Which Factors Are Associated With Death, Local Recurrence, and Perioperative Complications After En Bloc Resection...

Primary malignant tumors of the mobile spine are rare but aggressive, and treatment decisions must balance oncologic control against neurologic risk, perioperative morbidity, and reconstruction durability. Because en bloc resection is technically demanding and associated with substantial morbidity, clinically important unanswered questions include which factors are associated with death and local recurrence after surgery, how prior procedure or local recurrence affects outcomes, and how often complications and revision for mechanical instrumentation problems occur.

(1) What were the overall survival outcomes after en bloc resection, and which factors were associated with death? (2) What were the local recurrence-free survival outcomes after en bloc resection, and which factors-particularly margin status and prior procedure or local recurrence-were associated with local recurrence? (3) What were the proportions and patterns of perioperative complications, postoperative neurologic deterioration, and revision for mechanical instrumentation problems?

We performed a retrospective comparative study using a longitudinally maintained institutional registry at a tertiary referral musculoskeletal tumor center. The study end date was set at December 2023; two patients who underwent surgery after that date were excluded because they no longer met the temporal inclusion criteria. Between August 2007 and December 2023, we included 70 patients with primary malignant tumors of the mobile spine who underwent planned en bloc resection when an oncologically appropriate margin was judged technically feasible and durable local control was expected to provide clinical benefit. Among these 70 patients, 84% (59 of 70) achieved at least 2 years of surveillance, 13% (9 of 70) died within 2 years, and 3% (2 of 70) were lost to follow-up before 2 years for reasons other than death. In addition, 14% (10 of 70) had not been seen within the last 5 years, although 9 of those 10 had already died. The mean ± SD age was 36 ± 17 years, 66% (46 of 70) were male, and chondrosarcoma was the most common histologic subtype, accounting for 33% (23 of 70) of patients. Tumors were staged using the Enneking and Weinstein-Boriani-Biagini systems. Overall survival and local recurrence-free survival were estimated using Kaplan-Meier methods, and factors associated with these outcomes were evaluated using Cox regression. Factors associated with perioperative complications were evaluated using multivariable logistic regression.

Overall survival after en bloc resection was 67.0% at 5 years (95% confidence interval [CI] 54.3% to 79.7%). In the multivariable Cox model, high-grade tumor (HR 8.9 [95% CI 1.1 to 70.3]; p = 0.04), marginal margin (HR 4.4 [95% CI 1.1 to 17.5]; p = 0.04), and intralesional margin (HR 11.3 [95% CI 2.8 to 46.1]; p = 0.001) were associated with a higher hazard of death. Local recurrence-free survival was 67.5% at 5 years (95% CI 55.0% to 80.0%), and local recurrence developed in 31% (22 of 70) of patients. In the multivariable Cox model, prior intralesional procedure, open biopsy, or local recurrence (HR 2.9 [95% CI 1.2 to 7.1]; p = 0.02), marginal margin (HR 4.6 [95% CI 1.2 to 17.8]; p = 0.03), and intralesional margin (HR 9.7 [95% CI 2.6 to 36.5]; p = 0.001) were associated with a higher hazard of local recurrence. Perioperative complications occurred in 50% (35 of 70) of patients, including 18 major complication events. Postoperative neurologic deterioration occurred in 17% (12 of 70) of patients and was more frequent in patients with prior procedure or local recurrence than in those with no prior procedure other than core needle biopsy (31% [8 of 26] versus 9% [4 of 44], OR 4.4 [95% CI 1.2 to 16.7]; p = 0.02). Revision for mechanical instrumentation problems occurred in 6% (4 of 70) of patients; all four revisions were performed for posterior rod fracture. Although revision was more frequent in patients reconstructed with titanium mesh cages than in those reconstructed with three-dimensionally-printed vertebral body prostheses (30% [3 of 10] versus 2% [1 of 52], OR 21.9 [95% CI 2.0 to 240.2]; p = 0.01), no anterior column construct failed or required revision.

For selected patients with primary malignant tumors of the mobile spine, en bloc resection can provide durable local control, but the likelihood of benefit depends on achieving an oncologically appropriate margin and avoiding unplanned intralesional procedures or open biopsy before referral. These findings support early referral to specialized centers, image-guided biopsy rather than unplanned open procedures, and multidisciplinary planning before definitive surgery. Future multicenter studies should incorporate tumor-specific analyses and longitudinal functional outcomes to better guide patient selection and counseling.

Level III, therapeutic study.