Co-morbid OSA and Depression: What We Know and What We Need to Know
Obesity is a common medical condition complicating psychiatric illnesses. Obstructive sleep apnea is a closely related disorder to obesity with hazardous consequences in terms of both morbidity and mortality. In this article, we have made a qualitative search of both primary studies and systematic reviews pertaining to co-morbid depression and OSA. Under a structured format, the article discusses the epidemiology, pathophysiology, therapeutic significance of the association followed by a conclusion for clinical, as well as research approaches to co-morbidity.
KeywordsObstructive sleep apnea Sleep apnea Apnoea Depression MDD Depressive Co-morbid Association Link
Obstructive sleep apnea is a chronic sleep disorder with 5 or more apnea or hypopnea per hour of sleep occurring due to collapse of upper airway despite the presence of central ventilatory drive . It is usually a result of obesity or craniofacial malformations. Depression, on the other hand, is a psychiatric disorder of similar lifetime risk (8–12% in most populations), and multi-factorial origin with strong genetic and environmental predispositions . OSA and depression have considerable overlap in both day-time and night-time symptomatology. Because there is a biological plausibility in the association of the two conditions which we shall delve in this article, chances of drawing spurious associations or prescribing inaccurate treatment are both possible. There have been studies looking into the relationship between the two conditions. In this article, we have summed up available literature to provide insight into the relationship between the two conditions, so that clinical implications for sleep physicians, psychiatrists, psychologists, and general practitioners can be drawn.
OSA has grown into a common health condition by the turn of the last century [3, 4, 5]. There has been an exponential growth in obesity all over the globe. As a result of multiple etiological factors, the prevalence of obstructive sleep apnea is also on a rise . Few population-based studies place the prevalence of OSA at 4–7% for men and 2–4% for women [3, 4]. An Indian study reports OSA prevalence to be as high as 7.5% in adult urban population . The portion of OSA patients remaining untreated or undertreated is more than half. The age distribution of OSA shows gradual increase in incidence with age, peaking after 60 years of age . Years may pass from the time of symptom onset before a diagnosis of OSA is made in most patients. These patients may suffer from depression over time due to increasing constrains which they find in day-to-day living as well as due to metabolic derangements and other socio-economic factors.
2.1 Depression in OSA
An observational study in China (n = 1327 OSA patients) showed prevalence of depression as high as 47% on Symptom Checklist 90 and Self-Rating Depression Scale. The study revealed that depressive status is significantly associated with single status, high AHI, i.e., OSA, and low family and social supports among those OSA patients.
In another study of newly diagnosed OSA, patients (n = 121) screened for depression, and prevalence of 44.6% on Beck Depression Inventory (BDI) score (>10) was seen . Another retrospective study of 100 OSA patients puts the proportion to 19% . This tendency is adequate a reason for sleep specialists to screen for depression in an otherwise clear-cut OSA case. However, a causal relationship has not been established as further studies have shown that AHI values from sleep studies have not been shown to correlate with BDI scores (for depression) . Therefore, only future studies can clarify whether or not a causal relationship exists in this significant association.
2.2 OSA in Depression
The prevalence of OSA in patients of major depressive disorder (MDD) has consistently been reported to be high. In a recent systematic review for prevalence of OSA among psychiatry patients, data were pooled from six clinical studies (n = 522) with MDD . A pooled prevalence of OSA of 36.3% (95% CI 19.4–57.4, p < 0.01) was found. After excluding two studies [10, 11] with potential selection bias, the pooled prevalence of OSA among 378 people with MDD was recalculated at 25.2% (95% CI 10.2–50.2, Q = 58, p < 0.01). This was still higher than that reported for Schizophrenia and BPAD (24.5%, 95% CI 10.6–47.1, n = 681) in the same review. The increased association of OSA in BPAD has also been ascribed to the sleep disturbances seen in both disorders . In the review, an AHI/RDI > 5 was taken as diagnostic for OSA, while MDD was diagnosed based on DSM-IV (2000) or ICD-10 (1993), all being standard nomenclature. Increasing age and BMI were associated with OSA more significantly in these MDD patients on pooled analysis in the review.
3 Possible Mechanisms for Association
3.1 Theory of Sleep Fragmentation and Hypoxia
Beebe and Gozal  came up with a theory that sleep fragmentation and hypoxia during sleep cause disruption of sleep-related restorative effects on the prefrontal cortex. This, in turn, may result in ‘executive dysfunction,’ which refers to the abilities crucial for organization, planning, and adaptation . In addition, a functional magnetic resonance imaging (fMRI) study has shown that OSA is associated with reduced lateral prefrontal activation during working memory tasks . Similarly, functional neuroimaging abnormalities within the prefrontal cortex have also been associated with Major Depressive Disorders [15, 16].
The association of sleep fragmentation with depression has been evaluated in a study. This study reported a significant increase in major flow limitations and desaturations in 19 depressed patients (compared to control subjects) selected without regard to OSA symptoms . The authors hypothesized that respiratory-related sleep fragmentation of hypoxia-induced prefrontal dysfunction may predispose to mood disorder. However, the lack of correlation of parameters of OSA and depression, i.e., AHI and BDI, respectively, as shown in few studies argues against a causal relation .
3.2 Overlapping Symptomatology
Overlap in symptomatology between OSA and depression
Symptoms of depression (ICD-10)
Same symptom with respect to an OSA patient
Marked loss of interest or pleasure in activities that were normally pleasurable
In OSA, this is an obvious sign due to inability to concentrate, sleepiness, and fatigue
Absence of emotional reactions to events that normally produce a response
Sleepiness in inappropriate places can be interpreted as emotional apathy
Waking in the morning 2 h or more before the usual time
Frequent sleep fragmentation can lead to early departure from bed
Depression worse in the morning
Patient waking up with extreme fatigue, bodyache, headache can be interpreted as signs of depression
Evidence of psychomotor retardation or agitation
Poor quality of sleep leads to agitation or retardation in OSA
Marked loss of appetite
Weight loss of at least 5% in the last month
Notable decrease of libido
Also present in OSA
Considering the above scenario, patients with OSA syndrome may satisfy many somatic criteria for depression and vice versa, hence, an association.
3.3 Socio-economic and Metabolic Causes
OSA patients may suffer from cognitive and memory decline along with constant day-time sleepiness and fatigue. Metabolic profile of untreated OSA patients shows higher than normal prevalence of diabetes, dyslipidemia, and hyperuricemia. There is also an increased incidence of hypertension and adverse cardiac events, heart failure, and stroke in such patients. With these physical and mental disabilities, obesity with general inactivity, and their financial repercussions, OSA patients are under some psychological stress beyond normalcy. Biochemically, there is a constant rise in inflammatory cytokines like Il-6, and TNF alpha in both OSA and depression [19, 20]. This may lead to increased incidence of depression over long term. On the other hand, depression is associated with incident and prevalent weight gain, especially in women . Many patients also gain a significant amount of weight due to use of tricyclic antidepressants, SSRIs, or antipsychotics use. These patients can develop OSA while being on treatment. In this context, another hypothesis that OSA and depression both have independent correlation with obesity (or metabolic syndrome) could explain their mutual correlation without dose–response relationship .
4 Therapeutic Trials on Management of Co-morbid OSA and Depression
4.1 Primarily OSA with Depressive Symptoms
There have been some prospective cohort studies and fewer RCTs evaluating role of CPAP on depressive symptoms in OSA patients. A systematic review evaluated 26 of these open labelled studies which showed evidence in favour of improvement in depressive symptoms with adequate PAP therapy . The evidence cannot be considered of high quality or of high effect size because of absence of double-blinded placebo controlled trials. In addition, a significant minority of trials still show no benefit of PAP in terms of depressive symptoms [24, 25, 26, 27].
In patients with residual sleepiness despite adequate management of OSA, it is a usual practice among sleep physicians to screen for depression if not done initially followed by behavioural and pharmacological treatments.
4.2 Primarily Depression with OSA
Though there is scarcity of studies addressing OSA in common cases of depression, some studies in resistant depression have addressed OSA. In a study of 118 consecutive treatment resistant depression cases in Turkey, 27% were already diagnosed OSA and 11% more were found to be high risk for OSA . These 38% patients benefitted significantly with treatment for OSA in terms of depressive symptoms. The authors concluded that in cases of treatment resistant depression, especially with BMI ≥ 30 kg/m2, presence of hypertension, and/or dyslipidemia, OSA should be positively screened for. However, placebo effect cannot be ruled out. Nevertheless, treatment was effective and hazards of OSA were also managed which justifies screening and management for OSA in these subset of depressed patients.
5 Summary of Evidences
There is increased prevalence of OSA in Depressive disorder and vice versa suggestive of a bidirectional association
There exists no dose–response relationship to signify a causal relationship
Treating for OSA may improve depressive symptoms significantly in most prospective cohort and pre-post studies but not in sham-controlled trials. This may imply either for a strong placebo effect in depression or hint towards depressive symptoms being a part of OSA itself, primary depression not being associated altogether.
OSA and depression have a complex etiological and demographical overlap with similar symptomatology. At this point, we know that both the conditions have an association that requires enquiring for the other in the presence of one. We do not yet know the exact reason for the significant association. Most likely, it is a multi-factorial association with each of the above hypothesis being applicable in various subsets of patients. It is, therefore, necessary to appreciate the bidirectional relationship and be vigilant at both ends to quickly diagnose and treat the co-morbidities.
6.1 Research Perspective
Studies showing clear treatment benefit in OSA complicated by depression are lacking. Hence, double-blinded studies comparing effects of PAP therapy with placebo versus PAP therapy with antidepressants in patients of OSA with co-morbid depression may provide insight into the therapeutic implications of this association. The outcomes to look for in these studies should be time to and extent of restoration of EDS, fatigue and other symptoms, safety outcomes like weight gain, and compliance to treatment. The study should stratify cases of depression based on whether they are (1) primary depression, (2) reactive to OSA-induced psycho-social stressors, (3) severe OSA-induced excessive sleepiness and cognitive impairment, or (4) other stressor induced causes. Only with all this information, we can have a clearer idea into depression complicating OSA.
More studies to look for OSA (with appropriate screening questionnaires followed by Sleep study) in new and treatment resistant patients of depression should be carried out to identify significant predictors of OSA in depressed patients.
Studies should also focus on protocol to identify and manage depression-induced residual sleepiness in OSA patients on PAP therapy. Likewise, sham-controlled RCTs evaluating PAP therapy on resistant depression cases with mild OSA or flow-limitation will help to identify the importance of screening for sleep-disordered breathing in the patient group.
6.2 Clinical Management Perspective
An OSA patient with co-morbid depression is unlikely to show resilience to withstand fear of PAP therapy. He/she may soon become non-compliant to PAP therapy. Therefore, in a vigilant search for depressive symptoms, arrangement of psychological evaluation followed by watchful waiting in mild cases to counselling/drug therapy based on psychiatry opinion is prudent.
From psychiatry practice point of view, a patient identified with depression only and untreated for OSA may not gain any respite from his fatigue and lack of interest, but rather suffer from side effects of TCAs or SSRIs. Likewise, a previously responding patient of depression may become resistant to drug treatment after significant weight gain and development of OSA. Therefore, patients with depression and excessive day-time sleepiness or fatigue must be screened with OSA screening questionnaires. Appropriate referrals for sleep study should be made based on the same. Periodic reassessment for OSA is required because of the dynamics of obesity in these cases.