Validation of Adult Spinal Deformity Surgical Outcome Prediction Tools in Adult Symptomatic Lumbar Scoliosis.

STUDY

DESIGN

A post hoc analysis.

OBJECTIVE

Advances in machine learning (ML) have led to tools offering individualized outcome predictions for adult spinal deformity (ASD). Our objective is to examine the properties of these ASD models in a cohort of adult symptomatic lumbar scoliosis (ASLS) patients.

SUMMARY OF BACKGROUND DATA: ML algorithms produce patient-specific probabilities of outcomes, including major complication (MC), reoperation (RO), and readmission (RA) in ASD. External validation of these models is needed.

METHODS

Thirty-nine predictive factors (12 demographic, 9 radiographic, 4 health-related quality of life, 14 surgical) were retrieved and entered into web-based prediction models for MC, unplanned RO, and hospital RA. Calculated probabilities were compared with actual event rates. Discrimination and calibration were analyzed using receiver operative characteristic area under the curve (where 0.5=chance, 1=perfect) and calibration curves (Brier scores, where 0.25=chance, 0=perfect). Ninety-five percent confidence intervals are reported.

RESULTS

A total of 169 of 187 (90%) surgical patients completed 2-year follow up. The observed rate of MCs was 41.4% with model predictions ranging from 13% to 68% (mean: 38.7%). RO was 20.7% with model predictions ranging from 9% to 54% (mean: 30.1%). Hospital RA was 17.2% with model predictions ranging from 13% to 50% (mean: 28.5%). Model classification for all three outcome measures was better than chance for all [area under the curve=MC 0.6 (0.5-0.7), RA 0.6 (0.5-0.7), RO 0.6 (0.5-0.7)]. Calibration was better than chance for all, though best for RA and RO (Brier Score=MC 0.22, RA 0.16, RO 0.17).

CONCLUSIONS

ASD prediction models for MC, RA, and RO performed better than chance in a cohort of adult lumbar scoliosis patients, though the homogeneity of ASLS affected calibration and accuracy. Optimization of models require samples with the breadth of outcomes (0%-100%), supporting the need for continued data collection as personalized prediction models may improve decision-making for the patient and surgeon alike.