Endocrine Abstracts

Current clinical presentation of PHPT is mild disease with an increased risk of fragility fractures. PHPT predominantly occurs in postmenopausal women, who have an increased risk of osteoporosis and fractures due to ageing and loss of estrogen.

Aims: To explore skeletal phenotypes in postmenopausal women affected by PHPT with a wide clinical and biochemical spectrum of disease.

Patients: Postmenopausal (at least 5 years from last menses) women with PHPT were retrospectively evaluated at three third level Italian centers for management of osteoporosis and mineral metabolism disorders (n=120 from Milan, n=134 from Cuneo, n=132 from Pisa).

Methods: Data were collected from clinical records and analyzed by nonparametric Spearman correlation and multiple linear regression. Hierarchical clusterization by Wards’ method and Euclidean similarity index identified skeletal phenotype clusters; differences among clusters were detected by Kruskall-Wallis ANOVA.

Results: Considering the whole 386 PHPT women [aged 68.0 (61.0-74.0) years], fractures (clinical and morphometric vertebral, femur, humerus, pelvic, wrist fractures) positively correlated with phosphatemia and negatively with lumbar and neck T-scores. Cluster analysis based on fractures number, phosphatemia, lumbar and neck T-scores, identified 4 clusters: the most frequent phenotype (n=307) included women with lumbar osteoporosis and neck osteopenia with a prevalence of fractures of 29.5% (0-3 fractures, minimum-maximum) and higher 24-hours urinary calcium corrected for body weight (UCa). The second phenotype included women (n=16) with more important lumbar and neck osteoporosis who all experienced multiple (3-6) fractures. The third phenotype included women (n=37) with neck osteopenia/osteoporosis and mild lumbar osteopenia with a prevalence of fractures of 23.3% associated with lower UCa and estimated glomerular filtration rate (eGFR). The forth phenotype included women (n=19) with lumbar and neck osteopenia with a very low prevalence of fractures (2.1%, no more than 1 fracture). In this cluster, women were younger and heavier, with lower PTH levels. Unexpectedly, ionized and total calcium, phosphate, bone and total alkaline phosphatase, 25hydroxyvitamin D (25OHD) levels, and kidney stones prevalence (ranging 21.1-30.2%) were similar among the 4 clusters. Fractures were predicted by lumbar T-score. Besides, lumbar T-score was predicted by body mass index, 25OHD, eGFR and neck T-score.

Conclusions: Skeletal involvement in this large retrospective series of postmenopausal women with PHPT presented with heterogeneous phenotypes associated with different prevalence of fractures. Fractures were related with bone mineral density, but not with PHPT severity, suggesting that other factors besides PHPT should be considered in the evaluation of the bone involvement in postmenopausal women with PHPT.