Future Directions in Hypercalcemic and Normocalcemic Primary Hyperparathyroidism: FRAXplus for 10-Year Fracture Risk Assessment (A Retrospective Study).

Background

Osteoporosis/osteoporotic fractures are identified in both hypercalcemic (HC-HPT) and normocalcemic variant (NC-HPT) of primary hyperparathyroidism (HPT) at various rates.

Objective

Noting the need of modern society to easily assess the osteoporotic fracture risk amid the diagnosis of HPT, we aimed to address this gap by analyzing the 10-year fracture risk assessment based on traditional FRAX (Fracture Risk Assessment Tool) model in comparison to the novel algorithm (FRAXplus), according to the adjustment for the presence of HPT, as well as for the use of lumbar bone mineral density (BMD) in menopausal women with HPT versus controls (non-HPT), respectively, between HC-HPT versus NC-HPT.

Methods

For each patient, the latest algorithms of FRAX and FRAXplus provided the 10-year fracture risk for major osteoporotic fractures (MOF) and for hip fracture (HF) amid a single-center, retrospective, real-life study.

Results

In total, 131 subjects were included: 51.15% had HPT (64.18% of them had HC-HPT) versus age-, menopause duration-, and body mass index-matched (HPT-free) controls. As a result, 10-year fracture risk for MOF and HF was statistically significantly higher in HPT versus controls only for the calculation with femoral neck BMD. FRAXplus showed that for both estimations (MOF and HF) with introduction of lumbar BMD remained higher than controls (4.55% vs. 3.7%, p = 0.004, respectively, 1.05% vs. 0.5%, p = 0.002). In HPT group, 10-year fracture risk for MOF and HF were higher if adjustment for HPT was applied. The highest 10-year fracture risk for MOF was obtained for HPT adjustment with femoral neck BMD (5.9%) versus the estimation without using femoral neck BMD (5.25%, p = 0.001), respectively, versus the probability with adjustment for lumbar BMD (4.55%, p < 0.001). The same observation was for HF: 1.4% versus 1.2% (p = 0.028), respectively, versus 1.05% (p < 0.001). In HPT group, parathormone level positively correlated with 10-year hip fracture risk with HPT adjustment, without femoral neck BMD (r = 0.257, p = 0.049). Bone formation marker P1NP negatively correlated with 10-year fracture risk for MOF without femoral neck BMD (r = -0.416, p = 0.043), respectively, with the estimation including HPT adjustment without femoral neck BMD (r = -0.404, p = 0.05), and with the 10-year HF risk calculated without femoral neck BMD (r = -0.407, p = 0.049).

Conclusions

To our best knowledge, this is the first study to address the use of FRAXplus in HPT. The similar values between FRAX-based probabilities without the use of femoral neck BMD in HPT versus non-HPT controls suggested that this traditional estimation might not be so useful in HPT population, thus the need for novel models (HPT adjustment). HPT adjustment (FRAXplus) provided a higher MOF/HF risk versus non-adjustment (FRAX). All 10-year probabilities based on FRAX and FRAXplus models showed similar values in HC-HPT versus NC-HPT, which implies that current algorithms might not make a clear distinction between HPT subtypes, yet the statistically significant results within each of these subgroups sustain the FRAXplus application regardless of the variant.