“Time rate” of 24‐hour blood pressure variability

The characterization of associations between blood pressure (BP) phenotypes and health outcomes improves the clinical understanding of how BP may affect health and enables optimization of diagnostic and therapeutic strategies. Ultimately, BP phenotyping may enable precision medicine approaches by providing clinicians with a way to identify high‐risk patients and intervene to prevent adverse outcomes through pathophysiology‐based approaches. BP phenotypes include at least BP level and variability. The relationships between mean BP levels and adverse outcomes (eg, cardiovascular disease) have been well established. Variability of BP is an alternative BP phenotype that has been implicated in the pathogenesis of vascular damage and stroke independent of mean BP levels. BP variability (BPV) consists of short‐term BPV (eg, beat‐to‐beat variability and variability within a 24‐hour period) and long‐term BPV (eg, visit‐to‐visit BPV). Short‐ and long‐term BPV are weakly correlated, suggesting that their pathophysiology may not be identical.1, 2 Contributors to short‐term BPV include baroreflex and autonomic function, physical activity, large arterial stiffness, and responses to environmental and behavioral stimuli.3, 4, 5, 6 Conversely, contributors to long‐term BPV include lifestyle factors (eg, diet, physical activity, and sleep), socioeconomic status, adiposity, arterial stiffness, adherence to antihypertensive medication, and seasonal climatic change.3, 4, 5, 6

The SD, coefficient of variation (CV), and average real variability (ARV) based on three or more BP readings have been widely used to quantify both short‐ and long‐term BP variabilities.6 SD and CV were strongly correlated with ARV (Pearson's r=.7 to .8).7 However, their clinical implications may not be identical. For example, ARV weights the between‐reading time intervals and takes into account the order of the BP measurements, while SD is influenced by outliers or extreme BP values, which may be triggered by smoking, physical activity, orthostatic change, and emotional stimulation.8, 9 Of concern is the possibility that these parameters are partially dependent on the overall BP level. This concern may not be resolved even when mean BP level with multiple readings is used as an adjustment factor. Therefore, BPV independent of the mean1 and residual SD10, 11 have been proposed as alternative measures of BPV. These measures are less influenced by BP level with multiple readings compared with SD and ARV. In the current issue of the Journal of Clinical Hypertension, Kolyviras and colleagues¹² have provided an additional BPV index, called “time rate of BP variation” (Figure). The clinical implication of time rate of BP variation using noninvasive devices was first introduced from the same group in 2005,12 highlighting the potential effects of the magnitude and rate of BPV on vascular injury. They demonstrated that, among individuals with normotension and hypertension (n=514; mean age of 55 years), a 0.1‐mm Hg/min increase in time rate of 24‐hour systolic BP (SBP) variability was associated with an increment of 0.029 mm (95% confidence interval, 0.018–0.040) in common carotid artery intima‐media thickness, independent of 24‐hour mean SBP levels, 24‐hour SBP variability (assessed by SD), and the degree of nocturnal SBP dipping. In the current study, which recruited untreated Europeans (n=580; mean±SD age 51.8±11.0 years; 54% women; 30% with hypertension),13 the group further demonstrated that: (1) a higher time rate of 24‐hour or nocturnal SBP variability was associated with lower echogenicity of carotid plaque (ie, vulnerable plaque), independently of cardiovascular risk factors and 24‐hour mean SBP levels; (2) an SD of 24‐hour SBP, a commonly used index of BPV, was not associated with carotid plaque composition; and (3) a higher time rate of nocturnal diastolic BP (DBP) but not SBP variability was associated with the presence and amount of carotid plaque, independently of 24‐hour mean SBP levels.

Figure 1.

An example of computing time rate of blood pressure (BP) variability (r) of noninvasive ambulatory BP changes. S₀, S₁, S₂, and S₃ stand for 4 consecutive systolic blood pressure (SBP) readings at time indices of t₀, t₁, t₂, and t₃, respectively. Given 2 systolic BP (SBP) readings, S_i and S_i+1 at time indices t_i and t_i+1, respectively, the rate of BP change was defined as follows: r_i=S_i+1 – S_i/t_i+1 – t_i at time index t_ri=t_i+t_i+1/2. Given N recordings of SBP during the 24‐hour period of ambulatory BP monitoring, we can compute N‐1 values for the rate of SBP variation at N‐1 different time indices. Thus, for each patient with N SBP recordings, the following variable, to which we refer as 24‐hour rate of SBP variation (mm Hg/min), was calculated: $|r|=\sum _{i=1}^{N-1}|ri|$/N‐1. Reproduced from Hypertension. 2005;45:505–51212

The 24‐hour BPV assessed by ambulatory BP monitoring has been reportedly associated with carotid intima‐media thickness.3, 12 However, its association with carotid plaque composition is uncertain. The study by Kolyviras and coworkers may be the first to highlight the association of time rate of 24‐hour SBP variability with plaque composition. B‐mode ultrasound of carotid plaque composition, when compared with the surgically removed specimen, demonstrated that lower echogenic carotid plaque consists of hemorrhage and lipid (ie, vulnerable plaque).14 Therefore, plaque composition could better predict incident stroke and coronary artery disease compared with carotid intima‐media thickness.15 Higher short‐term BPV may lead to increased oscillatory shear stress to the vascular endothelium, potentially contributing to vascular injury by increasing expression of adhesion molecules and pro‐oxidant processes, and reducing nitric oxide synthase more than steady blood flow.16, 17 In a human study, greater BPV in the morning was associated with inflammation in the carotid plaque (ie, more macrophages and T lymphocytes, increased ubiquitin‐proteasome activity, and increased levels of tumor necrosis factor‐α and nuclear factor‐kB).18 However, these are observational studies, and thus it remains to be determined whether greater BPV assessed by ambulatory BP monitoring is a contributor to vascular injury or merely an epiphenomenon of vascular dysfunction or other pathophysiological conditions (eg, environmental and psychological stress). Further studies are warranted to assess whether reductions in time rate of SBP variability can contribute to carotid plaque regression and plaque stabilization.

Patients with hypertension have not only higher BP levels but also steeper rises and falls in BP.19 Ideally, the degree and rate of short‐term BP change should be assessed by beat‐to‐beat BPs using intra‐arterial BP monitoring. However, this approach may not be feasible in practical situations. Mean BP level measurements over 24 hours estimated by intra‐arterial continuous BP monitoring are comparable to estimated intermittent BP measurements taken at 30‐minute intervals. However, the comparability of estimates of BPV derived from continuous vs intermittent BP measurements has never been assessed.20 Kolyviras and colleagues calculated the time rate of 24‐hour BPV based on intermittent BP measurements taken at 15‐minute intervals. If the measurement interval is fixed (ie, no missing ambulatory BP data), the time rate of BPV is the absolute difference between successive BP measurements. Consequently, the correlation between time rate of BPV and ARV would be strong. Unfortunately, this point was not tested in the study by Kolyviras and associates. If the time rate of BPV and ARV are strongly correlated, the former may not reflect the rate of short‐term BP change. In addition, the mechanisms that might contribute to the discrepancy between time rate of 24‐hour SBP variability and SD of 24‐hour SBP in relation to plaque composition are uncertain. Characterizing the distribution of each BP index within this population and assessing how clinical characteristics associated with time rate of 24‐hour SBP differ from those associated with the SD of 24‐hour SBP may clarify the mechanisms.

Different BP components reflect distinctive hemodynamics and pathophysiology.21 DBP, at least up to age 50 years, reflects a steady‐state load of BP and is representative of resistant vessel structure and function alterations.22, 23 In contrast, SBP is an integrated measure of steady and pulsatile pressure load and is representative of large arterial (aortic) stiffness and cardiac output.22, 24 Kolyviras and colleagues emphasized that time rate variability of both SBP and DBP should be assessed, because associations of the time rate of BPV with outcomes differ by BP components (ie, SBP vs DBP). These association analyses need to be replicated in other populations. To understand how hemodynamic physiology differs between SBP and DBP variability, and whether the effect of each BP metric on vasculature differs, will require further investigations.

In summary, the current study has shown that: (1) a higher time rate of 24‐hour or nocturnal SBP variability was associated with lower echogenicity of carotid plaque (ie, vulnerable plaque), independently of cardiovascular risk factors and 24‐hour mean SBP levels; (2) the SD of 24‐hour SBP, a commonly used index of BPV, was not associated with carotid plaque composition; and (3) a higher time rate of nocturnal DBP but not SBP variability was associated with the presence and amount of carotid plaque, independently of 24‐hour mean SBP levels.

CONFLICT OF INTEREST

None.

Yano Y. “Time rate” of 24‐hour blood pressure variability. J Clin Hypertens. 2017;19:1078–1080. 10.1111/jch.13068