天津医药 ›› 2026, Vol. 54 ›› Issue (11): 1183-1188.doi: 10.11958/20260606

• 临床研究 • 上一篇    下一篇

ARDS患者早期肠内营养不耐受的列线图预测模型构建

黄建会(), 李兰兰, 刘承丽△()   

  1. 泸州市中医医院重症医学科 (邮编 646000)
  • 收稿日期:2026-03-04 修回日期:2026-05-25 出版日期:2026-11-15 发布日期:2026-11-10
  • 通讯作者: 李兰兰,刘承丽 E-mail:15892944597@163.com;1012554788@qq.com
  • 作者简介:黄建会(1977),女,副主任医师,主要从事重症医学方面研究。E-mail:15892944597@163.com
  • 基金资助:
    四川省积极应对人口老龄化自筹研究项目(24SCLN0145)

Construction of a nomogram prediction model for early enteral nutrition intolerance in patients with ARDS

HUANG Jianhui(), LI Lanlan, LIU Chengli△()   

  1. Department of Critical Care Medicine, Luzhou Hospital of Traditional Chinese Medicine, Luzhou 646000, China
  • Received:2026-03-04 Revised:2026-05-25 Published:2026-11-15 Online:2026-11-10
  • Contact: LI Lanlan, LIU Chengli E-mail:15892944597@163.com;1012554788@qq.com

摘要:

目的 基于C反应蛋白/白蛋白比值(CAR)、中性粒细胞/淋巴细胞比值(NLR)、预后营养指数(PNI)以及临床特征构建急性呼吸窘迫综合征(ARDS)患者早期肠内营养不耐受的列线图预测模型。方法 采用回顾性队列研究设计,选取2021年5月—2025年5月我院收治的395例接受肠内营养支持的ARDS患者。依据住院期间早期肠内营养不耐受发生情况,分为不耐受组(139例)和耐受组(256例)。收集患者临床资料及实验室指标,计算CAR、NLR和PNI。多因素Logistic回归分析筛选独立影响因素,并据此构建列线图模型。采用受试者工作特征(ROC)曲线、校准曲线(Hosmer-Lemeshow检验)和决策曲线分析(DCA)分别评价模型的区分度、校准度和临床净收益。结果 不耐受组CAR、NLR高于耐受组,PNI低于耐受组(均P<0.05)。多因素Logistic回归分析结果显示,CAR、NLR、急性生理与慢性健康状况评估Ⅱ(APACHE Ⅱ)评分升高是ARDS患者早期肠内营养不耐受的危险因素(P<0.05),PNI、氧合指数(OI)、平均动脉压(MAP)升高是保护因素(P<0.05)。基于上述因素构建列线图预测模型,ROC曲线分析显示,模型预测早期肠内营养不耐受的曲线下面积为0.866(95%CI:0.830~0.902);校准曲线显示,模型拟合度良好(Hosmer-Lemeshow χ2=4.273,P=0.832);DCA结果表明,模型在0.02~0.80阈值范围内具有显著临床净收益。结论 基于CAR、NLR及PNI构建的列线图预测模型能有效预测ARDS患者早期肠内营养不耐受风险。

关键词: 呼吸窘迫综合征, 列线图, 营养评价, 肠内营养不耐受, C反应蛋白/白蛋白比值, 中性粒细胞/淋巴细胞比值, 预后营养指数

Abstract:

Objective To develop a nomogram prediction model for early enteral nutrition (EN) intolerance in patients with acute respiratory distress syndrome (ARDS) based on the C-reactive protein-to-albumin ratio (CAR), neutrophil-to-lymphocyte ratio (NLR), prognostic nutritional index (PNI) and clinical features. Methods A retrospective cohort study design was adopted. A total of 395 ARDS patients who received EN support in our hospital from May 2021 to May 2025 were consecutively enrolled. According to the occurrence of early EN intolerance during hospitalization, the patients were divided into the intolerance group (n=139) and the tolerance group (n=256). Clinical data and laboratory indicators were collected, and CAR, NLR and PNI were calculated. Multivariate Logistic regression analysis was used to identify independent influencing factors, and a nomogram model was constructed accordingly. The area under the curve (AUC) of receiver operating characteristic (ROC), calibration curve (Hosmer-Lemeshow test) and decision curve analysis (DCA) were employed to evaluate the discrimination, calibration and clinical net benefit of the model, respectively. Results Compared with the tolerance group, the intolerance group had significantly higher CAR and NLR, and significantly lower PNI (P<0.05). Multivariate Logistic regression analysis confirmed that CAR, NLR and Acute Physiology and Chronic Health EvaluationⅡ(APACHE Ⅱ) score were risk factors for early EN intolerance in ARDS patients (P<0.05), while PNI, oxygenation index (OI) and mean arterial pressure (MAP) were protective factors (P<0.05). Based on these factors, a nomogram prediction model was constructed. ROC analysis showed that the area under the curve of the model for predicting early EN intolerance was 0.866 (95%CI: 0.830-0.902). The calibration curve demonstrated good fit (Hosmer-Lemeshow χ²=4.273, P=0.832). DCA results indicated significant net clinical benefit when the threshold probability was within the range of 0.02 to 0.80. Conclusion The nomogram prediction model constructed based on CAR, NLR and PNI can effectively predict the risk of early EN intolerance in ARDS patients.

Key words: respiratory distress syndrome, nomograms, nutrition assessment, enteral nutrition intolerance, C-reactive protein-to-albumin ratio, neutrophil-to-lymphocyte ratio, prognostic nutritional index

中图分类号: