Compliance with continuous positive airway pressure (CPAP) treatment remains a primary concern for improving treatment outcomes of obstructive sleep apnea. There are few studies that have considered the role of upper airway anatomy on the compliance with CPAP. We hypothesized that upper airway anatomy would influence the compliance with CPAP.
One hundred out of 161 consecutive patients were enrolled in this study. The following possible determinants were tested against CPAP use: demographic and anthropometric data, minimal cross-sectional area on acoustic rhinometry, cephalometric and polysomnographic data, questionnaires of Epworth sleepiness scale and Beck depression index, and histories of previous upper airway surgery, degree of nasal obstruction, daily cigarette consumption, and weekly frequency of alcohol intake.
Univariate analysis showed that histories of previous upper airway surgery and less frequent alcohol consumption, and longer mandibular plane-hyoid length (MP-H) on cephalometry were associated with longer average daily CPAP use. After adjustment for the confounding factors with multiple linear regression analysis, alcohol consumption and MP-H were still associated with the compliance with CPAP significantly.
To improve compliance with CPAP, careful evaluations of upper airway problems and life style are important before initiating CPAP.
Obstructive sleep apnea syndrome (OSAS) is characterized by a periodic reduction or cessation in airflow during sleep and daytime sleepiness, with a prevalence of 4.5% in men and 3.2% in women of middle age [
Continuous positive airway pressure (CPAP) is the first-line treatment in moderate to severe OSAS [
However, there are few studies that have considered the role of upper airway anatomy or lifestyle factors such as smoking or alcohol consumption in the compliance of CPAP. In the present study, we investigate the predicting factors of CPAP compliance, in particular upper airway anatomy and lifestyle factors in OSAS patients.
Consecutive adult patients from August 2010 to December 2013, who were confirmed to have OSAS by overnight polysomnography and were recommended CPAP treatment by the senior author (HYK) in the Department of Otorhinolaryngology-Head and Neck Surgery, were enrolled in the present study. Patients were excluded from this study if they had evidence of reduced cardiac function and/or chronic obstructive pulmonary disease at the time of the study or a previous history of a CPAP trial. Patients without enough data regarding anatomic or lifestyle parameters were also excluded from this study. The study was approved by the Institutional Review Board of Samsung Medical Center (IRB No. 2015-07-068).
At the initial visit, a medical history including demographic data and subjective symptoms was collected, and a full otorhinolaryngological evaluation including nose, oral cavity, and oropharynx was conducted. Age, gender, body mass index (BMI), daily amount of cigarette smoking, and weekly frequency of alcohol consumption were evaluated. Patients were also asked to quantify the degree of nasal obstruction they experienced on a scale ranging from 0–7, according to the severity of the symptoms. A modified Mallampati score (MMS) and a tonsillar grade were assigned according to Friedman’s classification, which has been described previously [
In order to objectively evaluate the dimensions of the nasal cavity, acoustic rhinometry was performed [
CPAP was recommended as a first-line treatment for severe OSAS, and was indicated for moderate to severe OSAS according to the results of a sleep study. In addition, patients with mild OSAS who wanted to use CPAP as a treatment were also enrolled in this study.
All patients had unattended auto-CPAP titrations for 2−3 weeks in the present study. They all received detailed instructions and 30–60 minutes of exposure to the CPAP machine to ensure a proper mask fit and acclimatization before the titration. Different types of CPAP nasal masks or nasal pillows were used according to facial structure and individual preference. Heated humidifiers were applied in all cases. A monitoring chip was placed in all of the CPAP devices and used to collect and store the CPAP use data. All patients were followed-up for more than three months after the start of CPAP treatment.
The monitoring chips provided varying information about CPAP use, including the days of CPAP use, the hours of daily use, and the presence of air leaks. We calculated the average daily CPAP use time at the end of the data collection period according to the following formula: average daily CPAP use = total hours of CPAP use/number of follow-up days.
Using the data collected during the detailed interview, physical examination, cephalometry, acoustic rhinometry, and data from the monitoring chip, the following possible determinants were tested against CPAP use: demographic data (patient age, gender, and BMI), questionnaires and history (Epworth sleepiness scale [ESS], Beck depression index [BDI], history of upper airway surgery for snoring or nasal obstruction, presence of mouth breathing during polysomnography, degree of nasal obstruction, daily cigarette smoking, and weekly frequency of alcohol consumption), physical examination (tonsillar grade and MMS), acoustic rhinometry data (MCA), cephalometric data (PAS, MPH, PL) and total apnea and hypopnea index (AHI).
Statistical analyses were performed using SPSS ver. 13 (SPSS Inc., Chicago, IL, USA). Pearson correlation tests were used for continuous variables, and the Spearman correlation test was used to determine the relationship between categorical variables and the average daily CPAP use. A multiple regression test was used to statistically determine the relationships between average daily CPAP use and the variables that were found to be significant in the univariate analysis. The significance level was set at
During the study period, 161 consecutive patients were prescribed to start CPAP. Among the recruited patients, 26 declined to use CPAP after the first trial. Further, 6 more patients were excluded from this study due to follow-up loss and 29 patients were excluded because of incomplete history. Finally, a total of 100 patients were enrolled in the present study. The mean average CPAP use was 4.4±1.6 hours/day. Participants were mostly male (91%) with a mean age of 50.8±10.6 years old. The mean BMI was 26.6±3.1, and the total AHI ranged from 11 to 117 events/hour, with a mean of 44.8±21.6 events/hour. Twenty-four patients (24%) had a previous history of upper airway surgery for snoring or nasal obstruction.
Univariate analysis showed that certain variables were associated with average daily CPAP use, although age, gender, BMI, and AHI did not affect CPAP use. A history of upper airway surgery (
Even following adjustment for the confounding effects of each variable on others, using multiple linear regression analysis, two (weekly alcohol consumption and MP-H) of these three variables still had significant relationships with average daily CPAP use (
The present study shows that factors including upper airway anatomy, history of upper airway surgery, and alcohol consumption are associated with CPAP compliance according to univariate analysis. Using a multivariate regression method, MP-H and alcohol consumption were shown to be significant predictors of CPAP compliance following adjustment for the confounding effects of other variables.
Many researchers have studied the role of the upper airway in OSAS. Even though improved nasal breathing does not consistently improve OSAS itself, it can reduce daytime sleepiness and snoring [
Although still unclear, several factors have been suggested as the cause of this discrepancy. The first is the change in techniques and concepts for the surgical correction of OSAS. Previously, the main surgical treatment for OSAS was UPPP, a single-level surgery involving removal of the uvula and redundant tissue of the soft palate. However, it has evolved over the past decade from UPPP alone to a multilevel surgery, which addresses obstruction at the levels of the nose, palate, and hypopharynx. In addition, oral surgery in our department was performed in a modified way to retain part of uvula, unlike the classic UPPP, and preserve the sealing effect of the soft palate. Long-standing forced mouth breathing and subsequent mouth air leak can be relieved by this technique. Another assumption is that this discrepancy is due to the change in motivation to use CPAP. The patients in the present study were all managed by an ENT doctor who usually preforms OSAS surgery. Those who have undergone prior upper airway surgery may believe that another surgical treatment will not be useful for them, and may be more motivated regarding CPAP use because they realize that their treatment options are now limited [
We also found that a longer MP-H was associated with improved CPAP compliance. To the best of our knowledge, this is the first study focusing on the role of various upper airway structures in CPAP compliance. We do not know the reason for this. However, MP-H is considered to represent the size of the tongue or the length of the collapsible airway, and thus, patients with a longer MP-H may have a greater risk of obstruction during sleep in the supine position, thus an increased chance of benefit with CPAP.
Frequent alcohol consumption inhibited CPAP use in the present study, which is known from many studies to increase nasal resistance [
Even though cigarette smoking did not reach statistical significance, it is also known to be associated with increased total nasal resistance, and Russo-Magno et al. [
Many studies have attempted to identify the predicting factors of CPAP compliance. While some studies have found relationships between age, gender, AHI, daytime sleepiness and CPAP compliance, these findings are inconsistent [
No association was found between CPAP compliance and the severity of OSAS or daytime sleepiness in the present study. The general consensus is that the results of sleep studies or self-reported daytime sleepiness do not predict the use of CPAP [
A diversity of factors may influence CPAP compliance including severity of disease, psychological factors, demographic factors, or obesity. The present study found that anatomical variations of the upper airway and other factors associated with the upper airway may influence CPAP compliance, suggesting a new direction of research. The upper airway, including the nose and pharynx, is the interface between the CPAP device and the lower airway. Structural variations or modifications will therefore have an effect on the communication between the upper and lower airways, which is illustrated by the report that preexisting nasal problems are related to lower adherence to the treatment [
In conclusions, a multitude of factors including anatomical variations of upper airway and frequent alcohol consumption may influence CPAP compliance, which empathizes the need for careful evaluations of upper airway problems and life style to improve CPAP compliance.
▪ Multiple factors may influence continuous positive airway pressure (CPAP) compliance.
▪ In anatomical factors, longer mandibular plane-hyoid length is associated with better CPAP compliance.
▪ In life style factors, frequent alcohol consumption is associated with worse CPAP compliance.
No potential conflict of interest relevant to this article was reported.
Cephalometric parameters for obstructive sleep apnea syndrome. PL, palatal length; MP, mandibular plane; PAS, posterior airway space; H, hyoid bone; Go, gonion; Gn, gnathion; PNS, posterior nasal spine; U, uvula.
Univariate analysis of the possible predictors of the mean CPAP daily use
Variable | Sample size | Mean±SD | Correlation coefficient | |
---|---|---|---|---|
Age (yr) | 100 | 50.8±10.6 | 0.121 | 0.156 |
Body mass index (kg/m2) | 100 | 26.6±3.1 | –0.088 | 0.309 |
Apnea and hypopnea index (/hr) | 100 | 44.8±21.6 | 0.040 | 0.636 |
Mild | 6 | 13.3±1.4 | ||
Moderate | 28 | 23.5±4.2 | ||
Severe | 66 | 56.7±18.3 | ||
Epworth sleepiness scale | 100 | 10.2±4.4 | 0.012 | 0.894 |
Beck depression index | 100 | 7.6±5.6 | 0.187 | 0.138 |
Nasal obstruction | 100 | 3.2±1.8 | –0.127 | 0.169 |
Daily cigarette smoking | 100 | 3.8±7.5 | –0.203 | 0.053 |
Weekly alcohol consumption | 100 | 1.6±1.6 | –0.343 | 0.010 |
Minimal cross-sectional area (cm2) | 100 | 1.3±0.4 | –0.035 | 0.812 |
Posterior airway space (mm) | 100 | 12.4±3.4 | –0.193 | 0.052 |
MP-H (mm) | 100 | 18.9±5.6 | 0.243 | 0.015 |
Palatal length (mm) | 100 | 42.5±4.5 | –0.124 | 0.110 |
CPAP, continuous positive airway pressure; MP-H, distance from the mandibular plane (MP) to the hyoid bone (H).
Mean daily CPAP use according to categorical variables
Characteristic | No. | CPAP (hr) | |
---|---|---|---|
Sex | 0.712 | ||
Male | 91 | 4.4±1.8 | |
Female | 9 | 4.4±1.6 | |
Previous surgery | 0.012 |
||
Yes | 24 | 5.0±1.8 | |
No | 76 | 4.3±1.6 | |
Mouth breathing | 0.054 | ||
Yes | 27 | 5.0±1.7 | |
No | 73 | 4.3±1.6 | |
Tonsillar grade | 0.252 | ||
0 | 6 | 5.6±1.5 | |
1 | 53 | 4.5±1.7 | |
2 | 32 | 4.3±1.6 | |
3 | 9 | 4.3±1.9 | |
Modified Mallampati score | 0.850 | ||
1 | 6 | 4.8±1.8 | |
2 | 38 | 4.3±1.6 | |
3 | 45 | 4.3±1.8 | |
4 | 11 | 4.9±1.9 |
Values are presented as mean±SD.
CPAP, continuous positive airway pressure.
Multiple regression analysis of the mean daily CPAP daily use
Variable | Coefficient | 95% confidence interval | |
---|---|---|---|
Constant | 4.148 | 3.562 to 6.634 | 0.002 |
PAS | –0.095 | –0.089 to –0.001 | 0.054 |
MP-H | 0.079 | 0.024 to 0.134 | 0.005 |
Daily cigarette smoking | –0.049 | –0.046 to –0.001 | 0.042 |
Weekly alcohol consumption | –0.326 | –0.519 to –0.132 | 0.001 |
Only significant data are shown.
CPAP, continuous positive airway pressure; MP-H, distance from the mandibular plane (MP) to the hyoid bone (H); PAS, posterior airway space.