摘要
目的探讨老年胆囊结石病人发病的高危风险因素并构建Nomogram预测模型。方法选取2019年1月至2022年11月中国航天科工集团七三一医院收治的160例老年胆囊结石病人为研究组,另选取同期进行健康体检的160例老年人为对照组。记录两组研究对象的临床资料,探索老年胆囊结石病人发病的相关因素;受试者工作特征曲线分析差异有统计学意义的因素对老年胆囊结石病人发病的预测价值;logistic回归分析影响老年胆囊结石病人发病的独立危险因素;R语言软件4.0“rms”包构建老年胆囊结石病人发病的Nomogram预测模型,校正曲线对Nomogram预测模型进行内部验证,决策曲线对Nomogram预测模型进行临床预测效能评估。结果与对照组相比,研究组病人体重指数(body mass index,BMI)较高(P<0.05),合并糖尿病及收缩压(SBP)≥140 mmHg(1 mmHg=0.133 kPa)的比例较高(P<0.05),总胆固醇(TC)、三酰甘油(TG)、低密度脂蛋白胆固醇(LDL-C)、空腹血糖(FBG)水平较高(P<0.05),高密度脂蛋白胆固醇(HDL-C)水平较低(P<0.05)。BMI、TC、TG、HDL-C、LDL-C、FBG的曲线下面积分别为0.846、0.722、0.814、0.825、0.832、0.857,最佳截断值分别为23.24 kg/m^(2)、5.01 mmol/L、1.33 mmol/L、0.86 mmol/L、3.8 mmol/L、5.6 mmol/L。BMI、糖尿病、SBP、TC、HDL-C、FBG是影响老年胆囊结石病人发病的独立危险因素。Nomogram模型预测老年胆囊结石病人发病的C-index为0.725(95%CI:0.638,0.816),校准曲线趋于理想曲线;Nomogram模型预测老年胆囊结石病人发病的阈值>0.19,Nomogram预测模型提供临床净收益,且临床净收益均高于BMI、糖尿病、SBP、TC、HDL-C、FBG独立预测因子。结论该研究以BMI、糖尿病、SBP、TC、HDL-C、FBG作为预测因子,构建了老年胆囊结石病人发病的Nomogram预测模型,这有利于老年胆囊结石的早期预防及干预治疗。
Objective To explore the high risk factors of senile cholecystolithiasis and construct a nomogram prediction model.Methods From January 2019 to November 2022,160 elderly patients hospitalized with cholecystolithiasis were selected as case group.Another 160 elders receiving health examination at the same time were selected as control group.The relevant clinical data of two groups were recorded to explore the related factors of elders with cholecystolithiasis.Receiver operating characteristic curve was plotted for examining the statistically significant factors;Logistic regression analysis was utilized for examining the independent risk factors;R language software 4.0"rms"package was utilized for constructing a nomogram prediction model.The correction curve verified the nomogram prediction model internally and the decision curve was employed for evaluating the clinical prediction efficiency of nomogram model.Results As compared with control group,body mass index(BMI)of study group was higher(P<0.05),the proportion of diabetes and systolic blood pressure(SBP)≥140 mmHg was higher(P<0.05),the levels of total cholesterol(TC),triglyceride(TG),low-density lipoprotein-cholesterol(LDL-C)and fasting blood glucose(FBG)were higher and the level of high-density lipoprotein-cholesterol(HDL-C)was lower(P<0.05).The area under the curve(AUC)of BMI,TC,TG,HDL-C,LDL-C and FBG was 0.846,0.722,0.814,0.825,0.832 and 0.857 and the optimal cutoff values were 23.24 kg/m,5.01 mmol/L,1.33 mmol/L,0.86 mmol/L,3.8 mmol/L and 5.6 mmol/L respectively.BMI,diabetes mellitus,SBP,TC,HDL-C and FBG were independent risk factors for an onset of cholecystolithiasis in elders.Nomogram model predicted that C-index was 0.725(95%CI:0.638-0.816)and calibration curve was almost ideal.The threshold value of nomogram model for predicting the incidence of senile cholecystolithiasis was>0.19.Nomogram model provided clinical benefits and net benefit was higher than the independent predictors of BMI,diabetes mellitus,SBP,TC,HDL-C and FBG.Conclusion BMI,diabetes me
作者
林大鹏
赵斌
陈豪
王丽娜
代明岩
Lin Dapeng;Zhao Bin;Chen Hao;Wang Lina;Dai Mingyan(No.731 Hospital,China Aerospace Science&Industry Group,Beijing 100074,China)
出处
《腹部外科》
2023年第6期473-477,共5页
Journal of Abdominal Surgery
基金
七三一医院院内科研课题(2022-QSYN-12)。
