Ask anyone who treats patients with sleep apnea what they consider the biggest challenge in their management and most will agree: convincing patients to regularly use positive airway pressure (PAP) therapy. Despite our best efforts to improve PAP tolerability, a significant proportion of patients remains unable to use it. Estimates of adherence rates range from 30% to 70% across studies.
The February 2014 issue of CHEST Physician highlighted a recent study exploring a novel treatment alternative for certain patients with sleep apnea. Investigators in the Stimulation Therapy for Apnea Reduction (STAR) trial enrolled a select group of sleep apnea patients with difficulty accepting or adhering to PAP therapy and surgically implanted a hypoglossal nerve stimulator designed to open the upper airway (Strollo et al. N. Engl. J. Med. 2014;370[2]:139).
Dr. Lauren Tobias
The study was sponsored by the manufacturer, which recently secured U.S. Federal Drug Administration (FDA) approval for the device and is expected to make it commercially available later this year.
The principle behind this therapy is straightforward. It is known that sleep apnea patients lose tone in many of the upper airway muscles that normally maintain a patent airway. The genioglossus muscle is the largest upper airway dilator muscle, and its contraction results in protrusion of the tongue and stiffening of the anterior pharyngeal wall. Augmenting stimulation of this muscle may, therefore, increase airway patency, making it a reasonable therapeutic target for obstructive sleep apnea (OSA).
How the device works
The stimulator sits in a pocket under the skin in the right mid-infraclavicular area. It connects to two tunneled leads: a sensing lead extending to the intercostal muscles to detect ventilation efforts and a stimulating lead extending superiorly to the hypoglossal nerve under the tongue. Stimulation of the nerve moves the tongue forward slightly, opening the upper airway. Patients are able to turn the device on before sleep and off upon awakening with a handheld remote control.
Their results
The trial’s results were impressive. Among a total of 126 patients enrolled, the median apnea-hypopnea index (AHI) at 12 months decreased by 68%, from 29 to 9 events per hour. The oxygen desaturation index (ODI) also decreased significantly, from 25 to 7 events per hour. Using the definition of success often employed in the surgical literature, namely a reduction in AHI by at least 50% to a residual level less than 20 per hour, they found that two-thirds of patients were "responders" to the therapy. In order to prove that the device was responsible for this improvement rather than changes in the patients’ disease over time, a randomly selected group of responders had their devices turned off for 1 week, allowing these subjects to serve as their own longitudinal controls. Indeed, patients’ disease worsened significantly with the devices deactivated, with AHIs rising to pre-implantation levels. The device use also appeared to result in modest improvements in both sleepiness and sleep-related quality of life.
Although adverse effects were reported as minor, they occurred fairly often. Forty percent of participants reported discomfort when the device was active, and 21% reported tongue soreness, likely exacerbated by movement of the tongue across the teeth during periods of stimulation. Eighteen percent of subjects reported tongue weakness after surgery, resolving over a period of days to weeks. Two patients experienced discomfort severe enough to necessitate repositioning and fixation of the device, and one elected to have the device removed.
A closer look
The major limitations of this study stem from its participant selection. Patients were overwhelmingly male (83%), averaging 55 years of age. They were significantly leaner than typical patients with OSA, with a mean BMI of 28.4 kg/m2. It seems reasonable to assume that more obese patients would have less of a response to nerve stimulation, since excessive laryngeal soft tissue is less likely to be the main contributor to their obstruction; moving the tongue forward should not significantly affect their pathophysiology. The authors admit as much, stating that their exclusion of patients with a BMI over 32 was based on feasibility studies showing that such patients were less favorable candidates for the procedure.
Dr. Christine Won
Furthermore, of the nearly 1,000 patients initially recruited for this study, about half were excluded for having AHIs outside the study’s defined range (AHI 20-50). An additional 5% each were excluded for having concurrent central apnea (at least 25% of episodes), positional OSA (nonsupine AHI less than 10), or complete concentric collapse of their palate during sleep endoscopy. Based on the exclusion of the majority of screened subjects, we should be cautious in generalizing the results of the trial to our own clinical practices. Careful patient selection will be key to ensuring comparable success in real-world patients.