Instrumentation failure after lumbar spondylectomy for spinal tumors: a systematic review and pooled individual patient analysis

PURPOSE: Instrumentation failure (IF) is a major complication after lumbar spondylectomy for spinal tumors, yet risk factors remain poorly defined. The present study aimed to determine the prevalence of IF and identify variables associated with IF and revision surgery using pooled individual patient data. METHODS: A systematic review and individual patient data analysis were performed per PRISMA guidelines. PubMed, CDSR, and Epistemonikos were searched through January 2025, and additional patient-level data were obtained from prior series. Studies were included if they reported outcomes after lumbar spondylectomy for primary or metastatic tumors. Demographic, tumor, surgical, and (neo)adjuvant therapy variables were extracted. Statistical analyses included chi-square tests, t-tests, and Firth's penalized logistic regression. Variables with p < 0.1 on univariable analysis and considered clinically relevant were entered into a penalized multivariable model, with a sensitivity analysis incorporating estimated blood loss as a surrogate of operative burden. RESULTS: A total of 169 patients (mean age 40.7 ± 17.6 years) were included after screening. IF occurred in 14% (23/169) of patients, with 96% of those requiring revision. On univariable regression, risk factors for IF included combined approach (OR 4.72, p = 0.01), staged procedures (OR 5.51, p < 0.001), pelvic fixation (OR 5.17, p < 0.001), multilevel spondylectomy (OR 2.74, p = 0.036), longer operative time (OR 1.17 per hour, p = 0.001), and greater blood loss (OR 1.16 per liter, p = 0.034). On primary multivariable analysis incorporating clinically relevant variables, no variable retained statistical significance, although pelvic fixation demonstrated the strongest trend (OR 2.19, p = 0.12). On sensitivity analysis adjusting for operative burden, pelvic fixation was significantly associated with IF (OR 4.95, p = 0.043). Median time to IF was 27 months. CONCLUSION: IF after lumbar spondylectomy occurs in roughly one in seven patients and is associated with multiple markers of procedural complexity. Although no single independent predictor was identified in the primary multivariable analysis, pelvic fixation demonstrated the strongest association across models, likely reflecting the increased biomechanical demands of lumbosacral constructs. Careful planning, reinforcement strategies, and long-term follow-up are critical to mitigate failure risk.