Validation and comparison of risk prediction models in patients with cardiogenic shock complicating acute myocardial infarction
10.3760/cma.j.issn.1671-0282.2020.07.005
- VernacularTitle:急性心肌梗死合并心源性休克人群临床预后评分的验证与比较
- Author:
Dejing FENG
1
;
Yu LIU
;
Lefeng WANG
;
Xinchun YANG
;
Shengli DU
;
Chuang LI
;
Zhen ZHAI
;
Yanyan LI
Author Information
1. 首都医科大学附属北京朝阳医院心脏中心 100020
- From:
Chinese Journal of Emergency Medicine
2020;29(7):914-920
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To externally validated the intra-aortic balloon pump (IABP) shockⅡ score and CardShock score for predicting in-hospital mortality in patients with acute myocardial infarction complicated by cardiogenic shock (AMI-CS) and compared them with the Acute Physiology and Chronic Health EvaluationⅡ (APACHEⅡ) score.Methods:According to the inclusion and exclusion criteria, patients admitted to the cardiac care center (CCU) of our center from December 2010 to May 2019 were enrolled in this study. Patients’ baseline characteristics, in-hospital interventions, and outcomes were collected. The APACHEⅡ score was calculated during hospitalization by clinicians and collected by researchers. Two researchers independently calculated the IABP-ShockⅡ score and CardShock score; any disagreement was discussed with the third researcher. The performance of risk scores was evaluated by discrimination and calibration. The discriminative ability of risk scores was evaluated using the area under the receiver operating characteristic curve (AUC) and compared by the Delong method. The calibration of these risk scores was examined by the Hosmer-Lemeshow goodness-of-fit test. The calibration plot was also built.Results:A total of 150 patients enrolled in our study, and the in-hospital mortality was 60%. According to the IABP-ShockⅡ score, patients scored as low risk (0-2), moderate risk (3-4), and high risk (5-9) had in-hospital mortality of 29%, 68%, and 80%, respectively. According to the CardShock score, patients scored as low risk (0-3), moderate risk (4-5), and high risk (6-9)had in-hospital mortality of 21%, 57%, and 82%, respectively. According to the APACHEⅡ score, patients scored<20, 20-30, and >30 had in-hospital mortality of 19%, 69%, and 93%, respectively. For predicting the in-hospital mortality, the APACHEⅡ score demonstrated excellent discrimination (AUC=0.90, 95% CI: 0.84-0.95). The IABP-ShockⅡ score and CardShock score showed good discrimination (AUC=0.76, 95% CI: 0.68-0.83 and AUC=0.79, 95% CI: 0.72-0.85). The discriminative ability did not significantly differ between the IABP-ShockⅡ score and the CardShock score (0.76 vs 0.79, P>0.05), but both of which were significantly lower than the APACHEⅡ score (0.76 vs 0.90, P<0.05, and 0.79 vs 0.90, P<0.05). At the same time, it was not significantly different between the IABP-ShockⅡ score and the CardShock score (0.76 vs 0.79, P>0.05). All of these three scores were adequately calibrated according to the Hosmer-Lemeshow goodness-of-fit test ( P>0.05).The calibration plot showed accurate calibration of these three scores. Conclusions:Although less accurate than the APACHEⅡ score, the IABP-ShockⅡ score and CardShock score can show accurate prediction for in-hospital mortality of AMI-CS patients.
