Abstract
Surgical treatment of sleep-disordered breathing (SDB) should be targeted at the suspected level of obstruction. 1 Nasal surgery is helpful in only a small percentage of patients. Palatal surgery such as uvulopalatopharyngoplasty can be effective, although the long-term results are not impressive, with less than 50% of patients responding to treatment. Physiologic studies 2 , 3 have shown the tongue base and hypopharynx to be the major sites of obstruction in up to 50% of patients with obstructive sleep apnea (OSA), possibly explaining the cause of failure in palatal surgery.
Surgical treatments of tongue-base and hypopharyngeal collapse include midline glossectomy, lingual plasty, hyoid suspension, mandibular osteotomy with genio-glossal advancement, and maxillary-mandibular advancement. Many of these procedures are associated with extreme morbidity and change in facial anatomy and may require a perioperative tracheotomy.
The Repose (Influence Inc, San Francisco, CA) is a new minimally invasive surgical kit that uses a bone–to–soft tissue anchor to stabilize the tongue base. Studies in an animal model and initial clinical results have been reported. 4 The purpose of this multicenter study was to evaluate the safety and initial clinical impact of this tongue-stabilization technique in patients with SDB.
METHODS AND MATERIAL
Patient Population
Sixteen patients (1 woman, 15 men) with SDB manifested by OSA were enrolled in this phase 1 study. Data were collected retrospectively as a case series with preoperative and postoperative comparison. Ages ranged from 35 to 74 years. The following data were collected before surgery: history and physical examination including Müller's maneuver. Snoring was quantified with a subjective scale based on bed partner observation. Complete preoperative and postoperative polysomnography (PSG) was performed and analyzed by sleep physicians. Institutional review board approval was attained from each participating institution, and informed consent was obtained from each patient before enrollment. The study included OSA patients with a respiratory distress index (RDI) of less than 70 events per hour. Patients with an RDI of less than 12.5 events per hour were not considered OSA patients and were excluded from this initial analysis. Patients with severe obesity, a body mass index of more than 35 kg/m 2 , were also excluded from the study, as were patients with previous radiation treatment to the head and neck or those in whom a general anesthetic was contraindicated. Tongue-base suspension was the only procedure performed in this group of patients.
70152-5-fig1.png)
Schema of Repose procedure: 1, insertion location, sagittal view; 2, passing of sutures to tongue base; 3, submucosal pass at tongue base with Mayo needle (pull-through loop in place); and 4, knot tie at mouth floor.
Device
The Repose is a presterilized disposable surgical kit that includes (1) a self-tapping screw with preattached double polypropylene No. 1 sutures, (2) a battery-operated screwdriver with a U-shaped end to allow insertion in the direction of the hand grip, and (3) a suture passer that facilitates passage of the suture through the tongue.
Procedure
The procedure was performed as described in the instructions for use. The patient is placed in the tonsillectomy position under general anesthesia with nasotracheal intubation. The screw inserter is placed so that the screw is inserted through the floor of the mouth in the midline, posterior to the Wharton's duct orifices. Once inserted, the screw is pressed against the mandible, below the roots of the teeth. The inserter is activated until the screw completely penetrates the mandible. The sutures now protrude through the entrance hole. The suture is then passed through the penetration hole at the mouth floor to an exit hole at the tongue base, about 1 to 1.5 cm lateral to the midline, with a special suture passer. An empty suture loop is passed in a similar manner exactly opposite the anchored suture. Care is taken not to pass the needle too laterally to avoid injury to the neurovascular bundle. The screw's suture protruding into the pharyngeal space is threaded onto a Mayo needle. The needle is passed into the exit point of the suture and then passed submucosally past the midline to the new exit hole, within the empty loop suture. As a result, the suture is now threaded onto the loop. The suture loop is pulled back to the anterior intraoral space, thus achieving a triangular suture pass, the base of which is in the posterior tongue base, which in turn is anchored to the mandible. At this point, the two sutures' ends are protruding from the same exit hole in the anterior mouth space. The two anchor suture ends are tied together below the tongue mucosa and are buried in the mouth floor, resulting in support of the tongue base. Tightness of the suspension was determined by digitally feeling tightness at the tongue base where an indentation could be felt. Overcorrection was avoided to prevent tissue strangulation, pain, and swelling. Figure 1 illustrates the steps of the surgery.
Patient Follow-up
Patients were evaluated clinically after surgery and at 1 week, 1 month, 2 months, and 6 months. Postoperative polysomnography and snoring observational questionnaires were obtained 2 to 3 months after surgery.
Preoperative and postoperative data from 14 patients
70152-5-table1.png)
Postoperative data were collected 2 to 3 months after surgery.
Data Analysis
At 2 to 3 months after surgery, 2 main end points were analyzed: RDI and subjective account of snoring by bed partner. The data were tabulated in an Excel (5.0) worksheet. Statistical analysis for RDI was performed with the 2-tailed t test.
RESULTS
Of the 16 patients, 2 had short-term complications of local infection and pain. In 1 of these patients there was a minor hematoma at the mouth floor. Both patients had the procedure reversed by removal of the suture with resolution of the infection and return to baseline. These 2 patients were excluded from the analysis. In the remaining 14 patients, pain, odynophagia, and speech difficulties persisted for 1 to 2 weeks and resolved without sequela. In 2 patients pain subsided after 4 weeks. No case of postoperative airway obstruction was encountered. In 1 patient 1 month after surgery a mouth floor cyst developed, probably because of sublingual gland obstruction. The cyst was marsupialized without further recurrence.
After surgery, snoring improvement was noted by bed partners in all 14 patients. Improvement was of various degrees. A validated questionnaire was not used, so the observations were not quantified. Table 1 shows the preoperative and postoperative snoring results.
Preoperative and postoperative RDIs are presented in Table 1. The mean preoperative RDI was 35 ± 16.5. The mean postoperative RDI at 2 to 3 months was 17 ± 8. This represents an improvement of 51.4% in RDI after surgery. This decrease was statistically significant (P = 0.001).
DISCUSSION
No specific criteria exist regarding the diagnosis of tongue-base collapse in SDB. At best, diagnosis is an educated guess. However, surgery targeted at the tongue base has shown various rates of success with even better results when combined with palatal and nasal surgery. 1 , 5 In this series of patients the clinical impression was of tongue-base collapse during Müller's maneuver or impression of a wide oropharynx without a redundant palate. This impression cannot be well quantified but probably suggests a patient selection bias.
Surgery targeted to alleviate tongue-base collapse in SDB is typically invasive with associated morbidity. 5 The Repose is a minimally invasive surgical procedure that anchors the tongue base to the anterior mandible with the purpose of stabilizing the tongue base during sleep. In this preliminary study a positive effect in sleep parameters was observed in short-term follow-up. A subjective improvement in snoring and a statistically significant decrease of 53% in mean RDI are similar to results of uvulopalatopharyngoplasty. 6 Patient tolerance of the procedure was good except in 2 patients, in whom the procedure was reversed because of pain and local infection. This significant improvement should be viewed with caution because of the relatively short follow-up. Because of the short-term follow-up and the small patient population, it is difficult to compare this treatment to other treatment modalities. The purpose of this study was to show safety and initial effect of the procedure. We found the procedure safe and fairly easy to perform. Postoperative discomfort lasted for 1 to 2 weeks. In 2 patients with prolonged pain and local infection, the procedure was reversed by suture removal. The ability to reverse the procedure is an advantage leaving no change in anatomy despite failure of the procedure.
The initial improvement in sleep parameters in these patients is encouraging. Long-term follow-up and a more controlled study are necessary to prove efficacy. In SDB multiple sites of collapse are likely to be involved; therefore tongue-base stabilization may be an important minimally invasive addition to other treatment modalities aimed at other levels of obstruction during sleep. A combined multilevel approach with minimal morbidity to the patient may be appropriate in surgical correction of SDB.