When Insomnia Meets Sleep Apnea: The Hidden Diabetes Link Scientists Are Chasing

For years, clinicians have treated insomnia and obstructive sleep apnea as two separate sleep disorders that occasionally show up in the same patient. A growing body of evidence now suggests that this framing misses something fundamental. When the two conditions co-occur, a state researchers call COMISA, or comorbid insomnia and sleep apnea, the consequences appear to be far greater than the sum of their parts, particularly for people living with type 2 diabetes. A new commentary published in the Journal of Clinical Sleep Medicine by pulmonologist Abdulghani Sankari of the John D. Dingell VA Medical Center and Wayne State University in Detroit argues that the diabetes–COMISA relationship deserves to be studied as a distinct clinical entity rather than an incidental overlap, and that unlocking this link could reshape how millions of patients are screened and treated.

The commentary accompanies a large epidemiological analysis drawn from TURKAPNE, a nationwide Turkish sleep apnea cohort, which found that diabetes mellitus is associated with both an increased prevalence and an increased severity of COMISA. That finding matters because COMISA is not a rare curiosity. Studies of sleep clinic populations and community cohorts suggest that a substantial share of patients diagnosed with obstructive sleep apnea also report chronic insomnia symptoms, and vice versa, yet the overlap has historically been under-recognized. Patients with COMISA often fall through diagnostic cracks: their insomnia complaint may mask the apnea during a routine visit, or the apnea diagnosis may lead clinicians to dismiss the insomnia as secondary, leaving the underlying sleep fragmentation untreated.

Part of what makes COMISA biologically plausible as a driver of metabolic disease is the distinct sleep architecture it produces. Research comparing patients with COMISA against those with obstructive sleep apnea alone or insomnia alone has shown that the combined disorder is associated with a unique pattern of sleep disruption, blending the repetitive upper airway collapses and oxygen desaturations of apnea with the prolonged sleep-onset difficulties and nocturnal awakenings characteristic of insomnia. In laboratory studies, acute intermittent hypoxia, the hallmark physiological insult of sleep apnea, has been shown to impair glucose metabolism even in awake, healthy volunteers, suggesting a direct mechanistic pathway from oxygen fluctuation to insulin resistance.

The insomnia component adds its own metabolic burden. A systematic review and meta-analysis focusing on insomnia disorder characterized by objectively short sleep duration found associations with hypertension, diabetes, and elevated body mass index. In other words, the subset of insomnia patients whose bodies cannot compensate for their sleep loss with longer time in bed appears to carry measurable cardiometabolic risk. When that short, fragmented sleep is layered on top of the cyclical hypoxemia of sleep apnea, the two insults may act synergistically on the sympathetic nervous system, the hypothalamic-pituitary-adrenal axis, and inflammatory pathways, all of which are central to glucose homeostasis.

Cardiovascular data reinforce the concern. An analysis of a large sleep registry found that type 2 diabetic patients with co-morbid insomnia and sleep apnea faced elevated cardiovascular risk compared with diabetic patients who had neither or only one of the two sleep disorders. Separately, work on the so-called hypoxic burden, a quantitative measure of the oxygen desaturation load carried during sleep, has shown that this metric predicts cardiovascular disease-related mortality better than the traditional apnea-hypopnea index in large community cohorts such as the Osteoporotic Fractures in Men Study and the Sleep Heart Health Study. Sankari’s own longitudinal research on nocturnal heart rate variability, measured through R-R interval changes in a community-based cohort, has similarly pointed to autonomic dysregulation during sleep as a harbinger of adverse cardiovascular outcomes.

What the TURKAPNE findings add is epidemiological weight to the idea that the arrow may point in both directions. Diabetes is associated with more prevalent and more severe COMISA, but the mechanistic literature suggests COMISA may also worsen glycemic control, creating a potential feedback loop. Hyperglycemia can promote nocturia, neuropathic pain, and restless discomfort that fragment sleep, while fragmented, hypoxic sleep promotes insulin resistance, appetite dysregulation through altered leptin and ghrelin signaling, and weight gain, which in turn worsens both apnea severity and metabolic control. Breaking this loop, the commentary argues, requires treating the two sleep disorders together rather than sequentially.

Treatment evidence is beginning to catch up with the biology. The most striking recent development comes from trials of tirzepatide, a dual incretin agonist originally developed for diabetes and obesity, which demonstrated significant reductions in the apnea-hypopnea index in patients with obstructive sleep apnea and obesity. Meta-analytic work on weight reduction more broadly confirms that even modest weight loss produces meaningful drops in apnea severity. For patients with COMISA and type 2 diabetes, these agents occupy a uniquely attractive position, because they simultaneously target the metabolic disease, the body weight that drives airway collapse, and potentially the sleep-disordered breathing that feeds back into glycemic instability. Whether they also ameliorate the insomnia component of COMISA remains an open question that researchers are only beginning to explore.

The pharmacological picture is complicated by safety concerns on the insomnia side. Longitudinal primary care data have linked prescriptions of hypnotics and anxiolytics with increased mortality, particularly in multimorbid patients, a category that describes many older adults with type 2 diabetes and coexisting sleep disorders. This makes non-pharmacological approaches, above all cognitive behavioral therapy for insomnia, the preferred first-line option for the insomnia component, while positive airway pressure therapy remains the cornerstone of apnea treatment. The challenge, as the commentary emphasizes, is that adherence to positive airway pressure is notoriously poor among patients with comorbid insomnia, who may find the mask and airflow intolerable when they already struggle to fall asleep. Integrated treatment pathways that address both disorders from the outset, rather than treating apnea first and hoping insomnia resolves, may improve adherence and outcomes for both.

Screening represents another frontier. Obstructive sleep apnea in women, particularly around reproductive aging and menopause, is frequently missed because presenting complaints differ from the classic male phenotype of loud snoring and witnessed apneas, and insomnia symptoms can further obscure the picture. Sankari and colleagues have highlighted these screening challenges in recent work, and the same logic applies to diabetic patients of both sexes: a patient who reports difficulty falling or staying asleep should prompt clinicians to consider whether sleep-disordered breathing is hiding underneath, and a patient prescribed continuous positive airway pressure who remains fatigued should be evaluated for persistent insomnia. Simple validated screening instruments for both conditions could be deployed in endocrinology and primary care clinics at minimal cost.

The commentary’s central message is ultimately one of reframing. Coexistence is not the right word for what happens when diabetes, insomnia, and sleep apnea converge in a single patient; the evidence points toward genuine pathophysiological entanglement, with each condition amplifying the others through hypoxic stress, autonomic activation, inflammatory signaling, and behavioral pathways such as sedentary fatigue and disordered eating. Large prospective cohorts that track sleep architecture, hypoxic burden, and glycemic outcomes simultaneously will be needed to establish causality and to determine whether treating COMISA aggressively improves diabetes complications and cardiovascular survival. Until then, the practical takeaway for clinicians is clear: in a patient with type 2 diabetes who sleeps badly, neither the insomnia nor the apnea should be treated as background noise, because together they may be quietly driving the disease that brought the patient to the clinic in the first place.

Subject of Research: The bidirectional relationship between type 2 diabetes and comorbid insomnia and sleep apnea (COMISA)

Article Title: Beyond coexistence: attempting to unlock the diabetes–COMISA Link

Article References: Sankari, A. (2026). Beyond coexistence: attempting to unlock the diabetes–COMISA link. Journal of Clinical Sleep Medicine, 22(1), Article 125. https://doi.org/10.1007/s44470-026-00135-2

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00135-2

Keywords: COMISA, type 2 diabetes, obstructive sleep apnea, insomnia, sleep medicine, intermittent hypoxia, glucose metabolism, cardiovascular risk, TURKAPNE cohort, tirzepatide, cognitive behavioral therapy for insomnia, sleep architecture