The Usefulness of Sleep Apnea Syndrome Screening using a Portable Pulse Oximeter in the Workplace

Transcription

1 J Occup Health 2007; 49: 1 8 Journal of Occupational Health The Usefulness of Sleep Apnea Syndrome Screening using a Portable Pulse Oximeter in the Workplace Kuniyuki NIIJIMA 1, Kazuhiko ENTA 1, Hiroko HORI 2, Shunsuke SASHIHARA 3, Tetsuya MIZOUE 4 and Yasuo MORIMOTO 5 1 Shizuoka Health Care Center, Central Japan Railway Company, 2 Department of Psychiatry, School of Medicine, University of Occupational and Environmental Health, Japan, 3 Health Care Center, Central Japan Railway Company, 4 Department of Preventive Medicine, Faculty of Medical Sciences, Kyushu University and 5 Department of Occupational Pneumology, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, Japan Abstract: The Usefulness of Sleep Apnea Syndrome Screening Using a Portable Pulse Oximeter in the Workplace: Kuniyuki NIIJIMA, et al. Shizuoka Health Care Center, Central Japan Railway Company Sleep Apnea Syndrome (SAS) is a condition characterized by sleep-disordered breathing resulting in health impairment and sleep problems. From the viewpoints of the prevention of health impairment, accidents at work and traffic accidents, active implementation of screening for SAS in workplaces is necessary. Using a portable pulse oximeter, we conducted SAS screening for workers, who applied for the screening or who were instructed to participate by occupational physicians based on their symptoms at the time of medical check-up, in order to evaluate the effectiveness of a portable pulse oximeter as a screening device in the workplace. During the 2- yr study period from April 2002 to March 2004, 380 workers underwent overnight pulse oximetry at home; data on 367 of them were considered valid and included in the analysis. We deemed screening-positive as oxygen desaturation of the peripheral artery of 4% or greater if 10 or more events were observed per hour (ODI4 10); or oxygen desaturation of the peripheral artery of 3% or greater if 15 or more events were observed per hour (ODI3 15). Eighty-three subjects were identified as screening-positive and 54 of them underwent polysomnography. All of them were diagnosed as having SAS with an apnea hypopnea index (AHI) 5, of which continuous positive airway pressure (CPAP) therapy was indicated in 48 cases Received Mar 6, 2006; Accepted Sep 8, 2006 Correspondence to: K. Niijima, Shizuoka Health Care Center, Central Japan Railway Company, 68 Kurogane-cho, Aoi-ku, Shizuoka city, Shizuoka , Japan ( niijima@jr-central.co.jp) (88.9%) with AHI 20. On the other hand, 11 of the 284 screening-negative subjects with mild sleep disordered breathing underwent polysomnography and all of them were diagnosed as having SAS with AHI 5. However, CPAP therapy was indicated only for 5 with AHI 20 of the 11. Therefore, in total, 65 subjects were diagnosed with SAS in this study and for 53 of them CPAP therapy was indicated. The simplicity of the SAS screening by pulse oximetry makes it easy to use for screening of workers, and this method was highly effective in detecting individuals with severe SAS for whom CPAP therapy was indicated. (J Occup Health 2007; 49: 1 8) Key words: Sleep Apnea Syndrome, Pulse oximeter, Screening, Health management, Occupational physician Sleep Apnea Syndrome (SAS) is a disorder characterized by 5 or more episodes of respiratory disturbance occurring in an hour due to temporary obstruction and stenosis of the airway. Symptoms of SAS are loud snoring, recurrent sleep interruption, excessive sleepiness and poor concentration during the day 1). These recurrent episodes of airway obstruction during sleep may cause nocturnal hypoxemia and subsequent arousals from sleep and have some health and sleep consequences. With regard to health effects, it has been suggested that sleep-disordered breathing (SDB) including habitual snoring and SAS is associated with the development of hypertension, ischemic heart disease, cerebrovascular disease and diabetes mellitus 2 4). Especially, the causal relationship between SAS and hypertension has been established, and SAS is also described as one of the identifiable causes of secondary hypertension 5). The 8-yr cumulative survival rate for patients with untreated severe SAS has been reported to

2 2 J Occup Health, Vol. 49, 2007 Table 1. Screening participants according to gender and reasons of participation Subjects Applicants Instructed by occupational physicians Total Male Female Total Applicants: Those who applied for the screening regardless of the presence or absence of any symptom after health education on SAS at each workplace. Instructed by occupational physicians: Those who agreed to participate in the screening after receiving an explanation from occupational physicians based on their symptoms of snoring and excessive daytime sleepiness and the results of a medical check-up. be 63% which is significantly lower than those with mild SAS 6). Sleep disorders due to SAS may cause excessive daytime sleepiness, and such patients are 700% more likely to be involved in traffic accidents due to drowsy driving; a figure which is even higher for those with severe SAS 7, 8). Prevalence of SAS in workers has been reported to be 4% in men and 2% in women 9). Therefore, it is of great significance to be able to make an early diagnosis and provide prompt treatment for SAS in workers, for worker health management and the prevention of accidents at work, including road traffic accidents. Although we can easily use questionnaires to screen for SAS, including the subjective symptom of sleepiness etc. in the workplace, the assessment provided based on the results of the self-administered items can be unreliable. On the other hand, testing for SDB using a pulse oximeter provides objective data and pulse oximetry has been reported to be a useful SAS screening tool 10 12). The pulse oximeter continuously monitors and records oxygen saturation of a peripheral artery (SpO 2 ) during sleep and it can detect a fall in SpO 2 which is generally caused by respiratory events. Thus, the presence or absence of irregular breathing events can be confirmed through this noninvasive monitoring. Recently a relatively compact, lightweight portable pulse oximeter has become widely used in a wide variety of clinical settings, in hospitals and the community. We conducted SAS screening of workers by using a portable pulse oximeter and examined the effectiveness of this screening device. Subjects We conducted SAS screening for workers in transport, construction, retail and security companies with prior permission from the health and safety committees of each workplace as part of health care services provided by occupational physicians. The screening was conducted for 380 subjects (373 men, 7 women) from April 2002 to March Of the subjects, 290 were interested applicants who applied for the screening regardless of the presence or absence of symptoms after health education on SAS at each workplace. The remaining 90 subjects who presented symptoms of snoring and excessive daytime sleepiness or had risk factors for SAS, such as obesity, agreed to participate in the screening after receiving an explanation from occupational physicians at the time of their medical check-up (Table 1). Because the changes in the hemoglobin quantity present obstacles for pulse oximetry 12), patients with severe anemia were not included. In addition, although the persons who had hypertension, diabetes mellitus, and arrythmia were included, there was no person who was not fit for work. Prior to the screening, the significance, method, freedom to drop out of the screening, and data handling etc. were fully explained to the subjects and informed consent was obtained from all subjects. Methods Pulse oximetry Oximetry was performed at night using a portable pulse oximeter (PULSOX-M24; Minolta, Osaka, Japan). Subjects were instructed on its usage and then given oximeters for unattended monitoring of SpO 2 and pulse rate during sleep at home. The pulse oximeter sensor was mounted on one fingertip and the pulse oximeter system was attached to the distal forearm. No specific limitations on daily life were imposed on the monitoring day; alcohol was allowed if quantities were moderate. SpO 2 and pulse rate were continuously measured and recorded every 5 s for 24 h. The pulse oximeters were collected and the recorded data stored in memory were input to a computer via a proprietary interface. The data was processed afterwards using appropriate software (DS- M Ver. 3.0; Minolta, Osaka, Japan). The pulse oximeter records pulse rate, SpO 2 values and other SAS related data such as the oxygen desaturation index (ODI: number of oxygen desaturation events per hour of measuring time), the lowest SpO 2 value, and CT 90 (the cumulative percentage of time spent at saturation below 90%). Falls in SpO 2 of 5% or greater per second were automatically

3 Kuniyuki NIIJIMA, et al. The Usefulness of SAS Screening by Oximetry in the Workplace 3 Fig. 1. Flow diagram of the screening. excluded because they could be affected by body movements. In addition to this, all cases were visually scrutinized and all periods of inadequate signal were manually excluded from the analysis. The effective measurement time was set to be more than 180 min. However, even if the valid measurement time was less than 180 min when inadequate data were excluded, reliable oxygen desaturation indexes were calculated when there were no inconclusive findings on the oximetry trend graph. Evaluation criteria (Fig. 1) The current study used ODI as screening diagnostic criteria for SAS. Subjects who had SDB events associated with 10 or more oxygen desaturation events of the peripheral artery of 4% or greater per hour (ODI4 10), or 15 or more oxygen desaturation events of the peripheral artery of 3% or greater per hour (ODI3 15) were defined as screening-positive and advised to visit a hospital. Among subjects who were defined as screening-negative, visiting a specialized medical institution was also advised to those who had ODI4 5 or ODI3 10, or those who fulfilled the criterion on CT 90 1%, which has been suggested as being indicative of SAS 13). All subjects were asked to answer a self-administered questionnaire at the time of taking the test. The Epworth Sleepiness Scale (ESS) 14), in which the subjects were asked to rate their daytime sleepiness, was included in the questionnaire. Those who scored more than 11 in total in the ESS were also advised to visit a hospital. Items in the questionnaire other than ESS were those related to the presence of snoring, apneas, urination during the night, arousals from sleep, morning headaches and excessive daytime sleepiness. Additional survey items For subjects who underwent polysomnography (PSG) at hospitals during the study period, we obtained information from the institution on their diagnostic results, apnea hypopnea index (AHI) levels and courses of treatment, particularly the presence of the indication of continuous positive airway pressure (CPAP) therapy which is commonly given to those with severe SAS, AHI 20, in Japan. Analysis To determine the relationship between ODI and AHI, we calculated a correlation coefficient and a regression line, from the ODI and AHI. To define the characteristics of the subjects to whom CPAP therapy was indicated, we compared the results of questionnaire items (snoring, apnea, urination, arousals, morning headaches, excessive daytime sleepiness), ESS, age, body mass index (BMI) for screening-positives with CPAP therapy indication (CPAP-P group), screeningnegatives with CPAP therapy indication (CPAP-N group) and control subjects (the rest of the screening-negatives after excluding those who were diagnosed as SAS). Questionnaire items were analyzed by using Pearson s chi-square test; and ESS, age and BMI were analyzed by using Student s t-test. For more detailed analysis on the characteristics of the CPAP-N group, questionnaire results, pulse oximetry results (ODI4, ODI3, CT 90 ), age, and BMI were compared between SAS patients screeningnegative but for whom CPAP therapy was not indicated (SAS-N group) and CPAP-N. Questionnaire items were

4 4 J Occup Health, Vol. 49, 2007 Table 2. Results of the screening Screening Positive Screening Negative Total Total number of subjects screened (%) 83(100.0%) 284 (100.0%) 367 (100.0%) No. of subjects who had a close examination (%) 54 (65.1%) 11 (3.9%) 65 (17.7%) No. of subjects who were diagnosed SAS with AHI 5 (%) 54 (65.1%) 11 (3.9%) 65 (17.7%) No. of subjects to whom CPAP therapy was indicated with AHI 20 (%) 48 (57.8%) 5 (1.8%) 53 (14.4%) SAS: sleep apnea syndrome, AHI: apnea hypopnea index, CPAP: continuous positive airway pressure, Screening positive: ODI4 10 or ODI3 15, Screening negative: ODI4<10 and ODI3<15. (%)=No. of relevant subjects/no. of subjects in a relevant group (screening positive group, screening negative group, whole). Table 3. Comparison between the groups of CPAP therapy indicated by screening result and the control group regarding the questionnaire results, age and BMI CPAP therapy indicated Control (N=273) Total (N=367) Screening positive (N=48) Screening negative (N=5) Witnessed snoring 97.9% # 100.0% 80.9% 85.0% Witnessed apnea 81.8% # 80.0% # 32.5% 43.5% Urination during the night 54.3% # 80.0% 37.7% 41.2% Arousals 50.0% 80.0% 35.3% 38.8% Morning headaches 23.9% 60.0% 23.5% 24.6% Daytime sleepiness 67.4% 100.0% 58.8% 61.6% ESS (mean ± S.D.) 8.1 ± 3.3* 11.4 ± 4.6* 6.6 ± ± 3.6 Age (mean ± S.D.) 45.8 ± 8.0* 33.4 ± ± ± 9.8 BMI (mean ± S.D.) ± 5.89* ± 4.70* ± ± 4.40 # : Significant probability of p<0.05 by Pearson s chi-square test for the control group (Fisher s exact probability test was used if expected frequency in any cell was less than 5). *: Significant probability of p<0.05 by Student s t test for the control group. Control: Screening negatives (excluding diagnosed SAS). N=maximum number (No. of subjects relevant to each item differs depending on valid number of responses). analyzed by using Pearson s chi-square test; ESS, age and BMI were analyzed by using Student s t-test; and ODI and CT 90 were analyzed by using the Mann-Whitney U test. To evaluate the effect of the positive involvement of occupational physicians in selecting screening subjects, we assessed the significance of differences in age, BMI, ESS, ODI4, ODI3, CT 90, AHI (of those who underwent PSG during the study period), snoring, witnessed apnea and the screening results between the subjects who applied for the screening and those who were instructed to participate in the screening by their occupational physicians. Age, BMI and ESS were analyzed by using Student s t-test, and ODI4, ODI3, CT 90 and AHI were analyzed by using the Mann-Whitney U test. Snoring, witnessed apnea and the screening results were analyzed by using Pearson s chi-square test. For statistical analysis, an appropriate software package (SPSS 12.0J for Windows) was used. Results Data valid for summing up and analysis were obtained from 367 subjects. The remaining 13 subjects were excluded because their valid measurement time was insufficient, and, therefore, the reliability of their ODI was low. The mean valid measurement time was ± 92.1 (SD: standard deviation) min. ( min). Characteristics of the 367 subjects were as follows: mean age 42.2 ± 9.8 (SD) yr (19 66 yr), mean BMI ± 4.40 (SD) kg/m 2 ( kg/m 2 ) (Table 3), and 54.2% of them were obese with BMI of 25 or greater. There were 83 screening-positives (Table 2). Seventysix of them exhibited ODI4 10 and ODI3 15, 6 of them exhibited ODI4 10 and ODI3<15, and 1 of them exhibited ODI4<10 and ODI3 15. Of the 83 screeningpositives, 54 underwent a close examination during the study period and all of them were diagnosed as SAS with AHI 5. Of the 54 who were diagnosed as SAS, CPAP therapy was indicated for 48 subjects with AHI 20, 88.9%

5 Kuniyuki NIIJIMA, et al. The Usefulness of SAS Screening by Oximetry in the Workplace 5 of the 54 subjects with SAS. On the other hand, 284 were determined as screening-negative. Of the 284 screening-negatives, 102 in total had one or more of the following: ODI4 5, ODI3 10, CT 90 1%, ESS 11 points. Of these 102, 11 of them underwent a close examination during the study period and all were diagnosed as SAS with AHI 5. In 5 of the 11 subjects (45.5%) with AHI 20 CPAP therapy was indicated. Accordingly, 65 subjects in total (17.7%) among the screened (valid 367) in our study were diagnosed with SAS, and CPAP therapy was Fig. 2. Relationship between ODI4 and AHI in patients with SAS (n=53). indicated in 53 subjects. A Pearson s correlation coefficient of (p<0.01) was obtained in the screening, indicating a good correlation between ODI3 and ODI4. The value of ODI3 which was equivalent to ODI4=10, as calculated from the regression line, was For 53 of 65 subjects who underwent a close examination, we evaluated a correlation between ODI values obtained in the screening and AHI levels obtained from hospitals; and obtained correlation coefficients of (p<0.01) (Fig. 2) and (p<0.01) for ODI4 and ODI3, respectively. Questionnaire results for the CPAP-P group, CPAP-N group and the control group are shown in Table 3. In the CPAP-P group, items of snoring, witnessed apnea and urination during the night were significantly higher than the control group; but arousals from sleep, morning headaches and excessive daytime sleepiness did not appear to be significantly different. In the CPAP-N group, only apnea was found to be significantly different from the control group. Snoring, urination during the night, arousals from sleep, morning headaches and daytime sleepiness were higher than the control group, although not significantly. ESS and BMI were significantly high in both the CPAP-P group and CPAP-N group compared to the control group. Age was significantly high in the CPAP-P group compared to the control group, but was lower than the control group in the CPAP-N group, although not significantly. The ages of the CPAP-N group were significantly lower than those of the SAS-N group. In the CPAP-N group, although ODI4, ODI3, CT 90, ESS, BMI, and all of the questionnaire items (snoring, witnessed apnea, urination during the night, arousals from sleep, morning headaches and excessive daytime Table 4. Comparison between SAS screening-negatives with indication of CPAP and those without indication of CPAP SAS patients screening negative CPAP therapy was indicated (N=5) CPAP therapy was not indicated (N=6) Significant difference Witnessed snoring 100.0% 100.0% * Witnessed apnea 80.0% 33.3% n.s.* Urination during the night 80.0% 33.3% n.s.* Arousals 80.0% 50.0% n.s.* Morning headaches 60.0% 16.7% n.s.* Daytime sleepiness 100.0% 50.0% n.s.* ODI4 (mean ± S.D.) 6.47 ± ± 2.40 n.s.** ODI3 (mean ± S.D.) ± ± 2.90 n.s.** CT 90 (mean ± S.D.) 5.58 ± ± 1.07 n.s.** ESS (mean ± S.D.) 11.4 ± ± 2.9 n.s.*** Age (mean ± S.D.) 33.4 ± ± 3.7 p<0.01*** BMI (mean ± S.D.) ± ± 3.74 n.s.*** *: Pearson s chi-square test (Fisher s exact probability test was used if expected frequency in any cell was less than 5). **: Mann- Whitney U test, ***: Student s t test.

6 6 J Occup Health, Vol. 49, 2007 Table 5. Comparison of the subject selection method (voluntarily or instructed) Subject Applicants (N=279) Instructed (N=88) Significant difference Age (Mean ± S.D.) 42.3 ± ± 10.2 n.s.* BMI (Mean ± S.D.) ± ± 5.71 p<0.01 * ESS (Mean ± S.D.) 7.0 ± ± 3.6 n.s.* ODI4 (Mean ± S.D.) 6.92 ± ± p<0.01** ODI3 (Mean ± S.D.) 9.25 ± ± p<0.01** CT 90 (Mean ± S.D.) 3.73 ± ± p<0.05** AHI # (Mean ± S.D.) ± ± n.s.** Witnessed snore 81.5% 97.5% p<0.01*** Witnessed apnea 42.2% 48.6% n.s.*** Screening positive 19.0% 34.1% p<0.01*** N=maximum number. No. of subjects relevant to each item differs depending on valid number of responses. *: Student s t test, **: Mann-Whitney U test, ***: Pearson s chi-square test. AHI # : Only the subjects who underwent PSG measurement (voluntarily: 30, Instructed: 23). sleepiness) were not significantly different, they were higher than the SAS-N group (Table 4). Although no significant differences were found for age, ESS, AHI and witnessed apnea, BMI, ODI4, ODI3, CT 90, snoring and the screening-positive rates were significantly higher in the subjects who underwent this screening on their occupational physician s recommendation than those who underwent this screening voluntarily (Table 5). Discussion Even though oximetry was performed at each worker s home in an unattended setting using a portable pulse oximeter, of 380 subjects screened, 96.6% (367) provided data which were valid for analysis, allowing a highly reliable measurement. The device used in the screening was easy to use; and the sensor was easily applied by the subject. The measurements obtained by using the device were thought to be useful and appropriate as a screening tool for workers. It has been reported that ODI4 15 identified patients with AHI 15 with lower sensitivities of 31% 15) and 40% 13) by home oximetry using a pulse oximeter for SAS screening. Chiner et al. reported that when patients with suspected SAS were examined, for ODI4 10, pulse oximetry identified patients with AHI 15 with a sensitivity of 71%, which was higher than that for ODI4 15 (sensitivity of 63%) 16). On the other hand, for ODI3 15 in such a study, home pulse oximetry identified patients with AHI 15 with a sensitivity of 73.5% which was higher than that for ODI4 15 (sensitivity of 52.9%) 17). Although these results were obtained by using types of oximeter different from the one used in this study, we used ODI4=10 and ODI3=15 as cut-off points based on these findings which, we think, should reduce false negative rates in screening for SAS at workplaces. In the screening, the number of subjects who exhibited ODI4 10 and were defined as screening-positive were 82, a higher number than those who exhibited ODI3 15 (77) and were also defined as screening-positive. Since the value of ODI3 equivalent to ODI4=10 calculated from the regression line was 12.7, we consider that the sensitivity of ODI3 15 might be lower than that of ODI4 10. On the other hand, our cut-off points (ODI4=10 and ODI3=15) detected severe cases of SAS with AHI 20 for whom CPAP therapy was indicated, which accounted for 88.9% of those who were screeningpositive and having SAS. The validity of the pulse oximetry was confirmed by synchronous overnight recordings of both pulse oximetry, using the same type of a pulse oximeter used in our screening, and PSG in a medical institution to which patients with suspected SAS had been referred. The sensitivity and specificity for detecting AHI 20 using a cut-off threshold of ODI4=10 were 88% and 100%, respectively, and for ODI3=15 they were 85% and 100%, respectively 11). Therefore, a portable pulse oximeter was indicated to be an effective tool for detecting severe degree SAS patients who need to be treated in the workplace. In addition, highly significant relationships between ODI values obtained in this screening and AHI values obtained from medical institutions were observed for both ODI4 and ODI3 in patients with SAS, indicating the usefulness of pulse oximetry for estimating the severity of SAS. In this screening, auxiliary criteria such as ODI4 5, ODI3 10, CT 90 1%, and ESS 11 were added to reduce the number of undetected SAS due to false negatives occurring in the screening. This lead to the identification of a further 11 SAS patients and CPAP was indicated for 5 of them. Statistical analysis suggested that the results of the questionnaire for CPAP-N (n=5) showed

7 Kuniyuki NIIJIMA, et al. The Usefulness of SAS Screening by Oximetry in the Workplace 7 significantly higher values of apnea, while the values of the rest of the items were higher, but not significantly, than the control group; mean values of ESS and BMI were significantly higher than the control group (Table 3). The CPAP-P group showed a similar trend. A study of SAS patients reported that age, obesity, snoring and apnea are the predictors of SAS and that more patients with severe SAS complain of daytime sleepiness than those with mild or moderate SAS, thus, our results appear to be supported 18). On the other hand, in contrast to the CPAP-P group in which mean age was significantly higher than the control group, no significant difference in mean age was found from the control group in the CPAP-N group, but its mean age was the lowest. In addition, for the 11 SAS patients screening-negative, the age of the CPAP-N group was significantly lower than the SAS-N group (Table4). Therefore, care should be taken over determination in young individuals who have SAS related symptoms, findings including obesity, and have any of ODI4 5, ODI3 10, CT 90 1%, ESS 11, even if they have been identified as screening-negative. The reasons for the false negatives in screening can be explained as follows: (1) since sleeping hours are not measured in oximetry, and ODI and CT 90 are calculated based on the total measuring time, these indicators may be underestimated; (2) slight respiratory disturbances not reaching desaturation of more than 4% or 3% from the baseline evaluated in this screening were not detected; (3) young individuals tend to have higher baseline SpO 2 19) and falls in SpO 2 are unlikely to occur 11). To deal with these factors, we are currently considering a method using a flow sensor detecting changes in airflow in the nose and mouth. Furthermore, in regard to the effect of the positive involvement of occupational physicians in selecting screening subjects, while no differences were found for age and sleepiness (ESS score) in the subjects who were instructed to participate in the screening, many of them were obese and frequently witnessed snoring, and their screening positive rate was also higher (Table 5). To eliminate occupational physicians subjective evaluation, we compared the sensitivities in detecting screeningpositive indicators of ODI4 10 or ODI3 15 for obesity (BMI 25), witnessed snoring, and ESS 11 in the 290 subjects who voluntarily underwent pulse oximetry regardless of symptoms. The results were 84.9%, 96.2%, and 26.9%, respectively (and for specificity, 59.3%, 22.1%, and 85.1%, respectively). The involvement of occupational physicians in the selection of subjects for screening was considered to be necessary for efficient screening, and selecting subjects particularly in relation to two of the above factors, obesity and witnessed snoring, was also considered to be useful. We think that these two factors are also important for reducing false negatives since a similar trend in the detection of false negatives in screening, identifying subjects for whom CPAP therapy was indicated, was described above. In the present screening, 17.7% of the subjects screened were found with SAS under close examination. Although it is not possible to compare strictly, the proportion of SAS cases in this screening was higher than the prevalence reported among the workforce 9). The overall characteristics of this screening were as follows: 54.2% were obese with BMI 25; and 85.0% answered to have witnessed snoring and 43.5% to have witnessed apnea in the questionnaire (Table 3). According to the National Nutrition Survey of 2002 in Japan, 28.9% of men and 23.0% of women over the age of 20 were obese with BMI 25 20). A study of about 7,000 Japanese factory workers reported that 64.7% of them presented with witnessed snoring and 8.2% of these workers were witnessed to have multiple episodes of partial or complete cessation of respiration during sleep 21). Therefore, the reason for the high proportion of SAS observed in the present screening is that it included the effect of the selection of some subjects by occupational physicians. Accordingly, more cases of witnessed respiratory disturbances and obesity may have been included in this screening than are present in the general population. Conclusion The simplicity of the SAS screening by pulse oximetry makes it easy to use for screening of workers. This method, using ODI as screening diagnostic criteria for SAS, was also highly effective at detecting individuals with severe SAS, AHI 20, for whom CPAP therapy was indicated. The need for caution should be emphasized for factors such as obesity, witnessed snoring and apnea to reduce false negatives especially in young individuals with mild desaturation. Acknowledgments: We wish to thank the following for their helpful assistance: the occupational health nurses who belong to Shizuoka Health Care Center of Central Japan Railway Company, other occupational health staff of Central Japan Railway Company, and medical specialists who accommodated us with reports of their diagnostic results and treatment plans. Part of this screening was supported by a grant given to the 2002 Occupational Health Research Project by the Occupational Health Promotion Foundation. A part of this report was presented at the 77th Annual Meeting of the Japan Society for Occupational Health, held in Nagoya, April References 1) The Report of an American Academy of Sleep Medicine Task Force: Sleep-related breathing disorders in adults: recommendations for syndrome definition

8 8 J Occup Health, Vol. 49, 2007 and measurement techniques in clinical research. Sleep 22, (1999) 2) Peppard PE, Young T, Palta M and Skatrud J: Prospective study of the association between sleepdisordered breathing and hypertension. N Engl J Med 342, (2000) 3) Shahar E, Whitney CW, Redline S, Lee ET, Newman AB, Nieto FJ, O Connor GT, Boland LL, Schwartz JE and Samet JM: Sleep-disordered breathing and cardiovascular disease, cross-sectional results of the Sleep Heart Health Study. Am J Respir Crit Care Med 163, (2001) 4) Elmasry A, Janson C, Lindberg E, Gislason T, Tageldin MA and Boman G: The role of habitual snoring and obesity in the development of diabetes: a 10-year follow-up study in a male population. J Intern Med 248, (2000) 5) The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (The JNC7 report). JAMA 289, (2003) 6) He J, Kryger MH, Zorick FJ, Conway W and Roth T: Mortality and apnea index in obstructive sleep apnea Experience in 385 male patients. Chest 94, 9 14 (1988) 7) Findley LJ, Unverzagt ME and Suratt PM: Automobile accidents involving patients with obstructive sleep apnea. Am Rev Respir Dis 138, (1988) 8) Findley LJ, Fabrizio M, Thommi G and Suratt PM: Severity of sleep apnea and automobile crashes. N Engl J Med 320, (1989) 9) Young T, Palta M, Dempsey J, Skatrud J, Weber S and Badr S: The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 328, (1993) 10) Tanigawa T, Horie S, Sakurai S and Iso H: Screening for sleep-disordered breathing at workplaces. Ind Health 43, (2005) 11) Nakamata M, Kubota Y, Sakai K, Kinefuchi S, Nakayama H, Ohdaira T, Satoh M, Shinoda H and Kohno M: The limitation of screening test for patients with sleep apnea syndrome using pulse oximetry. J Jap Soc Respir Care 12, (2003) (in Japanese) 12) Netzer N, Eliasson AH, Netzer C and Kristo DA: Overnight pulse oximetry for sleep-disordered breathing in adults. A review. Chest 120, (2001) 13) Gyulay S, Olson LG, Hensley MJ, King MT, Allen KM and Saunders NA: A comparison of clinical assessment and home oximetry in the diagnosis of obstructive sleep apnea. Am Rev Respir Dis 147, (1993) 14) Johns MW: A new method for measuring daytime sleepiness: The Epworth Sleepiness Scale. Sleep 14, (1991) 15) Ryan PJ, Hilton MF, Boldy DAR, Evans A, Bradbury S, Sapiano S, Prowse K and Cayton RM: Validation of British Thoracic Society guidelines for the diagnosis of the sleep apnoea/hypopnoea syndrome: can polysomnography be avoided? Thorax 50, (1995) 16) Chiner E, Signes-Costa J, Arriero JM, Marco J, Fuentes I and Sergado A: Nocturnal oximetry for the diagnosis of the sleep apnoea hypopnoea syndrome: a method to reduce the number of polysomnographies? Thorax 54, (1999) 17) Sano K, Nakano H, Ohnishi Y, Ishii Y, Nakamura T, Matuzawa K, Maekawa J and Narita N: Screening of sleep apnea/hypopnea syndrome by home pulse oximetry. Nihon Kokyuki Gakkai Zasshi 36, (1998) (in Japanese) 18) Viner S, Szalai JP and Hoffstein V: Are history and physical examination a good screening test for sleep apnea? Ann Int Med 115, (1991) 19) Gries RE and Brooks LJ: Normal oxyhemoglobin saturation during sleep. How low does it go? Chest 110, (1996) 20) Ministry of Health, Labour and Welfare, Japan. The national nutrition survey in Japan, Tokyo: Daiichi Shuppan, 2004 (in Japanese). 21) Kurono T, Niijima M, Sakuma T, Tatsumi K, Kimura H, Kunitomo F, Oota Y and Kuriyama T: Epidemiological study of snoring a questionnaire survey in factory workers. Nihon Kyobu Shikkan Gakkai Zasshi 31, (1993) (in Japanese)