Clustering Analysis of Traumatic Thoracic Spine Injuries in 64,000 Patients Using Machine Learning

Introduction: This study aims to identify unique clinical profiles in patients with thoracic spinal injury in the National Inpatient Sample (NIS) by leveraging machine learning-based clustering algorithms.

Methods: The 2015-2019 NIS was queried using ICD-10 PCS to identify patients with thoracic vertebral fracture and spinal cord injury. Machine learning clustering analysis evaluated the population based on 50 comorbidities, complications and clinical covariates. Optimal number of clusters was determined using the Davies-Bouldin Index and Calinski-Harabasz Index. Between-cluster multivariate logistic regression analysis was performed to assess risk of mortality and non-routine discharge. Kruskal-Wallis H-Testing was performed to assess variance in length-of-stay between clusters. Statistical analysis was performed using Python.

Results: 64,408 patients were clustered into 4 groups ranging from 497 to 57,457 patients. Mortality ranged from 3.23% in Cluster 1 to 21.1% in Cluster 4. Clusters 3 [OR 1.84, p< 0.001] and 4 [OR 6.59, p< 0.001] displayed significantly higher risk of mortality relative to Cluster 1. Individual cluster profiles are visualized in a heatmap. Cluster 1 had the lowest prevalence of comorbidities amongst groups. Cluster 4 had the greatest prevalence of arrhythmia, heart failure, obesity, sepsis, pneumonia, and mechanical ventilation use. Risk of non-routine discharge was lowest in Cluster 2 [OR 1.45, p< 0.001] and highest in Cluster 4 [OR 3.34, p< 0.001] compared to Cluster 1. Kruskall-Wallis H testing and post-hoc pairwise testing of length-of-stay distributions showed significant (p < 0.001) differences between all clusters, with the greatest test statistics occurring when comparing group 4 to all other groups.

Conclusion : Clustering analysis of patients with traumatic thoracic spinal injuries yielded four groups, characterized by unique comorbidity patterns. This method offers a refined understanding of how comorbid conditions interact, allowing for tailored clinical strategies and individualized treatment approaches.