TY - JOUR
T1 - Predicting Postoperative Neurological Outcomes in Metastatic Spinal Tumor Surgery Using Machine Learning
AU - JASA Study Group
AU - Maki, Satoshi
AU - Shiratani, Yuki
AU - Orita, Sumihisa
AU - Suzuki, Akinobu
AU - Tamai, Koji
AU - Shimizu, Takaki
AU - Kakutani, Kenichiro
AU - Kanda, Yutaro
AU - Tominaga, Hiroyuki
AU - Kawamura, Ichiro
AU - Ishihara, Masayuki
AU - Paku, Masaaki
AU - Takahashi, Yohei
AU - Funayama, Toru
AU - Miura, Kousei
AU - Shirasawa, Eiki
AU - Inoue, Hirokazu
AU - Kimura, Atsushi
AU - Iimura, Takuya
AU - Moridaira, Hiroshi
AU - Nakajima, Hideaki
AU - Watanabe, Shuji
AU - Akeda, Koji
AU - Takegami, Norihiko
AU - Nakanishi, Kazuo
AU - Sawada, Hirokatsu
AU - Matsumoto, Koji
AU - Funaba, Masahiro
AU - Suzuki, Hidenori
AU - Funao, Haruki
AU - Oshigiri, Tsutomu
AU - Hirai, Takashi
AU - Otsuki, Bungo
AU - Kobayakawa, Kazu
AU - Uotani, Koji
AU - Manabe, Hiroaki
AU - Tanishima, Shinji
AU - Hashimoto, Ko
AU - Iwai, Chizuo
AU - Yamabe, Daisuke
AU - Hiyama, Akihiko
AU - Seki, Shoji
AU - Kato, Kenji
AU - Miyazaki, Masashi
AU - Watanabe, Kazuyuki
AU - Nakamae, Toshio
AU - Kaito, Takashi
AU - Nakashima, Hiroaki
AU - Nagoshi, Narihito
AU - Inoue, Gen
N1 - Publisher Copyright:
© 2025 Wolters Kluwer Health, Inc. All rights reserved.
PY - 2025
Y1 - 2025
N2 - Study Design. Retrospective analysis of data collected across multiple centers. Objective. To develop machine learning models for predicting neurological outcomes one month postoperatively in patients with metastatic spinal tumors undergoing surgery, and to identify key factors influencing neurological recovery. Summary of Background Data. The increasing prevalence of spinal metastases has led to a growing need for surgical intervention to address mechanical instability and neurological deficits. Predicting postoperative neurological status, as assessed by the Frankel classification, can provide valuable insights for surgical planning and patient counseling. Traditional prognostic models have shown limitations in capturing the complexity of neurological recovery patterns. Methods. We analyzed data from 244 patients who underwent spinal surgery for metastatic disease across 38 institutions. The primary outcome was functional ambulation, defined as Frankel grades D or E at one month postoperatively. Four machine learning algorithms (Random Forest, XGBoost, LightGBM, and CatBoost) were used to build predictive models. Feature selection employed the Boruta algorithm and Variance Inflation Factor analysis to reduce multicollinearity. Results. Among the 244 patients, the proportion of ambulatory patients (Frankel grades D or E) increased from 36.8% preoperatively to 63.1% at one month postoperatively. The Random Forest model achieved the highest area under the receiver operating characteristic curve (AUC-ROC) of 0.8516, followed by XGBoost (0.8351), CatBoost (0.8331), and LightGBM (0.8098). SHapley Additive exPlanations analysis identified preoperative Frankel classification, transfer ability, inflammatory markers (C-reactive protein, white blood cell-lymphocyte), and surgical timing as the most important predictors of postoperative outcomes. Conclusions. Machine learning models showed strong predictive performance in assessing postoperative neurological status for patients with metastatic spinal tumors. Key factors including preoperative neurological function, functional ability, and inflammation markers significantly influenced outcomes. These findings could inform surgical decision-making and help set realistic postoperative expectations while potentially improving patient care through more accurate outcome prediction.
AB - Study Design. Retrospective analysis of data collected across multiple centers. Objective. To develop machine learning models for predicting neurological outcomes one month postoperatively in patients with metastatic spinal tumors undergoing surgery, and to identify key factors influencing neurological recovery. Summary of Background Data. The increasing prevalence of spinal metastases has led to a growing need for surgical intervention to address mechanical instability and neurological deficits. Predicting postoperative neurological status, as assessed by the Frankel classification, can provide valuable insights for surgical planning and patient counseling. Traditional prognostic models have shown limitations in capturing the complexity of neurological recovery patterns. Methods. We analyzed data from 244 patients who underwent spinal surgery for metastatic disease across 38 institutions. The primary outcome was functional ambulation, defined as Frankel grades D or E at one month postoperatively. Four machine learning algorithms (Random Forest, XGBoost, LightGBM, and CatBoost) were used to build predictive models. Feature selection employed the Boruta algorithm and Variance Inflation Factor analysis to reduce multicollinearity. Results. Among the 244 patients, the proportion of ambulatory patients (Frankel grades D or E) increased from 36.8% preoperatively to 63.1% at one month postoperatively. The Random Forest model achieved the highest area under the receiver operating characteristic curve (AUC-ROC) of 0.8516, followed by XGBoost (0.8351), CatBoost (0.8331), and LightGBM (0.8098). SHapley Additive exPlanations analysis identified preoperative Frankel classification, transfer ability, inflammatory markers (C-reactive protein, white blood cell-lymphocyte), and surgical timing as the most important predictors of postoperative outcomes. Conclusions. Machine learning models showed strong predictive performance in assessing postoperative neurological status for patients with metastatic spinal tumors. Key factors including preoperative neurological function, functional ability, and inflammation markers significantly influenced outcomes. These findings could inform surgical decision-making and help set realistic postoperative expectations while potentially improving patient care through more accurate outcome prediction.
KW - frankel classification
KW - machine learning
KW - metastatic spinal tumor
KW - postoperative neurological outcome
KW - prediction model
UR - http://www.scopus.com/inward/record.url?scp=105002415436&partnerID=8YFLogxK
U2 - 10.1097/BRS.0000000000005322
DO - 10.1097/BRS.0000000000005322
M3 - 学術論文
C2 - 40085125
AN - SCOPUS:105002415436
SN - 0362-2436
JO - Spine
JF - Spine
ER -