Do Proton Pump Inhibitors Increase Risk for Dementia?

Case: Ms. P is a 76 year old lady with a past medical history of GERD, osteoporosis status post right hip fracture, C. diff infection in 2011, uterine fibroids status post TAH-BSO, who comes in for a follow-up visit. Her medications include omeprazole 40 mg daily, vitamin D 800 IU daily, alendronate 70 mg weekly. Physical exam is unremarkable, except for BMI of 18. During medication reconciliation, she asks, “I heard that my omeprazole can cause dementia…”

I recalled the recent NY Times article about heartburn drugs related to increased risk of dementia and wondered if that could be possible. I told Ms. P, “I will look into that and let you know if you need to be concerned about that.”

The question was: In older patients, is the exposure to proton pump inhibitor (PPI) medications as compared to no exposure associated with increased risk of dementia? In order to answer this question, I performed a search for “PPI and dementia risk” in PubMed. I found two recent studies to review.

In 2015, Haenish and colleagues published the results of the German Study on Aging, Cognition and Dementia in elderly primary care patients (AgeCoDe). This was a longitudinal, multicenter cohort study involving 3,327 community-dwelling persons aged 75 years or older without any dementia at baseline. There were four 1.5 year follow up points with an overall follow up time of 6 years. Using time-dependent Cox regression, the authors found that PPI users had a significantly increased risk of any dementia [Hazard ratio (HR) 1.38, 95% confidence interval (CI) 1.04-1.83] and Alzheimer’s disease (HR 1.44, 95% CI 1.01-2.06) compared with nonusers. This risk persisted after adjusting for potential confounding factors: age, sex, education, ApoE4 allele, polypharmacy, depression, DM, ischemic HD and stroke.

In 2016, the same group published results of a larger prospective cohort study using routine claims data from the largest mandatory public health insurer in Germany. The study included 73,679 elderly patients, aged 75 years or older free of dementia at baseline that were followed for 7 years. After controlling for confounding factors (i.e. age, sex, polypharmacy, stroke, depression, ischemic HD, DM), the study found an association between regular PPI use and incident dementia compared to PPI non-users [hazard ratio, 1.44 [95% CI, 1.36-1.52]; P < .001]. Interestingly, the HR for occasional PPI users over non-users was 1.16.

These studies show a potential relationship between PPI use and increased risk of new dementia. However, there are several caveats. Cohort designs can only demonstrate an association, not causation. The authors’ conclusion that “the avoidance of PPI medication may prevent the development of dementia” must be interpreted with caution. This one is, at best, a weak association, with a wide confidence interval (in the first study). A randomized, placebo-controlled study would need to be performed to show a causal relationship. A dose–response relationship, if reliably demonstrated, might strengthen the association. It is also interesting to note that the prevalence of PPI use in Germany is about four percent, whereas it is higher in the US, especially given it’s availability over the counter.

It is fair to say however, that unnecessary, long-term use of any medications – including PPIs – should be avoided. Until we have further information, prescribers should remain judicious in the use of PPIs and look for inappropriate prescribing, especially in geriatric patients who are at risk for polypharmacy and the negative implications of medication over-prescription.

By Eugenia Shmidt, MD, GI Fellow at Icahn School of Medicine at Mount Sinai, New York

References:

1. Haenisch B, von Holt K, Wiese B, Prokein J, Lange C, Ernst A, Brettschneider C, König HH, Werle J, Weyerer S, Luppa M, Riedel-Heller SG, Fuchs A, Pentzek M, Weeg D, Bickel H, Broich K, Jessen F, Maier W, Scherer M. Risk of dementia in elderly patients with the use of proton pump inhibitors. Eur Arch Psychiatry Clin Neurosci. 2015 Aug;265(5):419-28.

2. Gomm W, von Holt K, Thomé F1, Broich K, Maier W, Fink A4, Doblhammer G, Haenisch B. Association of Proton Pump Inhibitors With Risk of Dementia: A Pharmacoepidemiological Claims Data Analysis. AMA Neurol. 2016 Apr 1;73(4):410-6.

Photo courtesy: Will Culpepper

This post is part of the #GeriCases series, in which we discuss a clinical case in geriatrics and palliative medicine and the attempt to provide patient-centered care with the use of best available evidence.

Ravishankar Ramaswamy, MD

@RavRamaswamy

Section Editor, #GeriCases