Outcomes Following Removal of Instrumentation After Posterior Spinal Fusion

BACKGROUND:Although there is good evidence to support the removal of instrumentation for infection following posterior spine fusion, there are few studies that report outcomes following removal for late operative site pain. The purpose of this study was 3-fold(1) to determine whether removal of instrumentation following posterior spinal fusion resolves preoperative pain, (2) to determine whether indolent infection not detected before removal of instrumentation is related to late operative site pain, and (3) to determine whether curve progression differs when spinal hardware is removed for infection versus late operative site pain. METHODS:A retrospective study of consecutive patients aged 10 to 21 years, who underwent removal of instrumentation after posterior spinal fusion over a 10-year-period was conducted. Patient demographics, preoperative and postoperative imaging results, laboratory studies, and operative findings were reviewed. All patients had a minimum 2-year follow-up. Statistical analysis was performed using 2-sample t test, bivariate analysis, and multivariate logistic regression models. RESULTS:Seventy-five patients were included. Indications for removal of spinal instrumentation were pain (57%), infection (28%), hardware failure (8%), and prominent hardware (7%). The mean time from index procedure to hardware removal was 2.8 years. The average loss of curve correction following complete hardware removal was 23.1 degrees. Patients who underwent removal of hardware because of infection had bigger changes in their curves than those without infection (mean, 33.8 degrees vs. 18.8 degrees). Of the 43 patients with pain, only 40% reported relief of their symptoms following removal of hardware. Sixteen of the 43 patients were found to have indolent infection confirmed by positive intraoperative culture results. CONCLUSIONS:Patients should be cautioned that hardware removal after posterior spinal fusion may not provide complete pain relief. Furthermore, there is risk for curve progression following removal of instrumentation, particularly in the setting of infection. Back pain may be an indicator of infection, and intraoperative cultures should be taken at the time of implant removal. LEVEL OF EVIDENCE:Level IV; retrospective case series.