# Concordance of the Fracture Risk Assessment Tool With and Without Bone Mineral Density and Its Role in Predicting Postoperative Proximal Junctional Kyphosis After Adult Spinal Deformity Surgery

## TL;DR

This study compares two methods for predicting fracture risk in spinal surgery patients and finds that a simpler method without bone density data works just as well for predicting complications.

## Contribution

The study shows that a simplified fracture risk model without bone density data is as effective as the more complex version for predicting postoperative complications.

## Key findings

- The MOF-BD and MOF + BD models showed excellent agreement in predicting major fractures.
- Baseline pelvic tilt and MOF-BD were independent risk factors for PJK with fracture.
- Age, prior fragility fracture, and secondary osteoporosis were moderately correlated with model differences.

## Abstract

To compare the 10-year risk of major osteoporotic fracture (MOF) without bone density (MOF-BD) data and MOF with BD (MOF + BD) data to determine their validity as risk prediction tools for postoperative proximal junctional kyphosis (PJK) with vertebral fracture after adult spinal deformity (ASD) surgery. ASD surgery can cause mechanical complications, such as PJK. Osteoporosis is linked to PJK; therefore, predicting bone fractures by BD is important. The fracture risk assessment tool was used to assess fracture risk, and the addition of BD information might have increased its accuracy.

This study obtained data from medical records, surgical factors, spinal alignment data, and imaging of patients with ASD.

Ninety-two patients were enrolled in the study (mean age at surgery: 67.8 years; mean follow-up: 8.0 years). A comparison of fracture risk assessment tool calculations in ASD patients demonstrated excellent agreement (r = 0.866; 95% CI: 0.801-0.910; Spearman correlation coefficient) for major fractures between the MOF-BD and MOF + BD groups. Independent risk factors for PJK with fracture after ASD surgery included baseline pelvic tilt (>30°), pelvic tilt at first postoperative standing (>30°), and MOF-BD (>15%). Age (r = 0.400, P < .001), prior fragility fracture (r = 0.467, P < .001), and secondary osteoporosis (r = 0.388, P < .001) were moderately correlated with dissociation between MOF-BD and MOF + BD (all Spearman correlation coefficient). No significant difference in the cumulative hazard of MOF + BD and MOF-BD was observed (P = .685; log-rank test).

The simpler model of 10-year risk of MOF without BD was in good agreement with BD and may aid general health screening as well as preoperative risk assessment for spinal reconstructive surgery.