切换至 "中华医学电子期刊资源库"

中华肾病研究电子杂志 ›› 2019, Vol. 08 ›› Issue (05) : 219 -225. doi: 10.3877/cma.j.issn.2095-3216.2019.05.006

所属专题: 文献

论著

腹膜透析相关腹膜炎危险因素研究的统计模型筛选
王宏1, 周建辉,1, 曹雪莹1, 黄静1, 蔡广研1, 陈香美,1   
  1. 1. 100853 北京,解放军总医院第一医学中心肾脏病科、解放军肾脏病研究所、肾脏疾病国家重点实验室(2011DAV00088)、国家慢性肾病临床医学研究中心、肾脏疾病研究北京市重点实验室
  • 收稿日期:2019-06-21 出版日期:2019-10-28
  • 通信作者: 周建辉, 陈香美
  • 基金资助:
    国家重点研发计划(2016YFC1103004、2018YFC0114503)

Screening of statistical models for risk factors of peritoneal dialysis-related peritonitis

Hong Wang1, Jianhui Zhou,1, Xueying Cao1, Jing Huang1, Guangyan Cai1, Xiangmei Chen,1   

  1. 1. Department of Nephrology, The First Medical Centre, Chinese PLA General Hospital, Chinese PLA Institute of Nephrology, State Key Laboratory of Kidney Diseases, National Clinical Research Center for Kidney Diseases, Beijing Key Laboratory of Kidney Diseases, Beijing 100853, China
  • Received:2019-06-21 Published:2019-10-28
  • Corresponding author: Jianhui Zhou, Xiangmei Chen
  • About author:
    Corresponding authors: Chen Xiangmei, Email:
    Zhou Jianhui, Email:
引用本文:

王宏, 周建辉, 曹雪莹, 黄静, 蔡广研, 陈香美. 腹膜透析相关腹膜炎危险因素研究的统计模型筛选[J/OL]. 中华肾病研究电子杂志, 2019, 08(05): 219-225.

Hong Wang, Jianhui Zhou, Xueying Cao, Jing Huang, Guangyan Cai, Xiangmei Chen. Screening of statistical models for risk factors of peritoneal dialysis-related peritonitis[J/OL]. Chinese Journal of Kidney Disease Investigation(Electronic Edition), 2019, 08(05): 219-225.

目的

筛选适用于腹膜透析相关腹膜炎危险因素研究的统计方法,为腹膜炎的临床研究提供新的思路。

方法

选择九种统计模型进行对比分析,包括:不含时间依存性协变量(时依协变量)的Logistic回归、泊松回归、负二项回归、COX回归、AG模型、PWP-CP模型6个模型,含时间依存性协变量的COX回归、AG模型、PWP-CP模型3个模型。统计数据来源于解放军总医院第一医学中心腹膜透析中心2013年1月至2016年12月开始腹膜透析治疗的终末期肾脏病患者,数据收集至2018年12月。分析腹膜炎的临床特点,对比各统计模型拟合情况和统计结果,筛选统计模型。

结果

从腹膜炎临床数据构成来看,作为因变量,腹膜炎事件由"发生时间"、"发生次数"、"发生顺序"三个要素构成,作为自变量,大部分协变量是随时间变化的,三维的因变量和变化的自变量共同构成了复杂的临床数据。从"腹膜炎三要素"纳入情况看,只有PWP-CP模型能够全部纳入,最大限度保证了数据的完整性。从统计模型拟合程度看,泊松模型存在过离散,不适用于本研究数据;使用时依协变量的模型较使用基线数据的模型拟合好。从统计结果看,含时间依存性协变量的模型避免了单纯使用基线数据造成的偏差,使统计结果更接近于真实情况。

结论

在腹膜炎危险因素的研究中,只有尽可能地保证数据信息的完整性和准确性,才能得到更真实的研究结果。本研究通过剖析腹膜炎的数据构成,对比各模型的统计结果,发现含时依协变量的PWP-CP模型更适用于腹膜炎危险因素的研究。

Objective

To screen statistical methods that are suitable for studying risk factors of peritoneal dialysis-related peritonitis so as to provide new ideas for clinical research of peritonitis.

Methods

Nine statistical models were selected for comparative analysis, including six models without time-dependent covariates: logistic regression, Poisson regression, negative binomial regression, COX regression, AG model, and PWP-CP model; and three models with time-dependent covariates: COX regression, AG model, and PWP-CP model. The statistical data were obtained from patients with end-stage renal disease who started peritoneal dialysis treatment from January 2013 to December 2016 in the Peritoneal Dialysis Center of The First Medical Center, Chinese PLA General Hospital. The data were collected until December 2018. Through the analysis of the clinical characteristics of peritonitis, and the comparison between the statistical models and results, the suitable statistical model was selected.

Results

From the clinical data of peritonitis, as a dependent variable, the peritonitis event consists of three elements: "time of occurrence" , "number of occurrences" , and "order of occurrence" . As independent variables, most of the covariates were time-dependent. The three-dimensional dependent variables and the varying independent variables together constituted complex clinical data. In terms of data integrity, only the PWP-CP model could include the "three elements of peritonitis" . In terms of statistical models fitting, the Poisson model was too discrete to enter the data of this study. The model based on the time-dependent covariates was better to fit than the model using the baseline data. In terms of statistical results, the model with time-dependent covariates avoided the bias caused by the simple use of baseline data, making the statistical results closer to the real situation.

Conclusion

In the study of risk factors of peritonitis, more realistic research results could be obtained only by ensuring the integrity and accuracy of the data as much as possible. In this study, by analyzing the data of peritonitis and comparing the statistical results of each model, it was found that the PWP-CP model with time-dependent covariates was more suitable for studying the risk factors of peritonitis.

表1 六种统计模型可纳入的腹膜炎要素
表2 基线特征和组间比较[±s,中位数(Q1-Q3),例数(%) ]
变量 全部患者(n=296) 非腹膜炎组(n=227) 腹膜炎组(n=69) t/z/χ2 P值
男性 63.2% 63.9% 60.9% 0.206 0.650
年龄(岁) 47(34-57) 47(34-57) 46(34-58) -0.215 0.830
年龄(≥65岁) 12.8% 13.2% 11.6% 0.124 0.724
继往血液透析史 28.7% 29.1% 27.5% 0.061 0.805
身高(cm) 167.29±8.39 167.5±8.21 166.61±8.98 0.770 0.442
体重(kg) 65.86±13.80 65.96±13.04 65.52±16.14 0.229 0.819
BMI(kg/m2) 23.39±3.79 23.39±3.61 23.40±4.36 -0.019 0.985
肥胖(BMI≥28 kg/m2) 10.8% 8.8% 17.4% 4.041 0.044
糖尿病 14.5% 14.5% 14.5% <0.001 0.993
血糖(mmol/L) 4.96±1.45 4.84±1.25 5.34±1.94 -1.990 0.049
心血管疾病 52.7% 54.6% 46.4% 1.444 0.229
收缩压(mmHg) 144.65±18.35 144.65±18.60 144.65±17.66 -0.002 0.998
舒张压(mmHg) 88.21±13.51 88.06±13.23 88.71±14.50 -0.351 0.726
平均动脉压(mmHg) 106.66±14.73 106.45±14.93 107.36±14.12 -0.448 0.655
血红蛋白(g/L) 100.51±18.96 100.05±19.48 102±17.22 -0.745 0.457
贫血 31.1% 32.6% 26.1% 1.048 0.306
血清白蛋白(g/L) 35.3(32.0-38.70) 35.4(32.08-38.7) 35(31.55-38.85) -0.220 0.921
低白蛋白血症 47.3% 46.3% 50.7% 0.424 0.515
谷草转氨酶(mmol/L) 13.3(9.7-18.7) 13.2(9.5-18.3) 14.4(10.3-21.0) -1.177 0.238
谷丙转氨酶(mmol/L) 12.9(9.60-19.50) 12.7(9.45-19.75) 14.6(10.75-19.13) -1.362 0.220
总胆红素(mmol/L) 5(4.10-6.60) 5.1(4.10-6.68) 4.6(3.90-5.88) -1.583 0.347
甘油三酯(mmol/L) 1.4(0.99-1.94) 1.44(1.04-1.95) 1.28(0.95-1.87) -0.819 0.469
总胆固醇(mmol/L) 4.39±0.98 4.35±0.93 4.51±1.13 -1.143 0.254
低密度脂蛋白(mmol/L) 2.69(2.18-3.34) 2.70(2.22-3.29) 2.59(2.10-3.47) -0.170 0.865
高密度脂蛋白(mmol/L) 1.06(0.85-1.35) 1.06(0.86-1.33) 1.07(0.82-1.51) -0.500 0.546
血钾(mmol/L) 4.13±0.64 4.13±0.62 4.13±0.70 0.069 0.945
低血钾 15.9% 15.9% 15.9% <0.001 0.987
血钠(mmol/L) 141.07±3.21 140.86±3.32 141.76±2.77 -2.033 0.053
血氯(mmol/L) 100.47±18.05 100.65±20.53 99.86±4.15 0.318 0.751
血镁(mmol/L) 0.93(0.84-1.02) 0.93(0.84-1.04) 0.93(0.84-1.01) -0.345 0.568
血钙(mmol/L) 2.13(2.02-2.23) 2.13(2.02-2.23) 2.15(2.02-2.26) -0.758 0.315
血磷(mmol/L) 1.53(1.28-1.82) 1.54(1.30-1.85) 1.48(1.16-1.78) -1.344 0.164
全段甲状旁腺激素(μg/ml) 191.4(104.85-333.40) 200.3(112.90-335.65) 179.5(80.88-305.65) -1.238 0.245
血尿素氮(mmol/L) 19.81±6.58 19.96±6.78 19.31±5.90 0.721 0.471
血肌酐(μmol/L) 809.9±287.8 818.0±283.5 783.0±301.9 0.886 0.376
残余肾功能(ml/min) 4.43±3.35 4.28±3.24 4.90±3.65 -1.290 0.198
尿量(L/d) 1.24±0.67 1.23±0.63 1.26±0.76 -0.378 0.706
净超滤量(L/d) 0.11±0.53 0.11±0.37 0.09±0.88 0.248 0.805
腹膜尿素清除指数(Kt/V) 1.1(0.89-1.37) 1.09(0.89-1.37) 1.15(0.91-1.45) -0.215 0.846
肾脏尿素清除指数(Kt/V) 0.76(0.41-1.17) 0.73(0.39-1.14) 0.77(0.46-1.30) -0.902 0.682
每周总尿素清除指数(Kt/V) 1.91(1.55-2.43) 1.91(1.51-2.37) 1.98(1.58-2.55) -0.770 0.782
腹膜Ccr (L/1.73 m2) 34.78(31.46-39.11) 34.86(31.78-39.31) 33.27(30.86-38.41) -1.159 0.306
肾脏Ccr (L/1.73 m2) 37.87(17.23-66.41) 36.68(16.67-64.67) 40.59(19.83-80.91) -1.107 0.205
每周总Ccr (L/1.73 m2) 72.07(54.44-99.95) 71.59(53.75-98.79) 73.06(57.83-109.54) -0.865 0.545
腹膜转运功能 0.72±0.11 0.72±0.11 0.71±0.11 0.932 0.352
表3 共线性诊断
表4 使用不含时依协变量的6种模型进行腹膜炎危险因素的多因素分析
表5 使用含时依协变量的模型进行腹膜炎危险因素的多因素分析
表6 各模型拟合统计量
[1]
Pérez Fontan M, Rodríguez-Carmona A, García-Naveiro R, et al. Peritonitis-related mortality in patients undergoing chronic peritoneal dialysis [J]. Perit Dial Int, 2005, 25(3): 274-284.
[2]
Nessim SJ, Bargman JM, Austin PC, et al. Predictors of peritonitis in patients on peritoneal dialysis: results of a large, prospective Canadian database [J]. Clin J Am Soc Nephrol, 2009, 4(7): 1195-1200.
[3]
Teixeira L, Rodrigues A, Carvalho MJ, et al. Modelling competing risks in nephrology research: an example in peritoneal dialysis [J]. BMC Nephrol, 2013, 14(1): 110.
[4]
Fan X, Huang R, Wang J, et al. Risk factors for the first episode of peritonitis in Southern Chinese continuous ambulatory peritoneal dialysis patients [J]. PLoS One, 2014, 9(9): e107485.
[5]
Kumar VA, Sidell MA, Yang WT, et al. Predictors of peritonitis, hospital days, and technique survival for peritoneal dialysis patients in a managed care setting [J]. Perit Dial Int, 2013, 34(2): 171-178.
[6]
Tian Y, Xie X, Xiang S, et al. Risk factors and outcomes of early-onset peritonitis in Chinese peritoneal dialysis patients [J]. Kidney Blood Press Res, 2017, 42(6): 1266-1276.
[7]
胡良平. SAS常用统计分析教程(第2版) [M]. 电子工业出版社,2015: 495-513.
[8]
高峻,董伟,高尔生,等. 多结局生存分析模型与Cox模型的随机模拟比较 [J]. 中国卫生统计,2007,24(3): 248-250,254.
[9]
Sarkar K, Chowdhury R, Dasgupta A. Analysis of survival data: challenges and algorithm-based model selection [J]. J Clin Diagn Res, 2017, 11(6): LC14-LC20.
[10]
Li PK, Szeto CC, Piraino B, et al. ISPD peritonitis recommendations: 2016 update on prevention and treatment [J]. Perit Dial Int, 2016, 36(5): 481-508.
[11]
中国肥胖问题工作组. 中国成人超重和肥胖症预防与控制指南(节录) [J]. 营养学报,2004,26(1): 1-4.
[1] 明昊, 肖迎聪, 巨艳, 宋宏萍. 乳腺癌风险预测模型的研究现状[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(05): 287-291.
[2] 庄燕, 戴林峰, 张海东, 陈秋华, 聂清芳. 脓毒症患者早期生存影响因素及Cox 风险预测模型构建[J/OL]. 中华危重症医学杂志(电子版), 2024, 17(05): 372-378.
[3] 黄鸿初, 黄美容, 温丽红. 血液系统恶性肿瘤患者化疗后粒细胞缺乏感染的危险因素和风险预测模型[J/OL]. 中华实验和临床感染病杂志(电子版), 2024, 18(05): 285-292.
[4] 罗文斌, 韩玮. 胰腺癌患者首次化疗后中重度骨髓抑制的相关危险因素分析及预测模型构建[J/OL]. 中华普通外科学文献(电子版), 2024, 18(05): 357-362.
[5] 贺斌, 马晋峰. 胃癌脾门淋巴结转移危险因素[J/OL]. 中华普外科手术学杂志(电子版), 2024, 18(06): 694-699.
[6] 林凯, 潘勇, 赵高平, 杨春. 造口还纳术后切口疝的危险因素分析与预防策略[J/OL]. 中华疝和腹壁外科杂志(电子版), 2024, 18(06): 634-638.
[7] 杨闯, 马雪. 腹壁疝术后感染的危险因素分析[J/OL]. 中华疝和腹壁外科杂志(电子版), 2024, 18(06): 693-696.
[8] 周艳, 李盈, 周小兵, 程发辉, 何恒正. 不同类型补片联合Nissen 胃底折叠术修补食管裂孔疝的疗效及复发潜在危险因素[J/OL]. 中华疝和腹壁外科杂志(电子版), 2024, 18(05): 528-533.
[9] 张伟伟, 陈启, 翁和语, 黄亮. 随机森林模型预测T1 期结直肠癌淋巴结转移的初步研究[J/OL]. 中华结直肠疾病电子杂志, 2024, 13(05): 389-393.
[10] 王涛, 刘静, 高玉伟, 王兴华, 胡秀红, 崔红蕊, 徐保振, 杨洪娟. 抗生素耐药背景下中医药防治腹膜透析相关性腹膜炎研究进展[J/OL]. 中华肾病研究电子杂志, 2024, 13(06): 340-344.
[11] 崔文鹏. 腹膜透析在老年终末期肾脏疾病患者中的应用[J/OL]. 中华肾病研究电子杂志, 2024, 13(05): 300-300.
[12] 司楠, 孙洪涛. 创伤性脑损伤后肾功能障碍危险因素的研究进展[J/OL]. 中华脑科疾病与康复杂志(电子版), 2024, 14(05): 300-305.
[13] 颜世锐, 熊辉. 感染性心内膜炎合并急性肾损伤患者的危险因素探索及死亡风险预测[J/OL]. 中华临床医师杂志(电子版), 2024, 18(07): 618-624.
[14] 李文哲, 王毅, 崔建, 郑启航, 王靖彦, 于湘友. 新疆维吾尔自治区重症患者急性肾功能异常的危险因素分析[J/OL]. 中华卫生应急电子杂志, 2024, 10(05): 269-276.
[15] 刘志超, 胡风云, 温春丽. 山西省脑卒中危险因素与地域的相关性分析[J/OL]. 中华脑血管病杂志(电子版), 2024, 18(05): 424-433.
阅读次数
全文
0
HTML PDF
最新录用 在线预览 正式出版 最新录用 在线预览 正式出版
0 0 0 0 0 0


摘要
38
最新录用 在线预览 正式出版
0 0 38
  来源 本网站 其他网站
  次数 19 19
  比例 50% 50%