Background/Purpose: There are no therapies available that specifically target CPPD, with treatment limited to symptomatic management. Hypomagnesemia is a recognized risk factor for CPPD, and its treatment has been advocated for CPPD management. Proton pump inhibitors (PPIs) are an under-appreciated cause of hypomagnesemia, and may therefore be an important risk factor for CPPD. We examined the relation of PPI use to the development of CPPD.

Methods: We conducted a time-stratified propensity score(PS)-matched cohort study in The Health Improvement Network (THIN), a general practitioner (GP) EMR database representative of the UK population. Using greedy matching, we matched incident PPI users to incident H2 blocker users by PS (see table) to reduce confounding by indication in 1-year cohort accrual blocks to account for secular trends. We identified incident cases of individuals diagnosed with pseudogout or chondrocalcinosis (hereafter labelled CPPD) using READ codes from 1995-2015 among those aged 50-89 who were enrolled in a GP practice for ≥1 year prior to study entry. We excluded subjects with a gout diagnosis. We compared the risk of incident CPPD among incident PPI users vs. incident users of H2 blockers using Cox proportional hazard models, censoring at time of drug switch, additionally adjusting for the PS model variables. Because of the small number of cases, we also conducted a nested case-control study within the same cohort, matched 1:4 by age and gender using risk-set sampling, evaluating incident use of PPI vs. non-use and vs. H2-blockers prior to incident CPPD.

Results: We identified 81,011 PPI initiators, who were PS-matched 1:1 to H2 blocker initiators, with mean age 66, mean BMI 26.7, 59% female. Overall, covariates were well-balanced in the two groups (SMDs <0.1). There were 118 and 58 incident cases of CPPD among the PPI and H2 blocker initiators, respectively. Incident PPI use was associated with a HR of 1.19 (95% CI 0.86-1.64) of incident CPPD compared with incident H2 blocker use. In the case-control study, when comparing PPI initiators to non-users, the adjusted OR was 1.58 (95% CI 1.39-1.79). When also considering H2 blocker use, the adjusted ORs were elevated for both PPI and H2 blocker users compared with non-users (ORs 1.79 and 1.52, respectively (see Table)); the ratio of ORs for PPI:H2 blocker initiators was similar to PS-matched study HR (ratio of ORs ~1.2; HR was ~1.2).

Conclusion: In this population-based study of GP-recorded CPPD, compared with incident H2 blocker use, incident use of PPIs was associated with a modest but nonsignificant increased risk of CPPD. We were limited by small number of cases. The elevated risk with H2 blocker use may represent residual confounding (e.g., NSAID with gastroprotective use for undiagnosed CPPD) or potentially a true biologic effect, e.g., due to general gastric acid suppression affecting Mg absorption. Nonetheless, the risk appears higher with PPIs than H2 blockers.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/proton-pump-inhibitors-and-risk-of-calcium-pyrophosphate-deposition-cppd-in-a-population-based-study/