• Alice Baroncini, Louis Boissiere, Daniel Larrieu, Sleiman Haddad, Ferran Pellisé, Ahmet Alanay, Frank Kleinstueck, Javier Pizones, Anouar Bourghli, Ibrahim Obeid, and European Spine Study Group (ESSG).
• IRCCS Ospedale Galeazzi Sant'Ambrogio, Milano, Italy.
• Spine. 2024 Oct 1.

Study DesignMulticentric, retrospective analysis of prospectively collected data.ObjectiveTo utilize machine-learning (ML) for clustering and management prediction (conservative vs. operative) in surgically treated adult spine deformity (ASD) patients, and to compare the attainment of the Minimum Clinically Important Difference (MCID) between predicted surgical and conservative patients.Summary Of Background DataManagement choice in ASD is complex. ML can identity patient clusters and predicted treatment, but it is unclear whether patients treated according to the prediction also show better clinical outcomes.MethodsASD patients (2-year follow-up) were divided into groups using k-means clustering. Management choice was predicted among operated patients in each cluster. The MCID for the Oswestry Disability Index (ODI) and the Scoliosis Research Society-22 (SRS-22) was calculated and compared between patients with and without surgical prediction.ResultsIn Cluster 1 (idiopathic scoliosis, n=675, 150 surgeries), 57% of patients had a conservative prediction. Of these, 52% and 49% achieved MCID for ODI and SRS-22, respectively, compared to 68% and 75% for those with surgical predictions (OR=2 and 3.1, respectively).In Cluster 2 (moderate sagittal imbalance, n=561, 200 surgeries), 12% had a conservative prediction. Of these, 29% and 46% achieved MCID for ODI and SRS-22, respectively, compared to 47% and 56% for those with surgical predictions.In Cluster 3 (significant sagittal imbalance, n=537, 197 surgeries), 17% had a conservative prediction. Of these, 12% and 15% achieved MCID for ODI and SRS-22, respectively, compared to 37% and 45% for those with surgical predictions (OR=4.2 and 4.5, respectively).ConclusionPatients with a concordant surgical prediction and management had higher odds of achieving the MCID, indicating a good correlation between prediction and clinical outcomes. In Cluster 3, the low percentage of patients with conservative prediction achieving the MCID suggests that machine learning could well identify patients with poor clinical outcomes.Copyright © 2024 Wolters Kluwer Health, Inc. All rights reserved.

9 keywords...

hide…