Obstruction level associated with outcome in hypoglossal nerve stimulation

PURPOSE: Selective hypoglossal nerve stimulation (sHNS) constitutes an effective surgical alternative for patients with obstructive sleep apnea (OSA). sHNS results in tongue protrusion and consecutive alleviation of obstructions at the tongue base level (lower obstructions). Furthermore, obstructions at the soft palate level (upper obstructions) may be prevented through palatoglossal coupling as seen on sleep endoscopy. However, it has not been studied if the distribution of obstruction level during a whole night measurement is a relevant factor for the treatment outcome. METHODS: Obstruction levels were measured with a manometry system during a whole night of sleep in 26 patients with OSA (f = 1, m = 25; age 59.4 ± 11.3; BMI = 29.6 ± 3.6) either before (n = 9) or after sHNS implantation (n = 12). Five patients received a measurement before and after implantation. Obstructions were categorized into velar (soft palate and above), infravelar (below soft palate), and multilevel obstructions. An association between obstruction level and treatment outcome was calculated. RESULTS: The mean distribution of preoperative obstruction level could be divided into the following: 38% velar, 46% multilevel, and 16% infravelar obstructions. Patients with a good treatment response (defined as AHI < 15/h and AHI reduction of 50%) had fewer preoperative velar obstructions compared to non-responder (17% vs. 54%, p-value = 0.006). In patients measured after sHNS implantation, a significantly higher rate of multilevel obstructions per hour was measured in non-responders (p-value = 0.012). CONCLUSIONS: Selective hypoglossal nerve stimulation was more effective in patients with fewer obstructions at the soft palate level. Manometry may be a complementary diagnostic procedure for the selection of patients for HNS.