Safety of cardiac magnetic resonance and contrast angiography for neonates and small infants: a 10-year single-institution experience.

BACKGROUND: With increasing applications of cardiac magnetic resonance (CMR) and magnetic resonance angiography (MRA) for evaluation of congenital heart disease (CHD), safety of this technology in the very young is of particular interest.
OBJECTIVE: We report our 10-year experience with CMR in neonates and small infants with particular focus on the safety profile and incidence of adverse events (AEs).
MATERIALS AND METHODS: We reviewed clinical, anesthesia and nursing records of all children ≤120 days of age who underwent CMR. We recorded variables including cardiac diagnosis, study duration, anesthesia type and agents, prostaglandin E1 (PGE1) dependence and gadolinium (Gd) use. Serially recorded temperature, systemic saturation (SpO(2)) and cardiac rhythm were analyzed. Primary outcome measure was any AE during or <24 h after the procedure, including minor AEs such as hypothermia (axillary temperature ≤95 °F), desaturation (SpO(2) drop ≥10% below baseline) and bradycardia (heart rate ≤100 bpm). Secondary outcome measure was unplanned overnight hospitalization of outpatients.
RESULTS: Children (n = 143; 74 boys, 69 girls) had a median age of 6 days (1-117), and 98 were ≤30 days at the time of CMR. The median weight was 3.4 kg (1.4-6 kg) and body surface area 0.22 m(2) (0.13-0.32 m(2)). There were 118 (83%) inpatients (108 receiving intensive care) and 25 (17%) outpatients. Indications for CMR were assessment of aortic arch (n = 57), complex CHD (n = 41), pulmonary veins (n = 15), vascular ring (n = 8), intracardiac mass (n = 8), pulmonary artery (n = 7), ventricular volume (n = 4), and systemic veins (n = 3). CMR was performed using a 1.5-T scanner and a commercially available coil. CMR utilized general anesthesia (GA) in 86 children, deep sedation (DS) in 50 and comforting methods in seven. MRA was performed in 136 children. Fifty-nine children were PGE1-dependent and 39 had single-ventricle circulation. Among children on PGE1, 43 (73%) had GA and 10 (17%) had DS. Twelve children (9%) had adverse events (AEs)-one major and 11 minor. Of those 12, nine children had GA (10%) and three had DS (6%). The single major AE was respiratory arrest after DS in a neonate (resuscitated without sequelae). Minor AEs included desaturations (n = 2), hypothermia (n = 5), bradycardia (n = 2), and bradycardia with hypoxemia (n = 2). Incidence of minor AEs was 9% for inpatients (vs. 4% for outpatients), and 8% for neonates (vs. 9% for age ≥30 days). Incidence of minor AEs was similar between PGE1-dependent infants and the non-PGE1 group. There were no adverse events related to MRA. Of 25 outpatients, 5 (20%) were admitted for overnight observation due to desaturations.
CONCLUSION: CMR and MRA can be accomplished safely in neonates and infants ≤120 days old for a wide range of pre-surgical cardiac indications. Adverse events were unrelated to patient age, complexity of heart disease, type of anesthesia or PGE1 dependence.