Edited by: Mohamed Abu-Farha, Dasman Diabetes Institute, Kuwait
Reviewed by: Javier Ena, Hospital Marina Baixa, Spain; Zhichao Feng, Albert Einstein College of Medicine, United States
This article was submitted to Diabetes, a section of the journal Frontiers in Endocrinology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Obstructive sleep apnea (OSA) is a common disorder in Type 2 diabetes (T2D) patients further increasing their already high cardiovascular risk. As T2D patients typically not report OSA symptoms, systematic screening for OSA in this population is warranted. We aimed to determine the readiness of T2D patients to undergo screening and to compare their adherence to continuous positive airway pressure (CPAP) therapy with “regular” sleep clinic patients who typically seek medical advice on their own initiative. We therefore recruited 494 consecutive T2D patients and offered them OSA screening using home sleep monitoring (type IV device). All participants in high risk of moderate-to-severe OSA were recommended home sleep apnea testing (HSAT) followed by CPAP therapy. Patients were followed-up for 12 months and outcomes compared to 228 consecutive sleep clinic patients undergoing HSAT. Among 307 screened T2D patients, 94 (31%) were identified at high risk of moderate-to-severe OSA. Subsequently, 54 patients underwent HSAT, 51 were recommended, and 38 patients initiated CPAP (acceptance 75%). Among 228 sleep clinic patients, 92 (40%) were recommended and 74 patients initiated CPAP (acceptance 80%). After 1 year, 15 (39%) T2D and 29 (39%) sleep clinic patients showed good CPAP adherence (use ≥ 4 h/night ≥ 70% nights). In conclusion, 20 T2D patients needed to be screened in order to obtain one successfully treated patient. OSA screening in T2D patients identified 31% with moderate-to-severe OSA. Once diagnosed, their CPAP acceptance and adherence did not differ from sleep clinic patients. However, the reasons for the high dropout during the screening-diagnostic process impacting the overall success of the screening program need to be identified and addressed.
Obstructive sleep apnea (OSA) is a common treatable disorder known to increase cardiovascular mortality (
Given the impact of unrecognized and therefore untreated OSA in Type 2 diabetes patients, the International Diabetes Federation recommends screening Type 2 diabetes patients for OSA (
Even though the consequences of untreated OSA and health benefits of CPAP are well-known and are easy to explain to patients, sleeping with a CPAP mask might present the patient with both psychological and physical discomfort. CPAP treatment requires a high degree of patient cooperation which becomes even more pronounced in screening-eligible Type 2 diabetes patients who were hitherto unaware of an additional health problem or did not consider OSA symptoms to be significant enough to, seek medical help and, in this case, were simply asked to follow their doctor's advice. Low CPAP acceptance and adherence is an ongoing challenge despite efforts to improve patient comfort and support, ranging from 25 to 73% (
In this study we hypothesized that Type 2 diabetes patients will show markedly lower acceptance and adherence to CPAP therapy as the OSA diagnosis resulted from a systematic screening performed in subjects who did not seek health care specialist for OSA related symptoms, in contrast to sleep clinic patients, who started CPAP treatment because they actively approached sleep physician. Additionally, we aimed to provide real-life data on the suitability, effectiveness and outcomes of systematic OSA screening in Type 2 diabetes patients by determining the adherence of patients with Type 2 diabetes to the systematic screening-diagnosis-therapy process in an outpatient clinics.
Subjects were recruited in diabetes care outpatient clinics providing routine care to unselected patients with diabetes located in Prague, Czech Republic between March 2014 and March 2015. In total, 494 consecutive patients fulfilled the inclusion criteria–diagnosis of Type 2 diabetes mellitus and age 18–80 years. Eleven patients were subsequently excluded due to unstable psychiatric disorders (5 patients) or already diagnosed OSA (6 patients) resulting in 483 subjects included in the study.
During regular scheduled visits (regular diabetes care), a physician educated all the subjects about OSA and the health risks associated with OSA as well as about the treatment options. At the same time, OSA screening by home sleep monitoring was offered in the form of a type IV device (ApneaLink, ResMed, San Diego, CA, United States) that recorded hemoglobin saturation, heart rate and nasal airflow during sleep. The respiratory event index (REI) was determined and subjects were stratified into low or high risk (REI ≥ 15) for the presence of moderate-to-severe OSA. Patients in the high-risk group were referred to a sleep clinic for a diagnostic home sleep study using a portable sleep monitor that recorded hemoglobin saturation, heart rate, nasal airflow, chest and abdominal respiratory effort and ECG (Nox T3, Nox Medical, Reykjavik, Iceland). Subsequently, patients with confirmed moderate-to-severe OSA were offered CPAP treatment according to AASM guidelines in an outpatient sleep center (Inspamed Prague, Czechia).
The collaborating outpatient sleep center (Inspamed, Prague, Czechia) retrospectively included 252 consecutive patients who underwent diagnostic home sleep study in 2014 and were 18–80 years old. Subsequently, 24 patients were excluded for previous experience with OSA diagnosis or CPAP therapy, leading to 228 patients being evaluated in the study. Patients with moderate-to-severe OSA defined as REI ≥ 15 (none of them treated for diabetes) were offered CPAP treatment according to AASM guidelines. This study was carried out in accordance with the recommendations of the Ethical Committee of the Third Faculty of Medicine, Charles University, Prague with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the Ethical Committee of the Third Faculty of Medicine, Charles University, Prague.
CPAP usage data of both, Type 2 diabetes and sleep clinic patients were analyzed at 3 and 12 months after establishing optimal mask fit, treatment pressure and regime (titration). Acceptance was defined as the patient's agreement to CPAP therapy after titration. Adherence was assessed using reports downloaded from CPAP machines. Patients using CPAP ≥4 h per ≥70% of nights were considered having “Good” adherence, while lower CPAP usage was considered as “Poor” adherence.
Screening for the presence of OSA was performed using a type IV device (ApneaLink, ResMed, San Diego, CA, United States) that recorded hemoglobin saturation, heart rate and nasal airflow during sleep in a home setting. Subjects were instructed to set-up the device and keep regular sleep habits. Support in the form of a non-stop phone help-line was established and the devices were returned to investigators the next morning. Automatic scoring of respiratory events with a 4% desaturation threshold was performed, apneas defined as a ≥90% reduction in airflow for at least 10 s and hypopneas defined as a ≥30% reduction in airflow for at least 10 s together with hemoglobin desaturation of ≥4%. Patients with REI ≥15 were considered as being at high risk of moderate-to-severe OSA. For 13 patients the oxygen desaturation index (ODI) was used due to a poor airflow signal.
Sleep recordings were performed using a type III device that recorded hemoglobin saturation, heart rate, nasal airflow, ECG, chest and abdominal respiratory efforts (Nox T3, Nox Medical, Reykjavik, Iceland) in a home setting. The recordings were evaluated by a board-certified sleep medicine physician according to AASM criteria (apnea defined by a ≥ 90% reduction in airflow for at least 10 s and hypopnea defined as a ≥ 30% reduction in airflow for at least 10 s together with ≥ 4% desaturation). Patients with moderate-to-severe OSA (REI ≥ 15) were recommended to initiate CPAP treatment.
Statistical analysis was performed using Prizm 5 for Windows Software (GraphPad Software Inc., La Jolla, CA, United States). Differences in anthropometrical parameters between the patient groups were analyzed using a
Out of 483 consecutive Type 2 diabetes patients who fulfilled the inclusion criteria, 321 patients consented to undergo OSA screening, resulting in 307 analyzed sleep recordings of an acceptable quality. Among successfully screened patients, 31% (63 men and 31 women) were identified as being in a high risk of moderate-to-severe OSA and thus invited for a diagnostic sleep study. However, such a sleep study was performed for only 60% of them due to the unwillingness of patients to further continue with the diagnostic process (Figure
Flow diagram of Type 2 diabetes patients in the screening study T2D, Type 2 diabetes; OSA, Obstructive sleep apnea; REI, respiratory event index.
The Type 2 diabetes patients who accepted the diagnostic home sleep apnea testing (HSAT) were characterized by 42% higher REI (32.6 ± 2.4 vs. 22.9 ± 1.5,
Characteristics of type 2 diabetes patients screened for OSA by home sleep monitoring.
Patients, |
307 (100%) | 213 (69%) | 94 (31%) | 37 (39%) | 56 (60%) |
Men, |
177 (58%) | 114 (54%) | 63 (67%) | 24 (65%) | 38 (68%) |
Age (years) | 64.0 ± 0.5 | 63.7 ± 0.6 | 64.8 ± 1.0 | 65.1 ± 1.6 | 64.3 ± 1.3 |
BMI (kg/m2) | 31.2 ± 0.3 | 30.4 ± 0.3 | 33.0 ± 0.6 |
32.4 ± 0.8 | 33.4 ± 0.9 |
Hypertension, |
255 (83%) | 172 (81%) | 83 (88%) | 32 (86%) | 46 (82%) |
Dyslipidemia, |
262 (85%) | 182 (85%) | 80 (85%) | 32 (86%) | 41 (73%) |
CV disease, |
46 (15%) | 28 (13%) | 18 (19%) | 6 (16%) | 10 (18%) |
ESS, |
3.2 ± 0.2 | 5.8 ± 0.3 | 6.6 ± 0.4 | 5.1 ± 0.5 | 7.6 ± 0.6 |
REI_screening study, |
12.5 ± 0.8 | 5.3 ± 0.3 | 28.7 ± 1.6 | 22.9 ± 1.5 | 32.6 ± 2.4 |
REI_diagnostic study, |
37.7 ± 2.4 | ||||
T90, % | 23.6 ± 3.4 |
Based on the results of home sleep apnea testing, 51 Type 2 diabetes patients were recommended to initiate CPAP treatment. However, 13 patients dropped out before or during the CPAP titration, resulting in a CPAP acceptance rate of 75% (38 treated patients). A similar acceptance rate of 80% (
Flow diagram of Type 2 diabetes and sleep clinic patients undergoing diagnostic home sleep apnea testing.
Type 2 diabetes patients who were recommended to initiate the CPAP treatment were, in comparison to sleep clinic patients who received same recommendation, older (64.4 ± 1.3 vs. 52.3 ± 1.4,
CPAP acceptance – comparison of type 2 diabetes and sleep clinic patients.
Patients, |
51 (100%) | 38 (75%) | 13 (25%) | 92 (100%) | 74 (80%) | 18 (20%) |
Men, |
34 (67%) | 23 (61%) | 10 (77%) | 75 (82%) | 64 (86%) | 11 (61%) |
Age (years) | 64.4 ± 1.3 | 62.9 ± 1.6 | 68.8 ± 2.3 | 52.3 ± 1.4 |
51.9 ± 1.4 | 53.7 ± 3.0 |
BMI (kg/m2) | 33.8 ± 0.9 | 35.0 ± 1.1 | 30.2 ± 1.0 |
34.3 ± 0.8 | 34.8 ± 0.8 | 32.1 ± 1.4 |
Hypertension, |
46 (90%) | 35 (92%) | 11 (85%) | 60 (65%) |
49 (66%) | 11 (61%) |
Dyslipidemia, |
42 (82%) | 32 (84%) | 10 (77%) | 41 (45%) |
35 (47%) | 6 (33%) |
CV disease, |
10 (20%) | 8 (21%) | 2 (18%) | 7 (8%) | 5 (7%) | 2 (11%) |
ESS (score) | 7.7 ± 0.7 | 7.6 ± 0.8 | 8.2 ± 1.5 | 10.3 ± 0.7 |
10.7 ± 0.7 | 8.6 ± 1.0 |
REI, |
39.7 ± 2.5 | 40.8 ± 3.0 | 36.3 ± 3.7 | 41.9 ± 2.4 | 46.0 ± 2.4 | 24.8 ± 2.5 |
T90, % | 25.5 ± 3.6 | 31.3 ± 4.2 | 5.7 ± 2.3 |
20.8 ± 3.0 | 24.7 ± 3.0 | 4.9 ± 1.8 |
Sleep clinic patients who accepted CPAP exhibited more severe OSA than patients not accepting CPAP (REI 46.0 ± 2.4 vs. 24.8 ± 2.5,
The CPAP recordings obtained after a 1-year follow-up showed “good” adherence to CPAP treatment defined as CPAP usage ≥ 4 h in ≥ than 70% of nights in 15 out of 38 Type 2 diabetes patients and 29 out of 74 sleep clinic patients who initiated CPAP treatment resulting in a 39% adherence rate in both groups (Table
CPAP adherence 1 year after initiating treatment —comparison of Type 2 diabetes and sleep clinic patients.
Patients, n (%) | 38 (100%) | 15 (39%) | 23 (61%) | 74 (100%) | 29 (39%) | 45 (61%) |
Men, n (%) | 23 (61%) | 9 (60%) | 14 (61%) | 64 (86%) |
28 (97%) | 36 (80%) |
Age (years) | 62.9 ± 1.6 | 64.6 ± 2.2 | 61.8 ± 2.1 | 51.9 ± 1.4 |
50.4 ± 2.4 | 52.9 ± 1.7 |
BMI (kg/m2) | 35.0 ± 1.1 | 35.7 ± 1.7 | 34.6 ± 1.5 | 34.8 ± 0.8 | 35.3 ± 1.2 | 34.5 ± 1.2 |
Hypertension, n (%) | 35 (92%) | 12 (80%) | 23 (100%) |
49 (66%) |
19 (66%) | 30 (67%) |
Dyslipidemia, n (%) | 32 (84%) | 13 (87%) | 19 (83%) | 35 (47%) |
18 (62%) | 17 (38%) |
CV disease, n (%) | 8 (21%) | 2 (13%) | 6 (26%) | 5 (7%) | 4 (14%) | 1 (2%) |
ESS (score) | 7.6 ± 0.8 | 5.9 ± 0.8 | 8.7 ± 1.1 | 10.7 ± 0.7 |
10.6 ± 1.2 | 10.8 ± 0.8 |
REI, n | 40.8 ± 3.0 | 44.0 ± 5.1 | 38.7 ± 3.8 | 46.0 ± 2.4 | 52.5 ± 4.1 | 41.8 ± 2.8 |
T90, min | 31.3 ± 4.2 | 38.1 ± 7.4 | 26.8 ± 4.9 | 24.7 ± 3.0 | 32.8 ± 5.1 | 19.5 ± 3.6 |
T90 ≥ 10%, n (%) | 29 (76%) | 12 (80%) | 17 (74%) | 42 (57%) | 21 (72%) | 21 (47%) |
CPAP ADUall, h | 3.8 ± 0.4 | 5.8 ± 0.3 | 2.6 ± 0.3 |
3.7 ± 0.3 | 6.1 ± 0.2 | 2.1 ± 0.3 |
CPAP ADU1year, h | 4.5 ± 0.3 | 5.8 ± 0.3 | 3.3 ± 0.2 |
4.8 ± 0.2 | 6.1 ± 0.2 | 3.4 ± 0.3 |
The proportion of patients with good adherence after 3 months and 1 year of using CPAP did not significantly differ (44.7 vs. 39.5%
Differences in CPAP adherence 3 months and 1 year after initiating the CPAP treatment.
All initiating CPAP | 38 (100%) | 74 (100%) |
Good adherence in 3 months, |
17 (44.7%) | 30 (40.5%) |
Good adherence in 1 year, |
15 (39.5%) | 29 (39.2%) |
Good adherence in 3 months but not in 1 year, |
4 (10.5%) | 6 (8.1%) |
Good adherence in 1 year but not in 3 months, |
2 (5.3%) | 5 (6.8%) |
CPAP adherence in T2DM and sleep clinic patients 1 year after CPAP initiation.
All patients initiating CPAP treatment after titration | 38 (100%) | 74 (100%) |
Patients using CPAP ≥ 4 h/night ≥ 70% of nights, |
15 (39%) | 29 (39%) |
Patients using CPAP ≥ 4 h/night on average, |
18 (47%) | 38 (51%) |
CPAP ADUall, h (% of desired 7 h/night/patient) | 3.77 (54%) | 3.66 (52%) |
CPAP ADU1year, h (%of desired 7 h/night/patient) | 4.47 (64%) | 4.75 (68%) |
The present study showed that systematic OSA screening identified 31% of consecutive Type 2 diabetes patients as having a high risk of moderate-to-severe OSA, nevertheless, only 16% of them demonstrated a measurable clinical benefit of such screening by accepting and adequately adhering to CPAP treatment. However, the present study did not observe differences in CPAP acceptance and adherence rates in patients with Type 2 diabetes when compared to a sleep clinic population.
Although such results seem disappointing, they are in line with a study reporting that 17% of patients with Type 2 diabetes in high-risk of moderate-to-severe OSA initiated CPAP treatment (
Epidemiological studies have provided evidence that adequate CPAP use is crucial for improving health outcomes such as sleepiness (
Barriers preventing better compliance with OSA treatment were shown to be of a complex nature ranging from patient characteristic, disease severity, technological factors, means of OSA diagnosis and CPAP delivery to psychological and cultural variables (
The significance of this study is that it implemented systematic OSA screening in consecutive Type 2 diabetes patients and followed its outcome all the way through the diagnostic process to measure the effectiveness of the subsequent CPAP treatment. However, the limitations of the study should be noted. First, although a considerable number of 483 patients with Type 2 diabetes were included in the study, a severe dropout during the diagnostic process led to a relatively modest number of 38 patients who initiated the CPAP treatment and on whom the adherence to CPAP was followed. Second, some of the drop out during the diagnostic process might have been prevented if the first step of using a type IV device for screening was skipped and all patients directly underwent diagnostic monitoring using a type III device. Another way to lower the drop out during the diagnostic process might be to describe the alternative treatment option available to the patient—an oral appliance—in the event that clinically significant OSA is diagnosed. Oral appliances are not routinely available in the Czechia but might represent a more acceptable prospective way of treatment than CPAP for some of the patients and therefore could encourage some of those who dropped out to continue in the diagnostic process. Third, OSA was diagnosed using a type III device in a home setting and not by deploying the gold standard polysomnography. Nevertheless, such a diagnostic attitude is far more accessible than polysomnography and more likely reflects practice in everyday life. Fourth, the patients enrolled in the study, both Type 2 diabetes patients and sleep clinic patients, were diagnosed and treated at one sleep clinic, so CPAP acceptance and adherence can partly reflect the attitude of that particular clinic. On the other hand, the attitude of healthcare professionals to both groups of Type 2 diabetes patients and sleep clinic patients was by default the same as they were not aware of any grouping and therefore the comparison of CPAP acceptance and adherence between the groups is reliable. Finally, in the present study metabolic outcomes (e.g., glucose tolerance and HbA1c) and microvascular complications were not assessed, even though adequate CPAP adherence clearly represents a key factor determining its beneficial effects on glucose tolerance, insulin sensitivity (
In conclusion, every third to fourth patient with Type 2 diabetes suffers from clinically significant obstructive sleep apnea syndrome indicated for CPAP treatment whilst not being aware of his/her condition. Once diagnosed, their acceptance and adherence to CPAP did not differ from sleep clinic patients even though originally they did not actively seek medical advice regarding OSA symptoms. As almost half of the patients identified to be at high risk of OSA through screening were unwilling to undergo the subsequent diagnostic process, our study implies that factors preventing Type 2 diabetes from further medical evaluation should be targeted, probably through more extensive patient education combined with complex psychological approaches. A consideration should be also given to the possibility of conducting screening and diagnostic sleep studies in a single step using appropriate home sleep apnea testing devices.
KW and AP recruited subjects with Type 2 diabetes, performed screening sleep studies, gathered data, participated in data analysis, participated in manuscript preparation. VD participated in subject recruitment, served as a study coordinator, helped in performing diagnostic sleep studies, participated in data interpretation, participated in manuscript drafting. MP scored sleep studies, was responsible for CPAP therapy in all subjects, recruited non-diabetic subjects from the sleep clinic, participated in data interpretation and manuscript drafting. JP designed and supervised the study, analyzed data and participated in interpretation of the results, edited the manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This study was supported by grants from Ministry of Health of the Czech Republic, grant no. 15-30155A and Charles University grant Progres Q36.