Ambulatory Blood Pressure Variability in Prediction of Target Organ Injury: The SHIP AHOY Study

Abstract

Background:

Ambulatory hypertension is associated with elevated left ventricular mass index (LVMI), cardiac dysfunction and increased arterial stiffness in adolescents. Whether addition of measures of BP variability improve prediction of subclinical cardiovascular target organ damage (TOD) over mean BP measures is not known.

Methods:

We assessed clinic and ambulatory BP (ABP), anthropometrics, and TOD in 397 adolescents. ABP means, standard deviation (SD), BP and heart rate (HR) dipping were calculated; coefficients of variability (CV) were calculated (SD/mean) to assess ABP variabilities. Measures of TOD included LVMI, left ventricular hypertrophy (LVH), LV systolic shortening, LV diastolic function (e’/a’); and pulse wave velocity. General linear models were used to determine if increased ABP variability measures were significant determinants of TOD in models containing mean ABP percentiles, age, sex, race/ethnicity, BMI z-score, and HR.

Results:

Mean participant age was 15.6 ± 1.7 years (63% white, 59% male) with mean casual BP 122.6/71.6 mm Hg ±12.4/10.5, and mean awake systolic ABP 124.2/72.0 ± 11.3/7.7 mm Hg. In linear models, increased awake CV-DBP and HR dipping were significant determinants of LVMI. CV-HR was an independent determinant of diastolic (e’/a’) but not systolic dysfunction. Using logistic regression, the combination of awake and asleep diastolic ABP variability and awake systolic ABP percentile improved prediction of LVH.

Conclusions:

Consideration of ABP variability in addition to ABP percentile may aid in identifying adolescents at risk for LVH.

Graphical Abstract

Introduction

Recent studies have demonstrated that blood pressure (BP) levels considered within the normal range in children and adolescents are associated with subclinical indicators of incipient cardiovascular (CV) disease, including left ventricular hypertrophy (LVH), increased arterial stiffness, and left ventricular (LV) dysfunction.[1-5] Additionally, data indicate that youth with BP values in the upper realm of the range that is currently considered,“ normal”, are at increased risk for CV adverse outcomes such as heart failure, myocardial infarction, and ischemic stroke.[6,7] LVH, the most studied intermediate marker, has been documented in about 40% of children and adolescents with hypertension at diagnosis, depending on the study population and definition of LVH. [2] While many factors such as obesity and length of exposure to hypertensive BPs influence the development of LVH, the LV response to BP levels is not uniform, and BP level alone does not consistently identify children and adolescents with LVH.[8] Studies in adults suggest that assessment of BP variability (BPV) may aid in the detection of individuals at risk for adverse outcomes. Increased BPV has been shown to independently influence CV morbidity and mortality in hypertensive.[9,10] As CV events are rare during childhood, studies in pediatric populations have considered the effects of higher BPV on intermediate markers of target organ damage (TOD) including left ventricular mass index (LVMI), arterial stiffness, neurocognition, and albuminuria.[11-17] Looking specifically at cardiac effects, findings regarding the possible association of BPV assessed by ambulatory blood pressure monitoring (ABPM) with increased LV mass index (LVMI) and LVH in children and adolescents.[15-17]

Although less studied, heart rate variability (HRV) has also been linked with hypertension, LVH, LV diastolic dysfunction, and adverse cardiovascular events in adults.[18-23] Lower HRV has been linked with higher BPs in normotensive male adolescents and with increased odds of having elevated BP and hypertension in healthy children.[24-27] However, information on the possible effect(s) of lower HRV on cardiovascular intermediate outcomes in hypertensive youth is limited.[28]

Here we sought to assess the effect of BP and HR variabilities on LVMI, LVH, and LV function utilizing ABPM data from the multi-center Study of High Blood Pressure in Pediatrics: Adult Hypertension Onset in Youth (SHIP AHOY). We hypothesized that increased BPV and lower HRV would be associated with higher LVMI and subclinical changes in LV function.

Methods

Data previously collected in the collaborative, multicenter SHIP AHOY study were analyzed for this sub-study, methods of which have been published.[29] Briefly, investigators at six academic centers in the United States enrolled healthy adolescents (11-<19 years) in a cross-sectional study to determine the BP level associated with target-organ effects. Exclusion criteria eliminated adolescents with secondary hypertension or on medications altering BP. A complete list of exclusions was published previously.[29] Participants and their parents/guardians provided written assent/consent as per local requirements. Each center obtained approval from the Cincinnati Children’s Institutional Review Board (Study ID#2015-2555, see supplementary file). Although not publicly available, data providing the basis for this analysis can be provided upon request.

Blood pressure measurement

Auscultatory clinic BP values were obtained by study personnel trained in proper technique by the Principal Investigators (JF, EU). Measurements were obtained at two study visits using an aneroid sphygmomanometer (Mabis^® Medkit 5, Mabis Healthcare, Inc., Lake Forest, IL, USA) per protocol.[29] Clinic BP percentile was determined using the 2017 American Academy of Pediatrics Clinical Practice Guidelines.[30] ABPM was performed at the second visit utilizing an oscillometric device for determination of awake, asleep and 24-hour SBP, DBP, and HR and dipping parameters (Spacelabs model 90927 “OnTrak,” Spacelabs Healthcare, Snoqualmie, WA, USA). The device was placed on the non-dominant arm as recommended by the manufacturer. An appropriate cuff size was selected as specified in the 2022 statement from the American Heart Association (AHA) on ambulatory blood pressure monitoring in children and adolescents.[31] The monitoring period was extended to 26 hours to ensure that a full 24 hours of data would be obtained. Participants were asked to complete a diary indicating unusual events and activities during the procedure. The time of day for the awake-asleep periods was set upon review of the participant’s diary. Readings were obtained every 20 minutes throughout the monitoring period. Adequacy of ABPMs was assessed according to the AHA statement on ABPM.[31] Data from the clinical sites were sent by secure email to the ABPM coordinating center in Seattle. Studies were interpreted by a single investigator (JTF). ABPM data were analyzed according to normative ABPM data as reported by Wühl et al. to allow for determination of individual ABPM percentile.[32]

Target organ assessment

Heart:

Echocardiography was performed at the same visit as ABPM, and images were interpreted utilizing the Cardiology Analysis System (Digisonics, Houston, TX). LV mass was calculated using the Deveraux equation[33] and indexed (LVMI) to ht^2.7.[34] Measures of LV function included global longitudinal shortening (GLS), and LV ejection fraction (LVEF) for systole (TOMTEC Corporation, Chicago IL). Diastolic dysfunction, reflected by impaired relaxation and elevated filling pressures, was assessed with the following measurements obtained by doppler and tissue doppler: 1) the ratio of early and late LV filling velocities (E/A) across the mitral valve; 2) average (septal/free wall) E/e ’and e’/a ’ velocities.[35] Diastolic dysfunction is suggested by increased E/e’, or reductions in E/A or e’a.

Vascular Stiffness:

Measurement of carotid-femoral pulse wave velocity (PWV) was performed at the second visit with the SphygmoCor CPV device (AtCor Medical, Sydney, Australia) as detailed before.[29] Briefly, the pulse transit time was calculated as follows; the difference in time from the peak of the R-waves detected by ECG to the foot of the pressure wave in the carotid artery (measured by tonometry) was subtracted from the time elapsed from the peak of R-waves to the foot of the pressure wave in the femoral artery. The PWV was then determined by the SphygmoCor CPV device (AtCor Medical, Sydney, Australia) using the distance from the carotid to femoral artery divided by the pulse transit time. [33] Measurement of the velocity provides an estimate of arterial stiffness with higher values indicating stiffer arteries.

Statistical analysis

The percentage of BP and HR dipping was calculated by dividing the difference between mean awake and asleep ABP values by the awake value x 100. Means and standard deviations (or frequencies and percents) were calculated for all parameters. Coefficients of variability (CV) were calculated for ambulatory systolic BP (SBP), diastolic BP (DBP) and heart rate (CV-SBP, CV-DBP, CV-HR) during the awake and asleep periods by dividing the standard deviation of the parameter by the mean (i.e., awake CV-SBP CV = awake SBP SD in mmHg / SBP awake mean in mmHg). General linear models were constructed to evaluate whether CV-BP and CV-HR parameters were significant determinants of target organ injury (i.e., elevated LVMI, reduced systolic and diastolic function, increased arterial stiffness). The full model contained: age, sex, race, ethnicity, waist/height ratio, BMI z-score, awake and asleep ABP systolic and diastolic percentiles, awake and asleep systolic and diastolic CV-BP, awake and asleep CV-HR and HR dipping. Logistic regression was performed to determine if addition of CV-BP and CV-HR parameters to mean ABP parameters enhanced prediction of LVH and diastolic dysfunction. The same covariates from the linear model were included. LVH was defined as LVMI >=38.6 g/m^2.7, the pediatric definition.[36] Diastolic dysfunction was defined as e’/a ’<10th% of normotensive participants (e’/a ’< 1.60974) (Figure 1). For both the linear and logistic regression, non-significant variables were removed until all remaining variables were significant at p ≤ 0.05.

Fig. 1.

ROC curves for prediction of LVH (LVM > 38.6 g/m2.7)1 and diastolic dysfunction (e’/a ’< 10th% for normotensive participants) using same variables as in linear regression. Full model for LVH included the sex, BMI z-score, mean awake ambulatory SBP%, mean ambulatory awake CV-DBP and asleep CV-BP and HR dipping. BP model only included sex, BMI z-score and mean awake ambulatory SBP%. BP variability model included sex, BMI z-score, both CV values and HR dipping. P value for difference between full model and BP model = 0.002; between BP and BP variability model = 0.008. Full model for e’/a ’included BMI z-score, mean awake ambulatory DBP%, and asleep CV-HR. There was no significant difference between C-statistics for full model and BP or between BP and HR variability models. The figures compare BP and BPV (left panel) and BP and HRV (right panel) only and are not adjusted for covariates.

Results

The study population consisted of 397 participants as shown in Table 1. The mean age was 15.6 years +/− 1.7 years with 250 participants identifying as white (63%) and 63 (16%) as Hispanic. Representation of other racial groups was limited (data not shown). Males constituted 59% (235/396) of the study population. The mean BMI z-score was 28.31± 7.6 kg/m² and the mean BMI z-score was 1.34 + 1.08. Mean clinic SBP was 123 mmHg, with a mean clinic percentile of 75%. Successful ABPM was accomplished in 375 participants. Data from the sleep period were excluded for 6 adolescents due to inadequate data for this time period. Mean ambulatory awake SBP was 124 mmHg with a mean percentile of 69%. Mean systolic and diastolic BP dipping was 12% and 19%, respectively. HR dipping was 12%. Mean LVMI was 32.69 ± 7.0 g/m^2.7. Other characteristics of the study population are detailed in Table 1.

Table 1.

Demographic and Physiologic Characteristics of Participants (N=397)

Parameter	Mean (frequency %) or Mean ± SD
Demographics
Age (years)	15.6 ± 1.7
Sex (male)	235 (59%)
Race (White)	250 (63%)
Ethnicity (Hispanic)	63 (16%)
Anthropometrics
BMI (kg/m2)	28.1 ± 7.6
BMI z-score	1.34 ± 1.08
Height (cm)	169.1 ± 9.8
Waist/Height ratio	0.53 ± 0.11
Clinic Hemodynamics
SBP (mm Hg)	122.6 + 12.4
SBP percentile*	75.2 ± 27.1
DBP (mm Hg)	71.6 ± 10.5
DBP percentile*	65.0 ± 28.8
HR (beats/min)	71.4 ± 12.2
ABPM Measures
SBP awake, mm Hg	124.2 ±11.3
SBP awake percentile*	68.7 ± 19.8
DBP awake, mm Hg	72.0 ± 7.7
DBP awake percentile*	61.5 ± 18.1
HR awake, beats/min	80.3 ± 10.6
SBP asleep, mm Hg	108.9 ± 11.0
SBP asleep percentile*	70.4 ± 19.5
DBP asleep, mm Hg	58.3 ± 7.4
DBP asleep percentile*	66.3 ± 19.5
HR asleep, beats/min	68.1 ± 10.2
SBP 24-hours, mm Hg	119.3 ± 10.9
SBP 24-hours percentile*	68.4 ± 20.4
DBP 24-hours, mm Hg	67.6 ± 7.2
DBP 24-hours percentile*	62.4 ± 18.0
HR 24-hours, beats/min	76.2 ± 9.9
SBP dipping, %	12.0 ± 6.0
DBP dipping, %	18.7 ± 8.0
HR dipping, %	12.03 + 7.25
Target Organ Assessments
LVMI (g/m2.7)	32.69 ± 7.00
Peak Longitudinal Shortening (%)	−20.34 ± 3.38
E/A ratio	2.25 ± 0.67
E/e’ ratio	6.22 ± 1.46
e'/a’ ratio	2.38 ± 0.71
PWV, (m/sec)	5.1 ± 0.8

^*Percentile based on sex and height [29,32]

LVMI; left ventricular mass index; PWV, pulse wave velocity

^*BMI = Body Mass Index; SBP = Systolic Blood Pressure, DBP = Diastolic Blood Pressure, HR = Heart Rate; LVMI = Left Ventricular Mass Index.

Percent dipping calculated by dividing the difference between awake and asleep value by awake value x 100.

Mean and standard deviation values for awake and asleep CV-BP and CV-HR are shown in Table 2. The mean CV was similar between awake and asleep periods for both systolic and diastolic BP. The mean CV-HR was 0.14 for the awake period and 0.12 for the asleep period. In multivariable linear modeling awake and asleep diastolic CV-BP and HR dipping were the only variability measures demonstrated to be significant predictors of LVMI (Table 3). Increasing BMI z-score was also positively associated with LVMI while male sex showed a negative association with LVMI. Similar analyses for other target organ effects demonstrated that increased asleep CV-HR was an independent determinate of e’/a’. No BP or HR variability measures proved to be determinants of LV systolic function, additional measures of LV diastolic function or arterial stiffness (PWV).

Table 2.

Blood Pressure and Heart Rate Coefficients of Variability

Variables	Mean	SD
CV-SBP awake	0.088	0.02
CV-SBP asleep	0.088	0.03
CV-DBP awake	0.15	0.05
CV-DBP asleep	0.14	0.05
CV-HR awake	0.14	0.04
CV-HR asleep	0.12	0.06

CV = coefficient of variation

Calculation of CV: CV = standard deviation of parameter by mean; i.e, CV-SBP awake = awake SBP standard deviation in mmHg/awake SBP mean in mmHg

Table 3.

Multivariable Models with Mean Ambulatory and BP Variability Parameters Predicting Measures of Target Organ Injury^*.

Variable	LVMI Higher adverse	SE	e'/a' Lower adverse	SE
Intercept	32.46	2.07	3.04	0.15
Sex	−2.41	0.63
BMI z-score	3.18	0.30	−0.19	0.03
SBP awake Avg%	0.04	0.02
DBP awake Avg%			−0.0090	0.002
CV-DBP awake	16.41	7.34
CV-DBP asleep	−27.32	6.87
CV-HR asleep			1.27	0.57
HR dipping	−0.170	0.04
R2	0.36		0.16

Full model contained: age, sex, race, ethnicity, waist/height ratio, BMI z-score, ABP awake systolic %, ABP asleep systolic%, ABP awake diastolic%, ABP asleep diastolic%, ABP awake systolic CV-BP, ABP asleep systolic CV-SBP, ABP awake diastolic CV-BP, ABP asleep diastolic CV-BP , awake CV-HR, asleep CV-HR, HR dipping.

^*Only beta estimates that were significant (p≤0.05) are displayed. Blank fields correspond to insignificant values.

CV, coefficient of variability; ABP, ambulatory blood pressure; SE, standard error

Logistic regression was performed to assess whether consideration of variability measures improved prediction of LVH and diastolic dysfunction over ABP measures alone. For prediction of LVH a model including the independent variables identified by linear regression (awake CV-DBP, asleep CV-DBP, HR dipping, awake SBP percentile, sex and BMI) was compared to a model composed only of the ABP parameter (plus BMI and sex). As shown in Figure 1, left panel, the combined model containing variability measures and the ABP parameter performed significantly better in the prediction of LVH compared with model comprised of the ABP parameter alone (C-statistic combined model = 0.8036, APBM model = 0.7525, p≤0.002). Similarly, a model composed of the variability measures (as defined above) did not differ significantly in predictive performance compared to the ABP model.

Evaluation of the predictive value of variability measures and ABP parameters for diastolic dysfunction (e’/a’) is shown in Figure 1, right panel. A combined logistic regression model composed of awake DBP percentile, asleep CV-HR and BMI did not improve prediction of diastolic dysfunction over consideration of models limited to either the ABP parameter or HRV measure.

Discussion

These data demonstrate that diastolic BPV as assessed by ABPM augments the prediction of LVH based solely on mean BP values. While indices of BPV did not influence LV systolic or diastolic function, asleep HRV showed a positive correlation with LV diastolic function. Our finding that BP and HR variability have effects on intermediate markers of cardiac target organ damage in a pediatric cohort constitutes a novel set of observations that increase the evidence that BP and BPV in children and adolescents contribute to the risk for adverse cardiovascular outcomes in adulthood.[6, 7, 37]

Many factors including neural, autonomic, behavioral, and environmental triggers induce variation in BP over seconds to years. It is well documented that BPV is greater in hypertensive compared with normotensive adults.[10] Investigations focusing on short term (ABP) BPV in adult populations demonstrate an association between increased BPV and enhanced risk for adverse cardiovascular outcomes and all-cause mortality.[38-42] Moreover, consideration of nocturnal BPV has been shown to improve prediction of adverse CV events and mortality in middle-aged adults when combined in models with nocturnal BP.[40] Regarding intermediate markers of TOD, short term BPV has been linked with hypertensive TOD damage, including increased arterial stiffness, reduction in LV diastolic function and increased rates of LVH.[18, 39, 43, 44] Moreover, the degree and prevalence of TOD is greater for any level of BP with increasing BPV.[39] Beyond consideration of dipping and morning BP surge, variability in both systolic and diastolic pressures while awake and asleep have been associated with increased cardiovascular risk.[40, 45] Many studies in adults have demonstrated the ill effect of increased diastolic BPV. An early longitudinal study conducted in a general population and thus, including normotensive individuals showed the importance of diastolic BPV in conveying increased risk for CV mortality.[42] Similarly, in another longitudinal study of over 9000 adults with untreated hypertension, investigators demonstrated an association between diastolic BPV and adverse CV events, CV mortality, and all-cause mortality.[38]

Long term effects of BPV in children have only recently been assessed. Analysis of data from the Bogalusa Heart Study showed that increased BPV in children and BP levels predicted hypertension in adulthood.[37] In a longitudinal study spanning 30 years, increased BPV in childhood and adolescence was associated with albuminuria in adulthood independent of BP means and cumulative BP exposure.[13] Literature considering more immediate outcomes, such as the effect of BPV on TOD, is limited in the pediatric population.[11, 12, 14-17, 46] Although some investigators have demonstrated that increased BPV in ABP is independently associated with LVMI[15, 16] others have not.[17] Differences in the methodology for setting awake and asleep periods, choice of ABP indices, and variability parameters utilized in the studies may have contributed to the discrepant results.[15-17] In a small sub-group of 40 children with primary hypertension, Lesiman et al., [16] found an independent association between awake systolic and diastolic BPV and LMVI as assessed by SD and average real variability but not with LVH. In contrast, other investigators demonstrated no relation between diastolic BP indices and LVMI or limited their analysis to systolic indices only.[15, 17] Moreover, pediatric investigations to date have been conducted in children with known hypertension or concern for hypertension and with limited participant numbers. In contrast, the SHIP AHOY study by design sought to include adolescents with BP values ranging from normal to stage 1 hypertension; thus, the study population included many healthy normotensive (> 50% of entire cohort) participants as well as youth with newly diagnosed hypertension. In this population we did not find a correlation between any measures of systolic BPV and LVMI or LV dysfunction. Similarly, none of the variability measures predicted an increase in PWV again, possibly related to the substantial number of normotensive participants and the relative distensibility of arteries in this age group. The differing results for awake and asleep diastolic variability with the former showing a positive and the latter a negative association with LVMI are puzzling and support the need for further investigation. While BPV parameters showed a limited influence on LVMI and diastolic function overall, the enhanced prediction of LVH by the combination of BPV and BP values demonstrated here are in accordance with evidence from studies conducted in adults.[43] Considered together, the available literature on the effects of increased BPV and its intersection with BP levels on LV structure and function in children is sparse.

This study also evaluated the possible association of HRV on LVMI, LVH and LV systolic function. HRV reflects the activity of the cardiac autonomic nervous system and a shift in favor of sympathetic versus parasympathetic input manifests as a reduction in HRV. The detrimental effect of such an imbalance is well recognized in the adult population where it is associated with HTN, LV dysfunction and adverse cardiovascular outcomes.[20, 22, 23] Tadic et al.[22] demonstrated an independent association between HRV parameters obtained by 24-h Holter monitor and LV diastolic function and LV longitudinal strain in a population consisting of 45 hypertensive and 63 normotensive middle-aged adults. Furthermore, hypertensives showed reduced HRV, increased LVM, reduced LV diastolic function and evidence of LV strain by 2DE and 3DE echocardiography compared with normotensives. A secondary study of data from a large prospective study evaluated 144 hypertensive middle-aged adults with 24-h ABPM, 24-h Holter, and Echocardiography and documented an association between LV diastolic dysfunction and reduced HRV.[23] Interestingly, these hypertensive adults demonstrated improvement in LV diastolic function, reduction in LV afterload and HRV one year after initiation of antihypertensive therapy.[23]

Although less studied in children, increased HR and reduced HRV are linked with high BP, clustering of cardiovascular risk factors, reduced physical activity, and obesity in children.[24-27, 46-48] Regarding TOD and HRV, Kowalewski et al.[28] assessed HRV and LVMI in 30 hypertensive children and 30 age- and gender-matched controls; though not controlled, weight and height did not differ significantly between groups. Enhanced sympathetic activity (decreased HRV) during sleep correlated with LVMI in the hypertensive children but not in controls.[28] In our cross-sectional study, asleep HRV showed an independent and positive correlation with a marker of LV diastolic function (e’/a’) but not with LVMI. This parallel relationship between HRV and e’/a ’indicates that reduced HRV was modestly associated with reduced diastolic function in this cohort. However, the impact of this association was not sufficient to improve prediction of diastolic dysfunction when combined with the ABP indices. These findings suggest that even before BP values reach the threshold for hypertension, imbalance in the autonomic nervous system may develop and engender subclinical TOD [6, 32]

Our study has certain strengths including the large number of adolescents who completed ABPM and echocardiography. However, given its cross-sectional design, causality was not established. The approach utilized for calculation of HRV parameters was novel. HRV is typically assessed from ECG recordings with measurement of R-R intervals on ECG followed by analysis with spectral density often preferred.[19] We utilized HR measurements recorded by an oscillometric ABPM device and analyzed HRV separately for awake and asleep periods. Although these measurements were obtained over a short period of time (< 60 seconds), an average of 70 readings were obtained over the course of 24 hours. Furthermore, children with heart disease were excluded and thus, the potential risk for interference with oscillometric measurements due to arrhythmias was minimal.

As our analysis considered short term BPV and HRV, it does not offer insight into the effects of BPV and HRV on youth over days, seasons, and years. The determinants of variability and the resulting impact on cardiovascular health and prognosis appear to differ when longer time periods are assessed.[9, 10, 41] Furthermore, BP measurement frequency during ABPM may influence detection of BP variability. In a recent consensus paper on blood pressure variability, the European Society of Hypertension suggested an interval of no more than 15 minutes between ABP readings when assessing BP variability.[10] While our protocol specified a measurement frequency of every 20 minutes, longer intervals are often utilized in adult and pediatric studies to minimize patient intolerance.[14, 16-18, 38, 41, 44, 45] Based on pediatric data it seems likely that the potential benefit of implementing this recommendation would likely result in an increase in the number of participants removing the monitor prior to completion. [49]

Lastly, the mechanisms generating BPV may differ across the age spectrum studied.[38, 41] We analyzed BPV across a narrow age range at one time point in normotensive and mildly hypertensive adolescents. Evidence suggests that primary hypertension in youth and young adults is due to heightened sympathetic activity and is characterized by increased cardiac output.[50] In contrast, the main driver of hypertension in older adults is increased systemic vascular resistance.[50] These differences likely influence the genesis, effects, and prognostic implications of BPV across the lifespan.[9, 38] Indeed, in older adults, impaired baroreceptive function and increased arterial stiffness are considered to be the predominant factors inducing BPV; however, in young adults, increased reactivity to stressors is implicated as the major contributor to BPV.[45] Although more research is required, recent studies conducted in young adults (mean age 36 years) indicate that BPV is a better predictor of adverse CV effects than ABP level alone – at least in this age group.[45] In contrast, evidence from cohorts of older individuals (≥ 50years of age) suggest that increased BP values are the superior predictor of CV risk with increased BPV reflecting TOD from CV disease.[45]

Conclusions

In the present study BPV assessed by ABPM appeared to add to the effect of a given BP level on the risk for TOD in adolescents. These findings may explain inconsistencies in the response of the target organs to the same BP level. Additionally, as shown here, both BPV and HRV when analyzed from ABP data were associated with deterioration in intermediate markers of TOD in adolescents. However, research on the effects of BPV and HRV on target organs is limited. Further studies are needed to confirm these results, determine which measures of variability are most useful, and establish thresholds for variability.