sustained PPI use and subsequent dementia

 A new article (preprint, pre-peer review) found that a high cumulative dose of proton pump inhibitors (PPIs) is associated with increased risk of dementia (see ppi and dementia neurol2023 in dropbox, or DOI: 10.1212/WNL.0000000000207747)

 

Details:

-- 5712 dementia-free participants were evaluated from the community-based ARIC study (Atherosclerosis Risk in Communities Study), enrolled from 1987-89 and followed through until 2017

    -- the ARIC study recruited patients from four US communities: Washington County, Maryland; Forsyth County, North Carolina; suburbs of Minneapolis, Minnesota; and Jackson Mississippi.

-- mean age 74, 58% female, 22% Black/78% white, BMI 29, hypertension 73%, diabetes 34%, APOE e4 carrier 25%, smoking never 38%/former 48%/current 5%, aspirin use 72%, anticholinergic use 1%, vitamin B12 use 5%, and H₂ receptor antagonists\ use 7%

-- baseline characteristics were very similar for PPI users versus nonusers, though the users were more likely to be female, white, have hypertension/diabetes, and not have the APOE4 genotype

-- PPI use was assessed visually at clinic visit one (1987-89) through clinic visit five (2017), as well as by annual phone calls. The current study used ARIC clinic visit five as the baseline, since this was the first visit in which PPI use was common

    -- PPI use was assessed two ways: current use at visit five, and duration of use up to visit five, with exposures potentially back to visit one.

        -- exposure categories were: zero days, 1 day to 2.8 years, 2.8 to 4.4 years, and >4.4 yrs

        -- over-the-counter medications were not included (the first PPI drug became available over-the-counter in 2003)

-- incident dementia status was assessed through neuropsychological examination, twice yearly telephone calls using a standardized screening tool, and surveillance of hospital discharge codes and death records

-- covariates assessed included sex, age, smoking status, race, ARIC center, BMI, sitting blood pressure, diabetes, and APOE4 genotyping when available; also assessed were medications that might have influenced cognition (antihypertensives, H₂receptor antagonists, anticholinergics, aspirin, and vitamin B12)

    -- the reference group was those not using PPIs at visit five

 

-- Main outcome: incident dementia after visit five

-- secondary analysis: H₂ receptor antagonist use as an active comparator, using the same methodology as determining PPI use noted above

 

-- median follow-up 5.5 years

 

Results:

-- minimum cumulative PPI dose was 112 days and maximum was 20.3 years

-- incident dementia cases: 585 cases (10.2%)

 

-- 1490 cumulative PPI users (26.1%) were identified from visit one in 2011, and 1450 current users (25.4%) at visit five, with median use of 3.5 years

-- PPI use >4.4 years prior to visit five, versus no PPI use: 33% increased risk of developing dementia during follow-up, HR 1.33 (1.00-1.77), in the fully adjusted model (which included the APOE genotype)

-- no evident association in those using lesser exposures of PPIs, at either the 1 day to 2.8 years or >2.8 to 4.4 year evaluations

-- no significant difference in the risk of dementia when H₂ receptor antagonists were assessed either as current exposure or as cumulative exposures over time

-- there was also no association between vitamin B12 deficiency and incident dementia (B12 deficiency has been postulated as a mechanism for potential PPI-induced dementia, likely related to decreased B12 absorption)

 

Commentary:

-- PPIs are used remarkably often: globally,  23.4% of the adult population take PPIs, and 40% of those >70yo (see ppi use review EurJClincPharm2023 in dropbox, or or https://doi.org/10.1007/s00228-023-03534-z )

-- one quarter of those taking PPIs continue to use them for >1 year and 28% of them took PPIs >3years

-- a large observational study from the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Medical Care Survey found that PPIs were used  in the US in 4.0% of ambulatory visits in 2002, increasing to 9.9% in 2009. And, 62.9% of patient visits did not have gastrointestinal complaints, gastrointestinal diagnoses, or other indicated reason for their use (see Proton Pump Inhibitor Use in the U.S. Ambulatory Setting, 2002–2009 | PLOS ONE ) . This study and others find that about 2/3 of those on PPIs do not have a clear indication for them

-- another study found that PPIs were dispensed in the US over 115 million times in 2016 

-- these numbers of PPI users have likely increased dramatically, but with difficulty assessing actual use of PPIs since they have been over-the-counter in the United States since 2003 and in European Union since 2013. And many other countries in the world have pretty easy access to many medications over the counter (i could not find clear information on which countries, but anecdotal experience confirms this)

 

-- the main finding of the study was that cumulative PPI usage of at least 4.4 years by individuals age 45 and older was associated with a higher incidence of newly diagnosed dementia

    -- some strengths of this study include the rigor of data collection (including physician-adjudicated cases of dementia), longer-term retroactive assessment of PPI use, the multi-ethnic database, and results adjusted for APOE4 genotype

 

-- Several prior studies have found an association between PPI use and dementia, though there have been mixed results. Several concerns have been raised about these studies including non-representative participants (largely white or Asian), determinations of dementia (using ICD codes relies on proper coding by clinicians), not having sufficient follow-up time to detect dementia, and variability of intra-individual PPI use over follow-up (instead of documenting long-term exposures). This study specifically assessed cumulative PPI use over time, likely a better marker for longer-term adverse effects

-- A claims-based study found that PPI use in the elderly was associated with an increased risk of dementia (see http://gmodestmedblogs.blogspot.com/2019/04/ppi-use-and-dementia.html ) 

-- for review of the clinical practice guidelines of the American Gastroenterological Association on de-prescribing PPIs, including commentary on several potentially serious adverse effects of PPIs: http://gmodestmedblogs.blogspot.com/2022/07/ppis-way-overprescribed.html 

-- there may also be important changes from PPIs in neurologic functioning: some studies have found that PPIs may be associated with decreased cognition, perhaps related to decreased vitamin B12 absorption and/or increased brain burden of amyloid-beta, the latter has been found in mouse studies (see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4696341/ ).  

-- there may also be a relationship between some other potential associations found with PPIs (eg CKD and stroke) and their associated increased risk of developing dementia.  

-- Additionally, there are documented major changes in the gut microbiome associated with PPIs (possibly because of hypochlorhydria), with documented decreased microbial diversity and dysbiosis/alterations in a healthy composition of the microbiome (of note, decreased microbial diversity and microbiome dysbiosis is also found in Alzheimer’s and in animal models of dementia, perhaps related to neuro-inflammation, oxidative stress or triggering amyloid-beta accumulation) as well as perhaps distortions of the important gut/brain axis, where there is bidirectional communication between the central nervous system and the gut through neural, endocrine, immune, and humoral links (many neurotransmitters are synthesized in the gut.  https://gut.bmj.com/content/65/5/740 ) has more details  on the gut/brain axis)

    --the gut/brain axis (see microbiome gut brain axis JClinInvest2015 in dropbox, or doi:10.1172/JCI76304), diocuments that the gut microbiome can affect emotional behavior, stress- and pain-modulation, and brain neurotransmitters. 

 

-- Chronic PPI use has been associated with numerous adverse events, including stroke, cardiovascular disease, chronic kidney disease, all-cause mortality, osteoporosis/fractures, C. difficile infections, community acquired pneumonia, increased risk of active tuberculosis (http://gmodestmedblogs.blogspot.com/2022/07/ppis-way-overprescribed.html , for the littany of potential adverse effects), though these associations have not been found in all studies 

        -- there also may be PPI-associated distortions in the respiratory microbiomes, which could predispose people to the increased risk of pneumonia and TB as found in some studies (see PPI affects lung microbiome JGastro2023, or doi.org/10.1007/s00535-023-02007-5)

 

Limitations: 

-- this study involved a retrospective analysis of the much larger observational study, thereby limiting our ability to determine causality because of potentially unmeasured confounders. However, assessing an active comparator, H₂ receptor antagonists in this case, and finding a negative result, does decrease the risk of residual unmeasured confounding 

-- unclear what the PPI taken was, or what the PPI dose was over this cumulative period of time 

-- though they did have annual assessments of continued PPI use, they did not have detailed use between these annual assessments 

-- this study did not include over-the-counter PPIs, only those prescribed, limiting the validity of the results some. However, it would be difficult to ascertain individual-based cumulative PPI exposure by numbers of PPI drugs purchased over-the-counter. And it is possible that some people in the control group did take OTC PPIs, which might have decreased the level of their association with dementia (although ARIC participants were asked not to take over-the-counter medications) 

    -- also, it is possible that some in the control group were put on prescription PPIs after visit 5 (it appears that the researchers did not assess PPIs taken after the 5^th visit), and the dementia risk of PPIs would be less striking in the study if many in the control group were put on PPIs after the 5^th visit (a likely occurrence since the overall use of PPIs continued to increase since then....)

-- though the study did include different racial diversity from prior studies, in particular including a large number of Black participants, it did not include other populations, limiting the generalizability of the results 

-- many participants in ARIC died prior to visit five, were not included in this analysis, and also might have distorted the results found 

 

So, a few issues: 

-- this study adds to the likelihood that higher cumulative doses of PPIs is in fact associated with incident dementia, and  there are quite plausible mechanisms to supportthis association 

-- it certainly makes sense (to me) to use the least potent medication for any indication whenever possible. It seems intuitively likely that the most potent ones tend to have the most disruptions of normal physiologic/homeostatic mechanisms in the body. The above-noted American Gastroenterological Society clinical practice guidelines on de-prescribing PPIs reinforce this concern

    -- and, this is especially true for PPI-associated disruptions of the different microbiomes (eg gut, respiratory tract), which seem to have a plethora of effects throughout the body (with new ones found pretty frequently)

-- so, this leaves us in primary care with yet another task: while taking care of patients' multiple medical/social/psych issues as we do routinely, another one is to try to de-prescribe PPIs in patients who are taking them and doing well 

    -- in that light, my anecdotal experience is that most patients are willing to try changing the PPIs (especially on hearing the potential of the many potential long-term adverse effects of PPIs), and some even find that H2 blockers taken regularly or as needed, or even just calcium as needed, seem to work quite well.

  

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs:

 

 go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

if you would like to see the articles in this blog, please email me. 

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list