Clinical Predictors of Effective Continuous Positive Airway Pressure in Patients With Obstructive Sleep Apnea/Hypopnea Syndrome

The Laryngoscope VC 2015 The American Laryngological, Rhinological and Otological Society, Inc. Clinical Predictors of Effective Continuous Positive Airway Pressure in Patients With Obstructive Sleep Apnea/Hypopnea Syndrome Chi-Chih Lai, MD; Michael Friedman, MD, FACS; Hsin-Ching Lin, MD, FACS; Pa-Chun Wang, MD; Michelle S. Hwang, BS; Cheng-Ming Hsu, MD; Meng-Chih Lin, MD; Chien-Hung Chin, MD Objectives/Hypothesis: To identify standard clinical parameters that may predict the optimal level of continuous positive airway pressure (CPAP) in adult patients with obstructive sleep apnea/hypopnea syndrome (OSAHS). Study Design: This is a retrospective study in a tertiary academic medical center that included 129 adult patients (117 males and 12 females) with OSAHS confirmed by diagnostic polysomnography (PSG). Methods: All OSAHS patients underwent successful full-night manual titration to determine the optimal CPAP pressure level for OSAHS treatment. The PSG parameters and completed physical examination, including body mass index, tonsil size grading, modified Mallampati grade (also known as updated Friedman s tongue position [uftp]), uvular length, neck circumference, waist circumference, hip circumference, thyroid-mental distance, and hyoid-mental distance (HMD) were recorded. Results: When the physical examination variables and OSAHS disease were correlated singly with the optimal CPAP pressure, we found that uftp, HMD, and apnea/hypopnea index (AHI) were reliable predictors of CPAP pressures (P 5.013, P 5.002, and P <.001, respectively, by multiple regression). When all important factors were considered in a stepwise multiple linear regression analysis, a significant correlation with optimal CPAP pressure was formulated by factoring the uftp, HMD, and AHI (optimal CPAP pressure 5 1.01 uftp 1 0.74 HMD 1 0.059 AHI 2 1.603). Conclusions: This study distinguished the correlation between uftp, HMD, and AHI with the optimal CPAP pressure. The structure of the upper airway (especially tongue base obstruction) and disease severity may predict the effective level of CPAP pressure. Key Words: Snoring, sleep apnea, obstructive sleep apnea/hypopnea syndrome, continuous positive airway pressure. Level of Evidence: 4. Laryngoscope, 125:1983 1987, 2015 INTRODUCTION Continuous positive airway pressure (CPAP), first introduced in 1981, is usually the primary treatment for obstructive sleep apnea/hypopnea syndrome (OSAHS). 1,2 However, despite its efficacy in reversing sleep apnea, CPAP has variable compliance. High levels of nonadherence to CPAP treatment remains a major problem in treating OSAHS patients. 3 Any factors that significantly affect a patient s initial experience with CPAP can have an impact on subsequent compliance. 4 The optimal CPAP pressure level is an important factor, because a lower CPAP level may result in insufficient treatment effects and/or unintentional mask removal, whereas a higher CPAP level may induce pressure intolerance and/ or mouth dryness. Thus, predicting the optimal CPAP level and how to simplify the CPAP titration process are essential for the course of OSAHS treatment. Previous literature has studied the issue of predicting effective CPAP level. 5 8 However, most of these studies reported on predictive formulas for optimal CPAP level based only on limited anthropometric variables or polysomnography (PSG) parameters. In recent years, From the Division of Laryngology, Department of Otolaryngology (C.-C.L., H.-C.L., C.-M.H.), Sleep Center (H.-C.L., M.-C.L., C.-H.C.), Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine (M.-C.L., C.-H.C.), Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, Kaohsiung, Taiwan; Department of Otolaryngology Head and Neck Surgery (M.F.), Rush University Medical Center, Chicago, Illinois, U.S.A.; Department of Otolaryngology (M.F., M.S.H.), Advanced Center for Specialty Care, Advocate Illinois Masonic Medical Center, Chicago, Illinois, U.S.A.; and the Department of Otolaryngology (P.-C.W.), Cathay General Hospital, Taipei, Taiwan. Editor s Note: This Manuscript was accepted for publication December 10, 2014. The authors were responsible for work on this article as follows: Chi-Chih Lai, drafting and data collection; Michael Friedman, data interpretation, critical revision of the manuscript for important intellectual content; Hsin-Ching Lin, study design, data collection/interpretation, drafting/revision; Pa- Chun Wang, statistical analysis, drafting/revision; Michelle S. Hwang, data interpretation and critical revision of the manuscript; Cheng-Ming Hsu, drafting and data collection/interpretation; Meng-Chih Lin, data collection and critical revision of the manuscript; Chien-Hung Chin, data collection and critical revision of the manuscript. Dr. Michael Friedman received a research grant from ImThera Medical, Inc., San Diego, California. However, ImThera Medical Inc. had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. This study was sponsored in its entirety by the principal investigator (Hsin-Ching Lin). The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Hsin-Ching Lin, MD, Department of Otolaryngology, Sleep Center, Kaohsiung Chang Gung Memorial Hospital, 123, Ta-Pei Rd., Niao-Sung District, Kaohsiung City, 833 Taiwan. E-mail: enthclin@aol.com DOI: 10.1002/lary.25125 1983

increasing numbers of studies have emphasized the correlations between upper airway anatomical abnormalities and both the presence and severity of OSAHS. 9,10 The evaluation of upper airway anatomy and the standardization of physical findings for the diagnosis and treatment in OSAHS patients are important. Friedman et al. demonstrated the value of a clinical staging system that was based on tonsil size grading, modified Mallampati grade (also know as updated Friedman s tongue position [uftp]), and body mass index (BMI) for predicting the success of OSAHS surgery. 11,12 The optimal CPAP pressure is most commonly determined using gradually titrated pressures to abolish obstructive breathing events during sleep, under an attended PSG in the sleep laboratory. This procedure is time consuming, expensive, and requires at least two overnight sleep laboratory studies (diagnostic PSG and CPAP titration) for every patient. Changing paradigms of OSAHS treatment is partially a result of the need to reduce the cost of healthcare. As part of this change, diagnostic and therapeutic PSGs are frequently combined in a split night study. With limited sleep time, titration studies are often uncomfortable. An incomplete titration study may require the additional cost of another study or may lead to CPAP settings that are less than optimal. Pretest calculation of predicted CPAP settings could result in obtaining an optimal pressure in a shorter time period. The purpose of this study was to evaluate the correlations between the optimal CPAP level and combination of polysomnographic and anthropometric variables in adult patients with OSAHS. MATERIALS AND METHODS One hundred twenty-nine adult patients with OSAHS were retrospectively reviewed. The institutional review board of Chang Gung Memorial Hospital Ethical Committee approved the study protocol (CGMH IRB No. 103 1198B). Patients with long-term usage of medications known to affect sleep or those with current alcohol or drug abuse, congestive heart failure, obstructive pulmonary disease, coronary or cerebral vascular disease, history of psychosis, central sleep apnea syndromes, or previous OSAHS surgery were excluded. All of the patients completed comprehensive diagnostic sleep studies at the Kaohsiung Chang Gung Memorial Hospital, in a temperature-controlled and sound attenuated room with staff in attendance all night. The OSAHS was defined as an apnea/hypopnea index (AHI) >5 per hour, whereas the severity of OSAHS was a combination of the severity of daytime sleepiness and the value of AHI during sleep. By definition, obstructive apnea was a reduction in airflow 90% of baseline for at least 10 seconds with effort to breathe during apnea. Obstructive hypopnea was defined as an abnormal respiratory event with at least a 30% reduction in thoracoabdominal movement or airflow when compared to baseline, lasting for at least 10 seconds, and with 4% oxygen desaturation. The AHI was defined as the total number of apneas and hypopneas per hour of electroencephalographic sleep. All PSG findings were scored by an experienced physician who was certified by the Taiwan Board of Sleep Medicine and was unaware of the study and therefore blinded to the patient s participation in the study. The AHI, snoring visual analogue scale (SVAS), Epworth Sleepiness 1984 Scale, lowest O 2 saturation (LSAT), and snoring index were included in the analysis. Anthropometric measures, including upper airway anatomical variables were recorded. BMI (kg/m 2 ) was calculated as the ratio of weight (kg) to the square of height (m 2 ). Tonsil size grading was classified by two clinicians using the Friedman scale 13 as follows: 0, surgically removed tonsils; 1, tonsils hidden within the pillars; 2, tonsils extending to the pillars; 3, tonsils were beyond the pillars but not to the midline; and 4, tonsils extended to the midline. The anatomical relationships between the tongue and tonsils/pillar, uvula, soft palate, and hard palate were also evaluated by uftp. 14 The uvular length was measured from the base to the distal free end of the uvula on the midline by using a straight long caliper, with the subject sitting in an upright position with mouth open and tongue inside. Neck circumference (NC) was measured in centimeters just below the laryngeal prominence, with the head positioned in the Frankfort horizontal plane. Waist circumference (WC) was measured in centimeters at midway between the last rib and the top of the iliac crest at the end of a normal expiration, and hip circumference (HC) was measured in centimeters at the level of the greater trochanter. With patients heads in a natural head posture in the upright seated position, thyroid-mental distance (TMD) from the thyroid notch to the mental prominence and hyoid-mental distance (HMD) from the hyoid bone to the mental prominence were both measured in centimeters at the end of the expiration phase and without swallowing. A simple right-angle caliper demonstrated in previous literature 9 was used to measure the vertical and horizontal distances first and then to get TMD and HMD. All patients underwent a successful standard manual titration for CPAP trial. To determine the optimal pressure level for CPAP treatment, full-night, manual titration with the same montage PSG was performed. The optimal pressure level was defined as the lowest effective pressure that controlled most respiratory disturbances including apnea, hypopnea, and snoring in all body positions and in all stages, especially in the supine position during rapid eye movement sleep, to determine the optimal pressure level for home use. Statistical analyses were conducted using the SPSS software (SPSS 15.0 for Windows; IBM, Armonk, NY). Pearson correlation coefficients were used to explore the relationships between optimal CPAP level and baseline data, including anthropometric and PSG variables. Statistical significance was set at P <.05. Once the variables were selected, stepwise multiple linear regression analysis was used to select independent predictive variables and develop a predictive equation for optimal CPAP level. RESULTS There were 129 adult OSAHS patients (117 males and 12 females) with a mean age of 46.2 years (range, 22 74 years). Baseline clinical data composed of effective CPAP level, anthropometric measures with upper airway anatomic variables, and PSG variables are shown in Table I. Correlations Between Optimal CPAP Pressure and Anthropometric and PSG Variables Pearson correlation coefficient analysis showed that significant positive correlations between effective CPAP pressure and anthropometric variables including BMI, subjective tonsil size grading, uftp, NC, WC, HC, and HMD. In addition, significant positive correlations were

TABLE I. Baseline Clinical Data (n 5 129 Patients). Variable Mean 6 SD* Range CPAP pressure (cm H 2 O) 7.48 6 2.6 3 18 Age (yr) 46.16 6 10.95 22 74 Anthropometric variables BMI (kg/m 2 ) 27.07 6 3.55 20.7 39.8 uftp 3.07 6 0.60 1 4 Tonsil size 1.83 6 0.85 0 4 Uvular length (cm) 1.21 6 0.47 0 3 NC (cm) 39.37 6 2.98 31.3 46.5 WC (cm) 92.75 6 11.41 72 115 HC (cm) 100.07 6 6.40 84 125 HMD (cm) 4.56 6 0.64 2.9 5.9 TMD (cm) 5.54 6 0.75 3.7 7.1 Polysomnographic variables AHI (/hr) 43.30 6 22.49 5.2 104.9 SVAS 9.22 6 1.81 0 10 ESS 10.05 6 4.82 0 22 LSAT (%) 74.5 6 13.15 31 94 SI (/hr) 287.67 6 213.13 1 852 *Data are expressed as mean 6 standard deviation. circumference. also noted between effective CPAP pressure and PSG variables, including AHI, SVAS, and LSAT (Table II). These variables were considered potentially independent predictors. Stepwise Multiple Linear Regression for Final Model The stepwise multiple linear regression analysis was used to examine potential variables for optimal CPAP pressure to determine the optimum number of independent variables in a CPAP prediction model. Three variables, uftp, HMD, and AHI were identified as independent predictors in the final model (Table III). The final CPAP predictive formula was: optimal CPAP pressure 5 1.01 uftp 1 0.74 HMD 1 0.059 AHI 2 1.603. DISCUSSION Determining the sites of upper airway anatomical abnormalities to predict the levels of obstruction is the principal goal in the diagnosis and treatment of OSAHS, and numerous methods have been presented. Simple visual inspection and systematic physical examination of the upper airway can be easily performed. Importantly, these clinical evaluations are inexpensive and easy to learn. Aside from the physical examination, other methods used to examine the upper airway include nasofibroscopy with or without Mueller s maneuver, drug-induced TABLE II. Pearson Correlation Between Optimal CPAP Level and Collected Variables. Variable Coefficient P Value Anthropometric BMI (kg/m 2 ) 0.504 <.001* uftp 0.315 <.001* Tonsil size 0.231.009* Uvular length (cm) 0.110.22 NC (cm) 0.396 <.001* WC (cm) 0.255.003* HC (cm) 0.365 <.001* HMD (cm) 0.264.003* TMD (cm) 0.100.26 Polysomnographic AHI (/hr) 0.541 <.001* ESS 0.163.07 LSAT (%) 20.470 <.001* SI (/hr) 0.088.32 SVAS 0.216.01* Others Age (yr) 20.006.95 Sex 0.023.79 *P <.05. circumference; sleep endoscopy, radiologic imaging examinations, and acoustic reflection techniques. These interventions are limited and often lead to additional costs. 15 The optimal CPAP level determined by predictive formulas may be useful in calculating the starting pressure for initiating CPAP titration at the sleep laboratories or for home use, thus reducing the time and cost burden associated with initial CPAP titration. Unlike previous studies on CPAP pressure using only PSG parameters and limited anthropometric variables, this study incorporated the use of easily performed upper airway anatomic evaluations, including tonsil size grading, uftp, uvular length, TMD, and HMD. The relationships between anthropometric measurements and OSAHS severity have been identified. 9 The uftp is a simple anatomical staging system that estimates the presence and severity of hypopharyngeal obstruction in OSAHS, and the importance of uftp in OSAHS severity is confirmed. 9,12 Increased HMD and inferior displacement of the hyoid bone are related to tongue base obstruction and have been shown to be associated with OSAHS. 16 Increased HMD may also increase upper airway instability and lead to destabilizing the upper airway during sleep. The AHI is a crucial factor of OSAHS and is included in most of previous predictive formulas. These three variables are independent and significant predictors in our final model. This study has some limitations. First, the clinical evaluation and physical examination is subject to 1985

TABLE III. Multiple Regression Analysis to Predict Continuous Positive Airway Pressure Optimal Pressure. Variable b (Standard Error) P Value Age, yr 20.01 (0.02).491 Gender Male 20.3 (0.87).730 Female 1.00 BMI, kg/m 2 0.22 (0.12).06 uftp 0.79 (0.31).013* Tonsil size 20.12 (0.26).649 Uvular length, cm 0.22 (0.40).587 NC, cm 0.08 (0.12).496 WC, cm 20.03 (0.02).131 HC, cm 20.01 (0.05).866 HMD, cm 0.69 (0.22).002* TMD, cm 20.12 (0.27).662 AHI, /hr 0.04 (0.01) <.001* SVAS 0.06 (0.10).570 ESS 0.03 (0.04).399 LSAT, % 20.02 (0.02).330 SI, /hr 0.001 (0.001).577 *P <.05. circumference. variability. BMI, uvular length, NC, WC, HC, TMD, and HMD are quantitative measures and are relatively objective. Tonsil grade and uftp are subjective grading systems, and interexaminer variability must be considered. Ng et al. demonstrated the acceptable intraexaminer and interexaminer reproducibility of tonsil size grading, 17 whereas Friedman et al. demonstrated a high level of interexaminer agreement of uftp by kappa statistical analysis and validated the widespread clinical use of uftp reliably and accurately. 12,14 Second, the differences in physiologic monitoring methods used in different sleep centers may have some impact on PSG results, affecting the AHI part of the equation presented in results. Thus, although the concept of determining what factors correlate with final CPAP pressure is valid and useful, readers should be advised that direct use of the reported formula may not be appropriate in all sleep centers. A better approach may be for individual centers to utilize their own data to apply the three measurements used in the formula, such that a center-specific formula can be determined. Third, previous studies have quantified the anthropomorphic differences between Asian, white, Hispanic, and black ethnicities. These data would immediately suggest that a specific formula for CPAP pressure determined with subjects of one ethnicity may not apply to another ethnicity. Further, even the finding that HMD, uftp, and AHI were the three measurements that correlate with final effective CPAP pressure may not hold with populations of different ethnicities, as completely different factors may correlate. Therefore, developing an ethnicity-specific formula for different ethnicities should be a better approach to clinical applicability. Fourth, the predicted CPAP level cannot account for all defects in the upper airway or for the many comorbidities that may influence CPAP levels. Moreover, the pathophysiology of OSAHS may not only be a static obstructive phenomenon but may involve dynamic events during sleep that are not readily explainable by tonsil sizes or uvula lengths. Even using more detailed anthropometric measures including upper airway anatomical variables, predicted CPAP level could not reflect all CPAP levels during sleep. After all, the mechanical effects of CPAP, or surgery for that matter, do not ameliorate all comers with OSAHS. Finally, the sample size was still limited and more reliable, and valid results might be obtained with larger sample sizes. The predictive formula may be useful in calculating a starting pressure for initiating CPAP titration and allowing the optimal pressure to be achieved with only few incremental changes. Further studies testing the use of this predictive formula in comparison to laboratory CPAP titration in clinical practice are needed. CONCLUSION Anatomic abnormalities of the upper airways are important not only in determining the severity of OSAHS but also in planning CPAP treatment. This study found a correlation between the uftp,hmd, and AHI with the optimal CPAP pressure. The structure of the upper airways (especially tongue base obstruction) and disease severity may help predict the level of optimal CPAP pressure. Pretitration prediction of optimal CPAP pressures using anatomic evaluations may hold important value in decreasing the time and cost burden of CPAP titration. Acknowledgments The authors thank Drs. Mao-Chang Su and Yung-Che Chen from the Sleep Center and the Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, Kaohsiung, Taiwan, for assistance in manuscript preparation. They did not receive any financial compensation for their contribution to this study. BIBLIOGRAPHY 1. Sullivan CE, Issa FG, Berthon-Jones M, Eves L. Reversal of obstructive sleep apnoea by continuous positive airway pressure applied through the nares. Lancet 1981;1:862 865. 2. Gay P, Weaver T, Loube D, et al. Evaluation of positive airway pressure treatment for sleep related breathing disorders in adults. Sleep 2006;29: 381 401. 3. Kohler M, Smith D, Tippett V, Stradling JR. Predictors of long-term compliance with continuous positive airway pressure. Thorax 2010;65:829 832. 4. Drake CL, Day R, Hudgel D, et al. Sleep during titration predicts continuous positive airway pressure compliance. Sleep 2003;26:308 311. 5. Miljeteig H, Hoffstein V. Determinants of continuous positive airway pressure level for treatment of obstructive sleep apnea. Am Rev Respir Dis 1993;147:1526 1530. 6. Akashiba T, Kosaka N, Yamamoto H, Ito D, Saito O, Horie T. Optimal continuous positive airway pressure in patients with obstructive sleep apnoea: role of cranio-facial structure. Respir Med 2001;95:393 397. 1986

7. Basoglu OK, Tasbakan MS. Determination of new prediction formula for nasal continuous positive airway pressure in Turkish patients with obstructive sleep apnea syndrome. Sleep Breath 2012;16:1121 1127. 8. Loredo JS, Berry C, Nelesen RA, Dimsdale JE. Prediction of continuous positive airway pressure in obstructive sleep apnea. Sleep Breath 2007; 11:45 51. 9. Friedman M, Tanyeri H, La Rosa M, et al. Clinical predictors of obstructive sleep apnea. Laryngoscope 1999;109:1901 1907. 10. Zonato AI, Bittencourt LR, Martinho FL, Junior JF, Gregorio LC, Tufik S. Association of systematic head and neck physical examination with severity of obstructive sleep apnea-hypopnea syndrome. Laryngoscope 2003;113:973 980. 11. Friedman M, Ibrahim H, Joseph NJ. Staging of obstructive sleep apnea/ hypopnea syndrome: a guide to appropriate treatment. Laryngoscope 2004;114:454 459. 12. Friedman M, Hamilton C, Samuelson CG, Lundgren ME, Pott T. Diagnostic value of the Friedman tongue position and Mallampati classification for obstructive sleep apnea: a meta-analysis. Otolaryngol Head Neck Surg 2013;148:540 547. 13. Friedman M, Ibrahim H, Bass L. Clinical staging for sleep-disordered breathing. Otolaryngol Head Neck Surg 2002;127:13 21. 14. Friedman M, Soans R, Gurpinar B, Lin HC, Joseph NJ. Interexaminer agreement of Friedman tongue positions for staging of obstructive sleep apnea/hypopnea syndrome. Otolaryngol Head Neck Surg 2008;139:372 377. 15. Stuck BA, Maurer JT. Airway evaluation in obstructive sleep apnea. Sleep Med Rev 2008;12:411 436. 16. Bucchieri A, Mastrangelo C, Stella R, Poladas EG. Cephalometric evaluation of hyoid bone position in patients with obstructive sleep apnea [in Italian]. Minerva Stomatol 2004;53:33 39. 17. Ng SK, Lee DL, Li AM, Wing YK, Tong MC. Reproducibility of clinical grading of tonsillar size. Arch Otolaryngol Head Neck Surg 2010;136: 159 162. 1987