Predictive factors associated with the onset of Kummell’s disease

BACKGROUND CONTEXT: Kummell's disease (KD) is a long-term complication of vertebral compression fractures, resulting in vertebral collapse and kyphosis. Despite its clinical significance, the underlying mechanisms remain poorly understood. PURPOSE: To identify the predictive factors for the onset of KD and provide clinical insights for early screening and intervention in KD patients. DESIGN: Retrospective study. PATIENT SAMPLES: A total of 170 patients were included, comprising 66 with KD and 104 with old compression fractures. OUTCOME MEASURES: Relevant clinical and imaging data were collected retrospectively. Spinal imaging indicators were also assessed. METHODS: We analyzed clinical data from patients admitted from May 2021 to April 2024 for vertebral compression fractures. Sixty-six diagnosed with KD based on clinical and imaging criteria were identified as the case group. The control group consisted of 104 patients with vertebral compression fractures who underwent conservative treatment and showed no signs of KD upon reexamination one year later. The clinical data included gender, age, bone mineral density (BMD), history of long-term smoking, alcohol abuse, hypertension, diabetes, coronary heart disease, osteoporosis, glucocorticoid use, previous vertebral compression fracture segment and vertebral compression ratio, Cobb angle, vertebral fracture morphology and disc degeneration grade. Independent predictive factors for KD were determined using multivariate binary logistic regression. Receiver operator characteristic (ROC) analysis and Kaplan-Meier plot were used to assess the diagnostic efficiency of parameters for predicting the occurrence of KD. RESULTS: T-tests and Chi-square tests identified significant differences between groups in age, BMD, alcohol abuse, history of hypertension, history of diabetes, history of osteoporosis, history of glucocorticoid use, vertebral compression segment, Cobb angle, vertebral compression ratio, vertebral compression morphology, and disc degeneration grade between the two groups. Binary logistic regression revealed six independent predictors of KD: age, BMD, history of osteoporosis, vertebral compression rate, vertebral compression morphology, and disc degeneration grade. ROC demonstrated that age ≥ 70.5, BMD (T-score) ≤ - 3.65 and a vertebral compression ratio ≥ 29.9% were strongly correlated with KD (P < 0.001). Kaplan-Meier plot showed that most cases of KD occurred within one year after initial vertebral compression fractures, with significant differences in KD incidence observed across different disc degeneration grades (Log-rank test, P < 0.001). CONCLUSIONS: The risk of developing KD is heightened in patients with the following predictive factors are present: (1) Age ≥ 70.5 years; (2) BMD (T-score) ≤ - 3.65; (3) History of osteoporosis; (4) Vertebral compression ratio ≥ 29.9%; (5) Wedge-shaped vertebral compression morphology; and (6) Grade III or higher disc degeneration. Early screening and regular follow-up of high-risk patients are recommended for timely preventive interventions.