Do proton pump inhibitors decrease calcium absorption?

Abstract

Proton pump inhibitors (PPIs) increase osteoporotic fracture risk presumably via hypochlorhydria and consequent reduced fractional calcium absorption (FCA). Existing studies provide conflicting information regarding the direct effects of PPIs on FCA. We evaluated the effect of PPI therapy on FCA. We recruited women at least 5 years past menopause who were not taking acid suppressants. Participants underwent three 24-hour inpatient FCA studies using the dual stable isotope method. Two FCA studies were performed 1 month apart to establish baseline calcium absorption. The third study occurred after taking omeprazole (40 mg/day) for 30 days. Each participant consumed the same foods during all FCA studies; study meals replicated subjects' dietary habits based on 7-day diet diaries. Twenty-one postmenopausal women ages 58 ± 7 years (mean ± SD) completed all study visits. Seventeen women were white, and 2 each were black and Hispanic. FCA (mean ± SD) was 20% ± 10% at visit 1, 18% ± 10% at visit 2, and 23% ± 10% following 30 ± 3 days of daily omeprazole (p = .07, ANOVA). Multiple linear regression revealed that age, gastric pH, serum omeprazole levels, adherence to omeprazole, and 25-hydroxyvitamin D levels were unrelated to changes in FCA between study visits 2 and 3. The 1,25-dihydroxyvitamin D(3) level at visit 2 was the only variable (p = .049) associated with the change in FCA between visits 2 and 3. PPI-associated hypochlorhydria does not decrease FCA following 30 days of continuous use. Future studies should focus on identifying mechanisms by which PPIs increase the risk of osteoporotic fracture.

Fig. 1

Study recruitment.

Fig. 2

We measured fractional calcium absorption in participants (n = 21) on three separate occasions. Each subject completed her third calcium absorption study after taking omeprazole 40 mg daily for 30 days. The study group's mean FCA was 20% at visit one, 18% at visit two, and 23% following omeprazole therapy (p = .07, ANOVA). In this figure, we illustrate absorption using the means and 95% confidence intervals.

Fig. 3

We assessed bone resorption by measuring participants' (n = 21) fasting morning urine CTX levels. Subjects' mean CTX values at visits 1, 2, and 3 were 1.9, 1.8, and 1.9 µg/mmol of creatinine, respectively (p = .80, ANOVA). In this figure we summarize CTX levels using the means and 95% confidence intervals.