The Rest of the Story

Abstract:
Objective: Patients with obstructive sleep apnea (OSA) may experience unfavorable psychologic symptoms such as depression and anxiety. The aim of this study was to confirm this hypothesis and to investigate whether the psychologic symptoms among OSA patients can be relieved by surgical intervention.

Study Design: Prospective, longitudinal intervention study.

Methods: The 5-Item Mental Health scale (MH-5) was used to evaluate the postoperative changes of mood after extended uvulopalatal flap (EUPF) surgery on 84 Taiwanese patients with OSA. The preoperative and postoperative MH-5 data obtained from these patients were compared with a Taiwanese population norm. 

Results: Before surgery, the MH-5 scores of the OSA patients were significantly worse than the Taiwanese population norm of 72.8 (P < .0001). Postoperatively, mean MH-5 scores significantly increased from 61.8 +/- 16.0 to 70.0 +/- 15.8 (P = .0006). The effect size of this score change was 0.51, indicating a moderate degree of mood improvement. However, this score was still inferior to that of the population norm (P = .0045). The mood improvement was not significantly associated with the changes in either sleep apnea events or the level of sleepiness. Neither the changes in respiratory disturbance index (P = .4382), maximum arterial oxygen saturation (P = .4866), nor the change in Epworth Sleepiness Scale scores (P = .4951) were predictive of the MH-5 score improvement (R2 = 0.07).

Conclusions: This study demonstrated that patients with OSA had a higher level of anxiety, depression, and probable behavior or personality changes than the population norm. EUPF surgery could significantly improve the mood status among OSA patients; the effect of surgery was mild but clinically relevant. However, the extent of mood improvement experienced by OSA patients receiving operations may not simply be attributable to the changes in sleep apnea events or a reduced level of sleepiness.

(C) The American Laryngological, Rhinological & Otological Society, Inc.

It makes sense to speculate that interruptions in sleep and decreases in blood oxygen could result in anxiety and/or depression.  However, if it were that simple, one would expect there to be a more or less linear correlation between the two.   At the very least, we would expect to see a relationship between the overall level of sleepiness and the psychiatric symptoms.  However, the Epworth scale improvements were not related to the improvements in anxiety and depression. 

I am not familiar with the instrument they used to assess psychiatric symptoms: the MH-5.  It is not widely used in psychiatry, so there is a chance that it is not really a valid measure of clinical syndromes.  Taking it at face value, though, assuming the MH-5 actually means something, it would be of great interest to look at the individual patients, see which ones got the greatest mental health boost from the surgery, and try to figure out what is different about them, as compared to the patients who had the surgery but did not have improvements in mood and anxiety. 

The reason this is important is that psychiatric disorders are defined by symptoms, not by underlying causes.  This is a known problem in psychiatric research.  Everyone in the field wants to come up with so way of subdividing the population of depressed patients into groups that would correlate with treatment responses.  That way, we would not end up given treatments that are not likely to work, to patients who really need some other kind of treatment. 

With the exception of atypical depression (which responds preferentially to MAOI's) and psychotic depression (which responds preferentially to a combination of an antidepressant and an antipsychotic), we have not had much success in subdividing the types of depression into clinically relevant groups.  There is hope that functional neuroimaging  -- fMRI and PET -- could be useful in this regard.  So far, however, all we have is tantalizing clues; nothing really useful has come of it so far.  If there is a relationship between sleep architecture  and subtypes of depression, that would be good to know.