关键词: 脊柱侧弯矫形术, 输血, 危险因素, 列线图模型

Abstract: Objective To analyze the risk factors of massive blood transfusion in adolescent scoliosis surgery, establish a nomogram early warning model, and verify the predictive effect of the model. Method The clinical data of 292 adolescent scoliosis patients treated in our hospital from March 2015 to February 2020 were retrospectively analyzed and used as a modeling set. The patients were divided into a massive blood transfusion group and a non-massive blood transfusion group according to the occurrence of massive intraoperative blood transfusion (Within 4 hours, the infusion volume of allogeneic red blood cells is ≥ 1 time the total circulating blood volume, which is 40 mL/kg), and the independent risk factors of intraoperative massive blood transfusion were screened out by Logistic regression. An early nomogram warning model for predicting massive intraoperative bleeding was established based on independent risk factors, the Bootstrap method was used for internal validation of the model, and the receiver operating curve (ROC) and calibration curve were used to evaluate the discrimination and calibration of the model. In addition, 68 cases of adolescent scoliosis orthopedics treated in our hospital from March 2020 to February 2023 were selected as the validation set for external validation. Results The incidence of massive intraoperative blood transfusion in 292 adolescent scoliosis patients was 20.89% (61/292). Multivariate logistic regression analysis showed that BMI≤18kg/m² (OR=2.584, 95%CI: 1.314~5.081), preoperative Cobb angle>48° (OR=3.680, 95%CI: 1.808~7.491), number of vertebral fusion segments>10 (OR=4.543, 95%CI: 2.210~9.340) and intraoperative blood loss >960 mL (OR=3.817, 95%CI: 1.805~8.070) were independent risk factors for massive intraoperative blood transfusion (P<0.05), and cyclic acid (OR=0.393, 95%CI: 0.193~0.803) was a protective factor (P<0.05). Based on this result, a nomogram prediction model was established. The area under the receiver operating characteristic of the modeling set was 0.841 (95%CI: 0.784~0.899), the sensitivity was 82.68%, the specificity was 75.41%, the risk threshold corresponding to the maximum Youden's J statistic 0.58 was 25%, and the predictive critical value score was 220 points, that is, those with predicted nomogram scores ≥ 220 points were high-risk groups for massive blood transfusion during surgery. The area under the Receiver operating characteristic of the validation set is 0.790 (95%CI: 0.720~0.850); The slope of the correction curve approximates the ideal curve; The probability of predicting the risk of massive intraoperative blood transfusion using the column chart warning model is consistent with the actual probability of occurrence. The Pt ranges were 0.38~0.9 and 0.42~0.9, respectively, and the patients could obtain positive net income. Conclusion The column chart warning model established based on the risk factors of massive blood transfusion during adolescent scoliosis correction surgery has good fitting and predictive effect. Those with a column chart score of ≥ 220 are high-risk groups for massive blood transfusion during surgery. Medical staff should prepare for large-scale blood transfusion plans in advance and provide timely blood transfusion treatment.

Key words: Scoliosis, Orthopedics, Blood transfusion, Risk factors, Nomogram model