ORIGINAL ARTICLE
Changes of snoring sound after relocation pharyngoplasty for obstructive sleep apnoea: the surgery reduces mean intensity in snoring which correlates well with apnoea–hypopnoea index Li, H.Y.,*† Lee, L.A.,*‡ Yu, J.F.,§ Lo, Y.L.,¶ Chen, N.H.,¶ Fang, T.J.,* Hsin, L.J.,* Lin, W.N.,* Huang, C.G.k & Cheng, W.N.** *Department of Otolaryngology, Sleep Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan, † Department of Sleep Medicine, Royal Infirmary Edinburgh, Edinburgh, UK, ‡Department of Otolaryngology, Xiamen Chang Gung Hospital, Xiamen, China, §Graduate Institute of Medical Mechatronics, Taiouan Interdisciplinary- Otolaryngology Laboratory, Chang Gung University, ¶Department of Pulmonary and Critical Care Medicine, Sleep Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, kDepartment of Laboratory Medicine, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, **Graduate School of Recreation and Sports Management, Taipei Physical Education College, Taipei, Taiwan Accepted for publication 5 October 2014 Clin. Otolaryngol. 2015, 40, 98–105
Objective: To investigate objective changes of snoring after surgery in patients with obstructive sleep apnoea (OSA) and correlate these with changes in the apnoea–hypopnoea index (AHI). Design: Prospective case series. Setting: A novel measurement, Snore Map, was used to analyse full-night snore sounds in terms of the maximal/ mean intensity, peak/mean frequency, snoring index and energy type (Snore Map type, 0–4). Snore sound was classified into three bands according to frequency energy spectrum: B1 (40–300 Hz), B2 (301–850 Hz) and B3 (851– 2000 Hz). Participants: Thirty-four male and two female OSA patients (mean age, 39 years; mean AHI, 53.1/h; mean body mass index, 26.8 kg/m2) with favourable anatomic structure were consecutively enrolled. Main outcome measures: Parameters of polysomnographies and Snore Maps at baseline and six months after
operation were compared. Statistical significance was set at P < 0.05. Results: Thirty-two patients completed this study. The mean reduction in the total-snoring index was insignificant but there were significant decreases in total mean intensity, total peak frequency, total mean frequency and Snore Map type after surgery. There were also significant decreases in the mean intensity in all three bands, the snoring index in B2/B3 and the mean frequency in B1 postoperatively. Changes in the total mean intensity, total mean frequency, B2 mean intensity and B3 snoring index positively correlated with change in the AHI. Conclusions: Relocation pharyngoplasty significantly decreases both the snoring sound intensity and snoring frequency. These reductions are directly proportional to the improvement of OSA.
Snoring is a well-known phenomenon and highly prevalent disorder in the general population.1 The breathing noise is characterised by vibrations of the soft palate, tongue, lateral pharyngeal wall and epiglottis in inspiratory and sometimes expiratory phase of the respiratory cycle during sleep.2 Snoring can also cause social embarrassment, daytime sleepiness and carotid artery atherosclerosis.3 It is generally based on the perception of the observer, and commonly used assessments are subjective questionnaires.4
Unfortunately, the perception of snoring is not only determined by the noise itself but also by psychological intricacies and the mental state of the observers, thereby often lacking reliability and reproducibility.5 The objective snoring index by polysomnography is not widely used due to the little information provided. To further understand the characteristics of snoring, other sophisticated methods such as snoring sound intensity and power spectrum have been implemented to objectively measure snoring sound and correlate it with sleep apnoea and site of obstruction.6,7 In previous studies, a novel snore sound analytic computer program, the Snore Map (Chang Gung Memorial Hospital, Taoyuan, Taiwan, R.O.C.), was developed to define, categorise and analyse the snoring sound.8,9
Correspondence: H.Y. Li, Department of Otorhinolaryngology-Head and Neck Surgery, Chang Gung Memorial Hospital. No. 5, Fu-Shin Street, Gueishan Township, Taoyuan County 333, Taiwan. Tel.: +886 3 3281200; Fax: +886 3 3979361; e-mail: [email protected]
98
© 2014 John Wiley & Sons Ltd Clinical Otolaryngology 40, 98–105
Changes of snoring sound after OSA surgery
Various medical and surgical options are provided to improve snoring and its detrimental effects on health.10,11 Among them, palatal surgery is commonly performed as the soft palate is considered to be the major snore sound generator.11,12 Relocation pharyngoplasty, a modified uvulopalatopharyngoplasty that advances the soft palate and splints the lateral pharyngeal wall, has been proven to be an safe and effective in the treatment of obstructive sleep apnoea (OSA).13–15 However, the effect of relocation pharyngoplasty on snoring is not fully understood. This study aimed to demonstrate the objective changes of snoring sound after relocation pharyngoplasty using the Snore Map and correlate these with changes in the apnoea–hypopnoea index (AHI) and subjective snoring questionnaire scores. Methods Ethical considerations
This prospective chart review study was approved by the Institutional Review Board at Chang Gung Memorial Hospital, Taoyuan, Taiwan, and informed consent was obtained from each subject.
99
None had undergone surgical treatment for snoring or were taking any sedative or hypnotic medications at the time of data collection. Evaluation
The body mass index, neck circumference, tonsil size and tongue position were measured according to Friedman’s staging.16 Subjective questionnaires such as the visual analogue scale of snoring severity, snoring outcome survey and spouse/bed partners survey,17 and the Epworth sleepiness scale18 were applied. Level I overnight inlaboratory polysomnographies (Nicolet UltraSom System, Madison, WI, USA) were performed to document sleep parameters in each patient. The polysomnography parameters used in this study were polysomnography-snoring index (/h), AHI (/h), apnoea index (/h), hypopnoea index (/h), arousal index (/h), mean arterial oxygen saturation (%), least arterial oxygen saturation (%) and percentage (%) of supine sleep. Apnoea was defined as a reduction in the peak thermal sensor excursion by at least 90% of baseline for at least 10 s. Hypopnoea was defined as a ≥30% decrease in nasal pressure signal excursions for at least 10 s, with desaturation of ≥4% from pre-event baseline or an arousal from sleep.19
Subjects
All patients with sleep-disordered breathing and seeking surgical treatment underwent complete history, physical examination, subjective snoring/daytime sleepiness, questionnaire scores and fibre-optic laryngoscopy. They all had poor response to conservative measures (i.e. lateral sleep, body weight reduction, medication for nasal obstruction) and intolerance or unwillingness to use continuous positive airway pressure therapy for polysomnography-diagnosed OSA (AHI >5/h). Favourable anatomic structures of relocation pharyngoplasty involved the following: a) webbed, thin and pliable posterior pillars and b) sagittal collapse of retro-palatal space ≥50% in M€ uller’s manoeuver during fibre-optic laryngoscopy.13 The exclusion criteria were old age (>60 years), morbid obesity (body mass index >35 kg/m2), gross maxillary and mandibular deformities, tongue base touching/pushing the epiglottis during M€ uller’s manoeuver and high modified Mallampati score (>IV).16 Thirty-six subjects (two women and 34 men; mean age, 39.0 8.6 years; mean body mass index, 26.8 2.6 kg/m2; mean AHI, 53.1 26.3; 9 Friedman stage I [25%], 25 stage II [69%] and two stage III [6%]) were prospectively enrolled and underwent simultaneous overnight polysomnography and snoring sound recording, relocation pharyngoplasty and follow-up examinations six months postoperatively. © 2014 John Wiley & Sons Ltd Clinical Otolaryngology 40, 98–105
Snore Map
All-night snoring sound recording was performed together with polysomnography examination using a consistent methodology that has been described previously.8,9 The testing room had a ambient noise level approximate 10 dB. The snoring sounds were detected by a central microphone (TEDS type 46AE, G.R.A.S. Corp., Holte, Denmark) positioned 100 cm above the patient’s head. The signals of the recorded snoring sounds were collected by portable data cards (PXI 4462, National Instruments Corp., Austin, TX, USA) and processed by digital recording software (Sound & Vibration Toolkit for Labview, National Instruments Corp., Austin, TX, USA) at sample rate of 44 100 Hz. The frequency power spectrum (range, 40–2000 Hz) was created by fast Fourier transformation.8 For analysing the all-night snoring signals, an automatic detection algorithm was developed to detect considerable snoring events based on the following principles: (i) energy higher than 0.05 au and (ii) duration between 0.6 and 4.0 s. The sensitivity and positive predictive value of this system for detecting snores in the preliminary study were 99.9% and 99.1%, respectively.9 A Snore Map (frequency energy spectrum of the all-night snoring sounds) was used for semiquantitative assessment of
100 H.Y. Li et al.
the snoring sound for each subject. The total frequency domain (40–2000 Hz) was classified into three bands: B1 (40–300 Hz), B2 (301–850 Hz) and B3 (851–2000 Hz) (Fig. 1).9 We previously defined four Snore Map types of snoring sounds in OSA patients: type 1 (monosyllabical B1 snore), type 2 (duplex B1 and B2 snore), type 3 (duplex B1 and B3) and type 4 (triplex B1 and B2 & B3 snore). We found that the energy sum of the all-night snoring sounds between 40 Hz and 2000 Hz in non-snoring subjects was always
Changes of snoring sound after relocation pharyngoplasty for obstructive sleep apnoea: the surgery reduces mean intensity in snoring which correlates well with apnoea–hypopnoea index Li, H.Y.,*† Lee, L.A.,*‡ Yu, J.F.,§ Lo, Y.L.,¶ Chen, N.H.,¶ Fang, T.J.,* Hsin, L.J.,* Lin, W.N.,* Huang, C.G.k & Cheng, W.N.** *Department of Otolaryngology, Sleep Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan, † Department of Sleep Medicine, Royal Infirmary Edinburgh, Edinburgh, UK, ‡Department of Otolaryngology, Xiamen Chang Gung Hospital, Xiamen, China, §Graduate Institute of Medical Mechatronics, Taiouan Interdisciplinary- Otolaryngology Laboratory, Chang Gung University, ¶Department of Pulmonary and Critical Care Medicine, Sleep Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, kDepartment of Laboratory Medicine, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, **Graduate School of Recreation and Sports Management, Taipei Physical Education College, Taipei, Taiwan Accepted for publication 5 October 2014 Clin. Otolaryngol. 2015, 40, 98–105
Objective: To investigate objective changes of snoring after surgery in patients with obstructive sleep apnoea (OSA) and correlate these with changes in the apnoea–hypopnoea index (AHI). Design: Prospective case series. Setting: A novel measurement, Snore Map, was used to analyse full-night snore sounds in terms of the maximal/ mean intensity, peak/mean frequency, snoring index and energy type (Snore Map type, 0–4). Snore sound was classified into three bands according to frequency energy spectrum: B1 (40–300 Hz), B2 (301–850 Hz) and B3 (851– 2000 Hz). Participants: Thirty-four male and two female OSA patients (mean age, 39 years; mean AHI, 53.1/h; mean body mass index, 26.8 kg/m2) with favourable anatomic structure were consecutively enrolled. Main outcome measures: Parameters of polysomnographies and Snore Maps at baseline and six months after
operation were compared. Statistical significance was set at P < 0.05. Results: Thirty-two patients completed this study. The mean reduction in the total-snoring index was insignificant but there were significant decreases in total mean intensity, total peak frequency, total mean frequency and Snore Map type after surgery. There were also significant decreases in the mean intensity in all three bands, the snoring index in B2/B3 and the mean frequency in B1 postoperatively. Changes in the total mean intensity, total mean frequency, B2 mean intensity and B3 snoring index positively correlated with change in the AHI. Conclusions: Relocation pharyngoplasty significantly decreases both the snoring sound intensity and snoring frequency. These reductions are directly proportional to the improvement of OSA.
Snoring is a well-known phenomenon and highly prevalent disorder in the general population.1 The breathing noise is characterised by vibrations of the soft palate, tongue, lateral pharyngeal wall and epiglottis in inspiratory and sometimes expiratory phase of the respiratory cycle during sleep.2 Snoring can also cause social embarrassment, daytime sleepiness and carotid artery atherosclerosis.3 It is generally based on the perception of the observer, and commonly used assessments are subjective questionnaires.4
Unfortunately, the perception of snoring is not only determined by the noise itself but also by psychological intricacies and the mental state of the observers, thereby often lacking reliability and reproducibility.5 The objective snoring index by polysomnography is not widely used due to the little information provided. To further understand the characteristics of snoring, other sophisticated methods such as snoring sound intensity and power spectrum have been implemented to objectively measure snoring sound and correlate it with sleep apnoea and site of obstruction.6,7 In previous studies, a novel snore sound analytic computer program, the Snore Map (Chang Gung Memorial Hospital, Taoyuan, Taiwan, R.O.C.), was developed to define, categorise and analyse the snoring sound.8,9
Correspondence: H.Y. Li, Department of Otorhinolaryngology-Head and Neck Surgery, Chang Gung Memorial Hospital. No. 5, Fu-Shin Street, Gueishan Township, Taoyuan County 333, Taiwan. Tel.: +886 3 3281200; Fax: +886 3 3979361; e-mail: [email protected]
98
© 2014 John Wiley & Sons Ltd Clinical Otolaryngology 40, 98–105
Changes of snoring sound after OSA surgery
Various medical and surgical options are provided to improve snoring and its detrimental effects on health.10,11 Among them, palatal surgery is commonly performed as the soft palate is considered to be the major snore sound generator.11,12 Relocation pharyngoplasty, a modified uvulopalatopharyngoplasty that advances the soft palate and splints the lateral pharyngeal wall, has been proven to be an safe and effective in the treatment of obstructive sleep apnoea (OSA).13–15 However, the effect of relocation pharyngoplasty on snoring is not fully understood. This study aimed to demonstrate the objective changes of snoring sound after relocation pharyngoplasty using the Snore Map and correlate these with changes in the apnoea–hypopnoea index (AHI) and subjective snoring questionnaire scores. Methods Ethical considerations
This prospective chart review study was approved by the Institutional Review Board at Chang Gung Memorial Hospital, Taoyuan, Taiwan, and informed consent was obtained from each subject.
99
None had undergone surgical treatment for snoring or were taking any sedative or hypnotic medications at the time of data collection. Evaluation
The body mass index, neck circumference, tonsil size and tongue position were measured according to Friedman’s staging.16 Subjective questionnaires such as the visual analogue scale of snoring severity, snoring outcome survey and spouse/bed partners survey,17 and the Epworth sleepiness scale18 were applied. Level I overnight inlaboratory polysomnographies (Nicolet UltraSom System, Madison, WI, USA) were performed to document sleep parameters in each patient. The polysomnography parameters used in this study were polysomnography-snoring index (/h), AHI (/h), apnoea index (/h), hypopnoea index (/h), arousal index (/h), mean arterial oxygen saturation (%), least arterial oxygen saturation (%) and percentage (%) of supine sleep. Apnoea was defined as a reduction in the peak thermal sensor excursion by at least 90% of baseline for at least 10 s. Hypopnoea was defined as a ≥30% decrease in nasal pressure signal excursions for at least 10 s, with desaturation of ≥4% from pre-event baseline or an arousal from sleep.19
Subjects
All patients with sleep-disordered breathing and seeking surgical treatment underwent complete history, physical examination, subjective snoring/daytime sleepiness, questionnaire scores and fibre-optic laryngoscopy. They all had poor response to conservative measures (i.e. lateral sleep, body weight reduction, medication for nasal obstruction) and intolerance or unwillingness to use continuous positive airway pressure therapy for polysomnography-diagnosed OSA (AHI >5/h). Favourable anatomic structures of relocation pharyngoplasty involved the following: a) webbed, thin and pliable posterior pillars and b) sagittal collapse of retro-palatal space ≥50% in M€ uller’s manoeuver during fibre-optic laryngoscopy.13 The exclusion criteria were old age (>60 years), morbid obesity (body mass index >35 kg/m2), gross maxillary and mandibular deformities, tongue base touching/pushing the epiglottis during M€ uller’s manoeuver and high modified Mallampati score (>IV).16 Thirty-six subjects (two women and 34 men; mean age, 39.0 8.6 years; mean body mass index, 26.8 2.6 kg/m2; mean AHI, 53.1 26.3; 9 Friedman stage I [25%], 25 stage II [69%] and two stage III [6%]) were prospectively enrolled and underwent simultaneous overnight polysomnography and snoring sound recording, relocation pharyngoplasty and follow-up examinations six months postoperatively. © 2014 John Wiley & Sons Ltd Clinical Otolaryngology 40, 98–105
Snore Map
All-night snoring sound recording was performed together with polysomnography examination using a consistent methodology that has been described previously.8,9 The testing room had a ambient noise level approximate 10 dB. The snoring sounds were detected by a central microphone (TEDS type 46AE, G.R.A.S. Corp., Holte, Denmark) positioned 100 cm above the patient’s head. The signals of the recorded snoring sounds were collected by portable data cards (PXI 4462, National Instruments Corp., Austin, TX, USA) and processed by digital recording software (Sound & Vibration Toolkit for Labview, National Instruments Corp., Austin, TX, USA) at sample rate of 44 100 Hz. The frequency power spectrum (range, 40–2000 Hz) was created by fast Fourier transformation.8 For analysing the all-night snoring signals, an automatic detection algorithm was developed to detect considerable snoring events based on the following principles: (i) energy higher than 0.05 au and (ii) duration between 0.6 and 4.0 s. The sensitivity and positive predictive value of this system for detecting snores in the preliminary study were 99.9% and 99.1%, respectively.9 A Snore Map (frequency energy spectrum of the all-night snoring sounds) was used for semiquantitative assessment of
100 H.Y. Li et al.
the snoring sound for each subject. The total frequency domain (40–2000 Hz) was classified into three bands: B1 (40–300 Hz), B2 (301–850 Hz) and B3 (851–2000 Hz) (Fig. 1).9 We previously defined four Snore Map types of snoring sounds in OSA patients: type 1 (monosyllabical B1 snore), type 2 (duplex B1 and B2 snore), type 3 (duplex B1 and B3) and type 4 (triplex B1 and B2 & B3 snore). We found that the energy sum of the all-night snoring sounds between 40 Hz and 2000 Hz in non-snoring subjects was always
