Article ID: CJ-24-0926
Time-space network hypertension is a data science approach that connects diverse information related to hypertension within a time-space framework. This field of academic research aims to predict disease onset and direct effective, individualized, optimized treatments by integrating and analyzing the variability of multiple internal biological and external environmental signals as they relate to blood pressure variability across different time phases. By linking time series changes in blood pressure and biological distribution with multi-environmental and physiological information, enabled by advances in digital technology, the time-space network hypertension approach contributes to “digital hypertension” research. This article from Jichi Medical University provides an update on research relating to the time-space network hypertension approach, which is designed to progress hypertension management towards achieving net zero cardiovascular events.
In recent years there have been rapid advances in digital information technologies. This has created expectations regarding the ability to collect data on a range of new clinical indicators that have the potential to add value based on the analysis of multifaceted information, which can then be applied to refine, improve, and personalize the management of hypertension.
Time-space network hypertension is a data science approach that connects diverse information related to hypertension within a time-space framework (Figure 1). This field of academic research aims to predict disease onset and direct effective, individualized, and optimized treatments by integrating and analyzing the variability of multiple biological signals from different organs (detected using signals such as electrocardiogram, pulse oximetry, and photoplethysmography) and external environmental signals (e.g., room temperature, light, humidity, and atmospheric pressure) and relating these to blood pressure (BP) variability across different time phases.
Time-space network hypertension and metrics. BP, blood pressure; ECG, electrocardiogram; K, potassium; Na, sodium.
Time-space network hypertension is positioned as a systematic academic concept that considers an individual patient’s BP-related disease states and surrounding environment as a BP-related system, using BP (which changes over time and propagates throughout the body) as an evaluation indicator. Conversely, digital hypertension is an academic research field that leverages the rapid advances in digital technology over recent years to pursue added value in healthcare and medicine, specifically in the treatment and prevention of hypertension. Although there is some overlap between the time-space network hypertension and digital hypertension, the former is primarily an academic concept whereas the latter primarily relates to a technological and interdisciplinary approach. In short, time-space network hypertension is a discipline that links time series changes in BP and biological distribution with environmental and physiological information, enabled by advances in digital technology, thus contributing to “digital hypertension” research.1–3 This article introduces the time-space network hypertension concept, presents our latest research results in this field, and discusses future prospects.
The development of life-threatening cardiovascular diseases events such as stroke, myocardial infarction, aortic dissection, and heart failure is a complex process that first involves aging combined with chronic risk factors, then progression of arteriosclerosis and the development of cardiac hypertrophy (Figure 2, Left). These conditions are exacerbated by hypertension, which is a powerful cardiovascular risk factor (along with diabetes, dyslipidemia, smoking, and chronic kidney disease). In high-risk individuals who have advanced arteriosclerosis and cardiac hypertrophy, an exaggerated BP surge can act as an acute trigger, leading to plaque rupture or hemorrhage, thereby causing atherosclerotic cardiovascular disease events. Furthermore, elevated BP increases cardiac afterload, and can therefore trigger acute heart failure.
Trajectory of cardiovascular disease and the resonance hypothesis of blood pressure surge. ABPM, ambulatory blood pressure monitoring; BP, blood pressure; BPM, blood pressure monitoring; CV, cardiovascular; PM2.5, particle pollution from fine particulates. Modified from Kario 2016,4 with permission of Oxford University Press.
Our time-space network hypertension research seeks to associate time series changes in cardiovascular risk factors with indicators of both short- and long-term BP variability, resulting in the development of cardiovascular diseases. The ultimate goal is to anticipate who will experience cardiovascular disease events, as well as when, where, and under what circumstances these events may occur.
One of the 2 key axes of time-space network hypertension is “time series”. In this context, the “resonance hypothesis” of BP surge (Figure 2, Right),1,4,5 which triggers cardiovascular events, is a fundamental hypothesis of time network hypertension. Environmental and lifestyle factors influence cardiovascular risk, and can promote the subclinical progression of cardiovascular disease, including atherosclerosis, cardiac hypertrophy, and organ damage. In addition, environmental factors, such as cold temperatures in winter and particle pollution from fine particulates (PM2.5), along with activities of daily living, such as acute exercise, mental stress, high salt intake, alcohol consumption, and poor sleep quality, can trigger exaggerated BP variability and surges (Figure 2, Right). When these acute pressor triggers cause BP surges that coincide with the peak of seasonal, day-to-day, and diurnal BP variations, the resonance of all these surges generates a pathologically large surge in BP that can trigger cardiovascular events. By predicting the occurrence of surge BP based on the time series characteristics of BP variability caused by various individual pressor triggers, it should be possible to anticipate the onset of cardiovascular events.
Another key axis of time-space network hypertension is “spatial distribution”. For this component of time-space network hypertension, the fundamental concept is SHATS (Figure 3).1,6,7 This facilitates the explanation for how surge BP triggers cardiovascular events at the coronary and cerebral arteries, especially in high-risk individuals. SHATS is a vicious cycle of hemodynamic stress and vascular disease, facilitating the power of surge BP to trigger cardiovascular disease events and advance organ damage.
Systemic hemodynamic atherothrombotic syndrome (SHATS): acceleration of the risk of cardiovascular events and organ damage via a vicious cycle of hemodynamic stress and vascular disease. An exaggerated morning surge in blood pressure (BP) and hemodynamic stress accelerate vascular disease, and advanced vascular disease augments BP variability. AI, augmentation index; BPM, blood pressure monitoring; BNP, B-type natriuretic peptide; BRS, baroreceptor sensitivity; CAVI, cardio-ankle vascular index; ECG, electrocardiogram; Echo, echocardiography; FMD, flow-mediated dilatation; MRI, magnetic resonance imaging; NT-proBNP, N-terminal pro B-type natriuretic peptide; PWV, pulse wave velocity; UACR, urinary to albumin creatinine ratio. Reprinted from Kario et al. 2020,7 with permission of Elsevier.
The heart generates pulse pressure approximately 100,000 times each day, and this is propagated throughout the body and its organs. The exaggerated power of surge BP is absorbed by the aorta when aortic compliance is good. Conversely, if aortic compliance is poor, the power of surge BP is not absorbed and is, instead, transmitted to peripheral arteries close to organs such as the brain, coronary arteries, and kidneys.
In SHATS, surge BP is transmitted to arteriosclerotic plaques in the coronary arteries and/or major cerebral arteries without any attenuation, triggering plaque rupture and causing an acute coronary event or atherothrombotic stroke. Even in the absence of arteriosclerotic plaques, frequent high-power surge BP can cause small artery disease, such as lacunar infarctions, cerebral hemorrhage, vascular dementia, and chronic kidney disease. Surge BP could also contribute to the development of acute heart failure due to increased afterload resulting from decreased aortic compliance, or to aortic dissection due to increased pressure load on the damaged aorta.
Digital technology connects chronological personal health records and facilitates digital health, behavioral modification, and digital therapeutics designed to promote lifestyle and environmental modifications (including a reduction in sodium intake, body weight control, exercise, stress management, and room temperature control; Figure 2, Left).
Cuffless wearable BP monitoring devices that detect BP surges during daily life, along with big data processing/transmission and artificial intelligence (AI), hold promise for estimating surge BP and predicting related cardiovascular events in the future. However, the accuracy of absolute BP values generated from cuffless BP monitoring devices is not yet good enough.8 Therefore, the best BP metrics are those measured using oscillometric home BP monitoring in clinical practice because there is a good body of evidence to support the use of this approach.9 The addition of genotype-based precision medicine and hemodynamic biomarker-based anticipation medicine should facilitate optimal personalized medicine, ultimately leading to the elimination of cardiovascular events.10,11
Time-space network hypertension uses chronological analysis of multidimensional big data to determine cardiovascular risk based on individual BP variability over time as a temporal axis and organ damage as a spatial axis, while also taking into account lifestyle, environmental, and biological signals.5
Diurnal BP VariabilityThe most well-established evidence regarding BP variability is the association between 24-h diurnal BP variability and cardiovascular diseases. Surge BP occurs on top of the circadian variations in BP and contributes to an individual’s BP variability risk (Figure 4).12
Synergistically accumulated trigger-specific blood pressure (BP) surges and diurnal variation. HTN, hypertension. Reprinted from Kario 2020.12
There are 4 types of diurnal BP variability: dipper (the normal pattern, where night-time BP is 10–20% lower than daytime BP); non-dipper (where night-time BP is 0–10% lower than daytime BP); riser (where night-time BP is higher than daytime BP); and extreme dipper (where night-time BP is >20% lower than daytime BP).13 Of these, the riser pattern of night-time BP is associated with the most significant progression of organ damage and the highest risk of cardiovascular diseases such as stroke and heart failure (Supplementary Figure 1).14 Potential mechanisms for these associations are as follows. The supine position during night-time sleep causes a shift of circulating blood to the upper body from the lower body, increasing cardiac preload.15 Furthermore, the absence of hydrostatic pressure in the brain leads to increased pressure on cerebral vessels. The combined effect of increased night-time heart rate and BP further exacerbates cardiac workload, thereby increasing the risk of heart failure.16
The extreme dipper pattern of night-time BP variability is also associated with a higher risk of stroke,13,14 with risk increasing as night-time BP decreases.14 The 2023 European Society of Hypertension guidelines include a Class I recommendation for evaluation of night-time BP and diurnal variability types using 24-h ambulatory BP monitoring (ABPM).17
Day-by-Day BP VariabilityDaily home BP monitoring captures BP variability over a longer period than a single 24-h period of ABPM. Our Japan Morning Surge Home Blood Pressure (J-HOP) study evaluated the prognostic impact of home BP variability using the following measures: coefficient of variation (CV), average real variability (ARV), peak BP (mean of the maximum of 3 readings), and time in therapeutic range.18–24 All these measures of BP variability were associated with cardiovascular risk (especially stroke risk) independent of age, other variables, and office BP.
Although home BP typically increases at lower room temperatures, increased BP variability (CV and ARV) during winter was particularly associated with an elevated risk of cardiovascular diseases.23 A home BP cardiovascular risk prediction score that includes home BP variability (ARV) has recently been developed (Supplementary Figure 2).20 Furthermore, the SHATS hypothesis indicates that the synergistic increased risk associated with home BP variability is augmented in high-risk individuals who have arterial stiffness.25–27
Peak BPPrevious prospective studies have shown that the highest (peak) values of office, home, and ambulatory systolic BP are strongly associated with an increased risk of cardiovascular diseases.18,21 The J-HOP study found that the peak home BP (morning and evening BPs) was a risk factor for stroke independent of office BP.21 In addition, the J-HOP nocturnal BP study demonstrated that peak night-time BP (the mean of the highest 3 night-time BPs) was associated with stroke risk independent not only office but also morning and evening BPs.28 Furthermore, the Japan Ambulatory Blood Pressure Monitoring Prospective (JAMP) study, which used ABPM, found that peak BP was a significant risk factor for stroke in patients with advanced atherosclerosis.29 In the UK transient ischemic attacks (UK-TIA) study of individuals with a history of transient ischemic attacks, peak office BP was found to be an extremely strong risk factor for stroke.30 These findings underscore the importance of SHATS in assessing cardiovascular risk.
BP ReactivityThe factors that trigger BP surges and the magnitude of BP surges vary between individuals. Therefore, we have developed a new type of “all-in-one” ABPM device that incorporates an actigraph and a thermometer and can measure office, home, and ambulatory BP (Figure 5).31 This led to the creation of 2 BP indices: actisenstivity (BP elevation in response to physical activity)1,11,32 and thermosensitivity (BP elevation in response to low temperatures).1,11,33 Actisensitivity and thermosensitivity vary depending on an individual’s medical characteristics and the combination of influencing stimuli. For example, in elderly individuals with a lean body type, temperature sensitivity is heightened, and reactivity increases in winter compared with summer (Figure 6, Left).11 These trends may explain the increase in cardiovascular diseases during the winter season.
Information communication technology-based multisensor ambulatory blood pressure monitoring (ABPM) and home blood pressure (BP) monitoring. BPV, blood pressure variability; CVD, cardiovascular disease. Modified from Kario 2017,31 with permission of Oxford University Press.
Actisensitivity and circadian rhythm of blood pressure (BP) in people with hypertension (“hyper” actisensitivity is a characteristic of hyper-reactive hypertension [e.g., exaggerated morning surge, orthostatic hypertension, and extreme nocturnal BP dipping], whereas negative actisensitivity is a characteristic of negative reactive hypertension [e.g., riser or reverse dipper pattern of nocturnal BP dipping, orthostatic hypotension, and blunted morning surge]). Reprinted from Kario 2017,11 with permission of Elsevier (Left) and Tomitani et al. 2024,32 with permission of Wolters Kluwer Health (Right).
BP reactivity is also a determinant of diurnal BP variability. Recently, we found that decreased reactivity is a determinant of the non-dipper pattern of nocturnal BP, whereas excessive reactivity is a determinant of morning BP surge (Figure 6, Right).32 In addition, research using wristwatch-style wearable BP monitors has shown that mental stress in the workplace can increase BP (Supplementary Figure 3).34 Based on these findings, there is a need to accumulate multiple sets of time series data and develop predictive algorithms that can objectively calculate an individual’s risk of experiencing BP surge based on the combination of season, location, and various pressor stimuli.
We define the ideal BP control state as “perfect 24-h BP control” (Figure 7).1,35 This requires strict BP control over each 24-h period with no excessive BP surges and stable BP variability throughout the day, characterized by a dipper-type pattern of nocturnal BP. In addition, if day-by-day BP variability remains relatively stable, this can be considered an ideal BP control state with minimal cardiovascular risk.
Triad of perfect 24-h blood pressure (BP) control. BP, blood pressure. Reprinted from Kario 2012.35
In principle, there are various methods that can be used to lower BP, including lifestyle and environmental modifications, pharmacotherapy, and renal denervation. However, currently, fewer than 50% of patients with hypertension using antihypertensive pharmacotherapy in Japan have their night-time or early morning BP adequately controlled.36,37 Antihypertensive drug therapies focus on increasing medication adherence by incorporating a combination of drugs that have different mechanisms of action into a single preparation to reduce the number of medications taken. Another approach is to ensure that individuals are aware of, and understand, their BP readings and BP targets with the aim of reducing clinical inertia. In addition, as described below, new approaches, such as digital therapeutics using smartphone applications,38–42 transcatheter renal denervation,43–55 and antibody and nucleic acid-based medications, are emerging as promising antihypertensive treatments.56 In addition to current standard therapies, all of these can contribute to a home BP-based approach to hypertension management (Supplementary Figure 4). Furthermore, non-steroidal mineralocorticoid receptor antagonists and angiotensin receptor–neprilysin inhibitors are novel agents that are now in clinical use,57–60 and endothelin receptor antagonists61 and aldosterone synthase inhibitors62 are undergoing clinical trials.
Key technologies and components in time-space network hypertension research that are crucial for achieving personalized anticipation medicine include wearable BP sensors capable of collecting longitudinal big data, the development of clinical indicators, and supporting digital application technologies (Figure 8).63 This also involves information and communication processing technologies for time series big data, and data analysis systems driven by advanced digital technologies, including AI.64 The development of new BP sensors is especially important. There is therefore a need for BP sensors that can more accurately obtain continuous BP readings and large-scale BP variation data, including early morning BP, daytime surge BP values, and nocturnal BP without sleep disturbance.65 These sensors should be capable of frequent measurements that can be evaluated using the current practice of home BP measurements or a single day of ABPM, as is the current clinical routine.
Time-space network hypertension research facilitates personalized anticipation medicine. AI, artificial intelligence; BP, blood pressure; EHR, electronic heath record; ICT, information and communication technology; IoT, internet of things; PHR, personal health record; SBP, systolic blood pressure.
The advancement of time-space network hypertension science, which connects diverse time series and physiological distribution digital information, is essential in the new era of digital hypertension management and transitions from a uniform population-based approach to more effective individualized anticipation medicine. To help achieve this goal, it is essential to develop BP measurement devices and clinical indicators, along with diagnostic algorithms, that used these tools and their data.
K.K. has received research grants from A&D, Omron Healthcare, Fukuda Denshi, CureApp, and Sanwa Kagaku Kenkyusho; and honoraria from Daichi Sankyo, Viatris, Novartis, Otsuka Pharmaceuticals, Medtronic, Otsuka Medical Device, and Omron Healthcare. S.H. has received manuscript fees from Novartis, outside of the submitted work. K.K. is a member of Circulation Journal’s Editorial Team. The other authors report no conflicts of interests.
Please find supplementary file(s);
https://doi.org/10.1253/circj.CJ-24-0926
