Development and validation of a novel predictive model and web calculator for evaluating transfusion risk after spinal fusion for spinal tuberculosis: a retrospective cohort study

Abstract

Objectives The incidence and adverse events of postoperative blood transfusion in spinal tuberculosis (TB) have attracted increasing attention. Our purpose was to develop a prediction model to evaluate blood transfusion risk after spinal fusion (SF) for spinal TB. Methods Nomogram and machine learning algorithms, support vector machine (SVM), decision tree (DT), multilayer perceptron (MLP), Naive Bayesian (NB), k-nearest neighbors (K-NN) and random forest (RF), were constructed to identified predictors of blood transfusion from all spinal TB cases treated by SF in our department between May 2010 and April 2020. The prediction performance of the models was evaluated by 10-fold cross-validation. We calculated the average AUC and the maximum AUC, then demonstrated the ROC curve with maximum AUC. Results The collected cohort ultimately was consisted of 152 patients, where 56 required allogeneic blood transfusions. The predictors were surgical duration, preoperative Hb, preoperative ABL, preoperative MCHC, number of fused vertebrae, IBL, and anticoagulant history. We obtained the average AUC of nomogram (0.75), SVM (0.62), k-NM (0.65), DT (0.56), NB (0.74), MLP (0.56) and RF (0.72). An interactive web calculator based on this model has been provided (https://drwenleli.shinyapps.io/STTapp/). Conclusions We confirmed seven independent risk factors affecting blood transfusion and diagramed them with the nomogram and web calculator.

Blood transfusion risk prediction in spinal tuberculosis surgery: development and assessment of a novel predictive nomogram

“…The high C-index was especially suitable for wide use in large sample data [ 21 ]. The nomogram in the present study achieved the best discriminatory ability with an AUC of 0.785, which was slightly higher than that of the previously established nomogram (AUC = 0.75) [ 16 ].…”

“…The requirement for blood transfusion in spinal tuberculosis surgery depends on several factors such as anaemia caused due to a deficiency of iron [ 11 ], vitamin B12, and folic acid [ 12 ], which are used to compound haemoglobin (HGB) and sex of the patient [ 13 – 15 ]. Dong et al exhibited that the factors such as preoperative mean corpuscular haemoglobin concentration, surgical duration, preoperative HGB, intraoperative blood loss, number of fused vertebrae, and anticoagulant history can predict the risk of blood transfusion after spinal tuberculosis fusion [ 16 ].…”

Development and validation of a carotid atherosclerosis risk prediction model based on a Chinese population

“…Nomograms can generate individual probabilities of clinical events by integrating different outcome and predictor variables and merging biological information with clinical prediction models ( 33 – 35 ). In recent years, it has been widely used as a prediction method in clinical settings ( 36 , 37 ), and it plays a role in promoting personalized medicine ( 38 , 39 ) and facilitates physicians in predicting disease risk ( 40 ). There have been many studies on CAS prevalence and risk factors ( 4 , 41 ), but few studies on CAS risk-prediction models have been reported.…”

A radiographic model predicting the status of the anterior cruciate ligament in varus knee with osteoarthritis

“…In clinical practice, the use of a single radiographic indicator to assess the status of ACL had great uncertainty, while a model constructed using multiple indicators could greatly improve the accuracy. At present, there have been many models with multiple indicators showing high predictive accuracy in other fields [ 36 – 38 ]. As far as we knew, this was the first study combining multiple imaging indicators to construct an ACLD prediction model.…”