Complications and Radiographic Outcomes of Posterior Spinal Fusion and Observation in Patients Who Have Undergone Distraction-Based Treatment for Early Onset Scoliosis.

STUDY DESIGN: Retrospective, multicenter.

OBJECTIVES: To compare surgical and radiographic outcomes of early-onset scoliosis (EOS) patients who had stopped lengthening for ≥2 years without additional surgery to those who had posterior spinal fusion (PSF) at the end of lengthening.

SUMMARY OF BACKGROUND DATA: Because of the risk of significant complications with PSF in patients with EOS, "watchful waiting" at the end of lengthening has been suggested as a viable alternative.

METHODS: Retrospective review of the Children's Spine Study Group (CSSG) database identified all patients with the diagnosis of EOS who had distraction-based treatment, who were ≥2 years from their last distraction, and who had complete records. Radiographic measures were obtained by a single unbiased trained observer. Treatment outcomes including curve correction, height and length gain, as well as complications were recorded.

RESULTS: The 37 patients (21 females and 16 males) had a mean age of 7.2 years; 12 were in the observation (OBS) and 25 in the PSF group. The PSF group had a slightly greater coronal Cobb angle and maximal kyphosis at the end of distraction. Although there was some correction of the coronal Cobb angle and maximal kyphosis following PSF, the differences between the two groups were not statistically significant at final follow-up. At final follow-up, the OBS group obtained 88% of T1-T12 height and 90% of T1-L1 length of that obtained by the PSF group. Twenty-six complications occurred in 15 patients, all in the PSF group.

CONCLUSIONS: Observation may be a viable alternative to PSF after distraction-based treatment in a subset of patients with EOS. PSF was found to provide no significant curve correction or gains in spine height and length compared to observation and carries a significant risk of complications.

LEVEL OF EVIDENCE: Level III, therapeutic.