The objective of this study was to determine the rate of patent ductus arteriosus (PDA) closure in premature infants using an adjustable indomethacin (INDO) dosing strategy, based on a second-dose peak plasma INDO level. We conducted a retrospective review of the medical records of premature infants that were treated with INDO for a PDA, had a second dose peak plasma NDO levels, and followed predetermined guidelines for INDO dosing adjustments, over a 4-year period (1995 to 1998). Of 103 infants treated with the adjustable INDO dosing strategy, 66 (64%) achieved PDA closure whereas 37 (36%) did not. No differences in the second-dose peak plasma INDO levels (830 +/- 339 versus 702 +/- 381 ng/mL), day of life treatment was started (4 +/- 3 versus 4 +/- 2 days), or the number of doses of INDO received (4 +/- 1 versus 5 +/- 2 dose) were observed between responders and nonresponders. However, fourth-dose peak plasma INDO levels, which were available from 38 of 66 (57%) of the responders and 20 of 37 (54%) of the nonresponders, were lower in nonresponders (1553 +/- 413 versus 1829 +/- 609 ng/mL, p < 0.05). Patient demographics, including birth weight and gestational age, were similar between these groups. Using an adjustable INDO dosing strategy, based on a second-dose peak plasma INDO level and estimated plasma levels, PDA closure rates of 64% can be achieved. Although a clear relationship between INDO plasma levels and PDA closure was evident form this study, the rate of PDA closure in our study was lower than has been observed in studies with serial plasma INDO level monitoring.