A preliminary study of sleep-disordered breathing in major depressive disorder

Abstract

Background and purpose

Individuals with obstructive sleep-disordered breathing (OSDB) commonly report symptoms of depression; however, the percentage of individuals with major depressive disorder (MDD) who experience OSDB is less clear. This study aimed to examine OSDB in a sample of individuals with MDD, unselected for sleep-related complaints, along a continuum of ventilatory and hypoxic abnormalities.

Patients and methods

The overnight sleep-related breathing of 19 individuals with MDD and 15 non-depressed controls was recorded using an unattended nasal pressure-based home sleep monitoring device. The device recorded nasal airflow, breathing effort, heart rate, oxygen saturation, and body position.

Results

The two groups varied significantly on three sleep-related breathing variables: major flow-limitation events, major flow-limitation events accompanied by a desaturation, and average saturation throughout the evening; and these groups approached significance on minor flow-limitation events accompanied by a desaturation and average number of desaturations throughout the evening. Sleep-related breathing variables predicted accurate grouping in 81.3% of those with MDD and 80.6% of the non-depressed participants.

Conclusions

These results suggest that OSDB may play a more important role in MDD than previously recognized. OSDB may contribute to or exacerbate the condition of someone predisposed to MDD, and the treatment of OSDB may ameliorate or possibly prevent depressive symptoms.

Introduction

Numerous studies demonstrate a high rate of depressive symptoms in individuals with obstructive sleep-disordered breathing (OSDB) [1]. The interaction of OSDB and mood is reciprocal and complex [2], [3]. In randomized controlled trials of positive airway pressure, pre-treatment depressed mood reportedly improved [4]. A population study using the Sleep-EVAL expert system questionnaire reported that after controlling for obesity and hypertension, the odds of having a Diagnostic and Statistical Manual of Mental Disorders—Fourth Edition (DSM-IV) breathing-related sleep disorder diagnosis was 5.6 for individuals with a major depressive disorder [5]. Two decades ago, a report [6] found a 15.1% prevalence rate of at least ‘some degree of sleep apnea’ (p. 566) within a depressed group of 86 inpatients. The authors concluded that the percentage is neither clinically significant nor an indication for routine OSDB screening. This conclusion may be premature given the following: (1) OSDB was described as present or absent, rather than studied along a continuum of disordered breathing events, (2) groups included a heterogeneous in-patient population that included individuals with unipolar major depression, bipolar, and schizoaffective disorder, (3) the study lacked a non-depressed control group, (4) the study employed thermistors to monitor airflow, an inaccurate method for detecting partial airflow obstructions (hypopneas) [7], and (5) only apneas were scored.

Patients with OSDB experience repetitive episodes of complete or partial airway obstruction, coupled with variable degrees of intermittent oxygen desaturation. The consequent sleep fragmentation may result in excessive daytime sleepiness, executive dysfunction and mood impairment [8]. Other symptoms include restless sleep, snoring, morning headaches, sexual dysfunction, irritability, impatience, and anxiety [9], [10], [11]. With the exception of snoring, all of the above symptoms have also been associated with MDD [12], [13], [14]

Deficits in concentration, working memory, and executive cognitive functioning are common to OSDB [15], [16] and MDD [17], [18], [19]. Auditory tone-induced sleep fragmentation causes changes in memory, attention and mood [20]. Beebe and Gozal (2002) theorize that sleep fragmentation and hypoxia during sleep lead to a disruption of sleep-related restorative effects on the prefrontal cortex. In turn, this may result in ‘executive dysfunction,’ which refers to the capacities crucial for organization, planning, and adaptation [21]. In addition, functional magnetic resonance imaging (fMRI) results suggest that OSDB is associated with reduced lateral prefrontal activation during working memory tasks [22]. Functional neuroimaging abnormalities within the prefrontal cortex have also been associated with MDD [24], [25].

Symptoms of depression in patients with OSDB may improve following treatment with continuous positive airway pressure [26], [27], though this has not always been observed [28]. The use of the wake-promoting drug Modafinil improves a range of cognitive symptoms and sleepiness in sleep apnea [29] and preserves prefrontal activation on working memory tasks following overnight sleep deprivation [30]. This drug is also effective as an augmenting agent in depression [31]. In summary, several clinical, neuropsychological, and functional neurobiological phenomena are common to OSDB and MDD, and treatment of sleep-disordered breathing has lead to a reduction in both depressive symptoms and neuropsychological impairment.

The purpose of the present study was to explore whether there was abnormal respiration and nocturnal oxygen desaturation within a group of individuals diagnosed with MDD. Unlike prior work on the relationship between MDD and OSDB, which has involved examining the number of individuals with MDD who meet certain limited criteria, the current study measures the relationship between MDD and OSDB along a continuum of ventilatory and hypoxic disturbances. A study with this methodology allows for an exploration of smaller, yet potentially significant, sleep-related breathing and desaturation phenomena. Our sample was comprised entirely of outpatient, unipolar-depressed individuals. In addition, a control group was included, providing critical normative data for the technique used to monitor airflow, a nasal cannula-pressure transducer technique; this method has been found to be a more accurate measure of abnormal sleep-related respiration than thermistors [7], including when it is used without sleep monitoring [32].