Background/Purpose: Patients with inflammatory rheumatic and musculoskeletal diseases (iRMDs) are at increased risk of osteoporosis and fragility fractures due to systemic inflammation and glucocorticoid (GC) use. Proton pump inhibitors (PPIs), frequently prescribed in this population, have also been linked to fracture risk, although findings remain inconsistent. Our previous cross-sectional analysis suggested a negative impact of PPIs on bone mineral density (BMD) but not on microarchitecture in iRMDs.This study aimed to assess the longitundinal effects of PPI use on BMD and bone microarchitecture.

Methods: The prospective Rh-GIOP cohort includes patients with iRMDs fulfilling German guidelines for osteoporosis check-up. Participants underwent dual-energy X-ray absorptiometry (DXA) and evaluation of microstructure by trabecular bone score (TBS) and 3D-DXA. This analysis included all patients with IRMDs who had data available up to the second follow-up visit (2015–2024). Follow-up timing was determined by clinical routine. Patients with manifest hyperparathyroidism and hyperthyroidism were excluded.Daily intake of PPI was ascertained via chart review and self-report. As-needed PPI use was combined with regular use to form the composite variable “any PPI use”. Primary outcomes were T-scores at the lumbar spine, total hip, TBS and 3D-DXA parameters of the femur. Linear mixed-effects models assessed the effect of any PPI use as well as daily PPI use on the primary outcomes and were weighted by propensity scores accounted for potential confounders such as age, sex, BMI, GC dose, C-reactive protein, disability and comorbidity. Missing data was handled by multiple imputation (predicitive mean matching).

Results: At baseline, 1,909 patients (74% women; mean age 63 ± 13 years) were included; 45% reported daily and 51% any PPI use. 713 had at least one follow-up (mean 2.7 ± 1.3 years), and 267 had a second (mean 4.6 ± 1.4 years).Any PPI use was associated with lower T-Scores at the lumbar spine (estimate: −0.115; standard error (SE): 0.052; P = .029). In contrast, total hip T-Score was not significantly affected by any PPI use (−0.027; SE: 0.024; P = 0.258) or daily use (−0.038; SE: 0.026; P = 0.37). TBS was related to a negative impact by any PPI use (−0.017; SE: 0.006; P = .004), and daily use (−0.021; SE: 0.006; P = .001). Femoral cortical thickness was also negatively affected by both any PPI use (−0.012; SE: 0.006; P = .050) and daily PPI use (−0.015; SE: 0.006; P = .017). Adjusted interaction analysis showed that PPI-related reduction in TBS also occurred in low-dose GC users (0-7.5 mg/day), suggesting a synergistic impact. For cortical thickness, PPI effects were particularly relevant in the absence of GC use. Daily PPI use was further linked to lower calcium levels (−0.010; SE: 0.004; P = .016) and higher parathyroid hormone levels (2.275; SE: 0.964; P = .019).

Conclusion: Our results indicate that PPIs may also impair bone microarchitecture in patients with iRMDs. This effect appears to be most prominent in individuals without GC use or on low-dose GCs, potentially linked to reduced calcium absorption. This underscores the need for cautious PPI use, especially in patients already at high risk for osteoporosis.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/effect-of-proton-pump-inhibitors-on-bone-mineral-density-and-microarchitecture-in-patients-with-inflammatory-rheumatic-and-musculoskeletal-diseases/