Insomnia as a cardiometabolic risk factor

Abstract

Dr. Grandner was in part supported by the National heart, Lung and Blood Institute (K23HL110216) and the National Institute for Environmental Health Sciences (R21ES022931), and a fellowship from the Institute for Translational Medicine and Therapeutics, funded by UL1RR024134 (Penn CTSA). Dr. Perlis was in part supported by National Institute of Mental Health (R01MH077900) and the National Center for Complementary and Alternative Medicine (R01AT003332). No off-label uses are described. No conflicts of interest are reported. Over the past two decades, many studies have challenged the conceptualization of insomnia as a primary or secondary disorder. Empirical studies show that insomnia tends to be a risk factor for, and a prodrome of new-onset illness and a risk factor for poor treatment response and illness recurrence.1,2 Theoretical papers suggest that it is difficult to establish that most cases of insomnia exhibit the temporal contiguity and variability needed to be classified as “secondary.”3,4 Accordingly, insomnia may be better conceptualized as a comorbid condition when it occurs concurrently with other disorders. This shift in perspective has been sufficiently persuasive that the next edition of the APA Diagnostic and Statistical Manual of Mental Disorders will abandon the primary/secondary distinction in favor of the diagnostic entity of “Insomnia Disorder.” The net effect of this change is to suggest that, as a disorder (as opposed to a symptom), more research is needed regarding the causes and consequences of insomnia. To date, there is a relatively large literature on the association between insomnia and mental illness, particularly major depression. This literature is primarily epidemiologic in nature and focused on insomnia as a risk factor. The literature on insomnia as a risk factor for medical illnesses is substantially more recent and sparse.5–7 The study by Strand and colleagues8 in this issue of SLEEP is important in that it represents a critical next step in the evaluation of how insomnia disorder may serve to predispose, precipitate, and/or perpetuate medical illness. The study presented by Strand et al. is exceptionally valuable because it adds to the nascent literature on a relationship between insomnia and cardiorespiratory function. Strand and colleagues utilized the HUNT-3 cohort to assess the relationship between (1) insomnia in terms of frequency of difficulty initiating sleep (initial insomnia), repeated awakenings (middle insomnia), and early morning awakenings (late insomnia), as well as with a global measure that summed these domains; and (2) cardiometabolic function in terms of peak oxygen uptake under heavy exercise conditions. They found that VO2peak (mL/kg/min) was negatively associated with insomnia severity. For example, values were 40.4 mL/kg/min in those with no insomnia and 35.1 mL/kg/min in those with initial, middle, and late insomnia. When specific symptoms/subtypes were explored, there was a linear trend for reduced VO2peak to be associated with the severity of middle insomnia, and to a lesser extent with initial insomnia. These effects persisted even after adjustment for a number of potential confounders. The study of Strand et al. has a number of important strengths. First is its large sample size (n = 3,489). Second, the sample is community-based (not self-selected for insomnia). Third, the use of an objective measure of cardiorespiratory fitness is exceptionally unusual to have such data on such a large sample. Fourth, the implementation of a challenge paradigm (assessing VO2 peak) essentially adopts a “testing the limits” approach that may be needed for insomnia, given the lack of findings in studies that fail to use such an approach when evaluating insomnia.9 Fifth, the authors attend to insomnia subtype (initial, middle, and late insomnia)—distinctions often ignored in favor of the broader categorization of primary or chronic insomnia. Clearly, this effort was fruitful, given that middle insomnia in particular was found to be associated with reduced respiratory fitness. The finding that middle insomnia was primarily responsible for the association of insomnia and respiratory fitness may be rationalized from a number of perspectives. One possibility is that middle insomnia describes a condition that captures the majority of the sleep state (i.e., a period ranging from 4 to 6 hours of sleep). Another possibility is that middle insomnia occurs after the initial sleep homeostatic response in the first few hours of sleep, and therefore it is, putatively, not as confounded by sleep homeostasis considerations, and thus allows for a “cleaner” assessment of the effects of sleep continuity disturbance. Yet another possibility is that compromised cardiometabolic function may be more disruptive to middle-of-the-night sleep, or, alternatively, middle insomnia is more disruptive to the processes that occur within sleep that serve to facilitate cardiometabolic function. A