Study design: Prospective radiographic and chart review of pulmonary function in patients who underwent 3 different anterior spinal surgery approaches for adolescent idiopathic scoliosis (AIS).
Objective: To assess the impact on pulmonary function in patients with AIS after anterior surgical approaches, including open thoracotomy, thoracoscopic with and without thoracoplasty, and thoracoabdominal 2 years after surgery.
Summary of background data: Potential advantages of anterior surgery in the treatment of AIS include saving of distal motion segments and improving kyphosis restoration in the thoracic spine, possibly at the cost of pulmonary function impairment. Although thoracoscopic spinal instrumentation and fusion has recently been shown to induce less pulmonary impairment compared with open thoracotomy, no study has evaluated the effect of thoracoplasty as an adjunct to thoracoscopic surgery, nor the effects of the thoracoabdominal approach for thoracolumbar curvature.
Methods: A multicenter spinal deformity database was queried for patients who underwent an anterior approach for either Lenke I or Lenke V idiopathic scoliosis. There were 68 patients in the thoracotomy group, 44 in the thoracoscopic group, and 19 in the thoracoabdominal group. Absolute and percent-predicted values of forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), and total lung capacity (TLC) were evaluated preoperatively and at 2-year follow-up, and comparisons were made within and between each group.
Results: Comparing between groups at 2 years postoperatively, the thoracotomy group demonstrated significantly greater decreases (-10.97% and -12.97%) in both percent-predicted FEV1 and FVC, respectively, when compared with the thoracoscopic group (-4.40% and -4.73%), respectively. Percent-predicted TLC in the thoracoscopic group increased (3.19%), but decreased in the thoracotomy group (-8.00%). Subanalysis of the thoracoscopic group at 2 years revealed that the addition of a thoracoplasty (3 to 5 ribs) significantly reduced percent-predicted FEV1 (-11.6%, P = 0.0013) and percent-predicted FVC (-16.0%, P = 0.017) from baseline. Patients who underwent thoracoscopy alone without a thoracoplasty experienced no significant detrimental changes in these parameters at 2 years, and instead experienced significant increases in absolute TLC (P < 0.001) and percent-predicted TLC (P = 0.035). There were no significant changes demonstrated in the thoracoabdominal group for all 3 parameters.
Conclusions: Slight declines in pulmonary function at 2-year follow-up were noted in both the thoracotomy and thoracoscopic groups, but to a significantly greater extent in those with an open thoracotomy. Significantly greater impairment in pulmonary function was seen in thoracoscopic patients who underwent thoracoplasty, whereas those without thoracoplasty either had no significant declines at 2 years or experienced slight but significant improvement. No significant diminishment was noted for the thoracoabdominal approach, despite disruption of the diaphragm.