Reference:
T. Makunts, Sama Alpatty, Kelly C. Lee, Rabia S. Atayee, and R. Abagyan. “Proton-Pump Inhibitor Use Is Associated with a Broad Spectrum of Neurological Adverse Events Including Impaired Hearing, Vision, and Memory.”
Scientific Reports, 2019.

This single study employed an observational design analyzing voluntary adverse event reports from the FDA FAERS database. The investigators selected monotherapy reports to minimize confounding from concurrent medications and validated cohort comparability through overlapping demographic parameters. The study examined six commonly prescribed PPIs and used H2RA users as a comparison group to control for the underlying indication (acid suppression therapy).

Effects
The study reported a substantial association between PPI use and peripheral neuropathy-related adverse events:

Outcome
Neuropathic/neurological
impairment adverse drug
reactions

Odds Ratio
8.68

95% Confidence Interval
[3.86, 19.49]

Interpretation
Over eight-fold increased reporting of neuropathy in PPI users compared to H2RA users

The odds ratio of 8.68 indicates that reports containing PPIs had substantially higher odds of including peripheral neuropathy adverse events compared to reports containing H2RAs. The confidence interval excludes the null value of 1.0, suggesting statistical significance. However, the study did not report specific dose-response relationships, duration-response relationships, or incidence rates in exposed versus unexposed populations. The number needed to harm was not calculated.

The study did not provide granular details about the neuropathy outcomes themselves, including specific types
of peripheral neuropathy (sensory, motor, sensorimotor, or autonomic), anatomical distribution, severity grading, diagnostic methods, specific symptoms, time to onset after PPI initiation, or reversibility upon discontinuation. The broad category of ”neuropathic/neurological impairment” encompasses cranial and peripheral neuropathies, sciatica, and nerve injury.

Regarding PPI exposure characteristics, the study did not report dosage, frequency, route of administration, indication for use, or timing of exposure relative to neuropathy onset. The authors noted that PPI treatment duration should not exceed two or three weeks according to recommendations, with long-term use beyond eight weeks warranting particular caution.

Methodological considerations
The pharmacovigilance design using FAERS data introduces several important considerations for interpreting these findings. The study did not ascertain a temporal relationship between PPI use and neuropathy onset, nor did it observe reversibility upon discontinuation, as these determinations would require longitudinal clinical data rather than cross-sectional adverse event reports. The absence of follow-up duration data and participant demographic information, including age distribution, gender, comorbidities, concurrent medications, clinical setting, and geographic location, limits assessment of effect modification and generalizability.

The investigators attempted to control confounding by restricting analysis to monotherapy reports, though they acknowledged that concurrent medications and comorbidities may be underreported in FAERS. No statistical adjustment methods for confounding or sensitivity analyses were reported. The authors explicitly characterized their findings as associative rather than causal, noting the inherent limitations of pharmacovigilance data for establishing causality.

The proposed biological mechanism involves vitamin B12 deficiency, which is plausible given that B12 deficiency can cause reversible peripheral neuropathy. However, the authors noted that few case reports in the literature directly link PPI use to peripheral neuropathy, indicating limited consistency with previous evidence. The study did not apply Bradford Hill criteria or other formal causal frameworks.