Introduction
The aim of the study was to investigate psychosocial function in patients with obstructive sleep apnea/hypopnea syndrome (OSAHS), before and after mandibular advancement splint (MAS) therapy.
Methods
In a prospective, parallel-group study, changes in psychosocial scores were compared before and after MAS therapy for OSAHS. Patients from the top (MAS treated, n = 40) or bottom (conservatively treated n = 35) of a MAS waiting list rated quality of life (Short Form 36 [SF-36] Health Survey) and sleepiness (Epworth Sleepiness Scale [ESS]). Both groups’ ratings at baseline were completed while on conservative therapy.
Results
Baseline variables did not differ between groups. At follow-up, psychosocial scores showed improvement in MAS-treated patients, with odds ratios (confidence interval) of 0.26 (0.09, 0.75) and 0.36 (0.14, 0.92) for sleepiness and energy/vitality domain, respectively. Statistically significant improvement in the median (range) sleepiness score (10 [1 to 18] to 6 [1 to 14]; P <0.001) and energy/vitality domain (18 [7 to 20] to 19 [14 to 20]; P = 0.03) were observed with MAS therapy. Reductions were detected in median apnea-hypopnea index (median [range] 16 [5.2 to 30] to 4.6 [0.8 to 17.2] events/hour; P <0.001), and oxygen desaturation index (ODI) (11 [3 to 16] to 0 [0 to 5] events/hour; P <0.001) at follow-up with MAS therapy. Similar changes were observed in the comparison group when they went on to be treated with MAS.
Conclusions
This study demonstrated that treatment with MAS produced statistically and clinically significant psychosocial and cardio-respiratory improvements.
Obstructive sleep apnea/hypopnea syndrome (OSAHS) has a considerable impact on quality of life, with excessive daytime sleepiness, depression, and cardiovascular insults all implicated. Continuous positive airway pressure (CPAP) is an effective treatment for OSAHS, with improvements in sleepiness and general health/function status. Specifically, it reverses excessive daytime sleepiness and has been shown to have large effects on energy/vitality and moderate effects on mental and physical health domains.
Mandibular advancement splint (MAS) therapy is increasingly being recognized as a treatment for mild to moderate OSA. The difficulty with prescribing MAS therapy is that there is large variability in response to treatment. A possible reason for this is that there are few biological markers to indicate which OSAHS patients will benefit from MAS therapy. In most cases, referrals for treatment are based on number of abnormal respiratory events per hour (apnea/hypopnea index; AHI). However, the AHI does not take account of the dynamics of upper airway narrowing. For example a high AHI can occur as a result of an enlarged uvula or mandibular retrognathia with retroglossal narrowing; MAS is likely to be more beneficial in the latter situation. In a small study, we recently showed that MAS therapeutic success (defined as a reduction in AHI <10 events/hour) can be improved using sleep nasendoscopy (SNE) to preselect the OSA patients who are most likely to respond to MAS therapy.
However, despite the increased recognition of the role of MAS in the treatment of OSAHS and the recognized debilitating effects of this condition, quantitative comparisons of the magnitude of effect of MAS therapy on psychosocial function is scant.
The primary aim of the current study was to compare changes in psychosocial function in MAS-treated patients with those in conservatively treated patients. Secondary aims were to determine the magnitude of change in cardiopulmonary outcome measures with MAS therapy.
Material and methods
Trial design
This trial was a repeated-measure, parallel-arm, intention-to-treat study of conservative vs MAS treatment in a clinical setting. Psychosocial questionnaires were completed at baseline, when all patients were being managed conservatively with advice on sleeping position, avoidance of alcohol in the evenings, and weight loss. The “conservative” patient group was still receiving conservative management only when followed-up 3 months later. The “MAS” patient group was followed-up 3 months after the patients’ MAS titration. The protocol was approved by the local ethics committee, and all patients gave written informed consent.
Study participants
A sample size calculation was determined from the nomogram described by Altman. Based on the findings of Bennett et al, the standard deviation for the ESS questionnaire was reported to be 3.4. It was estimated that a sample size of 40 patients in each group would be required to find a clinically relevant difference of 3 points on the Epworth Sleepiness Scale (ESS), at the 5% level of significance and a power of 80%, compared with controls. Eighty-five patients were recruited in total, with a greater number in the treatment (n = 45) than in the control (n = 40) group to allow for anticipated higher levels of dropouts.
The entry criterion for all patients was the clinical recommendation of MAS therapy. Inclusion criteria were mild to moderate OSA (AHI >5 <30 events/hour); favorable SNE findings (an improved airway patency plus snoring severity during simulated mandibular advancement ); healthy teeth and supporting periodontium, no evidence or history of temporomandibular joint disease, and age older than 18 years. Patients were excluded if they had metabolic disease, morbid obesity, obesity-hypoventilation syndrome, gross cardiovascular anomalies (history of myocardial infarction and/or stroke), or uncontrolled hypertension and AHI greater than 30 events per hour.
Position on the MAS treatment waiting list acted as the group selection criterion. Patients with a wait of 5 to 6 months were assigned to the conservative follow-up, which was 3 months later. Those with a less than 3-month wait were assigned to MAS follow-up, which was 3 months after titration. Eligibility criteria were met by 85 patients (MAS, n = 45; conservative, n = 40). Of these, 5 patients (2 MAS, 3 conservative) were not included, as they failed to return completed questionnaires. A further 5 (3 MAS, 2 conservative) patients were lost to follow-up. Three reported they were unable to cope with the MAS, and 2 failed to return for conservative follow-up at 3 months. The final sample sizes were 40 MAS and 35 conservative patients.
Mandibular advancement splint
A single operator treated all patients, with the same dental laboratory fabricating all the appliances. A removable Herbst appliance ( Fig 1 ), which comprised separate maxillary and mandibular full occlusal coverage acrylic splints connected by bilateral telescopic arms to permit the mandible to be advanced in a stepwise fashion, was used in the study. The Herbst MAS was constructed with a bite registration recording the patient’s maximum comfortable protrusion. Short intermaxillary elastics were attached anteriorly to limit jaw opening during sleep.
Overnight sleep studies
OSA was diagnosed in all patients prior to recruitment, and the effectiveness of MAS therapy was objectively assessed overnight in-hospital with the MAS in situ. An experienced sleep specialist, blinded to the purpose of the study, scored both the baseline and follow-up studies using standardized criteria: respiratory events were defined as apnea, the absence of airflow for more than 10 seconds; hypopnea, more than a 50% reduction in airflow for greater than 10 seconds in the presence ofmore than 4% desaturation (or an arousal from sleep—baseline study only). The oxygen desaturation index (ODI) was defined as the number of times the oxygen saturation dipped more than 4% per hour of sleep; the mean overnight Sa o 2 was also recorded.
Outcome measurements
Health status
Short Form 36 Health Survey . The Short Form 36 (SF-36) questionnaire provides a generic 36-item measure of subjective health status of the patient. It provides 3 summary scores for health transition (score range 1 to 5) and mental and physical components (standardization provides population means ± SD of 50 ±10). There are 8 domain subscores (role limitation: physical; role limitation: emotional; social function; mental health; energy/vitality; pain; general health; and physical function) which range from 0 to 100.
Sleepiness
Epworth Sleepiness Scale . The 8-item Epworth scale is widely accepted and provides a subjective measure of daytime sleepiness. The patient scores the likelihood of dozing in situations of differing somnolence, with a score range of 0 to 24 and a normative mean ± SD of 5 ± 3.
Statistical analysis
Data analysis was carried out using the SPSS for Windows (SPSS, Chicago, Il), with alpha-significance values set at 0.05. The baseline characteristics of the MAS vs conservative group were compared using independent sample t tests (normally distributed data) and the Mann-Whitney test (non-normally distributed data) as appropriate. The Wilcoxon matched-pairs signed rank test was used to analyze differences between AHI, apnea index, ODI, and mean desaturation overnight in the MAS group before and after therapy.
Univariate analysis was used to test the hypothesis that improvements in psychosocial function were associated with MAS therapy. Odds ratios and 95% confidence intervals for the improvement in quality of life domains and sleepiness were computed using logistic regression statistical tests. The data were not subsequently tested with multivariate analysis because of the similarity of baseline characteristics between the treatment and control groups ( P >0.20; Table I ).
| Baseline characteristics | MAS group ( n = 40) |
Conservative group (n = 35) |
P value |
|---|---|---|---|
| Age (years) | 49.3 (46.2, 52.4) | 51.4 (47.9, 54.5) | 0.51 |
| BMI (kg/m 2 ) | 25.9 (25.1, 26.9) | 26.2 (25.3, 27.2) | 0.65 |
| Neck circumference (cm) | 39 (39, 40) | 40 (39, 41) | 0.35 |
| Epworth sleepiness scale | 10 (1-18) | 12 (0-18) | 0.28 ∗ |
| AHI (events/hour) | 16 (5-30) | 15 (5-30) | 0.33 |
| Apnea index (events/hour) | 5 (0-16) | 4 (0-15) | 0.62 |
| ODI (events/hour) | 11 (3-16) | 10 (3-14) | 0.29 |
| Mean saturation overnight (%) | 95 (90-98) | 96 (93-98) | 0.28 |
| Sleep efficiency (%) | 87 (54-98) | 84 (55-94) | 0.65 |
∗ All parameters tested with independent sample t test, with the exception of the Epworth sleepiness scale, which was tested with Mann-Whitney test.
Results
Between group comparison at baseline
At baseline there were no differences in age, gender, body mass index, neck circumference, and daytime sleepiness between the groups ( Table I ). The 2 groups were also comparable in their baselines: AHI, apnea indices, ODI, mean desaturation levels, and sleep efficiency ( Table I ). At baseline there were no differences in any of the SF-36 domains or the ESS scores between the MAS and conservative groups ( Table II ).
| Variable | Baseline | P value | Follow-up | P value |
|---|---|---|---|---|
| Sleepiness (Epworth sleepiness scale) | ||||
| MAS | 10 (1-18) | 0.18 | 6 (1-14) | <0.001 |
| Conservative | 12 (0-18) | 11 (1-18) | ||
| Quality of Life (SF-36) domains | ||||
| Role limitation (physical) | ||||
| MAS | 23 (20-23) | 0.15 | 23 (20-23) | 0.74 |
| Conservative | 23 (20-23) | 23 (20-23) | ||
| Role limitation (emotional) | ||||
| MAS | 23 (20-23) | 0.57 | 23 (20-23) | 0.68 |
| Conservative | 23 (20-23) | 23 (20-23) | ||
| Social functioning | ||||
| MAS | 20 (13-20) | 0.58 | 20 (16-20) | 0.54 |
| Conservative | 20 (10-20) | 19 (11-20) | ||
| Mental health | ||||
| MAS | 19 (16-20) | 0.84 | 19 (17-20) | 0.57 |
| Conservative | 19 (13-20) | 19 (15-20) | ||
| Energy/vitality | ||||
| MAS | 18 (7-20) | 0.41 | 19 (14-20) | 0.03 |
| Conservative | 17 (7-20) | 17 (7-20) | ||
| Pain | ||||
| MAS | 20 (11-21) | 0.70 | 20 (14-21) | 0.94 |
| Conservative | 20 (11-21) | 20 (11-21) | ||
| General health perception | ||||
| MAS | 19 (13-20) | 0.12 | 19 (13-20) | 0.26 |
| Conservative | 18 (13-20) | 18 (12-20) | ||
| Physical function | ||||
| MAS | 20 (12-20) | 0.38 | 20 (17-20) | 0.26 |
| Conservative | 20 (12-20) | 19 (7-20) | ||
Between-group comparison at follow-up
Primary outcomes
Changes in the SF-36 domains were calculated between the baseline and follow-up period for the MAS and conservative groups. A significant difference ( P = 0.03) was detected only in relation to the energy/vitality domain ( Table II ). While the median score itself showed a small improvement (from 18 to 19), the range demonstrated a more significant improvement. As the scale of measurement was ordinal, the quality of life domain scores were transformed into a dichotomous scale to overcome the difficulties that arise in relation to analyzing questionnaire data—that is, no clear threshold exists that alone can signify treatment success—an increase in the score was noted as improvement and a decrease as no improvement. In the MAS group 68% of the patients showed an improvement in the energy/vitality domain ( P = 0.03; Table III ). A significant difference ( P <0.001) was detected in sleepiness between the MAS and conservative groups from the baseline to 3-month follow-up ( Table II ). The ESS scores were also transformed into a dichotomous scale; a decrease in sleepiness was recorded as improvement and an increase as no improvement. In the MAS group, 80% of patients showed improvement in sleepiness ( P = 0.01; Table III ).
