Echocardiography, not chest radiography, for evaluation of cannula placement during pediatric extracorporeal membrane oxygenation

• Published in: Pediatric critical care medicine Vol. 10; no. 1; p. 56
• Main Authors: Thomas, Timothy H, Price, Ryan, Ramaciotti, Claudio, Thompson, Marita, Megison, Steve, Lemler, Matthew S
• Format: Journal Article
• Language: English
• Published: United States 01.01.2009
• Subjects: Adolescent; Catheterization - adverse effects; Catheterization - methods; Child; Child, Preschool; Cohort Studies; Critical Care - methods; Critical Illness - therapy; Echocardiography, Doppler - methods; Extracorporeal Membrane Oxygenation - adverse effects; Extracorporeal Membrane Oxygenation - instrumentation; Extracorporeal Membrane Oxygenation - methods; Female; Foreign-Body Migration - diagnostic imaging; Humans; Infant; Infant, Newborn; Intensive Care Units, Pediatric; Male; Monitoring, Physiologic - methods; Radiography, Thoracic - methods; Retrospective Studies; Risk Factors; Sensitivity and Specificity; Young Adult
• ISSN: 1529-7535; 1947-3893
• DOI: 10.1097/PCC.0b013e3181937409

Optimal cannula position is essential during extracorporeal membrane oxygenation (ECMO). We hypothesize that echocardiography is superior to chest radiography in diagnosing abnormal cannula position during ECMO. Retrospective. Pediatric hospital. 100 pediatric patients requiring ECMO. We reviewed the medical records of all ECMO patients (n = 100), including reports of all echocardiograms (n = 326), during the years 2002-2004. Of the 91 patients who had echocardiograms while on ECMO, 33 had at least 1 echocardiogram for cannula-position evaluation. Of the remaining 58 patients with echocardiograms for reasons other than cannula-position evaluation, 4 (7%) were found to have abnormal cannula position. These included arterial cannula (AC) within 2-4 mm of the aortic valve (n = 2), AC across the aortic valve into the left ventricle (n = 1), and venous cannula (VC) abutting the atrial septum (n = 1). Of the 33 patients with echocardiograms for evaluation of cannula position, 8 (24%) required intervention. Of those 8 patients, 4 required cannula repositioning due to VC in the coronary sinus (n = 1), VC abutting atrial septum (n = 1), AC in left subclavian artery (n = 1), and AC within 3 mm of aortic valve (n = 1). The remaining 4 with normal cannula position required upsizing of the VC (n = 2), increased circuit flow (n = 1), or intravascular volume administration (n = 1). Overall, 12 of 91 patients (13%) required intervention based on echocardiographic findings. Chest radiography did not detect abnormalities of ECMO cannula position in any of the 8 patients with this problem, nor were any additional patients with abnormal cannula position identified by chest radiography. Echocardiography appears to be superior to chest radiography for assessing ECMO cannula position in our institution. A prospective study, including cost analysis, comparing chest radiography and echocardiography, is needed to definitely determine the preferred diagnostic test or sequence of tests to establish ECMO cannula position.