Original Research
The Effect of Tonsillectomy Alone in Adult Obstructive Sleep Apnea Andrew J. Senchak, DO, MS1, Alex J. McKinlay, MD2, Jason Acevedo, MD3, Brenda Swain, APRN, CNP3, Maitram Christine Tiu, PA-C, MPAS4, Brian S. Chen, MD4, Jon Robitschek, MD5, Douglas S. Ruhl, MD, MSPH6, Lawrence L. Williams, MD1, Macario Camacho, MD6, William C. Frey, MD7, and Peter D. O’Connor, MD, OD7
Otolaryngology– Head and Neck Surgery 1–5 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599815575721 http://otojournal.org
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
Keywords tonsillectomy, tonsils, adult, obstructive sleep apnea
Abstract Objective. The purpose of this study was to determine the effect of tonsillectomy as a single procedure in the treatment of adult obstructive sleep apnea (OSA).
Received October 16, 2014; revised February 5, 2015; accepted February 11, 2015.
Study Design. Prospective multi-institutional study evaluating adults with tonsillar hypertrophy scheduled to undergo tonsillectomy as an isolated surgery. Setting. Tertiary care medical centers within the US Department of Defense. Subjects and Methods. Adult subjects with tonsillar hypertrophy who were already scheduled for tonsillectomy were enrolled from October 2010 to July 2013. Subjects underwent physical examination, Epworth Sleepiness Scale, Berlin Questionnaire, and polysomnogram before surgery and after. Collected data included demographics, questionnaire scores, apnea-hypopnea index (AHI), and lowest saturation of oxygen. Results. A total of 202 consecutive subjects undergoing tonsillectomy were enrolled. The final analysis included 19 subjects testing positive for OSA. The mean age was 27.9 years; mean body mass index, 29.6; median tonsil size, 3; and most frequent Friedman stage, 1. The AHI before surgery ranged from 5.4 to 56.4 events per hour. The mean AHI decreased from 18.0 to 3.2 events per hour after surgery, a reduction of 82%. The responder rate—with subjects achieving at least a 50% reduction of AHI to a value \15—was 94.7%. Following tonsillectomy, there were statistically significant reductions in median lowest saturation of oxygen level and Epworth Sleepiness Scale and Berlin scores. Conclusions. Adult tonsillectomy alone has beneficial effect in OSA management, particularly in young overweight men with large tonsils, moderate OSA, and low Friedman stage.
O
bstructive sleep apnea (OSA) is a condition with significant morbidity and mortality, with an estimated prevalence of 24% of men and 9% of women.1 It has been linked to daytime hypersomnolence, fatigue, impaired work productivity, and development of chronic cardiovascular disease.2 While medical therapy with continuous positive airway pressure has been shown to reduce symptoms and decrease rates of mortality, many patients experience mask discomfort, aerophagia, mask air leak, and claustrophobia with use of a continuous positive airway pressure machine. In fact, objective data collected on patients show a \50% compliance rate, with only 4 hours of nightly continuous positive airway pressure use.3 Surgical procedures to correct OSA or those that serve an adjunctive role remain a crucial therapeutic option. Currently, there are limited data on tonsillectomy alone as treatment for adult OSA, and there is a need for studies with high-level evidence.4 The American Academy of Sleep Medicine practice guidelines for the surgical modification
1
Walter Reed National Military Medical Center, Bethesda, Maryland, USA Darnall Army Medical Center, Fort Hood, Texas, USA 3 Reynolds Army Community Hospital, Fort Sill, Oklahoma, USA 4 Madigan Army Medical Center, Tacoma, Washington, USA 5 Landstuhl Regional Medical Center, Landstuhl, Germany 6 Tripler Army Medical Center, Honolulu, Hawaii, USA 7 San Antonio Military Medical Center, San Antonio, Texas, USA 2
The views herein are the private views of the authors and do not reflect the official views of the Departments of the Army, Air Force, or Navy, nor the Department of Defense. Corresponding Author: Andrew J. Senchak, DO, MS, Department of Otolaryngology, Walter Reed National Military Medical Center, 8901 Wisconsin Ave, Bethesda, MD 20889, USA. Email: [email protected]
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
2
Otolaryngology–Head and Neck Surgery
of the upper airway to treat OSA do not mention tonsillectomy as an option, except when performed with a uvulopalatopharyngoplasty.5 These guidelines do not recommend uvulopalatopharyngoplasty, with or without tonsillectomy, as a sole procedure to treat OSA. As evidence of this debate, some third-party payors do not reimburse for tonsillectomy to treat OSA, as they consider it to be experimental for this indication.6 The lack of high-level evidence to support upper airway surgery may be due in part to patient selection, surgical technique selection, or specific outcomes measured in the published literature. This study was undertaken to determine if adult tonsillectomy alone improves OSA. As with any therapy to treat OSA, the outcome goals were to resolve clinical signs and symptoms, to normalize respiratory parameters (ie, apneahypopnea index [AHI] and oxygen saturation), to decrease sleepiness, and to restore sleep continuity.5 The objective of this study was to measure these outcomes in otherwise healthy adult subjects with enlarged tonsils who underwent tonsillectomy.
Methods A multi-institutional prospective trial was conducted at San Antonio Military Medical Center, Walter Reed National Military Medical Center, Darnall Army Medical Center, Reynolds Army Community Hospital, Bayne-Jones Army Community Hospital, Landstuhl Regional Medical Center, Madigan Army Medical Center, and Tripler Army Medical Center. The institutional review board at each medical facility approved the research protocol, and subjects were recruited by the otolaryngology service at each institution. Since the military employs a high percentage of young healthy adults, this setting served as an ideal group from which to recruit a homogeneous population of subjects with enlarged tonsils. Enrollment for the study commenced from October 2010 through July 2013. All subjects were recruited from adults aged 18 years with tonsils size 2 to 4 who were already scheduled to undergo tonsillectomy, independent of the research project, for reasons of chronic tonsillitis, recurrent tonsillitis, or halitosis. The Friedman classification system was used to measure tonsils size and palate position, and the Friedman stage was calculated.7 We excluded subjects with a history of pharyngeal surgery, significant nasal pathology (deviated nasal septum that obstructed the nasal cavity, nasal polyposis, or sinusitis), or dental malocclusion (as determined by the relationship of the maxillary first molar to the mandibular first molar). Subjects completed 2 questionnaires: the Epworth Sleepiness Scale (ESS)8 and category 1 of the Berlin Questionnaire.9 Each subject was then given an ambulatory sleep monitor (ApneaLink Plus; ResMed Corp, San Diego, California), with instructions on how to use the monitor once overnight. This initial baseline sleep monitor use was deemed time T0. Subjects then underwent complete removal of the bilateral palatine tonsils, performed with electrocautery in all cases, with no other concurrent procedures performed. Approximately 2 to 6
months after surgery, subjects again filled out both questionnaires and repeated use of the ambulatory sleep monitor, deemed time T1. Data collected for the study included patient characteristics, physical examination findings, ESS and Berlin Questionnaire scores, the AHI, and the lowest saturation of oxygen. The AHI and lowest saturation of oxygen were calculated by a board-certified, fellowship-trained sleep physician (either W.C.F. or P.D.O.) by manually reviewing the ApneaLink Plus signal data. Respiratory events were scored using definitions established by the American Academy of Sleep Medicine, and hypopneas were scored using the ‘‘recommended’’ definition, according to the 2007 guidelines.10 OSA was defined as an AHI 5.0. Severity of OSA was defined as follows: mild (AHI, 5.0-14.9), moderate (AHI, 15.0-29.9), and severe (AHI, 30.0). The ESS and the snoring section of the Berlin Questionnaire (category 1) were used to measure symptoms of sleepiness and snoring, respectively. ESS scores from 0 to 10 were defined as no daytime hypersomnolence, and scores .10 indicated hypersomnolence. Scores of 0 or 1 from category 1 of the Berlin Questionnaire were considered normal and from 2 to 6, positive for significant snoring. Resolution of OSA was based on a responder rate, defined as reduction of AHI by at least 50% to an absolute index of \15 events per hour at T1. Data were analyzed and described using number counts, proportions, medians with interquartile ranges, or means with standard deviations or 95% confidence intervals. A chi-square analysis was used to compare proportions. For continuous data, a paired t test was used to compare means for data that were normally distributed, and a Wilcoxon rank-sum test was used to compare medians for nonparametric data. Statistical significance was defined as a P value \.05. All statistical analyses were performed using Stata, version 11.0 (StataCorp LP, College Station, Texas).
Results A total of 202 subjects were enrolled in the study, and 46 were excluded for reasons listed in Figure 1. The demographic characteristics of subjects included versus excluded are shown in Table 1. There was no statistically significant difference between these groups. Of those subjects who were included, 17.9% (n = 28 of 156) tested positive for OSA on preoperative polysomnogram. After accounting for those lost to follow-up, a final total of 19 subjects with OSA completed postoperative polysomnogram evaluation and were available for outcome analysis. Demographic characteristics for the 19 subjects completing data analysis are shown in Table 2. The median tonsil size was 3, and the most frequent Friedman stage was 1. The mean time from baseline (T0) to follow-up (T1) was 77.2 6 5.2 days. Individual outcome data before and after surgery for all 19 subjects completing the study are listed in Table 3. The AHI at baseline time T0 ranged from 5.4 to 56.4 events per hour. All 19 subjects had a reduction of AHI by at least 50% following surgery, regardless of preoperative tonsil size, Friedman stage, OSA severity, or body
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
Senchak et al
3 Table 2. Characteristics of Subjects Completing Data Analysis (n = 19).a
Consecutive Subjects Enrolled (n = 202)
Characteristics Subjects Excluded (n = 46) -PSG with unusable data (n = 24) -Surgery canceled (n = 15) -Could not tolerate PSG (n = 7)
Subjects Included (n = 156) -All successfully completing pre-op PSG
PSG Negative for OSA (n = 128)
PSG Positive for OSA (n = 28)
Lost to Follow-up (n = 9)
Subjects Completing Post-Op PSG (n = 19)
Age,a y Body mass index,a kg/m2 Maleb Caucasianb Tonsil sizec Friedman palate positionc Friedman stageb I II III
Values 27.9 6 10.1 29.6 6 6.0 84.2 42.1 3 (3, 3) 1 (1, 2) 68.4 21.1 10.5
a
Figure 1. Flow of study subjects. OSA, obstructive sleep apnea; PSG, polysomnogram.
Table 1. Demographic Characteristics of Subjects: Included vs Excluded.
Age,a y Body mass index,a kg/m2 Menb Caucasianb
Included (n = 156)
Excluded (n = 46)
P Value
26.5 6 7.2 27.2 6 4.7 71.8 71.8
24.6 6 4.5 27.12 6 3.9 63.0 67.4
.094 .987 .253 .564
a
Mean 6 standard deviation. Percentage.
b
mass index. The responder rate was 94.7% (n = 18 of 19) with only 1 subject (No. 12) failing to achieve an AHI \15 events per hour at T1. This subject had a preoperative body mass index of 42.8 and tonsil size of 3 and was Friedman stage 3. The mean and median change in outcome measures from T0 to T1 is shown in Table 4. The mean body mass index did not significantly change after surgery. There was a statistically significant decrease in the mean AHI from 18.0 to 3.2 events per hour following surgery (Figure 2), a reduction by 82%. There was a statistically significant increase in the median lowest saturation of oxygen value after surgery, while there was a statistically significant decrease in the median ESS and Berlin scores.
Discussion Evidence has shown that removing tonsils and adenoids in children with OSA provides a significant role in its treatment.11-13 A consistent improvement in respiratory parameters measured by polysomnogram and in quality-of-life surveys is seen after adenotonsillectomy in children.14 Current surgical approaches for OSA in adults can include many options, such as nasal surgery, oropharyngeal expansion procedures, base of tongue reduction, and maxillomandibular advancement. The goal of this project was to prospectively
Mean 6 standard deviation. Percentage. c Median (interquartile range). b
evaluate subjects already scheduled for tonsillectomy to determine the degree to which OSA can be improved after surgery. Subjects already met indications for tonsillectomy independent of the research project and were then screened for OSA. As expected with such a population, a significant number of the subjects did not test positive for sleep apnea. It is well established that ambulatory sleep testing may underestimate obstructive breathing events and is typically not recommended as a screening procedure for general populations.15 Despite these limitations, however, 17.9% of subjects scheduled for tonsillectomy for routine indications were found to have some degree of OSA. The final sample size, while small, was powered adequately to demonstrate a statistically significant change in AHI following tonsillectomy and a considerable responder rate of 94.7%. This subject population on average had enlarged tonsils, was predominantly men and overweight, had moderate OSA, and a low Friedman stage. Of note, 2 subjects had size 2 tonsils. While the intent was to enroll subjects with tonsil size 3 or 4, these 2 subjects were reclassified as size 2 based on intraoperative examination findings and were kept in the study. Despite having smaller tonsils, these subjects still responded to surgery and had resolution of OSA. The only subject who failed to respond to surgery did have size 3 tonsils but was morbidly obese and had severe OSA. This suggests a possible correlation between these factors and surgical outcome. All subjects, however, had a reduction of AHI by at least 50% following surgery. Subjects had a median reduction in ESS scores after treatment, although the postoperative level did show mild persistent hypersomnolence at follow-up. Snoring, however, did improve with median Berlin Questionnaire score levels resolving to normal. A strength of the study was that the subject population was drawn from patients already scheduled for tonsillectomy and then screened for OSA. This was done to avoid bias in determining which should undergo tonsillectomy for
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
4
Otolaryngology–Head and Neck Surgery
Table 3. Outcome Measures (n = 19). Body Mass Index, kg/m2
Apnea-Hypopnea Index, Events/h
No.
Tonsil Size
Palate Position
Friedman Stage
T0
T1
T0
T1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4
1 1 1 1 1 1 2 1 3 1 2 2 3 1 2 2 2 1 1
2 2 1 3 1 1 1 1 2 1 1 3 2 1 1 1 1 1 1
25.0 27.0 23.0 41.8 26.7 30.8 25.0 24.6 32.7 29.1 30.4 42.7 34.8 21.7 29.6 24.5 29.6 25.7 36.8
24.7 27.2 23.2 41.8 27.4 30.9 23.7 22.9 32.5 31.9 29.9 41.8 34.9 21.6 30.2 24.6 28.2 26.5 35.2
17.0 6.3 7.0 11.1 22.5 10.1 5.4 29.4 37.0 16.9 15.4 56.4 33.3 7.6 26.5 7.2 9.5 7.1 16.4
1.7 0.6 3.2 4.4 0.9 3.1 1.4 2.5 2.4 1.1 5.8 24.8 2.0 0.7 2.7 0.0 3.7 0.4 0.2
Abbreviations: T0, before surgery; T1, after surgery.
Table 4. Change in Outcome Measures Before (T0) and After (T1) Surgery (n = 19). Time T0
T1
P Value
29.4 6 5.9 .558 Body mass index,a kg/m2 29.6 6 6.0 AHI,b events/h, baseline 18.0 (11.5, 24.5) 3.2 (0.6, 5.9) \.001 LSAT,c percentage O2 85 (82, 90) 91 (87, 94) .008 ESSc score 12 (10, 15) 9 (5, 12) .014 Berlinc score 4 (3, 4) 1 (0, 3) .013 Abbreviations: AHI, apnea-hypopnea index; ESS, Epworth Sleepiness Scale; LSAT, lowest saturation of oxygen. a Mean 6 standard deviation. b Mean (95% confidence interval). c Median (interquartile range).
treatment of OSA. One limitation of the study was an unexpectedly high rate of unusable ambulatory monitor data. This was due to improper use of the unattended portable monitor and subsequent poor signal tracing. However, in some subjects who were willing to repeat use of the monitor, additional demonstration on proper use was provided, and a second recording was successful at obtaining data. Several subjects were also lost to follow-up. They either failed to return for a repeat visit or did not return phone calls. The young age of the subjects, the voluntary nature of the research, or the improvement in symptoms after surgery could have contributed to lack of follow-up. In addition, the
Figure 2. Apnea-hypopnea index (AHI) values before and after surgery.
military is a highly mobile population, and overseas deployment or change of duty station may have also influenced the dropout rate.
Conclusion Adult tonsillectomy alone has beneficial effect in the treatment of OSA, particularly in young overweight men with moderate sleep apnea, large tonsils, and a low Friedman stage. In this patient population, the AHI can be normalized in a majority of patients over a short-term period, with an approximately 82% reduction in AHI seen in all-comers in
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
Senchak et al
5
this study. Oxygen desaturations, sleepiness, and snoring are likely to improve as well. Patients with morbid obesity or severe OSA may be less likely to respond to surgery despite having large tonsils. A long-term follow-up should be performed to determine if these results are sustained. Acknowledgments We would like to express our gratitude to Emily Chapa, Andrew Sanchez, Fayann Ridgely, RN, Tammy Brown, LPN, Lucille Day, RN, and Leslie Gladu, RN, for their hard work in subject coordination and data collection, as well as Terence Davidson, MD, for his contributions to the design of this study.
Author Contributions Andrew J. Senchak, conception and design, data collection and analysis, drafting and critically revising manuscript, final approval; Alex J. Mckinlay, design, data collection and analysis, critically revising manuscript, final approval; Jason Acevedo, design, data collection and analysis, critically revising manuscript, final approval; Brenda Swain, data collection and analysis, critically revising manuscript, final approval; Maitram Christine Tiu, data collection and analysis, critically revising manuscript, final approval; Brian S. Chen, design, data collection and analysis, critically revising manuscript, final approval; Jon Robitschek, data collection and analysis, critically revising manuscript, final approval; Douglas S. Ruhl, data collection and analysis, critically revising manuscript, final approval; Lawrence L. Williams, Data analysis, critically revising manuscript, final approval; Macario Camacho, data analysis, critically revising manuscript, final approval; William C. Frey, design, interpretation of sleep studies, drafting and critically revising manuscript, final approval; Peter D. O’Connor, conception and design, data collection and analysis, interpretation of sleep studies, drafting and critically revising manuscript, final approval.
Disclosures Competing interests: None. Sponsorships: None. Funding source: This work was supported by funds from the AMEDD Advanced Medical Technology Initiative of the Telemedicine & Advanced Technology Research Center, Fort Detrick, Maryland. These funds were used for the salary for a research coordinator and to purchase sleep apnea monitors and disposable equipment used with the monitors.
References
2. Nieto FJ, Young TB, Lind BK, et al. Association of sleepdisordered breathing, sleep apnea, and hypertension in a large community-based study: Sleep Heart Health Study. JAMA. 2000;283:1829-1836. 3. Kribbs NB, Pack AI, Kline LR, et al. Objective measurement of patterns of nasal CPAP use by patients with obstructive sleep apnea. Am Rev Respir Dis. 1993;147:887-895. 4. Elshaug AG, Moss JR, Hiller JE, et al. Upper airway surgery should not be first line treatment for obstructive sleep apnoea in adults. BMJ. 2008;336:44-45. 5. Aurora RN, Casey KR, Kristo D, et al. Practice parameters for the surgical modifications of the upper airway for obstructive sleep apnea in adults. Sleep. 2010;33:1408-1413. 6. Aetna. Obstructive sleep apnea in adults. http://www.aetna.com/ cpb/medical/data/1_99/0004.html. Published March 2014. Accessed August 25, 2014. 7. Friedman M, Ibrahim H, Bass L, et al. Clinical staging for sleep-disordered breathing. Otolaryngol Head Neck Surg. 2002;127:13-21. 8. Johns MW.A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14:540-545. 9. Netzer NC, Stoohs RA, Netzer CM, et al. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med. 1999;131:485-491. 10. Iber C, Ancoli-Israel S, Chesson A, et al. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. Westchester, IL: American Academy of Sleep Medicine; 2007. 11. Farber JM.Clinical practice guideline: diagnosis and management of childhood obstructive sleep apnea syndrome. Pediatrics. 2002;109:704-712. 12. Mora R, Salami A, Passali FM, et al. OSAS in children. Int J Pediatr Otorhinolaryngol. 2003; 67(suppl 1): S229-S231. 13. Marcus CL, Moore RH, Rosen CL, et al. A randomized trial of adenotonsillectomy for childhood sleep apnea. N Engl J Med. 2013;368:2366-2376. 14. Mitchell RB.Adenotonsillectomy for obstructive sleep apnea in children: outcome evaluated by pre- and postoperative polysomnography. Laryngoscope. 2007;117:1844-1854. 15. Collop NA, Anderwson WM, Boehlecke B, et al; Portable Monitoring Task Force of the American Academy of Sleep Medicine. Clinical guidelines for the use of unattended portable monitors in the diagnosis of obstructive sleep apnea in adult patients. J Clin Sleep Med. 2007;3:737-747.
1. Young T, Palta M, Dempsey J, et al. The occurrence of sleepdisordered breathing among middle-aged adults. N Engl J Med. 1993;328:1230-1235.
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
The Effect of Tonsillectomy Alone in Adult Obstructive Sleep Apnea Andrew J. Senchak, DO, MS1, Alex J. McKinlay, MD2, Jason Acevedo, MD3, Brenda Swain, APRN, CNP3, Maitram Christine Tiu, PA-C, MPAS4, Brian S. Chen, MD4, Jon Robitschek, MD5, Douglas S. Ruhl, MD, MSPH6, Lawrence L. Williams, MD1, Macario Camacho, MD6, William C. Frey, MD7, and Peter D. O’Connor, MD, OD7
Otolaryngology– Head and Neck Surgery 1–5 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599815575721 http://otojournal.org
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
Keywords tonsillectomy, tonsils, adult, obstructive sleep apnea
Abstract Objective. The purpose of this study was to determine the effect of tonsillectomy as a single procedure in the treatment of adult obstructive sleep apnea (OSA).
Received October 16, 2014; revised February 5, 2015; accepted February 11, 2015.
Study Design. Prospective multi-institutional study evaluating adults with tonsillar hypertrophy scheduled to undergo tonsillectomy as an isolated surgery. Setting. Tertiary care medical centers within the US Department of Defense. Subjects and Methods. Adult subjects with tonsillar hypertrophy who were already scheduled for tonsillectomy were enrolled from October 2010 to July 2013. Subjects underwent physical examination, Epworth Sleepiness Scale, Berlin Questionnaire, and polysomnogram before surgery and after. Collected data included demographics, questionnaire scores, apnea-hypopnea index (AHI), and lowest saturation of oxygen. Results. A total of 202 consecutive subjects undergoing tonsillectomy were enrolled. The final analysis included 19 subjects testing positive for OSA. The mean age was 27.9 years; mean body mass index, 29.6; median tonsil size, 3; and most frequent Friedman stage, 1. The AHI before surgery ranged from 5.4 to 56.4 events per hour. The mean AHI decreased from 18.0 to 3.2 events per hour after surgery, a reduction of 82%. The responder rate—with subjects achieving at least a 50% reduction of AHI to a value \15—was 94.7%. Following tonsillectomy, there were statistically significant reductions in median lowest saturation of oxygen level and Epworth Sleepiness Scale and Berlin scores. Conclusions. Adult tonsillectomy alone has beneficial effect in OSA management, particularly in young overweight men with large tonsils, moderate OSA, and low Friedman stage.
O
bstructive sleep apnea (OSA) is a condition with significant morbidity and mortality, with an estimated prevalence of 24% of men and 9% of women.1 It has been linked to daytime hypersomnolence, fatigue, impaired work productivity, and development of chronic cardiovascular disease.2 While medical therapy with continuous positive airway pressure has been shown to reduce symptoms and decrease rates of mortality, many patients experience mask discomfort, aerophagia, mask air leak, and claustrophobia with use of a continuous positive airway pressure machine. In fact, objective data collected on patients show a \50% compliance rate, with only 4 hours of nightly continuous positive airway pressure use.3 Surgical procedures to correct OSA or those that serve an adjunctive role remain a crucial therapeutic option. Currently, there are limited data on tonsillectomy alone as treatment for adult OSA, and there is a need for studies with high-level evidence.4 The American Academy of Sleep Medicine practice guidelines for the surgical modification
1
Walter Reed National Military Medical Center, Bethesda, Maryland, USA Darnall Army Medical Center, Fort Hood, Texas, USA 3 Reynolds Army Community Hospital, Fort Sill, Oklahoma, USA 4 Madigan Army Medical Center, Tacoma, Washington, USA 5 Landstuhl Regional Medical Center, Landstuhl, Germany 6 Tripler Army Medical Center, Honolulu, Hawaii, USA 7 San Antonio Military Medical Center, San Antonio, Texas, USA 2
The views herein are the private views of the authors and do not reflect the official views of the Departments of the Army, Air Force, or Navy, nor the Department of Defense. Corresponding Author: Andrew J. Senchak, DO, MS, Department of Otolaryngology, Walter Reed National Military Medical Center, 8901 Wisconsin Ave, Bethesda, MD 20889, USA. Email: [email protected]
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
2
Otolaryngology–Head and Neck Surgery
of the upper airway to treat OSA do not mention tonsillectomy as an option, except when performed with a uvulopalatopharyngoplasty.5 These guidelines do not recommend uvulopalatopharyngoplasty, with or without tonsillectomy, as a sole procedure to treat OSA. As evidence of this debate, some third-party payors do not reimburse for tonsillectomy to treat OSA, as they consider it to be experimental for this indication.6 The lack of high-level evidence to support upper airway surgery may be due in part to patient selection, surgical technique selection, or specific outcomes measured in the published literature. This study was undertaken to determine if adult tonsillectomy alone improves OSA. As with any therapy to treat OSA, the outcome goals were to resolve clinical signs and symptoms, to normalize respiratory parameters (ie, apneahypopnea index [AHI] and oxygen saturation), to decrease sleepiness, and to restore sleep continuity.5 The objective of this study was to measure these outcomes in otherwise healthy adult subjects with enlarged tonsils who underwent tonsillectomy.
Methods A multi-institutional prospective trial was conducted at San Antonio Military Medical Center, Walter Reed National Military Medical Center, Darnall Army Medical Center, Reynolds Army Community Hospital, Bayne-Jones Army Community Hospital, Landstuhl Regional Medical Center, Madigan Army Medical Center, and Tripler Army Medical Center. The institutional review board at each medical facility approved the research protocol, and subjects were recruited by the otolaryngology service at each institution. Since the military employs a high percentage of young healthy adults, this setting served as an ideal group from which to recruit a homogeneous population of subjects with enlarged tonsils. Enrollment for the study commenced from October 2010 through July 2013. All subjects were recruited from adults aged 18 years with tonsils size 2 to 4 who were already scheduled to undergo tonsillectomy, independent of the research project, for reasons of chronic tonsillitis, recurrent tonsillitis, or halitosis. The Friedman classification system was used to measure tonsils size and palate position, and the Friedman stage was calculated.7 We excluded subjects with a history of pharyngeal surgery, significant nasal pathology (deviated nasal septum that obstructed the nasal cavity, nasal polyposis, or sinusitis), or dental malocclusion (as determined by the relationship of the maxillary first molar to the mandibular first molar). Subjects completed 2 questionnaires: the Epworth Sleepiness Scale (ESS)8 and category 1 of the Berlin Questionnaire.9 Each subject was then given an ambulatory sleep monitor (ApneaLink Plus; ResMed Corp, San Diego, California), with instructions on how to use the monitor once overnight. This initial baseline sleep monitor use was deemed time T0. Subjects then underwent complete removal of the bilateral palatine tonsils, performed with electrocautery in all cases, with no other concurrent procedures performed. Approximately 2 to 6
months after surgery, subjects again filled out both questionnaires and repeated use of the ambulatory sleep monitor, deemed time T1. Data collected for the study included patient characteristics, physical examination findings, ESS and Berlin Questionnaire scores, the AHI, and the lowest saturation of oxygen. The AHI and lowest saturation of oxygen were calculated by a board-certified, fellowship-trained sleep physician (either W.C.F. or P.D.O.) by manually reviewing the ApneaLink Plus signal data. Respiratory events were scored using definitions established by the American Academy of Sleep Medicine, and hypopneas were scored using the ‘‘recommended’’ definition, according to the 2007 guidelines.10 OSA was defined as an AHI 5.0. Severity of OSA was defined as follows: mild (AHI, 5.0-14.9), moderate (AHI, 15.0-29.9), and severe (AHI, 30.0). The ESS and the snoring section of the Berlin Questionnaire (category 1) were used to measure symptoms of sleepiness and snoring, respectively. ESS scores from 0 to 10 were defined as no daytime hypersomnolence, and scores .10 indicated hypersomnolence. Scores of 0 or 1 from category 1 of the Berlin Questionnaire were considered normal and from 2 to 6, positive for significant snoring. Resolution of OSA was based on a responder rate, defined as reduction of AHI by at least 50% to an absolute index of \15 events per hour at T1. Data were analyzed and described using number counts, proportions, medians with interquartile ranges, or means with standard deviations or 95% confidence intervals. A chi-square analysis was used to compare proportions. For continuous data, a paired t test was used to compare means for data that were normally distributed, and a Wilcoxon rank-sum test was used to compare medians for nonparametric data. Statistical significance was defined as a P value \.05. All statistical analyses were performed using Stata, version 11.0 (StataCorp LP, College Station, Texas).
Results A total of 202 subjects were enrolled in the study, and 46 were excluded for reasons listed in Figure 1. The demographic characteristics of subjects included versus excluded are shown in Table 1. There was no statistically significant difference between these groups. Of those subjects who were included, 17.9% (n = 28 of 156) tested positive for OSA on preoperative polysomnogram. After accounting for those lost to follow-up, a final total of 19 subjects with OSA completed postoperative polysomnogram evaluation and were available for outcome analysis. Demographic characteristics for the 19 subjects completing data analysis are shown in Table 2. The median tonsil size was 3, and the most frequent Friedman stage was 1. The mean time from baseline (T0) to follow-up (T1) was 77.2 6 5.2 days. Individual outcome data before and after surgery for all 19 subjects completing the study are listed in Table 3. The AHI at baseline time T0 ranged from 5.4 to 56.4 events per hour. All 19 subjects had a reduction of AHI by at least 50% following surgery, regardless of preoperative tonsil size, Friedman stage, OSA severity, or body
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
Senchak et al
3 Table 2. Characteristics of Subjects Completing Data Analysis (n = 19).a
Consecutive Subjects Enrolled (n = 202)
Characteristics Subjects Excluded (n = 46) -PSG with unusable data (n = 24) -Surgery canceled (n = 15) -Could not tolerate PSG (n = 7)
Subjects Included (n = 156) -All successfully completing pre-op PSG
PSG Negative for OSA (n = 128)
PSG Positive for OSA (n = 28)
Lost to Follow-up (n = 9)
Subjects Completing Post-Op PSG (n = 19)
Age,a y Body mass index,a kg/m2 Maleb Caucasianb Tonsil sizec Friedman palate positionc Friedman stageb I II III
Values 27.9 6 10.1 29.6 6 6.0 84.2 42.1 3 (3, 3) 1 (1, 2) 68.4 21.1 10.5
a
Figure 1. Flow of study subjects. OSA, obstructive sleep apnea; PSG, polysomnogram.
Table 1. Demographic Characteristics of Subjects: Included vs Excluded.
Age,a y Body mass index,a kg/m2 Menb Caucasianb
Included (n = 156)
Excluded (n = 46)
P Value
26.5 6 7.2 27.2 6 4.7 71.8 71.8
24.6 6 4.5 27.12 6 3.9 63.0 67.4
.094 .987 .253 .564
a
Mean 6 standard deviation. Percentage.
b
mass index. The responder rate was 94.7% (n = 18 of 19) with only 1 subject (No. 12) failing to achieve an AHI \15 events per hour at T1. This subject had a preoperative body mass index of 42.8 and tonsil size of 3 and was Friedman stage 3. The mean and median change in outcome measures from T0 to T1 is shown in Table 4. The mean body mass index did not significantly change after surgery. There was a statistically significant decrease in the mean AHI from 18.0 to 3.2 events per hour following surgery (Figure 2), a reduction by 82%. There was a statistically significant increase in the median lowest saturation of oxygen value after surgery, while there was a statistically significant decrease in the median ESS and Berlin scores.
Discussion Evidence has shown that removing tonsils and adenoids in children with OSA provides a significant role in its treatment.11-13 A consistent improvement in respiratory parameters measured by polysomnogram and in quality-of-life surveys is seen after adenotonsillectomy in children.14 Current surgical approaches for OSA in adults can include many options, such as nasal surgery, oropharyngeal expansion procedures, base of tongue reduction, and maxillomandibular advancement. The goal of this project was to prospectively
Mean 6 standard deviation. Percentage. c Median (interquartile range). b
evaluate subjects already scheduled for tonsillectomy to determine the degree to which OSA can be improved after surgery. Subjects already met indications for tonsillectomy independent of the research project and were then screened for OSA. As expected with such a population, a significant number of the subjects did not test positive for sleep apnea. It is well established that ambulatory sleep testing may underestimate obstructive breathing events and is typically not recommended as a screening procedure for general populations.15 Despite these limitations, however, 17.9% of subjects scheduled for tonsillectomy for routine indications were found to have some degree of OSA. The final sample size, while small, was powered adequately to demonstrate a statistically significant change in AHI following tonsillectomy and a considerable responder rate of 94.7%. This subject population on average had enlarged tonsils, was predominantly men and overweight, had moderate OSA, and a low Friedman stage. Of note, 2 subjects had size 2 tonsils. While the intent was to enroll subjects with tonsil size 3 or 4, these 2 subjects were reclassified as size 2 based on intraoperative examination findings and were kept in the study. Despite having smaller tonsils, these subjects still responded to surgery and had resolution of OSA. The only subject who failed to respond to surgery did have size 3 tonsils but was morbidly obese and had severe OSA. This suggests a possible correlation between these factors and surgical outcome. All subjects, however, had a reduction of AHI by at least 50% following surgery. Subjects had a median reduction in ESS scores after treatment, although the postoperative level did show mild persistent hypersomnolence at follow-up. Snoring, however, did improve with median Berlin Questionnaire score levels resolving to normal. A strength of the study was that the subject population was drawn from patients already scheduled for tonsillectomy and then screened for OSA. This was done to avoid bias in determining which should undergo tonsillectomy for
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
4
Otolaryngology–Head and Neck Surgery
Table 3. Outcome Measures (n = 19). Body Mass Index, kg/m2
Apnea-Hypopnea Index, Events/h
No.
Tonsil Size
Palate Position
Friedman Stage
T0
T1
T0
T1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4
1 1 1 1 1 1 2 1 3 1 2 2 3 1 2 2 2 1 1
2 2 1 3 1 1 1 1 2 1 1 3 2 1 1 1 1 1 1
25.0 27.0 23.0 41.8 26.7 30.8 25.0 24.6 32.7 29.1 30.4 42.7 34.8 21.7 29.6 24.5 29.6 25.7 36.8
24.7 27.2 23.2 41.8 27.4 30.9 23.7 22.9 32.5 31.9 29.9 41.8 34.9 21.6 30.2 24.6 28.2 26.5 35.2
17.0 6.3 7.0 11.1 22.5 10.1 5.4 29.4 37.0 16.9 15.4 56.4 33.3 7.6 26.5 7.2 9.5 7.1 16.4
1.7 0.6 3.2 4.4 0.9 3.1 1.4 2.5 2.4 1.1 5.8 24.8 2.0 0.7 2.7 0.0 3.7 0.4 0.2
Abbreviations: T0, before surgery; T1, after surgery.
Table 4. Change in Outcome Measures Before (T0) and After (T1) Surgery (n = 19). Time T0
T1
P Value
29.4 6 5.9 .558 Body mass index,a kg/m2 29.6 6 6.0 AHI,b events/h, baseline 18.0 (11.5, 24.5) 3.2 (0.6, 5.9) \.001 LSAT,c percentage O2 85 (82, 90) 91 (87, 94) .008 ESSc score 12 (10, 15) 9 (5, 12) .014 Berlinc score 4 (3, 4) 1 (0, 3) .013 Abbreviations: AHI, apnea-hypopnea index; ESS, Epworth Sleepiness Scale; LSAT, lowest saturation of oxygen. a Mean 6 standard deviation. b Mean (95% confidence interval). c Median (interquartile range).
treatment of OSA. One limitation of the study was an unexpectedly high rate of unusable ambulatory monitor data. This was due to improper use of the unattended portable monitor and subsequent poor signal tracing. However, in some subjects who were willing to repeat use of the monitor, additional demonstration on proper use was provided, and a second recording was successful at obtaining data. Several subjects were also lost to follow-up. They either failed to return for a repeat visit or did not return phone calls. The young age of the subjects, the voluntary nature of the research, or the improvement in symptoms after surgery could have contributed to lack of follow-up. In addition, the
Figure 2. Apnea-hypopnea index (AHI) values before and after surgery.
military is a highly mobile population, and overseas deployment or change of duty station may have also influenced the dropout rate.
Conclusion Adult tonsillectomy alone has beneficial effect in the treatment of OSA, particularly in young overweight men with moderate sleep apnea, large tonsils, and a low Friedman stage. In this patient population, the AHI can be normalized in a majority of patients over a short-term period, with an approximately 82% reduction in AHI seen in all-comers in
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
Senchak et al
5
this study. Oxygen desaturations, sleepiness, and snoring are likely to improve as well. Patients with morbid obesity or severe OSA may be less likely to respond to surgery despite having large tonsils. A long-term follow-up should be performed to determine if these results are sustained. Acknowledgments We would like to express our gratitude to Emily Chapa, Andrew Sanchez, Fayann Ridgely, RN, Tammy Brown, LPN, Lucille Day, RN, and Leslie Gladu, RN, for their hard work in subject coordination and data collection, as well as Terence Davidson, MD, for his contributions to the design of this study.
Author Contributions Andrew J. Senchak, conception and design, data collection and analysis, drafting and critically revising manuscript, final approval; Alex J. Mckinlay, design, data collection and analysis, critically revising manuscript, final approval; Jason Acevedo, design, data collection and analysis, critically revising manuscript, final approval; Brenda Swain, data collection and analysis, critically revising manuscript, final approval; Maitram Christine Tiu, data collection and analysis, critically revising manuscript, final approval; Brian S. Chen, design, data collection and analysis, critically revising manuscript, final approval; Jon Robitschek, data collection and analysis, critically revising manuscript, final approval; Douglas S. Ruhl, data collection and analysis, critically revising manuscript, final approval; Lawrence L. Williams, Data analysis, critically revising manuscript, final approval; Macario Camacho, data analysis, critically revising manuscript, final approval; William C. Frey, design, interpretation of sleep studies, drafting and critically revising manuscript, final approval; Peter D. O’Connor, conception and design, data collection and analysis, interpretation of sleep studies, drafting and critically revising manuscript, final approval.
Disclosures Competing interests: None. Sponsorships: None. Funding source: This work was supported by funds from the AMEDD Advanced Medical Technology Initiative of the Telemedicine & Advanced Technology Research Center, Fort Detrick, Maryland. These funds were used for the salary for a research coordinator and to purchase sleep apnea monitors and disposable equipment used with the monitors.
References
2. Nieto FJ, Young TB, Lind BK, et al. Association of sleepdisordered breathing, sleep apnea, and hypertension in a large community-based study: Sleep Heart Health Study. JAMA. 2000;283:1829-1836. 3. Kribbs NB, Pack AI, Kline LR, et al. Objective measurement of patterns of nasal CPAP use by patients with obstructive sleep apnea. Am Rev Respir Dis. 1993;147:887-895. 4. Elshaug AG, Moss JR, Hiller JE, et al. Upper airway surgery should not be first line treatment for obstructive sleep apnoea in adults. BMJ. 2008;336:44-45. 5. Aurora RN, Casey KR, Kristo D, et al. Practice parameters for the surgical modifications of the upper airway for obstructive sleep apnea in adults. Sleep. 2010;33:1408-1413. 6. Aetna. Obstructive sleep apnea in adults. http://www.aetna.com/ cpb/medical/data/1_99/0004.html. Published March 2014. Accessed August 25, 2014. 7. Friedman M, Ibrahim H, Bass L, et al. Clinical staging for sleep-disordered breathing. Otolaryngol Head Neck Surg. 2002;127:13-21. 8. Johns MW.A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14:540-545. 9. Netzer NC, Stoohs RA, Netzer CM, et al. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med. 1999;131:485-491. 10. Iber C, Ancoli-Israel S, Chesson A, et al. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. Westchester, IL: American Academy of Sleep Medicine; 2007. 11. Farber JM.Clinical practice guideline: diagnosis and management of childhood obstructive sleep apnea syndrome. Pediatrics. 2002;109:704-712. 12. Mora R, Salami A, Passali FM, et al. OSAS in children. Int J Pediatr Otorhinolaryngol. 2003; 67(suppl 1): S229-S231. 13. Marcus CL, Moore RH, Rosen CL, et al. A randomized trial of adenotonsillectomy for childhood sleep apnea. N Engl J Med. 2013;368:2366-2376. 14. Mitchell RB.Adenotonsillectomy for obstructive sleep apnea in children: outcome evaluated by pre- and postoperative polysomnography. Laryngoscope. 2007;117:1844-1854. 15. Collop NA, Anderwson WM, Boehlecke B, et al; Portable Monitoring Task Force of the American Academy of Sleep Medicine. Clinical guidelines for the use of unattended portable monitors in the diagnosis of obstructive sleep apnea in adult patients. J Clin Sleep Med. 2007;3:737-747.
1. Young T, Palta M, Dempsey J, et al. The occurrence of sleepdisordered breathing among middle-aged adults. N Engl J Med. 1993;328:1230-1235.
Downloaded from oto.sagepub.com at NORTH CAROLINA STATE UNIV on April 18, 2015
