Predictors for short-term and long-term automatic PAP compliance
2022; American Academy of Sleep Medicine; Volume: 19; Issue: 1 Linguagem: Inglês
10.5664/jcsm.10236
ISSN1550-9397
AutoresSong I Park, Byung Kil Kim, Kyung Eun Lee, Sang Duk Hong, Yong Gi Jung, Hyo Yeol Kim,
Tópico(s)Neuroscience of respiration and sleep
ResumoFree AccessScientific InvestigationsPredictors for short-term and long-term automatic PAP compliance Song I. Park, MD, Byung Kil Kim, MD, Kyung Eun Lee, BS, Sang Duk Hong, MD, PhD, Yong Gi Jung, MD, PhD, Hyo Yeol Kim, MD, PhD Song I. Park, MD Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea , Byung Kil Kim, MD Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea , Kyung Eun Lee, BS Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea , Sang Duk Hong, MD, PhD Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea , Yong Gi Jung, MD, PhD Address correspondence to: Hyo Yeol Kim, MD, PhD, Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, Korea; Email: E-mail Address: [email protected]; and Yong Gi Jung, MD, PhD, Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, Korea; Email: E-mail Address: [email protected] Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea , Hyo Yeol Kim, MD, PhD Address correspondence to: Hyo Yeol Kim, MD, PhD, Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, Korea; Email: E-mail Address: [email protected]; and Yong Gi Jung, MD, PhD, Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, Korea; Email: E-mail Address: [email protected] Department of Otorhinolaryngology, Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea Published Online:January 1, 2023https://doi.org/10.5664/jcsm.10236Cited by:1SectionsAbstractEpubPDF ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTStudy Objectives:Positive airway pressure (PAP) is considered a standard treatment for obstructive sleep apnea (OSA), but there are compliance issues. As compliance to PAP tends to decrease with time, it is necessary to consider reasons affecting compliance at each period. Therefore, this study aimed to define factors affecting short-term and long-term compliance to PAP therapy.Methods:One hundred eighty-seven patients with OSA who started PAP treatment between July 2018 to March 2020 were included. Acceptance and compliance rates were monitored. Demographics, polysomnography (PSG) profiles, cephalometric data, and physical examination results were analyzed to identify factors predictive of PAP compliance at short-term (3 months) and long-term (12 months) periods.Results:The acceptance rate of PAP was 92.5%. Compliance at 3 months and 12 months was 79.1% and 51.3%, respectively. Higher apnea-hypopnea index (odds ratio [OR] 1.018, P = .049) and older age (OR 1.032, P = .039) were predictive factors of good automatic PAP (APAP) compliance at 3 months. However, long-term compliance was affected by the percentage of duration with O2 desaturation of < 90% (CT90; OR 1.032, P = .011) and baseline self-reported symptom scores such as nasal obstruction (OR 0.819, P = .038) and awakening (OR 0.796, P = .045).Conclusions:In PAP use, indicators of OSA severity such as apnea-hypopnea index affect short-term compliance. On the other hand, the mandibular plane to hyoid distance and self-reported symptoms such as nasal obstruction and awakening can affect long-term compliance.Citation:Park SI, Kim BK, Lee KE, Hong SD, Jung YG, Kim HY. Predictors for short-term and long-term automatic PAP compliance. J Clin Sleep Med. 2023;19(1):17–26.BRIEF SUMMARYCurrent Knowledge/Study Rationale: Compliance to positive airway pressure (PAP) did not increase over the past 20 years despite technological improvements and self-reported discomfort of upper airway has not yet been studied enough on the impact on PAP compliance. Identifying and treating factors that affect compliance rates over time are important in PAP therapy.Study Impact: The study included variables that were not covered in previous studies, such as cephalometry profiles or upper airway discomfort. The results of the importance of sleep apnea severity in the short term and self-reported symptoms in the long term suggest that different approaches are needed at different times.INTRODUCTIONObstructive sleep apnea (OSA) is common and reaches a prevalence of 15–45% in the general adult population depending on the diagnostic criteria.1,2 Effective treatments of OSA are important because of its impact on hypertension, diabetes, and cardiovascular and cerebrovascular diseases.3–10For the treatment of OSA, positive airway pressure (PAP) is recommended as first-line therapy and is effective in reducing the apnea-hypopnea index (AHI) and daytime sleepiness.11–13 However, the effectiveness of PAP is largely dependent on adherence and reported PAP compliance has been disappointingly low, especially in long-term usage.14–16 Loube et al17 found that approximately 50% of PAP users stop using it within the first year. Moreover, according to a recent systematic review, PAP adherence did not increase over the past 20 years despite technological improvements and support.18Therefore, identifying predictive factors for compliance has been emphasized and numerous factors have been discovered through clinical studies. The most consistent findings include OSA severity,19–21 self-reported daytime somnolence,22 and psychological factors such as initial self-efficacy.22,23 Some studies also found that a high level of socioeconomic status is associated with good compliance. Billings et al24 conducted a multicenter randomized trial for 135 patients with moderate to severe OSA and reported that lower socioeconomic residential areas are related to lower adherence to continuous PAP (CPAP). A cross-sectional study performed by Simon-Tuval et al25 reported that patients with low socioeconomic status showed significantly lower compliance rates.However, predictors for compliance have been characterized poorly, and some factors such as age, sex, and side effects are controversial among studies.21,26–28 In the analysis for long-term compliance conducted by Kohler et al,21 age and sex were not significantly associated with CPAP compliance. Also, Pelletier-Fleury et al27 did not find any association between age and compliance, while female sex and body mass index under 30 kg/m2 were related to noncompliance. On the other hand, May et al28 found that elderly age was a predictor of better adherence.There have also been reports of differences in predictors affecting PAP compliance over time.29 Kendzerska et al30 conducted a retrospective study in a post-stroke population with OSA and reported that having multiple comorbidities was related to early usage of CPAP while the presence of family support was important in long-term adherence. Nevertheless, clinical data on the differences in short- and long-term predictors in the same cohort are limited. Moreover, even though the role of upper airway anatomy on PAP compliance has been emphasized,31–34 no report has studied the effect of cephalometric profile as a predictor in PAP usage in patients with OSA as far as we know.Therefore, in this study, by retrospectively reviewing automatic PAP (APAP) data in our institution, we aimed to identify factors that affect compliance rates over time. In addition, based on our results, we suggest strategies to increase PAP usage over time.METHODSStudy participantsWe conducted a retrospective review of prospectively collected data of adult patients (≥ 18 years old) diagnosed with OSA and prescribed PAP between July 2018 and March 2020 at Samsung Medical Center by a single practitioner. The OSA was diagnosed with a full-night attended in-laboratory polysomnography (PSG). Since the Korean government started insurance for PAP devices in July 2018, the criteria of insurance coverage, either AHI ≥5 events/h with OSA symptoms (or with comorbidities) or AHI ≥15 events/h, were also applied in this study. Since all patients should start PAP therapy within 1 year from the onset of PSG for insurance coverage, we prescribed APAP for patients unless they had underlying medical conditions contradindicating APAP to meet the increased number of patients within the limited time. Therefore, we confined the study group to APAP patients to eliminate the effect of PAP mode. Patients were excluded if they had any of the following: incomplete data including PSG and medical records, missing self-questionnaires, and underlying medical conditions contraindicating APAP. A previous history of undergoing uvulopalatopharyngoplasty (UPPP) surgery was not an exclusion criterion in this study. Patients who did not attend the clinic during the follow-up period were retained for analysis and were considered PAP noncompliant. This study was approved by the Samsung Medical Center Institutional Review Board (IRB no. 2021-08-089-001).Outcome measuresThe primary outcome was to compare the predictors of 3-month APAP treatment compliance with that of 12 months. PAP usage information was objectively monitored on devices using a built-in compliance meter and downloaded by staff at each clinic visit. We defined "acceptance" as continuing the use of PAP without returning it after the first trial. Similar to previous studies, the criteria for "good compliance" were using PAP 4 hours per night on > 70% of nights.35 Adherence was defined as the use of a PAP device but that failed to achieve the criteria for good compliance.Factors associated with PAP usageThe following variables were examined to identify predictors for compliance: demographic profile, PSG profile, cephalometry profile, physical examination profile, and baseline self-reported symptom questionnaire scales. Demographic data included age, sex, body mass index (BMI), smoking/alcohol status, comorbidities, and previous operation/trauma history. PSG-related factors included OSA severity indexes (AHI, respiratory disturbance index [RDI], CT90 [the percentage of duration with O2 desaturation < 90%], desaturation nadir), quality of sleep (total bedtime, total sleep time, percentage of sleep stages, arousal index), and Epworth Sleepiness Scale (ESS) scores. Cephalometry variables included palatal length (PL; the distance from the posterior nasal spine to the tip of the uvula), posterior airway space (PAS; the minimal distance between the tongue base and the posterior pharyngeal wall at the level of the mandibular angle), and the mandibular plane to hyoid (MPH; the distance from the mandibular plane to the most anterosuperior point of the hyoid), as defined in a previous study (Figure 1).36 Lateral cephalometric radiographs were taken for all patients. Each cephalometric index was manually measured by a researcher who was blinded to each patient's compliance rate. Physical examination variables included modified Mallampati classification37 and Friedman tonsil grade.38 Baseline self-reported symptom questionnaire scales scored the following items on a scale of 0 (no discomfort at all) to 5 (very severe discomfort): snoring, apnea, nasal obstruction, headache, and awakening during sleep.Figure 1: Cephalometric parameters for obstructive sleep apnea syndrome.Gn = gnathion, Go = gonion, H = hyoid bone, MP = mandibular plane, PAS = posterior airway space, PL = palatal length, PNS = posterior nasal spine, U = uvula.Download FigureStatistical analysisDifferences between compliers and noncompliers were compared using the Mann-Whitney U test for continuous variables and the chi-square test for categorical data. Data were described using means and standard deviations or absolute and relative frequencies. Univariate and multivariate logistic regressions were performed to identify factors associated with PAP usage. Variables found on univariate analysis to have a P value < .10 were included in the multivariate analysis. Logistic regression results are presented as odds ratios (ORs) and 95% confidence intervals (CIs). A P value < .05 was considered to be statistically significant. The data were analyzed using the Statistical Package for Social Science (SPSS; IBM Corporation, Armonk, NY, USA).RESULTSStudy participants and the percentage of PAP usageAmong 230 initially identified patients, 43 were excluded for incomplete follow-up data in 7 patients, missing self-questionnaire in 28 patients, incomplete PSG data in 6 patients, and incomplete medical records in 2 patients. A total of 187 patients were included in this study (Figure 2).Figure 2: Flowchart of patients.APAP = automatic positive airway pressure, PSG = polysomnography.Download FigureOf the 187 patients, 173 (92.5%) accepted the use of PAP device, excluding those who refused PAP due to claustrophobia and other reasons. The numbers of patients who met the criteria for ≥ 4 hours of PAP use on 70% of nights in a consecutive 30-day period was 148 (79.1%) at 3 months and 96 (51.3%) at 12 months. The proportion of those who continued PAP use irrespective of compliance was 166 (88.8%) at 3 months but decreased to 136 (72.7%) at 12 months (Figure 3).Figure 3: The percentages of initial acceptance and adherence and compliance traces at 3 and 12 months.Download FigureBaseline characteristicsBaseline characteristics are presented in Table 1. The included study patients were predominantly male (90.9%), overweight, and with a mean AHI of 48.04 events/h. The mean age was 50.08 years, and the compliance group was older than the noncompliance group in both short- and long-term periods. However, the difference was not statistically significant.Table 1 Comparisons of characteristics between compliance and noncompliance groups at 3 and 12 months.FactorOverall (n = 187)Short-Term (3 Months)Long-Term (12 Months)Compliance (n = 148)Noncompliance (n = 39)PCompliance (n = 96)Non-Compliance (n = 91)PAge, y50.08 ± 12.04 [51.00]50.94 ± 11.81 [52.50]46.82 ± 12.49 [46.00].07250.59 ± 12.03 [51.50]49.54 ± 12.09 [49.00].496Sex, n (%).357>.999 Male170 (90.9%)136 (91.9%)34 (87.2%)87 (90.6%)83 (91.2%) Female17 (9.1%)12 (8.1%)5 (12.8%)9 (9.4%)8 (8.8%)PSG profile BMI, kg/m227.61 ± 4.50 [27.20]27.63 ± 4.63 [27.15]27.55 ± 4.04 [27.30].86527.75 ± 4.89 [27.20]27.47 ± 4.07 [27.20].981 AHI, events/h48.04 ± 21.12 [43.50]49.50 ± 21.12 [45.75]42.47 ± 20.42 [37.30].05649.78 ± 21.89 [46.35]46.19 ± 20.23 [43.10].317 ESS10.90 ± 5.17 [10.00]11.09 ± 5.29 [10.00]10.15 ± 4.68 [9.00].33010.60 ± 5.21 [10.00]11.21 ± 5.14 [11.00].401 RDI, events/h49.22 ± 20.50 [45.10]50.62 ± 20.64 [46.75]43.91 ± 19.28 [40.70].06350.57 ± 21.59 [47.60]47.80 ± 19.30 [44.40].446 CT90, %10.12 ± 15.33 [4.10]11.22 ± 16.38 [4.40]5.95 ± 9.51 [1.90].02513.25 ± 19.06 [5.00]6.83 ± 8.98 [3.10].091 Desaturation nadir, %77.83 ± 8.35 [79.00]77.55 ± 8.21 [79.00]79.37 ± 8.82 [82.00].15976.68 ± 9.21 [77.50]79.24 ± 7.15 [81.00].054 Total bedtime, min416.75 ± 46.27 [416.00]416.64 ± 46.15 [414.85]417.15 ± 47.33 [426.50].851416.62 ± 47.42 [416.15]416.88 ± 45.28 [415.00].793 Total sleep time, min347.64 ± 53.00 [349.00]349.25 ± 55.09 [349.50]341.53 ± 44.29 [333.50].218352.80 ± 53.49 [351.00]342.20 ± 52.22 [339.50].147 N2, %48.44 ± 12.16 [50.30]48.83 ± 12.30 [50.60]46.96 ± 11.65 [45.40].35149.09 ± 12.94 [51.50]47.75 ± 11.31 [47.80].261 REM, %18.31 ± 6.16 [18.40]18.35 ± 6.31 [18.40]18.17 ± 5.60 [18.30].87118.85 ± 6.59 [18.60]17.75 ± 5.65 [17.80].224 Arousal index38.50 ± 18.30 [34.50]39.29 ± 18.92 [35.55]35.54 ± 15.60 [31.10].32638.93 ± 19.38 [33.50]38.05 ± 17.17 [36.10].918Cephalometry, mm MPH18.98 ± 6.64 [18.93]18.79 ± 6.69 [18.78]19.69 ± 6.48 [20.22].44720.17 ± 6.73 [20.16]17.72 ± 6.33 [17.16].012 PL40.28 ± 5.16 [40.97]40.15 ± 5.50 [40.84]40.77 ± 3.63 [41.20].69740.00 ± 5.48 [40.69]40.57 ± 4.82 [41.12].406 PAS10.27 ± 2.78 [10.15]10.41 ± 2.72 [10.28]9.74 ± 2.95 [9.51].18110.18 ± 2.84 [10.15]10.36 ± 2.72 [10.15].660Physical exam Mallampati3.43 ± 0.84 [4.00]3.40 ± 0.89 [4.00]3.54 ± 0.60 [4.00].7513.41 ± 0.82 [4.00]3.45 ± 0.86 [4.00].520 Tonsil1.51 ± 0.79 [1.00]1.53 ± 0.82 [1.00]1.46 ± 0.68 [1.00].7921.57 ± 0.83 [2.00]1.45 ± 0.75 [1.00].255Self-reported symptoms questionnaire score Snoring4.07 ± 1.00 [4.00]4.07 ± 0.98 [4.00]4.08 ± 1.09 [4.00].7724.14 ± 0.88 [4.00]4.01 ± 1.11 [4.00].723 Apnea3.52 ± 1.27 [4.00]3.59 ± 1.13 [4.00]3.26 ± 1.68 [3.00].4703.43 ± 1.23 [4.00]3.63 ± 1.31 [4.00].185 Nasal obstruction1.91 ± 1.68 [2.00]1.86 ± 1.62 [2.00]2.13 ± 1.88 [2.00].4571.70 ± 1.61 [1.00]2.14 ± 1.72 [2.00].082 Headache1.29 ± 1.32 [1.00]1.37 ± 1.36 [1.00]1.00 ± 1.15 [1.00].1381.24 ± 1.25 [1.00]1.35 ± 1.40 [1.00].768 Awakening2.37 ± 1.42 [2.00]2.33 ± 1.41 [2.00]2.54 ± 1.45 [2.00].5342.15 ± 1.42 [2.00]2.61 ± 1.38 [2.00].026Comorbidity, n (%) Diabetes mellitus27 (14.4%)20 (13.5%)7 (17.9%).45414 (14.6%)13 (14.3%)>.999 Hypertension87 (46.5%)68 (45.9%)19 (48.7%).85747 (49.0%)40 (44.0%).558 Vascular disease7 (3.7%)5 (3.4%)2 (5.1%).6373 (3.1%)4 (4.4%).470Smoking, n (%).647.448 No153 (81.8%)122 (82.4%)31 (79.5%)81 (84.4%)72 (79.1%) Current smoker34 (18.2%)26 (17.6%)8 (20.5%)15 (15.6%)19 (20.9%)Alcohol (time/week)1.13 ± 1.37 [1.00]1.05 ± 1.28 [1.00]1.46 ± 1.64 [1.00].1561.10 ± 1.31 [1.00]1.16 ± 1.43 [1.00].995Trauma history, n (%)16 (8.6%)13 (8.8%)3 (7.7%)>.9999 (9.4%)7 (7.7%).796Nasal operation history, n (%)53 (28.3%)42 (28.4%)11 (28.2%)>.99929 (30.2%)24 (26.4%).627Pre-UPPP, n (%)19 (10.2%)17 (11.5%)2 (5.1%).37311 (11.5%)8 (8.8%).632Follow-up period, mo17.63 ± 8.47 [17.53]19.21 ± 7.76 [19.28]11.62 ± 8.48 [9.30]<.00120.39 ± 7.40 [20.12]14.72 ± 8.60 [14.23]<.001Median values are presented between square brackets. AHI = apnea-hypopnea index, BMI = body mass index, CT90 = percentage of duration with O2 desaturation < 90%, ESS = Epworth Sleepiness Scale, MPH = mandibular plane to hyoid, PAS = posterior airway space, PL = palatal length, PSG = polysomnography, RDI = respiratory disturbance index, REM = rapid eye movement, UPPP = uvulopalatopharyngoplasty.In the PSG variables, the overall severity-related indexes were more severe in the compliance group. But only the CT90 value was significantly higher in the short-term compliance group. The mean AHI was > 30 events/h in both groups with 150 patients (80.2%). In cephalometry, only the MPH at 12 months showed a statistically significant difference (compliance group 20.17 vs noncompliance group 17.72, P = .012). The measurements of PL and PAS were 40.28 mm and 10.27 mm, respectively. The average value of Mallampati grade was 3.43 and tonsil grade was 1.51 in physical examination, with no significant difference between groups.In baseline self-reported symptom questionnaires, snoring and apnea were the 2 most severe complaints, with average scores of 4.07 and 3.52, respectively. The mean values of nasal obstruction, headache, and awakening were 1.91, 1.29, and 2.37, respectively. In the short term, there was no association between the symptoms and the compliance. However, in the long term, the noncompliance group showed a tendency for higher baseline symptom scores in all items except for snoring. But only awakening was statistically significant.The most common comorbidity in the study population was hypertension (46.5%). The percentage of nonsmokers, low alcohol intake, and individuals who had UPPP was higher in the good-compliance group. But there was no significant difference between groups.The compliance group had a significantly longer follow-up period than the noncompliance group in both the short term and long term (19.2 vs 11.6 at 3 months with a P value < .001, 20.4 vs 14.7 at 12 months with a P value < .001). The shortest follow-up period was 11.62 months in the short-term, noncompliance group.In the short term, CT90 score was statistically higher in the compliance group (11.2 vs 5.95, P = .025), while other factors were not statistically significant between groups. In the long term, the compliance group had a longer length of MPH (P = .012) and lower awakening score (P = .026), which means that good PAP users are less likely to complain of awakening during their sleep.Predictors of PAP complianceUnivariate analysis showed that PAP compliance for the first 3 months was associated with age, AHI, RDI, CT90, and alcohol intake (Table 2). For every 1-year increase in age, compliance increased by 1.03 times. For each 1-unit increase in AHI, RDI, and CT90, compliance increased by 1.02, 1.02, and 1.03 times, respectively. On the other hand, compliance decreased by 0.791 for each increase in alcohol intake. A multivariate analysis was performed on these variables. In the case of AHI, RDI, and CT90, analysis was performed on AHI from among them due to multicollinearity. The final result showed that the use of PAP increased with increasing age (OR 1.032, P = .039) and with higher initial AHI values (OR 1.018, P = .049).Table 2 Predictors of PAP compliance at 3 and 12 months.PredictorShort-Term (3 Months)Long-Term (12 Months)Univariate AnalysisMultivariate AnalysisUnivariate AnalysisMultivariate AnalysisORPORPORPORPAge, y1.029 (0.999–1.060).0591.032 (1.002–1.063)0.0391.007 (0.984–1.032).548Sex MaleReferenceReference Female0.600 (0.198–1.819).3671.073 (0.395–2.914).890PSG profile BMI (kg/m2)1.004 (0.927–1.087).9241.014 (0.951–1.081).677 AHI (events/h)1.017 (0.999–1.036).0671.018 (1.000–1.037).0491.008 (0.994–1.022).245 ESS1.036 (0.967–1.111).3120.977 (0.924–1.034).424 RDI (events/h)1.017 (0.999–1.037).0711.007 (0.993–1.021).357 CT90 (%)1.034 (0.998–1.072).0661.033 (1.009–1.058).0071.032 (1007–1.058).011 Desaturation nadir0.972 (0.929–1.018).2270.962 (0.928-0.998).038 Total bedtime (min)1.000 (0.992–1.007).9511.000 (0.994–1.006).970 Total sleep time (min)1.003 (0.996–1.010).4181.004 (0.998–1.010).175 N2 (%)1.012 (0.984–1.042).3931.009 (0.986–1.033).449 REM (%)1.005 (0.949–1.064).8701.030 (0.982–1.080).224 Arousal index1.012 (0.991–1.033).2571.003 (0.987–1.019).743 CPAP pressure, cmH2O1.025 (0.830–1.266).8171.058 (0.892–1.254).517Cephalometry (mm) MPH0.965 (0.879–1.060).4591.062 (1.013–1.114).0131.056 (1.003–1.111).037 PL0.976 (0.909–1.048).5070.979 (0.925–1.035).452 PAS1.093 (0.959–1.246).1820.977 (0.881–1.084).658Physical exam Mallampati0.883 (0.469–1.664).7010.938 (0.664–1.325).716 Tonsil1.216 (0.623-2.375).5671.218 (0.844–1.757).291Self-reported symptoms questionnaire score Snoring0.719 (0.389–1.328).2921.135 (0.849–1.517).394 Apnea0.987 (0.641–1.519).9870.882 (0.701–1.109).283 Nasal obstruction0.959 (0.707–1.302).7900.851 (0.715–1.013).0700.819 (0.678–0.989).038 Headache1.129 (0.749–1.701).5620.937 (0.754–1.166).561 Awakening1.054 (0.718–1.547).7880.787 (0.639-0.970).0250.796 (0.636–0.995).045Comorbidity Diabetes mellitus0.895 (0.442–1.813).7581.024 (0.453–2.317).954 Hypertension0.714 (0.278–1.836).4851.223 (0.688–2.175).493 Vascular disease0.647 (0.121–3.468).6110.702 (0.153–3.225).649Smoking NoReferenceReference Current smoker0.551 (0.149–2.033).3710.702 (0.332–1.482).353Alcohol (time/week)0.719 (0.492–1.051).0880.719 (0.492–0.051).0880.968 (0.784–1.195).761Trauma history1.156 (0.312–4.275).8281.241 (0.442–3.485).681Nasal operation history1.009 (0.461–2.208).9831.208 (0.638–2.287).561Pre-UPPP2.401 (0.530–0.867).2561.343 (0.514–3.505).547AHI = apnea-hypopnea index, BMI = body mass index, CPAP = continuous positive airway pressure, CT90 = percentage of duration with O2 desaturation < 90%, ESS = Epworth Sleepiness Scale, MPH = mandibular plane to hyoid, OR = odds ratio, PAP = positive airway pressure, PAS = posterior airway space, PL = palatal length, PSG = polysomnography, RDI = respiratory disturbance index, REM = rapid eye movement, UPPP = uvulopalatopharyngoplasty.However, in the long term, univariate analysis showed that CT90, desaturation nadir, MPH, nasal obstruction, and awakening parameters had clinical significance. The higher CT90 (OR 1.033) and MPH values (OR 1.062) were associated with good compliance, while the higher desaturation nadir (OR 0.962), nasal obstruction (OR 0.851), and awakening values (OR 0.787) were related to lower compliance. Because of multicollinearity, CT90 was included in the multivariate analysis from among the CT90 and desaturation nadir. In subsequent multivariate analysis, the higher the CT90 value (OR 1.032, P = .011) the longer the MPH length (OR 1.056, P = .037); the less the patient self-reported a complaint nasal congestion (OR 0.819, 0.038) and awakening (OR 0.796, P = .045), the more likely the patient was to use a positive pressure device.DISCUSSIONTo our knowledge, this is the first study to analyze the predictors of short-/long-term PAP compliance in the same cohort including anatomical factors such as cephalometric profile. In this study, we found that 92.5% of patients with OSA accepted PAP usage, while short-term 3-month compliance was 79.1% and long-term 12-month compliance was 51.3%. To solve the compliance issues during the follow-up period, we scheduled the patients to visit the clinic every 3 months. Because this is national health insurance–mandated prescription interval in Korea, adhering to it allows the patient to be motivated for financial benefit while also allowing the patient to discuss the discomfort from PAP usage more frequently. We also encouraged patients to feel free to contact the device manager whenever they were in trouble with using the device. The clinical factors related to APAP use at 3 months were age and AHI, while CT90, MPH, nasal obstruction, and awakening were associated with 12 months of usage. These results imply that different factors can affect PAP usage depending on the period.In this study, we included upper airway anatomy factors of MPH, PL, and PAS, which were rarely discussed in previous studies. It is well known that upper airway anatomy influences the pathophysiology of OSA.39,40 Increased tongue dimensions can displace the hyoid bone inferiorly, increasing the length of the MPH.41,42 Therefore, a longer length of MPH has been considered a common anatomical feature in patients with OSA.43–45 However, few studies have discussed the effect of anatomical variables in PAP use. Jeong et al36 reported that longer MPH length on cephalometry was associated with good CPAP usage. As patients with longer MPH have a higher risk of obstruction during sleep, the benefit of using PAP will be significant in those patients. This is consistent with the present study and emphasizes the importance of anatomical factors. The higher MPH value, which can bring higher self-reported satisfaction in PAP usage, may act as a driving force for increased PAP compliance, especially in the long term.Studies have shown that the severity index of OSA including AHI, RDI, and oxygen desaturation index can affect PAP compliance. In the study by Campos-Rodriquez et al,20 which included 357 patients with OSA, AHI and the percentage of duration with oxygen desaturation being < 90% were significantly higher in the compliance group. Kohler and colleagues21 demonstrated that oxygen desaturation index was independently related to long-term CPAP compliance. Also, according to a study by Riachy et al,19 of 138 patients with OSA, the compliance group had significantly higher mean RDI and oxygen desaturation index and lower nadir oxygen saturation level after initiation of CPAP therapy. Similar to previous studies, we found that all 3 factors (AHI, RDI, CT90) indicative of OSA severity were associated with short-term compliance. Because of multicollinearity, only AHI was included in the multivariate analysis and showed clinical significance.However, AHI and RDI lost significance in the long term in this study, while CT90 maintained its significance. This might be explained by the insurance policies implemented in many countries, including Korea. Most insurance policies require compliance with PAP treatment to reimburse the device and associated disposable materials. In Korea, the national health insurance covers the cost of PSG and PAP for patients who meet the criteria including AHI. Therefore, with the awareness of the AHI value, AHI itself can act as a motivator for patients who start using PAP devices. But the convenience of using a PAP device can become more important to patients over time, leading to CT90 and desaturation nadir, factors related to the patient's self-reported symptoms, as predictors for long-term PAP usage.The result in our study that self-reported discomforts such as nasal obstruction or awakening were important in long-term compliance can also be seen as an issue similar to the convenience mentioned above. Many papers have reported the importance of nasal discomfort during CPAP treatment. Inoue et al46 emphasized the negative impact of nasal disease and total nasal resistance in early CPAP usage. Sugiura et al47 reported that higher nasal resistance was an important factor in CPAP nonacceptance. Nasal obstruction not only complicates CPAP use but is itself a risk factor for sleep-disordered breathing.48A low arousal threshold, which means being easily awakened from sleep, has also been considered as a predictor of poor PAP compliance.49,50 However, in this study, we found the arousal index was insignificant, while the self-reported awakening score of the self-questionnaire was significant. This might indicate that the degree of self-reported sleep depth perceived by the patient is more important than the objective data.We also found that lower alcohol intake was associated with good compliance in univariate
Referência(s)