Back to Annual Meeting Program
Blood Transfusions in Autologous Breast Reconstructions: Risk Factors, Complications, and an Evidence-Based Algorithm
John P. Fischer, MD, Brady Sieber, BA, Carrie Stransky, MD, Seema Sonnad, Phd, Stephen J. Kovach, MD, Suhail Kanchwala, MD, Joseph M. Serletti, MD, Liza C. Wu, MD.
Hospital of the Univesity of Pennsylvania, Philadelphia, PA, USA.
Purpose: Free tissue transfer requires lengthy operative times and can be associated with significant blood loss and morbidity. Perioperative transfusions in autologous breast reconstruction have been cited to occur at rates upwards of 25-95% and may be associated with significant complications (Appleton et al, 2011). However, there is currently no risk assessment tool or evidence-based approach to guide perioperative blood transfusions. The goals of our study are to delineate perioperative risk factors for transfusion, determine transfusion-related complications and costs, and provide an evidence-based approach to optimize patient selection, minimize morbidity, and conserve resources.
Methods: We reviewed our prospectively maintained free flap database and identified all breast reconstruction patients receiving blood transfusions. These patients were compared to those not receiving a postoperative transfusion. Specific variables examined included: baseline patient co-morbidities, preoperative and postoperative hemoglobin levels, intra-operative and postoperative complications, and blood transfusions. Statistical analysis was performed and all tests were two-tailed, and statistical significance was defined as p<0.05.
Results: We examined 849 patients undergoing 1265 free tissue transfers (395 unilateral and 454 bilateral flaps). Immediate tissue transfer was performed in 637 patients and delayed transfer in 212 patients. Flaps used included: msTRAM (66.8%), DIEP (22.4%), SIEA (6.4%), and IGAP/SGAP (4.4%). A total of 70 (8.2%) postoperative transfusions occurred with 2.1 units per episode. Several perioperative risk factors were associated with transfusions (Table 1.) Multivariate analysis revealed preoperative anemia (18.1% vs. 6.3%, p=0.001), length of surgery (559 min vs. 458 min, p=0.04), surgeon experience (11.2% vs. 6.0%, p=0.05), intra-operative arterial thrombosis (19.2% vs. 7.9%, p=0.025), and delayed venous thrombosis (45.5% vs. 7.8%, p=0.006) were associated with transfusion.
Patients receiving transfusions experienced higher rates of complications, including: fat necrosis (16.3% vs. 7.3%, OR=2.3, p=0.034), seroma (19.4% vs. 7.7%, OR=2.9, p=0.02), and flap loss (40.0% vs. 7.3%, OR=8.5, p=0.0001). Additionally, these patients experienced longer hospitalizations (5.5 vs. 4.1 days, p<0.0001) and higher rates of medical complications (24.3% vs. 4.3%, OR=7.5, p<0.0001). Medical complications included: arrhythmias (7.1% vs. 2.4%, OR=3.3, p=0.03) and CHF (5.7% vs. 0.1%, OR=47, p=0.001). A cost analysis demonstrated that transfusions added a direct cost of \,659 during our study period.
Conclusion: We provide a large review of perioperative factors associated with blood transfusions in autologous breast reconstruction and identify a useful, evidence-based algorithm geared toward optimizing blood transfusions and minimizing morbidity. Overall 8% of patients undergoing free tissue transfer required blood transfusions and we report several perioperative factors associated with blood transfusion. These included patient co-morbidities, immediate reconstructions, preoperative anemia, thrombotic events, and surgeon experience. Blood transfusions are additionally associated with added hospital costs and greater surgical and medical complications. We provide a tool to guide preoperative risk assessment and management of high-risk patients undergoing autologous breast reconstructions (Figure 1.).
Back to Annual Meeting Program