A novel risk score for severe ARDS patients undergoing ECMO after retrieval from peripheral hospitals. Issue 1 (23rd October 2017)

• Record Type: Journal Article
• Authors: Lazzeri, C.; Cianchi, G.; Mauri, T.; Pesenti, A.; Bonizzoli, M.; Batacchi, S.; Chiostri, M.; Socci, F.; Peris, A.
• Is Part Of: Acta anaesthesiologica scandinavica. Volume 62:Issue 1(2018:Jan.)
• Page Start: 38
• Page End: 48
• Publication Date: 2017-10-23
• Subjects: Anesthesiology -- Periodicals; Critical care medicine -- Periodicals; 617.9605
• DOI: 10.1111/aas.13022
• Languages: English
• ISSNs: 0001-5172

Abstract

Abstract : Background: Extracorporeal Membrane Oxygenation in severe ARDS unresponsive to conventional protective ventilation is associated with elevated costs, resource and complications, and appropriate risk stratification of candidate patients could be useful to recognize those more likely to benefit from ECMO. We aimed to derive a new outcome prediction score for patients retrieved by our ECMO team from peripheral centers, including systematic echocardiographic evaluation before ECMO start. Methods: Sixty‐nine consecutive patients with refractory ARDS requiring ECMO transferred from peripheral centers to our ICU (a tertiary ECMO referral center), from 1 October 2009 to 31 December 2015, were assessed. Results: All patients were transported on ECMO (distance, median 77, range 4–456 km) The mortality rate was 41% (28/69). Our new risk score included age ≥ 42 years, BMI < 31 kg/m 2, RV dilatation, and pH < 7.35. The proposed cut off (Youden's index method) of nine had a sensitivity of 96% and a specificity of 30% (AUC‐ROC: 0.85, 95% CI: 0.76–0.94, P < 0.001). When assessing the discriminatory ability of our risk score in the population of local patients, survivors had a mean value of 15.4 ± 8.6, whereas non‐survivors showed a mean value of 20.1 ± 7.4 ( P < 0.001). Conclusions: Our new risk score shows good discriminatory ability both in patients retrieved from peripheral centers and in those implanted at our center. This score includes variables easily available at bedside, and, for the first time, a pathophysiologic element, RV dilatation.