Oral findings and antiresorptive therapy exposure duration in postmenopausal women with osteoporosis: a cross-sectional cohort study

Abstract Background Oral findings are common among older adults, but it is unclear whether they vary with the duration of antiresorptive therapy (ART) exposure or relate to the severity of osteoporosis in postmenopausal women. Methods In this cross-sectional cohort ( n = 129), postmenopausal women undergoing osteoporosis assessment/follow-up were stratified by ART exposure duration: no current ART control (G1, n = 50), short ART (G2, 3–5 years; n = 39), and long ART (G3, > 5 years; n = 40). The prespecified primary oral endpoint was any mucosal finding ( oral_mucosal_any ), defined as ≥ 1 prespecified mucosal lesion category at oral examination. The oral_mucosal_any endpoint was analysed by component to distinguish traumatic from non-traumatic mucosal findings. Secondary oral endpoints included tooth and molar counts, periodontal findings, and the number of mucosal lesion categories. Group differences were analysed using age- and BMI-adjusted regression (G1 reference). Secondary analyses compared oral findings across fracture-burden strata and hip BMD/T-score categories and evaluated whether oral variables improved discrimination for hip osteoporosis (total hip T-score ≤ − 2.5) beyond age, BMI, and ART group. Results Mucosal findings were common and non-monotonic across ART groups (G1: 52.0%, G2: 30.8%, G3: 52.5%). Among the 59 participants with any mucosal finding, traumatic changes were present in 43 (72.9%); in a mutually exclusive classification, 38 (64.4%) had traumatic-only findings. In age- and BMI-adjusted models, short ART was associated with lower odds of oral_mucosal_any versus controls (OR 0.38, 95% CI 0.16–0.93; p = 0.035), whereas long ART did not differ (OR 0.94, 95% CI 0.38–2.33; p = 0.894). Tooth and molar counts, periodontal burden, and non-traumatic mucosal findings showed no robust graded pattern with ART duration. Oral findings were broadly similar across fracture-burden strata and hip BMD categories. Adding total teeth and oral_mucosal_any increased apparent Area Under the Curve (AUC) for hip osteoporosis from 0.795 to 0.820, but optimism-corrected AUC changed only from 0.736 to 0.741 (corrected ΔAUC = 0.005). Conclusions In this cross-sectional cohort, common oral findings did not show a robust graded duration–response pattern across ART exposure groups and were not consistently associated with fracture burden or hip BMD strata. The broad mucosal endpoint was largely driven by common traumatic findings and should be interpreted as a clinically pragmatic but biologically heterogeneous endpoint. Oral variables added negligible optimism-corrected discrimination beyond age, BMI, and ART group.