AUTHOR=Xiong Tuotuo , Wang Ben , Qin Wanyuan , Yang Ling , Ou Yunsheng 

TITLE=Development and validation of a risk prediction model for cage subsidence after instrumented posterior lumbar fusion based on machine learning: a retrospective observational cohort study

JOURNAL=Frontiers in Medicine

VOLUME=Volume 10 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2023.1196384

DOI=10.3389/fmed.2023.1196384

ISSN=2296-858X

ABSTRACT=Background: Interbody cage subsidence is a common complication after instrumented posterior lumbar fusion surgery, several previous studies have shown that cage subsidence is related to multiple factors. But the current research hasn’t combined these factors to predict the subsidence, there is a lack of an individualized and comprehensive evaluation of the risk of cage subsidence following the surgery. So we attempt to identify potential risk factors and develop a model that can predict the risk of subsidence by providing a Cage Subsidence Score (CSS), and evaluate whether machine learning-related techniques can effectively predict the subsidence. 
Methods: The imaging data of patients who underwent intervertebral fusion in our hospital from 2014 to 2019 were collected retrospectively. They were divided into a subsidence group and a non-subsidence group according to whether the interbody fusion cage subsidence occurred during follow-up. The conventional statistical analysis method was used to analyze whether the risk factors are related, and the prediction model was trained based on the logistic regression model and machine learning algorithms respectively, the diagnostic efficiency of prediction is further verified. 
Results: Univariate analysis showed differences in pre-/postoperative intervertebral disc height, postoperative L4 segment lordosis, postoperative PT, and postoperative SS between the subsidence and the non-subsidence (P < 0.05). The CSS was trained by stepwise regression: 2 points for postoperative disc height > 14.68 mm, 3 points for postoperative L4 segment lordosis > 16.91°, and 4 points for postoperative PT > 22.69°. If the total score is > 0.5, it is high-risk. The score obtains the area under the curve (AUC) of 0.857 and 0.806 in the development and validation set respectively. The AUC of the GBM model based on the machine learning algorithm to predict the risk in the training set is 0.971 and the validation set is 0.889 respectively. The AUC of the avNNet model reached 0.931 and 0.868 respectively. 
Conclusion: The machine learning algorithm has advantages in some indicators, and we have preliminarily established a CSS that can predict the risk of postoperative subsidence after lumbar fusion and confirmed the important application prospect of machine learning in solving practical clinical problems.