Compliance in sleep apnoea therapy: influence of home care support and pressure mode

Obstructive sleep apnoea syndrome (OSAS) is a major public health burden, with a worldwide prevalence of 2–4% among the adult population 1. OSAS is associated with excessive daytime sleepiness, cognitive dysfunction, impaired quality of life, hypertension and an increased cardiovascular morbidity and mortality 2. Continuous positive airway pressure (CPAP) is an effective treatment for OSAS symptoms and also decreases cardiovascular morbidity and mortality 3–6.

However, the effectiveness of this symptomatic therapy mainly depends on regular use. The use/comfort of CPAP therapy may be limited by side effects, such as mucosal irritation, mask dislodgement, mask leak or difficulty in exhaling, which may critically impair compliance 7–12 or lead to discontinuation of therapy; 23% of patients quit CPAP within 5 yrs, most of them during the first year 13. In other publications, discontinuation rates ranged from 8 11 to 46% 14.

Most of the side-effects are associated with the airflow generated by the CPAP device and tend to worsen with higher pressures or airflow rates. The pressure required to achieve airway patency often varies throughout the night, i.e. it may be higher during REM sleep or in the supine position.

Autoadjusting-CPAP (APAP) devices were developed for continuous adjustment of pressure, which yields a lower mean pressure needed to achieve airway patency; pressure is increased when airflow limitations or an increase in upper airway resistance are identified, whereas pressure is decreased when the airway is open. However, no significant improvement of comfort and compliance has been demonstrated, while the clinical outcomes seem to be similar for both pressure modes 10, 15.

Intensive patient support is usually aimed at early minimising adverse factors and focussing patients' awareness on the necessity of the treatment and the benefit eventually derived from it. The first few weeks seem to be crucial for patients' ability to cope with the disease and their acclimatisation to therapy 16, 17; and thus for long-term compliance. Hence, most of the interventions studied were focussed on this period of time. Problems caused by mask, airflow or an increase in pressure can be addressed by re-adjustment of the interface and/or the CPAP device, whereas prejudice, embarrassment or technical problems in dealing with the equipment are a possible target to behavioural/educational measures. Investigators have used education about disease and therapy by different media 18, 19 and a closer monitoring by means of phone calls, letters 16 or telemedicine 20. One study was based on home visits and could show an increase in the hours of use, with an improvement in subjective daytime symptoms and mood 18; however, more reliable clinical outcome parameters regarding the disease severity, such as the apnoea/hypopnoea index (AHI), were not compared.

The primary aim of the present study was to test whether an intensive support of OSAS patients during the first 6 months of CPAP therapy can enhance CPAP adherence and improve clinical outcomes compared with standard support. In addition, the effectiveness and compliance of APAP and fixed-pressure therapy was compared. It is the first long-term interventional study, with 100 patients being followed over 9 months. A total of four different surrogate parameters for compliance and a set of clinical parameters, which allows for an estimation of cardiovascular impairment as well as sleep fragmentation and daytime somnolence, were used.

METHODS

Study design

A total of 100 patients (78 male and 22 female; mean±sd age 57±12 yrs; BMI 31±5 kg·m⁻²) with newly diagnosed OSAS were enrolled (inclusion criteria: AHI ≥15, with or without corresponding daytime symptoms). Exclusion criteria were as follows: 1) any global respiratory failure; 2) central sleep apnoea syndrome; and 3) any severe mental or psychological impairment. Before enrolment, all patients gave their written informed consent and answered a questionnaire regarding daytime sleepiness (Epworth Sleepiness Scale (ESS)). The study was approved by the local ethic committee of the Albert Ludwigs University of Freiburg (Freiburg, Germany).

Patients underwent standard diagnostic overnight polysomnography using a Heinen and Löwenstein SIDAS (Heinen and Löwenstein, Herrsching, Germany) or Jaeger Sleeplab 1000 (Jaeger, Würzburg, Germany) polysomnography system. Sleep measurements during polysomnography included standard electroencephalography (EEG), electro-oculography (EOG) and electromyography (EMG). Nasal airflow and oxygen saturation were measured by thermistor and pulse oxymetry, respectively, and patients were monitored and recorded by infrared video camera. Respiratory effort was determined by thoracic and abdominal gauges. Following polysomnography, sleep stages and respiratory events were visually analysed and edited, page by page. The diagnostic night (DIAG) was followed directly by a second full-night polysomnography with CPAP therapy (THER_INI), during which the CPAP adjustment was performed by attended autotitration (pressure-limits 6-14 mbar). Thereafter, patients were allocated to following groups: 1) standard support and APAP therapy (n = 25); 2) standard support and CPAP therapy (n = 25); 3) intensive support and APAP therapy (n = 25); and 4) intensive support and CPAP therapy (n = 25). Fixed CPAP pressure was obtained choosing the pressure level with lowest RDI during the polysomnography. In all four groups a further polysomnographic monitoring and interview was performed after 3 months (3M) and 9 months (9M) of CPAP treatment.

The intensive support groups were further visited by specially trained members of the present authors’ Sleep Laboratory of the Department of Pneumology at 1, 2, 4, 5 and 6 months after CPAP therapy (fig. 1⇓). Intensive support included optimising the equipment such as mask fit, the use of moisturers as indicated and an early identification of patients with low compliance, as well as educational support and counselling.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1—

Schematic of the support protocols and course of the study. A total of 100 patients with newly diagnosed obstructive sleep apnoea syndrome were included and 78 completed the study. Home visits were, from top, at months 1, 2, 4, 5 and 6 after diagnostic polysomnography. CPAP: continuous positive airway pressure; APAP: autoadjusting CPAP; 3M: follow-up at 3 months; 9M: follow-up at 9 months.

RESULTS

DISCUSSION

This controlled parallel group trial shows that an intensive support, especially during the first few months from therapy initiation can enhance long-term adherence to CPAP in obstructive sleep apnoea for up to 20%, whereas the use of APAP had no such effect. It is the first home-based interventional study to follow 100 patients over 9 months, using several different parameters for compliance and clinical outcome.

Overall compliance was high in continuing users in both groups, with >75% of days used and a mean nightly usage of over 5 h. Comparable data on CPAP adherence have been published before. Pieters et al. 21 reported on a mean use of 5±1.8 h·night⁻¹, remaining stable for 2 yrs. Kribbs et al. 14 found an utilisation rate of 66% of the days monitored. A recent study by Sucena et al. 13 showed an increase in compliance in continuing users in the long term with 5.35 h after 1 yr and 6.55 h after 10 yrs. 23% of patients quit CPAP within 5 yrs, most of them during the first year.

In the present study, the most evident effect of the intervention was a lower drop-out rate among patients who had received an intensive support, where compliance data were monitored closely and therapy problems could be addressed immediately. 88% of these returned for follow-up after 9 months versus 68% of patients receiving standard support, which means that more patients were kept from discontinuing their therapy by the intervention performed. This suggests that the general acceptance of CPAP as a key aspect of therapy efficiency is an important, clear-cut and perhaps more-promising target for specific interventions.

Mean daily usage, the percentage of days used and the proportion of individual sleep time spent with adequate mask pressure were higher for up to 20% in the intensive support group, when analysed on an intention-to-treat basis. However, a possible limitation of these data might be a conservative missing value imputation, which could increase the statistical error.

Notably, even the “standard support” applied to the control group was more intensive than common support at the sleep unit with an explanation about the disease and its possible sequels, a thorough introduction to the device and equipment, and a brief diurnal CPAP trial before the second polysomnography. This effect may account for part of the high overall compliance, compared with literature 14 and has already caused difficulties in observing differences in several interventional CPAP trials 18, 22–24.

Another common problem in interventional CPAP trials is that not specific interventions but rather “packages” 25 are studied, so that the most relevant measure cannot be identified distinctively. Hoy et al. 18 used an intensive education, a 3-night CPAP trial and home nursing. They reported a 40% higher CPAP use with intensive support, sustained for over six months, in terms of usage in h·night⁻¹. Percentage of days or other parameters of usage were not monitored. Jean Wiese et al. 19 showed a higher retention rate for 1-month follow-up after video education with 72.9 versus 48.9% in the control group. Chervin et al. 16 and colleagues performed education, telephone calls and letters. The present study shows the effectiveness of regular home visits in the first 6 months from therapy initiation. It could not be clarified, in this study, whether the higher frequency of encounters or the home setting led to a higher adherence to CPAP. A combination of home visits at therapy initiation for a shorter period than in the present study, followed by more frequent follow-ups at the sleep units, might be a feasible alternative strategy.

Other measures, such as optimising mask-fit and educational support, were provided to patients in both groups, as mentioned above, although more frequently to the intensive support group. This may also have led to higher material costs in some patients of that group, contributing, alternatively, to their acclimatisation to CPAP. This suggests that, even if inferior to psycho-educational measures, mechanical interventions in terms of equipment optimisation (humidifiers, different interfaces, etc.) at therapy initiation should also be employed when necessary, in order to prevent impairment of CPAP adherence.

In the present authors’ experience, increased mask discomfort and equipment requirements, resulting in frequent interface change at therapy initiation, are the manifestation of difficulties in acclimatisation to CPAP and coping, in some patients. Intensive support, addressing the underlying problems and helping the patient to achieve a positive “cost–benefit calculation”, despite all discomfort, will help to reduce these material costs. These savings, however, could be allocated to personal costs again.

Additional savings could be achieved by the delay of definite purchase of CPAP devices by health insurers until proper assessment of patient compliance.

Therapy efficiency should be monitored closely, using data storage in CPAP devices and personal supervision. This will help to identify patients at risk of abandoning CPAP and to intervene, if necessary. Therefore, current considerations to simplify treatment and follow-up to save money in the short term cannot be supported from this “compliance” point of view.

As to the measurement techniques applied, the set of different parameters used in the present study allows for a more-detailed analysis of different usage patterns and, therefore, yields a closer description of compliance as a construct.

Clinical outcome parameters, i.e. the AHI, as well as ESS Scores tended to decrease continuously. This suggests some curative aspects of CPAP in the long term because the values used were measured at defined time points, which are not directly influenced by usage rates. Mean values over time could help in detecting a true continuous improvement. One possible explanation for such an effect is the reduction of mucosal oedema by minimising vibrational trauma, which is known to account for upper airway collapsibility, in part 26. Another possible explanation is a reduction in sleep fragmentation, reflected by a further decrease in the arousal-index in the present study; sleep fragmentation is known to increase upper airway collapsibility 27, 28. A decrease in daytime somnolence may also be due to a better acclimatisation to CPAP and therefore less sleep fragmentation. The BMI did not change significantly in the course of the study, so weight loss cannot explain the improvements noticed in this case.

The reduction of the AHI appeared to be stronger with intensive support. Here, a direct effect of the intensive support must be postulated, rather than a statistical one, through a higher compliance. In the present authors’ opinion, optimising the equipment, such as mask fit, minimising side-effects such as rhinorrhoea by the use of humidifiers, an early identification of patients with low compliance and again a better acclimatisation to CPAP accounted for a better quality and effectiveness of therapy, and in terms of clinical outcome. ESS mean values at baseline were significantly different, showing a possible bias for compliance outcome. However, they were comparatively low in both groups, indicating that overall symptom severity was not as evident as the high RDI would have suggested. Additionally, they decreased in the same manner in both groups. Therefore, even if slightly increased or normal ESS values were considered to influence CPAP therapy adherence 11, this did not influence compliance outcomes in the current study.

The clinical significance of these improvements remains unclear. A decrease in the AHI may well reflect amelioration of disease severity and hence its cardiovascular sequels through CPAP, as has been reported 3, 6. Hoy et al. 18 also demonstrated a reduction in sleepiness, better mood and improvements in reaction time. However, these parameters mainly pertain to quality-of-life issues. Parameters more related to intermittent hypoxia and its sequels, such as the AHI, were not discussed. Therefore, more studies on long-term cardiovascular outcomes, also considering costs arising from an increased morbidity, mortality and occupational injuries are needed to confirm relevant benefits from a higher therapy adherence and an intensive support, respectively. A consistent definition of effective CPAP use, in terms of usage in h·night⁻¹, percentage of days etc., is still lacking; “how much is enough?” 29 still remains an important question.

Secondly, APAP and CPAP were compared. P_mean was about 1 cmH₂O lower after 3 and 9 months, as has been reported for APAP previously: Teschler et al. 30 found a 23% reduction in APAP compared with CPAP.

Variation of pressure in the APAP device did not cause higher air-leaks or more arousals during sleep, i.e. the Arousal-Index was similar in both groups. However, the association between arousals and single pressure changes was not analysed. Fuchs et al. 31 found that an average of 20% of microarousals (MA) were preceded by a significant change in pressure (at least 0.5 mbar within 30 s.) delivered by APAP but the relative amount of “pressure-associated MA” was not significant in most individuals 31.

The lower P_mean among APAP patients did not lead to a higher compliance or therapy adherence, respectively. Patients used their device for more than 5 h·day⁻¹ and for about 75% of the days during the first 3 months in both groups, retention rate was even lower under APAP.

In their comparison between APAP and CPAP, Randerath et al. 15 also found no difference in compliance despite a lower P_mean under APAP. The duration of use was 5.25 h·day⁻¹. However, 75% of the patients preferred APAP for long-term treatment at home. Konermann et al. 32 found a higher amount of nights·week⁻¹ with nasal CPAP use for >4 h with APAP, after 6 months of therapy but without a significant difference in daily use. The authors themselves state that the slightly higher patient compliance with the self-adjusting device needs further confirmation, which could not be supported by the present study. Additionally, in a recent meta-analysis by Ayas et al. 33 a similar patient adherence for APAP and CPAP was found. The specific reasons for subjective preference of APAP remain unclear. The effective reduction in P_mean is a possible explanation but it has to be questioned whether patients have registered this reduction directly; the possibility would increase with a greater inter-individual variability of pressure throughout the night. An indirect benefit by the reduction of pressure- or airflow-associated side-effects can be assumed as a factor of influence but these questions were not evaluated in the current study, so far.

As to the clinical outcomes, both CPAP and APAP were equally effective in reducing the AHI, compared with the AHI at diagnosis, with an even further decrease in the long term. Concurrently, daytime sleepiness was reported to decrease significantly.

Autoadjusting continuous positive airway pressure as a mechanical type of intervention, showed no relevant impact on compliance or clinical outcomes in the therapy of obstructive sleep apnoeas syndrome compared with continuous positive airway pressure. Therefore, a switch to autoadjusting continuous positive airway pressure devices, which are more expensive currently, cannot be recommended in general, but to a more-selected group of patients might benefit from an automatic adjustment of pressure. As a result, more-precise differential indications for the use of autoadjusting continuous positive airway pressure need to be identified. Furthermore, even the use of autoadjusting continuous positive airway pressure for unattended therapy initiation at home needs to be questioned because the setting at the time of diagnosis and start of therapy (sleep unit or patients' home) also has shown to be a predictor of long-term compliance 34, 35. However, some costs may be reduced by using autoadjusting continuous positive airway pressure for attended autotitration for determination of fixed continuous positive airway pressure.

Statement of interest

Statements of interest for M. Idzko and the study itself can be found at www.erj.ersjournals.co.uk/misc/statements.dtl