|
|
Application of lasso-logistic model in the analysis of septic shock influencing factors after percutaneous nephrolithotripsy |
Xiang Fengming, Zhou Jie, Zhang Danyun, Zhang Xian |
Department of Urology, Wenzhou Central Hospital, Wenzhou 325000, China |
|
|
Abstract Objective To analyze the influencing factors of septic shock after percutaneous nephrolithotomy based on lasso of logistic regression variable selection method, and provide valuable guidance for reducing the occurrence of septic shock. Methods The clinical data of 802 patients who underwent percutaneous nephrolithotripsy in Wenzhou Central Hospital from March 2009 to January 2020 were retrospectively analyzed. Lasso logistic model was performed to identify the influencing factors of septic shock after percutaneous nephrolithotomy. Cross validation method was used to choose λ for lasso-logistic model.In addition, the fitting accuracy and prediction effect of lasso-logistic regression model, full-variable logistic regression model and stepwise logistic regression model were compared using Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). Results Lasso-logistic regression model showed that sex, stone size, preoperative urine culture, decrease in hemoglobin, and operation time were independent influencing factors of septic shock after percutaneous nephrolithotomy. The AIC and BIC of lasso-logistic regression model were 143.747 and 169.854, respectively, which were lower than those of full-variable logistic regression model and stepwise logistic regression model. Conclusion Lasso-logistic regression model of septic shock after percutaneous nephrolithotomy based on the above five factors has good fitting accuracy and predictive effect.
|
Received: 03 June 2020
|
|
|
|
[1]KHAN S R, CANALES B K. Genetic basis of renal cellular dysfunction and the formation of kidney stones[J]. Urol Res, 2009, 37(4):169-180.
[2]司亚晨,谌卫,郭志勇. 雌激素与泌尿系结石发生的研究进展[J]. 中国中西医结合肾病杂志, 2019, 20(6): 553-555.
[3]LI Z Q, WU A M, LIU J J, et al. Risk factors for hemorrhage requiring embolization after percutaneous nephrolithotomy: A meta-analysis[J]. Transl Androl Urol, 2020, 9(2):210-217.[4]潘生玉,程帆,叶芃. 经皮肾镜碎石取石术的研究进展[J]. 医学综述, 2018, 24(24): 4874-4879.
[5]倪大伟, 席俊华, 吴畏,等. 经皮肾镜取石术和输尿管软镜碎石术治疗孤立肾结石有效性和安全性的Meta分析[J]. 中国医药导报, 2019, 16(3):63-67.
[6]赵俊琴,王彤,王慧,等. Lasso惩罚计分检验在小样本回归模型自变量筛选与统计推断中的应用[J]. 中华疾病控制杂志, 2015, 19(5): 507-509.
[7]周佩敏,苏忠良,徐青, 等.采用决策树和logistic回归模型分析全髋关节置换术后手术部位感染的影响因素[J].中国医院统计,2019,26(6):404-407.
[8]LEE T F, CHAO P J, TING H M, et al. Using multivariate regression model with least absolute shrinkage and selection operator (LASSO) to predict the incidence of Xerostomia after intensity-modulated radiotherapy for head and neck cancer[J]. PLoS One, 2014, 9(2):e89700.
[9]RHODES A, EVANS L E, ALHAZZANI W, et al. surviving sepsis campaign: International guidelines for management of sepsis and septic shock: 2016[J]. Intensive Care Medicine, 2017, 43(3):304-377.
[10]TIBSHIRANI R. Regression shrinkage and selection via the lasso[J]. J Royal Stat Soc: Ser B Methodol, 1996, 58(1):267-288.
[11]李敏捷.LassoLogistic与Group LassoLogistic模型在出生缺陷研究中的应用[D].太原:山西医科大学,2016.
[12]BANSAL S, PAWAR P, SAWANT A, et al. Predictive factors for fever and sepsis following percutaneous nephrolithotomy: A review of 580 patients[J]. Urol Ann, 2017, 9(3):230.
[13]曹松强.经皮肾镜碎石术后感染性休克危险因素预测模型的建立[D].郑州:郑州大学,2018.
[14]王羿翔.中性粒细胞/淋巴细胞比值与血小板/淋巴细胞比值对经皮肾镜碎石取石术后发生脓毒血症的预测作用[D].沈阳:中国医科大学,2019.
|
[1] |
Huang Danhong, Yao Liping, Shang Qinfen. Intervention effect evaluation on septic shock patients carrying out comprehensive intervention[J]. journal1, 2017, 24(1): 27-29. |
[2] |
Zhong Jianliang, Chen Caier, Pei Jingbo, Cao Lijun, Lu Youchuan, Wang Gang. Comparative analysis for outpatient waiting time based on logistic regression model[J]. journal1, 2015, 22(4): 261-264. |
|
|
|
|