Predictors of flexibility and pain patterns in thoracolumbar and lumbar idiopathic scoliosis.

STUDY DESIGN

A retrospective evaluation of radiographs in patients with idiopathic scoliosis was undertaken to assess predictors of flexibility.

OBJECTIVE

To evaluate potential predictors of flexibility in patients with thoracolumbar and lumbar scoliosis.

SUMMARY OF BACKGROUND DATA

Curve flexibility is an important consideration in the operative management of idiopathic scoliosis. Flexibility of the major curve is a useful predictor of expected surgical correction, and flexibility of compensatory curves determines whether they are structural or nonstructural. An accurate assessment of curve flexibility has important implications on surgical approaches and planning for deformity correction. The role of age and curve magnitude in predicting curve flexibility has not been well defined. A quantitative assessment of changes in curve flexibility with age and progression of deformity may yield important insight into the change in surgical management options over time.

METHODS

A retrospective review of 75 patients with idiopathic thoracolumbar and lumbar scoliosis (age range 13-78 years) was undertaken. Preoperative standing and side-bending radiographs of thoracolumbar and lumbar curves were evaluated. Cobb angles of structural and fractional curves, curve flexibility, presence of lateral listhesis, and axial and radicular pain were documented. Predictors of structural and fractional curve flexibility were evaluated with correlation and regression analysis. Correlation analysis was used to demonstrate an association between radiographic findings and the clinical presentation.

RESULTS

Seventy-five patients had an average major curve magnitude of 56 degrees (range 34-82 degrees ) with flexibility averaging 55% (range 20-93%). Structural curve flexibility was highly inversely correlated with both curve magnitude (r = -0.7; P< 0.001) and with age (r = -0.6; P< 0.001). Lumbar fractional curve (L4-S1) flexibility showed a high inverse correlation with age (r = -0.65; P< 0.001) but did not show correlation with Cobb angle. Thoracic compensatory curves showed a moderate correlation with Cobb angle (r = 0.53). Structural and fractional curve flexibility showed high correlation with each other (r = 0.5-0.66). Regression analysis yielded a formula to predict the flexibility of the structural curve (FSC): FSC = 130 - (Cobb + Age/2). Axial pain was correlated with age (r = 0.63); however, it was not correlated with curve magnitude.

CONCLUSION

We have shown that curve magnitude and patient age are the main predictors of structural flexibility. Every 10 degrees increase in curve magnitude over 40 degrees results in a 10% decrease in flexibility; every 10-year increase in age decreases flexibility of the structural curve by 5% and the lumbosacral fractional curve by 10%. Curve magnitude and age of the patients are significant predictors of curve flexibility. The demonstration of this association offers useful information in estimating how surgical options for deformity correction may change over time.