Survival Predictors for Severe ARDS Patients Treated with Extracorporeal Membrane Oxygenation: A Retrospective Study in China

Abstract

Extracorporeal membrane oxygenation (ECMO) is increasingly being applied as life support for acute respiratory distress syndrome (ARDS) patients. However, the outcomes of this procedure have not yet been characterized in severe ARDS patients. The aim of this study was to evaluate the outcomes of severe ARDS patients supported with ECMO and to identify potential predictors of mortality in these patients. A total of 38 severe ARDS patients (aged 51.39±13.27 years, 32 males) who were treated with ECMO in the specialized medical intensive care unit of Guangzhou Institute of Respiratory Diseases from July 2009 to December 2014 were retrospectively reviewed. The clinical data of the patients on the day before ECMO initiation, on the first day of ECMO treatment and on the day of ECMO removal were collected and analyzed. All patients were treated with veno-venous ECMO after a median mechanical ventilation duration of 6.4±7.6 days. Among the 20 patients (52.6%) who were successfully weaned from ECMO, 16 patients (42.1%) survived to hospital discharge. Of the identified pre-ECMO factors, advanced age, a long duration of ventilation before ECMO, a higher Acute Physiology and Chronic Health Evaluation II (APACHE II) score, underlying lung disease, and pulmonary barotrauma prior to ECMO were associated with unsuccessful weaning from ECMO. Furthermore, multiple logistic regression analysis indicated that both barotrauma pre-ECMO and underlying lung disease were independent predictors of hospital mortality. In conclusion, for severe ARDS patients treated with ECMO, barotrauma prior to ECMO and underlying lung disease may be major predictors of ARDS prognosis based on multivariate analysis.

Factors associated with intraoperative extracorporeal membrane oxygenation support during lung transplantation

“…ECMO has been our preferred method of support since March 2012, and most patients were transplanted on intraoperative venoarterial ECMO. The ECMO implantation technique, circuits, and cannula types used at our institution have been reported elsewhere [13]. We used centrifugal pumps (Bioline, Maquet, Hirrlingen, Germany) at a flow rate of 3-5 L/min in all patients.…”

Outcomes and survival prediction models for severe adult acute respiratory distress syndrome treated with extracorporeal membrane oxygenation

“…For instance, chronic immunosuppression was associated with increased mortality both in Enger et al’s [54] score (OR 2.6, 95% CI 1.3–5.2) and the VV-ECMO [56] mortality score (OR 2.9, 95% CI 1.1–7.9). Liu et al [55] found an underlying lung disease (i.e. COPD, interstitial lung disease and lung cancer) to be an independent risk factor for mortality (OR 12.2, 95% CI 1.2–122.2; p  = 0.033).…”

A novel risk score for severe ARDS patients undergoing ECMO after retrieval from peripheral hospitals

“…The use of prone position was not reported in this data collection, despite the recently documented beneficial of this intervention . Similarly, other scores included variables which were either quite uncommon (such as a preexisting lung disease in the score by Liu et al . or difficult to assess in these patients (such as the pre‐ECMO neurological status assessed by the Glasgow Coma Scale included in the score by Cheng et al.…”

“…The seven risk scores published so far were all derived from patients already on ECMO support, thus being appropriate for predicting survival in these patients and not for selecting which patients should or should not be submitted to ECMO. The PRESERVE (PRedicting dEath for Severe ARDS on VV‐ECMO) study was published in 2013 and was constructed with eight pre‐ECMO parameters .…”

“…To our knowledge, seven risk scores have been suggested but they were derived from quite heterogeneous population and only three have been externally validated. Moreover, intrinsic heterogeneity of ARDS, differences in management and criteria to start ECMO, referral from peripheral centers make the exploitation of an easy and accurate scoring model quite challenging.…”