Sleep architecture parameters that predict postoperative resolution of nocturnal enuresis in children with obstructive sleep apnea.

OBJECTIVES

We performed a prospective cohort study in a pediatric tertiary care center to determine whether preoperative sleep architecture is associated with complete resolution of nocturnal enuresis (NE) after adenotonsillectomy.

METHODS

Thirty-seven pediatric patients with primary NE who underwent adenotonsillectomy for obstructive sleep apnea (OSA) were evaluated. Preoperative polysomnograms, as well as preoperative and postoperative reports of NE, were recorded. We performed chi2 analysis, Fisher's exact test (for p values), and t-tests to evaluate the impact of multiple demographic characteristics on sleep architecture, comparing children with resolved NE to those with unresolved NE after adenotonsillectomy.

RESULTS

The patients' mean age was 8.0 years (SD, 2.32 years). All children had presurgical primary NE. No age or gender differences were identified between children with resolved NE and those with unresolved NE. After surgery, more than half of the participants had resolution of NE. A higher percentage of boys had unresolved NE (chi2 = 3.63; p = 0.06). Improvement of NE was identified in children with a higher obstructive apnea-hypopnea index and more desaturation events. Eleven of the 12 children with prolonged stage 2 sleep reported resolution of NE (p = 0.001). Children with an obstructive apnea-hypopnea index of greater than 10 had a significantly greater rate of resolution of NE (p = 0.01). Logistic regression demonstrated that an elevated body mass index and the interaction of severe OSA and prolonged stage 2 sleep predicted resolution of NE. All 10 children with severe OSA and an abnormal total time spent in stage 2 sleep had resolution of NE.

CONCLUSIONS

Adenotonsillectomy is a treatment option for children with OSA and NE. Postoperative resolution of NE was seen in 51.4% of patients who underwent adenotonsillectomy. The children with both severe OSA and prolonged stage 2 sleep were 3.4 times as likely to have postoperative resolution of NE. These results suggest that there are significant differences in preoperative sleep architecture between children whose NE resolves after adenotonsillectomy and those whose NE does not resolve.