印度重症监护患者中的持续性危重病:一项注册登记嵌入式队列研究
该研究基于印度重症监护注册登记(Indian Registry of Intensive Care)56个ICU的成人危重患者数据,评估急性病情特征(入ICU诊断和急性生理变量)、既往特征(年龄、性别、Charlson合并症指数)及其组合对医院死亡的关联,使用广义混合效应模型分析,并在事后分析中比较ICU住院时间超过7天与未超过7天的患者。共纳入42925例患者,医院死亡率为17.4%。急性病情特征的预后判别能力随时间下降(AUROC从入ICU时0.76降至第7天0.69),既往特征的判别能力始终较差(入ICU时和第7天AUROC均为0.56)。ICU住院超过7天的患者病情更重(APACHE II中位数14 vs 10,p<0.001),医院死亡率更高(33.1% vs 14.6%;OR 2.17,95%CI 2.03-2.31),ICU死亡率更高(30.3% vs 13.2%),接受有创通气(51.7% vs 19.5%)、无创通气(18.4% vs 7.6%)、血管加压药(34.7% vs 19.1%)和肾脏替代治疗(11.4% vs 6.1%)的比例均显著更高。作者结论认为,在该印度危重患者队列中,随着ICU住院时间增加,急性病情特征的预后判别能力下降,而既往特征的判别能力始终较差;尽管ICU住院超过7天的患者死亡率更高且接受更多器官支持,但高收入国家队列的持续性危重病定义未
要点
- 在印度56个ICU的42925例成人危重患者中,医院死亡率为17.4%,急性病情特征的预后判别能力从入ICU时AUROC 0.76降至第7天0.69。
- 既往特征(年龄、性别、Charlson合并症指数)的预后判别能力始终较差,入ICU时和第7天AUROC均为0.56。
- ICU住院超过7天的患者病情更重(APACHE II中位数14 vs 10,p<0.001),医院死亡率33.1% vs 14.6%(OR 2.17,95%CI 2.03-2.31),ICU死亡率30.3% vs 13.2%。
- ICU住院超过7天的患者接受有创通气(51.7% vs 19.5%)、无创通气(18.4% vs 7.6%)、血管加压药(34.7% vs 19.1%)和肾脏替代治疗(11.4% vs 6.1%)的比例均显著更高。
- 作者结论认为,高收入国家队列的持续性危重病定义在印度该队列中未能被复制。
适用人群与场景
- 印度重症监护注册登记(Indian Registry of Intensive Care)所属56个ICU的成人危重患者。
研究设计与关键结果
- 研究设计为队列研究,纳入印度56个ICU的成人危重患者42925例,无干预措施。
- 医院死亡率为17.4%;急性病情特征的预后判别能力随时间下降(AUROC入ICU时0.76,第7天0.69),既往特征判别能力始终较差(AUROC入ICU时和第7天均为0.56)。
- ICU住院超过7天者病情更重(APACHE II中位数14 vs 10,p<0.001),医院死亡率33.1% vs 14.6%(OR 2.17,95%CI 2.03-2.31),ICU死亡率30.3% vs 13.2%。
- ICU住院超过7天者接受有创通气(51.7% vs 19.5%)、无创通气(18.4% vs 7.6%)、血管加压药(34.7% vs 19.1%)和肾脏替代治疗(11.4% vs 6.1%)的比例均显著更高。
- 作者结论认为,随着ICU住院时间增加,急性病情特征的预后判别能力下降而既往特征始终较差;高收入国家队列的持续性危重病定义未能被复制。
证据类型、等级与局限
- 证据类型:队列研究(注册登记嵌入式)。
- 局限:原文未报告局限性。
对床旁实践的改变
- 原文未据此给出推荐。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
OBJECTIVES: Among patients admitted to an ICU in high-income countries, persistent critical illness arises when the admission diagnosis and acute physiologic derangements no longer predict outcomes better than antecedent characteristics. We evaluated this phenomenon in India.
DESIGN: Cohort study.
SETTING: Fifty-six ICUs in India belonging to the Indian Registry of Intensive Care.
PATIENTS: Adult critically ill patients.
INTERVENTIONS: None.
MEASUREMENTS AND MAIN RESULTS: We evaluated associations of acute illness characteristics (admission diagnoses and acute physiology variables), antecedent characteristics (age, sex, and Charlson Comorbidity Index), and their combination on hospital mortality. Analyses used generalized mixed-effects models. In post hoc analyses, we compared patients with or without a long ICU stay. We included 42,925 patients; hospital mortality was 17.4%. Prognostic performance of acute illness characteristics declined over time (area under the receiver operating characteristic curve [AUROC], 0.76 at admission; 0.69 at day 7) and was consistently poor for antecedent characteristics (AUROC, 0.56 at admission and day 7). Patients with ICU stay greater than 7 days were sicker (median [interquartile range (IQR)]; Acute Physiology and Chronic Health Evaluation II, 14 [9, 19] vs. 10 [6, 15]; p < 0.001), with higher hospital mortality (33.1% vs. 14.6%; odds ratio, 2.17; 95% CI, 2.03-2.31) and ICU mortality (30.3% vs. 13.2%); receipt of invasive ventilation (51.7% vs. 19.5%), noninvasive ventilation (18.4% vs. 7.6%), vasopressors (34.7% vs. 19.1%), and kidney replacement therapy (11.4% vs. 6.1%) were all significantly higher among those staying in ICU greater than 7 days.
CONCLUSIONS: In this cohort of critically ill patients in India, as ICU stay increased, the prognostic performance of acute illness characteristics decreased whereas that of antecedent characteristics remained poor. Although patients with ICU stay greater than 7 days had higher mortality and received more organ support, the definition of persistent critical illness from high-income country cohorts was not replicated.
参考来源
- Tirupakuzhi Vijayaraghavan BK, Rashan A, Ramakrishnan N, et al. Persistent Critical Illness Among Intensive Care Patients in India: A Registry-Embedded Cohort Study. Crit Care Med. 2025. PMID: 40521972. doi:10.1097/CCM.0000000000006743
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。