Obstructive sleep apnea after dynamic sphincter pharyngoplasty.

INTRODUCTION

In patients who require additional surgery for velopharyngeal insufficiency (VPI), a higher incidence of obstructive sleep apnea (OSA) may be incurred. Although this phenomenon has been demonstrated with the posterior pharyngeal flap, the effect of dynamic sphincter pharyngoplasty (DSP) on OSA is less clear. The purposes of this case series were to (1) determine the incidence of OSA after DSP, (2) assess the changes in polysomnography after DSP, and (3) identify risk factors for the development of OSA after DSP. Our global hypothesis is that OSA and VPI exist on a continuum and that speech outcomes should not be considered in isolation.

METHODS

For a 13-year period, 146 patients with idiopathic VPI, submucous cleft palate, cleft palate only, or cleft lip and palate underwent DSP for VPI. The diagnosis of OSA was defined as the prescription of continuous positive airway pressure therapy by a pediatric sleep medicine physician. The incidence of OSA preoperatively and postoperatively was compared using Fisher exact test. When available, preoperative and postoperative apnea-hypopnea indices (AHIs) were compared using the pairwise, 2-tailed, Student's t-test. Patient factors, such as obesity (body mass index ≥ 95th percentile), the presence of a craniofacial syndrome, surgical history, and a preexisting OSA diagnosis, were noted. A multiple logistic regression was performed to elucidate risk factors for the development of OSA.

RESULTS

The average age at surgery was 9.2 years (range, 4-40 y), and the mean follow-up time was 4.5 years (range, 1 mo to 12 y). The incidence of OSA increased after DSP, from 2 to 33 patients (1.4%-22%, respectively; P = 0.05). In 23 patients (16%), both preoperative and postoperative AHIs were available. There was a significant increase in AHI after DSP, from 3.1 to 8.4 episodes per hour of sleep (P = 0.001). Previous tonsillectomy/adenoidectomy was predictive of OSA after DSP (relative risk = 2.4; P = 0.04).

CONCLUSIONS

We report an increased incidence of OSA and higher-than-average AHIs postoperatively after DSP. Preoperative tonsillectomy/adenoidectomy predicted the development of OSA after DSP. A high index of suspicion for development of OSA must be maintained in patients who undergo secondary speech operations for VPI. Clinical screening for OSA should be used in this population, with a low threshold for polysomnographic evaluation. The surgeon must be wary that improvements in speech after DSP may change airway dynamics and increase the risk of OSA.