Summary
Mortality Prediction in Femoral Neck Fracture Patients, an Intercontinental Validation
Abstract
Background
Mortality prediction in femoral neck fracture patients aged 65 years or above is valuable in treatment decision-making. The previously developed and internally validated clinical machine learning prediction model (SORG femoral neck fracture mortality algorithm) shows promise in identifying patients at risk of 90-day and 2-year mortality. Validation in an independent cohort is required to assess the generalizability of the model; especially in geographically distinct regions.
Question:
Is the SORG femoral neck fracture mortality algorithm externally valid in an Israeli cohort to predict 90-day and 2-year mortality?
Methods
The validation cohort included 2,033 patients that were 65 years of age or older and underwent femoral neck fracture surgery at the Sheba Medical Center in Israel (level I trauma center). Pathologic fractures were excluded. Most patients were female 64.8% (1,317 of 2,033), the median age was 81 years (interquartile range=75-86), and 80.4% (1,635 of 2,033) patients sustained a displaced femoral neck fracture (Garden III/IV). The 90-day mortality was 9.4% (190 of 2,033) and 2-year mortality was 30.0% (610 of 2,033). Generalizability of the SORG femoral neck fracture mortality algorithm was assessed using discrimination, calibration, Brier-score, and decision curve analysis.
Results
Despite numerous baseline differences, the SORG femoral neck fracture mortality algorithm performed acceptably to the validation cohort on discrimination (c-statistic 0.67 for 90-day, 0.67 for 2-year), calibration, Brier-score, and decision curve analysis.
Conclusion
The previously developed SORG femoral neck fracture mortality algorithm demonstrated good performance in an independent intercontinental population. The current iteration of SORG should not be relied on for patient care, though suggesting potential utility in assessing patients at low risk for 90-day or 2-year mortality. Further studies should evaluate this algorithm in a prospective setting and evaluate its feasibility and efficacy in clinical practice. The algorithm can be freely accessed at https://sorg-apps.shinyapps.io/hipfracturemortality/.