Common origin of the carotid arteries (COCA) is a normal anatomic variant reported to occur in approximately 10 percent of the general population. To determine whether this variant places venoarterial ECMO patients at a higher risk for adverse neurologic sequela due to potential occlusion of both carotid arteries by the arterial cannula, we retrospectively examined the clinical records and echocardiograms of our initial 100 ECMO patients. All echocardiograms were reviewed for aortic arch morphology by a pediatric cardiologist, blinded to all other data. We excluded patients on ECMO beyond 30 days of age (n=2), infants with congenital heart disease (n=6), and infants for whom the echocardiogram was not available at our institution or the arch vessels were not adequately visualized (n=24). The remaining 68 patients were separated into 2 groups based on the aortic arch branching pattern; those with separate origin of the carotids (n=59) and those with COCA (n=9). There was no statistically significant difference between groups in terms of birth weight, estimated gestational age, age at onset of ECMO and length of ECMO support. The neurologic outcome variables studied included the results of MRI, CT, EEG, BAER, head ultrasound and Bayley Scales of Infant Development, reported as Physical Development Index (PDI) and Mental Development Index (MDI). Multiple regression analysis indicated no predictive value of COCA in determining PDI and MDI outcomes (mean age 24 months, range 6-30 months, n=39 normal arch, n=9 COCA). A logistical regression analysis revealed that COCA showed no significance in predicting adverse neurologic sequela based on MRI, CT, EEG, BAER or head ultrasound. In conclusion, our study confirms that COCA is a common aortic arch variant (13%) and that this variant does not increase the risk of neurologic injury in patients undergoing venoarterial ECMO.
