ABSTRACT
Objective:To explore the changing trend of cardiac troponin T (cTnT) in patients with cardiogenic shock (CS) receiving veno-arterial extracorporeal membrane oxygenation (V-A ECMO) and its predictive value.Methods:A retrospective study was conducted. The data of patients with CS receiving V-A ECMO admitted to the First Affiliated Hospital of Nanjing Medical University from March 2015 to May 2020 were enrolled. The baseline data, ECMO related parameters, serum cTnT levels at 1, 2, 3 days after ECMO and intensive care unit (ICU) prognosis were recorded. The parameters with clinical significance and significant difference in univariate analysis were analyzed by binary multivariate Logistic regression analysis. Meanwhile, receiver operating characteristic (ROC) curve was drawn, area under ROC curve (AUC) was analyzed, and the threshold, sensitivity and specificity of serum cTnT level and its reduction rate for predicting clinical outcome were evaluated.Results:A total of 72 patients were enrolled, of which 42 survived and 30 died at ICU discharge, and the ICU mortality was 41.7%. Univariate analysis results: compared with the survival group, the patients in the death group had higher acute physiology and chronic health evaluation Ⅱ (APACHE Ⅱ) score [32 (30, 34) vs. 29 (25, 30)], and the incidence of cardiac arrest before ECMO (70.0% vs. 31.0%), the ratios of invasive mechanical ventilation and continuous renal replacement therapy during ECMO were higher (96.7% vs. 66.7%, 83.3% vs. 42.9%), and the differences were statistically significant (all P < 0.05). Serum cTnT levels (ng/L) at 2 days and 3 days after ECMO in the death group were significantly higher than those in the survival group [2 days: 6 373.5 (898.3, 15 251.5) vs. 1 760.5 (933.0, 4 257.8), 3 day: 6 202.0 (758.9, 16 554.3) vs. 1 678.0 (623.3, 3 407.8), both P < 0.05], and the decrease rates of cTnT within 2 days and 3 days after ECMO were significantly lower than those in the survival group [2 days: 17.3% (-44.2%, 34.7%) vs. 36.8% (18.1%, 60.6%), 3 days: 32.4% (-30.0%, 55.5%) vs. 53.2% (38.3%, 72.3%), both P < 0.05]. Binary multivariate Logistic regression analysis showed that cardiac arrest before ECMO [odds ratio ( OR) = 4.564, 95% confidence interval (95% CI) was 1.212-17.193, P = 0.025] and the decrease rate of cTnT level within 2 days after ECMO ( OR = 1.617, 95% CI was 1.144-4.847, P = 0.026) were independent prognostic risk factors for the ICU death of CS patients receiving V-A ECMO. ROC curve analysis showed that the decline rate of cTnT within 2 days after ECMO transfer had a certain predictive value for the ICU death of CS patients receiving V-A ECMO. The AUC was 0.704 (95% CI was 0.584-0.824). The optimal diagnostic threshold was 40.0%, the sensitivity was 86.7%, the specificity was 52.4%, the positive predictive value was 66.9%, and the negative predictive value was 89.1%. Conclusions:The early decline rate of cTnT in CS patients who received V-A ECMO treatment in death group was lower than that of survival patients. The cTnT decline rate 2 days after ECMO was an independent risk factor for the death of such patients.
ABSTRACT
Objective To determine the predictive values of different critical scoring systems for survival rate after discharge in critically ill patients supported by extracorporeal membrane oxygenation (ECMO). Methods The clinical data of 34 critically ill patients supported by ECMO admitted to Department of Emergency of the First Affiliated Hospital of Nanjing Medical University (Jiangsu Provincial People's Hospital) from July 2015 to September 2017 were retrospectively analyzed. The general information and the worst values of vital signs and related pathophysiological indicators within 12 hours before ECMO treatment of patients were collected, and sequential organ failure assessment (SOFA), multiple organs dysfunction score (MODS), simplified acute physiology score Ⅱ (SAPSⅡ), and acute physiology and chronic health evaluation Ⅳ(APACHEⅣ) scores were calculated. The patients were divided into survival group and non-survival group according to 28-day survival after hospital discharge. General clinical characteristics and aforementioned scores were compared between the two groups. Scoring systems for predicting prognosis were assessed by using the receiver operating characteristic (ROC) curve. The Kaplan-Meier method was used to depict the surviving curve. Results Thirty-four patients were finally enrolled, 13 of whom were dead at the follow-up period of 28 days after hospital discharge, and 21 survived. Duration of ECMO support in non-survival group was significantly shorter than that in survival group (hours: 101.4±7.8 vs. 134.4±12.6), SOFA, SAPSⅡ, and APACHEⅣ scores were significantly higher than those of survival group (SOFA score: 10.6±3.6 vs. 8.8±3.3, SAPSⅡscore: 38.7±14.3 vs. 31.8±12.5, APACHEⅣ score: 46.5±15.5 vs. 38.1±11.3, all P < 0.05). There was no significant difference in gender, age, body mass index (BMI), vital signs or related pathophysiological indicators within 12 hours before ECMO treatment, or MODS score between the two groups. ROC curve analysis showed that the area under ROC curve (AUC) of SAPSⅡ score for predicting 28-day survival rate was the highest, which was significantly higher than that of SOFA, MODS, and APACHEⅣ score (0.880 vs. 0.694, 0.654, 0.682, all P < 0.05). When the best cut-off value of SAPSⅡ score was 43, the sensitivity was 81.2%, and the specificity was 77.9%. Kaplan-Meier survival analysis showed that 28-day survival rate after hospital discharge in patients with SAPSⅡ score < 43 (n = 18) was significantly higher than that in patients with SAPSⅡ score ≥43 (n = 16; χ2= 2.444, P = 0.018). Conclusions Four critical scoring systems of SOFA, MODS, SAPSⅡand APACHEⅣ have been proved to have good prognostic ability to predict 28-day survival after hospital discharge in critically ill patients supported by ECMO. Among them, SAPSⅡ score system has more accurate prediction value.