儿童重症肺炎预后不良的影响因素分析及Nomogram模型构建

Analysis of influencing factors of poor prognosis of severe pneumonia in children and the construction of Nomogram model

摘要

目的 分析儿童重症肺炎(SP)预后不良的影响因素,构建预测模型。方法 选取2020年3月至2023年3月金华市妇幼保健院378例SP患儿,按照1∶1对照原则随机分为建模人群(189例)、验证人群(189例).比较建模人群、验证人群患病后3个月预后情况,在建模人群中分析预后不良影响因素、构建Nomogram模型,并对Nomogram模型进行外部验证。结果 SP患儿预后不良发生率为24.60%(93/378),其中建模人群预后不良发生率(23.81%)与验证人群(25.40%)对比差异无统计学意义(χ²=0.128,P=0.720);在建模人群中,预后不良组患病至就诊时间长于预后良好组(t=8.997,P<0.05),机械通气、有既往肺部感染病史、有肺部基础疾病、感染病原菌株数≥2株、有肺外并发症占比高于预后良好组(χ²值介于4.247~10.668之间,均P<0.05),血清PCT水平、急性生理学评分(APS)高于预后良好组(t值分别为10.353、5.278,P<0.05);患病至就诊时间、机械通气、既往肺部感染病史、感染病原菌株数、肺外并发症、血清PCT水平、APS均为预后不良的独立危险因素(OR值介于2.653~5.955之间,P<0.05);在验证人群中对该模型进行外部验证显示,Nomogram模型的C-index为0.948,具有良好的区分度,曲线下面积(AUC)为0.948(95%CI:0.910~0.988),预测预后不良的校准度与临床效用良好。结论 SP患儿预后受多方面因素影响,基于急性生理学评分,结合相关影响因素,构建的Nomogram模型,在预后预测方面具有良好的区分度、预测效能、校准度和临床效用.

Abstract

Objective The influencing factors of poor prognosis of severe pneumonia (SP) in children were analyzed, and a prediction model was constructed. Methods A total of 378 children with SP in Jinhua Maternal and Child Health Hospital from March 2020 to March 2023 were randomly divided into the modeling population (189 cases) and the verification population (189 cases) according to the 1∶1 control principle. The prognosis of the modeled population and the verified population at 3 months after the disease was compared, the adverse prognostic influencing factors were analyzed in the modeled population, the Nomogram model was constructed, and the Nomogram model was externally verified. Results The incidence of poor prognosis in SP was 24.60% (93/378), and there was no significant difference between the incidence of poor prognosis in the modeling population (23.81%) and the validated population (25.40%) (χ²=0.128, P=0.720). In the modeled population, the poor prognosis group had a longer time from disease to consultation (t=8.997, P<0.05), the proportion of mechanical ventilation, previous lung infection, underlying lung disease, ≥2 strains of infectious pathogens, and extrapulmonary complications in the poor prognosis group was higher than that in the good prognosis group (χ² values ranged from 4.247 to 10.668, P<0.05), and the serum procalcitonin level and acute physiology score (APS) were higher than those in the good prognosis group (t=10.353 and 5.278; P<0.05). Time from illness to presentation, mechanical ventilation, history of previous pulmonary infection, number of infectious strains, extrapulmonary complications, serum procalcitonin level, and APS were all independent risk factors for poor prognosis (OR=2.653-5.955, P<0.05). External validation of the model in the validation population showed that the C-index of the Nomogram model was 0.948, which had good discrimination, and the area under the curve (AUC) was 0.948 (95%CI: 0.910-0.988), indicating good calibration and clinical utility for predicting poor prognosis. Conclusion The prognosis of children with SP is affected by many factors, and the Nomogram model constructed based on acute physiological scores and related influencing factors has good discrimination, predictive efficacy, calibration and clinical utility in prognostic prediction.

关键词

儿童 / 重症肺炎 / 急性生理学评分 / 预后 / 影响因素 / Nomogram模型

Key words

child / severe pneumonia / acute physiology score / prognosis / influencing factor / Nomogram model

参考文献

[1] Wang H, Lu Z, Bao Y, et al. Clinical diagnostic application of metagenomic next-generation sequencing in children with severe nonresponding pneumonia[J]. PLoS One, 2020, 15(6): e0232610.

[2] Purimahua S L, Arsin A A, Daud A, et al. Biomarkers of pneumonia disease in under five children in "grill culture"[J]. Gac Sanit, 2021, 35(S2): S435-S437.

[3] 孙永烽, 杨勤, 陈敏, 等. 198例小儿重症肺炎病原学分析[J]. 贵州医科大学学报, 2020, 45(3): 345-349, 355.

[4] Gillet Y, Tristan A, Rasigade J P, et al. Prognostic factors of severe community-acquired staphylococcal pneumonia in France[J]. Eur Respir J, 2021, 58(5): 2004445.

[5] Fan F, Lv J, Yang Q Y, et al. Clinical characteristics and serum inflammatory markers of community-acquired mycoplasma pneumonia in children[J]. Clin Respir J, 2023, 17(7): 607-617.

[6] Gong W H, Gao K J, Shan Z M, et al. Research progress of biomarkers in evaluating the severity and prognostic value of severe pneumonia in children[J]. Front Pediatr, 2024, 12(1): 1417644.

[7] Mamun G M S, Zou M, Sarmin M, et al. Derivation and validation of a clinical prediction model for risk-stratification of children hospitalized with severe pneumonia in Bangladesh[J]. PLOS Glob Public Health, 2023, 3(8): e0002216.

[8] 康平, 李奕, 汤昱. 高迁移率族蛋白B1联合急性生理学评分对重症肺炎患儿诊断价值及预后价值的研究[J]. 中国全科医学, 2022, 25(8): 924-929, 936.

[9] 中国医师协会急诊医师分会. 中国急诊重症肺炎临床实践专家共识[J]. 中国急救医学, 2016, 36(2): 97-107.

[10] Okumura K, Latifi R, Smiley A, et al. Direct peritoneal resuscitation (DPR) improves acute physiology and chronic health evaluation (APACHE) IV and acute physiology score when used in damage control laparotomies: prospective cohort study on 37 patients[J]. Surg Technol Int, 2022, 41(1): 85-93.

[11] Mehta R, G Ptrainee, Chinthapalli K, et al. Glasgow coma scale explained[J]. BMJ, 2019, 65(1): 11296.

[12] Zhang R, Wang H, Tian S, et al. Adenovirus viremia may predict adenovirus pneumonia severity in immunocompetent children[J]. BMC Infect Dis, 2021, 21(1): 213.

[13] Boettiger D C, An V T, Lumbiganon P, et al. Severe recurrent bacterial pneumonia among children living with HIV[J]. Pediatr Infect Dis J, 2022, 41(5): e208-e215.

[14] 李洁, 孙欣荣, 樊亚丽, 等. 儿童腺病毒肺炎的临床特征分析[J]. 中国妇幼健康研究, 2022, 33(1): 141-145.

[15] 刘彬娜. 重症肺炎患儿支气管肺泡灌洗液病原菌分布、药物敏感性及预后危险因素分析[J]. 医学检验与临床, 2021, 32(4): 39-42.

[16] 李重锦, 苏新星, 蒋丽君, 等. 重症肺炎患儿预后的危险因素及其列线图预测模型构建研究[J]. 实用心脑肺血管病杂志, 2021, 29(9): 47-52.

[17] Wilkes C, Bava M, Graham H R, et al. What are the risk factors for death among children with pneumonia in low-and middle-income countries? a systematic review[J]. J Glob Health, 2023, 13(1): 05003.

[18] 李佳, 刘超, 胡姝雯, 等. 血清降钙素原、高迁移率族蛋白1及急性生理与慢性健康评分系统Ⅱ评分对重症肺炎患儿预后的评估[J]. 中国临床医生杂志, 2021, 49(12): 1494-1497.

[19] Yu H, Luo J, Ni Y, et al. Early prediction of extubation failure in patients with severe pneumonia: a retrospective cohort study[J]. Biosci Rep, 2020, 40(2): BSR20192435.

[20] 唐先俊, 姚欣. 老年重症肺炎患者血清中活化蛋白C、髓样分化因子88水平与不良预后的关系[J]. 河北医药, 2021, 43(1): 53-56, 61.

[21] 高蓓, 蔡晓玉, 廖蕾, 等. 血Scr与肺炎严重指数在评估重症肺炎患者预后的临床价值[J]. 标记免疫分析与临床, 2022, 29(2): 276-280.

[22] 张慧芳, 张雪, 沙玉霞, 等. 血清及支气管肺泡灌洗液中sTREM-1水平、APACHE Ⅱ评分及SOFA评分对重症肺炎患儿病情及预后评价[J]. 中国当代儿科杂志, 2020, 22(6): 626-631.

基金资助

2021年度公益类金华市科学技术研究计划项目(2021G4G232)