Posttraumatic stress disorder (PTSD) may enhance the risk for obstructive sleep apnea (OSA) in older male veterans, the results of a cross-sectional twin study suggested. However, additional high-quality research is needed and may yield important mechanistic insights into both conditions and improve treatment, experts said.
“The strength of the association was a bit surprising,” said study investigator Amit J. Shah, MD, MSCR, Emory University, Atlanta, Georgia. “Many physicians and scientists may otherwise assume that the relationship between PTSD and sleep apnea would be primarily mediated by obesity, but we did not find that obesity explained our findings.”
The study was published online in JAMA Network Open.
A More Rigorous Evaluation
“Prior studies have shown an association between PTSD and sleep apnea, but the size of the association was not as strong,” Dr. Shah said, possibly because many were based on symptomatic patients referred for clinical evaluation of OSA and some relied on self-report of a sleep apnea diagnosis.
The current study involved 181 male twins, aged 61-71 years, including 66 pairs discordant for PTSD symptoms and 15 pairs discordant for PTSD diagnosis, who were recruited from the Vietnam Era Twin Registry and underwent a formal psychiatric and polysomnography evaluation as follow-up of the Emory Twin Study.
PTSD symptom severity was assessed using the self-administered Posttraumatic Stress Disorder Checklist (PCL). OSA was mild in 74% of participants, moderate to severe in 40%, and severe in 18%.
The mean apnea-hypopnea index (AHI) was 17.7 events per hour, and the mean proportion of the night with SaO2 less than 90% was 8.9%.
In fully adjusted models, each 15-point within-pair difference in PCL score was associated with a 4.6 events-per-hour higher AHI, a 6.4 events-per-hour higher oxygen desaturation index, and a 4.8% greater sleep duration with SaO2 less than 90%.
A current PTSD diagnosis is associated with an approximate 10-unit higher adjusted AHI in separate models involving potential cardiovascular mediators (10.5-unit; 95% CI, 5.7-15.3) and sociodemographic and psychiatric confounders (10.7-unit; 95% CI, 4.0-17.4).
The investigators called for more research into the underlying mechanisms but speculated that pharyngeal collapsibility and exaggerated loop gain, among others, may play a role.
“Our findings broaden the concept of OSA as one that may involve stress pathways in addition to the traditional mechanisms involving airway collapse and obesity,” Dr. Shah said. “We should be more suspicious of OSA as an important comorbidity in PTSD, given the high OSA prevalence that we found in PTSD veterans.”
Questions Remain
In an accompanying editorial, Steven H. Woodward, PhD, and Ruth M. Benca, MD, PhD, VA Palo Alto Health Care Systems, Palo Alto, California, noted the study affirmatively answers the decades-old question of whether rates of OSA are elevated in PTSD and “eliminates many potential confounders that might cast doubt on the PTSD-OSA association.”
However, they noted, it’s difficult to ascertain the directionality of this association and point out that, in terms of potential mechanisms, the oft-cited 1994 study linking sleep fragmentation with upper airway collapsibility has never been replicated and that a recent study found no difference in airway collapsibility or evidence of differential loop gain in combat veterans with and without PTSD.
Dr. Woodward and Dr. Benca also highlighted the large body of evidence that psychiatric disorders such as bipolar disorder, schizophrenia, and, in particular, major depressive disorder, are strongly associated with higher rates of OSA.
“In sum, we do not believe that a fair reading of the current literature supports a conclusion that PTSD bears an association with OSA that does not overlap with those manifested by other psychiatric disorders,” they wrote.
“This commentary is not intended to discourage any specific line of inquiry. Rather, we seek to keep the door open as wide as possible to hypotheses and research designs aimed at elucidating the relationships between OSA and psychiatric disorders,” Dr. Woodward and Dr. Benca concluded.
In response, Dr. Shah said the editorialists’ “point about psychiatric conditions other than PTSD also being important in OSA is well taken. In our own cohort, we did not see such an association, but that does not mean that this does not exist.
“Autonomic physiology, which we plan to study next, may underlie not only the PTSD-OSA relationship but also the relationship between other psychiatric factors and OSA,” he added.
The study was funded by grants from the National Institutes of Health (NIH). One study author reported receiving personal fees from Idorsia, and another reported receiving personal fees from Clinilabs, Eisai, Ferring Pharmaceuticals, Huxley, Idorsia, and Merck Sharp & Dohme. Dr. Benca reported receiving grants from the NIH and Eisai and personal fees from Eisai, Idorsia, Haleon, and Sage Therapeutics. Dr. Woodward reported having no relevant conflicts of interest.
A version of this article first appeared on Medscape.com.