Comparison of cardiopulmonary function in patients with scoliosis at rest and during exercise: a case-control study

Preoperative cardiopulmonary assessment in scoliosis is typically dominated by resting spirometry and the Cobb angle, yet resting tests may understate functional impairment and curve magnitude predicts it only weakly. This study integrated resting echocardiography, static pulmonary function testing, CT lung volume, and cardiopulmonary exercise testing (CPET) in the same preoperative patients to characterize cardiopulmonary impairment and to separate the structural determinants of lung volume from those of exercise reserve.

In this single-center case-control study, 33 consecutive preoperative patients with scoliosis [21 adolescent idiopathic scoliosis (AIS), 12 congenital scoliosis (CS); aged 10-20 years] were compared with 20 age- and sex-matched healthy controls. Predicted spirometric values were derived from arm span with the GLI Global equations. Analyses used effect sizes with false-discovery-rate (FDR) correction, body-size-adjusted hierarchical regression, and cross-validated ROC analysis.

Resting cardiac function was preserved (left ventricular ejection fraction, cardiac output and cardiac index all non-significant after FDR), whereas restrictive ventilation and exercise reserve were reduced: FVC 72.0% vs. 92.8% predicted (q < 0.001), peak oxygen uptake 21.8 vs. 29.6 mL/min/kg (q < 0.001), O₂ pulse 5.9 vs. 9.7 mL/beat (q < 0.001), and VE/VCO₂ slope 32.4 vs. 27.5 (q < 0.001). Among 13 patients normal at rest, 10 (77%) were abnormal on CPET. Thoracic height independently predicted restrictive impairment beyond the Cobb angle, kyphosis, and body size (FVC %predicted B = 3.65 per cm, 95% CI   1.64-5.65; AUC 0.90, cross-validated 0.84; threshold 19.6 cm), whereas no deformity parameter independently predicted exercise capacity. At matched thoracic height, congenital scoliosis had lower CT lung volume than idiopathic scoliosis (1,881 vs. 2,356 cm³; p = 0.030).

Preoperative scoliosis is characterized by preserved resting indices but a frequently occult reduction in exercise reserve, including an exercise cardiac limitation revealed by O₂ pulse that resting echocardiography does not show. Thoracic height is a body-size-independent predictor of restrictive lung volume with a clinically usable threshold, while exercise reserve is multifactorial and requires direct CPET assessment. The two dimensions of preoperative cardiopulmonary risk should therefore be measured separately.