Search

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.

Introduction: The effect of pain on HRQoL scores in ASD patients is not well studied. Disability is a major factor on decision and outcomes. On the other hand, little is known about the effect of perceived and reported pain on these parameters, especially in the elderly population. We hypothesized that baseline back and leg pain would not affect the treatment decision whereas may have a negative effect on outcomes., Research Question: To determine the correlation between preoperative ODI and VAS scores; and to identify the effect of baseline VAS score on treatment decision and ODI improvement following treatment., Material and Methods: In this retrospective study, patients with a follow-up duration of minimum 2 years were enrolled from a prospective multicentric ASD database. Pearson and Spearman correlation tests were used to evaluate the correlation between ODI and VAS scores; univariate binary logistic regression method was used to analyze the effect of VAS on treatment decision as well as the outcomes., Results: 1050 patients (mean age 48.2) were analyzed. Baseline ODI and back, leg pain VAS scores were significantly correlated ( P  ​< ​0.001). One unit increase in baseline back and leg pain VAS scores, increased the probability of improvement in ODI by 1.219 (P ​= ​0.016) and 1.182 times (P ​= ​0.029), respectively in surgically treated patients; and reduced it by 0.894 times (P ​= ​0.012) for conservatively treated patients. For patients >70 years old, one-unit increase in baseline leg pain VAS score increased the probability of deciding on surgical treatment by 1.121 times (p ​= ​0.016)., Discussion and Conclusions: Preoperative back and leg pain VAS scores were found to be significantly correlated with the preoperative ODI scores. Additionally, preoperative baseline back and leg pain VAS scores were useful in predicting the improvement in disability as assessed by ODI. Another important finding was that, higher baseline leg pain (but not back pain) VAS scores increased the rate of elderly patients preferring surgical treatment., Competing Interests: The authors of this manuscript have no competing interests that influence the results and discussion of this paper., (© 2022 The Authors.)