What are the risks and incidence of increasing memory loss associated with PPI use? Are there any recommendations to decrease risk or alternative treatment options for GERD?

Comment by InpharmD Researcher

The majority of relevant studies, including the most recent meta-analyses, have found no significant association between use of PPIs and increased risk of cognitive impairment, including dementia. While some studies have observed a mild increased incidence of dementia in PPI cohorts, the risk was not determined to be statistically significant. Most data are presented in the form of observational or cohort studies with significant heterogeneity. While the incidence of memory loss with PPI use has not been well characterized, an analysis of the FAERS database reported 1.29% of PPI administrations reporting memory-associated adverse events; though this may be an unreliable source to estimate true incidence. Some studies do suggest H2RAs may be associated with cognitive decline, although the data are not yet definitive. Further studies are needed to explore the true incidence of memory loss with PPI use and the association of cognitive impairment with alternative GERD treatment options.

Background

A 2023 review article analyzed various studies and meta-analyses to determine whether there was an association between dementia and proton pump inhibitors (PPI) in adult patients without a previous diagnosis of cognitive impairment or dementia. A total of 28 different studies were included for analysis, mostly cohort studies (n= 15) and meta-analyses (n= 7). Only one randomized controlled trial (RCT) was included which found a non-significant increase in dementia with pantoprazole 40 mg versus placebo in atherosclerotic disease patients (odds ratio [OR] 1.20; 95% confidence interval [CI] 0.81 to 1.78; p= 0.36). One meta-analysis suggests a slightly higher risk of dementia from PPI use (hazard ratio [HR] 1.29; 95% CI 1.12 to 1.49) but the others did not support an association, including the most recent meta-analysis consisting of 11 observational studies (N= 642,949) with a HR for dementia from all-cause mortality at 1.11 (95% CI 0.88 to 1.37). Based on these findings, the authors do not suggest there’s a statistically significant association between PPI use and development of cognitive impairment, including dementia. [1]

A 2023 meta-analysis reviewed prospective studies that examined the association between PPI use (compared to non-use or intake of histamine-2 receptor antagonists [H2RA]) and incident dementia (all-cause or Alzheimer’s disease). Nine observational studies were included (N= 3,302,778 individuals; 839,940 PPI users, 2,462,838 non-users; n= 204,108 [6.2%] diagnosed with dementia). The primary meta-analysis showed no evidence of association between PPI use and all-cause dementia, with significant heterogeneity between studies (risk ratio [RR] 1.16; 95% CI 1 to 1.35; I^2= 99%). No differences were seen after adjusting for small-study bias using two different methods as well. A subgroup analysis (n= 3,109,769) did not identify a difference in risk of Alzheimer’s disease between PPI users and non-users (RR 1.15; 95% CI 0.89 to 1.5). A secondary analysis (n= 847,399) showed no association of dementia risk and PPI use vs. H2RA use (RR 1.03; 95% CI 0.66 to 1.62). The authors concluded that the meta-analysis showed no clear evidence for an association between PPI intake and risk of dementia, however they cannot rule out that a potential risk exists. The meta-analysis is limited by the inclusion of mostly observational studies, introducing risk for selection bias and limiting ability to control for confounding variables that may have affected the reported outcomes. [2]

A 2019 meta-analysis included six cohort studies (N= 106,599) that reported the risk of dementia or Alzheimer’s dementia among PPI users vs. non-PPI users. The analysis found a mild increased risk of dementia with PPI use with the pooled relative risk of dementia in PPI users being 1.23 (95% CI 0.9 to 1.67) compared to 1.01 (95% CI 0.78 to 1.32) in non-PPI users, though this difference was not statistically significant (p= 0.19) and significant heterogeneity was observed (p<0.00001, I^2= 95%). A subgroup analysis found increased risk with shorter follow-up duration of <5 years (relative risk 1.62; 95% CI 1.4 to 1.86; p<0.001; I^2= 22%), however in pooled studies of longer follow up (≥ 5 years) found no statistically significant increased risk of dementia with PPI use (relative risk 0.98; 95% CI 0.75 to 1.27; p= 0.87; I^2= 87%). Additionally, subgroups of sample size (<10,000 and ≥ 10,000), did not show an increased risk of dementia in PPI users (relative risk 1.25; 95% CI 0.86 to 1.84; p= 0.23; I^2= 72% and 1.17; 95% CI 0.75 to 1.82; p= 0.48; I^2= 97%, respectively). Overall, the results of analysis suggested no significant association between PPI use and risk of dementia or Alzheimer’s disease. This analysis is limited by the use of only cohort studies due to the risk for selection bias in these studies and the substantial heterogeneity among the included studies. [3]

A 2019 study analyzed the Food and Drug Administration Adverse Event Reporting System database (FAERS/AERS) to provide evidence of an increased association of memory impairment among PPI reports compared to H2RA use as a control. In reports of PPI administration without concurrent medications, there was a significant increase in memory impairment adverse drug reactions (ADRs) compared to H2RA administration (1.29% vs. 0.4%, respectively; OR 3.28; 95% CI 2.31 to 4.67); outcomes included memory impairment, amnesia, Alzheimer’s dementia, and non-Alzheimer's dementia. The H2RA cohort (n= 8,309) did not have any reports of Alzheimer’s dementia, while the PPI cohort (n= 42,537) had 80 reports of Alzheimer’s dementia. PPI use was also significantly associated with a number of neurological ADRs as well, including migraine, peripheral neuropathies, and visual and auditory neurosensory abnormalities. Data provided by FAERS reports should be interpreted with caution, due to over or under reporting, incomplete information, risk of biases, and lack of control over potential confounding variables. A causal relationship cannot be established based on FAERS reports. [4]

A 2022 analysis of the Wisconsin Registry for Alzheimer’s Prevention study, investigated the effects of PPI use and apolipoprotein e4 carrier status on changes in neuropsychological functioning in healthy adults with familial risk factors for dementia. Overall, 1,573 middle aged participants without any symptoms of cognitive decline during the first study visit were included. Data were analyzed for the first four collection periods in 2001, 2006, 2009, and 2012; participants submitted blood samples and completed self-reported questionnaires which inquired about diagnosis of different medical conditions, prescription and non-prescription drug use, tobacco use, and physical activity during each study visit. The Verbal Intelligence Quotient (VIQ) of the Wechsler Abbreviated Scale of Intelligence was used to estimate verbal intelligence. Standardized scores were used for the VIQ only, that compared the participants scores to a normative sample stratified by age, gender, ethnicity, and geographic location. The Mini-Mental State Examination (MMSE) was used to measure orientation and mental status. Working memory was assessed using the Digit Span subtest of the Wechsler Adult Intelligence Scale- Third Edition. The Auditory Verbal Learning Test (AVLT) was used to measure immediate verbal learning (AVLTI) and delayed memory (AVLTD). Psychomotor speed was utilized to assess the Trail Making Test, Part A (TMTA) and cognitive flexibility measured using the Trail Making Test, Part B (TMTB). Study participants were divided into four groups: PPI non-user, stopped PPI, started a PPI, and consistent PPI user. [5]

In terms of covariates evaluated, consistent PPI users were more likely than non-users and those who started or stopped a PPI to report taking an antihypertensive medication. Non-users were more likely than other groups to engage in physical activity at least four times per month. Participants who reported stopping a PPI were more likely than other groups to report using a H2RA. Participants who started or consistently used a PPI were more likely to non-users or those who stopped a PPI to report a history of diabetes, depression, or lung disease. Additionally, consistent PPI users were older than non-users and had lower cholesterol than those who started a PPI. The four groups did not differ in performance on any of the neuropsychological tests during the initial baseline observation. [5]

The linear mixed model results that tested the main effects of H2RA use, PPI use, and APOe4 carrier status did not show that H2RA use was significantly associated with change in performance on any of the neuropsychological tests. In regards to PPI use, main effects assessing PPI use on change in Digit Span, AVLTI, TMTA, and TMTB scores were not significant; change in AVLTD score was significant, but not after adjusting for false discovery rates. In terms of PPI use interaction with APOe4 status, only a significant interaction effect was seen on change in performance on TMTB. APOe4 carriers who stopped PPIs had lower predicted mean TMTB changes score than non-carriers who stopped PPIs. The models did not show any significant interaction effects between PPI use and time or APOe4 carrier status on change in test performance. [5]

The authors concluded that although stopping PPI use may have mild protective effects on executive functioning for non-APOe4 carriers, PPI use was not associated with memory decline in a sample of subjects with familial risk factors for dementia. This study is limited by its retrospective design, inability to differentiate between different PPIs, and that it relied on self-reported data for baseline characteristics and medication history. [5]

A 2017 prospective cohort study investigated the association between PPI use and cognitive function in women. Data were collected from 13,864 participants who completed a self-administered computerized neuropsychological test battery. Psychomotor speed and attention, learning and working memory, and overall cognition was assessed as primary outcome measures. Patients were of a mean age of 61 years old (range 50-70 years) at the time of cognitive testing. Patients who used PPIs tended to be slightly older with higher incidence of chronic medical conditions compared to patients who did not report regular PPI use. PPI users were also less physically active, with higher BMI, lower educational status, and slightly lower dietary quality scores. Initially, a modest association was observed between duration of PPI use and scores for psychomotor speed and attention (mean score difference for PPI use of 9-14 years vs. never used was -0.06; 95% CI -0.11 to 0; p-trend= 0.03). However, after controlling for H2RA use, the score difference was decreased. Additionally, for participants who did not regularly use PPIs, the duration of H2RA use was associated with poorer cognitive scores, with the strongest association apparent for learning and working memory (mean score difference for H2RA users of 9-14 years vs. never used was -0.2; 95% CI -0.32 to -0.08; p-trend<0.001). Ultimately, a convincing association between PPI use and cognitive function was not observed, and the authors concluded that this data does not provide evidence to suggest that PPI use increases dementia risk. [6]

References:

[1] Caetano C, Veloso M, Borda S. Proton pump inhibitors and dementia: what association?. Dement Neuropsychol. 2023;17:e20220048. Published 2023 May 29. doi:10.1590/1980-5764-DN-2022-0048
[2] Ahn N, Nolde M, Krause E, et al. Do proton pump inhibitors increase the risk of dementia? A systematic review, meta-analysis and bias analysis. Br J Clin Pharmacol. 2023;89(2):602-616. doi:10.1111/bcp.15583
[3] Li M, Luo Z, Yu S, Tang Z. Proton pump inhibitor use and risk of dementia: Systematic review and meta-analysis. Medicine (Baltimore). 2019;98(7):e14422. doi:10.1097/MD.0000000000014422
[4] Makunts T, Alpatty S, Lee KC, Atayee RS, Abagyan R. Proton-pump inhibitor use is associated with a broad spectrum of neurological adverse events including impaired hearing, vision, and memory. Sci Rep. 2019;9(1):17280. Published 2019 Nov 21. doi:10.1038/s41598-019-53622-3
[5] Collin BG, Raju D, Altman J, Katsikas S. The effects of proton pump inhibitors on neuropsychological functioning. Appl Neuropsychol Adult. 2022;29(6):1403-1412. doi:10.1080/23279095.2021.1883019
[6] Lochhead P, Hagan K, Joshi AD, et al. Association Between Proton Pump Inhibitor Use and Cognitive Function in Women. Gastroenterology. 2017;153(4):971-979.e4. doi:10.1053/j.gastro.2017.06.061

Literature Review

A search of the published medical literature revealed 2 studies investigating the researchable question:

What are the risks and incidence of increasing memory loss associated with PPI use? Are there any recommendations to decrease risk or alternative treatment options for GERD?

Level of evidence

C - Multiple studies with limitations or conflicting results  Read more→



Please see Tables 1-2 for your response.


 

Neuropsychological Assessment After Long-term Omeprazole Treatment

Design

Observational study

N= 60

Objective

To investigate the cognitive impairments associated with long-term omeprazole treatment, with a focus on memory, attention, and executive functions

Study Groups

Omeprazole (n= 30)

No omeprazole (n= 30)

Inclusion Criteria

At least 18 years old, no older than 82 years, born in Brazil, and have at least one year of formal education

Exclusion Criteria

Neurological or psychiatric disorders; history of alcohol or drug abuse; use of benzodiazepines and/or antipsychotics; uncorrected sensory impairments; individuals who discontinued treatment with omeprazole, those who were under treatment with histamine 2 receptor antagonists (H2RAs) as alternatives for gastric-acid-related disorders, and those receiving vitamin B12 and/or antioxidant supplementation

Methods

Participants in the experimental group received continuous treatment with omeprazole for at least six months. Control participants had never been treated with the drug. All data was obtained using a self-report questionnaire and a quality of life scale, which were used to collect additional details concerning sample characterization. Participants were evaluated by previously trained psychology students. Cognitive assessments were carried out in specialized rooms and conducted individually in a single session, lasting about 1 h and 30 min each assessment.

Duration

Treatment period: 6 months 

Outcome Measures

Verbal Fluency, Rey Auditory-Verbal Learning Test (RAVLT), Attention Assessment Battery, Five Digit Test (FDT), Hayling Test (HT), and Sentence-word Span test from the Brief Neuropsychological Assessment Instrument (NEUPSILIN)

Baseline Characteristics

 

Omeprazole (n= 30)

No omeprazole (n= 30)

 

Age, years

61.97 ± 7.47 64.33 ± 7.89  

Female

25 26  

Number of years of education

7.07 ± 3.3 5.80 ± 2.23  

The mean duration of omeprazole treatment in the experimental group was 9.57 ± 5.55 years. 

Results

Endpoint

Omeprazole (n= 30)

No omeprazole (n= 30)

p-value

Attention

BPA FA—correct

BPA FA—commission errors

BPA FA—omission errors

BPA DA—correct

BPA DA—commission errors

BPA DA—omission errors

BPA AA—correct

BPA AA—commission errors

BPA AA—omission errors

 

58.89 ± 16.56 

0.37 ± 0.62

2.65 ± 2.55

47.72 ± 14.92

1.24 ± 2.69

19.58 ± 15.92

49.13 ± 14.35

0.75 ± 1.29

6.37 ± 5.62

 

78.20 ± 21.72

0.76 ± 2.01

4.23 ± 11.69

55.93 ± 23.64

3.83 ± 10.26

17.36 ± 13.58

56.1 ± 23.45

1 ± 1.72

6.34 ± 5.64

 

0.000

0.325

0.48

0.118

0.193

0.566

0.176

0.547

0.981

Memory

RAVLT—A6

RAVLT—A7

RAVLT—total

RAVLT—LOT

RAVLT—PI

RAVLT—RI

Neupsilin span—total

Neupsilin span—longest correct seq.

 

5.33 ± 2.42 

5.46 ± 2.8

36.16 ± 7.74

13.33 ± 7.12

0.93 ± 0.32

0.6 ± 0.27

19.45 ± 3.58

1.48 ± 0.78

 

6.76 ± 2.68 

6.33 ± 2.3

36.10 ± 9.27

12.26 ± 7.88

0.92 ± 0.34

0.8 ± 0.37

18.3 ± 6.13

1.83 ± 1.14

 

0.034

0.196

0.976

0.585

0.882

0.019

0.386

0.177

Executive functions

UVF—total

LF—total

CF—total

FDT reading—time

FDT reading—errors

FDT interference—time

FDT interference—errors

FDT counting—time

FDT counting—errors

FDT shifting—time

FDT shifting—errors

FDT inhibition

FDT flexibility

Hayling—part A time

Hayling—part A errors

Hayling—part B time

Hayling—part B errors/15

Hayling—part B errors/45

Hayling—part B-A time

Hayling—part B/A time

 

35.63 ± 14.96 

14.34 ± 8.45

17.21 ± 6.65

56.6 ± 20.28

1.03 ± 5.28

103.19 ± 43.6

3.83 ± 7.36

64.51 ± 18.99

0.23 ± 0.67

140.68 ± 55.34

5.63 ± 6.41

7.91 ± 11.83

84.08 ± 54.81

30 ± 21.83

0.93 ± 0.98

85.94 ± 49.37

5.57 ± 3.37

16.67 ± 10.07

49.53 ± 32.26

3.1 ± 1.26

 

42.8 ± 21.5

21.3 ± 8.51

22.73 ± 6.29

44.69 ± 12.15

0 ± 0

91.8 ± 38.4

2.33 ± 2.97

49.94 ± 12.32

0.06 ± 0.25

154.24 ± 89.7

4.86 ± 5.13

5.25 ± 5.95

109.54 ± 84.96

23.14 ± 15.09

0.4 ± 0.67

79.52 ± 60.83

3.5 ± 3.09

10.5 ± 9.28

45.12 ± 33.73

3.46 ± 2.28

 

0.139 

0.003

0.002

0.008

0.289

0.287

0.305

0.001

0.213

0.484

0.611

0.275

0.173

0.162

0.017

0.623

0.016

0.017

0.606

0.45

Rey Auditory-Verbal Learning Test (data presented in graph): No significant group differences were observed in trials A1 through A5, suggesting that both groups were able to learn the words list after a series of repetitions, as expected. However, scores in trial A6 pointed to a significant difference between the learning curve of each group. These findings suggested that, although learning ability (trials A1–A5) did not differ between groups, participants did differ with regards to short-term episodic memory and forgetting speed (trial A6).
Correlation between duration of Omeprazole treatment (months) and neuropsychological assessment: These results suggest that longer treatments are associated with larger impairments in sustained attention, short-term episodic memory, and executive functioning. The duration of omeprazole treatment was also significantly associated with delayed episodic memory (RAVLT trial A7, [r= -0.308]). The analyses also showed a negative correlation between the duration of omeprazole treatment and scores on the Attention Assessment Battery (Focused attention— Number correct [r= -0.419]), the Rey Auditory-Verbal Learning Test (RAVLT trials A6 [r= -0.329], and A7 [r= -0.308]), and the Verbal Fluency Test (Category fluency—Total score [r= -0.349] and Letter fluency—Total score [r= -0.307]). These findings suggest that individuals exposed to the drug for longer periods of time were less accurate and recalled fewer words during the application of these instruments. The duration of treatment was also positively correlated with scores on the FDT (Reading—Time [r= 0.354], Interference—Errors [r= 267], Counting—Time [r= 0.424], and Errors [r= 380]) as well as the Hayling Test (Errors Part B/15 [r= 0.310] and Errors Part B/45 [r= 0.309]). This suggests that patients exposed to the drug for longer periods are likely to make more mistakes on the Hayling Test and take longer to complete the FDT.
Abbreviations: BPA: Psychology Battery Tests of Attention; FA: Focused attention; DA: Divided Attention; AA: Alternating Attention; RAVLT: Rey Auditory-Verbal Learning Test; LOT: Learning Over Trials; PI: Proactive Interference; RI: Retroactive Interference; Correct seq: correct sequence; UVF: Unconstrained Verbal Fluency; LF: Letter Fluency; CF: Category Fluency; FDT: Five Digit Test

Adverse Events

See results

Study Author Conclusions

The results showed significant group differences in verbal fluency, short-term episodic memory, selective attention, and executive functions. The duration of omeprazole treatment was also positively associated with the magnitude of cognitive impairment.

InpharmD Researcher Critique

The study's limitations include a small sample size, lack of a calculated sample, convenience sampling, reliance on self-reported measures of cognitive impairments, focus only on omeprazole without considering other medications or confounding factors, and restriction to the Brazilian population, limiting generalizability. Further research is needed to replicate and expand upon these findings to establish a stronger understanding of the potential cognitive impairments associated with omeprazole use.



References:

Haefliger R, Dries LS, Perassolo MS, Cardoso CO. Neuropsychological assessment after long-term omeprazole treatment [published online ahead of print, 2022 Aug 5]. Appl Neuropsychol Adult. 2022;1-9. doi:10.1080/23279095.2022.2106570

 

Gastric acid suppressants and cognitive decline in people with or without cognitive impairment

Design

Retrospective review

N= 4,485

Objective

To examine longitudinal associations between histamine‐2 receptor antagonists (H2RA) versus proton pump inhibitors (PPI) use and cognitive decline

Study Groups

H2RA (n= 835)

PPI (n= 3,650)

Inclusion Criteria

Use of H2RA or PPI

Exclusion Criteria

Diagnosis of cancer or receiving chemotherapy, use of both H2RA and PPI at baseline

Methods

Patient data were compiled from National Alzheimer's Coordinating Center Database.

Analyses were performed separately in people with clinically diagnosed mild–moderate Alzheimer's disease (AD), mild cognitive impairment (MCI), and normal cognition.

Participants who had a global Clinical Dementia Rating (CDR) score of 1 or 2 and met NIA and Alzheimer's Association (NIA‐AA) or National Institute of Neurological and Communicative Disorders and Stroke/Alzheimer's Disease and Related Disorders Association criteria were included in the group of clinically diagnosed mild–moderate AD.

Participants who had a global CDR score of 0.5 and met Petersen's criteria were included in the MCI group.

Normal cognitive status was based on clinician judgment, and all participants included in the normal cognition group had a global CDR score of 0.

Duration

September 2005 to February 2021

Outcome Measures

Primary outcome: time to MCI or all‐cause dementia (if dementia was diagnosed without a preceding MCI diagnosis)

Baseline Characteristics

 

Normal cognition

Mild cognitive impairment 

 

H2RA users

(n= 547)

PPI users

(n= 2,237)

H2RA users

(n= 288)

PPI users

(n= 1,413)

Age, years

73.11 73 76.22 75.7

Female

67% 66% 55% 52%

White

81% 83% 78% 83%

Apolipoprotein E 𝜀4 carriers

30% 28% 39%  40% 

Results

Endpoint

H2RA users

PPI users

HR (95% CI)

p-value

Progression to cognitive impairment/dementia in patients with normal cognition

Number of patients

Number of events

Follow-up, years

Events/person-year

5-year risk

 


547

62

2.27 ± 1.88

0.050

-

 


2,237

312

2.80 ± 2.09

0.050

-

 


-

-

-

-

0.937 (0.709 to 1.239) 

 


-

-

-

-

0.648 

Progression to cognitive impairment/dementia in patients with MCI

Number of patients

Number of events

Follow-up, years

Events/person-year

5-year risk

 

288

75

1.48 ± 1.58 

0.175

-

 

1,413

355

1.99 ± 1.79

0.126 

-

 

-

-

-

-

1.402 (1.085 to 1.811)

 

-

-

-

-

0.010

Memory decline in mild-moderate AD

Number of patients

Follow-up, years

 

157

1.53 ± 1.72 

 

471

1.77 ± 1.73 

RR 0.78 (0.67 to 0.925)

-

0.002

Post hoc analyses continued to indicate a trend suggesting that H2RA users exhibited earlier MCI‐to‐dementia progression, but the association did not reach significance (HR [95% CI] = 1.285 [0.997–1.656], p= 0.053).

HR, hazard ratio; RR, risk ratio; CI, confidence interval

Adverse Events

NA

Study Author Conclusions

No differential risk for incident clinical cognitive impairment or memory decline was seen between H2RA and PPI use in older people with normal cognition; however, among people with MCI, H2RA users had earlier progression to dementia over 5 years compared to PPI users. Also, H2RA use was associated with faster memory decline in people with mild–moderate AD but not in people with normal cognition. The current study refutes previous evidence that PPI use is associated with higher dementia risk, and it highlights that people with MCI or AD who require an acid suppressant may be more vulnerable to cognitive harms related to H2RA exposure than to exposure to a PPI.

InpharmD Researcher Critique

Some variables could not easily be controlled for, including the use of prescription versus over‐the‐counter H2RAs/PPIs, disease severity, ulcer diagnosis, and socioeconomic status, potentially confounding results. Furthermore, drug exposure prior to entry into the database was not available, and reporting of over‐the‐counter drugs was optional. 



References:

Wu CY, Xiong LY, Ouk M, et al. Gastric acid suppressants and cognitive decline in people with or without cognitive impairment. Alzheimers Dement (N Y). 2022;8(1):e12243. Published 2022 Feb 10. doi:10.1002/trc2.12243