Volume 16 Number 2 April - June 2015

• 3-YearPCCP PositionReview of Statement OSA in the on Lung E- Cigarettes Center of the Philippines • PrevalenceMDMA-induced and pneumothoraxConsequences of Shift Work Disorder among Nurses • A Predictive Formula for Optimal CPAP Level in Filipino OSA Patients • ANC score, STOP-BANG, and Berlin Questionnaire • OSA Symptoms and Excessive Daytime Sleepiness among Bus Drivers • Decoding the Polysomnogram

PHILIPPINE JOURNAL OF CHEST DISEASES AN OFFICIAL PUBLICATION OF THE PHILIPPINE COLLEGE OF CHEST PHYSICIANS

Editor-in-Chief Evelyn Victoria E. Reside, MD, FPCCP

Managing Editor Camilo C. Roa, Jr., MD, FPCCP

Issue Editor Irene Salve D. Joson-Vergara, MD, DPCCP

Reviewers Emily T. Aventura, MD, DPCCP Patrick Gerard L. Moral, MD, FPCCP Ma. Encarnita Blanco-Limpin, MD, FPCCP Albert L. Rafanan, MD, FPCCP Richmond. B. Ceniza, MD, FPCCP Ronald S. Reodica, MD, FCCP Manuel C. Jorge II, MD, FPCCP Michael Alexius A. Sarte, MD, FPSOHNS Norman L. Maghuyop, MD, FPCCP Rommel N. Tipones, MD, FPCCP Jennifer Ann M. Wi, MD, FPCCP

Editorial Assistant Ivan Noel G. Olegario, MD, MDC

PHILIPPINE COLLEGE OF CHEST PHYSICIANS OFFICERS 2015-2016 President Patrick Gerard L. Moral, MD, FPCCP

Vice President Vincent M. Balanag, Jr., MD, FPCCP

Secretary Charles Y. Yu, MD, FPCCP

Treasurer Lenora C. Fernandez, MD, FPCCP

Board Members Malbar G. Ferrer, MD, FPCCP Ivan N. Villespin, MD, FPCCP Gregorio P. Ocampo, MD, FPCCP Imelda M. Mateo, MD, FPCCP Eileen G. Aniceto, MD, FPCCP

Immediate Past President Chad Rey V. Carungin, MD, FPCCP

The opinions and data expressed in the Philippine Journal of Chest Diseases (PJCD) are those of the individual authors. They are not attributable to the editors or editorial board of the PJCD and should not be regarded as the ofﬁcial stand of/or endorsement by the Philippine College of Chest Physicians. References may be made in the articles regarding drug usage, which may not be included in the current prescribing information. The reader is, thus, urged to check the full prescribing information of drugs. No part of the PJCD may be reproduced without the written permission of the publisher.

Address all communication and manuscripts for publication to the following: The Editor, Philippine Journal of Chest Diseases, 84-A Malakas St., Pinyahan, Quezon City. Email: [email protected]. Phone: (+632) 924 9204. INSTRUCTIONS TO AUTHORS

The Philippine Journal of Chest Diseases 4. Disclosure of funding received for this work publishes scientific papers in the field of from any organization or company. pulmonary medicine. These papers may be in the 5. State if the paper has been presented in any form of collective and current reviews (state of convention and whether any awards have the art, meta-analyses), original investigations, been conferred on the paper. case reports, editorials or letters to the editor. All manuscripts must be submitted electronically to Abstract. The abstract should not be longer than [email protected]. Manuscripts should be 250 words. It should contain a summary of what single spaced and left-justified, including was done in the study, including objectives, study references. Use 10-point type, approximately 1- design, important results and conclusions. Only inch margins, and format for 8 ½ x 11 paper. The findings restricted to the study should be editorial staff requires files that can be opened and mentioned in the abstract. For research reports manipulated in Word 2004-2009, PowerPoint or only, abstracts must be in the structured form of Excel. four paragraphs, with headings Purpose, Methods, Results, and Conclusions; and must Accepted manuscripts become the property of the include the year of the study. The authors should Philippine College of Chest Physicians and are also provide three key words under which the published with the understanding that they are not article can be indexed. for publication elsewhere without approval. These manuscripts are subject to editorial Headings modification. For all manuscripts. Use main headings and short subheadings as needed. Do not create a Generally, write using the first person, active heading at the very top of the manuscript (e.g., voice; for example, “We analyzed data,” not “Introduction”), since layout constraints make “Data were analyzed.” The Abstract such headings unworkable. Text should be set in and acknowledgments or disclaimers are the Times New Roman font, 10 point in size, and exceptions to this guideline, and should be written single-spaced. The main heading of the online- in the third person, active voice; “The authors only text should be in 12 point and boldface; analyzed,” “The authors wish to thank.” subheadings should be in 10-point and boldface. If subheadings are used, two or more such Supply a title page as the first page of the headings must be used, as in outline style. manuscript with the following information: 1. The manuscript’s full title which should For research reports. Structure the body of the provide sufficient information regarding the manuscript using the headings Introduction, contents of the manuscript. Methods, Results, and Conclusions. At least a full 2. All authors should provide their complete paragraph of text must precede the Introduction names, professional titles, and institutional heading, for layout reasons. affiliations. Include an author byline that lists all authors’ full names and academic degrees For articles. Create headings that are substantive above a Masters; for example, “Juana Cruz, and interesting and that will give readers a sense MD, PhD, and Juan Ramos, MD”. Also of the article’s organization. Make headings as include sentence-style bios for each author short as is feasible. At least a full paragraph of than list position(s) or title(s) and institutional text must precede the initial heading, for layout affiliation(s); for example, “Dr. Cruz is reasons. assistant professor, Section of Pulmonary Medicine, Department of Internal Medicine, Text. Formal scientific or technical style shall be State University College of Medicine”. followed in writing the manuscripts. All 3. Contact information (address and email abbreviations should be spelled out when used address, plus telephone and/or fax) for the for the first time. For standard terminology, corresponding author. such as chronic obstructive pulmonary disease INSTRUCTIONS TO AUTHORS

(COPD) or forced vital capacity (FVC), only “live” figures that can be opened and formatted. standard abbreviations should be used. Information PDFs and JPGs are not accepted. Figures should or data that is best described in tables should be be two-dimensional; black-and-white or presented as such. Tables which duplicate grayscale; and without gridlines or background information provided in the text shall be removed. shading. X and Y axes, if present, must be Generic names of drugs shall be used except in labeled. instances where trade names are vital, such as in clinical trials. Figure legends should make the figure sufficiently understandable independent of the Tables and Figures. Only tables cited in the text manuscript. Legends should be placed on the last should be included. All tables should be called out page in the manuscript. All figures should be in the text and shall be numbered in ascending separated from the text file, yet bundled into a order depending on the sequence they were common file, if possible, with individual figures referred to in the text. A different order for tables separated by page breaks. and figures is to be used. Symbols are * † ‡ § ¶. The editorial staff reserves the right to determine A single table or figure with the appropriate labels whether the graphical instruments are appropriate should be printed on a single page. The text and for the information being imparted and modify or data in online tables should be Arial font, 10 point request modification/s for inappropriate in size, and single-spaced. The table title should be illustrations. The editorial staff reserves the right set in Arial font 12 point, and bold. Headings to generate illustrations compatible with the within tables should be set in 10 point bold. professional standards of the journal.

Explanatory notes or legends should be written References. Authors are responsible for the at bottom of the table or figure. Table titles should accuracy and completeness of their references and make the table sufficiently understandable for correct text citations. All references should be independent of the manuscript. Typically, include identified at the appropriate parts of the text using type of data, number and type of respondents, place Arabic numerals enclosed in parentheses. All of study, year of study. Titles should be placed references should then be typed double-spaced at directly above the table, not in a data cell. Columns the end of the manuscript and numbered should be clearly labeled, including unit of according to the order they were cited in the text. measure. Journal references should include the names of all the authors and inclusive page numbers. Footnotes: If information is needed to make the Abbreviations of names of journals should table understandable that won’t easily fit into the conform to those used in the Index Medicus. table title or data cells, create one or more footnotes. Table footnotes should be set in 8 point For world wide web citations, follow the and single-spaced. Place footnotes at the bottom of following format: <<URL>> <date of document> should also be explained. (accessed <date accessed>). You may break URLS across lines, but if possible, arrange for Figures. Only figures (or pictures) cited in the text breaks to occur only at punctuation separators should be included. All figures should be called out (but not on hyphens, and don’t ever add hyphens). in the text and shall be numbered in ascending order depending on the sequence they were Samples of the style to be followed in the listing referred to in the text. A different order for tables references are enumerated below: and figures is to be used. JOURNAL ARTICLE: Tanchuco JQ, Young J. Figures are acceptable as Excel, PowerPoint Normal standards for spirometric tests in Filipino or Word 2004-2009 files. All files supplied must be children. Chest Dis J 1989. 16:93-100. INSTRUCTIONS TO AUTHORS BOOK: Kelley MA, Fishman AP. Exercise Testing. invitations to forthcoming symposia or convention In: Pulmonary Diseases. 2 edition. Fishman AP, for publication at minimal cost depending on (ed.). McGraw-Hill Book Co.; 1989. pp. available space. 2525-2532. Reprints. Requests for additional reprints of WORLD WIDE WEB: Horton M, Adams R. individual articles should be addressed to: The Standard for interchange of USENET messages Editor-In-Chief, Philippine Journal of Chest Request for comment s 1036, Network Working Diseases, PCCP Secretariat, 84-A Malakas St., Group. <ftp://ftp.demon.co.uk/pub/doc/rfc/rfc1036. Brgy. Pinyahan, Diliman, <a href="/tags/Quezon_City/" rel="tag">Quezon City</a> (Telephone txt> Dec.1987 (Accessed 19 June 1995) No. 924-9204 and Fax No. 924-0144). Author/s of each manuscript are entitled to 25 copies of the Personal communications, unpublished data or article. These shall be sent to the major author. manuscripts in preparation should not be used as Requests for reprints should be addressed to the numbered reference. Instead, these may be cited in senior author. Reprints of entire issues may be parentheses or as a footnote on the page where they provided at cost, depending on availability of are mentioned. Authors assume responsibility for copies. verifying the accuracy of their cited reference. Subscriptions. All requests for subscriptions should Advertisements. All requests for rates should be be addressed to: The Business Manager, Philippine add-ressed to: The Business Manager, Philippine Journal of Chest Diseases, PCCP Secretariat, 84-A Journal of Chest Diseases, PCCP Secretariat, 84-A Malakas St., Brgy. Pinyahan, Diliman, Quezon City Malakas St., Brgy. Pinyahan, Diliman, Quezon (Telephone No. 9249204 and Fax No. 924-0144. E- City (Telephone No. 924-9204 and Fax No. mail address secretariat@philchest.org. One issue 924-0144). The journal also accepts (P120.00). Back issues (depending on availability announcements from institutions or professional P120.00). APRIL-JUNE 2015 VOLUME 16 NUMBER 2 1 EDITORIAL 3 3-Year Review of Patients Diagnosed with Obstructive Sleep Apnea in the Lung Center of the Philippines Marie Charisma L. de la Trinidad, MD, FPCP; Virginia de los Reyes, MD, FPCP, FPCCP 10 Prevalence and Consequences of Shift Work Disorder among Nurses at Lung Center of the Philippines Maria Cecilia I. Jocson, MD, FPCP; Virginia de los Reyes, MD, FPCP, FPCCP 19 Development and Validation of a Predictive Formula for Optimal CPAP Level in Filipino Patients with Obstructive Sleep Apnea seen in Lung Center of the Philippines Emelie Grace J. Bautista, MD, FPCP, DPCCP; Virginia de los Reyes, MD, FPCP, FPCCP 27 Adjusted Neck Circumference Score, STOP-BANG and Berlin Questionnaire As Screening Tools For Obstructive Sleep Apnea Caroline M. Armas, MD, FPCP; Patrick Gerard L. Moral, MD, FPCP, FPCCP 36 Prevalence of Obstructive Sleep Apnea Symptoms and Excessive Daytime Sleepiness among Commercial Long Distance Bus Drivers in the Philippines Albert B. <a href="/tags/Albay/" rel="tag">Albay</a> Jr., MD, FPCP, FPCCP; Manuel C. Jorge II, MD, FPCP, FPCCP; Ricardo M. Salonga, MD, FPCP, FPCCP 44 PRACTICE GUIDE Decoding the Polysomnogram (PSG) Irene Salve D. Joson-Vergara, MD, DPCCP; Manuel Peter Paul C. Jorge, MD, FPCCP 51 Las rinde el sueño (Sleep overcomes them) EDITORIAL Sleep Medicine: The Dream Issue Evelyn Victoria E. Reside, MD, FPCCP Editor-in-Chief“There is a time for many words, and there is Sleep Medicine has grown by leaps and also a time for sleep.” bounds, and it still continues to grow. Since the first known sleep laboratory opened in the Many generations ago, at around the 8th U.S. in the 1970’s, the field has now century BC, Homer was said to have written encompassed the realms of Pulmonary these words as part of the epic Greek poem, Medicine, Otorhinolaryngology, Psychiatry, “The Odyssey” which details the adventures Neurology and even Dental Medicine. of the main character Odysseus (also known Worldwide, there are societies of experts in as Ulysses), as he journeyed home to Ithaka sleep medicine (so-called “sleep doctors”), as after the Trojan War. well as formal training and accreditation in the field. It is a prime example of And here is this second issue for this year, a multidisciplinary care at its best, of how combination of words and sleep, finally medical and non-medical interventions can home to Ithaka after a laborious journey for make a difference, and of how we still must all writers, reviewers and editors, a special listen to our bodies and their needs for it to issue on Sleep Medicine which, ironically, fully function. left many of those involved with sleepless nights. This collection of scientific papers on Now considered a considerable cause of and the various aspects of sleep disorders are contributor to morbidity and mortality, sleep products of our PCCP sleep specialists and disorders have significantly evolved into a the Sleep Council (not to mention our field on its own, though it can be said to be tireless Issue Editor and Editorial Assistant) still lacking in awareness, both among the who may have long dreamt of a vehicle general public and the medical community. A through which they may communicate their flurry of research in the 1990’s have shown passion for the field. that sleep disorders figure in the development 1 Phil J Chest Dis 2015 of a variety of medical illnesses such as hypertension and heart disease, so much so that “getting enough sleep” has become, now more than ever, a requirement; though it is silent in the typical doctor’s prescription, it must be considered a fundamental human need that must be filled. Indeed, gone are the days when sleep was looked upon as simply a natural part of the physiologic cycle, and when snoring was typical for the exhausted person at the end of a long and tiring day. As the youngest Council in the Philippine College of Chest Physicians, the Philippine Journal of Chest Diseases is proud to feature the works of the newest members of the PCCP family, and fittingly so, as the Philippines is home to the bangungut, which is the quintessential sleep disorder among all others. “Daytime sleep is like the sin of the flesh; the more you have the more you want, and yet you feel unhappy; sated and unsated at the same time.” -Umberto Eco, The Name of the Rose Vol. 16 | Issue 02 | June 2015 2 Obstructive Sleep Apnea in Lung Center of the Philippines RETROSPECTIVE STUDY 3-Year Review of Patients Diagnosed with Obstructive Sleep Apnea in the Lung Center of the PhilippinesMarie Charisma L. de la Trinidad, MD, FPCP; Virginia de los Reyes, MD, FPCP, FPCCP ABSTRACTBackground: Despite the increasing relevance of obstructive sleep apnea (OSA) to important morbidities, it remains under-recognized worldwide. There is a need to better characterize the disease in the Asian population. At present, there is lack of published data regarding the prevalence and characteristics of OSA among Filipinos. Objectives: The main objective of this study was to describe the clinical profile of Filipino patients diagnosed with OSA. Its secondary objective was to determine factors associated with severe OSA in Filipinos.Methods: We retrospectively analyzed records of Filipino patients, at least 16 years old, diagnosed to have OSA, who underwent polysomnography, seen at the Lung Center of the Philippines from January 2011 to December 2013. We summarized their characteristics in general and the association of these characteristics with OSA severity. Result: A total of 100 patients were included in the study. Majority had severe OSA (n= 74). Their mean apnea-hypopnea index (AHI) was 77.0 (range of 5.5-191.5) events/hour. The mean lowest oxygen saturation was 78.0% (range of 41% to 93%) during sleep. Majority were male (73%) with age ranging from 22 to 80 years old (mean=48 years); obese (91%), with increased neck circumference (54%), had increased daytime naps (64%), and had hypertension (60%) and a family history of snoring (79%). Almost half had a history of smoking (46%). Most patients complained of snoring (97%), abnormal breathing pattern during sleep (80%), and waking up with dry mouth or sore throat (79%). The mean Epworth sleepiness scale score was 10.7 with a range of 2 to 21. This was poorly correlated with the severity of OSA. Characteristics associated with severe OSA were obesity (OR 12.250), abnormal breathing pattern (OR 4.444), smoking history (OR 2.443) and tonsillar grade of at least 3 (OR 9.259).Conclusion: We conclude that for Filipinos with OSA, the clinical profile was similar to those of other Asian populations. Majority were middle aged, obese, male, hypertensive, with an increased tonsillar grade and a family history of snoring. Mostly, these patients complained of snoring; abnormal breathing pattern during sleep; waking up with dry mouth or sore throat; and a need for daytime naps. Patients suspected to have OSA presenting with an abnormal breathing pattern during sleep, obesity, smoking history, and enlarged tonsils have an increased likelihood of having severe OSA. Keywords: Obstructive sleep apnea, clinical profile, Filipino3 Phil J Chest Dis 2015 Dela Trinidad MCL et al INTRODUCTION patients at risk for OSA among Asians highlighted Obstructive sleep apnea (OSA) is a common the lack of data, only a few studies providing an medical disorder associated with important estimate of the OSA burden in Asian communi- morbidity. Its relevance in the medical field has ties.4 Only India, Korea, Thailand, Japan, been markedly recognized in the past few decades, Malaysia, Turkey, China, Singapore, Taiwan and when descriptions made by Gastaut and associates Iran were included in the review. in 1965 showed that cessation of respiration may be The importance of published data in the due to obstruction of the upper airway.1 OSA is a Asian population is stressed by a study made in breathing disorder during sleep that involves a 2005 showing results indicating differences of soft decrease or complete discontinuation in airflow tissue and skeletal abnormalities among patients despite efforts to breathe. with different ethnicity. It was noted that Asian During sleep, muscle relaxation causes patients with OSA had higher Mallampati Score, collapse of soft tissues in the back of the throat, shorter thyromental distance and larger blocking the upper airway. This may lead to thyromental angle, and tended to have more severe reductions, also known as hypopneas, or complete OSA than white patients.5 This implies that data pauses, also known as apneas, in breathing. These involving Caucasians may not be applicable to the episodes last at least 10 seconds, and may lead to Asian population. reductions in oxygen saturation in the blood. The The main objective of this study was to lack of oxygen in the brain may trigger the body, describe the clinical profile of Filipino patients causing a brief arousal from sleep, which restores diagnosed with OSA. Its secondary objective was normal breathing. This may occur for several to determine the factors associated with severe episodes during the night and may cause sleep OSA in Filipinos. disturbance. Patients with OSA commonly present with METHODOLOGY snoring and abnormal breathing pattern during We retrospectively analyzed records of sleep. This includes pauses in breathing, choking, Filipino patients, at least 16 years old, diagnosed and snorting during sleep. Patients also may have to have OSA who underwent polysomnography daytime sleepiness and poor sleep quality.2 (PSG) at the Lung Center of the Philippines from OSA has also been increasingly recognized January 2011 until December 2013. The diagnosis as an important health issue as evidence of OSA was done based on the following increasingly shows that OSA may be an diagnostic criteria (with the presence of A and B; independent risk factor for hypertension, glucose or C; to satisfy the criteria): intolerance/diabetes mellitus, cardiovascular A. The presence of one or more of the following: diseases, and stroke. (1) The patient complains of sleepiness, non- The prevalence rates of OSA have been restorative sleep, fatigue or insomnia estimated in the range of 2% to 10% worldwide, symptoms; (2) the patient wakes with breath and the risk factors of OSA include advanced age, holding, gasping or choking; (3) the bed male sex, obesity, family history, craniofacial partner or other observer reports habitual abnormalities, smoking, and alcohol consumption.3 snoring, breathing interruptions or both during Despite its increasing relevance to important the patient’s sleep; or (4) the patient has been morbidity, OSA remains under-diagnosed diagnosed with hypertension, a mood disorder, worldwide and in Asia, where most people live in cognitive dysfunction, coronary artery disease, underdeveloped or developing nations. A stroke, congestive heart failure, atrial systematic review on the prevalence of OSA and fibrillation or type 2 diabetes mellitus. Vol. 16 | Issue 02 | June 2015 4 Obstructive Sleep Apnea in Lung Center of the Philippines B. PSG demonstrates five or more predominantly ing forms in patient’s records; 131 patients were obstructive respiratory events (obstructive and younger than 16 years old; and 9 patients had a mixed apneas, hypopneas or respiratory effort- diagnosis other than OSA. These patients were all related arousals [RERA]) per hour of sleep excluded. during PSG. A total of 100 patients were included in the C. PSG demonstrates 15 or more predominantly analysis. Most patients were male (73%) and obstructive respiratory events (apneas, middle aged with a mean age of 47 years (range of hypopneas or RERAs) per hour of sleep during 22 to 80 years) (Table 1). Majority were PSG. diagnosed to have severe OSA (74%), based on A review of the patients’ sleep questionnaire their apnea-hypopnea index (AHI) (Table 1), with and sleep study reports was also conducted. The an AHI of 5 to 15 for mild, 15 to 30 for moderate, clinical and demographic characteristics of these and greater than 30 for severe OSA. patients, including their signs, symptoms, sleep The mean AHI for this study was 77.0 schedule, sleep hygiene, physical characteristics, (range of 5.5 to 191.5) events/hour. The mean lifestyle, underlying medical condition, cardiac lowest oxygen saturation was 78.0% (range of arrhythmias, pressure applied during sleep study, 41% to 93%) during sleep. and family history were recorded, analyzed, and Table 2 shows the prevalence of signs and correlated to the likelihood of severe OHA by symptoms of the included patients, and the binary logistic regression. Data analysis used SPSS correlation of these signs and symptoms with using a level of significance of 5%. severe OSA. Most patients presented with snoring At 95% confidence level, a precision of 0.05, (97%), abnormal breathing pattern during sleep and an estimated true proportion of 4%, the (80%), and waking up with dry mouth or sore computed sample size was 60. throat (79%). Abnormal breathing pattern during sleeping was associated with a 4.44-fold greater RESULTS likelihood of severe OSA. From January 2011 to December 2013, 389 The Epworth sleepiness scale (ESS) was patients underwent sleep study at the Lung Center used to screen for daytime sleepiness, with a score of the Philippines Sleep Laboratory. Of these, 249 of 0 to 7 indicating abnormal sleepiness; 8 to 9 as patients were excluded due to incomplete or miss- average amount of daytime sleepiness; 10 to 15 as excessively sleepy and may seek medical attention; and 16 to 24 as having excessive Table 1. Baseline demographic data and OSA severity sleepiness and medical attention is warranted. The Characteristics (n=100) N mean ESS score of patients was 10.7 + 5.0 (range of 2 to 21).6 The Figure shows that ESS scores Gender were highly variable and correlated poorly with Male 73 Female 27 severe OSA vs non-severe OSA (coefficient of 0.001; OR 0.99 [95% CI 0.915, 1.092]; p=0.99). Mean age (range) (years) 48 (22-80) Majority of the patients had daytime naps (64%). OSA severity Only some claimed to take medication to initiate Severe OSA 74 sleep (12%), and had difficulty in reinitiating Moderate OSA 12 sleep (24%). Daytime naps (coefficient of -0.524; Mild OSA 12 Unclassified OSA 2 OR 0.592 [95% CI 0.221, 1.588]; p=0.298); medication intake to initiate sleep (coefficient OSA, obstructive sleep apnea. 5 Phil J Chest Dis 2015 Dela Trinidad MCL et al of -0.390; OR 0.677 [95% CI 0.186, 2.469]; ditions of the included patients. Majority had p=0.554); and difficulty in reinitiating sleep hypertension (60%). Some patients had gastro- (coefficient of -0.460; OR 0.632 [95% CI 0.232, esophageal reflux disease (28%) and Type 2 1.718]; p=0.368) were not correlated with severe diabetes mellitus (24%). Table 6 shows the OSA. correlation between comorbidities and severe Majority of patients were obese (31%) with a OSA. BMI of at least 30, and extremely obese (60%) with Most of patients had a family history of a BMI of at least 40 (Table 3). Most patients either snoring in the family (79%), while 15% had a had a Modified Mallampati airway score of 2 (30%) family history of sleep apnea. Both of these or 3 (27%). Also, most had a tonsillar grade of 2 factors did not significantly correlate with the (38%) or 3(21%). An abnormally large neck occurrence of severe OSA (p=0.774 and 0.196, circumference (greater than 43 cm and 38 cm in respectively). men and women, respectively) were noted in 55% Among the 100 patients included, 73% (39/71) of men and in 54% (13/24) of women. underwent determination of positive airway Tonsillar grade and obesity were significantly pressures. Sixty-four patients used continuous correlated with severe OSA. positive airway pressure (CPAP) and 9 patients Majority of patients had intake of caffeinated used bi-level airway pressure (BiPAP). The drinks (84%), particularly coffee (75%) (Table 4). range of pressure for CPAP was from 6 to 16 cm Many patients had a history of smoking (46%), with H20, with a mean of 10 cm H20. For BiPAP, the 41 patients having tried to quit. Smokers were found mean inspiratory positive airway pressure was to be 2.443 times more likely to have severe OSA 17.7 cm H20 (range of 8 to 27 cm H2O), and the compared to those without smoking history. mean expiratory positive airway pressure was Table 5 shows any underlying medical condi- 13.9 cm H20 (range of 4 to 20 cm H2O). Table 2. Prevalence and correlation of signs and symptoms to severe OSA Signs and Symptoms % Coefficient P-value Odds Ratio 95% CI Difficulty falling asleep 20 -0.325 0.559 0.722 0.242, 2.152' Waking up during the night 45 -0.038 0.934 0.963 0.392, 2.363 Un-refreshing Sleep 54 -0.388 0.406 0.679 0.272, 1.692 Excessively sleepy/ falling asleep during the day 56 -0.278 0.552 0.758 0.30, 1.891 Hallucination on waking/falling asleep 8 -1.715 0.026 0.18 0.04, 0.817 Snoring 97 1.058 0.46 2.88 0.174, 47.784 Abnormal breathing pattern during sleep 80 1.492 0.006 4.444 1.549, 12.753 Waking up with morning headache 52 -0.338 0.464 0.714 0.289, 1.761 Waking up with dry mouth or sore throat 79 0.148 0.787 1.16 0.396, 3.397 Confusion in the mornings 23 -0.893 0.082 0.409 0.15, 1.119 Decreased concentration or alertness 50 -0.662 0.156 0.516 0.207, 1.287 Irritability, short temper 56 -0.333 0.476 0.717 0.287, 1.788 Forgetfulness 53 -1.191 0.017 0.304 0.114, 0.811 Anxiety 36 -1.061 0.024 0.346 0.138, 0.871 History of falling asleep while driving 30 -0.575 0.235 0.563 0.218, 1.452 History of road accidents or near miss accidents 24 0.318 0.575 1.375 0.453, 4.175 Vol. 16 | Issue 02 | June 2015 6 Obstructive Sleep Apnea in Lung Center of the Philippines Figure. Distribution of patients with severe and non- Furthermore, during the sleep study, 10% were severe OSA by Epworth Sleepiness Scale noted to have cardiac arrhythmias. CPAP level and cardiac arrhythmias were not correlated with 25 severe OSA (p=0.461 and 0.211, respectively). 20 Discussion Since its first descriptions in the 1960s, 15 OSA has been increasingly recognized in the Western countries. Studies done mainly in ESS Score ESS 10 Western countries have demonstrated OSA as an independent risk factor for hypertension, glucose 5 intolerance/diabetes mellitus, cardiovascular diseases and stroke. Western studies also showed 0 that the factors associated with OSA include age 0 10 20 30 of 30 to 60 years, male gender, obesity, family Patient Number history, craniofacial abnormalities, smoking, Non-Severe OSA Severe OSA endocrine diseases (hypothyroidism, acromegaly), and alcohol consumption. Symptoms usually Table 3. Prevalence and correlation of physical findings severe OSA Physical % Coefficient P-value Odds Ratio 95% CI Characteristics Neck circumference in 55 (39/71) 3.518 0.0004 0.2479 0.114, 0.5392 cm (male) Neck circumference in 54 (13/24) 0.023 0.919 1.03 0.3328, 3.0837 cm (female) Tonsillar Grade 2.226 0.035 9.259 1.176, 72.908 1 14 2 38 3 21 4 4 Modified Mallampati 0.521 0.058 1.683 0.982, 2.884 airway score 1 13 2 30 3 27 4 16 BMI Non-obese -1.995 0.006 0.136 0.032, 0.572 Underweight 2 Normal 2 Overweight 0 Obese 0.251 0 12.25 - Obese 31 Extreme>40 60 OSA, obstructive sleep apnea; BMI, body mass index. 7 Phil J Chest Dis 2015 Dela Trinidad MCL et al Table 4. Prevalence and correlation of smoking and alcohol/caffeine intake with severe OSA Lifestyle factor % Coefficient P-value Odds Ratio 95% CI Smoking history 46 0.893 0.046 2.443 0.944, 6.323 Smoking cessation 41 0.861 0.85 2.366 0.887, 6.311 Alcoholic Beverages 36 0.103 0.829 1.109 0.434, 2.831 Caffeinated beverage intake 84 Coffee 75 -0.391 0.489 0.677 0.224, 2.046 Tea 34 0.219 0.655 0.125 0.476, 3.253 Cola 43 0.219 0.655 1.245 0.476, 3.253 OSA, obstructive sleep apnea. include awakening with snoring, gasping, or greater neck circumference, increased choking; dyspnea upon awakening; and awaken- tonsillar grade, and presence of arterial ing with a dry mouth. The cardinal daytime hypertension. Most of our patients complaint of patients with OSA is excessive predominantly presented with snoring, abnormal daytime somnolence or fatigue. breathing pattern during sleep, and waking up However, this clinical profile for OSA was with dry mouth. Among these symptoms, derived largely from Caucasian populations. A abnormal breathing pattern during sleeping, meta-analysis of Asian studies showed that which included pauses in breathing, snorting and despite a lower prevalence of obesity among choking, were associated with severe OSA. Asians than in Caucasians, the prevalence of OSA was similar between the two ethnicities.4 The Table 5. Prevalence of underlying medical meta-analysis also showed that the clinical profile conditions in patients with OSA of an Asian patient with OSA was an average age of 30 to 50 years, male gender, obese, and Underlying Medical Conditions Count smoker. The presence of a greater neck Systemic hypertension 60 circumference and arterial hypertension were also noted in the Asian population. Symptoms of Asian GERD 28 patients with OSA were usually snoring and Type II diabetes mellitus 24 daytime sleepiness, as shown by a higher ESS Surgery to nose, throat or jaw 16 score. Furthermore, after controlling for age and Asthma 15 obesity, the severity of OSA tends to be higher in Asians than in whites. The increased risk for OSA Allergic rhinitis 14 among Asians could be related in part to Myocardial infarction 8 characteristics of the craniofacial anatomy. COPD 4 The meta-analysis however did not include Nocturnal cardiac arrhythmias 2 Filipino patients with OSA. In this study, we have Genetic Disorders 2 retrospectively studied Filipino patients diagnosed with OSA at the Lung Center of the Philippines. Depression 2 Our study revealed a clinical profile similar to Pulmonary hypertension 1 other Asians indicated in the meta-analysis. Psychiatric Illness 1 Majority had severe OSA, with severe Cerebrovascular accident 0 desaturation during sleep. Similarly, our patients’ clinical profile was: mean age of 47 years, male OSA, obstructive sleep apnea; COPD, chronic obstructive pulmonary disease; GERD, gastroesophageal gender, obese, with a positive smoking history, a reflux disease. Vol. 16 | Issue 02 | June 2015 8 Obstructive Sleep Apnea in Lung Center of the Philippines Table 6. Correlation between co-morbidities and severe OSA Underlying Medical Coefficient P-value Odds Ratio 95% CI Conditions Asthma -0.527 0.389 0.591 0.178, 1.959 Allergic rhinitis 0.308 0.658 1.36 0.348, 5.317 GERD -0.409 0.405 0.665 0.254, 1.74 Systemic hypertension 0.571 0.215 1.77 0.718, 4.364 Myocardial infarction -0.558 0.468 0.572 0.127, 2.584 Type II diabetes mellitus 1.062 0.112 2.893 0.78, 10.7062 Surgery to nose, throat or jaw -0.024 0.969 0.976 0.281, 3.384 OSA, obstructive sleep apnea; GERD, gastroesophageal reflux disease. Patients also generally had a family history of 2. American Academy of Sleep Medicine. snoring. In contrast to the Asian studies, the ESS Obstructive Sleep Apnea. Available at: was shown to be variable when contrasted against www.aasmnet.org. Accessed on: May 18, 2015. severe OSA occurrence. 3. Lam JCM, Sharma SK, Lam B. Obstructive sleep apnea: Definitions, epidemiology & Conclusion natural history. Indian J Med Res Most Filipinos with OSA were middle aged, 2010;131:165-170. obese, male, hypertensive, with an increased 4. Mirrakhimov AE, Sooronbaev T, Mirrakhimov tonsillar grade and a family history of snoring. EM. Prevalence of obstructive sleep apnea in Mostly, these patients complain of snoring, Asian adults: a systematic review of the abnormal breathing pattern during sleep and waking literature. BMC Pulm Med 2013;13:10. up with dry mouth. Characteristics of patients 5. Lam B, Ip MSM, Tench E, Ryan CF. associated with severe OSA were the following: the Craniofacial profile in Asian and white patients presence of abnormal breathing pattern with obstructive sleep apnea. Thorax characterized as choking, gasping and pauses in 2005;60:504–510. breathing during sleep; obesity; tonsillar grade of at 6. Nguyen AT, Baltzan MA, Small D, Wolkove N, least 3; and smoking history. This clinical profile is Guillon S, Palayew M. Clinical reproducibility similar to those of other Asian countries. of the Epworth Sleepiness Scale. J Clin Sleep Med 2006;2:170-174. References 1. Fishman’s Pulmonary Diseases and Disorders, 5th ed. New York City: McGraw-Hill Education; 2008. 9 Phil J Chest Dis 2015 Jocson MCI et al CROSS-SECTIONAL SURVEY Prevalence and Consequences of Shift Work Disorder among Nurses at Lung Center of the PhilippinesMaria Cecilia I. Jocson, MD, FPCP; Virginia de los Reyes, MD, FPCP, FPCCP ABSTRACTObjectives: This study aimed to determine the prevalence of SWD among nurses at LCP and the demographic and work profile of nurses with SWD; and to investigate the consequences of SWD in terms of work-related errors, occupational accidents, vehicular accidents, physical fatigue, mental fatigue, anxiety and depression.Methodology: This was a descriptive, quantitative, cross-sectional survey that included nurses working at who were engaged in shift work. The survey was conducted from October 2013 to January 2014. Demographic and work-related data were collected. All participants completed a comprehensive battery of sleep-related questionnaires. Results: A total of 133 nurses were included in the study. Majority was female (63.9%) and married (58.6%). The mean age was 33 years. Twenty-four nurses were identified to have SWD, for a prevalence of 18.0% (95% CI 12.4%, 25.5%). Those with SWD had significantly lower years of service than those without SWD (4.2 vs 8 years, respectively; p=0.027). Those with SWD had significantly higher rates of unplanned overtimes compared to those without SWD (100% vs 71.6%; p=0.003). They also had significantly higher consumption of coffee (mean of 1.6 vs 1 cup/day; p=0.029). Nurses with SWD had a significantly higher use of sleep medicines (16.7% vs 4.6%; p=0.033) and alcohol (16.7% vs 3.7%; p=0.016) for sleep compared with those without SWD. The odds of work-related error were four-fold higher among those with SWD than those without SWD (p=0.007). Physical and mental fatigue scores are significantly higher in nurses with SWD than those without SWD. Conclusion: The prevalence of SWD among nurses is 18.0%. Nurses with SWD were younger with fewer years in service. They were more likely to have developed maladaptive behaviors of using sleep medicines and alcohol to improve sleep. SWD was significantly associated with self-reported work-related errors. Nurses with SWD had significantly higher scores for physical fatigue and mental fatigue. INTRODUCTION management, entertainment, hospitality, cleaning, Shift work can be defined as the production, manufacturing, construction and organization of working time by different teams in tranportation services. succession to cover more than the usual 8-hour Studies indicate that significant health risk work day, up to a 24-hour period. Shift work is may be associated with shift work. Persons who found in many different types of industries, such do shift work are at increased risk of as food preparation, healthcare, security, police cardiovascular and gastrointestinal diseases.1,2 and firefighting, information technology, Data from the Nurses’ Health Study support an Vol. 16 | Issue 02 | June 2015 10 Shift work disorder among nurses association between working rotating night shifts important to understand the functioning and and an increased risk of ischemic stroke.3 But efficiency of health care and its impact on health perhaps the most imminent and persistent problems of employees. In the Philippines, nurses routinely that shift workers contend with are frequent sleep work long shifts and sometimes need to do disturbance and excessive sleepiness. Drowsiness mandatory or unplanned overtime. They work and fatigue in the work environment can lead to irregular day and night shifts with most of them poor concentration, absenteeism, accidents, errors, working more than 40 hours a week.6 As injuries, and even fatalities.4 members of healthcare teams, nurses tend to be Shift work disorder (SWD) is a sleep disorder exposed to extreme loads under situations characterized by sleepiness and insomnia, which handling complicated dynamic phenomena under can be attributed to a person's work schedule. The severe time pressure. Shift work in these diagnostic criteria for SWD, as defined by the professionals a potential problem for health, American Academy of Sleep Medicine International mental and social well-being, work performance Classification of Sleep Disorders-2 (ICSD-2), and patient safety. include: (1) complaints of insomnia or excessive At Lung Center of the Philippines (LCP), sleepiness temporally associated with a recurring the shift work schedule of staff nurses are work schedule in which work hours overlap with categorized as follows: morning shift (7 am to 3 the usual time for sleep; (2) symptoms must be pm); afternoon shift (3 pm to 11 pm); and night associated with the shift work schedule over the shift (11 pm to 7 am). Supervisors work 12-hour course of at least one month; (3) sleep log or shifts. Nurses assume rotating shifts, that is, shift actigraphic monitoring for at least 7 days schedule changes regularly. This study aimed to demonstrates circadian and sleep-time determine the prevalence of SWD among nurses misalignment; and (4) sleep disturbance is not better at LCP and the demographic and work profile of explained by another sleep disorder, mental nurses with SWD; and to investigate the disorder, a medical or neurological disorder, consequences of SWD in terms of work-related medication use or substance use disorder.5 errors, occupational accidents, vehicular Besides excessive sleepiness and insomnia, accidents, physical fatigue, mental fatigue, symptoms of SWD may include poor concentration, anxiety and depression. impaired performance, headache, irritability or depressed mood and lack of energy. Symptoms can METHODOLOGY be nonspecific and may mimic other sleep/wake This was a descriptive, quantitative, cross- disorders or other medical or psychiatric conditions. sectional survey that included nurses working at Importantly, SWD is underrecognized, who were engaged in shift work. The survey was underdiagnosed, undertreated and under-represented conducted from October 2013 to January 2014. in published clinical studies. Clinicians and patients Only nurses who were employed at LCP for at often do not recognize the risks, consequences, and least 3 months at the time of the study were comorbidities associated with and the significant included. Nurses who went on duty only in the impact that it can have on quality of life. If patients daytime were excluded. Demographic and work- are diagnosed and symptoms are effectively related data were collected. Information on work- managed, patients may experience improvement in related errors, occupational accidents, and health, functioning, job performance, safety, vehicular accidents were collected through self- personal relationship and overall quality of life. reporting. Shift work in nurses is one of the critical All participants completed a issues in hospital inpatient care, making it vitally comprehensive battery of questions and scales 11 Phil J Chest Dis 2015 Jocson MCI et al that included a questionnaire to confirm chance of dozing off). A clinical cut-off of at least symptoms of SWD; the Bergen Shift Work Sleep 11 distinguishes patients with various sleep Questionnaire (BSWSQ); the Epworth Sleepiness disorders from healthy subjects. Scale (ESS); a Global Sleep Assessment The GSAQ is a reliable and validated Questionnaire (GSAQ); a Fatigue Questionnaire; general sleep assessment tool that distinguishes and the Hospital Anxiety and Depression Scale between symptoms of different sleep disorders, (HADS). such as obstructive sleep apnea (2 questions), The questionnaire used to confirm restless legs syndrome, periodic limb movement, symptoms indicative of SWD was composed of and parasomnias. The response alternatives are three questions previously developed and used ‘never’, ‘sometimes’, ‘usually’, and ‘always’. specifically to assess or diagnose SWD in Subjects answering “always” to one or more of the epidemiologic studies. These questions were: (1) five questions were excluded, in accordance with Do you experience difficulties with sleeping or the assessment and diagnosis procedures for excessive sleepiness (yes/no)?; (2) Is the sleep or SWD. sleepiness problem related to a work schedule The 11-item Fatigue Questionnaire is a where you have to work when you would commonly used tool for measuring fatigue. Items normally sleep (yes/no)?; and (3) Has this sleep are scored on four-point Likert scale. The scale is or sleepiness problem related to your work divided into two dimensions: Physical Fatigue, schedule persisted for at least one month (yes/ based on the first seven items (scale range of 0 to no)? Respondents had to answer “yes” to all three 21), and Mental Fatigue, based on the last four questions in order to fulfill the criteria for being a items (scale range of 0 to 12). case of SWD. HADS is a self-assessment scale consisting The BSWSQ systematically assesses the of 14 items. The responses range from 0 (no insomnia symptoms and sleepiness/tiredness symptomatology) to 3 (severe symptomatology). separately for different work shifts (day, evening, It has been found to perform well in assessing night shifts) and rest days. The questionnaire is symptom load and case-ness of anxiety (7 items) composed of six question related to: (1) and depression (7 items) in the general population. difficulties falling asleep; (2) difficulties Based on these tools, a participant was maintaining sleep; (3) early morning awakening; indicated as a case of SWD if he or she answered (4) non-restorative sleep; (5) sleepiness/tiredness “yes” to all 3 questions in the questionnaire used during work hours and (6) during time off work to confirm symptoms of SWD but who did not on workdays. Each item is rated on a five-point have an answer of “always” in the GSAQ, and had scale, ranging from 0 to 4 (‘never’, ‘rarely’, scores above the clinical cut-off level for either ‘sometimes’, ‘often’ and ‘always’). A composite insomnia or sleepiness. score for each shift is calculated, ranging from 0 Results of responses to these various tools to 24. An increasing sum score indicates were tabulated and expressed via descriptive increased sleep-wake disturbance in relation to statistics, using percentages for categorical that shift. The clinical cut-off of at least 18 variables, and means and standard deviations for indicated severe problems. continuous variables. Results for SWD-negative The ESS is composed of eight items, each and SWD-positive participants were calculated one to describe a specific situation for which and compared using t test of two independent respondents are asked to assess the likelihood of samples for interval data, and Z test for two them falling asleep or dozing off on a scale proportion populations for categorical data. Chi- ranging from 0 (would never doze off) to 3 (high square test of independence was used in Vol. 16 | Issue 02 | June 2015 12 Shift work disorder among nurses Table 1. Demographic and Work Profile of Respondents with or without Shift Work Disorder (SWD) Total Without SWD Profile of Nurses With SWD (n=24) P-value (N=133) (n=109) Age, mean (sd) 33.0 (8.9) 28.3 (6.4) 34.0 (9.1) 0.004 Gender, n (%) Female 85 (63.9) 14 (58.3) 71 (65.1) 0.529 Male 48 (36.1) 10 (41.7) 38 (34.9) 0.529 Marital status, n (%) Single 53 (39.8) 8 (66.7) 45 (41.3) 0.472 Married 78 (58.6) 16 (33.3) 62 (56.9) 0.379 Widow/Separated 2 (1.5) 0 (0.0) 2 (1.8) 0.503 Employment status, n (%) Staff Nurse 126 (94.7) 24 (100.0) 102 (93.6) 0.201 Supervisor 7 (5.3) 0 (0.0) 7 (6.4) 0.201 Duration of employment, mean (sd) 7.3 (7.6) 4.2 (6.0) 8.0 (7.7) 0.027 Work schedule, n (%) Rotating Shift 126 (94.7) 24 (100.0) 102 (93.6) 0.201 12-Hour Shift 7 (5.3) 0 (0.0) 7 (6.4) 0.201 Co-morbid, n (%) 52 (39.1) 12 (50.0) 40 (36.7) 0.226 Unplanned overtimes, n (%) 102 (76.7) 24 (100.0) 78 (71.6) 0.003 16-Hour Duty, n (%) 120 (90.2) 23 (95.8) 97 (89.0) 0.308 Smoker, n (%) 12 (9.0) 1 (4.2) 10 (9.2) 0.418 Cups of coffee, mean (sd) 1.1 (1.2) 1.6 (1.4) 1.0 (1.2) 0.029 Sleep medicines, n (%) 9 (6.8) 4 (16.7) 5 (4.6) 0.033 Alcohol consumption for sleep, n (%) 8 (6.0) 4 (16.7) 4 (3.7) 0.016 determining the relationship of two categorical and that of those with and without SWD. data. Odds ratios were computed in the test of Twenty-four nurses were identified to have relationship to further describe associations. SWD, for a prevalence of 18.0% (95% CI 12.4%, 25.5%). Those with SWD had significantly lower RESULTS years of service than those without SWD (4.2 vs A total of 133 nurses were included in the 8 years, respectively; p=0.027). Those with SWD study. Majority was female (63.9%) and married had significantly higher rates of unplanned (58.6%). The mean age was 33 years. Table 1 overtimes compared to those without SWD describes the profile of the included participants, (100% vs 71.6%; p=0.003). They also had Table 2. Shift Work Disorder (SWD) and Work-related Errors Without work- With work-related related errors Chi Square Odds Ratio P-value errors (n= 72) (n=61) With SWD 19 (79.2%) 5 (20.8%) (n=24) 7.39 4.015 0.007 Without SWD 53 (48.6%) 56 (51.3%) (n=109) 13 Phil J Chest Dis 2015 Jocson MCI et al significantly higher consumption of coffee (mean with SWD. In contrast, no statistically significant of 1.6 vs 1 cup/day; p=0.029). Nurses with SWD association was found between SWD and had a significantly higher use of sleep medicines occupational accidents (p=0.069) (Table 3), and (16.7% vs 4.6%; p=0.033) and alcohol (16.7% vs between SWD and vehicular accidents (p=0.149) 3.7%; p=0.016) for sleep compared with those (Table 4). without SWD. There were no significant Physical and mental fatigue scores are differences between gender, marital status, significantly higher in nurses with SWD than presence of co-morbidities, employment status, those without SWD (Table 5). Furthermore, the work schedule (rotating 8-hour shifts or 12-hour odds of being a case or borderline case for shifts), performance of 16-hour duties, and anxiety or depression were higher among those cigarette smoking. with SWD (Table 6 and 7, respectively). We found that SWD was significantly associated with self-reported work-related errors DISCUSSION (p=0.007) (Table 2). The odds of work-related SWD is a type of circadian rhythm sleep error were four-fold higher among those disorder in which the endogenous circadian Table 3. Shift Work Disorder (SWD) and Occupational Accidents With Without Chi Square Odds Ratio p - value occupational occupational accidents (n=14) accidents (n=119) With SWD (n=24) 5 (20.8%) 19 (79.2%) 3.303 2.924 0.069 Without SWD (n=109) 9 (8.3%) 100 (91.7%) Table 4. Shift Work Disorder (SWD) and Vehicular Accidents With vehicular Without vehicular Chi Square Odds Ratio p - value accidents (n=12) accidents (n=121) With SWD (n=24) 4 (16.7%) 20 (83.3%) 2.085 2.525 0.149 Without SWD (n=109) 8 (7.3%) 101 (92.7%) Table 5. Shift Work Disorder (SWD) and Fatigue Physical Fatigue Mean score Mean Difference t computed p - value With SWD (n=24) 10.58 3.99 4.69 <0.001 Without SWD (n=109) 6.59 Mental Fatigue Mean score Mean Difference t computed p - value With SWD (n=24) 4.92 1.74 3.67 <0.001 Without SWD (n=109) 3.18 Vol. 16 | Issue 02 | June 2015 14 Shift work disorder among nurses Table 6. Shift work disorder (SWD) and Anxiety Non-case Borderline/ Chi Square Odds Ratio (n= 106) Case (n=27) p - value With SWD (n=24) 11 (45.8%) 13 (54.2%) 20.758 8.019 <0.001 Without SWD (n=109) 95 (87.2%) 14 (12.8%) Table 7. Shi� work disorder (SWD) and Depression Non-case Borderline/ Chi Square Odds Ratio p - value (n=109) Case (n=22) With SWD (n=24) 15 (62.5%) 9 (37.5%) 9.318 4.431 0.002 Without SWD (n=109) 96 (88.0%) 13 (12.0%) sleep-wake cycle is normal, and the disturbance erratic sleep-wake schedule than others.1,9-11 Age arises from conflict between the pattern of sleep is an important predictor of shift work and wakefulness generated by the circadian intolerance.11 In contrast to our findings of rhythm and the desired pattern of sleep and younger age with fewer years of service wakefulness required by shift work. SWD is associated with SWD, researches have shown characterized by excessive sleepiness and/or that older shift workers, compared to their insomnia for at least one month in association with younger counterparts, are less likely to adapt to a shift work schedule, provided the symptoms are abrupt changes in shifts, and take more time to not better explained by another sleep/wake physically recover after working several days on disorder. the night shift.12 Majority (83%) of the The existence of many different work respondents were aged 40 years and below, so schedules makes it difficult to estimate the actual there was likely a small sample size for subjects prevalence of clinically significant sleep who had shorter, fragmented sleep. There are a disturbances. Daytime sleepiness resulting from number of papers that substantiate the cause of working atypical schedules is most commonly the poor adjustment of the older shift workers to seen in relation to the night shift and early shift work to the greater amount of sleep morning shifts.5,7,8 The prevalence of SWD is not disturbances.13-16 The reason why the younger clear because most studies have not used nurses in our study have shift work disorder is standardized diagnostic criteria. According to a unclear. We hypothesize that, being relatively study that combined formal diagnostic criteria new to their employment and with less work with an epidemiologic sample (using experience, they might have poorer coping questionnaire data and not clinical evaluation), strategies to their work shifts. 32% of night workers and 26.1% of rotating According to ICSD-2, there is no gender workers met the minimal criteria for SWD.1 In our difference in the vulnerability to SWD.5 Other study, the prevalence of SWD among the nurses studies however show that women are more was 18 % (95% CI 12.4%, 25.5%). likely than men to experience the adverse effects The ability to adapt to shift work varies on of shift work. Female shift workers have the an individual basis, with some shift workers being tendency to sleep less and feel more drowsy and more susceptible to the negative effects of an fatigued than their male counterparts.12 This 15 Phil J Chest Dis 2015 Jocson MCI et al tendency may be explained by the fact that most al (2002) found that rotating shift work affects household tasks and familial duties tend to be the the amount of sleep, but not the rate of errors responsibility of the woman. among workers on a three-shift schedule.19 This study showed that nurses with SWD Suzuki and colleagues (2004) presented no significantly used sleep medicines and alcohol for association between shift work and occupational sleep. There was also significantly higher accidents, but rather found an association consumption of coffee among those with SWD. between mental health and medical errors.20 There was no significant difference for cigarette Drake et al showed that accidents due to smoking. Drug and alcohol dependency may result sleepiness among shift workers meeting the from attempts to alleviate the sleep and criteria for SWD was double that of non-affected wakefulness disturbances produced by shift work.5 shift workers.1 There is strong evidence that Two frequently used coping strategies that have night, evening, rotating and irregular shifts are ‘physiological’ effects are smoking and drinking associated with an elevated risk of occupational coffee during the night. Consumption of cigarettes injury. According to population-attributable risk and coffee increases physiological activation, calculations, 6% to 7% of workplace injuries can thereby helping to keep the worker awake. be attributed to shift work. These risks are However, smoking habits of shift workers cannot generally understood to arise from two primary be generalized. One has to keep in mind that factors: (1) worker fatigue due to sleep smoking habits may be related to the type of work disturbance, long work hours and their resulting and industry. Importantly, smoking is strictly effect on circadian rhythms; and (2) typically forbidden among healthcare workers on duty. lower levels of supervision and co-worker There was a significant difference in support during non-daytime work schedules.21 performance of unplanned or mandatory overtimes Safety is further jeopardized during the among nurses diagnosed with SWD versus nurses commute home, with shift workers being at an without SWD (p=0.003). All of the nurses who increased risk for motor vehicle accidents, developed SWD had undergone unplanned or especially after the night shift.11 In our study, mandatory overtimes. Extended work schedules only 12 out of the 133 nurses experienced and overtime prove to be hazards whatever the vehicular accidents as drivers and there was no workplace. In a study by Scott et al in 2006, significant difference between those with and nurses have reported decreased ability to be without SWD. Majority of the respondents take vigilant, and struggled to stay awake when work public transportation on their commute home, hours are extended.17 which would explain the lower incidence of Our study showed significant association of vehicular accidents. SWD with self-reported work-related errors SWD was significantly associated with (p=0.007), but no significant association with higher physical and mental fatigues scores occupational injuries (p=0.069) and vehicular (p<0.001). The effects of fatigue on performance accidents (p=0.149). The nurses with SWD had are manifested as decreased alertness, vigilance, four-fold higher odds of experiencing work-related concentration, judgment and mood. As sleep- errors compared to those nurses without SWD. deprived workers continue to perform routine Results of studies on the ability of individuals to work tasks while fighting the biologic drive for maintain adequate levels of performance over sleep, they may fail to recognize the warning work shifts, particularly those shifts spanning signs of fatigue, which can have grave nighttime hours, are mixed. Gold and colleagues consequences. Nurses on night shifts have (1992) reported that the main factor associated reported high levels of stress, physical with medical errors was work shift.18 Kawada et exhaustion, and mental exhaustion.22 According Vol. 16 | Issue 02 | June 2015 16 Shift work disorder among nurses to Mayo and Duncan, 16.5% of nurses identified ed. Westchester, IL: American Academy of fatigue as a reason for failure in the administration Sleep Medicine; 2005. of medication.23 Fatigue is associated with 6. De Castro AB, Fujishiro K, Rue T, Tagalog frequent injury of nurses and road trauma EA, et al. Association between work schedule (especially after a night shift).24-26 characteristics and occupational injury and There are significant associations between illness. Int Nursing Rev 2010;57:188-194. SWD and anxiety (p<0.001), and with depression 7. Lu BS, Zee PC. Circadian rhythm sleep depression (p=0.002). These findings are disorders. Chest 2006;130:1915-1923. consistent with the study of Haines et al (2008) 8. Biovan DB, Tremblay GM, James FO. wherein they cited several studies that pointed to Working on atypical schedules. Sleep Med an association between shift work and 2007;8:578-589. “psychological distress, depression, anxiety and 9. Barion A, Zee PC. A clinical approach to burnout.”27 Their study focused on the possibility circadian rhythm sleep disorders. Sleep Med of work-family conflict: that shift work may 2007;8:566-577. interfere with participation in family life, because 10. Richardson GS. Shift work sleep disorder. In: of both scheduling and fatigue, which may Lee-Chiong T. Sleep: A Comprehensive increase the risk of depression. Jamal (2004) noted Handbook. Hoboken, NJ: John Wiley & Sons that shift workers report significantly higher Inc; 2006. burnout, emotional exhaustion, job stress and 11. Drake CL, Wright KP. Shift work, shift-work psychosomatic health problems (such as disorder, and jet lag. In: Kryger MH, Roth T, headaches, upset stomach, difficulty falling Dement WC. Principles and Practice of Sleep asleep) than workers on a regular day schedule.28 Medicine. St. Louis, MO: Elsevier Saunders Inc; 2011. References: 12. Quigg M. Circadian rhythms: problems with the body clock. In: Watson NF, Vaughn BV, 1. Drake CL, Roehrs T, Richardson G, et al. Shift eds. Clinician’s Guide to Sleep Disorders. work sleep disorder: prevalence and New York, NY: Taylor & Francis Group; consequences beyond that of symptomatic day 2006. workers. Sleep 2004;27:1453-1462. 13. Koller M, Kundi M, Cervinka R. Field 2. Lu ZW, Ann Gwee K, Yu Ho K. Functional studies of shift work at an Austrian oil bowel disorders in rotating shift nurses may be refinery. I: Health and psychosocial related to sleep disturbances. Eur J wellbeing of workers who drop out of Gastroenterol Hepatol 2006;18:623-627. shiftwork. Ergonomics 1978;12: 835–847. 3. Brown DL, Feskanich D, Sanchez BN, et al. 14. Åkerstedt T, Torsvall L. In: Sleep (ed. Koella Rotating night shift work and the risk of WP) Basel: Krager; 1981. ischemic stroke. Am J Epidemiol 15. Foret J, Bensimon B, Benoit O, Vieux N. In: 2009;169:1370-1377. Night and Shift Work: Biological and Social 4. Shift work and sleep. National Sleep Aspects (Reinberg A, Vieux N, anAndlauer P, Foundation (NSF). Available at: eds.). Oxford: Pergamon Press; 1981. http://sleepfoundation.org/sleep-topics/shift- 16. Härmä M, Ilmarinen J. Sleep Res work-and-sleep. 1987;16:612. 5. American Academy of Sleep Medicine. 17. Scott LD, Hwang WT, Rogers AE. The International Classification of Sleep impact of multiple care giving roles on Disorders: Diagnostic and Coding Manual. 2nd fatigue stress and work performance among 17 Phil J Chest Dis 2015 Jocson MCI et al hospital nurses. J Nursing Adm 2006;23:1149-1163. 2016;36:86-95. 18. Mayo A, Duncan D. Nurse perceptions of 18. Gold DR, Rogacz S, Bock N, Tosteson TD, medication errors: What we need to know for Baum TM, Speizer FE, Czeisler CA. Rotating patient safety. J Nurs Care Qual shift work, sleep, and accidents related to 2004;19:209-17. sleepiness in hospital nurses. Am J Public 19. Sharma A, Verma A, Malhotra D. Job Health 1992;82: 1011-1014. performance and chronic fatigue syndrome in 19. Kawada T, Suzuki S. Monitoring sleep hours nurses. Asian Soc Sci 2010;6:167-71. using a sleep diary and errors in rotating 20. Barker L, Nussbaum M. Fatigue, shiftworkers. Psychiatr Clinical Neurosci performance and the work environment: a 2002;56:213-214. survey of registered nurses. J Adv Nur 20. Suzuki K, Ohida T, Kaneita Y, Yokoyama E, 2011;67:1370–82. Miyake T, Harano S, Yagi Y, Ibuka E, Kaneko 21. Scott L, Hwang W, Rogers A, Nysse T, Dean A, Tsutsui T, Uchiyama M. Mental health G, Dinges D. The relationship between nurse status, shift work, and occupational accidents work schedules, sleep duration, and drowsy among hospital nurses in Japan. J Occup driving. Sleep 2007;30:1801-7. Health 2004;46:448-54. 22. Haines VY, Marchand A, Rousseau V, 21. Mustard C. Shift work and risk of work injury. Demers A. The mediating role of work-to- Presented at: The Health effects of Shift Work family conflict in the relationship between (Scientific Symposium); Toronto; April 12, shiftwork and depression. Work Stress 2010. 2008;22:341-356. 22. Dorrian J, Lamond N, Van den Heuvel C, 23. Jamal M. Burnout, stress and health of Pincombe J, Rogers AE, Dawson D. A pilot employees on non-standard work schedules: study of the safety implications of Australian a study of Canadian workers. Stress Health nurses’ sleep and work hours. Chronobiol Int 2004;20:113-119. Vol. 16 | Issue 02 | June 2015 18 Predictive formula for optimal CPAP in OSA VALIDATION STUDY Development and Validation of a Predictive Formula for Optimal CPAP Level in Filipino Patients with Obstructive Sleep Apnea seen in Lung Center of the Philippines Emelie Grace J. Bautista, MD, FPCP, DPCCP; Virginia de los Reyes, MD, FPCP, FPCCP ABSTRACTBackground: Prevalence of obstructive sleep apnea (OSA) is increasing among Asians. Sleep studies (polysomnography) are required to diagnose and determine the effective pressure to eliminate apneas. However, performing sleep studies may be expensive and time-consuming, and may delay the initiation of treatment. This study aimed to predict the optimal continuous positive airway pressure (CPAP) level among Filipino OSA patients by developing a CPAP prediction formula derived from sleep measurements and anthropometric parameters. This formula was meant to aid in titration studies and provide empiric pressure during initiation of CPAP therapy when titration is not used, thus avoiding delays in CPAP treatment. This study also aimed to compare the applicability of this new formula to two other validated prediction formulas.Methods: We retrospectively reviewed the records of 202 Filipino patients with OSA who underwent successful CPAP titration. We divided the 202 patients into 2 groups: Group 1 (n=127) was used for model-setting. Group 2 (n=75) was used for validation. In Group 1, we used stepwise multiple linear regression analysis to develop a predictive formula for optimal CPAP. In Group 2, we then compared the predicted CPAP derived from the new formula to those derived using the Hoffstein equation and the Lin equation. Results: The resultant predictive formula was: predicted pressure (cm H2O) = 9.697 + (body mass index [kg/m2]) x 0.092) + (apnea-hypopnea index x 0.015) - (min SpO2 x 0.057). This LCP formula accounted for 33.5% of the total variance (P < .001). The rate of patients being optimally estimated CPAP (defined as at most 1 cm H2O difference between the measured pressure and the predicted pressure) were 52%, 22%, and 30% using this formula, the Hoffstein formula, and the Lin formula, respectively. Compared with the present formula, the Hoffstein formula significantly underestimated CPAP, but no significant difference was noted with the Lin formula. Conclusion: A formula in predicting optimal CPAP level for Filipino OSA patients was developed and validated in this study. Although the predictive formula developed in this study in may not be sufficient to be a substitute for manual CPAP titration, it may be used to set the minimum CPAP level during titration among Filipino OSA patients. It predicted the optimal CPAP level better than the Hoffstein formula, suggesting that race may be an important factor in predicting CPAP level. However, it was not able to accurately predict the prescribed CPAP level; hence its usefulness is limited in some clinical settings.Keywords: Predictive formula; Continuous positive airway pressure; Obstructive Sleep Apnea; Asia19 Phil J Chest Dis 2015 Bautista EGJ et al INTRODUCTION pressure. 6 It was also used during initiation of Obstructive sleep apnea (OSA) is a CPAP therapy when titration is not used. disabling condition characterized by repetitive Of these prediction formulas, the Hoffstein obstruction of the upper airway often resulting in formula (PHoff = [0.16 x BMI] + [0.13 x NC] + oxygen desaturation and frequent arousals from [0.04 x AHI] – 5.12) is more commonly used to sleep.1 This condition predisposes to medical predict optimal CPAP settings.7 However, the complications such as hypertension, coronary Hoffstein formula did not correlate well with the vascular disease, congestive heart failure, optimal CPAP level in Asians, which may be due cerebrovascular disease, glucose intolerance, and to differences in body sizes and craniofacial impotence. OSA is also considered as a social features of Asians versus Caucasians. hazard leading to daytime somnolence and One of the studies that addressed this cognitive impairment that could result in low concern was the Lin formula, which was based productivity and increased risk for automobile on a Taiwanese population. BMI and AHI were crashes.2 the two most important predictors of CPAP In Asia, the prevalence of symptomatic setting in the Lin formula (PLin = 0.52 + [0.174 x OSA in middle-aged men and women is 4.1% to BMI] + [0.042 x AHI]).8 In its cross-validation 7.5% and 2.1% to 3.2%, respectively, which are study in 32 patients, 86% were within ±2 cm H20 similar to Caucasian populations.2 However, some of the predicted pressure, which is higher than the authors conclude that OSA is likely under- rate acquired when using the Hoffstein formula recognized in Asian countries. (82%). Risk factors for OSA, such as snoring, age, Both formulas are validated studies, but gender, neck circumference (NC), body mass none have been validated among Filipinos. Thus, index (BMI), craniofacial features and excessive the purpose of this study was to develop and daytime sleepiness have been used to screen for validate our own prediction formula among OSA.2 Of these, there was a strong correlation Filipino OSA patients, and to compare the between anthropometric measurements (BMI, % applicability of this new formula with the two predicted body weight, neck and abdominal other formulas previously mentioned. circumferences) and apnea-hypopnea index (AHI). The definitive diagnosis of OSA is currently METHODOLOGY based on a combination of these clinical features This study was conducted from January and overnight polysomnography (PSG).3 2009 to April 2014 at the Sleep Laboratory and Laboratory continuous positive airway pressure Sleep Disorders Clinic of the Lung Center of the (CPAP) titration is typically performed to Philippines (LCP). We retrospectively reviewed determine the optimal pressure required to medical records of 202 consecutive patients with eliminate apnea,4 but this procedure is time- OSA (with an AHI greater than 15 events per consuming, expensive and often leads to delayed night) who were at least 18 years old and had CPAP initiation. undergone CPAP titration. We enrolled only Thus, several researchers have proposed patients with optimal or good titration. Optimal other methods to determine optimal CPAP settings titration was defined as a reduction of the for OSA patients, such as CPAP prediction respiratory disturbance index (RDI) to less than 5 formulas.5 These formulas were derived from events/hour for at least a 15-minute duration, sleep measurements and anthropometric including supine rapid eye movement (REM) parameters, and used to set the initial CPAP level sleep at the selected pressure that is not during CPAP titration, thus minimizing the risk of continually interrupted by spontaneous arousals inadequate time to determine the effective or awakenings. Good titration was defined as a Vol. 16 | Issue 02 | June 2015 20 Predictive formula for optimal CPAP in OSA a reduction of RDI to less than or equal to 10 of 15 min or more. CPAP was increased if there events/hour, or by 50% if the baseline RDI was were 2 obstructive apneas, 3 hypopneas, or 5 less than 15 events/hour, and should include respiratory-effort-related arousals, or at least 3 supine REM sleep.4 min of loud or unambiguous snoring. The titrated Demographic and anthropometric variables, CPAP level was defined as the lowest effective including age, gender, BMI, neck circumference, pressure. Mallampati score, and tonsillar grade were We then grouped the patients into two: recorded. Patients with incomplete or missing Group 1 was composed of the first 127 data, and those with suboptimal CPAP titration consecutive patients, and was used for the model- level were excluded from the study. setting set (minimum sample size of 100 to be All included patients had to have undergone representative, with a 95% margin of error). The PSG. This included the following: remaining patients comprised Group 2 (n=75), electrooculography (REOG-M1 and LEOG-M2), which was used for validation. Results are electroencephalography (C3-M2, C4-M1, F3-M2, presented as means and standard deviations. F4-M1, O1-M2, O2-M1), electromyography, Statistics software (SPSS) was used for the electrocardiography (modified V2 lead), oxygen statistical analysis, and a p-value of less than 0.05 saturation (SpO2), and a microphone to detect was considered significant. Chi-square test or snoring. Standard PSG was performed by a well- independent t tests were used to assess trained PSG technologist using the REMstar Plus differences of variables between groups. These w/C-flex (Respironics). Electromyography was variables included age, sex, anthropometric recorded on the chin and both anterior tibial parameters (NC, tonsillar grade, Mallampati muscles. Airflow assessment was performed with score and BMI), AHI, and min SpO2 during a nasal pressure transducer and an oronasal sleep. Multiple linear regression analysis was thermistor. Thoracoabdominal movement done in Group 1 to identify independent assessment was performed by respiratory predictive variables and develop a predictive inductance plethysmography. Pulse oximetry was formula for optimal CPAP. Using values from used to determine oxygen saturation. Sleep and Group 2, we compared the CPAP pressure associated events were scored according to the derived from the regression equation with those American Academy of Sleep Medicine manual by derived from the Hoffstein and Lin formulas a PSG technologist. An apnea was defined as a using Z test on two proportion population. drop in the peak thermal sensor excursion by at Prediction was classified as optimal estimation least 90% from baseline for at least 10 seconds. A (difference of 1 cm H2O or less between hypopnea was defined as a nasal pressure signal measured pressure and predicted pressure), drop of at least 30% from baseline for at least 10 underestimation (predicted pressure is greater seconds with at least 3% reduction of SpO2 from than 1 cm H2O lower), or overestimation the pre-event baseline or with an arousal. The AHI (predicted pressure is greater than 1 cm H2O was defined as the ratio of apneas and hypopneas higher). per hour. We recorded the AHI and minimum oxygen saturation (min SpO2) during sleep from RESULTS the PSG data. Medical records and PSG data of 313 All patients underwent manual CPAP patients were reviewed, of which 111 were titration until the lowest effective CPAP level was excluded due to lack of data and/or suboptimal reached based on clinical guidelines. The starting titration. The remaining 202 patients (165 men, CPAP was 4 cm H2O, and the pressure was 37 women) were included in the analysis. The increased by at least 1 cm H2O within an interval mean age was 45 ± 11.3 years. Mean NC was 21 Phil J Chest Dis 2015 Bautista EGJ et al 41.8 ± 5.6 cm. Mean BMI was 33.5 ± 7.6 kg/m2. Table 1. Anthropometric and Polysomnographic Data in Group 1 vs Group 2 Mean AHI was 73.7 ± 35.6. Mean min SpO2 was Group 1 Group 2 P-value 73.8 ± 14.0 cm H2O. Age (years) 46.2 (11.8) 43.7 (10.8) Patients were divided into 2 groups, with 0.145 (mean, std) 127 patients for Group 1 and 75 patients for Group Gender, n (%) 2. There were no significant differences between Female 26 (20.5) 11 (14.7) 0.303 the 2 groups in terms of age, sex proportions, NC, Male 101 (79.5) 64 (85.3) tonsillar grade, Mallampati score, BMI, and PSG Neck circum- 41.6 (5.7) 42.2 (5.3) data (AHI and min SpO during sleep) (Table 1). ference (cm) 0.494 2 (mean, std) Multiple linear regression revealed three Tonsillar Grade, factors significantly affecting the CPAP level: n (%) BMI (p=0.001), min SpO2 (p<0.001), and AHI 0 7 (5.5) 1 (1.3) (p=0.023) (Table 2). An increase in BMI would 1 34 (26.8) 11 (14.7) 0.119 likely increase CPAP level by 0.092 holding other 2 53 (41.7) 42 (56.0) 3 31 (24.4) 20 (26.7) variables constant. Likewise, an increase in AHI 4 2 (1.6) 1 (1.3) would likely result to an increase to their CPAP Mallampati level by 0.015. Min SpO2 had an inverse effect, score, n (%) decreasing CPAP level by 0.057 holding other 0 0 (0.0) 0 (0.0) variables constant. 1 21 (16.5) 14 (18.7) 0.907 2 46 (36.2) 29 (38.7) The resultant predictive formula were as 3 37 (29.1) 21 (28.0) follows: predicted pressure (Ppred) (cm H2O) = 4 23 (18.1) 11 (14.7) 9.697 + (body mass index [kg/m2]) x 0.092) + BMI (kg/m2) 33.8 (7.7) 33.1 (7.5) (mean, std) 0.528 (apnea-hypopnea index x 0.015) - (min SpO2 x AHI (per hour) 73.7 (34.8) 73.7 (37.1) 0.057). This formula was able to predict the CPAP >.999 (mean, std) level, accounting for 33.5% of the total variance Min SpO (cm 73.8 (13.6) 73.8 (14.8) 2 0.989 (p<0.001). Using this formula, 81.2% of the H2O) (mean, std) patients were within ± 2 cm H2O of the predicted BMI, body mass index. AHI, apnea-hypopnea index; pressure (Table 3) using this formula. SpO2, partial pressure of oxygen (oxygen saturation). The predicted pressure from our formula (Ppred) was compared to the pressures derived from the Hoffstein formula and the Lin formula formula, optimal pressure was estimated at a among the patients in Group 2. When compared to higher rate using the LCP formula (40.90%) the Hoffstein formula (Figure 1), our equation (Figure 2). Tendency to underestimate the CPAP provided an optimal estimation of CPAP pressure level was less in the LCP formula (29.90%), but for 52 patients (40.9%), higher than the Hoffstein the likelihood to overestimate the CPAP level equation (29.3%, 22 patients) at a 5% level of was higher (29.10%). However, these differences significance. There was underestimation in 38 were not statistically significant. patients (29.9%) using the LCP formula, lower than the underestimation rate using the Hoffstein DISCUSSION equation (56%). However the LCP formula This study focused in determining the overestimated CPAP pressure in 37 patients clinical parameters (gender, age, BMI, (29.1%) compared to 11 patients (14.7%) using Mallampati score, tonsillar grade, and NC) and the Hoffstein formula. PSG data that could predict the CPAP level for Then we compared the predicted pressure Filipino OSA patients. It included 202 patients using our formula with those derived from the Lin who had moderate to severe OSA, majority being Vol. 16 | Issue 02 | June 2015 22 Predictive formula for optimal CPAP in OSA Table 2. Multiple Linear Regression Analysis of younger than what is reported in other studies. Variables Associated with Titrated CPAP Increased awareness regarding the adverse consequence of OSA and access to insurance Beta t P- Model Coefficient computed value benefits may explain the frequency of OSA Constant 9.697 6.359 <0.001 diagnosis in the younger age group. Body mass index 0.092 3.282 0.001 Obesity may compress and/or alter the properties of the upper airway by the deposition Minimum oxygen -0.057 -3.84 <0.001 saturation of fat tissue. Furthermore, central obesity may induce reduction in lung volume, thus impairing a Apnea-hypopnea 0.015 2.301 0.023 index caudally directed pharyngeal stabilizing traction force directed via the trachea (the so called CPAP, continuous positive airway pressure. “tracheal-tug”).12 The mean BMI of our patients was 33.5 ± 7.6 kg/m2, which was similar to 2 Table 3. Difference of titrated pressure and comparable to the Hoffstein study (34 ± 8 kg/m ) predicted pressure in Group 2 and higher to the Lin study among Taiwanese patients (28.3 ± 4 kg/m2).7,8 In contrast to the Difference (cm H2O) Number of patients (%) studies done in China, BMI appears to be 0 33 (26%) relatively less important and variations in 1 30 (23.7%) craniofacial bony structure more relevant in the 2 40 (31.5%) development of OSA. In this study, obesity may 3 13 (10.2%) be considered as a major risk factor for OSA. ≥4 11 (8.7%) Regarding craniofacial features, Asians were noted to have higher Mallampati scores that may reflect a more severe posterior males. Similar to other studies, OSA is more oropharyngeal narrowing, and larger tonsils could common in males than in females. Several studies further narrow the oropharyngeal area.13 A NC of have shown that male sex has a significant 15.5 inches (39.37 cm) or more increases the risk association with OSA. Studies done by Popovic et of OSA, which may be due to the deposition of al demonstrated that OSA was less frequent in fat around the neck, narrowing and further females due to differences in the structure and closing the upper airway.11,14 In this study, physiological behavior of the upper airway, with majority of the patients have a Mallampati score females having increased activity of the of 2, tonsillar grade of 2, and had a mean NC of genioglossus muscle, which protects their airway 41.8 ± 5.6 cm. However, none of these emerged 9,10 from collapsing during sleep. Other reasons as significant variables in our predictive formula. included difference in upper airway caliber, the PSG studies are required to diagnose OSA, smaller neck size in women, and the smaller size and according to Deegan et al, combination of of the critical soft tissue structures in women. AHI, overnight oximetry data and anthropometric Female hormones also seemed to have a protective measurements could further correctly identify role. Central distribution of body fat in men may patients with or without OSA by 25.3%.3 PSG 11 also explain the predominance of OSA in males. data from our patients showed a mean AHI of th th OSA is common on the 5 and 6 decade. 73.7 ± 35.6 (consistent with severe OSA) and a In this study, the youngest patient diagnosed with mean min SpO2 of 73.8 ± 14.0 cm H2O. AHI and OSA was 18 years old and the oldest was 84 years min SpO during sleep correlated well with OSA th th 2 old. Majority of patients were in the 4 and 5 in this study. decade with a mean age of 45.3 ±11.5 years, After considering the clinical variables that 23 Phil J Chest Dis 2015 Bautista EGJ et al correlate with the development of OSA, BMI, AHI (Taiwan) considered only BMI and AHI in their 8 and min SpO2 during sleep were the parameters formula. The Lin formula was used as a that were used to create the LCP formula in comparison in this present study because of the predicting the CPAP level. Our formula accounted relative ease of obtaining the values of its for 33.50% (p<0.001) of the total variance. To necessary parameters. validate this formula, we compared it to Hoffstein The LCP formula predicted the optimal and Lin formulas using a different set of patients CPAP level among Filipino OSA patients at a (Group 2). Hoffstein formula is the most widely higher rate than the Hoffstein formula (40.9% vs used equation in predicting optimal CPAP level, 29.3%). It also has a lower tendency to and it was developed based on Caucasian OSA underestimate the predicted CPAP level, although patients. BMI, AHI and neck circumference were it has a tendency to slightly overestimate the considered in calculating the optimal pressure. predicted pressure. Overestimation is preferred Other studies used the Hoffstein formula for over underestimation because the PSG validating newer CPAP pressure prediction technologist is guided to reach the higher equations. In Asia, there were three validation predicted CPAP level during titration rather than studies on the prediction of optimal CPAP level setting the goal at the lower level, which could be (one each in Korea, Taiwan, and Japan); however, subtherapeutic. The Lin formula and the LCP none were conducted among Filipinos. BMI and formula has no significant differences in AHI or RDI were consistently included in these estimating the CPAP level, suggesting that race studies. The study done by Lee et al among may be an important factor in predicting CPAP Koreans also included min SpO2 and Epworth level. sleepiness scale (ESS) score in predicting optimal We therefore conclude that the LCP pressure, while Akakoshi et al (Japan) used mean formula is more likely to predict the optimal oxygen saturation and cephalometric measure- CPAP level than the Hoffstein formula. Although ments (these formulas were not used as the predictive equation formulated in this study in comparators, as cephalometric measurements were may not be sufficient to be a substitute for the not measured in local patients, and data on ESS manual CPAP titration, it may be used to improve were incomplete in our patients).15,16 Lin et al the manual titration success rate by estimating or Figure 1. Rate of optimal prediction, underestimation, or overestimation of continuous positive airway pressure using the Lung Center of the Philippines formula vs the Hoffstein formula 60.00% p<0.001 Our Equation Hoffstein 50.00% 56.00% p=0.099 40.00% 40.90% p=0.020 30.00% 29.90% 29.30% 29.10% 20.00% 10.00% 14.70% 0.00% Underestimation Optimal Overestimation Vol. 16 | Issue 02 | June 2015 24 Predictive formula for optimal CPAP in OSA Figure 2. Rate of optimal prediction, underestimation, or overestimation of continuous positive airway pressure using the Lung Center of the Philippines formula vs the Lin formula Our equation p=0.897 Lin equation 45.00% p=0.201 40.00% 35.00% 40.90% p=0.222 30.00% 40.00% 25.00% 29.90% 38.70% 29.10% 20.00% 21.30% 15.00% 10.00% 5.00% 0.00% Underestimation Optimal Overestimation increasing the starting pressure needed for the pressure and be more cost-effective. Finally, we titration (particularly in split-night studies) among recommend utilizing the LCP formula Filipino OSA patients. Moreover, in a resource- prospectively to a larger number of Filipino OSA limited setting, treatment of OSA with a fixed patients. CPAP machine using a predictive formula to identify the CPAP level may be a justified option. References: This strategy may reduce cost and facilitate faster 1. A Report of the American Academy of Sleep treatment of OSA. Furthermore, in institutions Medicine Task Force. Sleep-related breathing where full diagnostic or therapeutic studies, and disorders in adults: recommendations for split-night studies are done, after determining the syndrome definition and measurement BMI, estimated AHI and min during the diagnostic techniques in clinical research. Sleep PSG, the PSG technologist may now be guided to 1999;22(5):667-89. reach the computed predicted optimal pressure 2. Lam B, Lam DC, Ip MS. Obstructive sleep during the titration study. Through this, it would apnoea in Asia. Int J Tuberc Lung Dis avoid multiple pressure changes, risk of 2007;11(1):2-11. inadequate timing, and unnecessary expenses. 3. WT. Predictive value of clinical features for the Obstructive Sleep Apnea Syndrome. Eur CONCLUSION Respir J 1996;9:117-124. A formula in predicting CPAP level for Filipino 4. Kushida CA, Chediak A, Berry RB, Brown OSA patients was developed and validated in this LK, Gozal D, Iber C, et al. Clinical study. It predicted the optimal CPAP level better guidelines for the manual titration of positive than the Hoffstein formula. This prediction airway pressure in patients with obstructive formula does not intend to replace manual CPAP sleep apnea. J Clin Sleep Med 2008;4(2): titration, but rather, under the supervision of a 157-171. sleep specialist, it would be useful in calculating 5. Choi JH, Kim KW, Choi J, Kwon SY, Lee the starting pressure for initiating titration. HM, et al Optimal continuous positive airway Consequently it could eliminate the need for pressure level in Korean patients with multiple pressure changes, minimize the risk of obstructive sleep apnea syndrome. Clin Exp inadequate time in determining the optimal Otorhinolaryngol 2010;3(4):207-211. 25 Phil J Chest Dis 2015 Bautista EGJ et al 6. Makarawate P, Senthong V, Sawanyawisuth K. apnea, obesity, and bariatric surgery. In: CPAP Pressure by CPAP Titration or Kryger MH, Roth T, Dement WC, editors. Prediction Formula. J Sleep Disorders Ther In: Principles and practice of sleep 2013;2:4. medicine, 5th ed. St. Louis: Elsevier 7. Hoffstein V, Mateika S. Predicting nasal Saunders, 2011:1339-1348. continuous positive airway pressure. Am J 13. Li KK, Kushida C, Powell NB, et al. Respir Crit Care Med 1994;150:486-8. Obstructive sleep apnea syndrome: a 8. Lin IF, Chuang ML, Liao MF, Chen NH, Li comparison between Far East Asian and HY. Prediction effective continuous positive white men. Laryngoscope 2000;110:1689– airway pressure in Taiwanese patients with 93. obstructive sleep apnea syndrome. J Formos 14. Peruvemba HL, Thazhepurayil R , Med Assoc 2003;102:215-221. Ponneduthamkuzhi J, Chetambath R. Clinical 9. Popovic RM, White DP. Influence of gender Prediction of OSA in Tertiary Care Setting. J on the waking genioglossal electromyogram Clin Diagnostic Res 2012;6(5):835-838. and the upper airway resistance. Am J Respir 15. Lee GH, Kim MJ, Lee EM, Kim CS, Lee SA. Crit Care Med 1995;152:725-31. Prediction of Optimal CPAP Pressure and 10. Popovic RM, White DP. Upper airway muscle Validation of an Equation for Asian Patients activity in normal women: influence of the with Obstructive Sleep Apnea. Respir Care hormonal status. J Appl Physiol 2013;58(5):810-815. 1998;84:1055-62. 16. Akahoshi T, Akashiba T, Kawahara S, 11. Mortimore IL, Marshall I, Wraith PK, et al. Uematsu A, Nagaoka K, Kiyofuji K, Neck and total body fat deposition in Okamoto N, Hattori T, Takahashi N, nonobese and obese patients with sleep apnoea Hashimoto S. Predicting optimal continuous compared with that in control patients. Am J positive airway pressure in Japanese patients Respir Crit Care Med 1998;157:280–283. with obstructive sleep apnoea syndrome. 12. Olson EJ, Courcoulas AP. Obstructive sleep Respirology 2009;14(2):245-50. Vol. 16 | Issue 02 | June 2015 26 ANC Score, STOP-BANG, and Berlin questionnaire RETROSPECTIVE STUDY Adjusted Neck Circumference Score, STOP-BANG and Berlin Questionnaire As Screening Tools For Obstructive Sleep Apnea Caroline M. Armas, MD; Patrick Gerard L. Moral, MD, FPCP, FPCCP ABSTRACT Background: Obstructive sleep apnea (OSA) is the most common sleep breathing disorder that is increasingly recognized as an important co-morbidity. Overnight polysomnography (PSG) remains as the gold standard for its diagnosis. However, clinical prediction models have also been used to assess its presence or absence. The two common models are the STOP-BANG and Berlin questionnaires. Another tool is the Adjusted Neck Circumference (ANC) score, which is a 4-item questionnaire, which can be used for screening OSA. Objective: To compare the predictive ability of STOP-BANG questionnaire, ANC Score and Berlin questionnaire in screening presence or absence of OSA. Methodology: The records of those who underwent PSG studies from January 2011 to May 2013 through Respicare Enterprises was reviewed for data assimilation. Patients’ clinical profile; data needed for ANC score computation, STOP-BANG and BERLIN questionnaires risk assessment scores; and PSG results were obtained. All data gathered was entered in the data abstraction tool and statistically analyzed. Results: A total of 171 patients were included in the study. Of them, 19 patients had no OSA, 44 had mild OSA, 19 had moderate OSA, and 89 had severe OSA. STOP BANG has the highest sensitivity, positive and negative predictive value among the three tools, while it was tied with ANC scoring system in specificity. STOP-BANG has the highest over-all predictive ability with an area under the curve (AUC) of 0.6486. However, asymptotic Chi-square analysis for AUCs revealed no statistically significant differences among the three procedures (p=0.3132). Conclusion: There was no statistically significant difference in the over-all predictive ability of STOP- BANG questionnaire, ANC Score and Berlin questionnaires. INTRODUCTION oxyhemoglobin desaturations terminating in sleep Obstructive sleep apnea (OSA) is a sleep- fragmentations and decrease in amounts of slow disordered breathing defined by the occurrence of wave and rapid eye movement sleep.2 Clinically, complete or partial airway obstruction during it is defined by the occurrence of daytime sleep.1 The instability of upper airway results sleepiness, loud snoring, witnessed breathing markedly reduced or absent airflow in the nose or interruptions, or awakening due to gasping or mouth, typically accompanied by episodes of choking in the presence of at least five 19 27 Phil J Chest Dis 2015 Armas CM et al obstructive respiratory events (hypopneas, apneas, circumference, waist circumference, and/or body or respiratory effort related arousals) per hour of mass index (BMI), and witnessed breathing sleep.3,4 An apnea-hypopnea index of at least 15 abnormalities during sleep, are now being used to can also characterize OSA.3 assess risk of having OSA. They are found to Recognizing clinically significant OSA is have generally high sensitivities but low important, as the disease is associated with specificities, and thus can be used in excluding significant morbidity and mortality. The the diagnosis of OSA.2 occurrence of excessive daytime sleepiness can The STOP-BANG questionnaire lead to significant impairments in the quality of (Appendix 1) is composed of eight items, life, cognitive performance, and social answerable by yes or no. The first four item functioning. OSA is also considered an consists of question about Snoring, Tiredness, independent risk factor for the development of Observed apnea and presence of high blood cardiovascular diseases, such as hypertension, Pressure (STOP), while the last four questions coronary artery disease, congestive heart failure, covers demographics and anthropometric and stroke.2 Effective treatment has been shown to measurements: BMI, Age, Neck circumference improve psychosocial function and health related (NC) and Gender (BANG). Patients are quality of life among OSA patients. It also showed categorized as either high risk for those who improvement in blood pressure and a decrease in answered “yes” in 3 or more of the questions, or cardiovascular events.5-8 low risk for those who answered “yes” in less The gold standard for diagnosing OSA is a than 3 of the 8 questions. full overnight polysomnography (PSG), which The Berlin questionnaire (Appendix 2) is provides detailed information on sleep state, composed of 11 questions, grouped into three respiratory and gas exchange abnormalities, body categories. A certain point is given for each position, heart rate and rhythm, and muscle tone response. The first category is composed of five and contraction.2 It can also be used in assessing questions relating to snoring, and is positive if the severity of OSA when detected. The severity of total score is 2 or more points. The second OSA is based on the frequency of respiratory category is composed of four questions relating events during sleep, with mild having at least 5 to feeling of tiredness and daytime sleepiness. It and less than 15 events per hour of sleep; is positive if the total score is 2 or more points. moderate with 15 to 3; and severe with more than The third category is composed of two questions, 30 events per her of sleep.3 However, although one on the presence of hypertension and the other considered a gold standard, it has limitations. It is if BMI is over 30. This category is considered resource intensive and generally requires facilities positive if one or both is present. The patients are of a full sleep laboratory and a well-trained then categorized as high risk for those with 2 or physician, and is costly. The high prevalence of more categories that are positive, and low risk if OSA makes it necessary to consider other with only 1 positive category. simplified approaches for its diagnosis.2 Adjusted neck circumference (ANC) Deegan and McNicholas in 1996 studied the (Appendix 3) is a screening tool that consists of predictive value of clinical features for OSA and only four measures, each given a certain point. found that the individual signs and symptoms are The measures included are hypertension (4 of little clinical diagnostic help in the diagnosis of points), snoring (3 points), choking or gasping (3 OSA. However when these are combined, the points), and NC (1 point per centimeter). The predictive value increased.9 Clinical prediction patients are categorized based on the total points, models, which generally consist of combinations with high probability for over 48 points, of anthropometric variables such as neck moderate probability for 43 to 48 points, and low Vol. 16 | Issue 02 | June 2015 28 ANC Score, STOP-BANG, and Berlin questionnaire probability with less than 43 points. each predictive to provide summarizing statistics This study aimed to compare the predictive and representation of each of the method’s powers of the ANC score, STOP- BANG predictive ability. Specifically, area under the questionnaire, and Berlin questionnaire in curve (AUC) index was computed for each detecting the presence or absence of OSA. screening tool to allow comparability. All statistical analyses were performed at a METHODOLOGY significance level of 5%. Using Cohen’s formula This was a retrospective comparative study with the type I error rate set at 5%, type II error on the predictive ability of the ANC and the rate at 20%, effect size at 0.30, and estimated STOP-BANG and Berlin questionnaires. It non-response (“dropping”) rate of 30%, included adult patients aged 18 years and above minimum sample size was computed to be at who underwent overnight PSG (diagnostic or split least 135 patients. night) study (type 1) from January 2011 to May 2013, with the diagnosis of no, mild, moderate, or RESULTS severe OSA based on PSG result. Pediatric Table 1 shows the demographic data of patients, those who underwent titration studies patients. A total of 171 patients were included in alone, or those diagnosed with sleep breathing the study. Of them, 19 patients had no OSA, 44 disorders other than OSA (e.g., central had mild OSA, 19 had moderate OSA, and 89 sleep apnea), or those incomplete data needed for had severe OSA. Majority of the patients were computation of the ANC score and those without males. Across all levels of OSA, excessive or records of STOP-BANG and Berlin question- loud snoring was the most common reason for naires, were all excluded. doing a PSG test. The study was conducted from June 2014 to Tables 2, 3 and 4 show the frequency January 2015. The records of those who distributions of patients with or without OSA underwent PSG studies from January 2011 to May using the ANC score, STOP-BANG 2013 through Respicare Enterprises were questionnaire, and Berlin questionnaires, reviewed for data assimilation. Patients’ clinical respectively. Additionally, the statistical estimates profile (i.e., age, sex, and BMI), data needed for in the evaluation metrics, presented in Table 5, ANC score computation, STOP-BANG and show each questionnaire’s predictive power in BERLIN questionnaires risk assessment scores, determining the presence of OSA. and PSG results were obtained. All data gathered STOP-BANG had the highest performance was entered in the data abstraction tool per patient, in almost all diagnostic measures, except for and the resulting ANC score and STOP-BANG specificity, where it was tied with ANC scoring and Berlin questionnaire outcomes were system. The predictive power of STOP-BANG in statistically analyzed with the PSG results. determining presence and absence of OSA was 98.03% and 21.05% respectively. The respective Statistical analyses figures for ANC score were 86.18% and 21.05%. All analyses were carried out using Stata 12. The likelihood of having OSA if an individual Data was summarized using means, proportions, has a positive ANC score test was 89.73%. and standard deviations, whichever apply. Conversely, 16% of the time, an individual has no Sensitivity, specificity, negative predictive value OSA when the result of the test was negative. (NPV), positive predictive value (PPV), and hit Finally, the percentage of individuals correctly ratio (i.e., accuracy) for each screening tool were classified by the ANC is 78.95%. computed. Receiver operating characteristic Table 6 provides estimates of the overall (ROC) curves and analyses were presented for predictive ability of the three models in 29 Phil J Chest Dis 2015 Armas CM et al Table 1. Patient characteristics No OSA Mild OSA Moderate OSA Severe OSA Number of patients 19 44 19 89 Gender (N; %) Male 14 (73.68%) 36 (81.82%) 15 (78.95%) 76 (85.39%) 13 Female 5 (26.32%) 8 (18.18%) 4 (21.05%) (14.61%) Age (years) Mean 54.00 60.00 45.00 73.00 Range 44.00 49.09 56.63 50.10 Standard deviation 14.24 11.54 11.88 14.87 BMI Mean 28.08 30.23 30.84 32.95 Range 17.25 37.43 22.04 38.34 Standard deviation 4.92 7.3 5.62 9.03 NC (cm) Mean 39.99 39.98 41.16 43.08 Range 18.53 38.34 21.59 31.5 Standard deviation 4.57 6.14 5.07 4.13 Hypertension (N; %) 12 (63.16%) 29 (65.91%) 13 (68.42%) 51 (57.30%) Snoring (N; %) 18 (94.74%) 44 (100.00%) 18 (94.74%) 89 (100.00%) Reasons for undergoing PSG (N; %) Difficulty falling asleep 11 (61.11%) 12 (27.91%) 9 (50.00%) 26 (29.89%) Waking up during night 14 (77.78%) 24 (55.81%) 8 (44.44%) 44 (50.57%) Excessive/loud snoring 13 (72.22%) 37 (86.05%) 12 (66.67%) 73 (83.91%) Un-refreshing sleep 8 (44.44%) 23 (53.49%) 8 (44.44%) 38 (43.68%) Excessively sleepy during day 8 (44.44%) 20 (46.51%) 7 (38.89%) 51 (58.62%) Legs jerking 5 (27.78%) 19 (44.19%) 6 (33.33%) 21 (24.14%) Hallucinations 0 (0.00%) 3 (6.98%) 2 (11.11%) 12 (13.79%) OSA, obstructive sleep apnea; BMI, body mass index; NC, neck circumference; PSG, polysomnography. determining the presence of OSA as measured by standard in diagnosing OSA. However, this the AUC of their respective ROC curves as shown procedure entails a fully equipped sleep in the Figure. Numerically, STOP-BANG had the laboratory with highly trained technologists and largest AUC, followed by ANC score then Berlin. physicians, which may not be available in some However, asymptotic Chi-square analysis for parts of our country. AUCs revealed that the differences between the Clinical prediction models have been three were not statistically significant (p=0.3132). developed to aid in assessing an individual’s risk for OSA, which would indicate the need for DISCUSSION further work-up. Among the different clinical OSA is a sleep breathing disorder that has prediction models, the two most commonly used been increasingly recognized as an important co- in clinics are the STOP-BANG and Berlin morbidity. It has been associated with several questionnaires. In a study done by El-Sayed disorders, such as cardiovascular diseases. As which compared four sleep questionnaires such, its recognition and treatment is important in (namely STOP-BANG, Berlin, STOP, and the overall care of patients. Epworth Sleepiness Scale [ESS]) for screening An overnight PSG remains the gold OSA, STOP-BANG and Berlin had the highest Vol. 16 | Issue 02 | June 2015 30 ANC Score, STOP-BANG, and Berlin questionnaire Table 2. Frequency distribution of presence of OSA Table 4. Frequency distribution of presence of OSA using the ANC Score using the Berlin questionnaire Intermediate High Low TOTAL Low or High TOTAL Probability w/ OSA 139 13 152 Probability w/o OSA 17 2 19 w/ OSA 131 21 152 w/o OSA 15 4 19 TOTAL 156 15 171 TOTAL 146 25 171 OSA, obstructive sleep apnea. OSA, obstructive sleep apnea; ANC, adjusted neck circumference. Table 3. Frequency distribution of presence of OSA a positive relationship was noted between ANC using the STOP-BANG questionnaire score and OSA severity. Higher ANC scores were High risk Low risk TOTAL found to be associated with more severe levels of w/ OSA 149 3 152 OSA, and that the estimated odds that a person’s response is in the more severe direction increases w/o OSA 15 4 19 multiplicatively by almost 11% for every 1-point TOTAL 164 7 171 increase in ANC score.13 In a study done by OSA, obstructive sleep apnea. Sahlqvist et al (2011), participants who received the short version of the questionnaire were more likely to respond (OR 1.48, 95% CI 1.06 to 2.07) sensitivity in predicting OSA (97.55% and to the query items. Item non-response was 95.07%, respectively).11 In a study by Chung et al relatively low, but still significantly higher in the (2008), a lower sensitivity of 68.9% to 87.2% was long questionnaire than with the short version reported for the Berlin questionnaire.12 (9.8% vs 5.8%; p=0.04). They concluded that The ANC score is another clinical shortening a relatively lengthy questionnaire prediction model that can be used as a screening significantly increased response rate.14 tool for OSA. It is a quicker way of assessing the This study compared the predictive ability probability of OSA, having only four items vs the of the two commonly used screening tools, STOP-BANG and Berlin questionnaires with 8 STOP-BANG and Berlin questionnaires, with a and 11 questions, respectively. In a previous study, much shorter screening tool, which was the ANC Table 5. Comparison of accuracy between the ANC Score, STOP BANG, and Berlin questionnaires ANC Score STOP-BANG BERLIN 86.18% 98.03% 91.45% Sensitivity (79.66%, 91.24%) (94.34%, 99.59%) (85.82%, 95.37%) 21.05% 21.05% 10.53% Specificity (6.05%, 45.57%) (6.05%, 45.57%) (1.30%, 33.14%) 89.73% 90.85% 89.10% Positive Predictive Value (83.62%, 94.13%) (85.36%, 94.79%) (83.13%, 93.52%) 16.00% 57.14% 13.33% Negative Predictive Value (4.54%, 36.08%) (18.41%, 90.10%) (1.66%, 40.46%) 78.95% 89.47% 82.46 % Hit Ratio (72.07%, 84.80%) (83.87%, 93.64%) (75.91%, 87.84%) 31 Phil J Chest Dis 2015 Armas CM et al score. The latter showed the lowest sensitivity, pressure therapy. Eur Respir J 2001 Dec; numerically, of the three tools, while it was tied 18(6):996-1002. with STOP-BANG in specificity. However, the 6. Pichel F, Zamarrón C, Magán F, del Campo AUCs of the ROC curves (i.e., predictive ability) F, Alvarez-Sala R, Suarez JR. Health-related of the three screening tools were not significantly quality of life in patients with obstructive different (p=0.3132). sleep apnea: effects of long-term positive airway pressure treatment. Respir Med 2004 CONCLUSION Oct;98(10):968-76. This study showed that there was no statistically 7. Becker HF, Jerrentrup A, Ploch T, Grote L, significant difference in the overall predictive Penzel T, Sullivan CE, Peter JH. Effect of ability of the ANC Score and the STOP-BANG nasal continuous positive airway pressure and Berlin questionnaires. ANC scoring system treatment on blood pressure in patients with was a quicker way of assessing presence or obstructive sleep apnea. Circulation absence of OSA and can be used as an alternative 2003;107(1):68-73. to longer screening tools. 8. Doherty LS, Kiely JL, Swan V, McNicholas WT. Long-term effects of nasal continuous REFERENCES positive airway pressure therapy on 1. Redline S. Genetics of Obstructive Sleep cardiovascular outcomes in sleep apnea Apnea. In Meir TR, Kryger H. Principles and syndrome. Chest 2005 Jun;127(6):2076-84. Practice of Sleep Medicine. Canada: Elsevier 9. Deegan PC, McNicholas WT. Predictive Saunders; 2011. value of clinical features for the obstructive 2. McNicholas WT. Diagnosis of obstructive sleep apnoea syndrome. Eur Respir J 1996 sleep apnea in adults. Proc Am Thorac Soc Jan;9(1):117-24. 2008 Feb 15;5(2):154-60. 10. Boynton G, Vahabzadeh A, Hammoud S, 3. Epstein LJ, Kristo D, Strollo PJ Jr, Friedman Ruzicka DL, Chervin RD. Validation of the N, Malhotra A, Patil SP, Ramar K, Rogers R, STOP-BANG Questionnaire among Patients Schwab RJ, Weaver EM, Weinstein MD; Referred for Suspected Obstructive Sleep Adult Obstructive Sleep Apnea Task Force of Apnea. J Sleep Disord Treat Care 2013 Sep the American Academy of Sleep Medicine. 23;2(4). Clinical guideline for the evaluation, 11. El-Sayed IH. Comparison of four sleep management and long-term care of obstructive questionnaires for screening obstructive sleep sleep apnea in adults. J Clin Sleep Med 2009 apnea. Egyptian J Chest Dis Tuberculosis Jun 15;5(3):263-76. 2012; 61(4):433–441. 4. Fleetham J, Ayas N, Bradley D, Ferguson K, 12. Chung F, Yegneswaran B, Liao P, Chung SA, Fitzpatrick M, George C, Hanly P, Hill F, Vairavanathan S, Islam S, Khajehdehi A, Kimoff J, Kryger M, Morrison D, Series F, Shapiro CM. Validation of the Berlin Tsai W; CTS Sleep Disordered Breathing questionnaire and American Society of Committee. Canadian Thoracic Society Anesthesiologists checklist as screening tools guidelines: diagnosis and treatment of sleep for obstructive sleep apnea in surgical disordered breathing in adults. Can Respir J patients. Anesthesiology 2008 May;108(5): 2006 Oct;13(7):387-92. 822-30. 5. McFadyen TA, Espie CA, McArdle N, 13. Armas C, Moral PG. Adjusted Neck Douglas NJ, Engleman HM. Controlled, Circumference Score in Severity of prospective trial of psychosocial function Obstructive Sleep Apnea. Presented at: 2014 before and after continuous positive airway APSR Congress in Bali, Indonesia, 13-16 Vol. 16 | Issue 02 | June 2015 32 ANC Score, STOP-BANG, and Berlin questionnaire November 2014. Poster P-E-003. and reminder type on response rate to a 14. Sahlqvist S1, Song Y, Bull F, Adams E, complex postal survey: randomised controlled Preston J, Ogilvie D; iConnect consortium. trial. BMC Med Res Methodol 2011 May Effect of questionnaire length, personalisation 6;11:62. Appendix 1 STOP BANG Questionnaire Height _____ inches/cm Weight _____ lb/kg Age _____ Male/Female BMI _____ Collar size of shirt: S, M, L, XL, or _____ inches/cm Neck circumference* _____ cm 1. Snoring: Do you snore loudly (louder than talking or loud enough to be heard through closed doors)? (Yes/No) 2. Tired: Do you often feel tired, fatigued, or sleepy during daytime? (Yes/No) 3. Observed: Has anyone observed you stop breathing during your sleep? (Yes/No) 4. Blood Pressure Do you have or are you being treated for high blood pressure? (Yes/No) 5. BMI: BMI more than 35 kg/m2? (Yes/No) 6. Age: Age over 50 years? (Yes/No) 7. Neck circumference: Neck circumference greater than 40 cm? (Yes/No) 8. Gender: Gender male? (Yes/No) * Neck circumference is measured by staff High risk of OSA: answering yes to three or more items Low risk of OSA: answering yes to less than three items Adapted from: Chung F, Yegneswaran B, Liao P, Chung SA, Vairavanathan S, Islam S, Khajehdehi A, Shapiro CM. STOP Questionnaire. A Tool to Screen Patients for Obstructive Sleep Apnea. Anesthesiology 2008; 108:812–21 33 Phil J Chest Dis 2015 Armas CM et al Appendix 2 Berlin Questionnaire The questionnaire consists of 3 categories related to the risk of having sleep apnea. Patients can be classified into High Risk or Low Risk based on their responses to the individual items and their overall scores in the symptom categories. Category 1: BERLIN QUESTIONNAIRE Item 1: if ‘Yes’, assign 1 point Height (m) ______Weight (kg)______Item 2: if ‘c’ or ‘d’ is the response, assign 1 point Age______Male / Female Item 3: if ‘a’ or ‘b’ is the response, assign 1 point Item 4: if ‘a’ is the response, assign 1 point Please choose the correct response to each Item 5: if ‘a’ or ‘b’ is the response, assign 2 question. points CATEGORY 1 Add points. Category 1 is positive if the total Do you snore? score is 2 or more points _ a. Yes _ b. No CATEGORY 2 _ c. Don’t know 6. How often do you feel tired or fatigued after your sleep? If you snore: _ a. Nearly every day 2. Your snoring is: _ b. 3-4 times a week _ a. Slightly louder than breathing _ c. 1-2 times a week _ b. As loud as talking _ d. 1-2 times a month _ c. Louder than talking _ e. Never or nearly never _ d. Very loud – can be heard in adjacent rooms 7. During your waking time, do you feel tired, 3. How often do you snore? fatigued or not up to par? _ a. Nearly every day _ a. Nearly every day _ b. 3-4 times a week _ b. 3-4 times a week _ c. 1-2 times a week _ c. 1-2 times a week _ d. 1-2 times a month _ d. 1-2 times a month _ e. Never or nearly never _ e. Never or nearly never 4. Has your snoring ever bothered other 8. Have you ever nodded off or fallen asleep people? while driving a vehicle? _ a. Yes _ a. Yes _ b. No _ b. No _ c. Don’t Know If yes: 5. Has anyone noticed that you quit 9. How often does this occur? breathing during your sleep? _ a. Nearly every day _ a. Nearly every day _ b. 3-4 times a week _ b. 3-4 times a week _ c. 1-2 times a week _ c. 1-2 times a week _ d. 1-2 times a month _ d. 1-2 times a month _ e. Never or nearly never _ e. Never or nearly never Vol. 16 | Issue 02 | June 2015 34 ANC Score, STOP-BANG, and Berlin questionnaire Category 2: High Risk: if there are >2 Categories where the Item 6: if ‘a’ or ‘b’ is the response, assign 1 point score is positive Item 7: if ‘a’ or ‘b’ is the response, assign 1 point Low Risk: if there is 0-1 Category where the Item 8: if ‘a’ is the response, assign 1 point score is positive Item 9 should be noted separately. Additional question: item 9 should be noted Add points. Category 2 is positive if the total separately score is 2 or more points Adapted from: Netzer NC, Stoohs RA, Netzer CM,Clark K, CATEGORY 3: Strohl KP. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea 10. Do you have high blood pressure? syndrome. Ann Intern Med 1999 Oct 5;131(7): _Yes 485-91. _ No _ Don’t know Category 3 is positive if the answer to item 10 is ‘Yes’ OR if the BMI of the patient is greater than 30kg/m2 Appendix 3 Adjusted Neck Circumference (ANC) Score Parameter Point allocation Neck Circumference (cm) size in cm = points Hypertension 4 Snoring 3 Choking or gasping during sleep 3 Probability of Obstructive Sleep Apnea If the ANC Score is: < 43 = low probability 43 - 48 = intermediate probability >48 = high probability 35 Phil J Chest Dis 2015 Albay AB Jr et al CROSS-SECTIONAL STUDY Prevalence of Obstructive Sleep Apnea Symptoms and Excessive Daytime Sleepiness among Commercial Long Distance Bus Drivers in the PhilippinesAlbert B. Albay Jr., MD, FPCP, FPCCP; Manuel C. Jorge II, MD, FPCP, FPCCP; Ricardo M. Salonga, MD, FPCP, FPCCP ABSTRACT Introduction: More than one million people are killed worldwide every year as a result of road traffic crashes. It is a fact that rates are higher in low- and middle-income countries. The Land Transportation Office (LTO) of the Philippines in 2005 reported close to 26,000 commercial buses traversing our highways. Road traffic accidents, viewed erroneously as random events, are amenable to rational analysis. Research supports that obstructive sleep apnea (OSA) is a significant cause of road traffic accidents resulting in two- to sevenfold increased risk. Screening for OSA and excessive daytime sleepiness (EDS) is not part of the licensing requirements for our long-distance commercial bus drivers. The actual incidence of these conditions is not known. The responsibility of certifying long distance drivers is heavy on the physician as it directly affects public safety. Establishing the incidence of these conditions is of use in policy making. Objectives: This study aimed to determine the prevalence of OSA symptoms and EDS among commercial long distance bus drivers and their correlation with road accidents, driving history and performance. It also aimed to elicit risk factors associated with increased OSA symptoms and daytime sleepiness.Methodology: This was a cross-sectional study of legal adult drivers from Philippine commercial bus companies who are engaged in long-distance travel. The subjects completed a questionnaire containing details on health vices, sleep habits, driving history, OSA symptoms and the Epworth Sleepiness Scale. Statistical analyses were made through the latest version of STATA program.Results and Conclusions: The profile of the Filipino long-distance bus driver was that of a middle- aged male who has at least reached secondary level of education, with a propensity to be overweight. This population is considered high risk for developing OSA and EDS. He was likely a current or previous smoker who engages in moderate caffeine intake daily. He often compensated his inadequate 4 hours of sleep during workdays with frequent naps. He has been involved in this occupation for about a decade. He has driven almost daily, with each trip lasting 8 to 11 hours and covering 361.64 ± 140.90 kilometers per trip. Only 35% had a replacement driver during these trips. The prevalence of OSA symptoms (85.6%), daytime sleepiness (52.5%), and combined OSA symptoms and daytime sleepiness (46.25%) were higher than in the general population. Bus drivers with the combined condition had increased risk of being involved in an accident by almost three-fold. Increasing the number of days of the driver on the roads increased his propensity to encounter accidents.Vol. 16 | Issue 02 | June 2015 36 OSA and daytime sleepiness in bus drivers Recommendations: We recommend review of the existing LTO certifications for long-distance commercial bus drivers. We suggest that a simple questionnaire be included as a screening for those who potentially have OSA and EDS. A review of existing hours and days that bus drivers have to be engaged in this activity must be done, and regulations must be placed. Bus companies should adopt a healthy lifestyle program for their drivers that include smoking cessation, avoidance of vices, weight management, and sleep hygiene. BACKGROUND More than one million people are killed driving in monotonous and unfamiliar roads. worldwide every year as a result of road traffic Several investigations have found fatigue to be crashes. The World Health Organization reported especially frequent among commercial drivers.3-6 in 2004 that road traffic injuries are the 11th Medical research supports the finding that leading cause of death and the 9th leading cause of obstructive sleep apnea (OSA) is a significant disability-adjusted life years worldwide.1 cause of motor vehicle crashes resulting in 2- to In the Philippines, the actual number of road 7-fold increased risk.7-11 Recent reports indicate traffic accidents is unknown. However, the rates that OSA is present in a greater prevalence in are generally higher in low- and middle-income operators of commercial motor vehicle drivers countries than in high-income nations. Altogether, than in the general population.12 low-income and middle-income countries Many high-income countries have laws accounted for 90% of all road traffic deaths in restricting the number of hours commercial 2002.1 In 2005, the Land Transportation Office drivers can drive at a stretch. In September 2006, (LTO) of the Philippines reported close to 26,000 the American College of Chest Physicians, the commercial buses traversing our highways.2 This American College of Occupational and figure excludes the non-registered commercial Environmental Medicine and the National Sleep buses whose safety for public use is uncertain. Foundation published their statement regarding Historically, motor vehicle crashes have sleep apnea and commercial motor vehicle been viewed as random events that happen to operators.13 The task force recommends others and are an inevitable outcome of road screening to include an examination form asking transport. However, this is not the case. Road the commercial motor vehicle driver to respond traffic crashes are events amenable to rational to questions about sleep disorders, pauses in analysis and remedial action. Various studies have breathing while asleep, daytime sleepiness and found that the incidence of road crashes could be loud snoring. An inquiry whether the driver has reduced if people did not drive while feeling ever been told by any healthcare provider that he sleepy or after sleeping for less than 5 hours in the or she has sleep apnea or any sleep disorder preceding 24 hours. should also be made. The driver is also asked Studies in the United States identified the about their history of motor vehicle crashes and following risks for being involved in crashes: assessed if these were related to sleepiness or young drivers aged 16-29; male gender; shift sleep related disorder. workers who work at night or with long irregular The task force likewise recommended working hours; people with untreated sleep apnea screening for hypertension, obesity and increased or narcolepsy; driving long distances (more than 2 neck circumference. Inappropriate sleepiness hours of straight driving); driving after consuming during normal waking hours should be alcohol, in extreme weather, or during hours when determined through a general question about normally asleep; fatigue; poor quality sleep; and wake time alertness as well as by asking about 37 Phil J Chest Dis 2015 Albay AB Jr et al excessive sleepiness during activities that are in assessing daytime sleepiness. It distinguishes usually associated with vigilance. A sleep study is feeling asleep from just feeling tired. The also recommended as part of the screening tool. advantages of the ESS, aside from it being The task force recognized that further validated, is its reliability in assessing subjective research is needed in several areas of sleep apnea sleepiness and its ability to distinguish normal and commercial vehicle driving. Among these are patients from patients with sleep disorders. It is correlating screening tests with functional the most useful tool in assessing daytime outcomes including its use in predicting crashes, sleepiness in clinical practice because it relates to incorporating ethnic, gender and age-related common life situations. ESS is a standard, low- factors into screening and prediction models, as cost, quick, and easy-to-apply scale. It has been well as assessing factors related to continuous validated comparing to the gold standard of EDS, positive airway pressure (CPAP) use, that predict which is Multiple Sleep Latency Test (MST),15 It reduction in crashes. 13 has a high internal consistency, measured by the Screening for OSA symptoms and excessive Cronbach alpha index (0.88).16 daytime sleepiness (EDS) is not part of the The ESS has been translated into different licensing requirements for our long-distance languages (Italian, Chinese, Spanish and Greek) commercial bus drivers. The actual incidence of and has been found to have high construct these conditions is not known. Applying the Joint validity, test-to-test reliability and internal Task Force Recommendation in the Philippine consistency.17-20 In 2006, a Filipino translation setting will not be cost-effective as sleep studies was tested and found to have similar validity, here are expensive and available only in major reliability and applicability to the Philippine cities. The responsibility of certifying drivers as fit setting.21 to engage in long-distance driving is heavy on the This study aimed to determine the medical practitioner as its consequence directly prevalence of OSA symptoms and EDS among affects public safety. Hence, establishing the commercial long-distance bus drivers and their incidence of OSA symptoms and EDS is of use in correlation with road accidents, driving history policy making as well as in screening which and performance. It also aimed to elicit risk subset of patients need to undergo expensive sleep factors associated with increased OSA symptoms diagnostics. and daytime sleepiness. The Epworth sleepiness scale (ESS) was first published in 1991 and was designed to METHODOLOGY measure subjective sleep propensity as it occurs in This was a cross-sectional study involving ordinary life situations.14 It is currently the most licensed professional adult drivers from <a href="/tags/Manila/" rel="tag">Manila</a>- utilized subjective test of daytime sleepiness in based Philippine commercial bus companies clinical practice. The questionnaire describes eight engaged in long-distance travel. The presence of situations and the participant scores the likelihood illness explicitly defined by law as rendering a of inducing sleep in each situation on a scale of 0 person incapable of driving a motor vehicle; the to 3. A score of less than or equal to 8 indicates presence of neurologic or psychiatric condition normal sleep function; 8 to 10 indicates mild impeding participation in the study; and lack of sleepiness; 11 to 15 indicates moderate sleepiness; consent including due to inability to understand 16 to 20 indicates severe sleepiness; and a score of the questionnaires were exclusion criteria. 21 to 24 indicates excessive sleepiness. A score of All commercial bus driving companies 10 is defined as the upper limit of normal, while a located with stations in Quezon City were score greater than 10 signifies EDS. sampled. Convenient sampling of available bus The ESS is most useful in clinical practice drivers was done on several days of the week. Vol. 16 | Issue 02 | June 2015 38 OSA and daytime sleepiness in bus drivers Based on the incidence of EDS in the of questions pertaining to symptoms of sleep general population, our computed sample size was apnea and a Filipino version of the Epworth 156. We sent out 189 questionnaires to account for Sleepiness Scale. No assessment of sleep a possible rejection rate of 20%. adequacy was performed. After an informed consent is obtained, three Qualitative variables are expressed as sets of self-administered questionnaires were percentages and quantitative variables as means administered: (1) a general questionnaire and standard deviations. A p-value of 0.05 or less regarding personal habits, common co- and a 95% confidence interval was considered to morbidities, and use of medications; (2) a indicate statistical significance. Percentages were questionnaire containing details of traffic compared with the Chi-square test whenever accidents focused on the presence or absence of necessary. We computed for odds ratio on each drowsiness at the time of accident and other variable of interest. potential risk factors like sleep deprivation, fatigue All 160 bus drivers were males, with a and alcohol intake will be given; and (3) final set mean age of 42.9 years (Table 2). Majority was able to reach secondary level of education (73.53%). The average body mass index (BMI) Table 1. Breakdown of participants by bus company was 24.2 kg/m2. Almost half of the participants Drivers who were overweight by World Health Organization Company participated standards. (N=160) Majority of the drivers who participated Victory Liner 47 (29.4%) were non-smokers (46.5%), mild to moderate alcohol drinkers (48.4%), and mild to moderate Baliwag CBL 16 (10.0%) caffeine drinkers (49.4%). The most common co- Raymond Transportation 13 (8.1%) morbidity seen in the sampled population was hypertension. A small number of respondents JAM Liner 12 (7.5%) admitted to a history of illicit drug use (9.4%). Five Star 12 (7.5%) Review of the sleeping hygiene of respondents showed that majority slept in a room Dagupan Bus Company 11 (6.9%) with other members of the family (Table 3). Almost half slept on their side, followed by a Partas Bus Inc 10 (6.3%) third of them sleeping in the supine position. Use Ceres Tours 9 (5.6%) of television as a sleeping aid was prominent in 48% of the respondents. Only a small number of ES Transport 8 (5%) respondents have significant awakenings (11%). Genesis 6 (3.8%) However, almost half of the bus drivers who participated in the study expressed difficulty in RCG Transport 6 (3.8%) returning to sleep upon being abruptly awakened. JAC Liner 4 (2.5%) More than half of the respondents have inadequate sleep (less than 4 hours) during Isarog Bus 2 (1.3%) weekdays. Majority of them are able to obtain Penafrancia Tours 2 (1.3%) adequate sleeping hours (4 to 8 hours) during weekends or holidays. A small number (15%) Saulog Transit 1 (0.6%) admited to poor quality of sleep exhibited by feeling of tiredness despite several hours of sleep Buena Transit 1 (0.6%) and rest. Almost half of the respondents were 39 Phil J Chest Dis 2015 Albay AB Jr et al Table 2. Demographic and clinical characteristics Table 3. Sleeping habits of responders (N=160) of respondents (N=160) Sleeping habit Prevalence (%) Characteristics Percentage Sleeping partner Sex Alone 43.80% Male 100% Wife or kids 55.60% Female Pet 0.60% Level of education Primary 7.35% Secondary 73.53% Sleeping position Tertiary 11.76% Supine 27% Vocational 7.35% Side 45.30% BMI Prone 6.90% Mean (kg/m2) 24.2 Turner 17.60% Unaware 3.10% Underweight (BMI <18) 7% Normal (BMI 18-22.9) 24% Use of sleeping aids Overweight (BMI 23-24.9) 49% Television 48.30% Obese I (BMI 25-29.9) 17% Reading material 17.90% Obese II (BMI >30) 3% Radio 25.50% Smoking History Liquor 6.90% Non-smoker 46.50% None 1.30% Previous smoker 27.00% Current smoker 26.40% Multiple awakenings during 11.25% Co-morbidities sleep Hypertension 8 Difficulty returning to sleep 45.60% COPD 4 Tuberculosis 1 Hours of sleep on weekdays Bronchial asthma 1 Inadequate 56.20% Alcohol consumption Adequate 40% None 36% More than adequate 3.80% Mild to Moderate 49.40% Heavy 13.90% Hours of sleep on weekends Caffeine intake Inadequate 23.80% None 25.60% Adequate 51.20% Mild to Moderate 49.40% More than adequate 25% Heavy intake 25% Feeling of tiredness upon History of Illicit drug use 9.40% 15% waking up BMI, body mass index; COPD, chronic obstructive pulmonary disease. Able to take naps 43.13% able to take naps in between long driving trips. events. Five percent of these accidents were The average long-distance bus driver has obtained after prolonged hours of non-stop been engaged in this occupation for 10.4 years driving (more than 2 hours). More than half of (Table 4). On review, most of them engaged in the respondents admitted to be sleeping during long-distance driving almost daily, with each trip these accidents. Half of the respondents have lasting 8 to 11 hours, and covering 361.64 ± likewise been involved in near-traffic accidents. 140.90 kilometers. Only a third of these drivers As a result of poor traffic conditions in the had a replacement in these long trips, with at least Metro Manila, most of these bus companies one stopover during the entire trip. scheduled long-distance travels during less On inquiry into traffic related accidents, congested hours. These were usually late at night 52% admitted to have been involved in such and early in the morning. Bus drivers then had to Vol. 16 | Issue 02 | June 2015 40 OSA and daytime sleepiness in bus drivers Table 4. Occupational characteristics of adjust not only to prolonged hours of driving but responders (N=160) to also a shift in work schedule. This can be Occupational characteristic Result likened to alteration in sleep patterns among shift workers. This could be an added factor for the Years of Driving 15.7 years increased incidence of sleep related accidents in Years engaged in long distance this population. 10.4 years bus driving Our investigation showed that among Average number of days commercial long distance bus drivers, the engaged in long distance 6.4 days prevalence of OSA symptoms (85.6%), daytime driving sleepiness (52.5%), and combined OSA Number of hours per trip 8.04 + 3.05 hours symptoms and daytime sleepiness (46.25%) were higher than in the general population. Statistical With a replacement driver 35% analysis showed that the increased number of during trips days of driving was associated with a 1.48-fold Average Distance traveled per 361.64 + 140.90 km increased risk for the driver to be involved in an trip accident (Table 5). Likewise, the presence of Average number of Rest OSA, EDS and the combined condition greatly 1.7 periods during the trip increases the risk of having an accident by 2.9, Involved in Traffic accidents 52% 2.08, and 2.8 times, respectively. Further analysis showed that the type of % of accidents occurring after 2 5% accident that these bus drivers with combined hours OSA symptoms and EDS is likely to occur within Sleeping related accidents 54% 2 hours and are greatly sleep-related. Involvement in Near Traffic These numbers are alarming and are 58% Accidents beyond the levels noted in the general population. Table 5. Risk factors of accidents among long-distance bus drivers Risk Factor Odds Ratio Confidence Interval P-value Years of driving 0.970 0.926-1.016 0.199 Years of long-distance driving 0.936 0.886-0.988 0.016 Days of driving 1.477 1.140-1.912 0.003 Hours of driving 0.910 0.817-1.014 0.088 Replacement 0.560 0.288-1.092 0.089 Distance traveled 1.002 0.9995-1.004 0.121 Number of stopovers 1.053 0.647-1.713 0.835 OSA symptoms 2.900 1.119- 7.490 0.028 EDS 2.077 1.102-3.912 0.024 Combined EDS and OSA 2.798 1.465-5.344 0.002 OSA, obstructive sleep apnea; EDS, excessive daytime sleepiness. 41 Phil J Chest Dis 2015 Albay AB Jr et al Table 6. Correlation of combined OSA and EDS with accidents Event Odds Ratio Confidence Interval P-value Accidents 2.798 1.465-5.344 0.002 Within 2 hours 3.484 1.801-6.738 <0.001 Sleeping related accidents 3.307 1.700-6.435 <0.001 Near Accidents 2.386 1.230-4.629 0.01 OSA, obstructive sleep apnea; EDS, excessive daytime sleepiness. Long-distance bus drivers are high risk for RECOMMENDATIONS OSA and EDS. These two conditions are We recommend review of the existing LTO potentially treatable and can be modified. certifications for long distance commercial bus Subjecting these high-risk individuals to further drivers. We suggest that the simple questionnaire testing and adequate management may limit the be included as a screening for those who are risk of accidents that these two conditions carry. potentially with OSA and EDS. We recommend that drivers found to be at high risk should be CONCLUSION tested further, preferably with diagnostic sleep The profile of the Filipino long-distance studies, and be given appropriate treatment prior bus driver was a middle-aged male who has at to sending them out to our roads to drive long least reached secondary level of education, with a distances. We hope that this study will help us propensity to be overweight. This population is rehabilitate these high-risk individuals for them considered high risk for developing OSA and to be more productive and safe in their work. We EDS. The typical bus driver was likely a current likewise recommend that bus companies adopt a or previous smoker who engaged in moderate healthy lifestyle program for their drivers that caffeine intake daily. He often compensated for include smoking cessation, avoidance of vices, his inadequate four hours of sleep during weight management, and especially good sleep workdays with frequent naps. He had been hygiene. Likewise, we recommend review of involved in this occupation for about a decade. existing hours and days that bus drivers have to He has driven almost daily, with each trip lasting be engaged in this activity. Just as regulations are 8 to 11 hours, and covering 361.64 ± 140.90 placed in number of hours for work in offices, kilometers per trip. Only 35% had a replacement bus drivers ought to be given physiologic number driver during the trips. of hours to rest and engage in their job. The prevalence of OSA symptoms In terms of study conduct, future similar (85.6%), daytime sleepiness (52.5%), and studies may further quantify the temporal combined OSA symptoms and daytime sleepiness correlation of accidents and OSA/EDS to (46.25%) are higher than the general population. eliminate temporal and recall biases. It may also Analysis of our data showed that the bus driver explore the burden and interaction of co-morbid with the combined condition increases his risk of conditions such as hypertension to OSA/EDS in being involved in an accident by almost three- this subpopulation. fold. Increasing the number of days of the driver on the roads increases his propensity to encounter accidents. Vol. 16 | Issue 02 | June 2015 42 OSA and daytime sleepiness in bus drivers REFERENCES: disordered breathing and motor vehicle 1. World report on road traffic injury and accidents in a population based sample of prevention: summary. Geneva: World Health employed adults. Sleep 1997;20:608-613. Organization; 2004. 12. Moreno CR, Carvalho FA, Lorenzi C, et al. 2. Land Transportation Office of the Philippines High risk for obstructive sleep apnea in truck Annual Report 2005. drivers estimated by the Berlin Questionnaire: 3. Connor J, Norton R, Ameratunga S, Robinson prevalence and associated risk factors. E, Civil I, Dunn R, Bailey J, Jackson R. Chronobiol Int 2004;21:871-879. Driver sleepiness and risk of serious injury to 13. Statement from the joint task force of the car occupants: population based case control American College of Chest Physicians, the study. BMJ. 2002 May 11;324(7346):1125. American College of Occupational and 4. Drowsy driving and automobile crashes. Environmental Medicine and the National Sleep Washington, DC, National Center on Sleep Foundation. J Occupational Environmental Med Disorders Research/National Highway Traffic Suppl 2006:48;9. Safety Administration Expert Panel on Driver 14. Johns M. A new method of measuring daytime Fatigue and sleepiness. 1996. Available at: sleepiness: the Epworth Sleepiness Scale. Sleep www.nhtsa.dot.gov/people/injury/ 1991;14:540. drowsy_driving1/drowsy.html. 15. Johns MW. Sleepiness in different situations 5. Hartley LR, Penna F, Corry A, Feyer AM. measured by the Epworth Sleepiness Scale. Comprehensive review of fatigue research. Sleep 1994;17:703. Fremantle, Murdoch University, Institute for 16. Johns MW. Reliability and factor analysis of the Research in Safety and Transport, 1996. Epworth sleepiness scale. Sleep 1992;15(4): 6. Mock C, Amegeshi J, Darteh K. Role of 376-81. commercial drivers in motor vehicle related 17. Chiner E, Arriero JM, Signes-Costa J, Marco J, injuries in Ghana. Injury Prevention Fuentes I. Validation of the Spanish version of 1999;5:268–271. the Epworth Sleepiness Scale in patients with a 7. Howard ME, Desai AV, Grunstein RR, et al. sleep apnea syndrome. Arch Bronconeumol Sleepiness, sleep-disordered breathing and 1999;35(9):422-7. accident risk factors in commercial vehicle 18. Chung KF. Use of the Epworth Sleepiness Scale drivers. Am J Respir Crit Care Med in Chinese patients with obstructive sleep apnea 2004;170:1014-1021. and normal hospital employees. J Psychosom 8. Shiomi T, Arita AT, Sasanabe R, et al. Falling Res 2000;49(5):367-72. asleep while driving and automobile accidents 19. Vignatelli L, Plazzi G, Barbato A, Ferini- among patients with obstructive sleep apnea- Strambi L, Manni R, Pompei F, D'Alessandro R; hypopnea syndrome. Psychiatry Clin Neurosci GINSEN (Gruppo Italiano Narcolessia Studio 2002;56:333-334. Epidemiologico Nazionale. Italian version of 9. Horstmann S, Hess CW, Bassetti C, Gugger the Epworth sleepiness scale: external validity. M, Mathis J. Sleepiness related accidents in Neurol Sci 2003;23(6):295-300. sleep apnea patients. Sleep 2000;23:383-389. 20. Tsara V, Serasli E, Amfilochiou A, 10. Teran-Santos J, Jimenez-Gomez A, Cordero- Constantinidis T, Christaki P. Greek version of Guevarra J. The association between sleep the Epworth Sleepiness Scale. Sleep Breath apnea and the risk of traffic accidents. 2004;8:91-5. Cooperative Group Burgos-Santander. N Engl 21. Albay, A, Sison CM, Jorge M. Validation of the J Med 1999;340:847-851. Filipino version of the Epworth Sleepiness 11. Young T, Bluestein J, Finn L, Palta M. Sleep Scale. Chest Suppl 2007;4:132. 43 Phil J Chest Dis 2015 Joson-Vergara ISD et al PRACTICE GUIDE Decoding the Polysomnogram (PSG) Irene Salve D. Joson-Vergara, MD, DPCCP; Manuel Peter Paul C. Jorge, MD, FPCCP INTRODUCTION Stages N1 and N2 are called light sleep, Polysomnography is a sleep study that is an and stage N3 is called deep or slow-wave sleep. essential tool for diagnosing sleep disorders, the Each sleep cycle usually last from 90 to 120 result of which is called a polysomnogram (PSG). minutes. It is initially composed of a segment of It involves a comprehensive measurement and NREM sleep followed by REM sleep.3 As the recording of various biophysiological events that night progresses, NREM shortens while REM occur during sleep.1,2 Interpreting the PSG is a lengthens. daunting task because it entails meticulous During the PSG, the subject may also observation and keen attention for detail. awaken. Wakefulness is denoted as Stage W. Likewise, it requires knowledge not only of the interpretation of various test such as PARAMETERS IN THE PSG electroencephalogram (EEG) and electromyogram The standard PSG also includes the (EMG) but also correlation of specific changes or monitoring and evaluation of arousals, abnormalities in each parameter that points to a respiration, limb movements, snoring, oximetry, specific disease.3 body position and cardiac rhythm disturbance.3-5 It is the task of accredited sleep specialists Each parameter is described on the vertical axis to interpret the PSG. However, as pulmonologists, of the graph while the time of recording is it is our duty to recognize sleep disorders, refer to reflected on the horizontal axis (Figure 1). PSG is a sleep specialist, and understand the traditionally recorded and scored in 30 second interpretation of the PSG, which can still be “epochs”.4 daunting to the pulmonologist. It is the goal of this article to define the parameters (often presented in EEG codes) that are shown in the PSG as well as The EEG (Figure 2) is the primary variable describe the different waveforms that are seen in that records wakefulness, arousals and sleep the report. Hopefully, this may aid in stages. It is labelled according to the placement understanding PSG. This article is not meant in of the electrodes. The primary channels are as any way to provide a guideline in the follows: interpretation/scoring of the test. • Central channel referenced to an ear mastoid site (C4-M1); STAGES OF SLEEP • Frontal channel referenced to an ear mastoid Sleep is characterized by different stages: site (F4-M1); and. non-rapid eye movement (NREM) sleep, which is • Occipital channel referenced to a mastoid site further divided into stages N1, N2 and N3; and, (O2-M1). rapid eye movement (REM) sleep (stage R). These These are the recommended derivations; stages are determined by the electrical brain wave however, additional leads may be used to activity as recorded by the EEG; eye movements supplant in case of malfunction. The electrical which are recorded by the electrooculogram activity of the brain is reflected in the EEG as (EOG); and the electrical activity of the different waves characterized by the frequency in cycles muscles in the body as recorded by the EMG. per second or hertz (Hz), amplitude (voltage), Vol. 16 | Issue 02 | June 2015 44 Decoding the PSG Figure 1. The polysomnogram and its parameters EEG, electroencephalogram, EOG, electro-occulogram; EMG, electromyogram. and the direction of major deflection. The waves a positive component lasting at least 0.5 seconds. are typically classified as delta (<4 Hz), theta (4 K-complexes appear similar to vertex sharp to 7 Hz), alpha (8 to 13 Hz) and beta (>13 Hz). waves, but with longer duration. Sleep spindles These waves can be classified by counting the are waves with frequencies between 11 to 16 Hz, number of peaks per second of recording.4, 5 with duration of more than 0.5 seconds. Alpha waves resemble blades of grass and Stage N3 consists of delta waves (Figure are commonly seen when the patient is in an 6), which are high amplitude waves with a awake, but relaxed, state with the eyes closed frequency of oscillation between 0 to 4 Hz. (Figure 3). Alpha waves can also be seen during Stage R is characterized by LVMF theta REM sleep and arousals. waves. Some alpha waves may be present; During stage N1, theta waves (also known however, they appear slower than wake alpha. as low voltage mixed frequency [LVMF] waves) Sharply contoured serrated waves called saw are found (Figure 4). Near the transition from tooth waves are also evident during this stage, Stage N1 to stage N2, high-amplitude negative usually preceding a burst of rapid eye waves (upward deflection) followed by a positive movements seen in the EOG leads.4, 5 component (upward deflection) with a short duration (less than 0.5 seconds) occur and these EOG are called vertex sharp waves. The EOG monitors eye movements and During N2 stage, combination of theta thus facilitate identification of sleep onset by waves interspersed with sleep spindles and or K- detecting slow rolling eye movements that occur complexes are seen (Figure 5). K-complexes are during transition to N1, and REM that are well delineated negative sharp wave followed by characteristic of stage R (Figure 7). 4 At least two 45 Phil J Chest Dis 2015 Joson-Vergara ISD et al Figure 2. Electroencephalogram in the polysomnogram Count the number of peaks within a 1-second interval Figure 3. Alpha waves (8 to 13 Hz) Figure 5. K-complex and sleep spindle in N2 Photograph by Hugo Gamboa under a CC BY-SA 3.0 license. Signal noise has been filtered to show only the wave form. Figure 4. Theta waves (4 to 7 Hz) channels of EOG are recommended. Each EOG channel records from an electrode placed 1 cm above or below the outer canthus of the eye. Additional EOG combinations can be applied to distinguish between vertical and lateral eye Photograph by Hugo Gamboa under a CC BY-SA 3.0 movement. 4 license. Signal noise has been filtered to show only the wave form. EMG The EMGs are placed in the chin and both Figure 6. Delta waves (<4 Hz) legs. In the chin area, it is used for determining the level of muscle tone. Chin tone decreases during sleep onset. It is lowest during REM.3 Three electrodes are used to record chin EMG. One mental electrode placed in the midline 1 cm above the inferior edge of the mandible, and two Photograph by Hugo Gamboa under a CC BY-SA 3.0 submental electrodes placed 2 cm below the license. Signal noise has been filtered to show only the inferior edge and 2 cm to the left and to the right wave form. of the midline. 4 Vol. 16 | Issue 02 | June 2015 46 Decoding the PSG Figure 7. Electrooculogram indicating rapid eye during sleep. A single ECG channel is sufficient movement (encircled) for standard PSG monitoring. The two standard ECG electrodes are applied in a lead II format.4 Upper Airway Sound Recording Upper airway sound recording is done with the use of a snore sensor or microphone placed over the trachea or on the side of the neck. This is done to detect snore bursts that may aid in determining and verifying the nature of arousals. Respiration is measured by airflow and respiratory effort channels. These are used to monitor respiration particularly for the detection of apneas, hypopneas, respiratory effort-related EMG in the legs are used to detect other causes arousal (RERA) and other sleep related breathing of sleep disturbances such as periodic limb events. At least three respiratory parameters are movements of sleep (PLMS) (Figure 8). These recorded namely: nasal/oral airflow, thoracic movements are often visually detectable during effort and abdominal effort.4 the monitoring process; however, monitoring the Various transducers are used to document anterior tibialis muscles can establish the degree airflow. A thermal sensor is used in the detection of severity of the disorder by quantifying the rate of apnea while pressure transducers are used for of movements as well as the correlation with both hypopnea detection and RERA detection.4,5 EEG arousal. This is done by placing two Respiratory effort is monitored by using electrodes longitudinally on the anterior tibialis intercostal or diaphragmatic EMG electrodes, muscle of each leg 2 to 3 cm apart. Limb EMG of esophageal pressure monitoring or calibrated or the upper extremities may also be recorded if uncalibrated inductance plethysmography that clinically indicated.3-5 allows differentiation between abdominal and thoracic movement. Although the most accurate ECG measure of respiratory effort is esophageal The ECG monitors heart rhythm as well as pressure manometry, correct placement of the cardiovascular dysfunctions that may occur sensor is tedious and can cause patient discomfort Figure 8. Electromyogram of the legs in polysomnography 47 Phil J Chest Dis 2015 Joson-Vergara ISD et al and sleep disturbance. Hence, the recommended scoring. sensors are thoracoabdominal inductance A hypnogram, which is a graphical repre- plethysmography belts.4,5 sentation of the organization of sleep during the night is also provided. The various stages of sleep Oxygen saturation are seen in the vertical axis and the time of the The blood oxygen saturation (SpO2) levels night is seen on the horizontal axis (Figure 9). during the entire duration of the PSG provides crucial information about the severity of the sleep SLEEP CENTERS related breathing disorder. Pulse oximetry Sleep Centers in the Philippines are measurement is usually performed using a finger concentrated mainly in tertiary hospitals in Metro probe, although other placements such as the Manila. However, cities such as <a href="/tags/Cebu/" rel="tag">Cebu</a> City, earlobe or toe may be used depending on the Bacolod City, Iloilo City and Davao City also patient.4,5 have Sleep centers. Body position REFERENCES: The body position is also monitored on a 1. Caraskadon MA, Rechschaffen A. continuous basis. This is essential because a lot Monitoring and staging human sleep. In: of sleep disorders can be exacerbated by body Kryger MH, Roth T, Dement WC, eds. orientation during sleep. Body position can be Principles and Practice of Sleep Medicine. monitored with various commercially available Philadelphia, PA: WB Saunders, 2000:1197– body position monitoring devices or by direct 1215. observation by the sleep technologist.5 2. Somers VK, Javaheri S. Cardiavascular effects of sleep-related breathing disorders. PSG Report In: Kryger MH, Roth T, Dement WC, eds. The standard PSG consist of a full night Principles and practices of sleep medicine, test. If a sleep breathing disorder is diagnosed, 4th ed. Philadelphia: Elsevier Saunders; the patient comes back a second night for 2005:1182-1186. continuous positive airway pressure (CPAP) 3. Geyer JD, Talathi S, Carney PR. Introduction titration. However, if there is a high index of to sleep and polysomnography. Chapter 11. suspicion of a sleep breathing disorder, a split- Available at: http://com-peds- night study may be done. The first four hours neurodynamicslab.sites.medinfo.ufl.edu/files/ consist of the standard PSG followed in the 2012/02/Chap11.pdf second half of the night by CPAP titration. 4. American Association Sleep Technologist: After the test is done, the sleep specialist sleep technology: technical guideline. Darien, describes the sleep architecture and scores each IL: 2012. epoch according to sleep stages. Respiratory 5. Iber C, Ancoli-Israel S, Chesson A, et al. The events are enumerated and the Apnea Hypopnea AASM Manual for the Scoring of Sleep and Index (AHI) is given. An AHI of greater than 5 Associated Events: Rules, Terminology and per hour denotes a sleep breathing disorder. Technical Specifications. 1st Ed. Westchester, Periodic limb movements, snoring, oxygen IL: American Academy of Sleep Medicine, saturations and cardiac rhythms are described.5 2007. Recommended CPAP values are stated if it was determined. Table 1 presents a summary of the different parameters described in each part of Vol. 16 | Issue 02 | June 2015 48 Decoding the PSG Table. Scoring the polysomnogram I. Sleep Stage Scoring Sleep Stages Sleep Architecture · Stage W (Wakefulness) · Lights out clock time · Stage N1 (NREM 1) · Lights on clock time · Stage N2 (NREM 2) · Total sleep time (TST; stages 1,2,3 and R in mins) · Stage N3 (NREM 3) · Total recording time (lights out to lights on in min.) · Stage R (REM) · Sleep Latency (SLl; lights out to first epoch of any sleep in min.) · Stage R latency (sleep onset to first epoch of stage R in min.) · Wake after sleep onset (time awake after sleep onset up to lights on) · Percent sleep efficiency (TSTx100/TRT) · Time in each stage · Percent TST in each stage II. Respiratory Events Scoring Apnea - drop in the peak thermal sensor or airflow sensor excursion by >90% of baseline - duration of the event lasts at least 10 seconds. - at least 90% of the event’s duration meets the amplitude reduction criteria for apnea. · Obstructive Apnea - associated with continued or increased respiratory effort throughout the entire period of absent airflow · Central Apnea - associated with absent inspiratory effort throughout the entire period of absent airflow · Mixed apnea - associated with absent inspiratory effort in the initial portion of the event, followed by resumption of inspiratory effort in the second portion of the event Hypopnea - nasal pressure signal excursions drop by >30% or >50% of baseline. - lasts at least 10 seconds - > 4% desaturation from pre-event baseline (if with 30% drop in nasal pressure signal) or >3% desaturation from pre-event baseline or the event is associated with arousal (if with 30% drop in nasal pressure signal) - at least 90% of the event’s duration must meet the amplitude reduction of criteria for hypopnea Respiratory Effort-Related Arousal (RERA) - a sequence of breaths lasting at least 10 seconds characterized by increasing respiratory effort or flattening of the nasal pressure waveform leading to an arousal from sleep when the sequence of breaths does not meet criteria for an apnea or hypopnea. · Continuous O2 saturation, mean · O2 saturation, minimum · Cheyne-Stokes Breathing - at least three consecutive cycles of cyclical crescendo and decrescendo change in breathing amplitude AND at least ONE of the following: (1) 5 or more central apneas OR [central] hypopneas per hour of sleep (2) Cyclic crescendo and decrescendo change in breathing amplitude of at least 10 consecutive minutes. III. Arousal scoring · Number of arousals - abrupt shift of EEG frequency including alpha, theta, and/or frequencies > 16 Hz (but not spindles) that lasts at least 3 seconds, with at least 10 seconds of stable sleep preceding the change. · Arousal index (ArI) IV. Periodic limb movement in sleep (PLMS) scoring Significant leg movement event: - Minimum duration is 0.5 seconds - Maximum duration is 10 seconds - Minimum amplitude is an 8 µV increase in EMG voltage above resting EMG - Timing of the onset is defined as the point at which there is an 8 µV increase in EMG voltage above resting EMG. - Timing of the ending of a LM event is defined as the start of a period lasting at least 0.5 sec during which the EMG does not exceed 2 µV above resting EMG. 49 Phil J Chest Dis 2015 Joson-Vergara ISD et al Figure 9. Hypnogram indicating the stages of sleep over time Image by Tash510 under a CC BY-SA 3.0 license. Vol. 16 | Issue 02 | June 2015 50 Las rinde el sueño (Sleep overcomes them) by Francisco Jose de Goya (1797) Etching and aquatint on laid paper (21.9 × 15.4 cm). This is a faithful photo- graphic reproduction of a two-dimensional, public domain work of art. The work of art itself is in the public domain. 51 Phil J Chest Dis 2015 OUTSIDE BACK COVER The Philippine Journal of Chest Diseases An official publication of: Philippine College of Chest Physicians 84-A Malakas St., Pinyahan, Quezon City, Philippines Email: secretariat@philchest.org Phone: (+632) 924 9204 </div> </article> </div> </div> </div> <script src="https://cdnjs.cloudflare.com/ajax/libs/jquery/3.6.1/jquery.min.js" crossorigin="anonymous" referrerpolicy="no-referrer"></script> <script> var docId = '8da8d27bd88dcb1a99a6b51aefbd728b'; var endPage = 1; var totalPage = 60; var pfLoading = false; window.addEventListener('scroll', function () { if (pfLoading) return; var $now = $('.article-imgview .pf').eq(endPage - 1); if (document.documentElement.scrollTop + $(window).height() > $now.offset().top) { pfLoading = true; endPage++; if (endPage > totalPage) return; var imgEle = new Image(); var imgsrc = "//data.docslib.org/img/8da8d27bd88dcb1a99a6b51aefbd728b-" + endPage + (endPage > 3 ? ".jpg" : ".webp"); imgEle.src = imgsrc; var $imgLoad = $('<div class="pf" id="pf' + endPage + '"><img src="/loading.gif"></div>'); $('.article-imgview').append($imgLoad); imgEle.addEventListener('load', function () { $imgLoad.find('img').attr('src', imgsrc); pfLoading = false }); if (endPage < 5) { adcall('pf' + endPage); } } }, { passive: true }); if (totalPage > 0) adcall('pf1'); </script> <script> var sc_project = 11552861; var sc_invisible = 1; var sc_security = "b956b151"; </script> <script src="https://www.statcounter.com/counter/counter.js" async></script> </html><script data-cfasync="false" src="/cdn-cgi/scripts/5c5dd728/cloudflare-static/email-decode.min.js"></script>