Prognostic value of oxygen challenge test for patients with cardiogenic shock receiving extracorporeal membrane oxygenation / 中华危重病急救医学

ABSTRACT

Objective To investigate the prognostic value of oxygen challenge test (OCT) for patients with cardiogenic shock receiving extracorporeal membrane oxygenation (ECMO). Methods A retrospective analysis was conducted. Seventy-eight patients diagnosed with cardiogenic shock receiving veno-arterial (V-A) ECMO admitted to Department of intensive care unit (ICU) of Wuxi People's Hospital Affiliated to Nanjing Medical University from June 2012 to May 2017 were enrolled. Ten-minute OCT was implemented by transcutaneous oximetry at 6 hours after ECMO initiation. The basic data of patients (gender, age, primary disease); the acute physiology and chronic health evaluationⅡ (APACHE Ⅱ) score, sequential organ failure assessment (SOFA) score, left ventricular ejection fraction (LVEF), mean arterial pressure (MAP) at the start of ECMO treatment; arterial blood gas analysis index, dose of vasoactive agents, transcutaneous oxygen pressure (PtO2), 10-minute OCT value (OCT10), oxygen challenge index (OCI) at 6 hours after ECMO initiation; and the ECMO support time, duration of mechanical ventilation and its parameters, and application of intra-aortic balloon pump (IABP) within 60 days were recorded. Patients were divided into the survival group and the death group according to their 60-day mortality status, and the differences between the two groups were compared. Receiver operating characteristic curve (ROC) analysis was used to analyze the prognostic value of OCT10 and OCI. According to the best boundary value of OCT10and OCI, Kaplan-Meier survival curve was drawn and the 60-day cumulative survival rate was compared. The risk factors affecting prognosis were analyzed by multivariate Logistic regression. Results Sixty-seven patients were finally enrolled in the study, with 31 in the survival group and 36 in the death group. Compared with the survival group, APACHE Ⅱ score, SOFA score, use of IABP in death group were higher, PtO2, OCT10and OCI were lower, and duration of ECMO and ventilation were longer, but there was no significant difference in gender, age, primary disease, LVEF, MAP, ventilator settings, dose of vasoactive agents, or results of arterial blood gas between the two groups. OCT10, OCI, APACHE Ⅱ score and SOFA score were predictive values for 60-day deaths, and the area under ROC curve (AUC) of OCT10was 0.866±0.042 [95% confidence interval (95%CI) = 0.760-0.937], the AUC of OCI was 0.829±0.051 (95%CI = 0.717-0.910), the AUC of APACHE Ⅱ score was 0.860±0.043 (95%CI = 0.754-0.933), and the AUC of SOFA score was 0.821±0.049 (95%CI = 0.708-0.904) (all P < 0.01). The cut-off point for OCT10was ≥70.0 mmHg (1 mmHg = 0.133 kPa) with the sensitivity of 91.67% and the specificity of 67.74%. The cut-off point for OCI was ≥0.68 with the sensitivity of 88.68% and the specificity of 71.58%. According to the cut-off point for OCT10or OCI, the 60-day cumulative survival rate of patients with high OCT10was significantly higher than that of low OCT10[58.06% (18/31) vs. 36.11% (13/36), χ2= 5.425, P = 0.020];the survival rate in high OCI group was significantly higher than that in low OCI group [55.17% (16/29) vs. 39.47% (15/38), χ2= 5.119, P = 0.024]. It was shown by multivariate Logistic regression that OCT10[odds ratio (OR) = 0.883, 95%CI = 0.791-0.965, P = 0.006] and OCI (OR = 0.011, 95%CI = 0.001-0.087, P = 0.005) were independent risk factors for 60-day mortality. Conclusion OCT could predict the prognosis of patients with cardiogenic shock receiving ECMO.