Accuracy of Home Blood Pressure Monitoring in Youth: A Systematic Review and Meta-Analysis

Abstract

Objective

To evaluate the accuracy of home blood pressure monitoring (HBPM) in youth compared with the reference standard of ambulatory BP monitoring (ABPM).

Study design

MEDLINE, Embase, Web of Science, and Cochrane Library were systematically searched from inception to November 13, 2024. Two authors identified studies testing the accuracy of HBPM compared with ABPM in youth ≤18 years. We assessed outcomes on the continuous scale (mean difference [MD]) or categorical scale (sensitivity, specificity, κ). We conducted a meta-analysis using hierarchical random effects weights.

Results

Of 2966 titles screened, 26 studies were included (total N=1992 patients, median 59 per study). Daytime home BP was lower than awake ambulatory BP (systolic MD, −6.4 mmHg [95% CI, −10.7 to −2.0]; diastolic MD, −3.3 mmHg [95% CI, −6.0 to −0.7]). Diagnosing daytime ambulatory hypertension by HBPM had modest sensitivity (0.52-0.64) and good specificity (0.82-0.97), with moderate agreement (κ 0.50-0.65). Accuracy was higher for white coat hypertension (sensitivity, 0.77-0.89; specificity, 0.85-0.94; κ, 0.66-0.73) and specificity was slightly higher for masked hypertension (sensitivity, 0.23-0.38; specificity, 0.92-0.96; κ, 0.27-0.36). Nocturnal home BP was higher than asleep ambulatory BP (systolic MD, 2.6-5.0 mmHg; diastolic MD, 2.2-3.2 mmHg) with moderate agreement (κ, 0.33-0.49). Limitations included different hypertension thresholds and HBPM protocols.

Conclusions

The accuracy of HBPM may not be sufficient to replace ABPM as a single test. However, HBPM may be a practical alternative when ABPM is not available, particularly when clinic and home BPs are concordant. Future work is needed to determine the best approach to HBPM in clinical practice.

Hypertension affects 4% of children globally¹ and is associated with premature cardiovascular morbidity and mortality in adulthood.² Although screening for hypertension in children is recommended at annual well-visits starting at 3 years old,³ the accuracy of conventional office blood pressure (BP) measurement is limited in children due to the high prevalence of white coat hypertension (high BP in clinic with normal BP outside clinic)^4–6 and masked hypertension (normal BP in clinic with high BP outside clinic).⁷ Therefore, the 2017 American Academy of Pediatrics (AAP) clinical practice guideline on hypertension recommends 24-hour ambulatory BP monitoring (ABPM) to confirm hypertension for any child with high office BP and suggests strongly considering ABPM in children with risk factors for masked hypertension like obesity or chronic kidney disease.³

However, ABPM is not widely available in pediatric primary care and typically requires referral to subspecialists,⁸ which may not be feasible for the >20% of US children with obesity who need evaluation for masked hypertension.⁹ In a large, national sample of children diagnosed with hypertension after the 2017 AAP guidelines, only 8% had ever completed ABPM.¹⁰ In addition, ABPM can be poorly tolerated for the full 24 hours, limits a child’s activities, requires time to obtain and return the device, involves expensive hardware and software, and is poorly reimbursed by insurance. In contrast to ABPM, in which a programmed device automatically takes repeated BP measurements over 24 hours, home BP monitoring involves a patient measuring their BP multiple times per day over several days. Home BP monitoring is now recommended to confirm hypertension in adults.^{11, 12} Due to the convenience and ease of self-monitoring, adult patients prefer home BP monitoring over ABPM.¹³ Although home BP monitoring is used by 70% of pediatric nephrologists,^{14, 15} the 2017 AAP guideline states that additional data on the accuracy of home BP monitoring are needed before recommending it for diagnosing hypertension in children.³

The objective of this systematic review and meta-analysis is to evaluate the accuracy of home BP monitoring in children and adolescents compared with the reference standard of ABPM. We hypothesized that home BP monitoring is an accurate out-of-office modality but may miss children with masked hypertension if nighttime measurements are not included in the home BP monitoring protocol. This work could inform whether home BP monitoring can provide a feasible and effective modality to measure out-of-office BP in the pediatric primary care and subspecialty settings.

Methods

This systematic review protocol has been registered in the International Prospective Register of Systematic Reviews database (registration number: CRD42023463208). The systematic review is being reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement.¹⁶

Search Strategy

MEDLINE, Embase, Web of Science, and the Cochrane Library were systematically searched from inception to November 13, 2024. No date or language limits were imposed on the search. Non-English articles were translated to English using Google Translate.¹⁷ An experienced medical librarian was consulted on methodology and performed a medical subject heading (MeSH) analysis of known key articles provided by the research team [mesh.med.yale.edu]. Scoping searches were performed in each database and an iterative process was used to translate and refine the searches. To maximize sensitivity, the formal search used controlled vocabulary terms and synonymous free-text words. The search strategies for each database are shown in Table I. Reviewers checked for additional relevant cited and citing articles using included studies.

Table 1 Online.

Search Strategy and Number of Studies Identified by Database.

Ovid MEDLINE(R) ALL
1	(self-measur* or “out of office” or Home or Homes or House or Domicil*).mp.	456146
2	exp pediatrics/ or exp child/ or exp adolescent/ or exp young adult/ or (adolescen* or child* or early adult* or juvenile* or kid or kids or paediatric* or pediatric* or preadolescen* or pubescence or pubescent or schoolchild* or teen or teens or teenager* or toddler* or youth* or young adult* or young patient* or young people or young person*).mp.	4848531
3	blood pressure determination/ or blood pressure monitoring, ambulatory/ or ((blood pressure or BP) and (measure* or monitor* or reading* or evaluat* or determine* or record*)).mp.	356387
4	1 and 2 and 3	1180
Embase (Ovid)
1	home monitoring/ or (self measur* or “out of office” or Home or Homes or House or Domicil*).mp.	669264
2	exp pediatrics/ or exp child/ or exp adolescent/ or young adult/ or (adolescen* or child* or early adult* or juvenile* or kid or kids or paediatric* or pediatric* or preadolescen* or pubescence or pubescent or schoolchild* or teen or teens or teenager* or toddler* or youth* or young adult* or young patient* or young people or young person*).mp.	5549913
3	blood pressure measurement/ or blood pressure monitoring/ or ((blood pressure or BP) and (measure* or monitor* or reading* or evaluat* or determine* or record*)).mp.	563178
4	1 and 2 and 3	2082
Web of Science Core Collection (Clarivate)
1	TS=(“self measur*” or “out of office” or Home or Homes or House or Domicil*)	982,283
2	TS=(adolescen* or child* or “early adult*” or juvenile* or kid or kids or paediatric* or pediatric* or preadolescen* or pubescence or pubescent or schoolchild* or teen or teens or teenager* or toddler* or youth* or “young adult*” or “young patient*” or “young people” or “young person*”)	3,504,116
3	TS=((“blood pressure” or BP) and (measure* or monitor* or reading* or evaluat* or determine* or record*))	332,674
4	1 and 2 and 3	802
Cochrane (Wiley)
1	(self NEXT measur*):ti,ab,kw OR (“out of office”):ti,ab,kw OR (Home or Homes or House or Domicil*):ti,ab,kw	71200
2	(adolescen* or child* or early NEXT adult* or juvenile* or kid or kids or paediatric* or pediatric* or preadolescen* or pubescence or pubescent or schoolchild* or teen or teens or teenager* or toddler* or youth* or young NEXT adult* or young NEXT patient* or young NEXT people or young NEXT person*):ti,ab,kw	405987
3	(blood NEXT pressure or BP) NEXT/5 (measure* or monitor* or reading* or evaluat* or determine* or record*)	26612
4	1 and 2 and 3	257
	Cochrane Database of Systematic Reviews – 9 results  Cochrane Central Register of Controlled Trials – 248 results

All searches were run on November 13, 2024.

Eligibility Criteria

We included observational studies testing the accuracy of home BP monitoring compared with the reference standard of ABPM in children and adolescents ≤18 years old. Studies were included if they compared the results of home BP monitoring and ABPM on either the continuous scale or the categorical scale (eg, hypertensive vs normotensive). Conference abstracts were eligible for inclusion if the necessary data were available in the abstract.

Study Selection

Search results were pooled in EndNote and de-duplicated. This set of search results was uploaded to Covidence for screening. Every title and abstract were screened by two independent authors (JN, VC). Any title and abstract identified for inclusion by at least one author was reviewed in the full text stage. One author (JN) first reviewed the full text of eligible articles for inclusion in the systematic review. A second author (VC) then reviewed the full text in turn and verified inclusion or exclusion of each article.

Data Extraction

We collected the following information from each study: first author, publication year, year(s) the study was conducted, study country, study setting, retrospective or prospective study design, study sample, inclusion and exclusion criteria, age and sex of participants, number of participants, type of home BP monitor, home BP monitoring schedule, type of ABPM device, ABPM schedule, number of participants with hypertension, mean BP on home BP monitoring, mean BP on ABPM (including daytime, nighttime, and/or 24-hour mean BP values as available). For each home BP monitor, we searched stridebp.org, validatebp.org, and any pertinent references cited by the included study to determine if the device had been validated in children, adults, or both. These websites provide listings of validated BP devices and are sponsored by the European Society of Hypertension and American Medical Association, respectively.^{18, 19} For studies that evaluated the performance characteristics of home BP monitoring at detecting the categorical outcome of hypertension, we reported the study’s diagnostic criteria for hypertension by the index test of home BP monitoring and by the reference standard of ABPM. For these studies, we extracted a 2x2 table if available containing the number of true positives (hypertension on home BP monitoring and ABPM), false positives (hypertension on home BP monitoring but normal on ABPM), false negatives (normal on home BP monitoring but hypertension on ABPM), and true negatives (normal on home BP monitoring and ABPM). For the categorical outcomes, we included studies that reported any of the following: sensitivity, specificity, positive predictive value, negative predictive value, overall agreement, and/or Cohen’s kappa (κ). Data from each article were extracted by one author (JN) then verified by a second author (VC). Disagreements in article selection or data extraction were resolved by discussion to consensus. If two studies assessed the accuracy of home BP monitoring in an overlapping cohort of children, the study with the larger sample size was used to avoid duplication. For one study,²⁰ we used PlotDigitizer²¹ to extract mean BP data from a figure. We evaluated methodological quality using the Quality Assessment of Diagnostic Accuracy Studies, version 2 (QUADAS-2) tool.²²

Statistical Analysis

For studies that compared home BP monitoring and ABPM on the continuous scale, we identified the paired mean difference (as home BP value minus ABPM value) and corresponding 95% confidence interval. Depending on the data available in each study, we reported mean differences between daytime home BP results and (1) awake mean BP on ABPM; and/or (2) 24-hour mean BP on ABPM. We extracted the reported mean difference with 95% confidence interval when available. When such data were not available, we used established methods^{23, 24} to calculate the mean difference and 95% confidence interval based on reported means, standard errors or standard deviations, and the reported correlation coefficient between home BP and ambulatory BP (with r=0.7 imputed^25–27 if not reported in the individual study). For studies that performed nocturnal home BP monitoring, we reported the paired mean difference between nighttime mean home BP and asleep mean BP on ABPM using the same methods as above. Because most studies were conducted at the same hypertension center in Greece, we used robust variance estimation with hierarchical random effects weights to account for correlated effect sizes from clustering by center.^{28, 29} We performed univariate meta-regression to explore the effect of study-level covariates (sample size, mean home BP, mean ambulatory BP, publication year, mean age) on effect size, with P<0.01 considered statistically significant as is recommended when calculating meta-regression coefficients with robust variance estimation on <40 studies.²⁹ Consistent with past literature using robust variance estimation,^{29, 30} we report the I² statistic for illustrative purposes without significance levels. The risk of bias assessment was not used for any weighting of the effect estimates.^{23, 31}

For studies that evaluated accuracy on the categorical scale and provided data for a 2x2 table, we calculated the sensitivity, specificity, and positive and negative likelihood ratios for each study. When available, we also evaluated the accuracy of combining office and home BP monitoring results to diagnose hypertension as defined by the reference standard of ABPM. Due to heterogeneous study methods and hypertension thresholds, we did not pool the results for the categorical outcomes and we present the results as ranges. When available, we also report ranges for overall diagnostic agreement and κ. Statistical analyses were conducted using Stata/SE version 17.0 (StataCorp) and Review Manager version 5.4 (Cochrane Collaboration).

Results

Study Selection

The PRISMA flow diagram for search results is shown in Figure 1. Of the 2966 unique titles and abstracts screened, 196 full-text articles were assessed for eligibility, and 26 studies met inclusion criteria for our systematic review, involving 1992 patients. The most common reasons for excluding full-text articles were wrong study design (n=72), inclusion of adults in the study sample, (n=36), and no comparison with ABPM (n=23).

Figure 1.

Flow Diagram for the Selection of Studies Evaluating the Accuracy of Home Blood Pressure Monitoring Compared With Ambulatory Blood Pressure Monitoring in Youth.

Study Samples

The median number of patients who underwent home BP monitoring and ABPM in each study was 59 (min-max range, 6-215). Table II presents the methodology, sample, exclusion criteria, BP device data, and hypertension thresholds for each study. Of the 26 included studies, 16 were conducted in Greece, 4 in the US, 3 in Brazil, 1 in India, 1 in United Kingdom, and 1 as part of a multicenter study in 13 European countries. Most studies assessed the accuracy of home BP monitoring compared with ABPM in children and adolescents referred to a hypertension clinic for high BP.^{25, 26, 32–45} Three studies described patients with type 1 diabetes mellitus.^46–48 Glenn et al and Wühl et al described patients with chronic kidney disease taking antihypertensive medications^{49, 50} and Adalat et al described patients with end-stage kidney disease on peritoneal dialysis.⁵¹ Two studies were single-arm trials evaluating the efficacy of amlodipine (Tallian et al)⁵² and candesartan (Franks et al)⁵³ that used home BP monitoring and ABPM to determine BP control. The study by Brady et al was part of a quality improvement collaborative in six pediatric primary care clinics in which children with elevated BP were identified to undergo home BP monitoring.⁵⁴ Vitetzaki et al evaluated home BP monitoring and ABPM in normotensive children and adolescents with overweight or obesity.⁵⁵

Table 2.

Study Characteristics for Articles Included in the Systematic Review (N=26).

Source	Country	Study Type	Setting	No. Patients	Age Range (Mean±SD)	Study Population	BP Devices	HBPM Schedule	HBPM Thresholds	ABPM Schedule	ABPM Thresholds	HBPM-ABPM Timing	HBPM Device Validation
Tallian, 1999	USA	P	Pediatric nephrology clinic	19	1-18y (13.1±3.6)	HTN confirmed by ABPM then enrolled in single-arm trial on amlodipine. ABPM repeated when home BP was <95%ile.	ABPM: SpaceLabs 90202, 90207 HBPM: Starline manual BP for small cuffs; Omron HEM-412C for larger cuffs	BID	≥95%ile for age/sex (reference norms NS)	q20min	NS	NS	NS
Stergiou, 2004	Greece	P	HTN center	55	6-18y (12.3±2.9)	Referred for HTN on ≥2 clinic visits. Exclude: hx of HTN, chronic disease, anti-HTN Rx	ABPM: SpaceLabs 90207, 90217 HBPM: Omron HEM-705CP, IC; Omron 711 IS if AC <20cm	Duplicate readings BID on 3 workdays/wk x2wks. Valid: ≥12 readings, ≥4 days	No thresholds used	q20min. Valid: ≥30 awake and ≥12 sleep readings	Awake BP ≥90%ile for sex/height1	<2-3 weeks	✓ adults
Wühl, 2004	13 European countries	P	Pediatric nephrology units in ESCAPE trial	118	3-19y (11.2±4)	Mild-moderate CKD, mean 24h SBP >50%ile and/or controlled by anti-HTN Rx, on ramipril. HBPM optional in trial.	ABPM: SpaceLabs 90207 HBPM: Omron MX 1 BP	≥3 HBP readings over 2-7 days around ABPM.	≥95%ile for sex/height2	Day: q15min Night: q30min	Awake BP ≥95%ile for sex/height1	<7 days	✓ children
Stergiou, 2005 (Blood Press Monit)	Greece	P	HTN center	16	8-17y (13.3±2.9)	Referred for HTN on ≥2 clinic visits. Exclude: hx of HTN, clinic BP>160/110, chronic disease, anti-HTN Rx	ABPM: SpaceLabs 90207, 90217HBPM: Omron HEM-705CP, IC; Omron 711 IS if AC <20cm	Duplicate readings BID on 3 workdays/wk x2wks. Valid: ≥12 readings, ≥4 days	No thresholds used	q20min.Valid: ≥30 awake and ≥12 sleep readings	NS	<2-3 weeks	✓ adults
Stergiou, 2005 (Am J Hypertens)	Greece	NS	NS	58 total (n=29 age 12-19y)	12-19y (NS)	Normotensive with type 1 DM	NS	6 workdays	NS	Daytime	NS	NS	NS
Franks, 2008	USA	P	Single-center uncontrolled trial on candesartan	11	6-18y (14.2±3.4)	≥20kg, HTN (≥3 OBP ≥95%ile or anti-HTN Rx). Exclude: ARB hypersensitivity, acute HTN, renal/hepatic dysfunction, coarct, transplant, pregnancy, breastfeeding.	ABPM: SpaceLabs 90207 HBPM: Omron HEM-741C	BID	NS	Day: q15min Night: q30min	NS	NS	NS
Stergiou, 2008 (J Hypertens)	Greece	P	HTN center	102	6-18y (12.8±2.9)	Referred for HTN. Exclude: anti-HTN Rx, DM, chronic disease, secondary HTN, stage 2 HTN at 2 visits	ABPM: SpaceLabs 90207, 90217 HBPM: Omron 705IT, HEM-705CP, IC, 711 IS	Duplicate readings BID on 6 school days in 2wks. Valid: ≥12 readings, ≥4 days	Arsakeion School study3	q20min. Valid: ≥20 awake readings	Awake BP ≥95%ile for sex/height4	<2 weeks	✓ children and adults
Stergiou, 2008 (Am J Hypertens)	Greece	P	HTN center	100	6-18y (13.0±2.8)	Referred for HTN. Exclude: DM, chronic disease, secondary HTN, stage 2 HTN at 2 visits, any BP-affecting med within 4 weeks	ABPM: SpaceLabs 90207, 90217 HBPM: Omron HEM-705CP, IC, 705 IT; Omron 711 IS or 705IT for AC <20cm	Duplicate readings BID on 6 school days in 2wks. Valid: ≥12 readings, ≥4 days	NS	q20min. Valid: ≥20 awake and ≥8 sleep readings	NS	<2 weeks	✓ children and adults
Stergiou, Nasothimiou, 2009	Greece	P	HTN center	48	6-17y (11.3±3.1)	Clinic BP ≥90%ile and/or home BP ≥95%ile.3 Exclude: DM, systemic disease, acute illness, anti-HTN Rx	ABPM: SpaceLabs 90207, 90217 HBPM: Omron 705IT	BID for 3 workdays over 1wk. Valid ≥ 6 readings	≥95%ile for sex/height3	q20min for 24 hrs; Valid: ≥20 awake readings	German Working Group on Pediatric HTN (1997)4	NS	✓ children and adults
Stergiou, Alamara, 2009	Greece	P	HTN center	50	12-28y (20±3.8)	Type 1 DM. Exclude: hx HTN, Rx BP-affecting med, proteinuria, serious chronic disease, fever, UTI	ABPM: SpaceLabs 90207, 90217 HBPM: Omron HEM-705CP, IC	BID x6 workdays in 2 wks. Valid ≥ 12 readings, ≥ 4 days	Adolescents: Arsakeion School study;3 Adults: ESH thresholds5	q20min. Valid: ≥30 awake and ≥12 sleep readings.	Adolescents: ≥95%ile1 Adults: ESH thresholds6	<2 weeks	✓ adults
Furusawa, 2011	Brazil	P	Outpatient clinic	40	NS (12.1±3.6)	HTN. Exclude: stage 2 HTN, arrhythmia, DBP=0, auscultatory gap; <80% successful HBPM + ABPM readings	ABPM: SpaceLabs 90207 HBPM: Omron HEM-705 CP	Triplicate readings BID x14 days, preferably left arm	4th Task Force age/sex office thresholds7	Day: q10min; Night: q15min. Valid: ≥80% successful	German Working Group on Pediatric HTN (1997)4	ABPM 1 day before HBPM	✓ adults
Salgado, 2011	Brazil	P	HTN clinic	109	5-15y (9.8±2.6)	Referred for HTN (by 4th Report).7 Exclude: BP-affecting meds, AC >30cm	ABPM: SpaceLabs 90207 HBPM: Omron-705 CP	Duplicate readings BID x7 days. Valid: ≥24 readings; 1st day discarded	NS	Day: q20min Night: q30min. Valid: ≥70% successful, ≥1/hr	NS	NS	✓ adults
Stergiou, 2011	Greece	P	HTN center	81	6-18y (13±3)	Pt in Arsakeion study3 with clinic/home BP >90%ile. Exclude: DM, anti-HTN Rx, secondary HTN, stage 2 HTN, renal, cardiac, or systemic disease.	ABPM: SpaceLabs 90207, 90217 HBPM: Omron 705IT	BID x6 workdays in 2wks. Valid: ≥12 readings	≥95%ile by Arsakeion School Study3	q20min. Valid: ≥11 awake and ≥7 sleep readings	Mean 24h ≥95%ile8	<2 weeks	✓ children and adults
Vitetzaki, 2011	Greece	P	NS	48	4-14y (median 11.1)	Normotensive with overweight/obesity	NS	6 days	NS	24 hours	NS	NS	NS
Ntineri, 2015	Greece	NS	HTN center	186	5-20y (12.3±3.1)	Referred for HTN.	Validated oscillometric device	Duplicate readings BID x7 days	NS	24 hours	Awake ABPM (norms NS)	<6 weeks	NS
Stergiou, 2015	Greece	R	HTN clinic	642 total (n=82 age 5-12y; n=95 age 13-17y)	5-12y (9.7±2.2); 13-17y (14.5±1.3)	Referred for HTN. Exclude: systemic disease, DM, secondary HTN, acute illness, BP-affecting meds	ABPM: SpaceLabs 90207, 90217 HBPM: Omron 705IT, 705CP, 711 IS	BID x6 workdays in 2 wks. Valid: ≥12 readings	NS	q20min. Valid: ≥20 awake and/or ≥7 sleep readings	≥95%ile1	<4 weeks	✓ children and adults
Póvoa, 2017	Brazil	P	School screening	133	12-17y (15.0±1.6)	n=100 BP ≥90%ile,7 n=33 normal BP. Exclude: handicap, pregnancy, chronic disease, anti-HTN Rx, antidepressant, anxiolytic, antiinflammatory, contraceptive, Tanner stage 1	ABPM: SpaceLabs 90207 HBPM: Omron HEM-705 CP	BID duplicate readings x6d; total 24 readings	NS	Day: q15min Night: q20min. Valid: ≥70% successful	NS	<1 week	✓ adults
Ntineri, 2019	Greece	P	HTN center	89	12-25y (18.4±4.4)	Referred for elevated BP and healthy volunteers	Oscillometric device	BID duplicate readings x7d	12-15y: ≥95%/<95%ile or ≥135/<85; 16y: ≥135/<85 [isolated systolic HTN]	q20min	12-15y: ≥95%/<95%ile (or ≥130/<80); ≥16y: ≥130/<80 [isolated systolic HTN]	NS	NS
Zeniodi, 2020	Greece	R	HTN center	251 total (n=78 age 6-11y; n=137 age 12-17y)	6-11y (10.1±1.5); 12-17y (14.4±1.5)	Healthy volunteer or referred for HTN. Had OBP, ABP, HBP, and TOD eval. Exclude: DM, chronic disease, secondary HTN, acute illness, BP-affecting Rx w/i 4w	ABPM: SpaceLabs 90207, 90217; Microlife WatchBP O3 HBPM: Omron 705IT; Microlife WatchBP Home	BID duplicate readings x7 workdays in 2 wks. Valid: ≥12 readings, ≥3 days	<16y: Arsakeion School study;3 ≥16y: ≥135/855	q20min. Valid: ≥20 awake and ≥7 sleep readings	<16y: 2016 ESH Guidelines1,9 ≥16y: 24h ≥130/80, awake ≥135/85, sleep ≥120/706	<2 weeks	Omron: ✓ children and adults Microlife: ✓ adults
Stambolliu, 2021	Greece	P	HTN center	91	6-18y (13.3±2.8)	Referred for HTN. Exclude: systemic disease, BP-affecting med, DM, secondary HTN, acute illness	ABPM: Microlife WatchBP O3 HBPM: Microlife WatchBP Home N	Day: BID duplicate readings x7d in 2wks; Night: hourly readings x3 starting 2hrs after bedtime x3 nights. Valid: ≥12 day and ≥3 sleep readings	Night: 6-15y: ≥95%ile for sex/height1,9 ≥16y: ≥120/70	q20min. Valid: ≥20 awake and ≥7 sleep readings	Night: 6-15y: ≥95%ile for sex/height1,9 ≥16y: ≥120/70	<2-3 weeks	✓ adults
Stergiou, 2021	Greece	P	HTN Center	58	6-18y (13.0±2.9)	Referred for HTN. Exclude: secondary HTN, systemic disease, acute illness, BP-affecting med w/i 4w, AC ≥42cm, stage 2 HTN	ABPM: SpaceLabs 90217, 90207; Microlife WatchBP O3 HBPM: Omron 705IT; Microlife WatchBP Home N	BID duplicate readings x7d in 2wks. Valid ≥12 readings over 3d	<16yo: Arsakeion School study3 16-18yo: adult thresholds10	q20min. Valid: ≥20 awake and ≥7 sleep readings.	<16yo: German Working Group on Pediatric HTN (2002)1 16-18yo: adult thresholds10	<4 weeks	Omron: ✓ children and adults Microlife: ✓ adults
Brady, 2022	USA	P	6 urban pediatric primary care clinics	6	3-22y (10.0±5.1)(for the 92 patients that consented)	English-speaking pts w/ h/o elevated BP invited for HBPM. Only pts ≥8yo w/discordant OBP and HBPM had ABPM.	NS	BID triplicate readings x7d.	2017 AAP clinic BP thresholds11	NS	NS	NS	NS
Glenn, 2022	USA	P	3 pediatric nephrology clinics	103	11-19y (15.0±2.7)	Had CKD and on anti-HTN Rx x6mo. Exclude: non-English speaking, sibling in study, cognitive impaired	ABPM: SpaceLabs 90217A-1 HBPM: Omron HEM-907XL	RA attended pt’s home q6mo x2y (5 visits total) to perform triplicate BP readings	BP index: 95%ile by 2017 AAP norms11	Day: q20min Night: q30min Valid: ≥40 readings, ≥1/hr12	BP index ≥1 and/or BP load ≥25%1	ABPM <2 weeks after HBPM	✓ adults
Menti, 2022	Greece	NS	NS	24	12-30y (19.8±5.3)	Type 1 DM	ABPM: Microlife WatchBP O3 HBPM: Microlife WatchBP Home N	BID duplicate readings x7d. Night: 3x/night for 3 nights	<16y: ≥95%ile ≥16y: ≥120/70	q20min	NS	NS	✓ adults
Adalat, 2023	UK	R	Pediatric nephrology clinic	24	5-17y (median 13.2)	On peritoneal dialysis (2-11mo post-initiation); 71% on anti-HTN Rx.	NS	7-day mean auscultatory post-dialysis BP performed by parents	>90%ile per 2016 ESH guidelines9	NS	>90%ile per 2016 ESH guidelines9	NS	NS
Puthukara, 2024	India	P	Pediatric nephrology and HTN clinic	60	5-18y (10.7±2.8)	BMI z-score ≥2. Exclude: congenital heart disease, CKD, systemic illness w/HTN, congenital malformation, malignancy, bone marrow transplant, BP-elevating Rx.	ABPM: SpaceLabs 90227 Home: Omron HEM 7121	TID triplicate readings (with 5-min intervals) x7d	Arsakeion School study3	Day: q20min. Night: q30min. Valid: >40 readings	5-12y: mean day or night BP >95%ile1 13-18y: lower of either adult thresholds or 95%ile13	ABPM <1 week after HBPM	✓ adults

Abbreviations: NS, not specified; R, retrospective; P, prospective; y, years; w, weeks; SD, standard deviation; BP, blood pressure; ABPM, ambulatory blood pressure monitoring; HBPM, home blood pressure monitoring; AC, arm circumference; ESCAPE, Effect of Strict Blood Pressure Control and ACE Inhibition on the Progression of Chronic Renal Failure in Pediatric Patients; SBP, systolic blood pressure; DBP, diastolic blood pressure; HTN, hypertension; BID, twice a day; ARB, angiotensin receptor blocker; DM, diabetes mellitus; Rx, treatment; OBP, office blood pressure; TOD, target organ damage; AAP, American Academy of Pediatrics; Pts, patients; ESH, European Society of Hypertension; RA, research assistant; CKD, chronic kidney disease; mo, months

^1.Wühl E, Witte K, Soergel M, Mehls O, Schaefer F. Distribution of 24-h ambulatory blood pressure in children: normalized reference values and role of body dimensions. J Hypertens. 2002;20:1995-2007.

^2.de Man SA, André JL, Bachmann H, Grobbee DE, Ibsen KK, Laaser U, Lippert P, Hofman A. Blood pressure in childhood: pooled findings of six European studies. J Hypertens. 1991 Feb;9(2):109-14.

^3.Stergiou GS, Yiannes NG, Rarra VC, Panagiotakos DB. Home blood pressure normalcy in children and adolescents: the Arsakeion School study. J Hypertens. 2007;25:1375-9.

^4.Soergel M, Kirschstein M, Busch C, Danne T, Gellermann J, Holl R, et al. Oscillometric twenty-four-hour ambulatory blood pressure values in healthy children and adolescents: A multicenter trial including 1141 subjects. The Journal of Pediatrics. 1997;130:178-84.

^5.Parati G, Stergiou GS, Asmar R, Bilo G, De Leeuw P, Imai Y, et al. European Society of Hypertension guidelines for blood pressure monitoring at home: a summary report of the Second International Consensus Conference on Home Blood Pressure Monitoring. Journal of Hypertension. 2008;26:1505-26.0

^6.O’Brien E, Asmar R, Beilin L, Imai Y, Mancia G, Mengden T, Myers M, Padfield P, Palatini P, Parati G, Pickering T, Redon J, Staessen J, Stergiou G, Verdecchia P; European Society of Hypertension Working Group on Blood Pressure Monitoring. Practice guidelines of the European Society of Hypertension for clinic, ambulatory and self blood pressure measurement. J Hypertens. 2005 Apr;23(4):697-701.

^7.National High Blood Pressure Education Program Working Group on High Blood Pressure in children and Adolescents. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics 2004; 114:555–579.

^8.Lurbe E, Cifkova R, Cruickshank JK, Dillon MJ, Ferreira I, Invitti C, et al. Management of high blood pressure in children and adolescents: recommendations of the European Society of Hypertension. J Hypertens. 2009;27:1719-42.

^9.Lurbe E, Agabiti-Rosei E, Cruickshank JK, Dominiczak A, Erdine S, Hirth A, et al. 2016 European Society of Hypertension guidelines for the management of high blood pressure in children and adolescents. Journal of Hypertension. 2016;34:1887-920.

^10.Williams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, Burnier M, Clement DL, Coca A, de Simone G, Dominiczak A, Kahan T, Mahfoud F, Redon J, Ruilope L, Zanchetti A, Kerins M, Kjeldsen SE, Kreutz R, Laurent S, Lip GYH, McManus R, Narkiewicz K, Ruschitzka F, Schmieder RE, Shlyakhto E, Tsioufis C, Aboyans V, Desormais I; Authors/Task Force Members:. 2018 ESC/ESH Guidelines for the management of arterial hypertension: The Task Force for the management of arterial hypertension of the European Society of Cardiology and the European Society of Hypertension: The Task Force for the management of arterial hypertension of the European Society of Cardiology and the European Society of Hypertension. J Hypertens. 2018 Oct;36(10):1953-2041.

^11.Flynn JT, Kaelber DC, Baker-Smith CM, Blowey D, Carroll AE, Daniels SR, et al. Clinical Practice Guideline for Screening and Management of High Blood Pressure in Children and Adolescents. Pediatrics. 2017;140:e20171904.

^12.Flynn JT, Daniels SR, Hayman LL, Maahs DM, McCrindle BW, Mitsnefes M, et al. Update: Ambulatory Blood Pressure Monitoring in Children and Adolescents. Hypertension. 2014;63:1116-35.

^13.Flynn JT, Urbina EM, Brady TM, Baker-Smith C, Daniels SR, Hayman LL, Mitsnefes M, Tran A, Zachariah JP, null. Ambulatory blood pressure monitoring in children and adoles-cents: 2022 update: a scientific statement from the American Heart Association. Hypertension 2022; 79:e114–e124.

Study Quality

The results of the quality assessment of individual studies are shown in Figure 2A Online and summarized across all included studies in Figure 2B Online. Fifteen of 26 studies had unclear or high applicability concerns around patient selection due to inclusion of specific study samples that may not represent the general pediatric population. Nearly all studies (24 of 26) had low risk of bias around the conduct and interpretation of the index test. Twelve studies had unclear or high applicability concerns around the definition of ambulatory hypertension by the reference standard used in the study. Ten studies had unclear or high risk of bias related to patient flow due to lack of information about the timing of home BP monitoring and ABPM.

Figure 2 Online.

Risk of Bias and Applicability Concerns (A) for Each Included Study and (B) Summarized Across All Included Studies.

BP Devices

Most studies used Omron home BP devices, although four studies used Microlife WatchBP home BP devices,^{42–44, 46} including the two studies by Stambolliu et al and Menti et al that incorporated nocturnal home BP readings.^{43, 46} For 24-hour ABPM, most studies used SpaceLabs ambulatory BP devices, yet four studies used Microlife WatchBP ambulatory BP devices.^{42–44, 46} According to the available validated device listings, most studies used home BP devices and ambulatory BP devices that had been validated in adults but not in children.

BP Monitoring Schedules

Studies used a variety of home BP monitoring schedules. Studies based in Greece generally used duplicate morning and evening readings (with a 1-minute rest between readings), with an adequate home BP monitoring result defined as having ≥12 readings. The two single-arm antihypertensive efficacy trials by Tallian et al and Franks et al used twice daily home BP readings over an unspecified time period.^{52, 53} Triplicate readings in the morning and evening were used in two studies.^{36, 56} One study in Brazil (Furusawa et al) used triplicate readings in the morning and evening for 14 days;³⁶ two other studies in Brazil (Salgado et al and Póvoa et al) used duplicate readings in the morning and evening for 6-7 days and defined an adequate study as having ≥24 valid readings.^{25, 40} The one study in India (Puthukara et al) used triplicate readings, with 5-minute intervals, thrice daily at fixed times for 7 days.⁴⁵ One US-based study by Glenn et al used attended home BP measurements in which a research assistant obtained three BP readings 30 seconds apart during a home visit, which was completed every 6 months for 2 years.⁴⁹ In the UK study of children on peritoneal dialysis by Adalat et al, parents measured an auscultatory home BP post-dialysis for 7 days.⁵¹ For the two studies with nocturnal home BP monitoring, the nocturnal home BP device (Microlife Watch BP Home N) was programmed to take 3 BP measurements at 1-hour intervals starting 2 hours after going to sleep.^{43, 46} Patients were advised to complete nocturnal home BP readings for 3 nights for a total of 9 nocturnal home BP readings.^{43, 46} In studies that provided information about the ABPM schedule, ABPM readings were obtained every 15 to 20 minutes during the daytime and every 20 to 30 minutes overnight.

Hypertension Thresholds on Home BP Monitoring

In the 11 studies that categorized hypertension status by daytime home BP and ABPM results (Table III), the thresholds for hypertension varied across studies. For home BP readings, five studies applied the home BP norms from the 2007 Arsakeion School study⁵⁷ to define hypertension as an average daytime home BP ≥95^th percentile for sex and height in four studies^{34, 37, 42, 45} and >90^th percentile in one study.⁵¹ Four studies defined hypertension at home using office BP norms.^{27, 36, 49, 56} Ntineri et al focused on isolated systolic hypertension, defined as a home systolic BP ≥95^th percentile with home diastolic BP <95^th percentile for children aged 12-15 years and a home BP ≥135/<85 for ages 16-25 years but did not provide the reference used for the percentiles.⁴¹ One study did not specify home thresholds.³⁸

Table 3.

Results of Studies Reporting Diagnostic Test Characteristics of Home BP Monitoring Compared With ABPM.

Studies Reporting Sensitivity and Specificity
Study	Sample Size (% HTN on ABPM)	Index Test	Reference Standard	TP	FP	FN	TN	Sensitivity (95% CI)	Specificity (95% CI)	LR+ (95% CI)	LR− (95% CI)	Other Outcomes
Wühl, 2004	118 (22.9%)	7-day home BP	Awake ABPM	14	16	13	75	0.52 (0.32-0.71)	0.82 (0.73-0.90)	2.95 (1.66-5.24)	0.58 (0.39-0.87)	Clinic + Home BP: Sens 0.41, Spec 0.92 Clinic or Home BP: Sens 0.81, Spec 0.60 Clinic BP x3 + Home: Sens 0.33, Spec 0.98
Stergiou, 2008	102 (30.4%)	6-day home BP	Awake ABPM	17	6	14	65	0.55 (0.38-0.71)	0.92 (0.83-0.96)	6.49 (2.83-14.88)	0.49 (0.33-0.73)	HTN: κ 0.50 (0.31-0.69) WCH: Sens 0.89, Spec 0.92, κ 0.73 (0.56-0.89) MH: Sens 0.36, Spec 0.96, κ 0.36 (0.07-0.66)
Ntineri, 2015	186 (9.7% sustained HTN in clinic and ABPM; 3.8% MH; 9.1% WCH)	7-day home BP	Awake ABPM					0.64	0.97			Sustained HTN: Agreement 92%, κ 0.65 WCH: Sens 0.77, Spec 0.94, Agreement 92%, κ 0.66 MH: Sens 0.38, Spec 0.92, Agreement 87%, κ 0.27 24h ABPM: “similar” results
Zeniodi, 2020	251 (27.1%)a	7-day home BP	24-hour ABPM	48	18	20	165	0.71 (0.58-0.81)	0.90 (0.85-0.94)	7.18 (4.51-11.42)	0.33 (0.23-0.47)	Agreement 84.9%, κ 0.61
Glenn, 2022	103 patients, 251 paired HBPM-ABPM results (41.8%)	Attended home BP by RA on 1 day	24-hour ABPM	92	34	13	112	0.88 (0.81-0.94)	0.77 (0.69-0.86)	3.76 (2.78-5.09)	0.16 (0.10-0.27)	Agreement 81.8% Awake ABPM: Sens 0.96, Spec 0.83, Agreement 88.5%
Puthukara, 2024	60 (43.3%)	7-day home BP	24-hour ABPM	8	4	18	30	0.31 (0.17-0.50)	0.88 (0.74-0.95)	2.62 (0.88-7.75)	0.78 (0.59-1.04)	Agreement 63.3% 3-day HBPM: Sens 0.15, Spec 0.91 WCH: Sens 0.88, Spec 0.85 MH: Sens 0.23, Spec 0.94
Studies Reporting Other Measures of Accuracy or Agreement
Study	Sample Size (% HTN on ABPM)	Index Test	Reference Standard	Other Outcomes
Furusawa, 2011	40 (75%)	14-day home BP	Awake ABPM	κ 0.56
Stergiou, 2011	81 (19.8%)	6-day home BP	24-hour ABPM	Agreement 85%, κ 0.53
Ntineri, 2019	89 (23.6% systolic HTN on HBPM)	7-day home BP	24-hour ABPM	Age 12-15y (N=55): Agreement 81.8%, κ 0.44 Age 16-25y (N=34): Agreement 79.4%, κ 0.40
Stambolliu, 2021	91 (22.0% nocturnal HTN)	3-night home BP	Asleep ABPM	Nocturnal HTN: Agreement 82%, κ 0.49 Non-dipping status: Agreement 57%, κ 0.19
Brady, 2022	6	7-day home BP	ABPM	Agreement 33.3% Note: Patients had ABPM only if office BP and home BP results were discordant
Menti, 2022	24 (13%)	3-night home BP	Asleep ABPM	Nocturnal HTN: Agreement 88%, κ 0.33 Non-dipping status: Agreement 38%, κ 0.032
Adalat, 2023	24 (25%/44% sys/diastolic HTN)	7-day home BP	24-hour ABPM	Systolic HTN: PPV 65%, NPV 83% Diastolic HTN: PPV 60%, NPV 50%

BP, blood pressure; ABPM, ambulatory BP monitoring; HTN, hypertension; TP, true positive; FP, false positive; FN, false negative; TN, true negative; CI, confidence interval; LR+, positive likelihood ratio; LR−, negative likelihood ratio; Sens, sensitivity; Spec, specificity; κ, kappa statistic; WCH, white coat hypertension; MH, masked hypertension; HBPM, home BP monitoring; RA, research assistant

^aTotal of 251 patients includes 78 patients aged 6-11 years, 137 patients aged 12-17 years, and 36 patients aged 18-25 years.

Hypertension Thresholds on ABPM

For defining hypertension on ABPM, five studies^{27, 37, 42, 45, 49} used the 95^th percentile cutoff from the 2002 ambulatory BP thresholds developed by the German Working Group on Pediatric Hypertension,⁵⁸ with only Puthukara et al⁴⁵ using adult thresholds for patients ≥13 years as is recommended in current guidelines.⁵⁹ Adalat et al⁵¹ used the 90^th percentile cutoffs from the German Working Group norms.⁵⁸ Two studies^{34, 36} used the earlier 1997 ambulatory BP thresholds from the German Working Group,⁶⁰ which do not account for the skewed distribution of the data. Ntineri et al’s study on isolated systolic hypertension used an ABPM threshold of systolic/diastolic BP ≥95^th percentile/<90^th percentile for ages 12-15 years and ≥130/<80 for ages 16-25 years but did not specify the references used for percentiles.⁴¹ Two studies did not specify ABPM thresholds.^{38, 56} Of the 11 studies that compared daytime home BP monitoring with ABPM, four studies^{27, 34, 36, 38} used awake ABPM as the primary outcome, six studies^{37, 41, 42, 45, 49, 51} used 24-hour ABPM as the primary outcome, and one study did not specify.⁵⁶ In the two studies that reported the accuracy of home BP monitoring for both awake and 24-hour ABPM, the sensitivity and specificity of home BP monitoring was similar (Ntineri et al)³⁸ to slightly higher (Glenn et al)⁴⁹ when the reference standard was awake ABPM compared with 24-hour ABPM. For the two studies by Stambolliu et al and Menti et al that reported nocturnal hypertension status by home BP and ABPM results,^{43, 46} nocturnal hypertension was defined using the same criteria for both modalities as a mean nighttime systolic and/or diastolic BP ≥95^th percentile for sex and height⁵⁸ in patients <16 years and as a mean nighttime systolic and/or diastolic BP ≥120/70 for patients ≥16 years.

Home BP vs Ambulatory BP on the Continuous Scale

Among 1376 patients with paired daytime home and awake ambulatory BP readings, daytime systolic home BP was consistently lower than awake systolic ambulatory BP (paired mean difference, −6.4 mmHg [95% CI, −10.7 to −2.0], I² 98%) (Figure 3A). Daytime diastolic home BP was also lower than awake diastolic ambulatory BP, but the magnitude of difference was smaller (paired mean difference, −3.3 mmHg [95% CI, −6.0 to −0.7], I² 94%) (Figure 3B). Studies comparing the daytime mean home BP with the 24-hour mean ambulatory BP found a smaller difference in magnitude than when the awake ambulatory BP was used (paired mean difference for systolic BP, −1.3 mmHg [95% CI, −3.0 to 0.3], I² 94%; paired mean difference for diastolic BP, 2.0 mmHg [95% CI, −0.4 to 4.4], I² 88%) (Figure 4). In the two studies that compared nighttime home BP with asleep ambulatory BP readings, nighttime home BP was slightly higher than asleep ambulatory BP (paired mean difference for systolic BP range, 2.6 to 5.0 mmHg; paired mean difference for diastolic BP range, 2.2 to 3.2 mmHg) (Figure 5). Exploratory analyses in meta-regression did not find a statistically significant association between the following study-level covariates and the paired mean difference for daytime home systolic BP and awake ambulatory systolic BP: number of patients (β −0.04 mmHg per 1-patient increase in sample size [95% CI, −0.2 to 0.1], P=0.22), mean home BP (β 0.5 mmHg increase per 1-mmHg increase [95% CI, −2.0 to 2.9], P=0.29), mean ambulatory BP (β −0.5 mmHg per 1-mmHg increase [95% CI, −5.5 to 4.5], P=0.45), publication year (β 0.5 mmHg increase per 1-year increase [95% CI, −0.1 to 1.0], P=0.06), mean age of patients (β 0.5 mmHg increase per 1-year increase in age [95% CI, −0.3 to 1.2], P=0.09).

Figure 3. Mean Difference in (A) Systolic and (B) Diastolic Blood Pressure Between Daytime Home Blood Pressure Monitoring and Awake Ambulatory Blood Pressure Monitoring.

^aStudy presents results for 5-12 year old patients and 13-17 year old patients separately.

^bStudy presents results for 6-11 year old patients and 12-17 year old patients separately.

HBP, home blood pressure; ABP, ambulatory blood pressure

Figure 4. Mean Difference in (A) Systolic and (B) Diastolic Blood Pressure Between Daytime Home Blood Pressure Monitoring and 24-Hour Ambulatory Blood Pressure Monitoring.

^aStudy presents results for 6-11 year old patients and 12-17 year old patients separately.

HBP, home blood pressure; ABP, ambulatory blood pressure

Figure 5. Mean Difference in (A) Systolic and (B) Diastolic Blood Pressure Between Nocturnal Home Blood Pressure Monitoring and Asleep Ambulatory Blood Pressure Monitoring.

Results were not pooled in meta-analysis due to only two studies available.

HBP, home blood pressure; ABP, ambulatory blood pressure

Hypertension Status by Home BP Monitoring vs ABPM

Five studies reported the sensitivity and specificity of daytime home BP monitoring to detect ambulatory hypertension on either awake ABPM (sensitivity, 0.52-0.64; specificity, 0.82-0.97)^{27, 34, 38} or 24-hour ABPM (sensitivity, 0.31-0.71; specificity, 0.88-0.90)^{42, 45} (Table III). Glenn et al⁴⁹ reported the accuracy of attended home BP measured by a research assistant on a single day compared with 24-hour ABPM (sensitivity, 0.88 [95% CI, 0.81-0.94]; specificity, 0.77 [95% CI, 0.69-0.86]). Wühl et al²⁷ also evaluated the accuracy of combining office and home BP results and found that having hypertensive BP at three clinic visits and on home BP monitoring had the highest specificity of 0.98 while having hypertensive BP at a single clinic visit or on home BP monitoring had the highest sensitivity of 0.81. Three studies^{34, 38, 45} reported the accuracy of home BP monitoring to detect white coat hypertension (sensitivity, 0.77-0.89; specificity, 0.85-0.94; κ, 0.66-0.73) and to detect masked hypertension (sensitivity, 0.23-0.38; specificity, 0.92-0.96; κ, 0.27-0.36). Overall, diagnostic agreement for hypertension status between home BP monitoring and ABPM was 63-92% with κ 0.40-0.65.^{34, 36–38, 41, 42, 45} For the diagnosis of nocturnal hypertension, Stambolliu et al and Menti et al^{43, 46} compared 3-night home BP monitoring with asleep ABPM and found overall agreement of 82-88% and κ 0.33-0.49. Detection of non-dipping status between nocturnal home BP monitoring and asleep ABPM had lower agreement of 38-57% and κ 0.032-0.19 in these two studies.^{43, 46}

Discussion

This systematic review and meta-analysis of 26 studies found that daytime home BP readings were consistently lower than awake ambulatory BP readings in youth, with a paired mean difference of −6.4 mmHg (95% CI, −10.7 to −2.0) for systolic BP and −3.3 mmHg (95% CI, −6.0 to −0.7) for diastolic BP. Classifying daytime ambulatory hypertension status by home BP monitoring had modest sensitivity (0.52-0.64) and good specificity (0.82-0.97). Limited data on a novel method of nocturnal home BP monitoring showed slightly higher BP readings than asleep ABPM (2.6 to 5.0 mmHg for systolic BP; 2.2 to 3.2 mmHg for diastolic BP), with moderate agreement in diagnosing nocturnal hypertension (κ 0.33-0.49). Future work is needed to define clinical situations in which daytime and nighttime home BP monitoring may replace 24-hour ABPM and the extent to which clinic and home BP results may be combined to rule-in or rule-out ambulatory hypertension.

The observed difference between daytime home BP and awake ambulatory BP in youth is progressively eliminated with increasing age.³⁹ Home BP and awake ambulatory BP become similar after the age of 30 years and daytime ambulatory BP eventually becomes lower than home BP after the age of 60 years.³⁹ The higher daytime ambulatory BP readings in youth have been attributed to the increased physical activity in children during the daytime that impacts ABPM readings but not home BP readings.³⁹ Thus, pediatric reference normative data from the Arsakeion School study⁵⁷ for home BP monitoring and the German Working Group on Pediatric Hypertension⁵⁸ for ABPM show that the 50^th percentile home BP is 4 to 8 mmHg lower than that for daytime ambulatory BP, although these differences are smaller at the 95^th percentile BP.⁶¹ Adult guidelines have set the same threshold of ≥130/80 for hypertension based on office BP, home BP, and daytime ABPM.¹¹ For pediatric home BP monitoring, consensus has not been established on the best threshold to define daytime hypertension, with some studies^{34, 42} applying the sex- and height-based 95^th percentile cutoffs from the Arsakeion School study⁵⁷ while other studies^{49, 50, 54} apply the office BP thresholds from the 2017 AAP guideline.³ For nocturnal hypertension, pediatric studies have used the same thresholds for home BP monitoring and ABPM as asleep BP is expected to be similar on both modalities.^{43, 46}

Although the included studies showed disagreement in hypertension status between home BP monitoring and ABPM in ~10-20% of patients, clinically meaningful differences in home and ambulatory BP were infrequent. Stergiou et al observed that clinically important disagreement occurred in only 8% of patients for systolic BP and none for diastolic BP, defined as a patient with home and ambulatory BP results in opposite directions around a 5-mmHg gray zone above and below the diagnostic threshold.³⁴ In addition, patients with diagnostic disagreement tended to have intermediate BP levels that fell between those with home and ambulatory normotension and those with home and ambulatory hypertension.³⁴ We found that home BP monitoring may be particularly useful for detecting white coat hypertension (sensitivity, 0.77-0.89; specificity, 0.85-0.94; κ, 0.66-0.73) and for ruling in masked hypertension (sensitivity, 0.23-0.38; specificity, 0.92-0.96; κ, 0.27-0.36); however, based on the low sensitivity for masked hypertension, patients with normal clinic BP and normal home BP with risk factors for masked hypertension may still require ABPM to make a definitive diagnosis.

There are several advantages to home BP monitoring that may overcome the potential limitations in accuracy compared with ABPM. First, completion of ABPM typically requires a referral to the pediatric subspecialist. Delays in referral are common⁶² and utilization of ABPM remains low.^{10, 63} Thus, the 2023 European Society of Hypertension guidelines have recommended that ABPM be performed whenever possible to diagnose pediatric hypertension but acknowledge that home BP monitoring may be more accessible and practical in pediatric primary care.⁶⁴ Second, ABPM devices cost $2000-$2500, with additional costs incurred for batteries, device maintenance, annual calibration, and ~1 hour of staff time needed to coordinate one patient’s ABPM.^65–67 Home BP devices cost $30-$200, although this cost is typically paid by patients out-of-pocket.⁶⁶ Third, pediatric data suggest that the association between home BP monitoring and target organ damage is similar to that for ABPM and target organ damage.^{42, 43, 68} Fourth, the reproducibility of home BP monitoring and ABPM results are comparable with each other and superior to office BP.^{35, 69, 70} Fifth, children and adolescents report that ABPM is more stressful and inconvenient than home BP monitoring and ~75% prefer home BP monitoring over ABPM.⁴³

Despite these advantages, there are several barriers to implementation of pediatric home BP monitoring. Validated pediatric home BP devices with pediatric cuff sizes are not widely available.⁷¹ Feasibility and adherence to a home BP monitoring protocol can also be challenging. A multicenter US study in pediatric primary care clinics found that only 1 in 3 children with high office BP that agreed to home BP monitoring completed any home BP readings and that only 1 in 6 children completed the full 7-day home BP protocol, despite up to 10 reminder calls from the research team.⁵⁴ In addition, insurance reimbursement is low for both ABPM and home BP monitoring, and may be less than the per-patient cost for a clinic to maintain an out-of-office BP monitoring program.⁶⁷ Future work is needed to develop and test strategies for implementation of home BP monitoring with high fidelity in the pediatric primary care setting.

Our study has several limitations. First, studies used different home BP monitoring protocols and applied different thresholds to define home and ambulatory hypertension. Only one study⁴⁵ used the updated thresholds from the American Heart Association’s 2022 Scientific Statement on pediatric ABPM,⁵⁹ which recommends applying the adult cutoffs for ambulatory hypertension to adolescents ≥13 years. Second, the sequence of home BP monitoring and ABPM was not randomized and was generally determined by device availability, which may introduce an order effect such that devices may have lower readings when used second than when they are used first.⁷² In addition, the time interval between home BP monitoring and ABPM ranged from 1 day to 6 weeks and was not reported in 10 studies. Third, most home BP devices used in the included studies had been validated in adults but not in children. According to the AAMI/ESH/ISO universal standard, a BP device should be validated in ≥35 children aged 3 to 12 years old for pediatric use.⁷³ Nevertheless, the mean age of patients for most studies in this systematic review was ≥13 years, and adolescents aged ≥13 years are considered part of the general adult population in BP device validation studies.⁷⁴ Fourth, most studies were cohorts of youth evaluated at referral hypertension centers, which demonstrated greater adherence to the home BP monitoring protocol than the one study based in pediatric primary care clinics.⁵⁴ Fifth, the nocturnal home BP monitor used in the two studies in this systematic review is not currently available in the US.^{43, 46} In addition, these two studies only compared nocturnal hypertension status on each modality but did not consider daytime hypertension status.^{30, 50} It is unclear how combining daytime and nighttime home BP results can predict hypertension status on 24-hour ABPM, as ambulatory hypertension can be diagnosed based on a mean daytime, nighttime, or 24-hour BP above the cutoff.⁵⁹ Puthukara et al found that ~80% of youth with obesity and masked hypertension had isolated nocturnal hypertension that was missed by daytime home BP monitoring but that may have been detected if nocturnal home BP monitoring had been performed.⁴⁵ Sixth, only one study assessed the accuracy of combining clinic and home BP readings. Of note, a novel method of clinic BP measurement called unattended automated office BP (AOBP) measurement, in which a device takes BP automatically after a rest period without any clinic staff in the room, has been shown to mitigate the white coat effect in children^{75, 76} and is now recommended in some adult guidelines.^{77, 78} Combining unattended AOBP results with nocturnal home BP monitoring may have the potential to improve the diagnosis of hypertension without ABPM. However, the AOBP device used in these two pediatric studies (BpTRU device) is no longer being manufactured, validated pediatric AOBP devices are not widely available, and AOBP devices cost 2- to 3-times more than conventional office BP devices.^{79, 80} Additional research is needed on the accuracy and optimal approach to implementation of unattended AOBP in pediatric settings.

In conclusion, we found that the sensitivity and specificity of pediatric home BP monitoring may not be sufficient to replace ABPM as a single, definitive diagnostic test. However, home BP monitoring may be a practical alternative for out-of-office BP assessment when ABPM is not available, particularly when office and home BP readings are in agreement and are ≥5 mmHg above or below the diagnostic threshold. ABPM may still be required for patients with intermediate home BP results near the threshold for hypertension and patients with normal home BP results that have risk factors for masked hypertension

Abbreviations

HBPM

home blood pressure monitoring

ABPM

ambulatory blood pressure monitoring

BP

blood pressure

MD

mean difference

CI

confidence interval

AAP

American Academy of Pediatrics

US

United States

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

MeSH

medical subject heading

QUADAS-2

Quality Assessment of Diagnostic Accuracy Studies, version 2

AAMI/ESH/ISO

Association for the Advancement of Medical Instrumentation/European Society of Hypertension/International Organization for Standardization