Scoliosis and chest cage deformity measures predicting impairments in pulmonary function: a cross-sectional study of 492 patients with scoliosis to improve the early identification of patients at risk.

Abstract

This study examines the correlations between pulmonary function tests (PFTs) and radiographical measures of spinal deformities in patients with scoliosis. To define the parameters that enable more accurate predictions of restricted pulmonary function (PF) in patients with scoliosis. The early identification of patients with scoliosis who are at risk of developing severe curve progression and restricted PF is critical to improving patient care. A total of 492 patients with a thoracic/thoracolumbar curve (TC) and preoperative PFTs as well as radiographical analysis of the TC and thoracic kyphosis (TK) were assessed. The forced vital capacity (FVC) was expressed as a percentage of the predicted value (FVC%). According to guidelines for the severity of pulmonary impairments, classifications were used: no impairment (FVC > 80%), mild (65 < FVC ≤ 80%), moderate (50 < FVC ≤ 65), and severe impairment (FVC ≤ 50%). Main radiographical parameters were included and statistical analyses were used to identify radiographical predictors and to develop prediction models. The average age was 17 years, 94% of the patients had adolescent idiopathic scoliosis, 20% had lordoscoliosis (TK <10°). The average FVC% was 75% ± 17%, their average TC was 57° ± 21°, and TK was 26° ± 25°. A total of 38% had no impairment, 37% exhibited mild impairment, 15% had moderate impairments, and 10% were severely impaired. Spinal deformities identified by radiographical analysis (TC, TC-flexibility, apical vertebral rotation, TK, and rib hump) significantly influenced FVC% (P < 0.01). TK logistic regression analysis revealed that 2 factors are predictive for an FVC 50% or less: increased scoliosis with a low TK less than 0° (odds ratio = 7.3) or a high TK more than 60° (odds ratio = 20). A model for predicting FVC values more than 50% was established and shown to achieve a correct classification rate of 94%. Patients with increasing coronal and sagittal plane deformities with a high thoracic scoliosis apex are at the highest risk for reduced FVC%. The models developed provided improved estimations of actual PF based on the magnitude of the radiographical deformity.