: expert review and best practice advice from the American Gastroenterological Association.Gastroenterology. 2017; 152: 706-715Abstract Full Text Full Text PDF PubMed Scopus (525) Google Scholar Are these results helpful for patients and providers? Yes. They provide high-quality, prospectively collected data from patients randomly assigned to the exposed versus unexposed groups. This type of trial data is extremely useful for decreasing the likelihood that any associations identified are solely due to confounding. Does this study settle the question regarding any causation or any harm? No. The study was not powered to detect all of the safety outcomes evaluated. The study also did not directly measure certain associations, such as decreases in micronutrient levels (eg, vitamin B12, magnesium, or iron) or the impact of extremely long-term exposures. For some end points, such as cognition, the study did not use sensitive, validated measures to identify small potential changes. However, as the authors note, the current study's sample size would have been expected to detect effects of similar magnitude to those previously reported using similar end points, especially given many prior associations used diagnoses from administrative databases (which would be expected to identify greater disease severity). Reassuringly, the current study also generally lacked any strong trends for these outcomes. Thus, even substantially larger sample sizes would be unlikely to detect the associations that varied by more than chance. The study's follow-up duration is also within most previously reported exposure–complication windows. For example, a recent observational study of 5254 proton pump inhibitor users versus a smaller number of never users found an increased risk of kidney disease after only 3 month's medication exposure.16Rodriguez-Poncelas A. Barcelo M.A. Saez M. Coll-de-Tuero G. Duration and dosing of proton pump inhibitors associated with high incidence of chronic kidney disease in population-based cohort.PLoS One. 2018; 13e0204231Crossref Scopus (34) Google Scholar This finding means that the current study's multiyear duration, in a larger population, would have been expected to detect comparable causal exposure–disease associations. So inquiring minds want to know: what are the downsides to taking long-term proton pump inhibitors? The current study would indicate, from the strongest study design available, that if you start a diverse group of patients on a proton pump inhibitor and continue it for a few years, there is an increased risk of gastrointestinal infections and no detectable excess risk for several other potentially important clinical events. These findings support acid suppression for persons with clear indications, such as bleeding prophylaxis, ulcer healing, esophagitis, and gastroesophageal reflux symptoms refractory to other measures; that is, those who would likely benefit from demonstrated randomized trial data. Given known problems with overprescribing and overuse, patients and clinicians should maintain appropriate vigilance in prescribing acid suppression only to persons with defined indications and at the lowest effective dose and duration. Safety of Proton Pump Inhibitors Based on a Large, Multi-Year, Randomized Trial of Patients Receiving Rivaroxaban or AspirinGastroenterologyVol. 157Issue 3PreviewProton pump inhibitors (PPIs) are effective at treating acid-related disorders. These drugs are well tolerated in the short term, but long-term treatment was associated with adverse events in observational studies. We aimed to confirm these findings in an adequately powered randomized trial. Full-Text PDF Covering the CoverGastroenterologyVol. 157Issue 3PreviewA double-blind, placebo-controlled trial in healthy volunteers showed that oral supplements of Bifidobacterium breve Bif195 decrease risk of aspirin-induced small intestinal enteropathy. Full-Text PDF Proton Pump Inhibitor Therapy and Fracture Risk: Discrepancy of Results Between Observational and Interventional StudiesGastroenterologyVol. 158Issue 4PreviewIn their timely and crucial placebo-controlled randomized trial with 53,152 patient-years of follow-up for a median of 3.01 years (mean age 67.6 years, 78% male), Moayyedi et al1 showed that proton pump inhibitor (PPI) therapy using pantoprazole (40 mg daily) was independent of bone fracture risk; odds ratios (95% confidence intervals [P values], incident events [%] of pantoprazole vs placebo) were 0.96 (0.79-1.17 [.71], 203 [2.3%] vs 211 [2.4%]) in the analysis of all participants and 0.89 (0.71-1.13 [.35], 136 [2.0%] vs 150 [2.2%]) in the analysis that excluded those who permanently discontinued either of them. Full-Text PDF
Referência(s)