5-Year Reoperation Risk and Causes for Revision After Idiopathic Scoliosis Surgery.

STUDY DESIGN: An actuarial "survivorship" analysis.

OBJECTIVE: The aim of this study was to define the incidence and cause of surgical revision 5 years after scoliosis surgery.

SUMMARY OF BACKGROUND DATA: Data on contemporary revision surgery rates after idiopathic scoliosis surgery beyond the 2 years postoperatively in the adolescent and young adult population are limited.

METHODS: Patients enrolled in a prospective, multicenter, idiopathic scoliosis surgical registry from 1995 to 2009 were reviewed. Any spine reoperation was defined as a "terminal event." An actuarial survivorship analysis that adjusts for patients lost to follow-up was performed to determine cumulative survival. Time intervals were defined as 0 to <3 months, 3 months to <1 year, 1 to <2 years, 2 to <5 years, and 5 to 10 years. Registry data and radiographs were reviewed and five categories for reoperation assigned: 1) implant failure and/or pseudarthrosis, 2) implant misplacement and/or prominence, 3) wound complication and/or infection, 4) residual deformity and/or progression, and 5) other.

RESULTS: One thousand four hundred thirty-five patients from 12 sites were included. The majority were female (80%), with major thoracic curves (76% Lenke 1-4), and average age of 15 ± 2 years (10-22) at surgery. Most had posterior spinal instrumentation and fusion (81%). At this time, 75 (5.2%) patients required reoperation. Twenty-two occurred within 3 months postop, 10 more before 1 year, 12 more before 2 years, another 20 by 5 years, and 10 more after 5 years. This corresponded to an actuarial cumulative survival of 98.3% at 3 months, 97.5% at 1 year, 96.6% at 2 years, 93.9% at 5 years, and 89.8% at the final interval (5-10 yrs).

CONCLUSION: Revisions for scoliosis continue to occur well after 2 years with a 5-year survivorship of 93.9%. Reasons for reoperation are not uniformly distributed over time, with implant-related issues and infection the leading cause for early revision, while late infection was the most common cause after 2 years. Long-term follow-up of these postoperative patients remains important.

LEVEL OF EVIDENCE: 3.