Multiple office blood pressure monitoring for the diagnosis of hypertension in children

Abstract

The measurement (M) of blood pressure (BP) in children may be challenging, and the ideal approach, free of critical issues (such as the interference due to the white coat effect or to the high variability), has yet to be identified. Herein, we compare BP as measured with multiple office BP monitoring (mOBPM) with the standard approach as suggested by guidelines. A cohort of healthy children underwent two mOBPMs 1 year apart. Mean systolic and diastolic values obtained by mOBPM were compared with the 1st, 2nd, 3rd, and 4th measurements (repeated measures ANOVA). mOBPMs with a coefficient of variation (CV) > 15% were excluded. The number of children with BP > 90th centile was determined based on (a) each of the initial three readings, (b) the mean of the 2nd and 3rd Ms, (c) the 4th M, (d) the mOBPM at baseline, and (e) 1 year apart. Out of 164 enrolled children, 13 (7.9%) were excluded because of a CV > 15%. The analysis on the remaining 151 children showed that the first three Ms provided a significantly higher BP than the mOBPM. The 4th M was the first one to be aligned with the results of the mOBPM. Based on the 1st, 2nd, 3rd, and 4th Ms, a BP > 90th centile was observed in 29, 20, 21, and 16 children, respectively. The mean of the 2nd and 3rd Ms identified 12 children with high BP, while the mOBPM revealed elevated BP in only 6 children and this finding was confirmed only in 3 of them 1 year later.

Conclusion: The first three readings systematically overestimate BP, while the 4th one better aligns with the mOBPM. If the 4th reading is abnormal, the complete mOBPM will likely offer a more reliable BP assessment.

What is Known: • Blood pressure measurement may be challanging and initial readings systematically overestimate real values.

What is New: • The 4th blood pressure reading better aligns with the mean of 10 measurements. In daily clinical practice, if the 4th reading is abnormal, the complete mOBPM (www.mobpm.com) will likely offer a more reliable BP assessment.

Introduction

Several guidelines and studies recommend to regularly measure blood pressure (BP) in children [1–5]. However, the US Preventive Services Task Force and the UK National Screening Committee recently underlined that there is insufficient evidence to assess the balance of benefits and harms of screening children for high BP given the difficulties to obtain reliable BP values [6, 7].

In fact, the measurement (M) of BP in children may be challenging.

Office-based BPM is inexpensive and widely available, and thus represents the first approach to BP assessment. Nevertheless, its low specificity may result in false positivity [6–8], leading to unnecessary investigations and/or inappropriate treatments. Consistently, current guidelines insist on performing multiple measurements over time [1, 2].

Ambulatory BP monitoring (ABPM), although potentially more accurate, is time-consuming, expensive, and of limited access. Moreover, being poorly tolerated by children, it can provide false positive results. In addition, ABPM should be performed only above 5 years of age [1, 2] due to the lack of reference values for heights < 120 cm. Finally, the reference values for ABPM are derived from a relatively small population including only central European white children; thus, they may not be applicable to non-Caucasian children [9, 10].

In the adult setting, automated office BP measurement (AOBPM), consisting of multiple BP readings recorded by an automated oscillometric device with the patient resting quietly and alone, was shown to reduce the white coat effect and to better correlate both with the mean awake ambulatory BP and with organ damage [11]. Hence, AOBPM is currently recommended and supported for the diagnosis of hypertension in adults [12–14]. In the pediatric setting, although it may have a role for BP measurement [15], AOBPM is not routinely used, and it has been associated with poor sensitivity [16]. This leaves room for research into an alternative, straightforward, cost-effective, and accessible in-office approach that provides more reliable BP values in children.

To address some of the troubles in measuring BP in children, we regularly use multiple office BP monitoring (mOBPM) [8]. Herein, we compare BP values obtained with mOBPM and with the standard method suggested by current guidelines in a cohort of healthy children.

Methods

mOBPM consists of 10 serial BP readings taken after 5 min of rest, on the non-dominant arm, at 3-min intervals, by mean of a validated automated oscillometric device (OMRON M3, Omron Healthcare, Hoofddorp, The Netherlands), possibly in the absence of healthcare providers and using a cuff of appropriate width (approximately 40% of arm circumference) and length (80–100% of arm circumference). After discarding outlier values (< 5th and > 95th centiles of the 10 recorded values), the coefficient of variation (CV) and the mean of the remaining systolic (S) and diastolic (D) readings are calculated by a custom-built software [8, 17].

In the context of a multicenter study, we performed two sets of mOBPMs, 1 year apart, in a cohort of children enrolled between October 1, 2022, and February 28, 2023, from 10 different primary schools located both in Italy and San Marino Republic, upon informed parental consent. All healthy children born at term, after a regular pregnancy, with a normal weight at birth, were considered eligible. Children with current or previous renal disease, suffering from any chronic systemic disease, or taking any drug were excluded. mOBPMs were performed in a quiet school setting familiar to the children, after having measured their height and weight, to minimize the anxiety induced by the presence of healthcare personnel and the potential effect on BP values [18]. The number of subjects with a BP > 90th centile was determined considering (a) each of the first three readings, (b) the mean of the 2nd and 3rd Ms, (c) the 4th M, (d) the mean BP value obtained with the mOBPM at baseline, and (e) 1 year later. BPMs were normalized for age, sex, and height¹. Statistical significance was tested by ANOVA for repeated measurements.

Ethical approval was granted by our Ethics Committee (Milano Area 2, approval no. 1119_2022). The study was performed in accordance with the ethical standards laid down in the Declaration of Helsinki and its amendments.

Results

Between October 1, 2022, and February 28, 2023, 164 healthy children were enrolled but 13 (7.9%) were excluded from the analysis for a CV > 15% in either SBP (n = 2) or DBP (n = 11). The remaining 151 (86 females, 57.0%) had a mean age of 8.6 ± 0.34 years, a mean weight of 31.1 ± 6.9 kg, a mean height of 132.7 ± 6.2 cm, and a mean BMI of 17.5 ± 2.82 kg/m² (28 with a BMI > 20 kg/m²). The mean CVs of included mOBPMs were 6% and 4% for SBP and DBP, respectively.

Figure 1, panels A and B, shows BP values obtained at each M and with the mOBPM (as box plot): the 1st, 2nd, and 3rd Ms were significantly higher than those obtained with the mOBPM, for both systolic and diastolic values, while the 4th M was not different from values obtained with the mOBPM. To better appreciate the differences in BP obtained with the different approaches to BP measurement, values are also provided as means (± SD) in Table 1 both in mmHg and in SDS. The statistically significant difference observed between BP as measured at baseline mOBPM compared with the one performed 12 months later is explained by the aging of children.

Fig. 1.

Systolic (panel A) and diastolic (panel B) blood pressure SDS as determined by the 1st, 2nd, 3rd, and 4th measurements, as mean of the 2nd and 3rd values and as mOBPM in a cohort of healthy children (the pink area correspond to the interquartile range of values obtained from the initial mOBPM). Prevalence of subjects with higher-than-normal BP at single measurements and with mOBPM at baseline and 12 months later (panel C). Legend: SDS, standard deviation score; mOBPM, multiple office blood pressure measurement. *Statistical significance by ANOVA; red line: mOBPM mean

Table 1.

Mean (± SD) values of BP in mmHg and SDS as measured with different approaches to BP determination in a cohort of 151 healthy children

	Systolic BP		Diastolic BP
	mmHg	SDS	mmHg	SDS
1st reading	98.1 ± 12.9*	− 0.25 ± 1.01*	66.1 ± 9.3*	0.54 ± 0.77*
2nd reading	95.6 ± 11.9*	− 0.55 ± 0.96*	64.6 ± 8.6*	0.37 ± 0.70*
3rd reading	94.3 ± 11.4*	− 0.69 ± 0.91*	64.6 ± 10.0*	0.32 ± 0.81*
4th reading	92.8 ± 11.0	− 0.77 ± 0.90	62.5 ± 9.5	0.19 ± 0.82
Mean of 2nd and 3rd readings	94.9 ± 10.1*	− 0.62 ± 0.80*	64.6 ± 7.9*	0.35 ± 0.64*
mOBPM at baseline	92.4 ± 8.0	− 0.81 ± 0.70	62.6 ± 6.7	0.20 ± 0.59
mOBPM at 12 months	94.5 ± 7.5*	− 0.78 ± 0.62*	63.7 ± 7.0*	0.21 ± 0.57*

Legend: BP, blood pressure; mOBPM, multiple office blood pressure measurement; SDS, standard deviation score; SD, standard deviation

^*Statistical significance compared with baseline mOBPM by ANOVA

Based on the 1st, 2nd, 3rd, and 4th Ms, a BP > 90th centile was observed in 29, 20, 21, and 16 children, respectively (Fig. 1, panel C). When the average of the 2nd and 3rd Ms was used, as suggested by guidelines, only 12 children were identified with higher-than-normal BP. However, the mOBPM indicated that only 6 had an elevated BP, and this finding was confirmed in only 3 of them, with the mOBPM performed 1 year apart. These 3 patients, with confirmed hypertension, were addressed to a global diagnostic assessment (physical examination, laboratory investigations, and echocardiography) to identify the cause of the high BP, whose results go beyond the scope of the present paper.

None of the children with a normal BP value at the 4th reading showed an abnormal mOBPM both at baseline and 1 year later.

Discussion

BPM is a fundamental aspect of the physical examination, and obtaining accurate and reliable values is critical for the diagnosis of hypertension. This can be particularly challenging in children, due to several factors including the white coat phenomenon that can cause significant variability in BP values and misleading results.

To address these challenges, we propose an analytical method that reduces variability in office-based BPMs, filters outlier values, and minimizes anxiety-induced interference. mOBPM is straightforward, simple, cost-effective, and accessible to any pediatrician. As such, mOBPM has the potential of identifying hypertensive patients and avoiding unnecessary investigations. Although mOBPM may resemble AOBPM, it differs because the former provides an elaborated average of multiple BP readings after outlier values have been filtered. In addition, mOBPM calculates the CV of introduced values offering clinicians an indication of the reliability of the obtained means. Finally, the custom-built software, freely available as webApp to anyone (www.mobpm.com), automatically normalizes the obtained systolic and diastolic BP means for age, sex, and height, thereby streamlining the clinical workflow and minimizing the risk of errors [17].

Current guidelines on hypertension recommend measuring BP three times [1, 2]. However, based on our results, the initial three readings are higher than subsequent ones, while the 4th reading better aligns with values obtained with the mOBPM [8]. This is consistent with the fact that BP level can decrease during a single visit and, consequently, repeated and averaged BPM can be more reliable. Thus, we suggest that the 4th M can better represent the real BP level of patients, but, in case of initial abnormal values, the complete mOBPM will likely offer a more reliable BP assessment (Fig. 2). Additionally, in research settings, mOBPM greatly reduces the sample size needed by minimizing BP variability.

Fig. 2.

Algorithm for the identification of suspected hypertensive patients with mOBPM (estimated from 151 healthy children)

Among the possible limitations of the present study, there is the lack of reference values collected with mOBPM. Moreover, we cannot count on a validation of our approach based on organ damage. However, it should be underlined that also the current approach, suggested by guidelines, is not based on organ damage [1]. Another possible limitation is that in the present study BP was determined by using an oscillometric device rather than the auscultatory method. However, our study was aimed at comparing BP values of different readings, all obtained with the same oscillometric device. Furthermore, nowadays, the oscillometric approach is widely used with the additional advantage of preventing observer bias and terminal digit preference [19], which can be even more critical in a research setting compared with the clinical practice [20].

The present study is not meant to be a validation of mOBPM, rather a comparison of BP values obtained with the current standard of care and those obtained by extending the measurements to 10 determinations. The comparison shows some systematic inconsistencies with measurements taken by the current method with the risk of overestimating the number of hypertensive children. These inconsistencies can be overcome by increasing the number of BP readings: the 4th reading minimizes the risk of misdiagnosing the child as hypertensive and, even more so, if the entire mOBPM is recorded (Fig. 2).

In conclusion, the first three readings systematically overestimate BP, while the 4th one better aligns with the mOBPM. In daily clinical practice, if the 4th reading is abnormal, the complete mOBPM will likely offer a more reliable BP assessment.

Abbreviations

M

Measurement

BP

Blood pressure

BPM

Blood pressure measurement

OBPM

Automated office blood pressure monitoring

ABPM

Ambulatory blood pressure monitoring

AOBPM

Automated office blood pressure measurement

mOBPM

Multiple office blood pressure measurement

CV

Coefficient of variation

S

Systolic

D

Diastolic

BMI

Body mass index

SDS

Standard deviation score

Author contribution

L.D. conceptualized and designed the study, performed data entry, carried out the initial analyses, drafted the initial manuscript, reviewed and revised the manuscript. M.C.M., T.R., L.V. and P.S. carried out the initial analyses, drafted the initial manuscript, reviewed and revised the manuscript. M.V. and C.O. contributed to the methodological aspects of the study, and reviewed and revised the manuscript. D.R., G.T., T.N., C.T., L.S., A.G. and E.S. collected data, and approved the final version of the manuscript. T.M. performed data entry, carried out the initial analyses, reviewed and revised the manuscript. G.A. conceptualized and designed the study, designed the data collection instruments, coordinated and supervised data collection, carried out the initial analyses, and critically reviewed and revised the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Funding

The study was supported by Associazione Bambino Nefropatico, Fondazione G. e D. De Marchi and Progetto Alice ETS. Associazione Bambino Nefropatico, Fondazione G. e D. De Marchi and Progetto Alice ETS had no role in the design and conduct of the study.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare no competing interests.