EVOLUTION OF THE TECHNIQUE OF CONGENITAL DIAPHRAGMATIC-HERNIA REPAIR ON ECMO

With the increase in popularity of delayed repair of congenital diaphragmatic hernia (CDH), many institutions are using extracorporeal membrane oxygenation (ECMO) to stabilize patients preoperatively. This practice has led to controversy regarding whether the repair should be performed while the patient is on ECMO or after decannulation. This report details the authors' experience with repair of CDH on ECMO. Of the 154 high-risk CDH patients treated at Children's Hospital, Boston, MA, since ECMO became available (1984), 97 received ECMO, including 31 who had repair performed while on ECMO. In group I (nine patients), repair was carried out only if the patients were unweanable from ECMO after 7 days. Activated clotting times (ACT) were maintained at 200 to 220 seconds. In group II (22 patients), repair was performed on ECMO electively, before decannulation. ACT were maintained at 180 to 200 seconds. Additionally, all patients in group II received aminocaproic acid before surgery. This was administered continuously for 72 hours postoperatively or until decannulation. Patients in group II had significantly less overall blood loss (P = .02) and lower transfusion requirement (P = .0003) than those in group I. Additionally, four of the nine patients in group I required reexploration because of hemorrhage; this was not required for any patient in group II (P = .005). Although the survival rates differed, this may have been because of a bias in patient selection between the two groups. From these preliminary data, the authors conclude that repair of congenital diaphragmatic hernia on ECMO can be performed safely, with a minimum of hemorrhagic complications. The improvement seen in this study is attributed to several factors: (1) increased experience of the surgical team in repairing CDH on ECMO, (2) a reduction in the intraoperative and postoperative ACT, and (3) use of aminocaproic acid, an antifibrinolytic agent. Whether repair on ECMO will have an impact on the overall survival of these patients remains to be determined; however, death secondary to hemorrhage need not contribute to the mortality. © 1994.