分析老年股骨颈骨折患者半髋关节置换术后输血的危险因素，建立老年股骨颈骨折患者半髋关节置换术后输血的列线图预测模型。

回顾性分析2016年1月至2020年6月江苏省苏北人民医院股骨颈骨折行半髋关节置换术的235例患者。纳入标准：年龄≥60岁，新发、单侧股骨颈骨折；手术方式为半髋关节置换术。排除标准：合并全身其他骨折；术前输血；合并凝血功能障碍或者其他血液系统疾病；身体状况较差无法耐受手术者；病历资料不完整者。根据术后是否输血，分为输血组和未输血组。收集其临床资料，包括性别、年龄、身体质量指数(BMI)、高血压、糖尿病、冠心病、脑卒中、吸烟、饮酒、术前血红蛋白(Hb)、术前血小板(PLT)、术前白蛋白、术前凝血功能、术前抗凝药使用、骨折Garden分型、受伤至手术时间、假体类型、术后是否引流、术后血钙浓度、美国麻醉师协会(ASA)分级、麻醉方式、手术时间、术中出血量等。应用单因素和多因素logistic回归模型筛选术后输血的独立危险因素；通过R软件构建列线图预测模型，并绘制出受试者工作特征(ROC)曲线及校准曲线来评价模型的区分度和准确度。

本研究共纳入235例研究对象，输血组60例，输血率为25.5%。两组患者在术前Hb(χ²＝62.831)、麻醉方式(χ²＝6.539)、手术时间(χ²＝79.392)、术中出血量(χ²＝74.515)、假体类型(χ²＝5.631)方面的组间差异有统计学意义(均为P<0.05)。多因素logistic回归模型分析显示：术前Hb水平(Hb<100 g/L)、手术时间延长(时间≥60 min)、术中出血量增多(出血量≥200 ml)是老年股骨颈骨折患者半髋关节置换术后输血的独立危险因素(均为P<0.05)。列线图预测模型曲线下面积AUC为0.95，校准曲线的斜率接近1，提示该预测模型具有良好的区分度和准确度。

基于术前血红蛋白、手术时间、术中出血量这3项独立危险因素构建的老年股骨颈骨折患者半髋关节置换术后输血预测的列线图模型具有良好的区分度和准确度，望为临床上早期甄别术后高风险输血患者提供指导意义。

To analyze the risk factors of blood transfusion after hemiarthroplasty in elderly patients with femoral neck fracture, and to establish a nomogram prediction model of blood transfusion after hemiarthroplasty in elderly patients with femoral neck fracture.

A total of 235 patients who underwent hemiarthroplasty for femoral neck fractures in the Northern Jiangsu People’s Hospital from January 2016 to June 2020 were analyzed retrospectively.Inclusion criteria: age ≥60 years, new onset, unilateral femoral neck fracture; surgical method is hemiarthroplasty. Exclusion criteria: other fractures; preoperative blood transfusion; coagulation disorders or other hematological diseases; poor physical condition; incomplete medical records. According to whether blood transfusion was performed or not, the patients were divided into transfusion group and non-transfusion group. The clinical data were collected, including sex, age, body mass index (BMI), hypertension, diabetes, coronary heart disease, stroke, smoking, drinking, preoperative hemoglobin (Hb), preoperative platelet (PLT), preoperative albumin, preoperative use of anti-coagulants, Garden classification of femoral neck fracture, time of injury to operation, the type of prosthesis, postoperative drainage, postoperative serum calcium concentration, American Association of anesthesiologists (ASA) grade, mode of anesthesia, operation time, intraoperative blood loss and so on. Single-factor and multi-factor logistic regression models were used to screen the independent risk factors of postoperative blood transfusion, and the nomogram prediction model was constructed by R software, and the receiver operating characteristic (ROC) curve and calibration curve were drawn to evaluate the differentiation and accuracy degree of the model.

A total of 235 subjects were included in this study, including 60 cases in the blood transfusion group, with a blood transfusion rate of 25.5%. There were significant differences between the two groups in preoperative Hb(χ²=62.831), mode of anesthesia(χ²=6.539), operation time(χ²=79.392), intraoperative blood loss(χ²=74.515) and the type of prosthesis(χ²=5.631)(all P<0.05). Multivariate logistic regression model analysis showed that lower preoperative Hb(Hb<100 g/L), prolonged operation time (time ≥60 min) and increased intraoperative blood loss (blood loss≥200 ml) were independent risk factors for blood transfusion after hemiarthroplasty in elderly patients with femoral neck fracture(all P<0.05). The area under the curve (AUC) of the nomogram prediction model was 0.95, and the slope of the calibration curve was close to one, indicating that the prediction model had good degree of differentiation and sccuracy.

The nomogram model for predicting blood transfusion after hemiarthroplasty in elderly patients with femoral neck fracture based on three independent risk factors(preoperative hemoglobin, operation time and intraoperative blood loss) has good differentiation and accuracy degree. It is expected to provide guidance for early screening of patients with high-risk blood transfusion after operation.