PPIs associated with COPD and asthma exacerbations

 A recent Belgian study found that the use of proton pump inhibitors (PPIs) is associated with increased exacerbations of COPD and asthma (which they refer to as chronic obstructive airway diseases, COADs) in a dose-response way: see copd inc with PPIs Chest2026 in dropbox, or DOI: https://doi.org/10.1016/j.chest.2026.01.002

 

Details:

-- 932,135 patients (with a total of 1,084,446 person-years of follow-up) were identified using the Belgian nationwide claims-based data of adult patients receiving long-term medication for COADs (ie, having at least two dispensings of medication for obstructive lung diseases) between 2017 and 2022

-- patients on histamine type 2 receptor antagonists (H2RAs) were excluded (at that time there was a carcinogen impurity found in ranitidine, the most prescribed H2 blocker, that was taken off the market),

-- 416,087 patients (44.6%) were also receiving PPIs:

-- mean age 61, 54% female, 29% low socioeconomic status (assessed by their medical coverage, with low SES defined as having increased medical reimbursements that is provided to individuals with low income)

-- 23% ever-smokers, normal weight 93%/obesity or overweight 6%

-- COAD exacerbation history: none in 35%/one moderate 28%/at least two moderate or one severe 37%

-- comorbidities: cardiovascular 35%, depression or anxiety 25%, diabetes 16%, frailty 13%, cancer 6%, sleep apnea 4%, GERD 3%, arthritis 2%, stomach ulcer 1%

-- medications: NSAID 42%, aspirin 23%, DOAC 7%, antacid 5%, clopidogrel 3%, vitamin K antagonist 3%

-- short acting bronchodilator use: appropriate use 86%/overuse 8%/heavy overuse 6% [of note, short acting bronchodilators are not considered appropriate asthma therapy in adults: https://gmodestmedblogs.blogspot.com/2024/12/asthma-reliever-therapy.html)

    -- however there were differences in the above demographics between the groups: patients receiving PPIs were more likely to be female (58% versus 51%), to be older (mean age 66 versus 57), to have lower SES (35% versus 24%), to have ever-smoked (31% versus 17%), to have more exacerbations at baseline (46% versus 30%) and to have experienced at least two moderate or one severe exacerbation in the previous year (48% vs 25%), and to have more cardiovascular comorbidities (40% versus 25%)

-- the overall use of PPIs in the whole cohort, where PPIs w use was quantified using the "defined daily doses (DDDs)" dispensed during the prior year:

    -- none: 516,048 patients, 55.4%

    -- <28 DDD: 57,540, 6.2%

    -- 29-180 DDD:128,017, 13.7%

    -- 181-365 DDD: 127,981, 13.7%

    -- >365 DDD: 102,549, 11.0%

-- the cumulative PPI use in those who were taking PPIs:

    -- 1-28 DDDs: 57,540 patients (13.8%)

    -- 29-180 DDDs: 128,017 patients (30.8%)

    -- 181-365 DDDs: 127,981 patients (30.8%)

    -- >365 DDDs: 102,549 patients (24.6%)

 

-- primary outcome: occurrence of COAD exacerbations with definitions based on the Global Initiative for Chronic Obstructive Lung Disease and the Global Initiative for Asthma reports. exacerbations were specifically categorized as moderate (outpatient prescription filled for an oral corticosteroid, guideline-recommended antibiotics, or both) and as severe (requiring hospitalization) 

 

Results:

-- the relationship between the number of patients in the different DDD categories and their accrued event rates of COAD exacerbations, displaying an evident dose-response curve with the more PPI use, the greater the exacerbations:

 

 

-- Adjusting for age, sex, smoking history, SES, exacerbation history, short acting bronchodilator use, frailty, Charlson Comorbidity, and the presence of GERD. the increased exacerbation risk was:

    -- COPD (n= 61,569): 16% increase, HR 1.16 (1.13-1.16), P<0.001

    -- asthma (n= 24,475): 16% increase, HR 1.16 (1.13-1.20), P<0.001

    -- asthma-COPD overlap (n= 8665): 24% increase, HR 1.24 (1.17-1.31), P<0.001

        -- assessment by IPTW (inverse probability of treatment weighting, a statistical method to equalize baseline differences between the groups): for the above three groups there was respectively a 13% increase, a 14% increase, and a 20% increase

 

-- based on the IPTW model, there was an increasing exacerbation risk across higher PPI use categories, versus no PPI use:

    -- <28 DDD: 9% increase, HR 1.09 (1.07-1.10)

    -- 29-180 DDD: 15% increase, HR 1.15 (1.14-1.16)

    -- 181-365 DDD: 19% increase, HR 1.19 (1.18-1.20)

    -- >365 DDD: 25% increase, HR 1.25 (1.23-1.26)

        -- stratification by COPD, asthma, or the asthma-COPD overlap: slightly stronger estimates of long-term PPI use in asthma and the overlap syndrome versus COPD by itself

 

-- supplementary analyses:

    -- GERD: associated with 15% increase with PPI use, HR 1.15 (1.08-1.23) in those with GERD, and 16% increase in those without GERD, HR 1.16 (1.15-1.16). ie, no significant difference

        -- however, the 1-28 DDDs and 29-180 DDDs PPI groups were not statistically associated with exacerbations among patients with GERD

        -- but it was significant in patients without GERD: respectively with 9% increase in the 1-28 DDDs, HR 1.09 (1.07-1.10); and 15% in the 29-180 DDDs group, HR 1.15 (1.14-1.16). And the dose-dependent association was more pronounced in patients without GERD

    -- stratification by age group: the strongest effective PPIs on exacerbations was observed in the youngest age group

    -- stratification by frailty: stronger associations in patients without frailty although associations in both strata were significant

    -- by being more stringent on actual PPI exposure at the index date of enrollment, there was a 26% relationship with COAD exacerbations, HR 1.26 (1.26-1.27)

    -- when evaluating PPI exposure for less than one week, there was no statistically associated significance

    -- for the 7952 (2.9%) active PPI users patients taking CYP2C19 inhibitors in the index analysis (mainly fluoxetine or fluconazole) there was a 17% increased exacerbation risk when on PPIs, HR 1.17 (1.14-1.20), likely since CYP2C19 inhibitors increase PPI concentrations by reducing their metabolism

    -- restricting the analysis just to severe exacerbations of COAD still resulted in a 9% increased risk, HR 1.09 (1.07-1.12)

 

Commentary:

-- COADs are highly prevalent worldwide with their attendant personal and societal burdens

-- exacerbations can be fatal but are also associated with accelerated lung function decline, impaired quality of life, and increased overall mortality risk

-- gastroesophageal reflux disease (GERD) is common in the setting of both COPD and asthma

    -- extraesophageal GERD symptoms can occur that include chronic cough, wheezing, and sore throat

    -- proton pump inhibitors (PPIs) are often used as first-line management for GERD

-- however, there have been many studies assessing the overuse of PPIs in both hospitalized patients and those seen in the primary care setting, with >50% of PPI prescriptions associated with inappropriate overuse (see PPI overuse EurJournalInternMed2017 in dropbox, or doi.org/10.1016/j.ejim.2016.10.007)

    -- one issue in the primary care setting is that we clinicians have relatively little time to spend with patients. Many patients have multiple important clinical problems, and when PPIs work to decrease their upper GI symptoms, we often need to move on to addressing their other problems; and it is easier to just renew the PPIs than to engage in a perhaps lengthy discussion with the patient about the potential adverse effects of PPIs (see below) and perhaps switching to the better tolerated histamine-2 receptor antagonists (eg famotidine)

 

-- this study found that there was an increased likelihood of COAD exacerbations in patients who were using PPIs, especially among long-term users, those who were younger, who did not have GERD, who did not have frailty, and were long-term PPI users who had asthma/COPD overlap syndrome in particular

    -- this was true even by minimizing confounding by indication through using IPTW

    -- and there was a pretty clear dose-response association: the more PPI usage, the more frequent the COAD exacerbations

 

-- a European study also found that PPI use was significantly associated with a shorter time to severe exacerbations in patients who had COPD (the prior thought was that treating the GERD would decrease the exacerbations: doi.org/10.1111/resp.12758)

-- an observational cohort study in 12 countries found an increased risk of COPD exacerbations with PPIs, also in the absence of GERD diagnoses: https://pubmed.ncbi.nlm.nih.gov/26970108/

-- a systematic review and meta-analysis also found that PPI use contributed to both the development and exacerbation of asthma

 

-- there are several proposed mechanisms of action that could explain the relationship between PPIs and increased COAD exacerbations:

   -- PPIs lead to a microbial imbalance that has been found to be associated with increased systemic and airway inflammation which could potentially influence pulmonary outcomes

        -- this may be mediated by gut dysbiosis (adverse gut microbiome changes) that can influence the respiratory microbiota through translocation of these adverse gut microbiota to the lungs, as well as through these adverse gut microbes altering circulating inflammatory cytokines leading to longer gut dysbiosis that might affect lung function for longer; this all highlights the importance of the gut-lung axis in the development and progression of COADs. the gut-lung axis also includes tryptophan metabolites and bile acids produced by the gut microbiome that can reach the lung through the circulation (for the gut-lung axis, see https://pubmed.ncbi.nlm.nih.gov/39301033/ and https://pubmed.ncbi.nlm.nih.gov/32158441/)

            -- of note, there are several microbiomes in the human body besides the largest one, the gut microbiome, including both the upper (including mouth) and lower respiratory tracts, the skin, and the vagina. and there are profound gut microbiome relationships with other body parts as well as the gut-lung one, including the gut-liver axis, and especially the gut-brain axis

        -- PPIs are present in endothelial lysosomes:

              -- PPIs impair lysosomal acidification, enzyme activity, and proteostasis leading to the accumulation of protein aggregates; this all is associated with oxidative stress, endothelial dysfunction, and accelerated endothelial aging which could lead to increased vascular permeability and contribute to poor COAD control

        -- a low pH is essential for converting pepsinogen into active pepsin which breaks down proteins, and this acidic pH is blocked by PPIs

            -- incompletely digested proteins from the non-acid upper GI condition from the PPIs that thereby reach the intestines then may act as antigens and may induce T-helper response and IgE sensitization of the immune system, potentially triggering asthma exacerbations in susceptible people

            -- vitamin and mineral deficiencies (calcium, iron, magnesium, vitamin B12) are associated with PPIs; there is clinical evidence that these are associated with lung function; iron deficiency and magnesium deficiency have the best clinical data of being associated with increased COAD exacerbation risk

-- from my perspective, PPIs are one of the major drug classes that should not be available over-the-counter, since they are associated with several major medical concerns (per the American Gastroenterology Association guidelines and their references: adverse effects Gastroent2017 in Dropbox, or doi.org/10.1053/j.gastro.2017.01.031);  and, as noted above, over 1/2 the time are inappropriately prescribed:

    -- chronic kidney disease: there is increased risk of acute interstitial nephritis but also chronic kidney disease that is not explained solely by the acute kidney injury, with a 1.8% absolute annual excess CKD risk comparing PPIs to H2 receptor antagonists (H2RAs), and a 17% and 50% increased relative risk respectively, adjusting for confounders in the 2 studies done on this

    -- dementia: 2 clinical studies in 3,327 non-institutionalized German adults aged 75 years or more with serial neuropsychiatric examinations: PPIs were associated with a 38% increased risk for dementia, with similar risk increases for Alzheimer’s and non-Alzheimer’s dementia. a retrospective study accessing an insurance database covering more than half of the German population over 75 years old found a 44% higher risk for dementia in regular users of PPIs compared to non-users; when occasional users of PPIs were compared to non-users, there was a 16% higher risk. In this study, adults selected for PPIs had strikingly higher baseline rates of depression, stroke, and polypharmacy (an important criterion, since patients taking PPIs tend to have more comorbidities than those who do not); however, perhaps those with lots of comorbidities have increased dementia risk when adding PPIs to their comorbidities???): https://gmodestmedblogs.blogspot.com/2023/08/sustained-ppi-use-and-subsequent.html

    -- Myocardial infarction: equivocal studies. But a data-mining study of patients with a low baseline risk for MI found an excess relative risk of 9-16% for MI after a median of four years of PPI use. The COGENT study provided the most important single piece of evidence related to PPIs and MI. If PPIs do cause vasoconstriction, such an effect would likely be most obvious in patients who, like the participants in COGENT, have a high baseline risk for MI. PPIs are associated with vasoconstriction per their effect on the endothelial nitric oxide synthase: https://www.ahajournals.org/doi/10.1161/circulationaha.113.003602

    -- Infections:

        -- Small intestinal bacterial overgrowth (SIBO): 2 studies performing duodenal aspirates for the diagnosis of SIBO and a rigorous, self-controlled study in which within-individual changes in bacterial counts were assessed before versus after PPIs. In these two studies, PPIs were associated with an over 20-fold relative risk for SIBO.

        -- Other infections:

        -- C difficile infection (CDI): PPIs have no direct effect on pH in the colon, though they appear to exert a significant “downstream” effect on colonic bacteria. Bacterial taxa associated with Clostridium difficile infection (CDI) were increased in the stool of healthy volunteers after 4-8 weeks of high-dose PPIs. Observational studies show an approximately 50% relative risk for CDI associated with PPIs, although CDI remains rare enough that there is little confidence in this estimate.

             -- Spontaneous bacterial peritonitis may be associated with a 3-fold increased risk particularly in patients with cirrhosis

             -- Campylobacter or Salmonella infection: may be associated with a 2- to 6-fold increased risk

        -- the risk for infections is likely related to the profound effects on the gut microbiome by PPIs: https://pmc.ncbi.nlm.nih.gov/articles/PMC7426480/, likely related to the strong PPI effects of reducing the acidity of gastric secretions, since the gastric acid provides an important barrier from bacteria ingested or from the respiratory tract from reaching the colon

            -- the alterations in gut bacteria due to hypochlorhydria may also lead to changes in intestinal permeability and translocation of bacteria across the gut wall

            -- the altered microbiome risk associated with PPIs is modest compared to traditional risk factors such as antibiotics; some studies suggest that PPIs may be more important in children. a recent blog reviewed the role of antibiotics in long-term microbiome dysbiosis (https://gmodestmedblogs.blogspot.com/2026/05/long-term-microbiome-effects-after.html). perhaps the increased infections from PPIs could lead to antibiotics that will further disrupt the microbiome and perhaps for many years??

     -- bone fracture: some mixed data on this association though there may well be a 4-fold increase. the link between PPIs and increased fracture risk is based on several potential mechanisms, including hypochlorhydria-associated malabsorption of calcium or vitamin B12 (B12 does affect bone health, since it has a direct effect on osteoblast activity), gastrin-induced parathyroid hyperplasia, and osteoclastic vacuolar proton pump inhibition (of course, there are many factors leading to osteoporotic fractures, and it is very hard to target PPIs as the sole contributor to the increase; the known mechanistic data, however, supports their association with increased risk)

Limitations

-- low-dose over-the-counter PPIs were not registered in the database and therefore not included in the study

-- this analysis was based on PPI use at the index date of the study; there may well have been changes in PPI usage over the next five years of the study

-- there also may well be changes over time in non-pharmacologic factors that affect lung function: smoking, exercise, occupational and environmental exposures, etc

so,

-- a pretty impressive study supporting other studies finding that the adverse effects of taking PPIs includes an increased risk of obstructive lung disease (COPD, asthma)

-- this adverse effect is an addition to the array of reasons to limit PPI use to clear-cut indications (and overuse seems to occur most of the time), and when appropriate, limiting the duration of the PPI use

    -- the issue for us in primary care is that PPIs work so well for GI symptoms that we may not have time to engage a patient who also has lots of other problems in a discussion about switching from PPIs to H2-receptor antagonists or even just calcium tablets when appropriate; and, of course, we should also be reinforcing the nonpharmacologic approaches to decrease GERD/gastritis, including avoiding things that decrease lower esophageal pressure such as caffeine, smoking, chocolate, alcohol. This all require a prolonged discussion, would take away from time to deal with the array of other important patient issues: it is just easier just to renew a medication (PPIs) that work and move on to the other issues

-- a very real problem is that PPIs are available over-the-counter and may be taken even without our knowledge or ability to reduce their use.

    -- PPIs, to me, are not my only OTC med of concern: NSAIDs are available OTC and are also associated with oh-so-many adverse effects including upper GI symptoms and potentially severe bleeding (made 3-fold worse in the 50% of people globally who have untreated H Pylori infections), chronic kidney disease, increased myocardial infarction risk, heart failure, hypertension, fluid retention, even higher likelihood of developing chronic pain when used for acute pain: (https://gmodestmedblogs.blogspot.com/2022/05/acute-pain-anti-inflammatories-lead-to.html, ...)

-- i have certainly found many patients who refuse to take prescribed meds after reading their litany of printed adverse effects packaged with them, yet they take OTC PPIs and NSAIDs that have many really severe problems that they are unaware of...

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

to get access to the dropbox: https://www.dropbox.com/scl/fo/pcf4vymtpo77f49wd2ljr/ANyJqB40CuBd-DMEr67XDoI?rlkey=5dsalmuhf8ekd0svsgvn93xm1&dl=0

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