Prediction model for spinal cord injury in spinal tuberculosis patients using multiple machine learning algorithms: a multicentric study.

Spinal cord injury (SCI) is a prevalent and serious complication among patients with spinal tuberculosis (STB) that can lead to motor and sensory impairment and potentially paraplegia. This research aims to identify factors associated with SCI in STB patients and to develop a clinically significant predictive model. Clinical data from STB patients at a single hospital were collected and divided into training and validation sets. Univariate analysis was employed to screen clinical indicators in the training set. Multiple machine learning (ML) algorithms were utilized to establish predictive models. Model performance was evaluated and compared using receiver operating characteristic (ROC) curves, area under the curve (AUC), calibration curve analysis, decision curve analysis (DCA), and precision-recall (PR) curves. The optimal model was determined, and a prospective cohort from two other hospitals served as a testing set to assess its accuracy. Model interpretation and variable importance ranking were conducted using the DALEX R package. The model was deployed on the web by using the Shiny app. Ten clinical characteristics were utilized for the model. The random forest (RF) model emerged as the optimal choice based on the AUC, PRs, calibration curve analysis, and DCA, achieving a test set AUC of 0.816. Additionally, MONO was identified as the primary predictor of SCI in STB patients through variable importance ranking. The RF predictive model provides an efficient and swift approach for predicting SCI in STB patients.

[Comparison of clinical efficacy of complete and incomplete radical debridement for spinal tuberculosis: a Meta-analysis].

OBJECTIVE: To compare clinical efficacy of complete and incomplete radical debridement for spinal tuberculosis by Meta-analysis. METHODS: The literatures of RCT or non-RCT with complete and incomplete radical debridement for spinal tuberculosis from Medline, EMBASE, Cochrane Library, Web of Science, CBM, CNKI and Wanfang were searched from the time of creating database to July, 2017. Two independent reviewers identified eligible studies, extracted data and evaluated risk of bias of included studies. Meta analysis were performed by Revman 5.3 and GRADE system were used to grade evidence. Recurrence rate, adverse effects, healing time, chemotherapy duration, spinal deformity by correction angle, bone fusion time in interface of intervertebral, erythrocyte sedimentation rate and C-reaction protein were compared between two groups. RESULTS: Totally 9 literatures were chosen, including 5 RCT and 4 non-RCT with 1 302 patients. Compared with incomplete radical debridement, complete radical debridement had lower recurrence rate [OR=0.14, 95%CI(0.08, 0.22), P<0.000 01], lower rate of adverse effects[OR=0.18, 95%CI(0.12, 0.27), P<0.000 01], shorter healing time[MD=-4.80, 95%CI(-5.14, -4.45), P<0.000 01]and chemotherapy duration [MD=-5.25, 95%CI(-5.64, -4.86), P<0.000 01], larger spinal deformity by correction angle[MD=4.88, 95%CI(3.55, 6.27), P<0.000 01], smaller erythrocyte sedimentation rate[MD=-8.74, 95%CI(-11.99, -5.49), P<0.000 01] and C-reaction protein [MD=-4.75, 95%CI(-8.61, -0.88), P=0.02] . However, there was no difference on bone fusion time in interface of intervertebral between two groups[MD=-0.19, 95%CI(-0.50, 0.12), P=0.23]. CONCLUSIONS: Compared with incomplete radical debridement, complete radical debridement has advantages of lower incidence of recurrence, lower rate of adverse reaction, shorten healing time and chemotherapy time, recovered faster. Techniques are selected according to indication of patients individual, complete radical debridement is recommended at the same indications.