Surgical treatment of obstructive sleep apnea in neurologically compromised patients.

Children with cerebral palsy are at risk of developing obstructive sleep apnea, which is initially managed by medical therapy but often requires tracheostomy for stabilization of the airway. We report preoperative and postoperative polysomnographic findings in a prospective series of 18 patients with cerebral palsy and obstructive sleep apnea who were refractory to medical management and underwent aggressive surgical treatment of upper airway obstruction. Fifteen of the 18 children (83 percent) in whom tracheostomy was recommended were spared the procedure. Eighteen children with cerebral palsy failed medical management of obstructive sleep apnea and were advised to have tracheostomy. There were 9 boys and 9 girls, ranging in age from 9 months to 17 years and 6 months at the time of operation. Tonsillectomy and adenoidectomy was performed in 9 patients, turbinectomy and/or septoplasty in 9, tongue-hyoid advancement in 13, uvulopalatoplasty in 13, conventional mandibular advancement in 2, distraction osteogenesis of the mandible in 2, and tongue reduction in 7. A concomitant Wilkes-Brody procedure for drooling was performed in 6 patients. Preoperative and postoperative polysomnographic data were compared by means of a paired t test. The mean preoperative apnea index, respiratory disturbance index, and lowest oxygen saturation were 3.61, 7.02, and 73.7, respectively. Mean postoperative apnea index, respiratory disturbance index, and lowest oxygen saturation were 0.67, 1.44, and 88.2, respectively. Lowest oxygen saturation and respiratory disturbance index were both improved significantly, with p values of 0.0367 and 0.0021, respectively. Fifteen patients are tracheostomy-free (83 percent) at a mean follow-up time of 30 months (range 14 to 49 months.) Two (11 percent) of the children ultimately required tracheostomy, and one (6 percent) died from respiratory failure following the parents' decision not to proceed with further treatment. Our results confirm the efficacy of an aggressive surgical approach to the treatment of obstructive sleep apnea in neurologically compromised children. Many children and their families may potentially avoid the long-term commitment and cumulative hazards of tracheostomy. Additional strategies that have been adopted include identification and aggressive management of seizures, esophageal reflux, and excessive oral secretions and the application of mandibular distraction and skeletal expansion whenever feasible. Close postoperative monitoring is necessary with reoperation for recurrent symptoms of obstructive sleep apnea if documented by sleep study and associated with evidence of recurrent or residual morphologic abnormalities.