Diagnosis of patent ductus arteriosus by different echocardiographic methods in very preterm infants

There is no consensus regarding the timing and diagnostic criteria for identifying hemodynamically significant patent ductus arteriosus (hsPDA). Our aim was to evaluate if the use of different diagnostic criteria at different times could be associated with a different incidence of hsPDA in very preterm infants. We studied 41 infants with gestational age < 32 weeks born in neonatal intensive care units (NICU) in Florence, Italy, or in Paris, France. They received the first echocardiography between 24 and 48 h of life and the second between 72 and 84 h to diagnose hsPDA using Florence and Paris criteria and PDA severity score. Concordance of diagnosis between criteria was evaluated with the Cohen unweighted κ statistic. The incidence of hsPDA diagnosed by the Florence (35%) or Paris (34%) criteria or by PDA severity score (35%) was similar. Concordance was substantial between Florence and Paris criteria and between Florence criteria and PDA severity score but was fair between Paris criteria and PDA severity score. Moreover, concordance significantly changed from the first to the second echocardiography. Conclusion: The studied diagnostic criteria showed important variations of concordance when applied at different times. This led to diagnose hsPDA in different patients at different times while leaving the overall percentage of hsPDA unchanged. Our results suggest that more attention should be paid to the choice of diagnostic criteria for individuating hsPDA in very preterm infants. (Table presented.)