Prognostic analysis of traumatic spinal cord injury based on LASSO-COX regression and construction of a Nomogram prediction model
LUO Cui-xiang
WU Xiao-ming
XING Dong-qian
Abstract：Objective To analyze prognostic factors of traumatic spinal cord injury(TSCI)using LASSO-Cox regression and construct a Nomogram prediction model for poor prognosis.Methods A total of 216 patients with TSCI treated at Hengshui Fourth People's Hospital from May 2021 to May 2023 were retrospectively enrolled.All patients were followed up for 12 months post-treatment and divided into poor prognosis and good prognosis groups accordingly.Periop-erative clinical data were compared between groups.LASSO-COX regression was used to identify independent predictors of poor prognosis.The predictive value of relevant indicators was assessed using ROC curve analysis.A Nomogram model was constructed based on the identified prognostic factors.Results The poor prognosis rate among TSCI patients was 38.57％.No significant differences were observed in demographic or baseline perioperative characteristics between the groups(P＞0.05).Compared to the good prognosis group,patients in the poor prognosis group had a higher incidence of preoperative spinal canal occupancy ≥50％,complete injuries,and multisegmental injuries,longer time from injury to treatment,and elevated levels of serum NLRP3 and cystatin C(CysC).In contrast,levels of TGF-β1,miR-133a,and serum response factor(SRF)were lower(P＜0.05).LASSO-Cox regression identified the following as significant risk factors for poor prognosis:time from injury to treatment,preoperative spinal canal occupancy ≥50％,complete inju-ry,elevated NLRP3 andCysC,and reduced TGF-β1,miR-133a,and SRF(P＜0.05).The Nomogram model based on these variables showed excellent predictive performance with an AUC of 0.929(95％CI:0.896-0.962)and good calibration.External validation revealed sensitivity,specificity,and accuracy of 83.33％,87.70％,and 86.00％,re-spectively.Conclusion Delayed treatment,preoperative spinal canal occupancy ≥50％,complete injury,and dysregu-lation of specific serum biomarkers(NLRP3,CysC,TGF-β1,miR-133a,SRF)are significant predictors of poor prog-nosis in TSCI patients.The nomogram model constructed from these factors demonstrates high clinical value for early prog-nostic assessment.
Keywords：traumatic spinal cord injuryprognosisLASSO regressionCOX regressionNomogram prediction model
Publication Date：2025-07-15
Online Publishing Date：2025-08-25(First online date of this platform, not the publication date of the document)
Pages：6( 1022-1027 )