When Insomnia Meets Sleep Apnea, Heart Risk Soars, Landmark Swiss Study Finds

Millions of people toss and turn at night, and millions more stop breathing in their sleep without knowing it. A growing body of research now suggests that when these two sleep disorders strike the same person at the same time, the consequences for the heart may be far worse than either condition alone. A new prospective analysis from the population-based HypnoLaus cohort in Lausanne, Switzerland, published in eClinicalMedicine, provides some of the strongest evidence yet that the combination of insomnia symptoms and moderate-to-severe obstructive sleep apnea — a phenotype researchers call COMISA — independently raises the risk of suffering a cardiovascular event, while neither disorder on its own shows the same effect.

The study followed 1,860 adults drawn at random from the resident population of Lausanne, aged between 35 and 75 at cohort entry. Unlike most previous work in this field, which relied on diagnostic codes or self-report, every participant underwent a full night of in-laboratory polysomnography, the gold standard for measuring sleep-disordered breathing. Obstructive sleep apnea was defined as an apnea-hypopnea index of at least 15 events per hour of sleep, while insomnia symptoms were captured with the Pittsburgh Sleep Quality Index, counting difficulty falling asleep within 30 minutes or waking during the night or early morning at least three times per week over the previous month. Cardiovascular outcomes — fatal and nonfatal myocardial infarction, ischemic and hemorrhagic stroke, central retinal artery occlusion, and coronary heart disease — were identified through medical records, hospital databases, and mortality registries, and then adjudicated by an expert committee of internists, cardiologists, and neurologists.

Over a median follow-up of 7.92 years, the pattern that emerged was strikingly specific. Participants with COMISA — 211 people, or 11.3 percent of the sample — had a crude hazard ratio of 3.68 for incident cardiovascular events compared with those who had neither condition. After statistical adjustment for age, sex, body mass index, diabetes, dyslipidemia, hypertension, smoking, alcohol intake, and prior cardiovascular events, the hazard ratio settled at 1.78, still clearly significant. In contrast, insomnia symptoms without sleep apnea showed no significant association with cardiovascular events, and neither did sleep apnea without insomnia symptoms. The adjusted eight-year absolute risk of a cardiovascular event was 12.9 percent in the COMISA group versus 8.1 percent in the reference group — an absolute difference of nearly five percentage points.

Perhaps the most intriguing statistical finding was a significant interaction between the two conditions. When the researchers modeled obstructive sleep apnea and insomnia as separate binary variables and included an interaction term, the hazard ratio for that interaction was 2.28, rising to 2.67 among participants free of cardiovascular disease at baseline. In plain terms, the combined hazard of having both disorders exceeded what would be expected from simply adding their individual effects — a signature of synergy rather than mere overlap. This statistical interplay is what elevates COMISA from a diagnostic curiosity to what the authors describe as a clinically relevant phenotype with its own prognostic weight.

The robustness of the result was tested from multiple angles. The team repeated the analysis using three alternative definitions of insomnia: one incorporating regular use of sleep medication, one requiring subjective wake after sleep onset of at least 30 minutes, and one requiring objectively measured wake after sleep onset of at least 30 minutes on polysomnography. COMISA remained significantly associated with cardiovascular events under every definition, with hazard ratios ranging from 1.86 to 2.14 in the full sample and up to 2.34 in participants without prior cardiovascular disease. Notably, the estimates were larger with the stricter wake-after-sleep-onset definitions, hinting that people whose apnea is punctuated by prolonged, genuine awakenings — rather than brief respiratory arousals — may represent the most vulnerable subgroup.

Sensitivity analyses reinforced the picture. Excluding participants who began continuous positive airway pressure therapy during follow-up did not eliminate the association, nor did including participants whose total sleep time fell below 240 minutes. When the outcome was restricted to cardiac events only, excluding the 35 cerebrovascular events, the association weakened and lost statistical significance — a finding the authors interpret cautiously given the small number of stroke events, but one that aligns with established links between both sleep apnea and insomnia and cerebrovascular disease.

Why should the combination of a racing mind and a collapsing airway be so much more dangerous than either alone? The authors lay out a plausible pathophysiological cascade. Insomnia is associated with physiological hyperarousal, dysregulation of the hypothalamic-pituitary-adrenal axis, elevated sympathetic tone, and metabolic disturbance. Obstructive sleep apnea contributes repetitive bouts of hypoxia, oxidative stress, dramatic swings in intrathoracic pressure, systemic inflammation, and further sympathetic activation. Together, these mechanisms could compound one another, keeping the cardiovascular system in a state of chronic nocturnal stress. A recent heart rate variability analysis of the Sleep Heart Health Study supports this view, reporting COMISA-specific autonomic dysfunction with reduced parasympathetic activity during wakefulness and heightened sympathetic activation during sleep.

Curiously, conventional markers of apnea severity told a different story. The apnea-hypopnea index, oxygen desaturation index, hypoxic burden, and respiratory arousal index were not significantly different when COMISA participants were directly compared with those who had sleep apnea alone. What did distinguish the COMISA group was sleep architecture: they spent a smaller proportion of the night in slow-wave, or N3, sleep — the deepest and most restorative stage. Previous HypnoLaus work has linked reduced delta spectral power to incident hypertension, suggesting that the microstructure of deep sleep may be a hidden determinant of cardiovascular vulnerability. The authors propose that alterations in both the macrostructure and microstructure of sleep, rather than respiratory severity per se, may drive the excess risk, and they call for advanced electroencephalographic analyses to identify the electrophysiological fingerprints of the phenotype.

The findings also carry an uncomfortable implication for sleep medicine. Large randomized trials have repeatedly failed to show that continuous positive airway pressure therapy prevents cardiovascular events in unselected patients with sleep apnea, and this study offers a possible explanation: comorbid insomnia may act as an effect modifier, diluting any cardiovascular benefit of treating the airway alone. The authors argue that systematic screening for insomnia symptoms in patients diagnosed with sleep apnea — and vice versa — could sharpen risk stratification. Encouragingly, cognitive behavioral therapy for insomnia has been shown to improve adherence to positive airway pressure therapy, and treating the apnea may in turn ease insomnia symptoms in some patients, opening the door to integrated treatment strategies whose cardiovascular impact now needs to be tested directly.

The study is not without limitations. Insomnia was assessed with a questionnaire that does not fully align with formal diagnostic criteria for chronic insomnia disorder, and symptoms were measured only at baseline, so the results reflect the prognostic value of COMISA as it existed at the start of follow-up rather than persistent insomnia over time. Residual confounding, as always in observational research, cannot be excluded. Yet the strengths are considerable: a randomly sampled general population, objective full-night polysomnography, expert-adjudicated outcomes, more than seven years of follow-up, and consistent results across multiple definitions and subgroups. Taken together with earlier, more heterogeneous studies, the work makes a compelling case that COMISA deserves recognition as a distinct, high-risk entity — one that clinicians can no longer afford to treat as two separate problems, and one that may hold the key to finally reducing the cardiovascular toll of disordered sleep.

Subject of Research: The association between comorbid insomnia and obstructive sleep apnea (COMISA) and incident cardiovascular events in a population-based cohort

Article Title: Comorbid insomnia and sleep apnea (COMISA) and incident cardiovascular risk: a prospective analysis from the HypnoLaus cohort

Article References: Solelhac, G., Imler, T., Marchi, N.-A., Baillieul, S., Van Den Broecke, S., Caetano, G., Heiniger, G., Waeber, A., Ericson, I., Bayon, V., Lombardi, A.-S., Haba-Rubio, J., Vollenweider, P., Marques-Vidal, P., & Heinzer, R. (2026). Comorbid insomnia and sleep apnea (COMISA) and incident cardiovascular risk: a prospective analysis from the HypnoLaus cohort. eClinicalMedicine, 100, Article 104227. https://doi.org/10.1016/j.eclinm.2026.104227

Image Credits: AI Generated

DOI: 10.1016/j.eclinm.2026.104227

Keywords: COMISA, insomnia, obstructive sleep apnea, cardiovascular risk, polysomnography, HypnoLaus, sleep medicine, hypertension, stroke, myocardial infarction, CPAP, cognitive behavioral therapy