Indomethacin tocolysis increases postnatal patent ductus arteriosus severity

Postnatally, therapeutic indomethacin administration is usually effective in mediating patent ductus arteriosus (PDA) constriction in premature infants. There are infants, however, who remain resistant to indomethacin and require more aggressive surgical intervention to facilitate ductal closure. Indomethacin tocolysis has been reported to increase the incidence of persistent PDA in premature infants. It was our impression that infants exposed to antenatal indomethacin not only suffered from an increased incidence of PDA, but that they were more symptomatic from PDA and that for them, PDA was more resistant to medical closure. It is this observation that we sought to examine in this study. Methods. Medical records of all mothers and premature neonates with birth weight less than or equal to 1500 g, admitted to the neonatal intensive care unit of the Shaare Zedek Medical Center during 1996 and 1997, who survived for at least 1 week, were reviewed retrospectively. Data on maternal indomethacin and steroid exposure, birth weight and gestational age, and ductus status and treatment were analyzed. In our obstetrics department, indomethacin is the medication of choice to inhibit premature labor. Mothers who arrive in premature labor are started on indomethacin therapy, if delivery is not imminent. All infants less than or equal to 1500 g were studied by a pediatric cardiologist between 24 and 72 hours of life using two-dimensional echocardiography with color flow mapping to assess ductal patency. Decisions to treat were based on echocardiographic evidence of PDA, along with any of the following clinical signs: bounding pulses, diastolic pressure of less than or equal to 25 mm Hg, pulmonary plethora and/or cardiomegaly on chest x-ray, or increasing oxygen requirement with no other explanation. Initial treatment is with indomethacin, if there are no contraindications. Our general approach is to begin therapy with a continuous indomethacin infusion, followed by a course of bolus indomethacin if the infant does not respond. However, each attending neonatologist may treat according to his/ her preference tie, bolus vs continuous). All infants with PDA are followed with serial echocardiographic examinations until the ductus is closed. Results. A total of 105 premature infants met the above criteria. Thirty-six of these 105 infants had echocardiographic signs of a PDA (34.3%). Those with PDA were less mature (gestational age, 28.9 +/- 2.6 vs 30.3 +/- 2.6 weeks, respectively) and tended to be smaller (1060 +/- 270 vs 1166 +/- 261 g). Of the 36 infants with PDA, 15 (42%) resolved spontaneously and 21 (58%) were symptomatic and required treatment with indomethacin. There were no differences in gestational age or birth weight between infants whose PDA resolved spontaneously and those requiring indomethacin therapy. Four of the 21 (19%) treated infants remained unresponsive to indomethacin and required ductal ligation. Of 17 infants with PDA who responded to indomethacin therapy, 1 (6%) was treated with a single course of bolus indomethacin, to which he responded, and 16 (94%) were treated with continuous indomethacin and responded promptly. The differences in therapeutic responsiveness to initial treatment with continuous vs bolus indomethacin were not significant. Of the 105 infants, 29 were exposed to indomethacin tocolysis. Those who were exposed to antenatal indomethacin and those who were not were well-matched with respect to birth weight and gestational age. Fifteen (52%) of the 29 exposed infants versus 18 (24%) of the 76 infants not exposed to antenatal indomethacin developed a PDA postnatally (relative risk = 2.1; 95% confidence interval: 1.22-3.74), and 45% of the antenatally exposed infants versus 12% of the nonexposed infants were symptomatic and required indomethacin (relative risk = 1.9; 95% confidence interval: 1.17-3.20). Four of the exposed infants versus none of the unexposed infants required surgical ligation. Among the indomethacin-exposed infants, the nonresponsive and responsive infants were well-matched with regard to birth weight, gestational age, antenatal steroid exposure, and day of life on which indomethacin therapy was initiated. Multiple regression analyses found prenatal indomethacin exposure to be the most significant antecedent variable associated with both the incidence and the severity of PDA, as indicated by the need for indomethacin treatment. Conclusions. We have demonstrated that prenatal indomethacin exposure increases both the incidence and the clinical severity of postnatal PDA, as manifested by increased need for therapeutic indomethacin and surgical ligation. Furthermore, we have shown it to be a more significant risk factor than gestational age, birth weight, or antenatal steroid exposure in both the development and the severity of postnatal PDA. These data should be considered in considerations as to choice of tocolytic therapy.