INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: obstructive sleep apnea; treatment; dental appliance

It has been well documented that anterior displacement of the mandible using a dental appliance (DA) reduces the severity of obstructive sleep apnea syndrome (OSAS) (1-8). Anterior advancement of the mandible increases the retropalatal airway patency as well as that at the base of the tongue in the passive pharynx of patients with OSAS (9). There has been growing interest, over recent years, in this principle of treatment and a number of devices have been developed. However, lack of standardization of the use of DA for the treatment of OSAS raises a number of problems (9). Published experiences with these devices indicate that they are effective to varying degrees in the treatment of obstructive respiratory events. At least two hypotheses can be proposed to explain these discrepancies. First, there are major differences in design between the various DA currently available, and this may have an impact on the success rate (10). Second, in most of the previous studies, the degree of mandibular advancement was progressively increased over several weeks until snoring ceased and symptoms were improved or until the patient could not tolerate any further advancement (3-6). Although Kato and coworkers recently described a marked dose-dependent effect of mandibular advancement on sleep-disordered breathing (11), this clinical approach evaluated the effect of a DA at a single mandibular position per patient and did not determine whether repositioning the mandible resulted in an increase or decrease in the efficacy of the device. Ideally, evaluation of the efficacy of a DA should be conducted during a single night of polysomnography, in the same way as titration of effective continuous positive airway pressure (12).

The objective of our study was to assess the feasibility of progressive mandibular advancement during sleep without arousing the patient by means of hydraulic propulsion of the mandibular arch of the DA.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

The study population consisted of seven snorers (six men) with OSAS (66.9 ± 32.4 apneas and/or hypopneas per hour) confirmed during an all-night diagnostic sleep study. The age of the patients ranged from 20 to 60 years (mean ± SD: 50 ± 17 years), and their body mass index ranged from 22 to 33 kg/m² (28 ± 4 kg/m²). All patients were intolerant of nasal continuous positive airway pressure (nCPAP). After being informed of the various therapeutic options, the patients were invited to freely participate in this prospective pilot study. Written informed consent was obtained from each patient. The investigation was approved by the local ethics committee. Patients with inadequate dental structures for anchoring of the DA, temporomandibular joint dysfunction and/or previous uvulopalatopharyngoplasty were excluded.

Dental Appliances

Two custom-made DAs were used for each patient: a temporary DA (TDA) and a permanent DA (PDA). After taking impressions of the upper and lower teeth and recording the central occlusion position, two upper (maxillary) and two lower (mandibular) separate full coverage acrylic appliances were constructed that could be clipped onto the two dental arches.

The two arches of the TDA were connected by two lateral hydraulic systems. They were composed of a 25-mm segment of a 1-ml syringe (tube and piston) fixed over the second upper molar and the lower canine. Each syringe was connected to a 1-mm diameter catheter connected to a single syringe (20 ml) filled with physiologic saline and situated outside of the patient's room. Mandibular advancement was induced by mobilization of the column of water in the syringe. The degree of advancement was controlled by the volume injected and by a paper graph plotter attached externally to the anterior part of the mandibular arch (Figure 1). Plaster models of the upper and lower teeth were mounted on a semi-adjustable articulator to construct a temporary DA in the initial central occlusion position.

View larger version (134K): [in this window] [in a new window]   Figure 1.   Prototype titration device.

The two arches of the permanent DA were connected by Herbst attachments (rod and tube devices) allowing for opening, protrusion, and some side-to-side movement but no retrusion of the mandible.

Study Design

Each patient underwent two consecutive nights of full polysomnography: a titration night and a treatment night. Measurements included sleep stage (electroencephalogram, electro-oculogram, and submental electromyogram), nasal airflow (nasal cannulae) and oral airflow (oral thermistance), rib cage and abdominal wall motion (respiratory inductance plethysmography), and arterial oxygen saturation (Sa_O2). All polysomnographic data were analyzed visually. Sleep staging was performed according to standard criteria (13). Apnea was defined as cessation of airflow for more than 10 seconds. Hypopnea was defined as a reduction of airflow, regardless of its amplitude, with a fall of Sa_O2 greater than 3% or arousal (14).

Titration night. For each patient, the maximum voluntary advancement of the mandible was determined before inserting the device. The greatest advancement measured on three consecutive maneuvers was adopted. As the patients had never previously used a DA, the TDA was adapted to the patient's mouth 30 minutes before the start of the recording. Mandibular advancement, by increments of 1 mm every 15 minutes, was started after the first episode of rapid eye movement (REM) sleep or after 2 hours of recording after sleep onset. This advancement was continued until a significant reduction of the incidence of the sleep-disordered breathing was obtained or until reaching the position of maximum advancement of the system or the position causing discomfort or pain, waking the patient and preventing any further progression.

Treatment night. The length of the Herbst attachments of the PDA was adjusted on the semi-adjustable articulator to obtain the final advancement reached during the previous titration night.

A paired t test was used to compare polysomnographic indices obtained during the diagnostic night and the treatment night in the same patient.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Titration Night

Maximum active protrusion by the conscious patient was 10.7 ± 2.5 mm (8-15 mm). Advancement was begun 153 ± 37 minutes (105-199 minutes) after the start of the recording. The mean mandibular advancement reached during the titration night was 12.6 ± 2.7 mm (9-17 mm) corresponding to 120 ± 20% (100-150%) of the maximum active protrusion. This level was reached after 71 ± 28 minutes (30-120 minutes). The total sleep time with the DA was 201.3 ± 54 minutes (93-256 minutes) with 21.6 ± 18.7% (0-49%) of slow-wave sleep and 10.2 ± 7.3% (0-21.4%) of REM sleep. The mean wakefulness time during sleep was 56 ± 28 minutes (27-92 minutes), but an awakening was never observed during or for 60 seconds following advancement maneuvers. Discomfort caused prolonged arousal, especially in patient no. 3, in whom only 93 minutes of sleep without REM were obtained. The apnea-hypopnea index (AHI) of the whole group decreased from 66.9 ± 32.4 to 26.1 ± 20.7 events per hour with the device. The most marked reduction was observed for apneas, as the apnea index decreased from 35.2 ± 27.1 to 6.9 ± 6.3 per hour. In patient no. 7, despite the persistence of sleep-disordered breathing, the advancement process was stopped at the maximum value allowed by the construction of the device, i.e., 17 mm. The patient was not woken by this degree of advancement.

All patients reported a feeling of painful tightness of the temporomandibular joint and muscles on waking. This feeling resolved in less than 60 minutes after removal of the DA in every case.

Treatment Night

All patients agreed to complete the second night of recording with the DA adjusted according to the data of the titration night.

The results of the titration night were confirmed in all patients (Figure 2). The individual polysomnographic data are shown in Table 1. The AHI decreased compared with the titration night in all but one of the patients. The mean AHI of the group during treatment was 19.6 ± 20.2 events per hour. When response was defined as a reduction of the AHI to below 10 per hour, 42.9% of patients were considered to be responders. When response was defined as a reduction of the AHI to below 20 per hour, 71.4% of patients were considered to be responders.

View larger version (19K): [in this window] [in a new window]   Figure 2.   Individual distribution of AHI during the diagnostic night, the titration night and the treatment night.

Patient Outcome

Patient no. 2, 19-years-old, preferred jaw advancement surgery. After a 12-mm surgical mandibular advancement, the AHI was 2 per hour. In patient nos. 4 and 7, further mandibular advancement was unsuccessful. The other four patients have been treated at the degree of mandibular advancement determined at the end of the titration night for an average of 288.8 ± 57.4 days (239-377 days).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This pilot study shows that it is possible to mobilize the mandible during polysomnography without waking the patient during the advancement maneuvers. The simple hydraulic propulsion system used in this study allows one-night titration of advancement of the DA away from the patient's bedside and prediction of the capacity of the DA to reduce the incidence of sleep-disordered breathing.

Among the various treatment options able to restore patent upper airways in OSAS, DA appears to be a promising alternative to nCPAP. In the majority of studies devoted to DA, clinical criteria have been the major determinants in the choice of the degree of final mandibular advancement without looking for any additional objective benefit in the case of polysomnographic failure. However, Kato and coworkers recently reported a dose-dependent effect of mandibular advancement on pharyngeal mechanics in patients with sleep-disordered breathing (11). They reported that mandibular advancement in paralyzed patients with OSAS under general anesthesia produced a dose-dependent reduction of closing pressure in all pharyngeal segments. In their study, each 2-mm mandibular advancement coincided with approximately 20% improvement in number and severity of nocturnal desaturations during normal sleep in these patients with OSAS. Like nCPAP (15), the DA must therefore be titrated, and the effective degree of advancement mandibular must be determined not simply on subjective clinical criteria, but also on objective data provided by polysomnographic recording. Like nCPAP, one-night polysomnographic titration would be the most appropriate solution to define the degree of advancement mandibular effective on sleep-disordered breathing or, on the contrary, to conclude on the inefficacy of this treatment. Raphaelson and colleagues reported the results of titration performed during polysomnography in six patients with OSAS (12). They used a DA adjusted by means of an anterior screw, the Silencer. A highly significant reduction of the AHI was obtained in the six patients studied during this titration night, and four of them continued to use the DA after 6 months. They reported little information about the quality of sleep of patients submitted to this titration procedure. Several awakenings were probably induced by adjustments of the Silencer by the technician because of the site of the adjustment system on the DA itself. As a result of the hydraulic propulsion system in our DA, mandibular advancement was controlled away from the patient's bedside, with no direct intervention on the DA. We did not observe any awakening occurring during the various mandibular advancement maneuvers. Patients with OSAS have a high arousal threshold (16). Possible explanations for a higher arousal threshold in patients with OSAS could be habituation to repeated arousal stimuli and/or decreased upper airway mechanoreceptor function (17). Our patients suffered from severe OSAS, and the absence of awakening observed during mandibular advancement maneuvers could be partly explained by a particularly marked elevation of their awakening threshold. However, awakenings were sometimes observed during the titration phase, independently of the brief periods of mandibular advancement. The fact that our patients had never previously used a DA, and the large intra-oral volume of this prototype titration device probably affected the results and certainly explain the short sleep time observed in one patient, and probably the short REM sleep time observed in another three patients during the titration night. Sleep instability could have contributed to respiratory instability generating hypopneas and certainly interfered with determination of the degree of mandibular advancement effective during REM sleep in these patients. Improved control of respiratory events during the second night with the DA could be explained by improved sleep stability. The results of a new treatment recording performed 7 months later in patient no. 1 support this hypothesis, as, although the AHI was only slightly modified (7 versus 2 per hour), the arousal index decreased from 22.3 to 9.1 per hour after regular use of the DA for 7 months. Adaptation to the presence of the DA in the central occlusion position for several nights before titration and miniaturization of the hydraulic propulsion system would be recommended during subsequent studies.

However, despite these methodological problems, a significant reduction of the incidence of the sleep-disordered breathing was observed during the titration night, as the AHI decreased by an average of 56%. Only one patient did not obtain any improvement during treatment. The amount of mandibular advancement was then progressively increased (1 mm per week) until the patient could not tolerate any further advancement and did not obtain any significant improvement of sleep-disordered breathing. A similar course was observed for patient no. 4. In every patient, the result of the titration night was predictive of the result of the treatment night, as demonstrated by the values of the AHI obtained during evaluation of the efficacy of the device. At the end of the second night of treatment, the AHI decreased by an average of 69% compared with the diagnostic examination and 71% of patients had an AHI of less than 20 per hour. However, these results were obtained in a small series of patients, as the objective of this study was not to evaluate the efficacy of the device, but to determine the feasibility of titration. Nevertheless, it should be noted that our study population was composed of patients with particularly severe OSAS, as the initial mean AHI was 66.9 per hour (range: 30-106 per hour). Lowe and coworkers recently reported a very similar success rate, 61%, in a group of 18 severe patients with an AHI ranging between 30 and 75 per hour (18). In this series, effective mandibular advancement (9-14 mm) was also close to that observed in our study (between 9 and 17 mm). These results show that the initial severity of the disease is not predictive of failure of the DA technique. The possibility of one-night polysomnographic titration could allow this device to be proposed as first-line treatment of OSAS, even in the most severe patients.

In conclusion, this pilot study shows that it is possible to perform nonarousing mobilization of the jaw during polysomnography. These results suggest that the efficacy of DA and the effective degree of mandibular advancement can be predicted during a polysomnographic recording. Further studies therefore need to be conducted on a larger number of patients previously accustomed to using a smaller DA.