Predicting self-image satisfaction after adult spinal deformity surgery: a machine learning approach using patient...

Area of Science:

• Orthopedics
• Spine Surgery
• Patient-Reported Outcomes

Background:

• Adult spinal deformity (ASD) significantly impacts patient self-image (SI).
• Surgical correction aims to improve functional outcomes and patient satisfaction.
• Understanding predictors of patient-acceptable symptom state (PASS) is crucial for evaluating surgical success.

Purpose of the Study:

• To analyze the changes in the SRS-22 SI domain after surgical correction in ASD patients.
• To identify preoperative and postoperative predictors of achieving PASS for SI.
• To stratify patients into phenotypic clusters for targeted analysis.

Main Methods:

• Retrospective analysis of a prospective, multicenter registry of 710 surgically treated ASD patients.
• Stratification into Young Coronal (YC), Old Moderate (OM), and Old Severe (OS) phenotypic clusters.
• Development of Random Forest models to predict SI PASS achievement (score ≥ 3.3).

Main Results:

• Preoperative SI PASS achievement was low (13%), varying by cluster (YC: 28%, OM: 23%, OS: 1%).
• Postoperative SI PASS achievement significantly increased to 65% (YC: 85%, OM: 68%, OS: 52%).
• Functional disability (ODI), pain (VAS), and physical/social functioning (SF-36) were stronger SI predictors than radiographic parameters.

Conclusions:

• Self-Image is significantly impaired in ASD but substantially improves with surgical intervention.
• Patient-reported functional status and pain are more critical to SI satisfaction than spinal alignment.
• SI is a vital indicator of the holistic patient experience in ASD, challenging purely radiograph-centric evaluations.