Soft Tissue Reconstruction after Concomitant Vertebrectomy and Chest Wall Resection for Spinal Tumors

Purpose: Oncologic resections involving both the spine and chest wall commonly require immediate soft tissue reconstruction to obliterate dead space, protect hardware, and separate the intrapleural space and exposed spinal cord with well-vascularized tissue.  Considering the increased magnitude of defects involving both the thoracic spine and chest wall, we hypothesized that these patients would have an increased complication rate compared to patients with defects of the thoracic spine alone. Additionally, we hypothesized that among the patients with spine and chest wall defects, those with muscle flap separation of the intrapleural space and spinal cord would have fewer complications.

Methods: We performed a retrospective review of prospectively-maintained data at a single center, identifying patients who underwent immediate reconstruction of thoracic spine wounds between 2006 and 2016. Patients were separated into two cohorts for comparison: resections limited to the thoracic spine (TS) and those also involving the chest wall (TS+CW). Surgical complications included wound infection, seroma, hematoma, skin edge separation, cerebrospinal fluid leak, hardware exposure/loss, and flap loss. Medical complications included pneumonia, venous thromboembolism, cardiac event, and neurologic event. We performed both univariate and multivariate logistic regression analyses to identify patient and surgical factors that were predictive or protective of postoperative complications.

Results: One-hundred patients were identified, 53 in the TS group and 47 in the TS+CW group. Mean follow up was 35 months and was equivalent between the two cohorts (38.5 months for TS+CW vs. 28.8 months for TS; p=0.35). Preoperative comorbidities were similar in both groups. Metastatic disease was more common for TS versus TS+CW patients (88.7% versus 38.3%;p=0.001). TS+CW patients were younger (48.7 vs. 58.2 years;p=0.001) and more frequently required: removal of two or more adjacent vertebra (70.2% vs. 17%;p=0.001), instrumentation of greater than 6 vertebral levels (76.6% vs. 26.4%;p=0.001), pleural space entry (87.2% vs. 3.8%;p=0.001), use of more than 3 flaps (74.5% vs. 43.3%;p=0.002), and postoperative mechanical ventilation (55.3% vs. 17%;p=0.001). Univariate and multivariate logistic regression analyses showed no difference between the two cohorts in surgical complications, complication requiring reoperation, or medical complications. However, mean 3-year survival was greater in the TS+CW group: 41.2% versus 23%. Subset analysis of the TS+CW group demonstrated that TS+CW patients that received muscle flap separation of the spinal cord from the intrapleural space developed fewer seromas (4.2% vs. 26.1%;p=0.04) and fewer overall complications (50% vs. 82.6%;p=0.03).

Conclusions: Despite the added morbidity of the resection and reconstruction of TS+CW defects, there is no increase in complications compared to TS patients. For TS+CW reconstructions, we advocate for separation of the intrapleural space and the spinal cord with a muscle flap, as this resulted in significantly fewer seromas and overall complications.