Association of Long-Term Proton Pump Inhibitor Use With Nutrient Deficiencies: A Retrospective Cross-Sectional Study.

Abstract

Background Proton pump inhibitors (PPIs) are generally employed to treat acid-related disorders. However, chronic acid blockade can reduce the absorption of essential micronutrients, such as vitamin B12, magnesium, calcium, and iron, potentially contributing to clinically significant deficiencies as defined by standard laboratory reference ranges. Methods A retrospective observational study was conducted using electronic medical records collected between July and December 2025. The study included adults (18 years and older) who had been taking PPIs continuously for six months or more. Nutrient deficiencies were defined according to institutional laboratory reference ranges, with the following cut-off values: vitamin B12 <200 pg/mL, magnesium <1.7 mg/dL, calcium <8.6 mg/dL, and ferritin <30 ng/mL. Both centers used identical reference ranges and standardized laboratory protocols, and all results were cross-checked to minimize inter-laboratory variability; any inconsistent or repeated values were excluded from analysis. The associations between PPI type/duration and deficiency status were analyzed using the Fisher-Freeman-Halton exact test, the chi-square test, and the Kruskal-Wallis test (p<0.05). Findings The sample consisted of 255 participants: 139 (54.5%) female patients and 116 (45.5%) male patients. The patients were on omeprazole (n=101; 39.6%), pantoprazole (n=64; 25.1%), esomeprazole (n=61; 23.9%), and lansoprazole (n=29; 11.4%). The prevalence of vitamin B12 deficiency varied by PPI type: 39 (38.6%) of omeprazole users and 36 (59.0%) of esomeprazole users were vitamin B12 deficient (p=0.018). Users of esomeprazole had the highest prevalence of magnesium deficiency (n=24; 39.3%) vs users of omeprazole (n=12; 11.9%; p=0.004). Calcium deficiency according to the type of PPI also differed, wherein eight (7.9%) patients on omeprazole and nine (31.0%) patients on lansoprazole were affected (p=0.021); the median serum calcium was found to be between 9.06 mg/dL and 8.71 mg/dL (H=11.29, p=0.010). There were no significant differences in iron/ferritin deficiency according to the type of PPI (33/101 (32.7%) in omeprazole users vs 29/61 (47.5%) in esomeprazole users) or PPI duration (p≈0.12). Conclusion Long-term use of PPIs was associated with a high prevalence of micronutrient deficiencies, especially B12, magnesium, and calcium, with substantial differences by PPI type. These results highlight the importance of routine nutritional monitoring in patients receiving long-term PPI therapy.