Construction and validation of a nomogram predictive model for safety-related influencing factors in unilateral...

To explore the influencing factors on the safety of unilateral puncture kyphoplasty in the treatment of spinal compression fractures, and to construct and verify the nomogram prediction model.

Relevant information about patients who received safe unilateral puncture kyphoplasty for the treatment of spinal compression fractures in our hospital from January 2022 to October 2023 was collected retrospectively. Independent predictive variables included in the Nomogram model were determined and modeled through univariate analysis and multivariate logistic analysis, for internal validation and evaluation of model accuracy, discrimination and calibration.

A total of 240 patients was enrolled and randomly divided into a training set (n = 168) and a validation set (n = 72) at a 7:3 ratio. The rates of ineffective treatment in training set and validation set were 18.45% and 16.67%. Age, cone leading edge height loss rate, sagittal Cobb angle of vertebral body, pedicle width, pedicle height, preoperative hemoglobin level and preoperative blood calcium level were significantly associated with ineffective treatment of spinal compression fractures (all P < 0.05). The receiver operating characteristic curves showed that the area under the curves for suboptimal treatment predicted by the nomogram model was 0.871 (95% CI 0.769-0.972) and 0.787 (95% CI 0.578-0.996), respectively.

The constructed nomogram demonstrated acceptable predictive ability for the safety of unilateral paracentesis kyphoplasty in the treatment of spinal compression fractures. This model shows promising potential for clinical application, though the generalizability requires further external validation.