Sleep Breath DOI 10.1007/s11325-014-1073-y

ORIGINAL ARTICLE

Endoscopic upper airway evaluation in obstructive sleep apnea: Mueller’s maneuver versus simulation of snoring Hong Huo & Wuyi Li & Xu Tian & Chunxiao Xu & Jian Wang & Dahai Yang

Received: 23 April 2014 / Revised: 1 October 2014 / Accepted: 23 October 2014 # Springer-Verlag Berlin Heidelberg 2014

Abstract Purpose This study aimed to compare fiberoptic nasopharyngoscopy during Mueller’s maneuver (FNMM) with fiberoptic nasopharyngoscopy with simulation of snoring (FNSS) for upper airway (UA) assessment in patients with obstructive sleep apnea and hypopnea syndrome. We also investigated the relationship between daytime endoscopic examinations and nocturnal pressure measurements. Methods We conducted a prospective, case-series study at Peking Union Medical College Hospital. All patients were evaluated by daytime FNMM and FNSS. The retropalatal and retroglossal regions were continuously video recorded during quiet breathing, FNMM, and FNSS. We calculated the narrowing rate and determined the level of obstruction and pattern of collapse (lateral, anterior-posterior, or concentric). Patients also underwent nocturnal pressure measurements to identify obstruction sites. Results Ninety-two patients were enrolled. FNMM and FNSS detected retropalatal obstruction in every case. Fifty-six and 38 patients had retroglossal obstruction detected by FNMM and FNSS, respectively. There was diagnostic agreement between FNMM and FNSS in 72 patients when diagnosing retroglossal obstruction, but the patterns of collapse were different using each technique. Pressure measurements showed that lower apnea and hypopnea index (AHI) and the proportion of lower AHI were significantly lower in the isolated retropalatal obstruction group than in the combined obstruction group diagnosed with either FNMM or FNSS (pfive events/h with snoring and excessive daytime sleepiness established by pre-study polysomnography; and reluctance to use, failure to tolerate, or lack of response to CPAP. Patients who had been acutely unwell in the previous 6 months, had unexplained fever or untreated infection, were pregnant, were unable to perform SS, or who had undergone previous surgery for OSAHS were excluded. All of the patients provided written consent for participation in the study. We recorded age, sex, height, and weight, and calculated body mass index (BMI) for each patient. All of the patients underwent daytime FNMM and FNSS examinations, and attended an overnight AG assessment.

AG pressure measurements An overnight hospital AG (ApneaGraph, MRA Medical, Gloucestershire, UK) assessment was performed in each subject between 23:00 pm and 05:00 am. Pharyngeal pressure and airflow and peripheral oxygen saturation were measured simultaneously and continuously to diagnose UA obstruction (apnea, hypopnea, and snoring) and to identify the segment of obstruction during sleep. This system comprises of a logger, catheter, pulse oximeter, and Apnea Analysis software (Apnea Analysis version 6.61, MRA Medical). The catheter contains two micropressure (P2 and P0) and two temperature sensors (TI and T0). The P2 sensor was positioned just below the soft palate, and the P0 sensor was placed in the esophagus to measure central respiratory drive; the difference between the two was used to determine the segment of obstruction. The catheter also measured changes in nasal airflow and mouth–nasal airflow with T1 in the nose and T0 behind the base of the tongue, respectively. Snoring was recorded with an internal microphone. Data were recorded into an internal memory card and then downloaded for analysis by an investigator who was blinded to the results of endoscopic examinations. Obstructive apnea was deemed to have occurred when there was cessation of breathing for >10 s associated with evidence of persistent respiratory effort. Hypopnea was defined as a >50 % decrease in airflow with a >4 % decrease in SaO2. Upper obstruction refers to the region above the P2 sensor, including the retropalatal region, nasopharynx and nose, and mainly the retropalatal region in adults. Lower obstruction refers to the region distal to the P2 sensor, including the retroglossal region and hypopharynx, and mainly the retroglossal region in adults [15]. The following parameters were used for analysis in this study: the AHI; lowest oxygen saturation (LSaO2); average oxygen saturation (ASaO2); the lower AHI (the number of apnea and hypopnea per hour caused by lower obstruction); and the proportion of the lower airway obstruction. Previous studies have reported that the use of continuous airway pressure and flow monitoring is reproducible [16], and that AG and polysomnography are not significantly different for measuring the AHI, the total number of apneic events, ASaO2, and LSaO2 [17]. We used the following definitions of OSAHS: mild, AHI ≥5 and ≤15; moderate, AHI >15 and ≤30; and severe, AHI >30. Endoscopic examinations Endoscopic examinations were performed by an experienced otorhinolaryngologist using a fiberoptic nasopharyngoscope with a 3.4-mm external diameter (Olympus Evis EXERA II 180, Olympus Medical Systems, Tokyo, Japan). Participants’ nasal passages were sprayed with aerosolized 1 % ephedrine and 1 % pontocaine before the examination. Patients were shown how to perform MM and SS. Patients were asked to

Sleep Breath

relax and snore in their usual way with their mouth open and were allowed to practice SS. Patients adopted a supine position in which the Frankfurt plane was perpendicular to the floor. A lubricated endoscope was inserted through a nostril and advanced until the epiglottis was visible. The entire UA was examined, with emphasis on the retropalatal and retroglossal levels. The uvula and tip of the epiglottis were used as landmarks for retropalatal and retroglossal levels. The physician ensured that the endoscope was held firmly at the nares during SS and MM. During MM, we used the right index finger fixing the endoscope between the index finger and alar rim. We simultaneously used the thumb and middle fingers kneading on both sides of the nose to guarantee no flow at the nose with the scope in place. SS and MM were repeated two or three times to ensure that stable images were recorded. Video images of the procedure were recorded (Endoscopic Videoimage Processing System, ZXMA Corporation, Beijing, China). Six still images of the retropalatal and retroglossal levels during quiet breathing (end of quiet expiration), MM, and SS were captured for each patient and saved in bitmap format.

lateral and AP dimensions were measured through the center of the airway. Collapsibility was calculated by dividing the difference in measurement obtained during quiet respiration (N) and during MM or SS, with the original measurement obtained during quiet respiration and expressed as a percentage according to the following formula:

Image analysis

Statistical analysis

Bitmap images were analyzed using ImageJ (National Institutes of Health [NIH], USA), which is a public domain Java image processing program derived from NIH Image. ImageJ was used in another study describing upper airway crosssectional area very recently [18]. One co-investigator who was blinded to the AG results analyzed each image five times, and the mean was calculated. Measurements of UA diameter in the lateral and anterior-posterior (AP) direction and crosssectional areas were determined for each image (Fig. 1). UA

Data were analyzed using SPSS Statistics version 17.0 (IBM, New York, NY). Continuous data are reported as mean±standard deviation (SD). If variables were not normally distributed, they are reported as medians with the interquartile range. The independent t test was used to compare groups when data were normally distributed, and the Wilcoxon W test was used when data were not normally distributed. The level of statistical significance (p value) was defined as 50 % of collapse as the cut-off for collapse at each site, defined obstruction of a level as >50 % reduction in the cross-sectional area, lateral or AP collapse as >50 % reduction in airway diameter in each direction, and concentric collapse as >30 % reduction in airway diameter in both directions.

Sleep Breath Table 2 Levels of obstruction and patterns of collapse during FNMM and FNSS

Results Patients’ characteristics and AG pressure measurements Ninety-six patients were examined, but four (one male and three females, 4.2 %) were excluded because they were unable to reproduce snoring while awake. Ultimately, 92 patients with a mean age of 40.5±8.5 years (range, 25–60 years) and a mean BMI of 28.1±3.8 kg/m2 (range, 20.0–49.8 kg/m2) were included in the analysis. Eighty-seven (94.6 %) participants had a BMI >24 kg/m2 and were thus considered overweight. The study population of Chinese patients was mainly men (88, 95.7 %). All of the patients were of Chinese Han ethnic origin, the predominant ethnicity in China. Of the 92 patients, 73 (79.3 %) had severe OSAHS, 11 (12.0 %) had moderate OSAHS, and eight (8.7 %) had mild OSAHS. The mean AHI was 49.3±20.8 events/h (range, 5.1– 79.7 events/h). The mean LSaO2 was 71.5±11.8 % (range, 43–89 %) and mean ASaO2 was 92.2±4.3 % (range, 79– 98 %). The median lower AHI was 8.2 events/h (range, 0.3– 71.2 events/h). The median proportion of lower airway obstruction was 21.0 % (range, 1.0–99.0 %). Video-endoscopic assessment Upper airway collapsibility at each level during FNMM and FNSS is shown in Table 1. The level of obstruction and patterns of lateral, AP, and concentric collapse are shown in Table 2. Retropalatal obstruction was detected in all of the participants by FNMM and FNSS. Agreement between the two techniques in the diagnosis of retropalatal obstruction was 100 %. Fifty-six and 38 patients had retroglossal obstruction during FNMM and FNSS, respectively. Thirty-seven of 92 patients had retroglossal obstruction during FNMM and FNSS, and 35 had no retroglossal obstruction during either FNMM or FNSS. There was diagnostic agreement between the techniques in 72 of 92 (78.3 %) patients in identifying whether there was retroglossal obstruction. Of 92 participants with retropalatal obstruction, 23 (25.0 %) had lateral collapse, Table 1 Upper airway collapsibility at the retropalatal and retroglossal levels during FNMM and FNSS

RP RP

Lat

RG AP

Conc

RG

Lat

AP

Conc

FNMM

92

23

4

65

56

49

1

6

FNSS

92

14

21

57

38

28

6

4

Abbreviations, FNMM fiberoptic nasopharyngoscopy with Mueller’s maneuver; FNSS fiberoptic nasopharyngoscopy with simulation of snoring; RP retropalatal obstruction; RG retroglossal obstruction; Lat lateral collapse; AP anterior-posterior collapse; Conc concentric collapse

four (4.3 %) had AP collapse, and 65 (70.7 %) had concentric collapse during FNMM. During FNSS, 14 (15.2 %) patients had lateral collapse, 21 (22.8 %) had AP collapse, and 57 (62.0 %) had concentric collapse. Of 56 participants with retroglossal obstruction detected by FNMM, 49 (87.5 %) had lateral collapse, one (1.8 %) had AP collapse, and six (10.7 %) had concentric collapse. Of the 38 patients with retroglossal obstruction detected by FNSS, 28 (73.7 %) had lateral collapse, six (15.8 %) had AP collapse, and four (10.5 %) had concentric collapse. Relationship between AG findings and obstructive level by endoscopy Because all of the patients were determined as having retropalatal obstruction by FNMM or FNSS, we focused on retroglossal obstruction. We detected no isolated retroglossal obstruction. Therefore, the patients were subdivided into two groups: those with isolated retropalatal obstruction and those with both retropalatal and retroglossal obstructions. We found that when using FNMM, the AHI, the lower AHI, and the proportion of the lower AHI were significantly lower in the isolated retropalatal obstruction group than in the combined obstruction group, but there were no significant differences in BMI, ASaO2, or LSaO2 between the groups (Table 3). For FNSS, the lower AHI and the proportion of the lower AHI were significantly lower in the isolated retropalatal obstruction group than in the combined obstruction group, but there were no significant differences in the AHI, BMI, ASaO2, or LSaO2 between the groups (Table 4).

Percentage of collapse

RP RG

FNMM FNSS FNMM FNSS

0–25 %

26–50 %

51–75 %

76–100 %

0 0 22 45

0 0 14 9

2 5 16 19

90 87 40 19

Abbreviations, RP retropalatal level; RG retroglossal level; FNMM fiberoptic nasopharyngoscopy with Mueller’s maneuver; FNSS fiberoptic nasopharyngoscopy with simulation of snoring

Discussion There is a wide variety of surgical approaches to reconstruct the UA or modify its anatomical structures to prevent collapse and obstruction during sleep. The objective of UA evaluation is to identify all sites of obstruction and match appropriate interventions with individual patients. Surgery is more likely

Sleep Breath Table 3 Comparison of AG findings in patients who had isolated retropalatal obstruction (n=36) with those who had combined obstruction (n=56) as detected by FNMM Variables

Isolated RP

Combined

t-statistic or p value z-statistic

BMI (kg/m2) AHI (/h)

28.4±3.3

27.9±4.2

0.581(t)

0.563

42.1±23.0

54.0±18.0

−2.778(t)

0.007 0.382

ASaO2

0.927±0.043

0.919±0.042

0.878(t)

LSaO2

0.731±0.137

0.705±0.104

1.041(t)

0.301

LAHI (/h)

2.2 (1.0, 11.5)

27.8(3.8, 49.8)

−4.273(z)

0.000

LOWER%

0.09(0.04, 0.20)

0.50 (0.12, 0.86) −3.966(z)

0.000

Abbreviations, AG ApneaGraph; FNMM fiberoptic nasopharyngoscopy with Mueller’s maneuver; isolated RP isolated retropalatal obstruction; combined combined obstruction; BMI body mass index; AHI apneahypopnea index; ASaO2 average oxygen saturation; LSaO2 lowest oxygen saturation; LAHI lower AHI; LOWER% the proportion of lower AHI. Normally distributed data are presented as means±standard deviation and were analyzed using the independent t test. Data that were not normally distributed are presented as the median (interquartile range) and were analyzed using the Wilcoxon W test

to be successful if the specific region of obstruction has been identified beforehand. However, identifying the level of obstruction can be challenging and there is no standard diagnostic technique to identify the exact site. Clinical judgments are presently informed by comprehensive analysis of the findings from a variety of examinations. This situation encourages development of new methods to obtain reliable information in UA evaluation. Borowiecki and Sassin [7] performed the first endoscopic evaluation of dynamic changes of the pharynx with MM in

Table 4 Comparison of AG findings in patients who had isolated retropalatal obstruction (n=54) with those who had combined obstruction (n=38) as detected by FNSS Variables

Isolated RP

Combined

BMI 27.9±2.9 28.3±4.9 (kg/m2) AHI (/h) 46.9±22.3 52.8±18.2 0.927±0.039 0.916±0.048 ASaO2 LSaO2 0.724±0.119 0.705±0.104 LAHI (/h) 3.8 (1.7,19.4) 31.1(3.7,49.7) LOWER% 0.12 (0.06,0.44) 0.65 (0.13,0.88)

t-statistic or p value z-statistic −0.380(t)

0.705

−1.361(t) 1.186(t) 0.865(t) −2.875(z) −3.027(z)

0.177 0.239 0.389 0.004 0.002

Abbreviations, AG Apnea Graph; FNSS fiberoptic nasopharyngoscopy with simulation of snoring; isolated RP isolated retropalatal obstruction; combined combined obstruction; BMI body mass index; AHI apneahypopnea index; ASaO2 average oxygen saturation; LSaO2 lowest oxygen saturation; LAHI lower AHI; LOWER% the proportion of lower AHI. Normally distributed data are presented as means±standard deviation and were analyzed using the independent t test. Data that were not normally distributed are presented as the median (interquartile range) and were analyzed using the Wilcoxon W test

awake patients with OSAHS. In addition to static narrowing, dynamic narrowing of the retropalatal and retroglossal regions contributes to the pathophysiology of OSAHS [19]. FNMM is widely used to evaluate the UA before surgery to improve patient selection and predict outcome and postoperative changes in the UA [8, 9]. These mean that FNMM during the wakeful state can grasp some features of obstructive sites during sleep. Snoring is a sign of increased UA resistance and is the most common symptom suggesting obstructive sleep apnea. Won and colleagues [20] found that the acoustic characteristics of snoring differed according to the site of UA obstruction as determined by sleep videofluoroscopy. Most people can simulate snoring, and endoscopic examination of the UA during SS can predict sleep-disordered breathing [11]. Taken together, these findings suggest that FNSS may yield important information on the likely level of UA obstruction during sleep. Endoscopic UA evaluation is performed during wakefulness, when muscle tone and respiratory drive may be different to that during sleep. Therefore, the physiological factors that contribute to OSAHS may not be accurately evaluated. Sleep nasendoscopy allows visualization of the site and mechanism of snoring and UA obstruction in patients with OSAHS. Early sleep nasendoscopy has been undertaken during natural sleep [21], but this technique is time consuming and labor intensive. Subsequently, Croft and Pringle assessed the UA in OSAHS patients under sedation [22], which allowed determination of the pattern of UA narrowing and obstruction under sedation. Since the first publication in 1991, drug-induced sleep endoscopy (DISE) has increased in popularity and become the most widespread diagnostic tool for UA endoscopic evaluation. Evidence is emerging that certain DISE findings are related to treatment outcome, and DISE is an important selection tool in treatment advice because there is high interest in the prospective prediction of non-CPAP treatment outcome [2, 4–6, 23]. While it has been argued that sedation would result in hypotonia of the genioglossus muscle, which may yield misleading results [24]. As described by Stuck and Maurer [25], some researchers [26, 27] consider that nasendoscopy during sedation is only partially equivalent to nasendoscopy during natural sleep. Furthermore, endoscopy under sedation is usually performed for a short time and cannot reflect UA changes at different sleep stages. In OSAHS, UA collapse occurs most often in the retropalatal and retroglossal levels. In our study, we found that retropalatal obstruction was more common than retroglossal obstruction in either FNMM or FNSS. Indeed, all of the patients had retropalatal obstruction. This finding is similar to the previous findings under DISE. Ravesloot et al. reported that retropalatal collapse was most frequently observed (83 %) in a cohort of 100 OSAHS patients [28] and Vroegop et al. reported retropalatal collapse (81 %) in 1249 SDB cases [29]. Although FNSS was as effective as FNMM

Sleep Breath

in detecting retropalatal obstruction, more patients with retroglossal obstruction were detected using FNMM (56 patients) compared with FNSS (38 patients). There was agreement between FNSS and FNMM in 72 of 92 patients when diagnosing retroglossal obstruction. We also found that FNSS and FNMM could detect patterns of collapse. Patterns of collapse under DISE are helpful for optimizing selection of patients for UA surgery [4, 6, 30]. In our study, FNMM detected more lateral and concentric collapse at the retropalatal and retroglossal levels than FNSS, and FNSS detected more AP collapse than FNMM. These differences may be explained by differing physiological consequences of the maneuvers: MM creates negative pressure in the UA, mimicking the situation of obstructive sleep apnea, while SS simulates vibration of anatomical structures in the narrowed UA. Both maneuvers lower downstream intraluminal pressure, although the pressure might become lower with MM than with SS because SS does not have complete obstruction. A difference between the methods is that MM is “static” (no flow) and SS is “dynamic”. To determine whether endoscopic evaluation can reflect airway collapse during the night, we compared it with AG. The AG technique has the advantage of being able to measure nocturnal UA obstruction and identify the segment of obstruction during sleep. It can detect retropalatal or retroglossal obstruction for each obstructive event during a whole night’s sleep and allow dynamic nocturnal studies of the UA in OSAHS. A limitation of AG is that pressure measurements can usually only determine the lowest obstructed airway level, and it is difficult to assess whether the retropalatal airway is also collapsed in retroglossal obstructive events. Furthermore, AG cannot identify the structure that is causing or contributing to the obstruction. We usually combine AG with morphological examinations in our clinical practice. We chose the lower AHI and the proportion of the lower AHI for analysis in our study because lower obstruction does not occur during AGmeasured upper obstruction. Of the 92 patients in our cohort, the lower AHI ranged from 0.3 to 71.2 events/h and the proportion of lower airway obstruction ranged from 1.0 to 99.0 %. This suggests that all of our patients had some degree of multilevel obstruction. Our findings showed that retroglossal obstruction, as determined by either FNMM or FNSS, was associated with lower airway obstruction as identified by AG, which is mainly the retroglossal region. This suggests that the level of airway obstruction, which was observed during FNMM or FNSS during wakefulness, can predict the level of obstruction during sleep. Patients with combined obstruction as determined by FNMM had more severe AHI than those with isolated retropalatal obstruction. This finding suggests that more severe OSAHS comes with more multi-level collapse, which is consistent with the previous findings under DISE [28, 29].

Daytime FNMM and FNSS appear to be reliable methods of evaluating the level of UA obstruction in OSAHS, and FNSS might also provide different information regarding patterns of collapse and retroglossal collapse. Findings on the level of collapse and patterns of collapse detected by these two techniques will help the surgeons to make decisions on the surgical technique in individual patients. Although FNSS and FNMM may not completely mimic an apneic episode on the extent of obstruction and patterns of collapse during daytime, FNMM and FNSS provide information on intrinsic soft tissue tone and collapsibility, which are associated with anatomical and physiological changes in the UA during sleep in OSAHS. A limitation of this study is that the pattern and extent of UA collapse that was observed with FNMM and FNSS were dependent on the patients’ effort. Although participants were taught to perform MM with maximal effort, their efforts were highly variable. Furthermore, a small proportion of patients could not simulate snoring and needed practice before the examination. Both of these factors may have affected the results. Despite of our findings that FNMM and FNSS were reliable and easy methods of evaluating the UA when patients were awake, neither can completely reflect the changes that occur during sleep. In the absence of sufficient data of postsurgical sleep studies, the success of surgeries using this approach has not been validated in this cohort. Further studies are required to examine the ability of FNSS to predict surgical success. Our findings are based on preliminary research of Chinese patients (mainly men), which need to be confirmed in other ethnic groups.

Conclusions FNSS and FNMM are low risk, widely available, and easy techniques. These techniques yield reliable information on the dynamic anatomy and physiology of UA obstruction in patients with OSAHS. FNSS may provide some different information regarding retroglossal obstruction and patterns of collapse from FNMM. Both techniques can help the surgeons to make decisions regarding the surgical technique in individual patients.

Disclosure The authors declare that they have no conflict of interest.

References 1. Ravesloot MJ, de Vries N, Stuck BA (2014) Treatment adherence should be taken into account when reporting treatment outcomes in obstructive sleep apnea. Laryngoscope 124:344–345

Sleep Breath 2. De Vito A, Carrasco Llatas M, Vanni A, Bosi M, Braghiroli A, Campanini A, de Vries N, Hamans E, Hohenhorst W, Kotecha BT, Maurer J, Montevecchi F, Piccin O, Sorrenti G, Vanderveken OM, Vicini C (2014) European position paper on drug-induced sedation endoscopy (DISE). Sleep Breath. doi:10.1007/s11325-014-0989-6 3. Strollo PJ Jr, Soose RJ, Maurer JT, de Vries N, Cornelius J, Froymovich O, Hanson RD, Padhya TA, Steward DL, Gillespie MB, Woodson BT, Van de Heyning PH, Goetting MG, Vanderveken OM, Feldman N, Knaack L, Strohl KP, STAR Trial Group (2014) Upper-airway stimulation for obstructive sleep apnea. N Engl J Med 370:139–149 4. Vanderveken OM, Maurer JT, Hohenhorst W, Hamans E, Lin HS, Vroegop AV, Anders C, de Vries N, Van de Heyning PH (2013) Evaluation of drug-induced sleep endoscopy as a patient selection tool for implanted upper airway stimulation for obstructive sleep apnea. J Clin Sleep Med 9:433–438 5. Johal A, Battagel JM, Kotecha BT (2005) Sleep nasendoscopy: a diagnostic tool for predicting treatment success with mandibular advancement splints in obstructive sleep apnoea. Eur J Orthod 27: 607–614 6. Koutsourelakis I, Safiruddin F, Ravesloot M, Zakynthinos S, de Vries N (2012) Surgery for obstructive sleep apnea: Sleep endoscopy determinants of outcome. Laryngoscope 122:2587–2591 7. Borowiecki BD, Sassin JF (1983) Surgical treatment of sleep apnea. Arch Otolaryngol 109:508–512 8. Hsu PP, Tan AK, Tan BY, Gan EC, Chan YH, Blair RL, Lu PK (2007) Uvulopalatopharyngoplasty outcome assessment with quantitative computer-assisted videoendoscopic airway analysis. Acta Otolaryngol 127:65–70 9. Sher AE, Thorpy MJ, Shprintzen RJ, Spielman AJ, Burack B, McGregor PA (1985) Predictive value of Müller maneuver in selection of patients for uvulopalatopharyngoplasty. Laryngoscope 95:1483–1487 10. Pevernagie D, Aarts RM, De Meyer M (2010) The acoustics of snoring. Sleep Med Rev 14:131–144 11. Herzog M, Metz T, Schmidt A, Bremert T, Venohr B, Hosemann W, Kaftan H (2006) The prognostic value of simulated snoring in awake patients with suspected sleep-disorder breathing: Introduction of a new technique of examination. Sleep 29:1456–1462 12. Hudgel DW (1986) Variable site of airway narrowing among obstructive sleep apnea patients. J Appl Physiol 61:1403–1409 13. Osnes T, Rollheim J, Hartmann E (2002) Effect of UPPP with respect to site of pharyngeal obstruction in sleep apnoea: Follow-up at 18 months by overnight recording of airway pressure and flow. Clin Otolaryngol Allied Sci 27:38–43 14. Tvinnereim M, Mitic S, Hansen RK (2007) Plasma radiofrequency preceded by pressure recording enhances success for treating sleep related breathing disorders. Laryngoscope 117:731–736 15. Yu R, Li W, Huo H, Shen P, Tian X (2011) Short daytime ApneaGraph for initial case selection of obstructive sleep apneahypopnea syndrome before surgery. Eur Arch Otorhinolaryngol 268:1663–1669

16. Rollheim J, Tvinnereim M, Sitek J, Osnes T (2001) Repeatability of sites of sleep-induced upper airway obstruction. A 2-night study based on recordings of airway pressure and flow. Eur Arch Otorhinolaryngol 258:259–264 17. Singh A, Al-Reefy H, Hewitt R, Kotecha B (2008) Evaluation of ApneaGraph in the diagnosis of sleep-related breathing disorders. Eur Arch Otorhinolaryngol 26:1489–1494 18. Safiruddin F, Vanderveken OM, de Vries N, Maurer JT, Lee K, Ni Q, Strohl KP (2014) Effect of upper-airway stimulation for obstructive sleep apnoea on airway dimensions. Eur Respir J. doi:10.1183/ 09031936.00059414 19. Abdullah B, Rajet KA, Abd Hamid SS, Mohammad WM (2011) A videoendoscopic evaluation of the upper airway in South East Asian adults with obstructive sleep apnea. Sleep Breath 15:747–754 20. Won TB, Kim SY, Lee WH, Han DH, Kim DY, Kim JW, Rhee CS, Lee CH (2012) Acoustic characteristics of snoring according to obstruction site determined by sleep videofluoroscopy. Acta Otolaryngol 132:S13–S20 21. Borowiecki B, Pollak CP, Weitzman ED, Rakoff S, Imperato J (1978) Fibro-optic study of pharyngeal airway during sleep in patients with hypersomnia obstructive sleep-apnea syndrome. Laryngoscope 88(8 Pt 1):1310–1313 22. Croft CB, Pringle M (1991) Sleep nasendoscopy: a technique of assessment in snoring and obstructive sleep apnoea. Clin Otolaryngol Allied Sci 16:504–509 23. Eichler C, Sommer JU, Stuck BA, Hormann K, Maurer JT (2012) Does drug-induced sleep endoscopy change the treatment concept of patients with snoring and obstructive sleep apnea? Sleep Breath 17: 63–68 24. Bonora M, St John WM, Bledsoe TA (1985) Differential elevation by protriptyline and depression by diazepam of upper airway respiratory motor activity. Am Rev Respir Dis 131:41–45 25. Stuck BA, Maurer JT (2008) Airway evaluation in obstructive sleep apnea. Sleep Med Rev 12:411–436 26. Sadaoka T, Kakitsuba N, Fujiwara Y, Kanai R, Takahashi H (1996) The value of sleep nasendoscopy in the evaluation of patients with suspected sleep-related breathing disorders. Clin Otolaryngol Allied Sci 21:485–489 27. Jones TM, Ho MS, Earis JE, Swift AC, Charters P (2006) Acoustic parameters of snoring sound to compare natural snores with snores during ‘steady-state’ propofol sedation. Clin Otolaryngol 31:46–52 28. Ravesloot MJ, de Vries N (2011) One hundred consecutive patients undergoing drug-induced sleep endoscopy: Results and evaluation. Laryngoscope 121:2710–2716 29. Vroegop AV, Vanderveken OM, Boudewyns AN, Scholman J, Saldien V, Wouters K, Braem MJ, Van de Heyning PH, Hamans E (2014) Drug-induced sleep endoscopy in sleep-disordered breathing: report on 1,249 cases. Laryngoscope 124:797–802 30. Vanderveken OM (2013) Drug-induced sleep endoscopy (DISE) for non-CPAP treatment selection in patients with sleep-disordered breathing. Sleep Breath 17:13–14