Background: Convex hemiepiphysiodesis has been reported to have mixed results in the treatment of congenital spine deformities. Multiple modifications of the original technique were suggested to improve the results. The purpose of this study is to report the results and complications of an instrumented convex growth arrest procedure modified with concave distraction.
Methods: The records of 11 patients with long sweeping congenital curves (involving multiple anomalous vertebrae) who underwent convex instrumented hemiepiphysiodesis with concave distraction were evaluated. Mean age at index operation was 58 months (range, 29 to 101 mo). Lengthening of the concave distraction rod was done every 6 months. The magnitude of coronal/sagittal deformity and T1-T12 height were measured on the preoperative, postoperative, and latest follow-up radiographs. Average follow-up was 44.9 months (range, 24 to 89 mo).
Results: In the coronal plane, the convex hemiepiphysiodesis segment was corrected from an average of 60.5 to 40.4 degrees postoperatively and further improved to 35.5 degrees at the latest follow-up. The distracted segment was corrected from 33.4 to 15.2 degrees postoperatively and to 12.7 degrees at the latest follow-up. Sagittal plane alignment was minimally affected. The average T1-T12 height was 157.1 mm in the early postoperative period and 181.1 mm at last follow-up. During follow-up, we identified partial pull-out of screws on the distraction side in 5 of the 11 patients and rod breakages in 3 patients. These were revised during planned lengthenings. There were no unplanned surgeries, deep wound infections, nor neurological complications.
Conclusions: Convex instrumented hemiepiphysiodesis with concave distraction resulted in good curve correction while maintaining the growth of thorax. The correction of the anomalous segment improved over time, proving the effectiveness of the hemiepiphysiodesis. Addition of a concave distraction construct appears to enhance spinal growth, thereby augmenting the hemiepiphysiodesis effect.