Insight into the 24‐hour ambulatory central blood pressure in adolescents and young adults

Abstract

This study attempted to investigate the behavior of 24‐hour central ambulatory blood pressure (ABP) in adolescents and young adults. Adolescents and young adults (age 10‐25 years) referred for elevated blood pressure (BP) and healthy volunteers had simultaneous 24‐hour peripheral (brachial) and central (aortic) ABP monitoring using the same automated upper‐arm cuff device (Mobil‐O‐Graph 24h PWA). Central BP was calculated by the device using two different calibration methods (C1SBP using peripheral systolic (pSBP)/diastolic BP and C2SBP using mean arterial/diastolic BP). A total of 136 participants (age 17.9 ± 4.7 years, 54% adolescents, 77% males, 25% volunteers, 34% with elevated peripheral ABP) were analyzed. Twenty‐four‐hour pSBP was higher than C1SBP, with this difference being more pronounced during daytime than nighttime (16.3 ± 4.5 and 10.5 ± 3.2 mm Hg, respectively, P < .001). Younger age, higher body height, and male gender were associated with greater systolic ABP amplification (pSBP‐C1SBP difference). C1SBP followed the variation pattern of pSBP, yet with smaller nighttime dip (8.4 ± 6.0% vs 11.9 ± 4.6%, P < .001), whereas C2SBP increased (2.4 ± 7.2%) during nighttime sleep (P < .001 for comparison with pSBP change). Older age remained independent determinant of larger nighttime BP fall for pSBP and C1SBP, whereas male gender predicted a larger nighttime C2SBP rise. These data suggest that the calibration method of the BP monitor considerably influences the diurnal variation in central BP, showing a lesser nocturnal dip than pSBP or even nocturnal BP rise, which are determined by the individual's age and gender.

1. INTRODUCTION

Peripheral blood pressure (BP) is known to exhibit a typical diurnal variation with a 10%‐20% fall during nighttime sleep compared to daytime values. ¹ Alterations in the diurnal rhythm, such as the absence of nocturnal fall, namely the “non‐dipping” status, or even a rising status, have been linked with increased cardiovascular risk in adults. ¹ , ² , ³ Relevant data in children and adolescents are scarce with most studies conducted in individuals with important underlying medical conditions like diabetes. ⁴ , ⁵ Several conditions and factors interfere and disrupt circadian BP variation, including increased sympathetic nervous system activity during the night, abnormal neurohormonal regulation, poor sleep quality, physical inactivity, obesity, smoking, chronic kidney disease, salt sensitivity, obstructive sleep apnea, and diabetes. ² , ⁶ , ⁷

Although there is considerable evidence on the circadian variation of peripheral BP, ¹ , ² , ³ , ⁷ , ⁸ the diurnal variation of central BP (ceBP) remains largely unexplored. Moreover, there are still no reference values for 24‐hour central ambulatory BP (ABP) monitoring and no cutoff points for defining nocturnal patterns of ceBP behavior. This is particularly important in young individuals, in whom the difference between peripheral and central systolic BP (SBP) is larger and 24‐hour ABP is regarded as mandatory for diagnosing hypertension. ⁸ This study aimed to investigate the behavior of the 24‐hour central ABP in adolescents and young adults, as estimated by a cuff‐based oscillometric device.

2. METHODS

2.1. Study design and participants

A cross‐sectional evaluation of adolescents and young adults (age 10‐25 years), referred to a Hypertension Center for suspected hypertension and apparently healthy volunteers, was performed. Participants were subjected to 24‐hour peripheral (brachial) and central (aortic) ABP monitoring. Exclusion criteria were cardiac, renal or lung disease, diabetes, evidence of secondary hypertension, severe systemic disease, acute illness, or use of drugs that might affect BP within 4 weeks before and during the study. The study protocol was approved by the Sotiria Hospital Scientific Committee. Signed informed consent was obtained from all adult participants and the adolescents' parents.

2.2. Ambulatory BP monitoring

Participants had 24‐hour ABP monitoring (measurements at 20‐minute intervals) on a routine school or workday on their non‐dominant arm. A single validated automated upper‐arm cuff oscillometric device was used, which provides simultaneous peripheral BP measurement and ceBP estimation with each reading (Mobil‐O‐Graph 24h PWA; IEM GmbH). ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ Cuff size was selected based on individual's arm circumference.

The Mobil‐O‐Graph device, upon completion of each oscillometric peripheral BP measurement, automatically re‐inflates cuff at the diastolic BP (DBP) level for about 10 seconds to record the brachial pressure waves. Upon downloading the data to the manufacturer's software (HMS version 5.1), the aortic waveform is generated by means of a generalized transfer function (in‐built ARC‐Solver algorithm). The device provides two options for calibration of brachial waveform using either brachial SBP and DBP (calibration method C1 [C1SBP]; default setting) or mean arterial BP (MAP) and DBP (calibration method C2 [C2SBP]) in order to estimate ceBP.

Participants were instructed to remain still during measurements with the monitored arm extended or relaxed. ABP monitoring was considered valid only if it had included at least 20 daytime and 7 nighttime readings. ¹ Only readings classified by the ABP monitor software as having excellent or good (grade 1‐2) signal quality were analyzed. ¹⁴ In addition, measurements with device failure to estimate central parameters together with the peripheral BP were excluded to have a direct comparison of peripheral and ceBP at the same time points. Average 24‐hour/daytime (awake)/nighttime (asleep) ABP were calculated according to the individual's reported sleeping times. The siesta measurements were not included in daytime ABP.

2.3. Definitions

In adolescents aged <16 years, 24‐hour ambulatory (or nocturnal) hypertension was defined using published normalcy tables as 24‐hour (or nighttime) peripheral SBP (pSBP) and/or DBP ≥ 95th percentile according to gender and height (as long as the values were lower than the adults' thresholds) and in older participants using the recommended thresholds for ABP in adults (24‐hour pSBP/DBP ≥ 130/80 mm Hg or nighttime ≥ 120/70 mm Hg). ² , ⁸

Based on available cutoffs for peripheral BP, participants were classified according to their percentage nighttime SBP change compared to daytime levels in two groups: (i) “dippers”: nocturnal fall ≥10% and (ii) “non‐dippers”: nocturnal fall <10% (or rise). ¹ , ⁸

2.4. Statistical analysis

The normality of continuous variables was checked using the Kolmogorov‐Smirnov test. Comparison of continuous variables in the same participants was performed with Student's paired t tests with Bonferroni correction for multiple comparisons applied where appropriate, and among subgroups with one‐way analysis of variance (ANOVA) or Kruskal‐Wallis test as appropriate. Chi‐squared test was used to compare categorical variables. Associations between quantitative variables were assessed with bivariate correlation analyses, computing either Pearson's or Spearman's correlation coefficients as appropriate. Nighttime SBP change was calculated either as difference (nighttime‐daytime BP) or percentage ([nighttime‐daytime BP]×100/daytime BP). Determinants of nighttime SBP change were assessed using multivariate linear regression analysis. Independent variables were age, gender, elevated body mass index (BMI ≥ 85th centile for individuals up to 20 years old or ≥ 25 kg/m² for older ones ¹⁵ ), and 24‐hour MAP. The diurnal variation of hemodynamic parameters was graphically presented using hourly means. SBP or pulse pressure (PP) amplification were expressed as difference (pSBP‐C1SBP or peripheral PP‐C1 PP) and ratio (pSBP/C1SBP or peripheral PP/C1 PP). The IBM SPSS Statistics 21 (SPSS Inc.) software was used. Results are expressed as means ± standard deviation (SD) or medians with interquartile range (IQR), as appropriate. A two‐sided probability value of P < .05 was considered statistically significant.

3. RESULTS

Out of 137 individuals recruited, one was excluded due to arrhythmia and 136 were analyzed. Characteristics of study participants are shown in Table 1. Mean age was 17.9 ± 4.7 years (range 10‐25 years), 74 (54.4%) were adolescents and 62 (45.6%) adults, 104 (76.5%) males, 34 (25%) healthy volunteers, 46 (33.8%) had elevated 24‐hour ABP. The median number of analyzed ABP measurements (after discarding those with failure to estimate central parameters) was 58 (IQR 49‐61; range 31‐71) for 24‐hour period, 36 (IQR 29‐42; range 13‐53) for daytime and 19 (IQR 16‐22; range 6‐35) for nighttime.

Table 1.

Participants' characteristics (mean ± SD)

	Total	Males	Females	Adolescents	Adults
Ν	136	104	32	74	62
Age (years)	17.9 ± 4.7	17.7 ± 4.4	18.2 ± 5.4	14.0 ± 1.8	22.4 ± 2.4*
Body mass index (kg/m2)	25.0 ± 4.9	25.1 ± 4.8	24.6 ± 5.4	24.7 ± 4.8	25.4 ± 5.1
Elevated body mass index (%)	73 (53.7)	56 (53.8)	17 (53.1)	45 (60.8)	28 (45.2)
Smokers (%)	10 (7.4)	7 (6.7)	3 (9.4)	2 (2.7)	8 (12.9)*
Healthy volunteers (%)	34 (24.8)	21 (20.2)	13 (40.6)*	10 (13.5)	24 (38.7)*
24 h ABP hypertension (%)	46 (33.8)	38 (36.5)	8 (25.0)	24 (32.4)	22 (35.5)
Nocturnal hypertension (%)	43 (31.6)	36 (34.6)	7 (21.8)	22 (29.7)	21 (33.9)
Peripheral 24 h systolic ABP	123.5 ± 10.6	125.5 ± 10.3	116.8 ± 8.7*	122.0 ± 8.3	125.2 ± 12.6
Peripheral 24 h diastolic ABP	70.8 ± 7.5	71.3 ± 7.8	69.1 ± 6.5	68.3 ± 6.4	73.8 ± 7.7*
Peripheral 24 h PP	52.7 ± 7.3	54.2 ± 6.7	47.7 ± 6.8*	53.7 ± 6.8	51.4 ± 7.7
Central‐1 24 h systolic ABP	109.3 ± 9.4	110.6 ± 9.6	105.2 ± 7.7*	107.2 ± 7.5	111.8 ± 10.8*
Central‐2 24 h systolic ABP	129.9 ± 13.7	133.7 ± 12.7	117.7 ± 8.5*	126.4 ± 11.2	134.1 ± 15.1*
Central‐1 24 h diastolic ABP	72.5 ± 7.6	73.1 ± 7.8	70.6 ± 6.6	70.1 ± 6.4	75.4 ± 7.9*
Central‐2 24 h diastolic ABP	72.7 ± 7.6	73.5 ± 7.8	70.3 ± 6.4*	70.0 ± 6.3	75.9 ± 7.9*
Central‐1 24 h PP	36.8 ± 4.8	37.5 ± 4.8	34.6 ± 4.4*	37.2 ± 4.5	36.3 ± 5.2
Central‐2 24 h PP	57.2 ± 11.2	60.2 ± 10.3	47.4 ± 8.0*	56.4 ± 11.0	58.1 ± 11.5
24 h mean ABP	94.9 ± 8.3	96.1 ± 8.3	90.9 ± 6.8*	92.9 ± 6.5	97.3 ± 9.5*
24 h heart rate (bpm)	71.4 ± 9.4	69.7 ± 8.5	77.0 ± 10.0*	73.8 ± 9.5	68.5 ± 8.4*
ABP amplification (difference)	14.2 ± 3.7	15.0 ± 3.5	11.5 ± 2.9*	14.8 ± 3.7	13.4 ± 3.5*
ABP amplification (ratio)	1.13 ± 0.03	1.14 ± 0.03	1.11 ± 0.03*	1.14 ± 0.04	1.12 ± 0.03*

Abbreviations: ABP, ambulatory blood pressure (mm Hg); PP, pulse pressure (mm Hg).

^* P < .05 for comparison between groups (males vs females; adolescents vs adults).

3.1. Comparison of 24‐hour peripheral vs central SBP

pSBP was higher than C1SBP during both daytime (difference 16.3 ± 4.5 mm Hg, range 7‐30 mm Hg, P < .001) and nighttime (10.5 ± 3.2 mm Hg, range 5‐24 mm Hg, P < .001, Table 2). However, C2SBP did not differ from pSBP during daytime (0.1 ± 7.0 mm Hg, p = NS), but exceeded pSBP by 18.4 ± 10.9 mm Hg during nighttime (P < .001; Table 2).

Table 2.

Hemodynamic parameters from 24‐h ambulatory blood pressure monitoring

Parameter	24‐h	Daytime	Nighttime
Peripheral SBP	123.5 ± 10.6	129.0 ± 10.7	113.5 ± 10.3*
Central‐1 SBP	109.3 ± 9.4	112.7 ± 9.8	103.1 ± 9.6*
Central‐2 SBP	129.9 ± 13.7	128.9 ± 13.3	132.0 ± 16.2*
Peripheral DBP	70.8 ± 7.5	76.2 ± 8.1	61.5 ± 6.9*
Central‐1 DBP	72.5 ± 7.6	78.1 ± 8.1	62.8 ± 7.1*
Central‐2 DBP	72.7 ± 7.6	77.8 ± 8.1	64.1 ± 7.3*
Heart rate (bpm)	71.4 ± 9.4	78.0 ± 10.4	59.3 ± 9.1*
Peripheral PP	52.7 ± 7.3	52.8 ± 7.8	52.0 ± 7.2*
Central‐1 PP	36.8 ± 4.8	34.6 ± 5.4	40.2 ± 5.7*
Central‐2 PP	57.2 ± 11.2	51.1 ± 11.1	67.9 ± 13.4*
Central‐1 SBP amplification (difference)	14.2 ± 3.7	16.3 ± 4.5	10.5 ± 3.2*
Central‐1 SBP amplification (ratio)	1.13 ± 0.03	1.15 ± 0.04	1.10 ± 0.03*
Central‐1 PP amplification (difference)	15.9 ± 4.0	18.2 ± 4.8	11.8 ± 3.5*
Central‐1 PP amplification (ratio)	1.48 ± 0.12	1.58 ± 0.16	1.32 ± 0.10*
Mean arterial pressure (mm Hg)	94.9 ± 8.3	100.4 ± 8.5	85.3 ± 7.9*
Stroke volume (mL)	76.5 ± 10.1	70.0 ± 11.1	88.0 ± 12.0*
Cardiac output (L/min)	5.2 ± 0.4	5.3 ± 0.5	5.1 ± 0.6*
Cardiac index (L/min/m2)	2.8 ± 0.4	2.8 ± 0.4	2.7 ± 0.5*
Augmentation index (%)	16.2 ± 5.2	15.8 ± 5.2	17.3 ± 7.5*
Total vascular resistance (s×mm Hg/mL)	1.1 ± 0.1	1.2 ± 0.1	1.1 ± 0.1*
Pulse wave velocity (m/s)	5.0 ± 0.4	5.1 ± 0.4	4.8 ± 0.3*

Abbreviations: PP, pulse pressure (mm Hg); SBP, systolic blood pressure (mm Hg); DBP, diastolic blood pressure (mmHg).

^* P < .001 for comparison between daytime and nighttime.

3.2. Determinants of 24‐hour peripheral‐central SBP amplification

Twenty‐four‐hour SBP amplification correlated inversely with age (r = −.27 for difference; −.35 for ratio, both P < .01) and positively with height (r = .20 for difference, P < .05; for ratio P = NS), but did not differ according to BMI groups (increased or normal BMI; for both difference and ratio P = NS). Males had higher SBP amplification than females (P < .001, Table 1), even after adjusting for age and height (ANCOVA P < .01). Hypertensives had higher SBP amplification (difference, not ratio) than normotensives (13.5 ± 3.6 vs 15.5 ± 3.4 mm Hg; P < .01 and 1.13 ± 0.03 vs 1.13 ± 0.03; P = NS).

3.3. Diurnal variation of central BP

Twenty‐four‐hour diurnal variation of ceBP, peripheral BP, and other parameters from pulse wave analysis is presented in Figures 1and 2 and Figures S1‐S3. Thirty‐seven (27.2%) participants reported daytime nap (siesta) during ABP monitoring.

Figure 1.

Diurnal variation of central and peripheral blood pressure and other parameters from pulse wave analysis. C1SBP, central systolic BP estimated using calibration with SBP/DBP; C2SBP, central systolic BP estimated using calibration with MAP/DBP; DBP, diastolic BP; MAP, mean arterial pressure; SBP, peripheral systolic BP

Figure 2.

Diurnal variation of systolic blood pressure and pulse pressure amplification

C1SBP followed the variation pattern of pSBP exhibiting a minor dip in the afternoon (3‐5 pm; concurrently with the daytime sleep [siesta] in some participants) and during the night (10 pm‐6 am; Figure 1). More specifically, there was a gradual decrease in both parameters (more pronounced for pSBP) from 10 pm, reaching nadir and subsequent plateau between 4 and 6 am, followed by increase thereafter. Similar variation patterns were observed for DBP, MAP, heart rate, total vascular resistance, SBP amplification, cardiac output, and cardiac index (Figure 2, Figure S2).

However, C2SBP slightly increased during nighttime (Figure 1). The pattern of diurnal change of C2SBP simulated those observed for augmentation index and stroke volume (Figures S2 and S3).

Peripheral PP showed a late minor nocturnal dip (Figure S1). C1 and C2 PP exhibited a clear nocturnal rise, more pronounced for the latter (Figure S1).

Systolic blood pressure or PP amplification exhibited a progressive fall starting earlier than peripheral and central BP, from 7 pm, that was mainly the result of relative increase of C1SBP, while pSBP reached plateau in the same time interval (Figures 1, 2).

3.4. Nocturnal behavior

Forty‐three participants (31.6%) had nocturnal hypertension. Τhe percentage nighttime changes in pSBP, C1SBP, and C2SBP were normally distributed (Kolmogorov‐Smirnov tests P > .05, Figure 3). Nighttime dip in C1SBP (9.6 ± 6.9 mm Hg or 8.4 ± 6.0%) was lower than that of pSBP (15.4 ± 6.1 mm Hg or 11.9 ± 4.6%, P < .001; Tables 2, 3). However, average C2SBP was higher during night (nocturnal change 3.1 ± 9.2 mm Hg or 2.4 ± 7.2%; Tables 2, 3). A non‐dipping behavior (nighttime dip < 10% or rise) was detected in 47 (34.6%), 80 (58.8%), and 132 (97.1%) participants based on pSBP, C1SBP, and C2SBP evaluation (P < .001 for comparison, Figure 4). In 13 (9.6%) and 80 (58.8%) participants, C1SBP and C2SBP increased during night indicating a “reverse dipping” pattern (Figure 3).

Figure 3.

Distribution of nocturnal systolic blood pressure change in the study participants

Table 3.

Nocturnal systolic BP (SBP) change (mm Hg; percentage in parentheses)

Nocturnal SBP change	Total	Males	Females	Adolescents	Adults
Peripheral	−15.4 ± 6.1 (−11.9 ± 4.6)	−15.9 ± 5.9 (−12.1 ± 4.2)	−13.9 ± 6.8 (−11.4 ± 5.5)	−14.1 ± 6.3 (−11.0 ± 4.8)*	−17.1 ± 5.5 (−13.1 ± 4.0)
Central‐1	−9.6 ± 6.9 (−8.4 ± 6.0)	−10.0 ± 6.8 (8.6 ± 5.7)	−8.6 ± 7.3 (−7.9 ± 6.9)	−7.7 ± 7.1 (−6.8 ± 6.3)*	−12.0 ± 5.9 (−10.3 ± 5.0)
Central‐2	3.1 ± 9.2 (2.4 ± 7.2)	4.0 ± 9.5 (3.1 ± 7.3)*	0.1 ± 7.4 (0.3 ± 6.3)	3.4 ± 9.1 (2.7 ± 7.1)	2.7 ± 9.5 (2.0 ± 7.3)

^* P < .05 for comparison vs the other subgroup.

Figure 4.

Patterns of nighttime dipping of systolic blood pressure. Chi‐square P < .001 among peripheral, central‐1 and central‐2 measurements; P < .001 between any two measurements

The magnitude of nocturnal fall of pSBP and C1SBP did not differ between males and females, but was larger in adults than adolescents (all P < .01). In contrast, nighttime change of C2BP was larger in males than females (P = .03), with no difference between adolescents and adults (Table 3). The magnitude of nocturnal change of peripheral and central SBP did not differ neither according to BMI group (increased vs normal) nor between normotensives and hypertensives (P = NS).

In multivariate regression analyses, older age remained an independent determinant of larger nighttime BP fall in pSBP (R ² = .09, β = −.29, P = .001) and C1SBP (R ² = .15, β = −.52, P < .001), whereas male gender predicted larger nighttime rise of C2SBP (R ² = .04, β = 3.27, P = .03; Table S1).

4. DISCUSSION

This study aiming to understand the behavior of 24‐hour central ABP in 136 untreated and apparently healthy young individuals, as assessed with a cuff‐based device, showed that: (a) pSBP was higher than C1SBP during daytime and nighttime (by 16.3 and 10.5 mm Hg, respectively); (b) younger age, higher body height, and male gender were associated with larger SBP amplification; (c) C1SBP followed the variation pattern of pSBP with dip during nighttime sleep, but C2SBP followed a different pattern; (d) both pSBP and central SBP exhibited a normal distribution in the nighttime change, but C1SBP had lower mean nighttime dip than pSBP (8.4% vs 11.9%) while C2SBP was increased during nighttime (by 2.4% compared to daytime); (e) older age was an independent determinant of larger nighttime dip for pSBP and C1SBP, whereas male gender predicted a larger nighttime rise of C2SBP.

Central diurnal SBP variation was investigated using C1 and C2 calibration methods. C2 is known to provide a more accurate estimation of central aortic BP, as shown by invasive validation studies in adults. ¹⁶ On the other hand, a study in children and adolescents showed greater overestimation of aortic SBP with the calibration method C2 instead of C1. ¹⁷ In addition, pSBP is known to be underestimated by the oscillometric device in adults, ¹² , ¹⁸ whereas relevant data in children are unclear. ¹⁷ , ¹⁹ A single study reported that the Mobil‐O‐Graph overestimated pSBP in children, yet intra‐arterial pSBP was indirectly measured. ¹⁷ These findings might be supported by a recent study in older adults highlighting the age effect on the difference between cuff BP and invasive BP measurements, which showed that cuff SBP overestimated aortic SBP in those aged 40‐49 years, but with each older age decade, there was a progressive shift toward increasing underestimation. ²⁰ Nevertheless, C2 method should not be used for evaluation of BP amplification, neither in pediatric nor in adult populations, because it leads to erroneous assessments (usually negative BP amplification which is biologically improbable). However, it has been suggested that ceBP obtained with C2 method is superior to C1 in terms of association with target‐organ damage and outcomes. ²¹ , ²² , ²³ , ²⁴ , ²⁵

Systolic blood pressure amplification is highly variable within and among individuals. ²⁶ Non‐modifiable (older age, female gender, and shorter height) and modifiable factors (hypertension, dyslipidemia, diabetes, smoking, and vasoactive drugs) can predict lower amplification. The underlying mechanism is alteration of “timing‐synchronization” of forward and reflected waves that is influenced by large arteries stiffness, peripheral arterial resistance, heart rate, and characteristics of wave reflections. ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ The present study confirmed that age, gender, and height correlate with ABP amplification.

In line with previous studies using tonometric or oscillometric 24‐hour ABP monitoring, this study showed that BP amplification is decreased during nighttime. ³¹ , ³² , ³³ , ³⁴ From a physiological point of view, this observation could be attributed to the postural change during nighttime, as supine position during sleep has been associated with increased pulse wave reflections (as reflected on the augmentation index, Figure S3). ³¹ , ³⁵ , ³⁶ Heart rate and arterial stiffness (pulse wave velocity) also decrease during sleep, contributing to lower BP amplification (Figures S2 and S3). A period of relaxation beginning in the early evening resulted in progressive decline of BP amplification from 7 pm that preceded nocturnal dip (Figure 2), similarly to previous relevant studies. ³¹ , ³²

The most intriguing finding is the conflicting results on the nocturnal behavior of ceBP derived from the two different calibration methods used for its estimation. The present study mostly confirms previous reports on the diurnal variation of hemodynamic parameters in adults (decrease of average pSBP, DBP, MAP, C1SBP, heart rate, peripheral PP, BP amplification, cardiac index, total vascular resistance, pulse wave velocity, and increase of C1PP, C2PP and augmentation index during nighttime). ³¹ , ³² , ³³ , ³⁴ , ³⁷ However, the different nocturnal pattern (dominant rise) of C2SBP has not been previously reported and, in fact, does not coincide with the dipping demonstrated in a study by Argyris et al ³⁴ in older adults (mean age 54 years), where the same device and calibration method were used. Nevertheless, the overall trend for the wake‐sleep changes among peripheral, C1 and C2 SBP is similar in the two studies. This unexpected result may reflect an age‐specific circadian rhythm, complex hemodynamic changes during resting in young individuals with elastic aorta and arterial tree, or technical insufficiency of the device in estimating ceBP, especially in ambulatory conditions or changing body posture. ³⁵ More specifically, considering multiple physiological differences between adults and children (body size, vascular wall properties, heart rate, and BP levels), transfer functions developed for non‐invasive ceBP estimation in adult population might not be accurate in young individuals. ³⁸ Furthermore, such concerns may also arise for the accuracy of brachial waveform recording via the cuffs and of the estimation of brachial BP and MAP through the oscillometric curve that can affect the validity of calibration methods. ¹⁶ , ¹⁷ More research is needed to ascertain and elucidate such findings. In any case, this study suggests that the established cutoffs for defining dipping patterns of peripheral BP are invalid for such classifications for ceBP.

The findings of this study should be interpreted in the light of important limitations. First, the relatively small sample size including a selected population of young individuals referred to a hypertension clinic and healthy volunteers. Second, peripheral and central BP values were obtained by the same device, which relies the estimation of the latter on data derived from peripheral waveform (also an advantage). Third, the device has been validated for ceBP measurements only in static (not ambulatory) conditions and in adult populations. Recently, two published studies in young individuals (mean age 8‐9 years) with history of heart disease evaluated the Mobil‐O‐Graph against invasive measurements and reported conflicting results about its accuracy on the ceBP approximation in children and adolescents. ¹⁷ , ¹⁹ Whether these findings are relevant for the population of our study, remains questionable.

In conclusion, this work attempted to “understand the behavior” of this method and device, rather than to suggest anything related to clinical practice—at least at the present time. Central hemodynamics in young individuals exhibited circadian variability. However, day‐to‐night BP changes seemed to be different from those of pSBP and the established thresholds for defining nocturnal BP behaviors cannot be used. More research is necessary to investigate certain patterns in the young individuals and determine their clinical relevance. Moreover, further investigation is needed to determine the validity of novel technologies for non‐invasive ceBP estimation in youth and the optimal calibration method.

CONFLICT OF INTEREST

The authors declare that they have no conflict of interest.

AUTHOR CONTRIBUTIONS

Angeliki Ntineri, MD, Anastasios Kollias, MD, PhD, and George S Stergiou, MD, PhD, FRCP involved in study conception and design. Angeliki Ntineri, MD, Maria Elena Zeniodi, MD, and George S Stergiou, MD, PhD, FRCP involved in acquisition of data. Angeliki Ntineri, MD, Andriani Vazeou, MD, and Alexandra Soldatou, MD, PhD involved in analysis and interpretation of data. Angeliki Ntineri, MD, Anastasios Kollias, MD, PhD, and George S Stergiou, MD, PhD, FRCP involved in drafting of manuscript. Angeliki Ntineri, MD, Anastasios Kollias, MD, PhD, Maria Elena Zeniodi, MD, Andriani Vazeou, MD, Alexandra Soldatou, MD, PhD, George S Stergiou, MD, PhD, FRCP involved in critical revision.

Supporting information

Supplementary Material

Ntineri A, Kollias A, Zeniodi ME, Vazeou A, Soldatou A, Stergiou GS. Insight into the 24‐hour ambulatory central blood pressure in adolescents and young adults. J Clin Hypertens. 2020;22:1789–1796. 10.1111/jch.13999