Management of Sleep Apnea

Original Research SLEEP MEDICINE

Management of Sleep Apnea* Concordance Between Nonreference and Reference Centers Lourdes Herna´ndez, MD; Marta Torrella, MD; Nu´ria Roger, MD; Antonia Llunell, MD; Eugeni Ballester, MD; Llorenc¸ Quinto, MD; Mario Serrano, MD; Fernando Masa, MD; and Josep M. Montserrat, MD

Background: When a disorder is as prevalent as sleep apnea-hypopnea syndrome (SAHS), different medical levels and approaches should be involved in facilitating the diagnosis, at least, of patients with symptoms that disrupt social or working life, and of risk groups such as professional drivers. We sought to analyze the degree of concordance between management and treatment decisions for SAHS patients at sleep reference centers (RCs) and at non-RCs (NRCs). Materials and methods: Eighty-eight consecutive patients with suspected SAHS were referred by family doctors to the NRC. The patients were studied randomly at the RC, using full polysomnography, and at the NRC, employing respiratory polygraphy. The concordance in the therapeutic approach between both centers was analyzed via the outcomes, and the physiologic variables of the sleep studies were also evaluated. Results: The concordance in the final clinical decision was substantial. There was a good agreement with respect to apnea-hypopnea index as a categoric variable and as a continuous variable. Conclusions: These results suggest that respiratory physicians with simple sleep studies are able to manage a large number of patients with SAHS. Trial registration: Clinicaltrials.gov Identifier: NCT 00424658. (CHEST 2007; 132:1853–1857) Key words: full polysomnography; obstructive apnea; respiratory polygraphy; simple sleep studies Abbreviations: AHI ⫽ apnea-hypopnea index; CPAP ⫽ continuous positive airway pressure; CT90 ⫽ cumulative time spent with oxygen saturation of ⬍ 90%; NRC ⫽ nonreference center; RC ⫽ reference center; SAHS ⫽ sleep apnea-hypopnea syndrome

apnea-hypopnea syndrome (SAHS) is a comS leep mon disorder and full-night polysomnography 1–3

is the recommended method for establishing the diagnosis.4,5 This technique and the evaluation of the patients with a clinical suspicion of SAHS are usually performed by physicians who specialize in sleep medicine at reference centers (RCs). Increasing awareness of SAHS in the media and in medical circles, in addition to the growing evidence of SAHS as a risk factor in traffic accidents,6,7 hypertension,8,9 cardiovascular disease,10,11 and cerebrovascular disease,12–14 have exponentially increased the number of patients for evaluation. This growth in demand has, however, not been accompanied by any improvements in the approach to this problem. At present, the techniques of evaluation, which are www.chestjournal.org

complex and time consuming, are usually restricted to the RC. There is, therefore, a growing interest in alternative diagnostic methods and approaches. One approach that could be helpful is the performance of simple sleep studies, partially supervised, in a hospital or at home.15–20 However, the underlying problem persists (ie, simple sleep studies and patient assessments continue to be performed in overcrowded sleep laboratories at the RC). Accordingly, when a disorder is as prevalent as SAHS, physicians at different medical levels should be involved in facilitating the diagnosis of a broader number of patients, or at least of those patients with moderate-to-severe symptoms or those who belong to a risk group (eg, professional drivers). One good option could be to transfer the patient assessment to non-RCs CHEST / 132 / 6 / DECEMBER, 2007

1853

(NRCs). The present study therefore seeks to evaluate this strategy by analyzing the degree of concordance between the RC and the NRC with respect to decisions on the treatment and management of SAHS patients. Materials and Methods Study Subjects The study population consisted of 88 consecutive subjects with a suspicion of SAHS (mean [⫾ SD] age, 50 ⫾ 11 years; 81% male; mean body mass index, 30 ⫾ 4 kg/m2) from the outpatient clinics of three NRCs in the area of Barcelona, Spain (Hospital Asil de Granollers, Hospital General de Vic, and Hospital de Terrassa). Study Design All patients were evaluated randomly within 1 month at the RC and at the NRC. The patients were assessed on the basis of the clinical history, a specific questionnaire about sleep-disordered breathing, and a sleep study in the hospital. The evaluation was performed blinded by a sleep physician at the RC (polysomnography) and by a respiratory physician trained in sleep medicine at the NRC (simple sleep study). The patients did not know the final results until both tests had been performed. In both cases, the choice of treatment was registered on an ordinal scale with the following four points: 1, no diagnosis of SAHS and patient is discharged from the hospital; 2, mild SAHS, in which the patient should follow a conservative treatment and clinical control; 3, moderate-to-severe SAHS, in which patient should begin continuous positive air pressure (CPAP) treatment; and 4, other sleep disorders are diagnosed by the RC or there is a need for full-night polysomnography owing to a discordance between clinical features and respiratory polygraphy at the NRC. The indication for CPAP treatment followed the national guidelines of the Sociedad Espan˜ola de Patologia Respiratoria21 summarized as follows: (1) patients with severe SAHS-related symptoms with an abnormal apnea-hypopnea index (AHI) of ⬎ 10); or (2) patients with mild-to-moderate clinical symptoms with a higher AHI of ⬎ 30. The Human Ethics Committee of our hospital approved the protocol and informed consent was obtained from all the patients. *From Institut Clı´nic del Torax (Drs. Herna´ndez, Ballester, Quinto, Serrano, and Montserrat), Hospital Clı´nic, Institut d’Investigacions Biome`diques Agustı´ Pi i Sunyer (IDIBAPS), CibeRes, Barcelona, Spain; Pneumologia (Dr. Torrella), Hospital Asil de Granollers, Granollers, Spain; Pneumologia (Dr. Roger), Hospital General de Vic, Vic, Spain; Pneumologia (Dr. Llunell), Hospital de Terrassa, Terrassa, Spain; and San Pedro de Alca´ntara Hospital (Dr. Masa), CibaRes, Ca´ceres, Spain. This research was supported by grants from the Ministerio de Educacion y Ciencia (2004-00684); and Sociedad Espan˜ola de Patologia Respiratoria, CibeRes (ISCiii CB06/06). [15.25]. The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Manuscript received January 27, 2007; revision accepted August 17, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Josep M. Montserrat, MD, Institut Clı´nic del To`rax (ICT), Hospital Clı´nic, Institut d’Investigacions Biome`diques Agustı´ Pi i Sunyer, CibeRes, Villarroel 170, 08034 Barcelona, Spain; e-mail: [email protected] DOI: 10.1378/chest.07-0250 1854

Sleep Studies RCs: Polysomnography was performed in the usual manner.22–24 Briefly, the variables registered were EEG (three channels), chin electromyogram, electrooculogram, tibial electromyogram, arterial oxygen saturation, rib cage and abdominal motion, and their sum. Airflow was measured by cannula/thermistor. An apnea was defined as the absence of airflow ⱖ 10 s. Hypopnea was defined by any discernible reduction in the amplitude of the airflow signal ending in an arousal and/or association with a 3% desaturation, with a duration of at least 10 s. Upper airway resistance syndrome was defined as short periods of flow limitation ending with an arousal. An expert technician controlled the study. The sleep stages and respiratory variables were scored manually. An AHI of ⱖ 10 was considered to be abnormal. NRCs: The respiratory polygraphy was carried out in a simple silent room adjacent to the ward area. The following variables were recorded: body position; rib cage and abdominal motion; snoring; arterial oxygen saturation; and airflow using a cannula/ thermistor. The respiratory physician performed a manual scoring of the recording. Respiratory physicians previously underwent a period of training of 6 weeks duration in sleep medicine at a RC and attended a sleep training course of national repute. A nurse or a technician controlled the study. The definitions of hypopnea and apnea resembled those mentioned above, except in the case of arousal. Statistical Analysis Data were entered using a statistical software package (SPSS, version 10.0; SPSS; Chicago, IL) and were imported to another program (Stata, version 7.0; StataCorp; College Station, TX) to perform the analysis. Descriptive Analysis: Data were expressed as the mean ⫾ SD or percentage, respectively, for quantitative and qualitative variables. For continuous variables, a logarithmic transformation was undertaken to normalize the distribution, if necessary. If the distribution was not symmetrical, data were expressed as percentiles. Concordance Analysis: Evaluation of the concordance of the final outcome and the choice of treatment at the RC and the NRC were undertaken by using the statistic ␬ balanced, penalizing extreme discrepancies, following the classification of Landis & Koch.25 This procedure was also used for the concordance in the result of the AHI of the sleep study, categorized as follows: ⬍ 10; 10 to 29; and ⱖ 30. Concordance in the results of the sleep study (AHI and the cumulative time spent with an oxygen saturation of ⬍ 90% [CT90]) was evaluated in accordance with the methodology of Bland and Altman26 and the concordance coefficient of Lin.27

Results The characteristics of the patients evaluated and the main results of the sleep studies are summarized in Table 1. Thirty patients (34%) were evaluated at the Hospital Asil de Granollers, 31 patients (35%) were evaluated at the Hospital General de Vic, and 27 patients (31%) were evaluated at the Hospital de Terrassa. Three patients who were initially seen were not subsequently studied in both centers due to nonmedical reasons. They were removed from the study. Table 2 shows the distribution of patients with respect to the final outcome (see the “Study Design” Original Research

Table 1—General Characteristics of Patients Studied* Characteristics

Values

Patients (n ⫽ 88) Male sex, % Age, yr BMI,† kg/m2 Full-night polysomnography Sleep efficiency, % Arousal index† Stage I, % TST Stage II, % TST Slow-wave sleep, % TST REM sleep,† % TST AHI, events/h CT90, % TST Respiratory polygraphy RDI, events/h CT90, %

80.6 50.3 ⫾ 11.6 29.6 ⫾ 4.2 79.5 (70–88.6) 31.5 ⫾ 19 6.4 (4–10) 57 (50–70) 18 (9.3–24) 15.2 ⫾ 7.8 21.5 (8–57.5) 1 (0–8.6) 15 (4.5–48.5) 4 (0–22)

*Values are given as the arithmetic mean ⫾ SD or the median (25th percentile to 75th percentile), unless otherwise indicated. TST ⫽ total sleep time; RDI ⫽ respiratory disturbance index; BMI ⫽ body mass index. †Values are given as the geometric mean ⫾ SD.

section for classification). At the NRC, the final outcomes (choice of treatment) were patient discharge from the hospital (30.7%), diagnosis of SAHS-ambulatory control (21.6%), and diagnosis of SAHS-CPAP treatment (34.1%). A full polysomnography was requested in the remaining 13.6% of cases. At the RC, the percentages were 23.9%, 33%, and 36.4%, respectively, and other sleep disorders were found in 6.7% of patients. There was a substantial concordance in the Landis classification, with an agreement of 86.99% (␬, 0.711) when the patient hospital discharge, control, and CPAP treatment were analyzed. The main discordances in the choice of treatment can be summarized as follows: (1) three patients were discharged from the NRC (AHI, ⬍ 10) or received a diagnosis of mild SAHS, but CPAP was indicated at the RC (high AHI); (2) in three patients, CPAP treatment was recommended at the NRC, but with follow-up at the RC; and (3) 12 patients were in

Table 2—Distribution of Patients According to Final Outcome* RC NRC

1

2

3

4

1 2 3 4

15 4 0 2

8 13 3 5

1 2 27 2

3 0 0 3

*The Landis classification of agreement was evaluated among categories 1, 2, and 3 (86.99%; ␬ , 0.711). www.chestjournal.org

other situations with no clinical relevance, the differences being mainly due to discrepancies in AHI with mild symptoms, or no symptoms other than snoring. Finally, the NRC requested a polysomnography in 13.6% of patients, given the inconsistency between the results of the simple sleep study and the clinical features. At the RC, the following disorders were diagnosed in these patients: SAHS, five patients; upper airway resistance syndrome, two patients; other sleep disorders (ie, periodic leg movements in sleep and idiopathic somnolence), three patients; and hospital discharges, two patients. In the sleep study evaluation, the NRC obtained an AHI of ⬍ 10 in 38.6% of patients, from 10 to 29 in 22.7%, and ⱖ 30 in 38.6% of patients. At the RC, these percentages were 29.5%, 26.1%, and 44.3%, respectively. The concordance in the categoric AHI between NRC and RC was substantial, according to the classification of Landis and Koch,25 with an agreement of 83.5% (␬, 0.649). Taking the AHI as a continuous variable, we obtained good concordance for the Lin coefficient of 0.826 (95% confidence interval, 0.759 to 0.894; p ⬍ 0.0001). In the BlandAltman evaluation, the mean average was 4.6 (95% limits of agreement, ⫺25.08 and 34.31) [Fig 1]. This mean difference and the 95% limits of agreement were in crescendo if we separately analyze the centiles according to the magnitude of the value observed (not shown). The CT90 yields a good Lin coefficient of 0.77 (95% confidence interval, 0.677 to 0.863; p ⬍ 0.0001). The average difference in the Bland-Altman analysis was ⫺4.3 (95% limits of agreement, ⫺29.27 and 20.65).

Discussion This study shows that a respiratory polygraph performed by respiratory physicians at the NRC is of considerable value in the management of SAHS patients. The clinical therapeutic decision taken at a NRC after a respiratory polygraphy concurs with the one taken at an RC with full polysomnography in most cases. Therefore, a large number of patients with a suspected SAHS could be correctly diagnosed and managed by using respiratory parameters at a NRC. It is estimated that ⬎ 70% of individuals with SAHS have not received a diagnosis. One possible explanation for this is that the demand for sleep studies has not been accompanied by a change in diagnostic strategy. In 1992, Douglas et al15 demonstrated the usefulness of simple diagnostic methods by scoring nighttime respiratory variables from full polysomnography. This reduction of the signals enabled us to perform sleep studies more easily.16 –20 CHEST / 132 / 6 / DECEMBER, 2007

1855

management of SAHS patients. Our findings suggest that a considerable number of patients should be managed by the NRC.

Difference

AHI 50

References

0

-50 0

20

40

60

80

Average

Figure 1. Bland-Altman plot. The mean average between the number of events scored by the two methodologies was 4.6 (95% limits of agreement, ⫺25.08 and 34.31).

However, the joint statement by American sleep societies (ie, the American Thoracic Society, the American Academy of Sleep Medicine, and the American College of Chest Physicians),5 which classified sleep equipment into four levels according to complexity still does not accept respiratory polygraphy as a diagnostic method for SAHS, barring a small number of cases. In the present study, we sought not only to examine the metrics, as has already been done,16 –20 but also to compare the final outcome of the patient after examination by full polysomnography at an RC or by respiratory polygraphy at a NRC. We have demonstrated the feasibility of diagnostic studies when different levels of health care are involved. We have focused on clinical outcomes given that SAHSrelated symptoms play a significant part in decisions about CPAP treatment.21 However, some factors need to be taken into consideration with the above approach. First, training in sleep medicine. It is of paramount importance that physicians spend enough time in the sleep department and incorporate sleep medicine into their habitual practice. Second, as night respiratory polygraphy does not measure sleep efficacy, it should always be taken into account that the AHI could be underestimated.11–12,14 Third, SAHS patients with symptoms due to other sleep or general disorders should be correctly evaluated as they could need full polysomnography at the beginning of the assessment. Finally, fourth, the characteristics of the population under study are of great importance. Indeed, the results of the simplified studies are expected to be better in populations not previously assessed for suspected SAHS because the percentage of patients with severe SAHS is higher. In summary, different levels of the health-care system working in collaboration would facilitate the 1856

1 Young T, Palta M, Dempsey J, et al. The occurrence of sleep disorders breathing among middle aged adults. N Engl J Med 1993; 328:1230 –1235 2 Marin JM, Gascon JM, Carrizo S, et al. Prevalence of sleep apnoea syndrome in the Spanish adult population. Int J Epidemiol 1997; 26:381–386 3 Dura´n J, Esnaola S, Rubio R, et al. Obstructive sleep apnea-hypopnea and related clinical features in a populationbased sample of subjects aged 30 to 70 yr. Am J Respir Crit Care Med 2001; 163:685– 689 4 American Thoracic Society, Medical Section of the American Lung Association. Indications and standards for cardiopulmonary sleep studies. Am Rev Respir Dis 1989; 139:559 –568 5 Chesson AL Jr, Berry RB, Pack A. Practice parameters for the use of portable monitoring devices in the investigation of suspected obstructive sleep apnea in adults. Sleep 2003; 26:907–913 6 Tera´n-Santos J, Jime´nez-Go´mez, Cordero-Guevara J. The association between sleep apnea and the risk of traffic accidents: Cooperative Group Burgos-Santander. N Engl J Med 1999; 340:847– 851 7 Barbe´ F, Pericas J, Mun˜oz A, et al. Automobile accidents in patients with sleep apnea syndrome: an epidemiological and mechanistic study. Am J Respir Crit Care Med 1998; 158:18 –22 8 Peppard PE, Young T, Palta M, et al. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000; 342:378 –384 9 Nieto FJ, Young TB, Lind BK, et al. Association of sleepdisordered breathing, sleep apnea and hypertension in a large community-based study: Sleep Heart Health Study. JAMA 2000; 283:1824 –1836 10 Peker Y, Hedner J, Kraiczi H, et al. Respiratory disturbance index: an independent predictor of mortality in coronary artery disease. Am J Respir Crit Care Med 2000; 162:81– 86 11 Marin JM, Carrizo SJ, Vicente E, et al. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet 2005; 365:1046 –1053 12 Gami AS, Howard DE, Olson EJ, et al. Day-night pattern of sudden death in obstructive sleep apnea. N Engl J Med 2005; 352:1206 –1214 13 Parra O, Arboix A, Bechich S, et al. Time course of sleep-related breathing disorders in first-ever stroke or transient ischaemic attack. Am J Respir Crit Care Med 2000; 161:375–380 14 Yaggi HK, Concato J, Kernan WN, et al. Obstructive sleep apnea as a risk factor for stroke and death. N Engl J Med 2005; 353:2034 –2041 15 Douglas NJ, Thomas S, Jan MA. Clinical value of polysomnography. Lancet 1992; 339:347–350 16 Lloberes P, Montserrat JM, Ascaso A, et al. Comparison of partially attended night time respiratory recordings and full polysomnography in patients with suspected sep apnoea/ hypopnoea syndrome. Thorax 1996; 51:1043–1047 17 Ayappa I, Norman RG, Suryadevara M, et al. Comparison of limited monitoring using a nasal-cannula flow signal to full polysomnography in sleep-disordered breathing. Sleep 2004; 27:1171–1179 18 Calleja JM, Esnaola S, Rubio R, et al. Comparison of a cardiorespiratory device versus polysomnography for diagnosis of sleep apnoea. Eur Respir J 2002; 20:1505–1510 Original Research

19 Parra O, Garcı´a-Esclasans N, Montserrat JM, et al. Should patients with sleep apnoea/hypopnoea sı´ndrome be diagnosed and manager on the basis of home sep studies? Eur Respir J 1997; 10:1720 –1724 20 Ballester E, Solans M, Vila X, et al. Evaluation of a portable respiratory recording device for detecting apnoeas and hypopnoeas in subjects from a general population. Eur Respir J 2000; 16:123–127 21 Montserrat JM, Amilibia J, Barbe´ F et al. Tratamiento del sı´ndrome de las apneas-hipopneas durante el suen˜o. Arch Bronconeumol 1998; 34:204 –206 22 Montserrat JM, Ferrer M, Hernandez L, et al. Effectiveness of CPAP treatment in daytime function in sleep apnea syndrome: a randomized controlled study. Am J Respir Crit Care Med 2001; 164:608 – 613

www.chestjournal.org

23 Rechtschaffen A, Kales A, eds. A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. Washington, DC: US Government Printing Office, 1968; NIH Publication No. 204 24 American Sleep Disorders Association. EEG-arousals: scoring rules and examples; a preliminary report from the Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. Sleep 1992; 15:173–184 25 Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159 –174 26 Bland MJ, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 8:307–310 27 Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics 1989; 45:255–268

CHEST / 132 / 6 / DECEMBER, 2007

1857