- Email: [email protected]
Oral Continuous Positive Airway Pressure for Sleep Apnea
Oral Continuous Positive Airway Pressure for Sleep Apnea* Effectiveness, Patient Preference, and Adherence Jaime Beecroft, BSc; Sandra Zanon, RPSGT; Dejan Lukic, RPSGT; and Patrick Hanly, MD
Background: Nasal continuous positive airway pressure (CPAP) is the most definitive medical therapy for obstructive sleep apnea (OSA). Many patients have difficulty tolerating nasal CPAP due to nasal airway problems, mouth leak, and general discomfort from the mask and headgear. These limitations may be overcome by an oral mask (Oracle; Fisher & Paykel Healthcare; Languna Hills, CA) that does not require headgear. We performed a study to compare the Oracle mask to conventional nasal and oronasal masks in the effectiveness of CPAP delivery and patient satisfaction and adherence. Methods: Ninety-eight, consecutive CPAP-naı¨ve patients with OSA diagnosed by overnight polysomnography (apnea-hypopnea index [AHI] > 5) were referred for CPAP therapy. All patients were presented with a variety of CPAP masks, including nasal, oronasal, and Oracle, and reasons for mask choice were documented. After 3 weeks of acclimatization to the mask of their choice, patients had a CPAP titration sleep study to determine their optimal CPAP level. Further follow-up was obtained 2 months and 6 months later with a subjective patient assessment of CPAP use and efficacy, mask comfort, and upper airway dryness. Results: Patients were predominantly male (70%), middle aged (50.6 ⴞ 11.7 years), and moderately obese (body mass index, 32.5 ⴞ 9.0) with severe OSA (AHI, 40.6 ⴞ 25.8/h) [mean ⴞ SD]. Patients were classified into three groups based on their choice of mask: nasal (66%), Oracle (27%), and oronasal (7%). Baseline characteristics did not differ significantly between groups. Optimal CPAP was not significantly different between mask groups (nasal, 7.7 ⴞ 2.1 cm H2O; Oracle, 8.0 ⴞ 2.0 cm H2O; oronasal, 9.7 ⴞ 3.2 cm H2O; p ⴝ 0.267). Subjective ratings of adherence, efficacy, and mask comfort were also similar between groups. However, the Oracle group had more complaints of upper airway dryness and “rain-out.” The oronasal group had a disproportionately greater number of dropouts from CPAP therapy than the Oracle group (57% vs 19%, p ⴝ 0.046). Nine patients changed from the Oracle mask to a nasal mask during the study, whereas no patients changed from their nasal or oronasal masks. Conclusions: The Oracle mask is an efficacious interface for long-term CPAP therapy in patients with OSA. The main limitations of the mask are upper airway dryness and rain-out associated with heated humidification, which may be improved by further technical modifications. Oracle may be more acceptable than oronasal masks for patients who cannot rely exclusively on the nasal airway for CPAP therapy. (CHEST 2003; 124:2200 –2208) Key words: adherence; oral continuous positive airway pressure; sleep apnea Abbreviations: AHI ⫽ apnea-hypopnea index; CPAP ⫽ continuous positive airway pressure; OSA ⫽ obstructive sleep apnea
sleep apnea (OSA) is a common O bstructive condition that is associated with excessive daytime sleepiness and increased risk of cardiovascular
disease, particularly hypertension.2 The medical therapy of choice for OSA is the application of continuous positive airway pressure (CPAP), which
*From the Sleep Disorders Clinic (Mr. Beecroft, Mr. Lukic, and Dr. Hanly), St. Michael’s Hospital, University of Toronto, Toronto; and Medigas/Praxair (Ms. Zanon), W Toronto, ON, Canada. This research was performed at Sleep Disorders Clinic, St. Michael’s Hospital, University of Toronto, and Medigas/Praxair. Supported by Fisher & Paykel Healthcare.
Manuscript received January 8, 2003; revision accepted July 1, 2003. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: [email protected]). Correspondence to: Patrick J. Hanly, MD, 6049 Bond, St. Michael’s Hospital, 30 Bond St, Toronto, ON, Canada M5B 1W8; e-mail: [email protected]
1
2200
Clinical Investigations
prevents upper airway closure by creating a “pneumatic splint.”3 Although CPAP therapy is effective, patients frequently stop using it because of discomfort predominantly associated with the mask.4 The conventional method for administration of CPAP therapy is through a nasal or oronasal interface. While oronasal masks have been reported to overcome upper airway dryness associated with nasal masks,5 they have been rated as less comfortable and have been associated with a greater rate of noncompliance and less improvement in subjective sleepiness than nasal masks.6 However, nasal masks have been associated with more symptoms related to mouth leak, such as dry throat and mouth,6 and have been associated with other complaints such as nasal dryness, nasal congestion, and rhinorrhea.7 An alternative approach is to deliver CPAP exclusively through the mouth. The Oracle (Fisher & Paykel Healthcare; Languna Hills, CA) is an oral CPAP mask that avoids nasal airway problems and does not require headgear. The objective of this study was to determine how often patients chose the Oracle mask over the current conventional CPAP masks, and how their adherence and satisfaction with the Oracle mask compared to that of nasal and oronasal masks.
Materials and Methods Patient Recruitment All patients with OSA diagnosed at St. Michael’s Hospital Sleep Laboratory were considered potential candidates for this study. Patients referred for assessment of sleep apnea underwent diagnostic polysomnography, and those with an apnea-hypopnea index (AHI) ⬎ 5/h were seen for clinical follow-up at the ambulatory sleep clinic. Following clinical assessment, specific management of sleep apnea was discussed. Patients in whom CPAP therapy was indicated were referred to a home-care company that provided this treatment. None of these patients had previously received CPAP. Patients received nondirected, standardized counseling on the use of CPAP. At the home-care company, all patients were seen for approximately 1 h by the same individual, an experienced registered polysomnographic technologist (S.Z.). Each patient was shown the same range of masks, which consisted of the Oracle mask, nasal masks, and oronasal masks. The features of each mask type were reviewed with the patient. They were assisted to try on as many masks as they wished and asked to choose one to take home. In addition, they were placed on 5 cm H2O CPAP through the mask of their choice for a few minutes, and any questions or concerns were addressed. The reasons for patients’ choice of mask were documented. Patients were provided with a CPAP unit of their choice. Those who chose the Oracle mask were given heated humidification. Those who chose a nasal mask or oronasal mask were given whatever level of humidity was required for their comfort (cold passover, heated, or none). All patients were then acclimatized on a CPAP level of 5 cm H2O at home for 3 weeks before returning to the laboratory for a CPAP titration study. www.chestjournal.org
Oracle Mask The Oracle oral mask is a strapless, butterfly-shaped interface that rests in the oral vestibule between the lips and teeth (Fig 1). It is comprised of a rigid inner framework, which depresses the tongue, a soft seal, which sits between the lips and teeth, an adjustable flap, which seals around the mouth on the skin surface, and tubing, which attaches to the CPAP unit. The features of the mouthpiece ensure the desired positive airway pressure is delivered to the patient with minimal leakage, and that the mouthpiece is retained in the mouth while asleep. The flexible tubing facilitates freedom of movement while maintaining circuit integrity, and contains an exhaust port adjacent to the mouthpiece, which provides a means to purge exhaled gases from the breathing circuit. Polysomnography Diagnostic overnight polysomnography was performed in a standardized fashion in the Sleep Laboratory at St. Michael’s Hospital. Recordings were performed by continuous monitoring of the EEG, electrooculogram, and submental electromyogram, ECG, nasal airflow (Ultima Dual Airflow Pressure Sensor; Braebon Medical Corporation; Kanata, ON, Canada), chest and abdominal respiratory movements (Respitrace; Ambulatory Monitoring; Ardsley, NY), oximetry (Mallinckrodt/Nellcor Puritan Bennett; Hazelwood, MO), anterior tibialis electromyogram, and body position. The recordings were performed and scored by experienced, registered polysomnographic technologists. Scoring of the polysomnographic data was done in a standardized fashion, according to published criteria.8 At the time of the diagnostic polysomnography, patients were required to complete a sleep history questionnaire and Epworth sleepiness scale, as well as provide demographic information. Following CPAP orientation and acclimatization, all patients underwent a CPAP titration study at the sleep laboratory 3 weeks after they received their CPAP mask and unit. All patients were instructed to bring their own mask, which they wore for the duration of the study. Recordings were performed in a similar manner to the diagnostic study, except that the patients were monitored on CPAP and the pressure was titrated to a level that resolved obstructive respiratory events and was comfortable (termed optimal CPAP). Patients also completed a CPAP questionnaire (Appendix). The CPAP units were set at this level the morning after their study. Follow-up One week following commencement of CPAP therapy, patients were contacted by the home-care company to address problems and record their adherence to therapy. If patients were unable to tolerate either the Oracle mask or other mask of their choice, they were offered the option of switching to an alternative mask. Reasons for mask preference were documented. One week following the CPAP titration study, a second telephone follow-up was performed by the home-care company to address problems and record adherence. Subsequently, patients were contacted by telephone at 2 months and 6 months following commencement of CPAP therapy. At these times, patients completed a CPAP questionnaire designed to give a subjective assessment of CPAP usage and efficacy, mask comfort, and upper airway dryness. Adherence was measured as the average number of hours per day and days per week the patient reported using CPAP. Patients were asked to rate the comfort of their mask on a 10-point scale. Within this scale, a rating of 1 indicated the mask was so uncomfortable it was unusable, while a 10 indicated the mask was so comfortable they CHEST / 124 / 6 / DECEMBER, 2003
2201
Figure 1. Oracle CPAP mask.
did not realize it was on. Patients were also asked to rate upper airway dryness. Patients were asked whether they experienced dryness in their nose or mouth when waking after using CPAP and, if so, instructed to rate the dryness on a 10-point scale. Within this scale, a rating of 1 indicated the dryness was so bad they could not use CPAP, while a rating of 10 indicated the dryness was noticeable but not uncomfortable. Efficacy was rated on a scale that ranged from 0 to 6, with 0 indicating that the patient felt worse than prior to using CPAP, and 6 indicating that the patient felt much better than prior to using CPAP. Analysis Prevalence of mask choice and reasons for mask preference were analyzed, and polysomnography data and patient satisfaction/adherence were compared for the nasal, oronasal, and Oracle mask groups. Furthermore, dropouts from CPAP therapy and changes in mask preference were compared between the nasal, oronasal, and Oracle mask groups. The means of data among these three groups were compared using one-way analysis of variance and post hoc Tukey tests to determine significant differences. Two-month and 6-month data were compared using paired Student t test, and gender differences were examined using unpaired Student t test. Ordinal and binary variables were compared using a 2 test. All p values ⬍ 0.05 were considered statistically significant. 2202
Results Study Cohort Ninety-eight patients with OSA were recruited from St. Michael’s Hospital Sleep Clinic over a 6-month period and referred for CPAP therapy. Patients were classified into three groups according to whether they chose a nasal, oronasal, or Oracle mask. The baseline demographics (Table 1) and polysomnography data (Table 2) of the three groups were similar. The “optimal CPAP” level, as defined in the “Materials and Methods,” was not significantly
Table 1—Patient Demographics* Variables
Nasal
Oronasal
Oracle
Patients, No. 65 7 26 Male patients, % 69.2 71.4 73.1 Age, yr 49.6 ⫾ 10.2 53.0 ⫾ 19.0 53.1 ⫾ 13.2 Body mass index 32.2 ⫾ 8.7 38.6 ⫾ 17.4 31.7 ⫾ 6.8
p Value 0.934 0.385 0.231
*Data are presented as mean ⫾ SD unless otherwise indicated. Clinical Investigations
Table 2—Baseline Polysomnography* Variables
Nasal
Oronasal
Oracle
p Value
Total sleep time, h Sleep efficiency, %† Sleep latency, min‡ Stage 1 NREM, %TST Stage 2 NREM, %TST SWS, %TST REM, %TST AHI, /h Mean Sao2, % Periodic leg movements, /h Arousal index, /h Epworth sleepiness scale
5.6 ⫾ 1.2 79.0 ⫾ 13.7 14.1 ⫾ 14.4 9.1 ⫾ 7.5 60.9 ⫾ 11.1 13.4 ⫾ 9.0 16.5 ⫾ 7.3 39.9 ⫾ 25.8 93.7 ⫾ 2.9 9.3 ⫾ 15.7 42.1 ⫾ 24.1 10.0 ⫾ 5.4
5.0 ⫾ 2.0 71.1 ⫾ 23.9 8.5 ⫾ 6.5 13.7 ⫾ 8.7 60.6 ⫾ 11.4 6.1 ⫾ 9.8 19.5 ⫾ 9.6 43.4 ⫾ 40.7 91.2 ⫾ 3.8 30.1 ⫾ 38.3 42.0 ⫾ 42.9 9.8 ⫾ 4.0
5.6 ⫾ 1.1 78.0 ⫾ 13.0 14.6 ⫾ 17.5 10.2 ⫾ 6.7 61.2 ⫾ 10.6 11.7 ⫾ 7.8 19.5 ⫾ 9.6 41.6 ⫾ 22.6 93.9 ⫾ 2.8 9.7 ⫾ 25.1 39.5 ⫾ 23.0 11.4 ⫾ 5.5
0.527 0.437 0.656 0.343 0.988 0.132 0.629 0.924 0.113 0.059 0.904 0.528
*Data are presented as mean ⫾ SD. NREM ⫽ non-rapid eye movement sleep; SWS ⫽ slow-wave sleep; REM ⫽ rapid eye movement sleep; %TST ⫽ percentage of total sleep time; Sao2 ⫽ oxygen saturation. †Total sleep time expressed as a proportion of total study duration. ‡Time from lights out to the first epoch of sleep.
different between mask groups (nasal, 7.7 ⫾ 2.1 cm H2O; oronasal, 9.7 ⫾ 3.2 cm H2O; Oracle, 8.0 ⫾ 2.0 cm H2O; p ⫽ 0.267). On optimal CPAP, there were no significant intergroup differences in AHI (nasal, 6.7 ⫾ 13.3/h; oronasal, 9.8 ⫾ 12.8/h; Oracle, 10.9 ⫾ 20.1/h; p ⫽ 0.555). In addition, the percentage of patients in each group whose AHI was ⬍ 5/h while on their optimal CPAP did not differ significantly (nasal, 61%; oronasal, 67%; Oracle, 39%; p ⫽ 0.185). Furthermore, when the Oracle group was compared individually to both the nasal and oronasal groups, no significant differences were found. Of the 98 patients enrolled in the study, 88 patients (90%) received heated humidification, while 9 patients (9%) received cold-passover humidification, and 1 patient (1%) chose not to use humidification. The proportion of patients who received heated humidification was similar between groups (nasal, 85%; oronasal, 100%; Oracle, 100%; p ⫽ 0.059).
found them to be the least invasive (51%); that nasal masks did not produce the sensation of claustrophobia, choking, or gagging they attributed to the Oracle mask (18%); that they were nasal breathers (14%); and that they found nasal masks to be generally more comfortable, to fit better, and anticipated less dryness than with other masks (17%). Fifty-seven percent of patients who chose an oronasal mask reported they were mouth breathers who thought the Oracle mask was too invasive or uncomfortable. During the course of the study, seven patients who began CPAP therapy using the Oracle mask changed to a nasal mask at the 2-month follow-up, and an additional two patients changed from the Oracle mask to a nasal mask at the 6-month follow-up. Patients who chose either nasal or oronasal masks did not change mask types during the course of the study (Table 3). Patient Follow-up
Mask Preference Although the majority of patients chose nasal masks, which included Breeze (Mallinckrodt/Nellcor Puritan Bennett), Contour Deluxe, Profile Lite, Simplicity, and Comfort Classic (Respironics; Murrysville, PA), 27% chose the Oracle mask and 7% chose an oronasal mask (Full Face Mirage; ResMed; Poway, CA) [Table 1]. There were no significant gender differences between groups in the choice of mask. Patients who chose the Oracle mask reported an appreciation for the lack of headgear (69%), a history of mouth breathing at night (15%), greater comfort, a sensation of “getting more air” compared to other masks, and a concern that a nasal mask would aggravate preexisting chronic sinusitis (16%). Patients who chose nasal masks reported that they www.chestjournal.org
Patient dropout rates were monitored at the time of the CPAP titration study, at 2 months and at 6 months after CPAP was initiated (Table 3). Of the 98 patients enrolled, 11 patients dropped out prior to undergoing their CPAP titration study, an additional 9 dropped out at the 2-month follow-up, and an additional 10 patients dropped out at 6 months. The proportion of dropouts at these various follow-up points did not differ between groups (Table 3). An additional two patients did not return for their CPAP titration study, but reported that they were using CPAP at their 2 month follow-up. At 6 months, a total of 30 patients had discontinued CPAP use, with a disproportionately greater number of patients within the oronasal group than the Oracle group (oronasal, four patients [57%]; Oracle, five patients CHEST / 124 / 6 / DECEMBER, 2003
2203
Table 3—Enrollment and Follow-up of Patients*
Enrolled Dropped out prior to CPAP titration Dropped out at 2-mo follow-up Dropped out at 6-mo follow-up Total dropped out Changed mask at 2 mo Changed mask at 6 mo Total changed Total changed and dropped out 2-mo data not obtained 6-mo data not obtained Unavailable for follow-up
Nasal
Oronasal
Oracle
Total
65 (66) 7 (11) 8 (12) 6 (9) 21 (32) 0 (0) 0 (0) 0 (0) 21 (32) 3 (5) 5 (8) 0 (0)
7 (7) 2 (29) 0 (0) 2 (29) 4 (57) 0 (0) 0 (0) 0 (0) 4 (57) 0 (0) 2 (29) 0 (0)
26 (27) 2 (8) 1 (4) 2 (8) 5 (19) 7 (27) 2 (8) 9 (35) 14 (54) 3 (12) 1 (4) 1 (4)
98 11 (11) 9 (9) 10 (10) 30 (31) 7 (7) 2 (2) 9 (9) 39 (40) 6 (6) 8 (8) 1 (1)
p Value 0.293 0.307 0.103 0.135/0.046†
0.103 0.361 0.103
*Data are presented as No. (%). †Oracle vs oronasal.
[19%]; p ⫽ 0.046) [Table 3]. However, when patients who changed mask types were considered, the proportion of patients who dropped out and/or changed mask type during the course of the study did not differ significantly between groups. There were no significant differences in the proportion of dropouts between the Oracle and nasal groups, nor were there significant differences between the nasal and oronasal groups. Patients who dropped out during the course of the study were found to have a shorter total sleep time during their diagnostic sleep study (dropout group, 5.1 ⫾ 1.5 h; control group, 5.7 ⫾ 1.1 h; p ⫽ 0.044), and a longer sleep latency (dropout group, 28.2 ⫾ 40.1 h; control group, 14.3 ⫾ 15.1 h; p ⫽ 0.023) during their CPAP titration study; however, their disease severity (AHI) did not differ (dropout group, 38.3 ⫾ 28.7/h; control group, 41.5 ⫾ 24.7/h; p ⫽ 0.582). Patients who dropped out prior to CPAP titration did not differ in baseline demographics or diagnostic polysomnography data from those who did not. The proportion of patients who used humidification was similar between those who dropped out and those who continued to use CPAP therapy. The reported reasons for dropping out included an inability to acclimatize to CPAP therapy (43.3%), coexisting illness (10%), worsened sleep (23.3%), difficulty acclimatizing to mask and/or head-
gear (13.3%), and unwillingness to tolerate the inconvenience (10%). There were no significant intergroup differences in terms of missing data unavailable for follow-up (Table 3). At the 6-month follow-up, dropouts and unavailable for follow-up reduced the oronasal group to one patient. As such, comparisons of the 6-month data were made only between the nasal and Oracle groups. Adherence The average reported CPAP usage for the various mask groups is presented in Table 4. There were no significant differences between groups in reported hours of use per night during acclimatization (nasal, 4.95 ⫾ 2.05 h; Oracle, 4.87 ⫾ 2.51 h; oronasal, 4.50 ⫾ 1.87 h; p ⫽ 0.90), although the oronasal group reported using CPAP fewer nights per week during this period (nasal, 5.88 ⫾ 1.72 nights; Oracle, 6.62 ⫾ 0.89 nights; oronasal, 3.80 ⫾ 3.03 nights; p ⫽ 0.01). There were no significant differences in adherence between mask groups at 2 months, nor between the nasal and Oracle groups at 6 months (Table 4). At the 2-month follow-up, 49% of the patients enrolled reported using CPAP for at least 4 h per night and 5 d/wk. At the 6-month follow-up, this fell to 41%.
Table 4 —Adherence Data* Variables
Nasal
Oronasal
Oracle
Overall
p Value
Nights/week, 2 mo Hours/night, 2 mo Nights/week, 6 mo Hours/night, 6 mo Involuntary removal, 2 mo Involuntary removal, 6 mo
5.9 ⫾ 1.9 5.5 ⫾ 1.8 5.9 ⫾ 1.7 5.5 ⫾ 1.8 34/54 (63) 24/48 (50)
4.6 ⫾ 2.9 4.0 ⫾ 2.3 7.0 6.0 3/5 (60) 0/1 (0)
6.1 ⫾ 1.6 5.8 ⫾ 2.0 4.8 ⫾ 2.7 4.8 ⫾ 2.7 10/12 (83) 6/10 (60)
5.9 ⫾ 1.9 5.4 ⫾ 1.9 5.7 ⫾ 1.9 5.4 ⫾ 2.0 47/71 (66) 30/59 (51)
0.314 0.179 0.223 0.423 0.384 0.501
*Data are presented as mean ⫾ SD or No. (%). 2204
Clinical Investigations
When examining those patients who continued to use CPAP at 6 months, it was found that average weekly usage decreased significantly from 2 to 6 months (6.4 ⫾ 1.3 d/wk vs 5.8 ⫾ 2.0 d/wk, p ⫽ 0.014), while average nightly usage remained the same. There were no significant differences between mask groups in reported involuntary mask removal. Patients who received heated humidification reported greater CPAP adherence at the 6-month follow-up in terms of hours per night than those who received cold passover (heated, 5.5 ⫾ 2.1 h; cold passover, 4.5 ⫾ 0.6 h; p ⫽ 0.034).
oronasal, 3.9 ⫾ 1.6; Oracle, 4.5 ⫾ 1.2; p ⫽ 0.594), nor between the nasal and Oracle groups at 6 months (nasal, 4.8 ⫾ 1.1; Oracle, 4.9 ⫾ 1.1; p ⫽ 0.801), nor were there any significant differences over time. Furthermore, there were no significant differences in efficacy ratings at the time of CPAP titration (nasal, 3.7 ⫾ 1.5; oronasal, 3.0 ⫾ 1.4; Oracle, 4.4 ⫾ 1.1; p ⫽ 0.124). However, patients who received heated humidification reported greater subjective ratings of efficacy at the 6-month follow-up than those who received cold-passover humidification (heated, 4.9 ⫾ 1.1; cold passover, 3.8 ⫾ 0.5; p ⫽ 0.010).
Satisfaction There were no significant differences in the comfort ratings between mask groups at 2 months, nor between the nasal and Oracle groups at 6 months following the CPAP titration study (nasal, 6.0 ⫾ 1.9 and 6.6 ⫾ 1.4; oronasal, 5.8 ⫾ 2.4 and 7.0 [single patient]; Oracle, 5.5 ⫾ 1.6 and 6.9 ⫾ 1.3, respectively). When examining those patients who continued to use CPAP at 6 months, it was found that mean comfort ratings increased from 2 to 6 months (6.1 ⫾ 1.6 to 6.8 ⫾ 1.3; p ⫽ 0.003). At 2 months, 55% of patients using CPAP reported upper airway dryness, and the proportion of patients with this complaint tended to be higher in the oronasal and Oracle mask groups (nasal, 46%; oronasal, 80%; Oracle, 83%; p ⫽ 0.033). At 6 months, 47% of the patients who continued to use CPAP reported dryness, and the proportion of patients with this complaint was higher in the Oracle group (nasal, 42%; Oracle, 73%; p ⫽ 0.295), although this did not reach statistical significance. Among patients who reported dryness, the Oracle group reported significantly worse dryness both at 2 months (nasal, 7.8 ⫾ 1.6; oronasal, 5.8 ⫾ 3.7; Oracle, 6.0 ⫾ 1.2; p ⫽ 0.026 nasal vs Oracle) and at 6 months (nasal, 7.2 ⫾ 2.2; Oracle, 5.1 ⫾ 1.8; p ⫽ 0.021). There were no significant differences in dryness ratings over time, and the severity of dryness complaints did not differ between those who dropped out and those who did not, nor were there significant differences in the severity of dryness complaints between dropout and nondropouts within the Oracle group. At both 2 months and 6 months, four patients from the Oracle group complained of “rain-out,” whereby excess condensation accumulated in the CPAP tubing and disturbed their sleep. At 2 months, there were no rain-out complaints from the nasal or oronasal group; at 6 months, one patient from the nasal group complained of rain-out. There were no significant differences in subjective ratings of efficacy between mask groups at 2 months (nasal, 4.5 ⫾ 1.2; www.chestjournal.org
Discussion The Oracle mask is a new and unique interface for the delivery of CPAP therapy to patients with OSA. Our study demonstrated a number of interesting findings. Firstly, 27% of patients chose the Oracle mask over other conventional masks for the initiation of CPAP therapy, and 46% of these patients continued to use the Oracle mask 6 months after their CPAP titration study. Secondly, CPAP therapy was equally effective and efficacious through the Oracle mask as through nasal and oronasal masks in the treatment of sleep apnea, reflected by the reduction in AHI, subjective improvement, and adherence with therapy. Thirdly, the main limiting factor of the Oracle mask was discomfort associated with upper airway dryness and rain-out associated with the use of heated humidification. Finally, the number of patients who persevered with CPAP therapy was significantly higher in those who used the Oracle mask than in users of oronasal masks, suggesting that Oracle masks may be more effective than oronasal masks in patients who cannot rely on exclusive use of the nasal airway. Although CPAP therapy has its limitations, it remains the most definitive way to correct OSA.7 One limitation is the difficulty some patients experience using nasal and oronasal masks, associated with structural or functional abnormalities of the nasal airway.9 An additional limitation is the inconvenience and discomfort associated with the headgear that is required to stabilize nasal and oronasal masks. The Oracle mask addresses both of these problems by offering an interface that exclusively uses the oral airway and that does not require headgear to stabilize it. We enrolled consecutive patients in whom CPAP therapy was indicated following clinical and polysomnographic assessment. We chose CPAP-naı¨ve patients to avoid the effect of previous experience with CPAP confounding our results. The initial appeal of the Oracle mask is reflected by the fact that CHEST / 124 / 6 / DECEMBER, 2003
2205
27% of patients chose it over other conventional masks. However, the fact that 46% of these patients continued to use the Oracle mask 6 months after their CPAP titration, compared to 68% who continued to use their nasal mask (Table 3), indicates that the conditions for continued use have not yet been optimized. Our study design may have introduced bias due to nonrandomization of mask use, and the presentation, acclimatization, and titration of CPAP. We did not randomly use all three mask types in each patient for several reasons. First, we believed that many patients would be unable or unwilling to use all mask types, which would preclude adequate comparison. Second, we believed that allowing patients to use a mask of their choice would be an important motivator and determinant of continued use, which was important in maintaining the feasibility of the study. To this end, we permitted patients to change their mask type if their initial choice did not meet their expectations. Third, we wanted to assess patients’ mask preference following a standardized presentation (as described in “Materials and Methods”). Consequently, we feel that the design we used was appropriate for the questions we wished to address, and that it was a closer reflection of what happens in daily practice. We do not feel that patients’ mask choice was biased by our study design. All patients were acclimatized to CPAP in a standardized fashion. We have analyzed subjective compliance data obtained at the time of CPAP titration, which reflects the acclimatization experience of all patients who used CPAP from the time of their initial set-up up to the time of their CPAP titration study. There were no significant differences between groups in reported efficacy and hours of use per night, although the oronasal group reported using CPAP fewer nights per week (nasal, 5.88 ⫾ 1.72 nights per week; Oracle, 6.62 ⫾ 0.89 nights; oronasal, 3.80 ⫾ 3.03 nights; p ⫽ 0.01). Consequently, we do not believe that the acclimatization procedure was a source of significant bias in our study. We also used a standardized CPAP titration protocol, which determined the CPAP level that abolished flow limitation and was comfortable for the patient. The fact that optimal CPAP level and AHI while receiving optimal CPAP did not differ between mask groups indicates a lack of systematic bias in the CPAP titration studies. CPAP controlled OSA equally well when it was administered through the Oracle mask as when it was administered through conventional nasal or oronasal masks. This finding is consistent with previous work, which found CPAP delivery through the mouth to be as effective as delivery through the nose.10 –11 The optimal CPAP level, determined on 2206
the CPAP titration study, was the same for all three mask groups. This indicates that the mean difference between the acclimatization pressure (5 cm H2O) and the optimal pressure was the same in all groups. Although it is possible that suboptimal pressure during the acclimatization period contributed to the poorer adherence and elevated dropout rate within the oronasal group, these findings do not support this hypothesis. Furthermore, the subjective response to CPAP among patients who continued to use it following their titration study was the same, regardless of whether they used the Oracle mask or a conventional nasal mask. Unfortunately, electronic monitoring of CPAP adherence was not performed. We would have liked to monitor the time spent at the prescribed CPAP level, but this was not available on all CPAP units that were used in the study. Although the units did have a chronometer, we were unable to find a reliable way to record hours of use on each CPAP unit as patients frequently forgot to read the time meter on their unit or neglected to bring their CPAP units to their follow-up visits; we did not have the resources to perform home visits. Ultimately, we believed that the best measurement of CPAP adherence that was available on all patients was to use self-reporting, acknowledging the limitations of that measurement. If there was a systematic error in this measurement, we anticipate that it applied equally to all patient groups. Heated humidification plays an important role in CPAP therapy,12 and it is particularly important in patients who are using the Oracle mask. Although heated humidification was not standardized across groups in terms of the level of humidity that was set, the vast majority of patients in all groups did use heated humidification. This study was not designed to address the efficacy of humidification in different CPAP masks. In fact, we were concerned that using the same level of humidification in all patients would confound the interpretation of our results. We know from previous experience that oral masks require a higher level of humidification than nasal masks; consequently, limiting the level of humidification to patients who chose the Oracle mask could have biased our results by increasing the level of discomfort due to suboptimal humidification alone. Similarly, excessive humidification may have produced rain-out in some patients (regardless of the mask they chose), which could have limited their use of CPAP for reasons not specifically related to the mask they chose. We wanted to assess the efficacy of different masks assuming that comfort was optimized by all conventional means including adequate humidification. We also feel this provides a more accurate reflection of patients’ experience in everyClinical Investigations
day practice, and consequently patients were given the opportunity to titrate their level of humidification to optimize comfort. Upper airway dryness and rain-out associated with heated humidification are the main factors that limit the use of the Oracle mask for CPAP therapy. Complaints of upper airway dryness tended to be more prevalent throughout the study in patients who used the Oracle mask. Furthermore, among those who complained of dryness, the severity was significantly greater among patients who used the Oracle mask than those who used other masks. It is possible that further modifications in mask design will reduce the prevalence and severity of this complaint. Interestingly, the severity of upper airway dryness did not differ between patients who discontinued CPAP and those who persevered with it, which was also true for the Oracle group alone, which suggests that patients discontinued CPAP therapy for reasons other than upper airway dryness. The treatment of upper airway dryness with heated humidification may cause excessive condensation in the CPAP tubing (termed rain-out), resulting in discomfort and sleep disruption. Rain-out was more common in patients who used the Oracle mask, which is most likely due to a requirement for higher humidity with this mask. Unfortunately, our study did not record the level of humidification that patients used, which would address this possibility. The recent addition of a thermostat to heated humidifiers on CPAP units reduces the heater temperature as room temperature falls, which has been shown to reduce the degree of rain-out at nighttime. Further studies are required to determine what role these modifications will play in adherence to CPAP therapy, and whether overall outcome is improved in patients receiving oral CPAP. Until recently, patients who cannot tolerate nasal CPAP were obliged to use an oronasal mask, which is often uncomfortable and technically difficult to use.6,9 In our study, the proportion of patients who stopped using CPAP (dropouts) was significantly higher among those who chose an oronasal mask than those who opted for the Oracle mask (Table 3). Notwithstanding the fact that the difference was no longer significant when patients who changed from the Oracle mask to another type of mask were included in the evaluation (Table 3), this raises the possibility that the Oracle mask may have some advantages over oronasal masks in this patient population. Future studies should address whether the Oracle mask is a more efficacious interface for patients who have occluded nasal airways or who require more sophisticated ventilatory support. The www.chestjournal.org
reason for the higher dropout rate in the oronasal group is not entirely clear. However, previous work by others13 has suggested that oronasal masks are less effective than nasal masks in overcoming the dynamic forces responsible for airway closure in patients with OSA. In summary, the Oracle mask was an efficacious interface for long-term CPAP therapy in approximately 20% of patients with clinically significant OSA. The anticipated problem with upper airway dryness was adequately controlled in most patients by heated humidification, although excessive humidification (rain-out) was observed in some patients. The Oracle mask may be more acceptable than oronasal masks for patients who cannot rely exclusively on the nasal airway for CPAP therapy. Appendix: CPAP Questionnaire Please answer the following questions based on your usage of CPAP during the last 3 weeks: 1. On average, how many nights per week do you use CPAP? nights 2. On average, how many hours per night do you use CPAP? hours 3. Are you taking your mask off during the night? Yes or No If Yes: (a) Does this happen every night? Yes or No (b) When you wake, do you put your interface back on? Yes or No (c) When you sleep, is the length of time you keep your mask on slowly increasing? Yes or No 4. On a scale of 1 to 10, how comfortable do you find your interface? (1 ⫽ unusable; 6 ⫽ comfortable; 10 ⫽ so comfortable that you do not realize that it is on) Oracle mask Nasal mask 5. Are you experiencing dryness in your nose or mouth when waking after using your nasal CPAP? Yes or No On a scale of 1 to 10, rate the dryness in your nose or mouth. (1 ⫽ so dry you cannot use nasal CPAP; 6 ⫽ uncomfortable but manageable; 10 ⫽ noticeable but not uncomfortable) 6. On a scale of 0 to 6, please rate how you feel now compared to before you started using CPAP. (0 ⫽ worse [feel worse than before]; 3 ⫽ same [feel the same as before]; 6 ⫽ better [feel much better than before]).
References 1 Young T, Palta M, Dempsey J, et al. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993; 328:1230 –1235 2 Peppard PE, Young T, Palta M, et al. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000; 342:178 –184 3 Sullivan CE, Issa FG, Berthon-Jones M, et al. Reversal of obstructive sleep apnoea by continuous positive airway pressure applied through the nares. Lancet 1981; 18:862– 865 4 Aldrich MS. Obstructive sleep apnea syndrome. In: Sleep medicine. New York, NY: Oxford University Press, 1999; 227 5 Martins de Araujo MT, Vieira SB, Vasquez EC, et al. Heated CHEST / 124 / 6 / DECEMBER, 2003
2207
6 7
8
9
humidification or face mask to prevent upper airway dryness during continuous positive airway pressure therapy. Chest 2000; 117:142–147 Mortimore IL, Whittle AT, Douglas NJ. Comparison of nose and face mask CPAP therapy for sleep apnea. Thorax 1998; 53:290 –292 Sanders MH, Strollo PJ, Stiller RA. Positive airway pressure in the treatment of sleep-related breathing disorders. In: Chokroverty S, ed. Sleep disorders medicine: basic science, technical considerations, and clinical aspects. Boston, MA: Butterworth-Heinemann, 1999; 355–377 Rechtschaffen A, Kales A, eds. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Los Angeles, CA: Brain Information Service/Brain Research Institute, 1968; 1– 60 Prosise GL, Berry RB. Oral-nasal continuous positive airway
2208
10
11
12
13
pressure as a treatment for obstructive sleep apnea. Chest 1994; 106:180 –186 Anderson FE, Smith NC, Kingshott RN, et al. Efficacy of oral CPAP in the management of obstructive sleep apnea [abstract]. Sleep 2002; 25:A220 Pedler H, Whyte KF, Smith NC, et al. Efficacy of an oral interface for the delivery of CPAP therapy in the treatment of OSA [abstract]. Sleep 2001; 24:A266 Massie CA, Hart RW, Peralez K, et al. Effects of humidification on nasal symptoms and compliance in sleep apnea patients using continuous positive airway pressure. Chest 1999; 116:403– 408 Smith PL, Wise RA, Gold AR, et al. Upper airway pressure flow relationships in obstructive sleep apnea. J Appl Physiol 1988; 64:789 –795
Clinical Investigations
Background: Nasal continuous positive airway pressure (CPAP) is the most definitive medical therapy for obstructive sleep apnea (OSA). Many patients have difficulty tolerating nasal CPAP due to nasal airway problems, mouth leak, and general discomfort from the mask and headgear. These limitations may be overcome by an oral mask (Oracle; Fisher & Paykel Healthcare; Languna Hills, CA) that does not require headgear. We performed a study to compare the Oracle mask to conventional nasal and oronasal masks in the effectiveness of CPAP delivery and patient satisfaction and adherence. Methods: Ninety-eight, consecutive CPAP-naı¨ve patients with OSA diagnosed by overnight polysomnography (apnea-hypopnea index [AHI] > 5) were referred for CPAP therapy. All patients were presented with a variety of CPAP masks, including nasal, oronasal, and Oracle, and reasons for mask choice were documented. After 3 weeks of acclimatization to the mask of their choice, patients had a CPAP titration sleep study to determine their optimal CPAP level. Further follow-up was obtained 2 months and 6 months later with a subjective patient assessment of CPAP use and efficacy, mask comfort, and upper airway dryness. Results: Patients were predominantly male (70%), middle aged (50.6 ⴞ 11.7 years), and moderately obese (body mass index, 32.5 ⴞ 9.0) with severe OSA (AHI, 40.6 ⴞ 25.8/h) [mean ⴞ SD]. Patients were classified into three groups based on their choice of mask: nasal (66%), Oracle (27%), and oronasal (7%). Baseline characteristics did not differ significantly between groups. Optimal CPAP was not significantly different between mask groups (nasal, 7.7 ⴞ 2.1 cm H2O; Oracle, 8.0 ⴞ 2.0 cm H2O; oronasal, 9.7 ⴞ 3.2 cm H2O; p ⴝ 0.267). Subjective ratings of adherence, efficacy, and mask comfort were also similar between groups. However, the Oracle group had more complaints of upper airway dryness and “rain-out.” The oronasal group had a disproportionately greater number of dropouts from CPAP therapy than the Oracle group (57% vs 19%, p ⴝ 0.046). Nine patients changed from the Oracle mask to a nasal mask during the study, whereas no patients changed from their nasal or oronasal masks. Conclusions: The Oracle mask is an efficacious interface for long-term CPAP therapy in patients with OSA. The main limitations of the mask are upper airway dryness and rain-out associated with heated humidification, which may be improved by further technical modifications. Oracle may be more acceptable than oronasal masks for patients who cannot rely exclusively on the nasal airway for CPAP therapy. (CHEST 2003; 124:2200 –2208) Key words: adherence; oral continuous positive airway pressure; sleep apnea Abbreviations: AHI ⫽ apnea-hypopnea index; CPAP ⫽ continuous positive airway pressure; OSA ⫽ obstructive sleep apnea
sleep apnea (OSA) is a common O bstructive condition that is associated with excessive daytime sleepiness and increased risk of cardiovascular
disease, particularly hypertension.2 The medical therapy of choice for OSA is the application of continuous positive airway pressure (CPAP), which
*From the Sleep Disorders Clinic (Mr. Beecroft, Mr. Lukic, and Dr. Hanly), St. Michael’s Hospital, University of Toronto, Toronto; and Medigas/Praxair (Ms. Zanon), W Toronto, ON, Canada. This research was performed at Sleep Disorders Clinic, St. Michael’s Hospital, University of Toronto, and Medigas/Praxair. Supported by Fisher & Paykel Healthcare.
Manuscript received January 8, 2003; revision accepted July 1, 2003. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: [email protected]). Correspondence to: Patrick J. Hanly, MD, 6049 Bond, St. Michael’s Hospital, 30 Bond St, Toronto, ON, Canada M5B 1W8; e-mail: [email protected]
1
2200
Clinical Investigations
prevents upper airway closure by creating a “pneumatic splint.”3 Although CPAP therapy is effective, patients frequently stop using it because of discomfort predominantly associated with the mask.4 The conventional method for administration of CPAP therapy is through a nasal or oronasal interface. While oronasal masks have been reported to overcome upper airway dryness associated with nasal masks,5 they have been rated as less comfortable and have been associated with a greater rate of noncompliance and less improvement in subjective sleepiness than nasal masks.6 However, nasal masks have been associated with more symptoms related to mouth leak, such as dry throat and mouth,6 and have been associated with other complaints such as nasal dryness, nasal congestion, and rhinorrhea.7 An alternative approach is to deliver CPAP exclusively through the mouth. The Oracle (Fisher & Paykel Healthcare; Languna Hills, CA) is an oral CPAP mask that avoids nasal airway problems and does not require headgear. The objective of this study was to determine how often patients chose the Oracle mask over the current conventional CPAP masks, and how their adherence and satisfaction with the Oracle mask compared to that of nasal and oronasal masks.
Materials and Methods Patient Recruitment All patients with OSA diagnosed at St. Michael’s Hospital Sleep Laboratory were considered potential candidates for this study. Patients referred for assessment of sleep apnea underwent diagnostic polysomnography, and those with an apnea-hypopnea index (AHI) ⬎ 5/h were seen for clinical follow-up at the ambulatory sleep clinic. Following clinical assessment, specific management of sleep apnea was discussed. Patients in whom CPAP therapy was indicated were referred to a home-care company that provided this treatment. None of these patients had previously received CPAP. Patients received nondirected, standardized counseling on the use of CPAP. At the home-care company, all patients were seen for approximately 1 h by the same individual, an experienced registered polysomnographic technologist (S.Z.). Each patient was shown the same range of masks, which consisted of the Oracle mask, nasal masks, and oronasal masks. The features of each mask type were reviewed with the patient. They were assisted to try on as many masks as they wished and asked to choose one to take home. In addition, they were placed on 5 cm H2O CPAP through the mask of their choice for a few minutes, and any questions or concerns were addressed. The reasons for patients’ choice of mask were documented. Patients were provided with a CPAP unit of their choice. Those who chose the Oracle mask were given heated humidification. Those who chose a nasal mask or oronasal mask were given whatever level of humidity was required for their comfort (cold passover, heated, or none). All patients were then acclimatized on a CPAP level of 5 cm H2O at home for 3 weeks before returning to the laboratory for a CPAP titration study. www.chestjournal.org
Oracle Mask The Oracle oral mask is a strapless, butterfly-shaped interface that rests in the oral vestibule between the lips and teeth (Fig 1). It is comprised of a rigid inner framework, which depresses the tongue, a soft seal, which sits between the lips and teeth, an adjustable flap, which seals around the mouth on the skin surface, and tubing, which attaches to the CPAP unit. The features of the mouthpiece ensure the desired positive airway pressure is delivered to the patient with minimal leakage, and that the mouthpiece is retained in the mouth while asleep. The flexible tubing facilitates freedom of movement while maintaining circuit integrity, and contains an exhaust port adjacent to the mouthpiece, which provides a means to purge exhaled gases from the breathing circuit. Polysomnography Diagnostic overnight polysomnography was performed in a standardized fashion in the Sleep Laboratory at St. Michael’s Hospital. Recordings were performed by continuous monitoring of the EEG, electrooculogram, and submental electromyogram, ECG, nasal airflow (Ultima Dual Airflow Pressure Sensor; Braebon Medical Corporation; Kanata, ON, Canada), chest and abdominal respiratory movements (Respitrace; Ambulatory Monitoring; Ardsley, NY), oximetry (Mallinckrodt/Nellcor Puritan Bennett; Hazelwood, MO), anterior tibialis electromyogram, and body position. The recordings were performed and scored by experienced, registered polysomnographic technologists. Scoring of the polysomnographic data was done in a standardized fashion, according to published criteria.8 At the time of the diagnostic polysomnography, patients were required to complete a sleep history questionnaire and Epworth sleepiness scale, as well as provide demographic information. Following CPAP orientation and acclimatization, all patients underwent a CPAP titration study at the sleep laboratory 3 weeks after they received their CPAP mask and unit. All patients were instructed to bring their own mask, which they wore for the duration of the study. Recordings were performed in a similar manner to the diagnostic study, except that the patients were monitored on CPAP and the pressure was titrated to a level that resolved obstructive respiratory events and was comfortable (termed optimal CPAP). Patients also completed a CPAP questionnaire (Appendix). The CPAP units were set at this level the morning after their study. Follow-up One week following commencement of CPAP therapy, patients were contacted by the home-care company to address problems and record their adherence to therapy. If patients were unable to tolerate either the Oracle mask or other mask of their choice, they were offered the option of switching to an alternative mask. Reasons for mask preference were documented. One week following the CPAP titration study, a second telephone follow-up was performed by the home-care company to address problems and record adherence. Subsequently, patients were contacted by telephone at 2 months and 6 months following commencement of CPAP therapy. At these times, patients completed a CPAP questionnaire designed to give a subjective assessment of CPAP usage and efficacy, mask comfort, and upper airway dryness. Adherence was measured as the average number of hours per day and days per week the patient reported using CPAP. Patients were asked to rate the comfort of their mask on a 10-point scale. Within this scale, a rating of 1 indicated the mask was so uncomfortable it was unusable, while a 10 indicated the mask was so comfortable they CHEST / 124 / 6 / DECEMBER, 2003
2201
Figure 1. Oracle CPAP mask.
did not realize it was on. Patients were also asked to rate upper airway dryness. Patients were asked whether they experienced dryness in their nose or mouth when waking after using CPAP and, if so, instructed to rate the dryness on a 10-point scale. Within this scale, a rating of 1 indicated the dryness was so bad they could not use CPAP, while a rating of 10 indicated the dryness was noticeable but not uncomfortable. Efficacy was rated on a scale that ranged from 0 to 6, with 0 indicating that the patient felt worse than prior to using CPAP, and 6 indicating that the patient felt much better than prior to using CPAP. Analysis Prevalence of mask choice and reasons for mask preference were analyzed, and polysomnography data and patient satisfaction/adherence were compared for the nasal, oronasal, and Oracle mask groups. Furthermore, dropouts from CPAP therapy and changes in mask preference were compared between the nasal, oronasal, and Oracle mask groups. The means of data among these three groups were compared using one-way analysis of variance and post hoc Tukey tests to determine significant differences. Two-month and 6-month data were compared using paired Student t test, and gender differences were examined using unpaired Student t test. Ordinal and binary variables were compared using a 2 test. All p values ⬍ 0.05 were considered statistically significant. 2202
Results Study Cohort Ninety-eight patients with OSA were recruited from St. Michael’s Hospital Sleep Clinic over a 6-month period and referred for CPAP therapy. Patients were classified into three groups according to whether they chose a nasal, oronasal, or Oracle mask. The baseline demographics (Table 1) and polysomnography data (Table 2) of the three groups were similar. The “optimal CPAP” level, as defined in the “Materials and Methods,” was not significantly
Table 1—Patient Demographics* Variables
Nasal
Oronasal
Oracle
Patients, No. 65 7 26 Male patients, % 69.2 71.4 73.1 Age, yr 49.6 ⫾ 10.2 53.0 ⫾ 19.0 53.1 ⫾ 13.2 Body mass index 32.2 ⫾ 8.7 38.6 ⫾ 17.4 31.7 ⫾ 6.8
p Value 0.934 0.385 0.231
*Data are presented as mean ⫾ SD unless otherwise indicated. Clinical Investigations
Table 2—Baseline Polysomnography* Variables
Nasal
Oronasal
Oracle
p Value
Total sleep time, h Sleep efficiency, %† Sleep latency, min‡ Stage 1 NREM, %TST Stage 2 NREM, %TST SWS, %TST REM, %TST AHI, /h Mean Sao2, % Periodic leg movements, /h Arousal index, /h Epworth sleepiness scale
5.6 ⫾ 1.2 79.0 ⫾ 13.7 14.1 ⫾ 14.4 9.1 ⫾ 7.5 60.9 ⫾ 11.1 13.4 ⫾ 9.0 16.5 ⫾ 7.3 39.9 ⫾ 25.8 93.7 ⫾ 2.9 9.3 ⫾ 15.7 42.1 ⫾ 24.1 10.0 ⫾ 5.4
5.0 ⫾ 2.0 71.1 ⫾ 23.9 8.5 ⫾ 6.5 13.7 ⫾ 8.7 60.6 ⫾ 11.4 6.1 ⫾ 9.8 19.5 ⫾ 9.6 43.4 ⫾ 40.7 91.2 ⫾ 3.8 30.1 ⫾ 38.3 42.0 ⫾ 42.9 9.8 ⫾ 4.0
5.6 ⫾ 1.1 78.0 ⫾ 13.0 14.6 ⫾ 17.5 10.2 ⫾ 6.7 61.2 ⫾ 10.6 11.7 ⫾ 7.8 19.5 ⫾ 9.6 41.6 ⫾ 22.6 93.9 ⫾ 2.8 9.7 ⫾ 25.1 39.5 ⫾ 23.0 11.4 ⫾ 5.5
0.527 0.437 0.656 0.343 0.988 0.132 0.629 0.924 0.113 0.059 0.904 0.528
*Data are presented as mean ⫾ SD. NREM ⫽ non-rapid eye movement sleep; SWS ⫽ slow-wave sleep; REM ⫽ rapid eye movement sleep; %TST ⫽ percentage of total sleep time; Sao2 ⫽ oxygen saturation. †Total sleep time expressed as a proportion of total study duration. ‡Time from lights out to the first epoch of sleep.
different between mask groups (nasal, 7.7 ⫾ 2.1 cm H2O; oronasal, 9.7 ⫾ 3.2 cm H2O; Oracle, 8.0 ⫾ 2.0 cm H2O; p ⫽ 0.267). On optimal CPAP, there were no significant intergroup differences in AHI (nasal, 6.7 ⫾ 13.3/h; oronasal, 9.8 ⫾ 12.8/h; Oracle, 10.9 ⫾ 20.1/h; p ⫽ 0.555). In addition, the percentage of patients in each group whose AHI was ⬍ 5/h while on their optimal CPAP did not differ significantly (nasal, 61%; oronasal, 67%; Oracle, 39%; p ⫽ 0.185). Furthermore, when the Oracle group was compared individually to both the nasal and oronasal groups, no significant differences were found. Of the 98 patients enrolled in the study, 88 patients (90%) received heated humidification, while 9 patients (9%) received cold-passover humidification, and 1 patient (1%) chose not to use humidification. The proportion of patients who received heated humidification was similar between groups (nasal, 85%; oronasal, 100%; Oracle, 100%; p ⫽ 0.059).
found them to be the least invasive (51%); that nasal masks did not produce the sensation of claustrophobia, choking, or gagging they attributed to the Oracle mask (18%); that they were nasal breathers (14%); and that they found nasal masks to be generally more comfortable, to fit better, and anticipated less dryness than with other masks (17%). Fifty-seven percent of patients who chose an oronasal mask reported they were mouth breathers who thought the Oracle mask was too invasive or uncomfortable. During the course of the study, seven patients who began CPAP therapy using the Oracle mask changed to a nasal mask at the 2-month follow-up, and an additional two patients changed from the Oracle mask to a nasal mask at the 6-month follow-up. Patients who chose either nasal or oronasal masks did not change mask types during the course of the study (Table 3). Patient Follow-up
Mask Preference Although the majority of patients chose nasal masks, which included Breeze (Mallinckrodt/Nellcor Puritan Bennett), Contour Deluxe, Profile Lite, Simplicity, and Comfort Classic (Respironics; Murrysville, PA), 27% chose the Oracle mask and 7% chose an oronasal mask (Full Face Mirage; ResMed; Poway, CA) [Table 1]. There were no significant gender differences between groups in the choice of mask. Patients who chose the Oracle mask reported an appreciation for the lack of headgear (69%), a history of mouth breathing at night (15%), greater comfort, a sensation of “getting more air” compared to other masks, and a concern that a nasal mask would aggravate preexisting chronic sinusitis (16%). Patients who chose nasal masks reported that they www.chestjournal.org
Patient dropout rates were monitored at the time of the CPAP titration study, at 2 months and at 6 months after CPAP was initiated (Table 3). Of the 98 patients enrolled, 11 patients dropped out prior to undergoing their CPAP titration study, an additional 9 dropped out at the 2-month follow-up, and an additional 10 patients dropped out at 6 months. The proportion of dropouts at these various follow-up points did not differ between groups (Table 3). An additional two patients did not return for their CPAP titration study, but reported that they were using CPAP at their 2 month follow-up. At 6 months, a total of 30 patients had discontinued CPAP use, with a disproportionately greater number of patients within the oronasal group than the Oracle group (oronasal, four patients [57%]; Oracle, five patients CHEST / 124 / 6 / DECEMBER, 2003
2203
Table 3—Enrollment and Follow-up of Patients*
Enrolled Dropped out prior to CPAP titration Dropped out at 2-mo follow-up Dropped out at 6-mo follow-up Total dropped out Changed mask at 2 mo Changed mask at 6 mo Total changed Total changed and dropped out 2-mo data not obtained 6-mo data not obtained Unavailable for follow-up
Nasal
Oronasal
Oracle
Total
65 (66) 7 (11) 8 (12) 6 (9) 21 (32) 0 (0) 0 (0) 0 (0) 21 (32) 3 (5) 5 (8) 0 (0)
7 (7) 2 (29) 0 (0) 2 (29) 4 (57) 0 (0) 0 (0) 0 (0) 4 (57) 0 (0) 2 (29) 0 (0)
26 (27) 2 (8) 1 (4) 2 (8) 5 (19) 7 (27) 2 (8) 9 (35) 14 (54) 3 (12) 1 (4) 1 (4)
98 11 (11) 9 (9) 10 (10) 30 (31) 7 (7) 2 (2) 9 (9) 39 (40) 6 (6) 8 (8) 1 (1)
p Value 0.293 0.307 0.103 0.135/0.046†
0.103 0.361 0.103
*Data are presented as No. (%). †Oracle vs oronasal.
[19%]; p ⫽ 0.046) [Table 3]. However, when patients who changed mask types were considered, the proportion of patients who dropped out and/or changed mask type during the course of the study did not differ significantly between groups. There were no significant differences in the proportion of dropouts between the Oracle and nasal groups, nor were there significant differences between the nasal and oronasal groups. Patients who dropped out during the course of the study were found to have a shorter total sleep time during their diagnostic sleep study (dropout group, 5.1 ⫾ 1.5 h; control group, 5.7 ⫾ 1.1 h; p ⫽ 0.044), and a longer sleep latency (dropout group, 28.2 ⫾ 40.1 h; control group, 14.3 ⫾ 15.1 h; p ⫽ 0.023) during their CPAP titration study; however, their disease severity (AHI) did not differ (dropout group, 38.3 ⫾ 28.7/h; control group, 41.5 ⫾ 24.7/h; p ⫽ 0.582). Patients who dropped out prior to CPAP titration did not differ in baseline demographics or diagnostic polysomnography data from those who did not. The proportion of patients who used humidification was similar between those who dropped out and those who continued to use CPAP therapy. The reported reasons for dropping out included an inability to acclimatize to CPAP therapy (43.3%), coexisting illness (10%), worsened sleep (23.3%), difficulty acclimatizing to mask and/or head-
gear (13.3%), and unwillingness to tolerate the inconvenience (10%). There were no significant intergroup differences in terms of missing data unavailable for follow-up (Table 3). At the 6-month follow-up, dropouts and unavailable for follow-up reduced the oronasal group to one patient. As such, comparisons of the 6-month data were made only between the nasal and Oracle groups. Adherence The average reported CPAP usage for the various mask groups is presented in Table 4. There were no significant differences between groups in reported hours of use per night during acclimatization (nasal, 4.95 ⫾ 2.05 h; Oracle, 4.87 ⫾ 2.51 h; oronasal, 4.50 ⫾ 1.87 h; p ⫽ 0.90), although the oronasal group reported using CPAP fewer nights per week during this period (nasal, 5.88 ⫾ 1.72 nights; Oracle, 6.62 ⫾ 0.89 nights; oronasal, 3.80 ⫾ 3.03 nights; p ⫽ 0.01). There were no significant differences in adherence between mask groups at 2 months, nor between the nasal and Oracle groups at 6 months (Table 4). At the 2-month follow-up, 49% of the patients enrolled reported using CPAP for at least 4 h per night and 5 d/wk. At the 6-month follow-up, this fell to 41%.
Table 4 —Adherence Data* Variables
Nasal
Oronasal
Oracle
Overall
p Value
Nights/week, 2 mo Hours/night, 2 mo Nights/week, 6 mo Hours/night, 6 mo Involuntary removal, 2 mo Involuntary removal, 6 mo
5.9 ⫾ 1.9 5.5 ⫾ 1.8 5.9 ⫾ 1.7 5.5 ⫾ 1.8 34/54 (63) 24/48 (50)
4.6 ⫾ 2.9 4.0 ⫾ 2.3 7.0 6.0 3/5 (60) 0/1 (0)
6.1 ⫾ 1.6 5.8 ⫾ 2.0 4.8 ⫾ 2.7 4.8 ⫾ 2.7 10/12 (83) 6/10 (60)
5.9 ⫾ 1.9 5.4 ⫾ 1.9 5.7 ⫾ 1.9 5.4 ⫾ 2.0 47/71 (66) 30/59 (51)
0.314 0.179 0.223 0.423 0.384 0.501
*Data are presented as mean ⫾ SD or No. (%). 2204
Clinical Investigations
When examining those patients who continued to use CPAP at 6 months, it was found that average weekly usage decreased significantly from 2 to 6 months (6.4 ⫾ 1.3 d/wk vs 5.8 ⫾ 2.0 d/wk, p ⫽ 0.014), while average nightly usage remained the same. There were no significant differences between mask groups in reported involuntary mask removal. Patients who received heated humidification reported greater CPAP adherence at the 6-month follow-up in terms of hours per night than those who received cold passover (heated, 5.5 ⫾ 2.1 h; cold passover, 4.5 ⫾ 0.6 h; p ⫽ 0.034).
oronasal, 3.9 ⫾ 1.6; Oracle, 4.5 ⫾ 1.2; p ⫽ 0.594), nor between the nasal and Oracle groups at 6 months (nasal, 4.8 ⫾ 1.1; Oracle, 4.9 ⫾ 1.1; p ⫽ 0.801), nor were there any significant differences over time. Furthermore, there were no significant differences in efficacy ratings at the time of CPAP titration (nasal, 3.7 ⫾ 1.5; oronasal, 3.0 ⫾ 1.4; Oracle, 4.4 ⫾ 1.1; p ⫽ 0.124). However, patients who received heated humidification reported greater subjective ratings of efficacy at the 6-month follow-up than those who received cold-passover humidification (heated, 4.9 ⫾ 1.1; cold passover, 3.8 ⫾ 0.5; p ⫽ 0.010).
Satisfaction There were no significant differences in the comfort ratings between mask groups at 2 months, nor between the nasal and Oracle groups at 6 months following the CPAP titration study (nasal, 6.0 ⫾ 1.9 and 6.6 ⫾ 1.4; oronasal, 5.8 ⫾ 2.4 and 7.0 [single patient]; Oracle, 5.5 ⫾ 1.6 and 6.9 ⫾ 1.3, respectively). When examining those patients who continued to use CPAP at 6 months, it was found that mean comfort ratings increased from 2 to 6 months (6.1 ⫾ 1.6 to 6.8 ⫾ 1.3; p ⫽ 0.003). At 2 months, 55% of patients using CPAP reported upper airway dryness, and the proportion of patients with this complaint tended to be higher in the oronasal and Oracle mask groups (nasal, 46%; oronasal, 80%; Oracle, 83%; p ⫽ 0.033). At 6 months, 47% of the patients who continued to use CPAP reported dryness, and the proportion of patients with this complaint was higher in the Oracle group (nasal, 42%; Oracle, 73%; p ⫽ 0.295), although this did not reach statistical significance. Among patients who reported dryness, the Oracle group reported significantly worse dryness both at 2 months (nasal, 7.8 ⫾ 1.6; oronasal, 5.8 ⫾ 3.7; Oracle, 6.0 ⫾ 1.2; p ⫽ 0.026 nasal vs Oracle) and at 6 months (nasal, 7.2 ⫾ 2.2; Oracle, 5.1 ⫾ 1.8; p ⫽ 0.021). There were no significant differences in dryness ratings over time, and the severity of dryness complaints did not differ between those who dropped out and those who did not, nor were there significant differences in the severity of dryness complaints between dropout and nondropouts within the Oracle group. At both 2 months and 6 months, four patients from the Oracle group complained of “rain-out,” whereby excess condensation accumulated in the CPAP tubing and disturbed their sleep. At 2 months, there were no rain-out complaints from the nasal or oronasal group; at 6 months, one patient from the nasal group complained of rain-out. There were no significant differences in subjective ratings of efficacy between mask groups at 2 months (nasal, 4.5 ⫾ 1.2; www.chestjournal.org
Discussion The Oracle mask is a new and unique interface for the delivery of CPAP therapy to patients with OSA. Our study demonstrated a number of interesting findings. Firstly, 27% of patients chose the Oracle mask over other conventional masks for the initiation of CPAP therapy, and 46% of these patients continued to use the Oracle mask 6 months after their CPAP titration study. Secondly, CPAP therapy was equally effective and efficacious through the Oracle mask as through nasal and oronasal masks in the treatment of sleep apnea, reflected by the reduction in AHI, subjective improvement, and adherence with therapy. Thirdly, the main limiting factor of the Oracle mask was discomfort associated with upper airway dryness and rain-out associated with the use of heated humidification. Finally, the number of patients who persevered with CPAP therapy was significantly higher in those who used the Oracle mask than in users of oronasal masks, suggesting that Oracle masks may be more effective than oronasal masks in patients who cannot rely on exclusive use of the nasal airway. Although CPAP therapy has its limitations, it remains the most definitive way to correct OSA.7 One limitation is the difficulty some patients experience using nasal and oronasal masks, associated with structural or functional abnormalities of the nasal airway.9 An additional limitation is the inconvenience and discomfort associated with the headgear that is required to stabilize nasal and oronasal masks. The Oracle mask addresses both of these problems by offering an interface that exclusively uses the oral airway and that does not require headgear to stabilize it. We enrolled consecutive patients in whom CPAP therapy was indicated following clinical and polysomnographic assessment. We chose CPAP-naı¨ve patients to avoid the effect of previous experience with CPAP confounding our results. The initial appeal of the Oracle mask is reflected by the fact that CHEST / 124 / 6 / DECEMBER, 2003
2205
27% of patients chose it over other conventional masks. However, the fact that 46% of these patients continued to use the Oracle mask 6 months after their CPAP titration, compared to 68% who continued to use their nasal mask (Table 3), indicates that the conditions for continued use have not yet been optimized. Our study design may have introduced bias due to nonrandomization of mask use, and the presentation, acclimatization, and titration of CPAP. We did not randomly use all three mask types in each patient for several reasons. First, we believed that many patients would be unable or unwilling to use all mask types, which would preclude adequate comparison. Second, we believed that allowing patients to use a mask of their choice would be an important motivator and determinant of continued use, which was important in maintaining the feasibility of the study. To this end, we permitted patients to change their mask type if their initial choice did not meet their expectations. Third, we wanted to assess patients’ mask preference following a standardized presentation (as described in “Materials and Methods”). Consequently, we feel that the design we used was appropriate for the questions we wished to address, and that it was a closer reflection of what happens in daily practice. We do not feel that patients’ mask choice was biased by our study design. All patients were acclimatized to CPAP in a standardized fashion. We have analyzed subjective compliance data obtained at the time of CPAP titration, which reflects the acclimatization experience of all patients who used CPAP from the time of their initial set-up up to the time of their CPAP titration study. There were no significant differences between groups in reported efficacy and hours of use per night, although the oronasal group reported using CPAP fewer nights per week (nasal, 5.88 ⫾ 1.72 nights per week; Oracle, 6.62 ⫾ 0.89 nights; oronasal, 3.80 ⫾ 3.03 nights; p ⫽ 0.01). Consequently, we do not believe that the acclimatization procedure was a source of significant bias in our study. We also used a standardized CPAP titration protocol, which determined the CPAP level that abolished flow limitation and was comfortable for the patient. The fact that optimal CPAP level and AHI while receiving optimal CPAP did not differ between mask groups indicates a lack of systematic bias in the CPAP titration studies. CPAP controlled OSA equally well when it was administered through the Oracle mask as when it was administered through conventional nasal or oronasal masks. This finding is consistent with previous work, which found CPAP delivery through the mouth to be as effective as delivery through the nose.10 –11 The optimal CPAP level, determined on 2206
the CPAP titration study, was the same for all three mask groups. This indicates that the mean difference between the acclimatization pressure (5 cm H2O) and the optimal pressure was the same in all groups. Although it is possible that suboptimal pressure during the acclimatization period contributed to the poorer adherence and elevated dropout rate within the oronasal group, these findings do not support this hypothesis. Furthermore, the subjective response to CPAP among patients who continued to use it following their titration study was the same, regardless of whether they used the Oracle mask or a conventional nasal mask. Unfortunately, electronic monitoring of CPAP adherence was not performed. We would have liked to monitor the time spent at the prescribed CPAP level, but this was not available on all CPAP units that were used in the study. Although the units did have a chronometer, we were unable to find a reliable way to record hours of use on each CPAP unit as patients frequently forgot to read the time meter on their unit or neglected to bring their CPAP units to their follow-up visits; we did not have the resources to perform home visits. Ultimately, we believed that the best measurement of CPAP adherence that was available on all patients was to use self-reporting, acknowledging the limitations of that measurement. If there was a systematic error in this measurement, we anticipate that it applied equally to all patient groups. Heated humidification plays an important role in CPAP therapy,12 and it is particularly important in patients who are using the Oracle mask. Although heated humidification was not standardized across groups in terms of the level of humidity that was set, the vast majority of patients in all groups did use heated humidification. This study was not designed to address the efficacy of humidification in different CPAP masks. In fact, we were concerned that using the same level of humidification in all patients would confound the interpretation of our results. We know from previous experience that oral masks require a higher level of humidification than nasal masks; consequently, limiting the level of humidification to patients who chose the Oracle mask could have biased our results by increasing the level of discomfort due to suboptimal humidification alone. Similarly, excessive humidification may have produced rain-out in some patients (regardless of the mask they chose), which could have limited their use of CPAP for reasons not specifically related to the mask they chose. We wanted to assess the efficacy of different masks assuming that comfort was optimized by all conventional means including adequate humidification. We also feel this provides a more accurate reflection of patients’ experience in everyClinical Investigations
day practice, and consequently patients were given the opportunity to titrate their level of humidification to optimize comfort. Upper airway dryness and rain-out associated with heated humidification are the main factors that limit the use of the Oracle mask for CPAP therapy. Complaints of upper airway dryness tended to be more prevalent throughout the study in patients who used the Oracle mask. Furthermore, among those who complained of dryness, the severity was significantly greater among patients who used the Oracle mask than those who used other masks. It is possible that further modifications in mask design will reduce the prevalence and severity of this complaint. Interestingly, the severity of upper airway dryness did not differ between patients who discontinued CPAP and those who persevered with it, which was also true for the Oracle group alone, which suggests that patients discontinued CPAP therapy for reasons other than upper airway dryness. The treatment of upper airway dryness with heated humidification may cause excessive condensation in the CPAP tubing (termed rain-out), resulting in discomfort and sleep disruption. Rain-out was more common in patients who used the Oracle mask, which is most likely due to a requirement for higher humidity with this mask. Unfortunately, our study did not record the level of humidification that patients used, which would address this possibility. The recent addition of a thermostat to heated humidifiers on CPAP units reduces the heater temperature as room temperature falls, which has been shown to reduce the degree of rain-out at nighttime. Further studies are required to determine what role these modifications will play in adherence to CPAP therapy, and whether overall outcome is improved in patients receiving oral CPAP. Until recently, patients who cannot tolerate nasal CPAP were obliged to use an oronasal mask, which is often uncomfortable and technically difficult to use.6,9 In our study, the proportion of patients who stopped using CPAP (dropouts) was significantly higher among those who chose an oronasal mask than those who opted for the Oracle mask (Table 3). Notwithstanding the fact that the difference was no longer significant when patients who changed from the Oracle mask to another type of mask were included in the evaluation (Table 3), this raises the possibility that the Oracle mask may have some advantages over oronasal masks in this patient population. Future studies should address whether the Oracle mask is a more efficacious interface for patients who have occluded nasal airways or who require more sophisticated ventilatory support. The www.chestjournal.org
reason for the higher dropout rate in the oronasal group is not entirely clear. However, previous work by others13 has suggested that oronasal masks are less effective than nasal masks in overcoming the dynamic forces responsible for airway closure in patients with OSA. In summary, the Oracle mask was an efficacious interface for long-term CPAP therapy in approximately 20% of patients with clinically significant OSA. The anticipated problem with upper airway dryness was adequately controlled in most patients by heated humidification, although excessive humidification (rain-out) was observed in some patients. The Oracle mask may be more acceptable than oronasal masks for patients who cannot rely exclusively on the nasal airway for CPAP therapy. Appendix: CPAP Questionnaire Please answer the following questions based on your usage of CPAP during the last 3 weeks: 1. On average, how many nights per week do you use CPAP? nights 2. On average, how many hours per night do you use CPAP? hours 3. Are you taking your mask off during the night? Yes or No If Yes: (a) Does this happen every night? Yes or No (b) When you wake, do you put your interface back on? Yes or No (c) When you sleep, is the length of time you keep your mask on slowly increasing? Yes or No 4. On a scale of 1 to 10, how comfortable do you find your interface? (1 ⫽ unusable; 6 ⫽ comfortable; 10 ⫽ so comfortable that you do not realize that it is on) Oracle mask Nasal mask 5. Are you experiencing dryness in your nose or mouth when waking after using your nasal CPAP? Yes or No On a scale of 1 to 10, rate the dryness in your nose or mouth. (1 ⫽ so dry you cannot use nasal CPAP; 6 ⫽ uncomfortable but manageable; 10 ⫽ noticeable but not uncomfortable) 6. On a scale of 0 to 6, please rate how you feel now compared to before you started using CPAP. (0 ⫽ worse [feel worse than before]; 3 ⫽ same [feel the same as before]; 6 ⫽ better [feel much better than before]).
References 1 Young T, Palta M, Dempsey J, et al. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993; 328:1230 –1235 2 Peppard PE, Young T, Palta M, et al. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000; 342:178 –184 3 Sullivan CE, Issa FG, Berthon-Jones M, et al. Reversal of obstructive sleep apnoea by continuous positive airway pressure applied through the nares. Lancet 1981; 18:862– 865 4 Aldrich MS. Obstructive sleep apnea syndrome. In: Sleep medicine. New York, NY: Oxford University Press, 1999; 227 5 Martins de Araujo MT, Vieira SB, Vasquez EC, et al. Heated CHEST / 124 / 6 / DECEMBER, 2003
2207
6 7
8
9
humidification or face mask to prevent upper airway dryness during continuous positive airway pressure therapy. Chest 2000; 117:142–147 Mortimore IL, Whittle AT, Douglas NJ. Comparison of nose and face mask CPAP therapy for sleep apnea. Thorax 1998; 53:290 –292 Sanders MH, Strollo PJ, Stiller RA. Positive airway pressure in the treatment of sleep-related breathing disorders. In: Chokroverty S, ed. Sleep disorders medicine: basic science, technical considerations, and clinical aspects. Boston, MA: Butterworth-Heinemann, 1999; 355–377 Rechtschaffen A, Kales A, eds. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Los Angeles, CA: Brain Information Service/Brain Research Institute, 1968; 1– 60 Prosise GL, Berry RB. Oral-nasal continuous positive airway
2208
10
11
12
13
pressure as a treatment for obstructive sleep apnea. Chest 1994; 106:180 –186 Anderson FE, Smith NC, Kingshott RN, et al. Efficacy of oral CPAP in the management of obstructive sleep apnea [abstract]. Sleep 2002; 25:A220 Pedler H, Whyte KF, Smith NC, et al. Efficacy of an oral interface for the delivery of CPAP therapy in the treatment of OSA [abstract]. Sleep 2001; 24:A266 Massie CA, Hart RW, Peralez K, et al. Effects of humidification on nasal symptoms and compliance in sleep apnea patients using continuous positive airway pressure. Chest 1999; 116:403– 408 Smith PL, Wise RA, Gold AR, et al. Upper airway pressure flow relationships in obstructive sleep apnea. J Appl Physiol 1988; 64:789 –795
Clinical Investigations