Association between proton pump inhibitor use and vitamin B12 deficiency: a retrospective cohort study in primary care centers in Riyadh, Saudi Arabia.

BACKGROUND

Proton pump inhibitors (PPIs) are widely prescribed, yet their long-term effects on vitamin B12 status remain debated with inconsistent findings across studies. Data from Saudi Arabia and the Middle Eastern primary-care context are limited.

OBJECTIVES

To evaluate changes in serum vitamin B12 levels following ≥6 months of PPI therapy and assess the association between PPI use and vitamin B12 deficiency.

DESIGN

Retrospective chart-review cohort study with paired pre-post design.

SETTINGS: Family Medicine clinics at King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia (2016-2022).

PATIENTS AND

METHODS

Adults (≥18 years) receiving PPI therapy for ≥6 months with paired vitamin B12 measurements were included. Patients with conditions or medications affecting B12 absorption (metformin, gastrointestinal surgery, atrophic gastritis, vegetarian diet) were excluded. Paired pre-post vitamin B12 levels were compared using Wilcoxon signed-rank and McNemar tests. Neurological symptoms were documented from clinical records.

MAIN OUTCOME MEASURES: Change in serum vitamin B12 concentration, incidence of vitamin B12 deficiency (<200 pg/mL), and prevalence of neurological symptoms.

SAMPLE SIZE: 376 adult PPI users.

RESULTS

Most participants were aged ≥56 years (n=282, 75%), female (n=211, 56.1%), and overweight/obese (n=308, 81.9%). Median vitamin B12 declined significantly from 312 to 297 pg/mL (P<.001), with vitamin B12 deficiency prevalence increasing from 8.2% (n=31) to 12.2% (n=46) (P=.05). New-onset deficiency developed in 9.6% [Author: Please insert n for context, e.g., 9.6% (n)] of patients, with lower baseline B12 predicting greater decline. Neurological symptoms were nearly four times more common among those who developed deficiency (42.4% [95% confidence interval, CI 27.2-59.2] vs. 10.9% [95% CI 7.9-14.8], P<.001). No significant associations were observed with age, sex, BMI, PPI type, or duration. The mean change in serum vitamin B12 was -41.3 pg/mL (95% CI -67.1 to -15.5) and the McNemar paired odds ratio for new-onset deficiency was 1.83 (95% CI 1.00 to 3.46).

CONCLUSIONS

PPI use for six months or longer was associated with significant vitamin B12 decline, with neurological symptoms more frequent among deficient individuals. These findings support periodic vitamin B12 monitoring and structured deprescribing strategies, particularly for high-risk individuals.

LIMITATIONS: Retrospective design, single-center setting, and inability to assess dietary B12 intake or PPI adherence.