Does REM AHI Predict Persistent OSA After Pediatric Adenotonsillectomy?

OBJECTIVE: The utility of REM AHI in managing pediatric obstructive sleep apnea (OSA) is not fully understood. This study aimed to evaluate the relationship of preoperative REM AHI to postoperative persistence of OSA in children who underwent adenotonsillectomy. METHODS: This retrospective chart review identified children under the age of 18 years that received an adenotonsillectomy for OSA and a preoperative and postoperative polysomnogram. Children with craniofacial or neuromuscular disorders or a tracheostomy were excluded. The primary outcome was the postoperative persistence of OSA, defined as a postoperative obstructive apnea-hypopnea index (oAHI) ≥ 1.5 events/hour. REM-predominant OSA was defined as a ratio of REM/NREM AHI ≥ 2. REM AHI minus NREM AHI and REM AHI minus oAHI helped to identify patients with a larger distribution of REM AHI. RESULTS: A total of 353 patients were included. Postoperative persistent OSA was seen in 232 (65.7%) children. The preoperative REM AHI, REM AHI minus NREM AHI, and REM AHI minus oAHI of children with persistent OSA did not differ significantly from children with resolution of OSA. Rates of persistence were not different between those with REM-predominant OSA and REM-independent OSA (63.8% vs 70.7%, P = .218). CONCLUSION: This study suggests that preoperative REM AHI may be a poor predictor of OSA persistence after adenotonsillectomy. Further study is needed to help characterize how pre-operative REM AHI should impact clinicians' decision making, family counseling and recommendations.

Ventricular assist devices and sleep-disordered breathing-A mechanical heart stimulating a sleepy brain.

Sleep-disordered breathing, including obstructive sleep apnea (OSA) and central sleep apnea (CSA), is common in severe heart failure (HF) patients. There is limited data on the effect of left ventricular assist devices (LVAD) on sleep apnea. We performed a retrospective review of 350 durable LVAD patients and found 5 with a history of pre- and post-LVAD sleep studies. All five patients had OSA, and three had concomitant CSA. We observed reduced apnea-hypopnea index following LVAD placement. This was due to a near abolishment of CSA in three mixed sleep apnea patients-as seen by a central apnea index improvement from an average of 25.9 ± 13.1 to 1.4 ± 2.5 events per hour (p = 0.063). LVAD placement was associated with an increase in thermodilution cardiac output from 2.7 ± 0.6 to 4.1 ± 1.1 L/min (p = 0.014). These findings support chemoreception physiology seen in patients with poor circulation and the effect of restoring this circulation with LVAD support.