• BACKGROUND
    • Congenital kyphosis and kyphoscoliosis are much less common than congenital scoliosis. However, they are potentially more serious because compression of the spinal cord and paraplegia sometimes develop. The goals of the present study were to document the natural history of congenital kyphosis and kyphoscoliosis and to determine the stage at which the natural progression should be interrupted by treatment.
  • METHODS
    • We reviewed the medical records and radiographs of the spine of 112 consecutive patients. Sixty-eight patients had a type-I kyphosis due to an anterior failure of vertebral-body formation, twenty-four had a type-II kyphosis due to an anterior failure of vertebral-body segmentation, and twelve had a type-III kyphosis due to a combination of anomalies; the deformities of the remaining eight patients could not be classified. Eighty-five skeletally immature, untreated patients were first evaluated at a mean age of six years and nine months (range, two months to sixteen years and three months), and twenty-seven patients were skeletally mature at the time of the first visit. Sixty-three of the eighty-five skeletally immature patients were observed without any treatment for a mean period of six years and six months (range, one to sixteen years) before skeletal maturity, and the remaining twenty-two patients had a posterior arthrodesis of the spine soon after the initial visit. At skeletal maturity, forty-one patients had not been treated and sixty-eight had had an arthrodesis of the spine. The remaining three patients had not yet reached skeletal maturity at the time of the most recent follow-up.
  • RESULTS
    • The apex of the kyphosis was seen at all levels but was most frequent between the tenth thoracic and the first lumbar level (seventy-four patients; 66 percent). There was no relationship between the severity of the kyphosis and its location in the spine. Progression of the curve was most rapid during the adolescent growth spurt and stopped only at skeletal maturity. Progression was most rapid and the magnitude of the curve was the greatest in type-III kyphosis (twelve patients) followed by type-I kyphosis due to a posterolateral quadrant vertebra (thirty-nine patients), a posterior hemivertebra (eight patients), a butterfly vertebra (fifteen patients), and a wedged vertebra (six patients). A kyphosis due to two adjacent type-I vertebral anomalies progressed more rapidly and produced a more severe deformity than did a similar single anomaly. The prognosis for type-II kyphosis was variable and was much more severe when an anterolateral unsegmented bar had produced a kyphoscoliosis (nine patients) than it was when a midline anterior bar had produced a pure kyphosis (fifteen patients), which usually progressed slowly. Spontaneous neurological deterioration due to compression of the spinal cord occurred in ten patients (seven of whom had a type-I kyphosis and three of whom had an unclassifiable anomaly) at a mean age of thirteen years and eight months, and one other patient (with an unclassifiable anomaly) had spastic paraparesis at the age of twenty-eight years.
  • CONCLUSIONS
    • Congenital kyphosis and kyphoscoliosis are uncommon deformities with the potential to progress rapidly, resulting in severe deformity and possible neurological deficits. A thorough knowledge of the natural history is essential in the planning of appropriate and timely treatment to prevent progression of the deformity and neurological complications.