Congenital scoliosis due to a unilateral failure of vertebral segmentation usually has a poor prognosis. However, not all curves progress to the same degree or develop the same spinal deformity.Methods:
The medical records and spine radiographs of 171 patients with a scoliosis due to unilateral unsegmented bar were reviewed retrospectively. The mean patient age at diagnosis was 9.3 years. Thirty-six patients had immediate surgery, 103 immature patients were followed untreated or before surgery for a mean duration of 3.6 years, twenty-eight patients were seen untreated at skeletal maturity, and four patients had no follow-up.Results:
The unsegmented bar occurred at all levels; the mean extent was three vertebrae (range, two to eight vertebrae). Before the age of ten years, patients had a mean rate of scoliosis progression without treatment for all regions of the spine of 2° to 3° per year. By the age of ten years, seventy-three patients who had been seen untreated or prior to treatment had a mean scoliosis of 50° (range, 18° to 100°). After the age of ten years, these patients had an increase in the mean rate of scoliosis progression, but this rate varied per year depending on the affected region of the spine: 7° for the thoracolumbar curve, 5° for the midthoracic curve, and 4° for the upper thoracic curve. Spine surgery was performed on seventy-four patients who had a mean age of 12.2 years and a mean scoliosis of 78° for thoracolumbar curves, 66° for midthoracic curves, and 54° for upper thoracic curves. Of the patients with midthoracic congenital scoliosis, 24% developed a lower structural compensatory curve, which became larger and more deforming than the congenital curve, 22% had congenital rib fusions, and 16% had occult intraspinal anomalies.Conclusions:
Prognosis depends on growth imbalance at the site of the unsegmented bar as well as the location and extent of the bar, age at diagnosis, and spinal growth remaining.Level of Evidence:
Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence.