Optimizing reconstruction of oncologic sternectomy defects based on surgical outcomes

Background: The optimal strategy for oncologic sternectomy reconstruction has not been well characterized. We hypothesized that the major factors driving the reconstructive strategy for oncologic sternectomy include the need for skin replacement, extent of the bony sternectomy defect, and status of the internal mammary vessels. Study Design: We reviewed consecutive oncologic sternectomy reconstructions performed at The University of Texas MD Anderson Cancer Center during a 10-year period. Regression models analyzed associations between patient, defect, and treatment factors and outcomes to identify patient and treatment selection criteria. We developed a generalized management algorithm based on these data. Results: Forty-nine consecutive patients underwent oncologic sternectomy reconstruction (mean follow-up 18 ± 23 months). More sternectomies were partial (74%) rather than total/subtotal (26%). Most defects (n = 40 [82%]) required skeletal reconstruction. Pectoralis muscle flaps were most commonly used for sternectomies with intact overlying skin (64%) and infrequently used when a presternal skin defect was present (36%; p = 0.06). Free flaps were more often used for total/subtotal vs partial sternectomy defects (75% vs 25%, respectively; p = 0.02). Complication rates for total/subtotal sternectomy and partial sternectomy were equivalent (46% vs 44%, respectively; p = 0.92). Conclusions: Despite more extensive sternal resections, total/subtotal sternectomies resulted in equivalent postoperative complications when combined with the appropriate soft-tissue reconstruction. Good surgical and oncologic outcomes can be achieved with defect-characteristic-matched reconstructive strategies for these complex oncologic sternectomy resections.