Improving diagnostic accuracy in the transport of infants with suspected duct-dependent congenital heart disease.

AIM To identify factors that distinguish duct-dependent congenital heart disease (DDCHD) from non-DDCHD in newborn infants. METHOD A retrospective, cohort study. The Newborn Emergency Transport Service, Victoria (NETS) is a retrieval service for all inter-hospital neonatal transfers, and the Royal Children's Hospital, Melbourne (RCH) is a paediatric cardiac referral centre for the state of Victoria, Australia. All infants ≤10 days and ≥34 weeks gestation with suspected CHD and/or persistent pulmonary hypertension of the newborn (PPHN), transferred by NETS from non-tertiary neonatal units to RCH, over a 4-year period. RESULTS Of 142 eligible infants, 81 had DDCHD and 61 had non-DDCHD, of whom 51 had PPHN. Diagnostic accuracy of DDCHD by the NETS team was 77%. Presence of a heart murmur, abnormal pulses, upper and lower limb blood pressure (BP) difference >10 mmHg, cardiomegaly, initial SpO2 of <92%, PaO2 <50 mmHg, and pre-post ductal SpO2 difference >10% were significantly associated with DDHCD on univariate analysis. No single clinical finding was significantly associated with DDCHD on multivariate analysis. Labile SpO2 , abnormal lung parenchyma, mean BP <40 mmHg, pH <7.25, lactate >5 and FiO2 >0.5 were significantly associated with non-DDCHD, but at multivariate analysis only labile SpO2 and mean BP <40 mmHg were associated with non-DDCHD. CONCLUSIONS Clinical diagnosis of DDCHD outside of a cardiac centre is challenging. No single factor predicts DDCHD. Combined interpretation of clinical, physiological and x-ray findings may assist.