Related Experiment Video
Updated: Sep 14, 2025
Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
Development of a Predictive Model for Kummell's Disease
Yiliang Song1, Houling Zhao1, Shisheng Liu1
1Central Hospital Affiliated to Shandong First Medical University, Shandong First Medical University & Shandong Academy of Medical Sciences, Jinan, Shandong, China.
Background:
Kümmell's disease (KD) is a recognized complication of osteoporotic thoracolumbar vertebral compression fractures (OVCFs), but the risk factors for progression from OVCFs to KD remain unclear.
Methods:
This retrospective study included 104 patients with OVCFs from a tertiary hospital between January 2020 and December 2023. Patients were divided into a case group (56 patients with KD) and a control group (48 patients without KD). Data on 24 variables, including demographics, fracture characteristics, and biochemical markers, were collected and analyzed using SPSS 25.0 software. Statistical methods included independent samples t-test, Mann-Whitney U test, chi-square test, Fisher's exact test, and binary logistic regression.
Results:
Univariate analysis identified 5 significant factors related to KD onset: age, vertebral length, intervertebral foramen width, alkaline phosphatase, and albumin. Binary logistic regression analysis revealed 3 independent risk factors for KD: vertebral length, alkaline phosphatase, and albumin.
Conclusions:
When treating OVCF patients, clinicians should consider comprehensive factors. Patients with high alkaline phosphatase levels, longer fractured vertebrae, and low albumin levels should be closely monitored. Targeted interventions, such as early surgery, enhanced nutrition, and optimized antiosteoporosis plans, may reduce KD risk and improve patient outcomes.
