Abstract 1412 Poster Session III, Monday, 5/3 (poster 43)

The incidence of NEC is highly variably ranging from 6 to 14% in very low birth weight (VLBW) infants. Although enteral indomethacin was implicated as a risk factor for NEC, the data was limited to case reports. Intravenous indomethacin, however, has been suggested as a risk factor in the development of NEC in several studies when used during either the pre or post natal period. Indomethacin has also been reported to decrease blood flow velocity in the mesenteric artery and increase intestinal permeability. Prior studies had not compared the incidence of NEC in medically and surgically treated patients. We hypothesized that neonates with a PDA treated with indomethacin would have a more frequent occurrence of NEC compared to patients who underwent ductal ligation. We conducted a case-control investigation using a prospectively collected database of neonates whose birth weight was 501 to 1500 grams and admitted to the neonatal intensive care unit (NICU) of Texas Children's Hospital in Houston, TX from July 1, 1990 through December 31, 1997. During this period, 1694 infants were admitted to the NICU within this birth weight range. 480 (28%) patients were diagnosed with a clinical PDA which was confirmed with an echocardiogram. 455 of the infants (95%) with a PDA received medical and/or surgical intervention, while 25 (5%) received no treatment for this diagnosis. Of those infants given treatment, 253 (56%) received indomethacin only, 154 (34%) surgical ligation only, and 48 (11%) indomethacin plus surgical ligation for ductal closure. The incidence of NEC (defined as Bell's stage IIa or greater) was significantly increased in the group treated with ligation only (n=36, 23%, p<0.0002) or with indomethacin plus ligation (n=10, 21%, p<0.0006) versus the group treated only with indomethacin (n=25, 10%). However, on average, those infants treated with ligation only were lighter in birth weight, younger in gestational age, and had significantly increased mortality, postnatal steroid usage, late-onset sepsis, early-onset sepsis, central line placement, and use of mechanical ventilation. When adjusted for all potential confounders identified above, the ligation only group still had a significantly greater incidence of NEC (Odds Ration 2.75, Confidence Interval 1.49-5.10). These data indicate that surgical ligation, as compared to indomethacin, increases the likelihood of a patient developing NEC once a PDA is diagnosed. Although future analysis of this data may be impacted by the amount of left to right shunt or other factors related to the PDA, clinicians should consider indomethacin closure of a PDA unless contraindicated. We suggest that clinicians utilize indomethacin closure of a PDA as their first therapeutic option.