BackgroundLong-term outcomes of patients undergoing extensive fusions for adolescent idiopathic scoliosis (AIS) have conflicting results. Moskowitz found uninstrumented scoliosis fusion patients had similar back pain as a normal age matched population. Recently, long-term outcomes of patients with Harrington rod instrumentation were reviewed and found similar functional outcome scores to non-scoliosis patients, with trending toward worse outcomes when fusions extended to L4. Our study examined long-term functional outcomes of patients treated with Cotrel-Dubousset (CD) instrumentation and determined whether distal level of instrumented fusion (L4 and L5) correlate with increased back pain or lower functional level.MethodsRetrospective review of AIS surgeries from 1986 to 1996 was undertaken. Patient demographics and surgical data were collected via case-note audit. Patients were contacted and asked to complete a series of functional outcome questionnaires including visual analog scales (VAS) for pain, Short-Form 36 (SF-36), Scoliosis Research Society 22 (SRS-22) and Oswestry Disability Index (ODI) for function. ANOVA technique categorically compared outcome scores to most distal levels of fusion. Linear regression compared patient reported outcomes to time elapsed since surgery. Statistical significance was p < 0.05.ResultsOne hundred twelve patients were identified, 50 patients were contacted, and 22 agreed to participation and completed a full assessment. Follow-up time since surgery ranged from 15 to 26 years and age ranged from 30 to 43 years. Six patients reported daily VAS back pain of ≥5; with a mean of 2.5. Back pain was not associated with level of distal fusion (p = 0.92). ODI was 15.36, with six patients' ODI >20. No relationship was shown between ODI and distal level of fusion (p = 0.72). SF-36 and SRS 22 values were also not related to distal level of instrumentation. Patient reported VAS back pain scores (r2 = 0.18, p = 0.05), ODI (r2 = 0.09, p = 0.17), and SF-36 and SRS-22 were not worse in patients with longer follow-up over time. Back pain and certain functional score subcategories of the SF-36 and SRS-22 trended toward improved results over time.ConclusionsMost patients who underwent multi-segment spinal fixation appeared to do well long-term, with minimal back pain. Lowest instrumented segment did not appear to be associated with increased back pain after 15 to 25 years follow-up.