The risk of progression of scoliosis in cerebral palsy patients after intrathecal baclofen therapy

STUDY DESIGN. Retrospective radiographic and medical chart review with matched control group. OBJECTIVE. To identify the effect of intrathecal baclofen on the incidence of scoliosis, rate of curve progression, and pelvic obliquity compared with a matched cohort. SUMMARY OF BACKGROUND DATA. Although intrathecal baclofen therapy (ITB) has been shown to be effective in decreasing spasticity, case reports have described some children receiving ITB in whom progressive scoliosis was noted; other authors have described no effect on the spinal column. A controlled study has not been performed. METHODS. All patients with spastic CP treated with ITB between 1997 and 2003 at a single institution were reviewed. A total of 107 patients undergoing ITB for a minimum of 2 years were identified, of which 26 patients subsequently developed or had progression of scoliosis. Twenty-five age, gender, and gross motor function classification system (GMFCS) score-matched quadriplegic CP patients with scoliosis who did not receive ITB constituted the control group used to compare the rate of curve progression and pelvic obliquity. RESULTS. The average curve progression for the baclofen group after pump implantation was 16.3° per year; and for the control group was 16.1° per year. Both groups' curves progressed over time during growth (P = 0.001), but baclofen did not have an independent effect on curve progression (P = 0.181). Average pelvic obliquity for the 2 groups increased over time (P = 0.001), but there was no difference between the groups (P = 0.536). Twelve of 57 patients (21%) developed scoliosis after pump implantation during a mean of 3.6 years of follow-up. Thirty of 92 matched control patients (32%) not treated with ITB within the same time interval had scoliosis by maturity. CONCLUSION. This study demonstrates that ITB has no significant effect on curve progression, pelvic obliquity, or the incidence of scoliosis when compared with an age, gender, and GMFCS score-matched control group of patients with spastic CP without ITB.