Abstract
Introduction
Primary tumors of the spine requiring enbloc resection are rare. Most published reports are small and focus on disease free survival and recurrence rates. Very few studies focus on the various anterior and posterior reconstructive options and subsequent outcomes with respect to fusion rates and need for revision due to hardware failure. The objective of this review was to (1) summarize the published literature and (2) report the failure rates of various anterior and posterior reconstructive techniques after enbloc resection of spinal tumors and (3) supplement the deficiencies in the available published literature with expert opinion for reconstructive options from a group of experienced international spine tumor surgeons.
Material and Methods
An electronic search of the literature was undertaken from January 1990 – December 2013 evaluating specific reconstructive techniques of the spine after primary tumor enbloc resection. Prospective/retrospective trials and case series were included in the final analysis when fusion rates or failure rates were reported. The data available for each reconstructive technique was then combined and construct survivorship was summarized. In addition, a questionnaire was administered to a group of 20 international spine tumor surgeons evaluating specific reconstructive preferences at different regions of the spine based on the number of vertebrae resected and whether post-operative radiation was planned.
Results
The initial search yielded 381 articles with 31 subsequently included for full text review. Fourteen articles were included in the final analysis. There were 146 patients included for final review. There were 2/9 (22%) patients revised from short to long segment constructs and 3 reports of broken pedicle screws with only one requiring revision in longer constructs. Rates of revision for anterior reconstruction were similar for autogenous strut grafts (10%), cages (7.7%) and allograft strut grafts (8.3%). No surgeons responding to the questionnaire recommended short segment posterior constructs. For anterior reconstruction, cages packed with morcellized allograft and autograft were preferred (75%, p < 0.05) while strut bone grafting was choosen more often at the cervicothoracic junction (65%, p < 0.05) and when more than one vertebrae was resected in the mid thoracic spine (75%, p < 0.05). Few surgeons changed their anterior reconstructive technique (15%) or posterior reconstructive technique (10%) when post-operative radiation was planned.
Conclusion
The literature and consensus opinion supports posterior reconstruction with at least two vertebral levels of support above and below. For anterior vertebral column reconstruction, structural allograft, autograft and cages packed with morcelized bone have shown similar rates of fusion and failure. Expert opinion, however, suggests that structural autograft or potentially vascularized strut grafts should be used when spanning a defect greater than 2 vertebral bodies especially at the cervicothoracic junction.