Insomnia and hypertension: A misty landscape

Insomnia is the most common sleep disorder, and it is present in about 30% of the adult population globally. Interestingly, one out of three affected individuals report substantial symptoms.1 Historically, insomnia has been recognized as a significant healthcare problem that needs treatment by ancient Greeks and Egyptians who recommended use of poppy opium and hemp plant.2 More than 2 millenniums later, Justus von Liebig introduced chloral hydrate in the treatment of insomnia. This agent exerted a γ‐aminobutyric acid (GABA)ergic activity, a mechanism of action that is still shared with the current widely used in clinical‐practice sleep‐aids.3 However, the multifactorial effect of GABAergic stimulation can lead to changes in mood, suicidal thoughts, amnestic effects, motor deficits, slowed reaction time in psychomotor vigilance tests, falls in the elderly, and even respiratory abnormalities, as well as physical tolerance that can result in addiction, withdrawal, and drug overdose. In fact, desire to control these adverse effects has been proven to be a substantial impetus for progress in the field of insomnia. Nevertheless, even the newer more selective with shorter half‐life, benzodiazepines and non‐benzodiazepines have been linked to hazardous activities such as dangerous driving that limit their administration in the short and long term.

In the late ‘90s, several scientific teams realized the key role of the neurotransmitter orexin on the manifestation of narcolepsy.4 Orexin A and Orexin B were initially found to promote feeding after experimental injection in the brain of rats; hence, they were named orexin after the Greek word “όρεξη” for appetite. Moreover, endogenous orexin levels feature circadian oscillations coinciding with wakefulness. This awaking effect seems to be mediated by orexinergic neurons localized to the lateral hypothalamus and is counterbalanced by the major sleep promoting effect of GABAergic neurons. Herein, several potential orexin antagonists have been evaluated in insomnia, but only suvorexant met the safety and efficacy criteria of the Food and Drugs Administration, and eventually was approved for the management of primary insomnia, as a novel and more safe agent than the GABAergic compounds (with some issues not fully addressed). Suvorexant is a dual orexin receptor antagonist (DORA) that inhibits the activity of both orexin type‐1 and type‐2 receptors and downgrades the threshold to transition into sleep and attenuates orexin‐mediated arousal. Moreover, suvorexant is characterized by pharmacokinetic properties that entail rapid sleep onset, efficacy at low maintenance doses, and low potential of the hangover phenomenon.

Accumulating evidence suggests that insomnia is associated with a wide cluster of cardiovascular and metabolic disorders, such as coronary heart disease, stroke, hypertension, diabetes mellitus, obesity, dyslipidemia, atherosclerosis, heart failure, and arrhythmias.5, 6 With regard to hypertension, few data suggest that both the diagnosis of insomnia and the severity of the sleep disturbance (such as cortical hyper‐arousal common in insomnia) are associated with over‐activation of the hypothalamic‐pituitary‐adrenal axis and over‐secretion of cortisol, and thus may result in abnormalities in circadian regulation of blood pressure (BP), and non‐dipping status.7, 8, 9 Furthermore, up‐regulation of the sympathetic system, the immune system, and the inflammatory cascade (altogether, the atherogenic trinity) in insomnia, might contribute to hypertension, as well. Nevertheless, it remains unknown whether insomnia causes hypertension and data are still inconclusive. Most relevant studies employed self‐reported assessment of insomnia symptoms (sometimes with a single question: do you have insomnia/sleep disorders?), and few utilize polysomnography in the quantitation of sleep and sleep disruption. More importantly, available studies used different definition of hypertension, and methodologies for BP measurements (home, office, and rarely ambulatory BP values). Table 1 summarizes the data from large cross‐sectional and longitudinal studies, published over the last 2‐years, assessing the risk of hypertension in patients with insomnia.10, 11, 12, 13, 14, 15, 16

Table 1.

Studies on insomnia conducted over the last 2 y, evaluating the impact of insomnia on hypertension, as well as the association of insomnia in combination with short sleep duration (<6 h defined with actigraphy or polysomnography) with hypertension

Study	N/Race	Condition	Risk for hypertension
Shivashankar et al10	16 244/Asian	Self‐reported insomnia with rare symptoms	OR: 1.41 (95% CI: 1.12 to 1.77)
		Self‐reported insomnia with occasional symptoms	OR: 1.39 (95% CI: 1.16 to 1.67)
		Self‐reported insomnia with frequent symptoms	OR: 1.34 (95% CI: 1.09 to 1.65)
Wang et al11	8017/ Asian	Self‐reported insomnia with daytime impairment	OR: 1.52 (95% CI: 1.22 to 1.88) for elevated BP
Wang et al12	3176/Asian	Self‐reported insomnia	OR: 1.29 (95% CI: 1.10 to 1.51)
Ramos et al13	2148/ Latin	Self‐reported insomnia with daytime impairment	Beta: −1.6 (95% CI: −2 to 11.2)
Lin et al14	1739/ Asian	Clinical diagnosis of insomnia	HR: 1.21 (95% CI: 1.01 to 1.76)
Bathgate et al15	255/56.5% Caucasians	Clinical diagnosis of insomnia and sleep duration <6 vs ≥6 h	OR: 3.59 (95% CI: 1.58 to 8.57)
Johann et al16	328/Caucasians	Clinical diagnosis of insomnia and sleep duration <6 h vs ≥6 h	OR: 0.79 (95% CI: 0.46 to 1.36)

BP, blood pressure; CI, confidence intervals; HR, hazard ratio; OR, odds ratio.

A recently published systematic‐review pooled data from 64 clinical studies with diverse designs, including more than 740 000 adults (46.4% male) with insomnia to investigate the relationship between insomnia and hypertension.17 It was found that when insomnia is frequent, chronic, and accompanied with short sleep or objective indices of arousal, there is a robust link with hypertension. In contrast, limited data showed that hypertension did not consistently predict the manifestation of insomnia, especially in middle‐aged adults. Of note, it was reported that in patients with insomnia, the risk of hypertension increased in parallel with the severity of this sleep disorder.

The most reassuring way of proving the effects of insomnia on hypertension is to document changes in BP by treating insomnia. Unfortunately, data are limited in this field, as well. Cognitive behavioral therapy that provides significant benefits in insomnia status has failed to consistently show BP‐lowering effects.18 Huang and associates demonstrated that a non‐benzodiazepine (zolpidem) significantly influenced the conversion of non‐dipping to dipping status when used for the management of insomnia.19 Li et al20 conducted a randomized, placebo‐controlled trial to evaluate the impact of estazolam on BP outcomes in patients with insomnia. Overall, 402 Chinese with insomnia and hypertension were randomly assigned to estazolam 1 mg/d (n = 202) and standard antihypertensive treatment or the later and placebo (n = 200). After 28 days of follow‐up, estazolam was associated with a significantly greater reduction in office BP compared with placebo (−10.5/−8.1 vs − 3.4/−2.7 mm Hg, respectively; P < 0.001). Moreover, estazolam compared with placebo showed more likely to achieve well‐controlled BP levels (74.8% vs 50.5%, respectively; P < 0.001), along with substantial improvements in several validated scales of sleep quality, anxiety, and depression.

In the current issue of the Journal, Kario and colleagues conducted the SUPER‐1 study to investigate whether the DORA suvorexant exhibited BP‐lowering effects in patients with insomnia and hypertension.21 After a 4‐week ruin‐in period, 82 Japanese were randomized to receive an age‐adjusted dose of suvorexant (15 mg/d for individuals ≥ 65‐years or 20 mg/d for younger) (n = 40) or placebo (n = 42), for 2‐weeks. In total, both groups had similar baseline characteristics and comorbidities, but patients in the placebo group were more likely to receive an angiotensin receptor blocker. At baseline, patients in suvorexant compared with placebo had a similar office BP (OBP) (138/84 vs 137/82 mm Hg; P > 0.05) and ambulatory BP monitoring (ABPM) (153/90 vs 150/88 mm Hg; P > 0.05), and daytime and nighttime ABPM, whereas, patients on suvorexant had higher home systolic BP (151 vs 146 mm Hg; P = 0.025). With regard the primary outcome, suvorexant failed to provide any significant reduction in nighttime systolic ABPM versus placebo. Similar outcomes were showed in office and 24 h ABPM results, while no difference was observed in reference to dipping status. Suvorexant achieved a higher satisfaction in sleep and improved some sleep parameters.

The study by Kario et al21 is a clinically relevant study addressing the impact of DORA on reducing nighttime BP through a potential reduction in sympathetic tone and the potential suppression of the hypothalamic‐pituitary‐adrenal axis. However, suvorexant failed to establish any significant BP‐lowering effect in this short‐term period of 2‐weeks. Of note, a post hoc analysis suggested that suvorexant resulted in a greater reduction in morning systolic ABPM compared with placebo among participants with baseline morning systolic ABPM > 150 mm Hg (P = 0.037). The most interesting finding of the study, however, may be the high prevalence of masked hypertension in this population (47% in both groups). This finding is based on the observation that the baseline ABPM levels were higher than the office BP ones. These findings could explain the lack of correlation between insomnia and hypertension that was found in some studies employing office BP measurements. Given the significant lack of data addressing the association of insomnia and hypertension employing ABPM, it would be clinically interesting to study whether a specific hypertensive‐phenotype dominates patients with insomnia. This is especially true, considering the significantly higher cardiovascular risk in patients with masked hypertension.22

Kario et al21 conducted a well‐designed, double‐blind, randomized study with significant strengths such as the well‐balanced groups, the appropriate diagnosis of insomnia, and more importantly the adequate design with a ruin‐in period, and appropriate BP measurements. We hope that this study will influence the field in order to include similar BP measurement protocols for future clinical trials, since for the time being the methodology of BP measurement seems to be the “Achilles Heel” of the association between insomnia and hypertension.

In addition, the short‐term duration of the study makes it difficult to interpret the results and certainly makes it premature to draw conclusions on the BP‐lowering effects of suvorexant. The relatively small sample size limits the generalization of the study outcomes. Furthermore, the use of non‐validated questionnaires limits the utility of the results.

The potential association between insomnia and hypertension is interesting and it should be viewed in parallel with the proven association of obstructive sleep apnea and hypertension. However, existing data are limited and more work needs to be done. For proper investigations, a number of variables need to be established: (1) wide‐adoption of accepted definition of insomnia, (2) use of contemporary definition of hypertension and accepted methods of BP measurement are needed. In order to produce high‐quality data, high‐quality studies are needed. These future studies should be clinically relevant in order to investigate (a) the epidemiological characteristics of hypertension in insomnia, and (b) the impact of the insomnia treatment on BP levels. Till then, we can make no conclusions about the association of insomnia with hypertension.

CONFLICT OF INTEREST

The authors report no specific funding in relation to this research and no conflicts of interest to disclose.