The Fenestrated Fontan procedure (FFP) has improved outcome in high risk patients. The technique is evolving, however, and complications are not fully known. Over a 3-year period 13 patients (mean age 35 +/- 29 months) underwent an FFP in our institution. In the first two patients the fenestration had to be created because of high right atrial pressure and low cardiac output; in 11 patients the FFP was planned. In three patients the sutures for the adjustable fenestration were crossing the defect. In 10 patients, purse-string sutures were placed around but not across the defect. Because large fenestrations were created in 11 patients (8-12 mm) Glenn shunts were performed to improve arterial saturation. The postoperative course was relatively uneventful, with chest tubes being removed 1-8 days (mean 4 +/- 3 days) postoperatively and the hospital stay ranging from 7 to 27 days (mean 14 +/- 6 days). One patient had bleeding and another had a mediastinal abscess. The first patient died (7.6%) because of hemodynamic instability due to prolonged cardiopulmonary bypass from the creation and enlargement of the fenestration. One patient had a paradoxical cerebral embolism from clots that formed on the sutures crossing the fenestration. Because of this problem the remaining patients were placed on salicylates while awaiting closure of their fenestration. All 12 patients had their fenestrations closed, performed under local anesthesia in 9, at mediastinal abscess drainage in 1, and spontaneously in 2. We conclude that creation of large fenestrations in combination with Glenn shunts and the use of adjustable fenestrations are viable modifications of the FFP. The use of purse-string sutures around the fenestration and antiplatelet drugs can probably minimize the occurrence of paradoxical embolism.