Sleep Apnea May Reduce Antidepressant Response, Increase Dementia Risk

NEW ORLEANS ― Obstructive sleep apnea (OSA) may have strong influence on antidepressant response and increase the risk for mild cognitive impairment in older patients with major depressive disorder (MDD), two new studies suggest.

Lauren Waterman

The first, an open-label trial involving more than 400 patients with MDD, showed that those with comorbid OSA were 1.5 times less likely to respond to 12 weeks of treatment with the antidepressant venlafaxine (multiple brands) than those without OSA.

Participants with both MDD and OSA also reported more severe and longer episodes of depression, decreased physical functioning, and lower quality-of-life scores.

"This shows that clinicians should definitely screen for sleep apnea when patients are presenting with late-life depression," lead author Lauren Waterman, from the University of Pittsburgh School of Medicine, Pennsylvania, told Medscape Medical News.

The second study, which shared first prize in the Early Investigator Poster competition at the meeting, showed that patients with MDD and a high probability of OSA had significantly lower cognitive performance than those with moderate to low probability of OSA.

"Both studies show that it's not a good thing to have sleep apnea," said Iqbal "Ike" Ahmed, MD, clinical professor of psychiatry and geriatric medicine at the University of Hawaii at Manoa and from the Department of Psychiatry at Tripler Army Medical Center, when asked for comment.

"One showed that it may influence response to antidepressants, and the other that it may block improvement in cognition ― even if there is some improvement in mood," added Dr Ahmed.

Both studies were presented here at the American Association for Geriatric Psychiatry (AAGP) 2015 Annual Meeting.

Lasting Damage

Waterman noted that previous research has shown that treating OSA in younger patients with MDD can improve response to antidepressant medications. However, there has not been a focus on whether this also occurs in older patients ― even though 1 in 5 older adults have both MDD and OSA.

"Sleep apnea has the potential to cause lasting damage in older adults suffering with late-life depression. Because the symptoms are very similar across the two conditions, we wanted to do this study," she said.

A total of 468 adults older than 60 years with MDD who had a score of at least 15 on the Montgomery-Åsberg Depression Rating Scale (MADRS) were enrolled at one of three centers.

At baseline, demographics and physical data were collected, and polysomnography was used to determine OSA status. A total of 17.1 of the MDD patients (n = 80) had comorbid OSA.

For the first 6 weeks, all participants received venlafaxine, titrated up to 150 mg/day; this continued for an additional 6 weeks for responders. The dose was increased to 300 mg/day for the last 6 weeks for nonresponders.

Treatment response was defined as achieving a score of 10 or lower on the MADRS for two consecutive visits. A "composite insomnia index" was also created; its use was based in part on questions from the 17-item Hamilton Rating Scale for Depression (HRSD-17).

Results showed that 27.9% of the patients with both MDD and OSA had a response to venlafaxine vs 43.4% of those without OSA (adjusted hazard ratio, 1.69; 95% confidence interval [CI], 1.07 - 2.66).

Screening Important

At baseline, the OSA group had significantly higher mean body mass index (BMI) (34.1 vs 29.0, respectively) and more physical comorbidities than the non-OSA group (for both, P < .001).

Significantly more members of this group were also obese (P < .001), had hypertension (P < .001), and had uncontrolled glucose levels (P = .04). They also had lower scores on the Health-Related Quality of Life MOS Physical scale (P = .01).

Interestingly, although the group with OSA had greater depressive symptom severity on the MADRS (27.31 vs 26.54, respectively; P = .03) and longer duration of current episode (P = .002) than the non-OSA group at baseline, they were more likely to report being treated adequately for the episode on the Antidepressant Treatment History Form (71.8% vs 59.1%, respectively; P = .02).

In addition, "the HRSD insomnia index did not predict time to response, nor was there a significant HRSD insomnia and apnea interaction," report the investigators.

Waterman noted that this could be because the primary complaint from most patients with sleep apnea is not insomnia but rather daytime sleepiness. "So in future studies we'd like to find a better measure, such as the Stop-Bang Questionnaire," she said.

Overall, "physicians should consider screening for OSA when assessing older adults for depression, as it may indicate a more difficult to treat depression," write the investigators.

Dr Iqbal Ahmed

"I don't think the take-home message is that surprising. What they're saying is that if you have sleep apnea, you'll be less likely to respond to antidepressants," commented Dr Ahmed, who was not involved with this research.

"This has to do with vascular depression. So if you have sleep apnea, you probably have vascular changes in your brain," he added.

Cognitive Impairment

The second study, which was led by Nancy Kerner, MD, from Columbia University, in New York City, examined data on 25 patients (52% men; mean age, 70.6 years) with MDD and cognitive impairment, as shown by scores of less than 11 on the Logical Memory II subscale. All participated in a 16-week antidepressant study that screened for possible OSA.

Dr Nancy Kerner

The patients were given 10 to 20 mg/day of citalopram for the first 8 weeks. This regimen was continued for the last 8 weeks by the responders ― those who had a 50% reduction in baseline HDRS score. The nonresponders were switched to a different antidepressant of their clinician's choice for the last 8 weeks.

The Stop-Bang Questionnaire was used to screen for probability of OSA. Ten of the participants were considered to have a high probability of OSA, determined on the basis of a Stop-Bang score of at least 5. Those with a score of 4 or less were considered to have a moderate to low probability of OSA (n = 15).

In addition to the 24-item HDRS to measure depression severity, Stroop subtests were used to measure executive function, and subtests from the Selective Reminding Test (SRT) were used to assess memory functions. Neuropsychological tests were administered to examine cognitive performance.

Results showed that the group with a high probability of OSA had lower scores on all cognitive tests than those with a lower probability of OSA. Although HDRS and Stroop scores improved from baseline to the 16-week mark for both groups, the 16-week HDRS scores were significantly lower in those with probable OSA (P < .05).

In addition, SRT delayed recall significantly improved in the group with a moderate to low probability of OSA (P < .02); it decreased for the high-probability group (P < .02 for between-group difference).

Although the mean HDRS baseline score was lower for the probable-OSA group than for the lower probability group (20.9 vs 24.5), the difference was not significant. However, at baseline, the probable-OSA group had a significantly higher BMI (P < .001) and a higher percentage of obesity (P < .05) and hypertension (P < .02).

Treating Depression Not Enough

"Further studies are needed to elucidate whether OSA is a risk factor for rapid cognitive decline in older adults with both depression and cognitive impairment, and whether improvement of oxygenation during sleep in these patients can reverse existing cognitive deficits," write the investigators.

In a comment to Medscape Medical News, Dr Kerner said that clinicians should consider using the Stop-Bang for all their older patients. "It's so easy, asks yes-no questions, and takes less than 2 minutes."

Dr Ahmed noted that the results of this study were not exactly a surprise, because sleep apnea has been shown before to be associated with cognitive impairment and mood changes.

"So there's value in asking about sleep changes in patients of an older age or with a certain body type, or with daytime somnolence," he said.

"The first study showed less antidepressant response in those with sleep apnea. The second paper said you may get some response, but it's not very robust. And although there might be improvement in mood, you don't get improvement in cognition," said Dr Ahmed.

"The take-away is that you need to not just treat the depression. It's important to assess patients at high risk for sleep apnea, especially in those with unresponsive depression."

The study authors and Dr Ahmed report no relevant financial relationships.

American Association for Geriatric Psychiatry (AAGP) 2015 Annual Meeting. Abstracts EI-19 and EI-20. Presented March 28, 2015.