Closer to an understanding of fate: the role of vascular complications in free flap breast reconstruction.

BACKGROUND

Flap loss in autologous breast reconstruction is almost invariably a result of a vascular event; however, not all events portend the same fate. The purpose of this study was to determine whether intraoperative vascular complications predict subsequent postoperative vascular thrombosis and flap loss.

METHODS

This was a retrospective review of all free flaps performed between 2005 and 2010 in the University of Pennsylvania Health System for breast reconstruction. Details of all vascular thromboses and intraoperative technical difficulties were recorded. Flaps with a routine intraoperative course were compared with those with an intraoperative complication.

RESULTS

Overall, 1173 free flaps were performed in 804 patients. In regression analysis, an intraoperative vascular complication was the only independent predictor of a subsequent delayed vascular complication (odds ratio, 3.11; 95 percent confidence interval, 1.25 to 7.73). In subanalysis, intraoperative arterial thrombosis was not associated with a subsequent delayed arterial thrombosis; however, intraoperative technical difficulties were associated with increased delayed arterial thrombosis (1.0 percent versus 4.2 percent, p=0.05) and partial flap loss (0.6 percent versus 4.2 percent, p=0.02). There was a trend toward increased delayed venous thromboses following intraoperative venous thromboses (1.1 percent versus 16.7 percent, p=0.07). In aggregate, there was a higher rate of complete flap loss following any intraoperative vascular complication (0.9 percent versus 3.5 percent, p=0.04).

CONCLUSIONS

In free flap breast reconstruction, an intraoperative vascular problem leads to increased risk of a subsequent postoperative vascular complication and flap loss. Postoperative vascular complications do not appear to be overtly affected by specific surgical intervention or choice of anticoagulation in the setting of a preceding intraoperative problem.

CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, III.