Ambulatory Blood Pressure Monitoring in Children and Adolescents: Recommendations for Standard Assessment
2008; Lippincott Williams & Wilkins; Volume: 52; Issue: 3 Linguagem: Inglês
10.1161/hypertensionaha.108.190329
ISSN1524-4563
AutoresElaine M. Urbina, Bruce S. Alpert, Joseph T. Flynn, Laura L. Hayman, Gregory A. Harshfield, Mark Z. Jacobson, Larry T. Mahoney, Brian W. McCrindle, Michele Mietus‐Snyder, Julia Steinberger, Stephen R. Daniels,
Tópico(s)Heart Rate Variability and Autonomic Control
ResumoHomeHypertensionVol. 52, No. 3Ambulatory Blood Pressure Monitoring in Children and Adolescents: Recommendations for Standard Assessment Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBAmbulatory Blood Pressure Monitoring in Children and Adolescents: Recommendations for Standard AssessmentA Scientific Statement From the American Heart Association Atherosclerosis, Hypertension, and Obesity in Youth Committee of the Council on Cardiovascular Disease in the Young and the Council for High Blood Pressure Research Elaine Urbina, Bruce Alpert, Joseph Flynn, Laura Hayman, Gregory A. Harshfield, Marc Jacobson, Larry Mahoney, Brian McCrindle, Michele Mietus-Snyder, Julia Steinberger and Stephen Daniels Elaine UrbinaElaine Urbina , Bruce AlpertBruce Alpert , Joseph FlynnJoseph Flynn , Laura HaymanLaura Hayman , Gregory A. HarshfieldGregory A. Harshfield , Marc JacobsonMarc Jacobson , Larry MahoneyLarry Mahoney , Brian McCrindleBrian McCrindle , Michele Mietus-SnyderMichele Mietus-Snyder , Julia SteinbergerJulia Steinberger and Stephen DanielsStephen Daniels Originally published4 Aug 2008https://doi.org/10.1161/HYPERTENSIONAHA.108.190329Hypertension. 2008;52:433–451Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: August 4, 2008: Previous Version 1 Epidemiology of HypertensionThroughout the world, 1 in every 4 adults suffers from hypertension,1 a disease that contributes to 49% of ischemic heart disease and 62% of strokes worldwide. Inadequately controlled hypertension is currently the number one attributable risk for death across the globe.2 Data from the Framingham Heart Study predict that 90% of people who are normotensive at age 55 years will go on to develop hypertension in their lifetime.3 Hypertension in youth is also being diagnosed with increasing frequency.4 The global obesity epidemic is leading to a shift in the blood pressure (BP) distribution toward increasing levels in children and adolescents.5 This is particularly relevant because BP levels in the higher end of the distribution track into adulthood,6 resulting in prehypertension, which marks individuals at high risk for progressing to sustained hypertension.7Autopsy studies such as the Bogalusa Heart Study and the Pathobiologic Determinates of Atherosclerosis in Youth (PDAY) Study have demonstrated increased atherosclerosis at higher BP levels in youth.8,9 Therefore, accurate assessment and management of BP is essential for the prevention of target organ damage.10 Ambulatory BP monitoring (ABPM), which can more precisely characterize changes in BP throughout daily activities,6 has been found to be superior to clinic BP (CBP) monitoring in predicting cardiovascular morbidity and mortality.11 For this reason, ABPM is seeing more widespread use in evaluation for hypertension and risk of end-organ damage in adults.In children and adolescents, ABPM is gaining acceptance as a useful modality for the evaluation of BP levels in both hypertension research and in the clinic setting.12,13 This statement summarizes the current research and clinical applications of ABPM in children and adolescents and offers recommendations on implementation of ABPM in practice and interpretation of results. Because no outcome studies are yet available relating ABPM levels in children to hard outcomes such as myocardial infarction or stroke, these guidelines are expert opinion–driven and not evidence based.ABPM and Risk for Target Organ DamageIn adults, ambulatory, rather than CBP, is correlated more strongly with left ventricular mass (LVM)14 in both hypertensive and normotensive individuals.15 Similar results have been published for children, with the relationship greatest between LVM and nighttime systolic BP (SBP)14 and BP load.16 A recent pediatric study using ABPM to confirm hypertension demonstrated a relationship between severity of BP elevation and odds for LVH.17Similarly, increased carotid intima-media thickness (c-IMT), a risk factor for stroke,18 is associated with ambulatory BP,19 and the relationship between ABPM and c-IMT remains significant even after adjusting for CBP, suggesting that ABPM provides an independent contribution to risk stratification.20 To the best of our knowledge, no studies in healthy children have examined the relationship between ambulatory BP levels and c-IMT, but hypertensive children do demonstrate a relationship between higher ABPM levels and thicker carotid arteries.21,22 Furthermore, Sorof et al found that children with more significant abnormality in their ABPM pattern (increase in BP levels and the percentage of readings greater than the 95th percentile, or the BP "load") were more likely to have LVH. This may relate to an increased afterload induced by vascular abnormalities resulting in cardiac hypertrophy in hypertensive youth.23 ABPM is also superior in identifying adults with increased arterial stiffness, whether measured in the carotid artery (ultrasound)24 or aorta (pulse wave velocity)25,26 and with decreased endothelial function (brachial flow–mediated dilation).27 Few data are available in children, although one study of pediatric kidney transplant recipients found deterioration in carotid distensibility associated with higher daytime ambulatory SBP load.28Ambulatory BP in adults is also more strongly correlated with renal damage (renal albumin excretion) than is CBP.29 Albumin to creatinine ratio also relates most strongly to diastolic BP (DBP) variability, which can only be measured with ABPM.30 Data relating ABPM to kidney damage in healthy children are less clear. One study found no relationship between ambulatory BP and either creatinine clearance or albumin excretion in hypertensive youth.14 Another investigation found that nighttime ambulatory SBP did relate to creatinine clearance, but only in African American subjects.31ABPM Is Superior to Self-Measurement of BPSelf-measurement of BP (SMBP) can be performed anywhere, not just at home, and has been suggested as an acceptable alternative in place of ABPM in adults.32 To investigate whether SMBP values provide a feasible and reliable alternative to ABPM in differentiating true from white coat hypertension (WCH; see definition below) and in monitoring antihypertensive therapy in children, a recent study in 118 pediatric patients (age 3 to 19 years) with chronic renal failure compared ABPM, SMBP, and CBP measurements.33 The data showed that SMBP was a valuable addition to CBP measurement, as it agreed with ABPM more closely and more consistently over the whole range of BP as compared with CBP alone. The addition of SMBP to CBP also offered a higher degree of diagnostic specificity than CBP alone. However, the diagnostic sensitivity reached by SMBP and CBP was only 81% as compared with ABPM as the reference method. Therefore, 1 of 5 children diagnosed as hypertensive by ABPM would have been missed, even when both CBP and SMBP were used in combination. In addition, the range of agreement of SMBP with ABPM, albeit narrower than that of CBP, was unacceptably wide.33 Consequently, these data do not support the replacement of ABPM by SMBP.Use of ABPM in Evaluation of Secondary HypertensionSecondary hypertension is more common in children than in adults. Hypertension detected in very young children, or in children or adolescents with clinical signs that suggest systemic conditions and the diagnosis of stage 2 hypertension, are all suggestive of secondary hypertension. A number of findings on the history and physical examination may be indicative of the etiology of secondary hypertension.34 Ambulatory BP readings may be useful in differentiating primary from secondary hypertension, as adolescents with secondary hypertension have been shown to manifest greater nocturnal SBP loads and greater daytime and nocturnal DBP loads than children with primary hypertension.35 These patterns were highly specific for differentiating between essential (primary) and secondary types of hypertension.35 Although confirmatory studies in this area are needed, the potential use of ABPM in differentiating between primary and secondary hypertension was also suggested in a study from the Czech Republic, which demonstrated decreased nocturnal dipping in children with secondary hypertension.36White Coat HypertensionWCH is another clinical condition in which ABPM data are critical. WCH is defined as BP levels that are the 95th percentile or higher when measured in the physician's office or clinic but are completely normal (average BP 10% above the 95th percentile) were infrequently found to have WCH, as they were more likely to be true hypertensives.42Continued follow-up for patients exhibiting the WCH pattern may be necessary. Although adult studies find that patients with WCH have lower LVM than those with sustained hypertension, their cardiac mass is higher than that of normal controls.43 Furthermore, other forms of target organ damage, such as endothelial dysfunction44 and increased c-IMT,45 are associated with WCH and may account for the increase in adverse cardiovascular disease outcomes noted with this condition.46 Data in children are sparse, but youth with WCH have been shown to have greater body mass index and a tendency toward elevated LVM index, strengthening the indications for ABPM follow-up of WCH.47,48Masked HypertensionAnother condition that may be uncovered with ABPM is masked hypertension, defined as normal CBP but elevated ambulatory levels. The prevalence of this condition is not known, with estimates ranging from 5.7% in an unselected group of 592 children49 to a high of 9.4% in a study of 85 consecutive patients referred for suspected hypertension.47 Masked hypertension may be suspected when multiple primary care providers report hypertension, yet resting BP levels are less than the 95th percentile in the hypertension clinic or the clinical presentation (ie, LVH) seems inconsistent with CBP. In adults, masked hypertension has been associated with an increased cardiovascular (CV) risk50 and with progression of chronic kidney disease.51 In children, it is associated with progression to sustained clinic hypertension49 and higher LVM.47 Although carefully conducted home BP monitoring could possibly be used to identify masked hypertension, ABPM is a superior technique and is considered the gold standard for evaluation of both WCH and masked hypertension.PrehypertensionABPM may be particularly useful in children with office BP within 20% of the 95th percentile.12 In these patients, ABPM can be very helpful in stratifying risk for target organ damage, because even with normal average ABPM values, increased BP variability is associated with target organ damage in adults.52 This may be especially relevant if there is a strong family history of hypertension, because BP variance is under substantial genetic control. Twin and adoptive studies suggest that as much as 50% to 79% of BP variation is due to heredity,53,54 although early perinatal events may also play a role.55 In fact, one investigation found a relationship between impaired fetal growth and higher ambulatory SBP at 12 years of age, although the major independent determinate of ABPM was current body size.56ABPM and Multiple CV Risk FactorsABPM also offers a sensitive window to identify the burden of CV risk in youth with obesity and the metabolic syndrome. The specific link between central fat distribution in obese youth and elevated ABPM has been well described,57,58 and total adiposity and insulin resistance have been correlated with a high prevalence of the nondipping phenomenon (inadequate decrease in BP at night) in youth.59 Obstructive sleep apnea, which is found more often in obese children with insulin resistance, has also been associated with greater mean BP variability while awake and less nocturnal dipping, conditions that can be diagnosed only with ABPM.60Children with prehypertension and adverse lifestyle habits may also benefit from evaluation with ABPM. Higher salt intake is associated with nondipper status in adolescents61; adult studies clearly demonstrate higher ambulatory BP levels in less active patients even after adjusting for age, body mass index, alcohol intake, and smoking.62 Psychosocial stress may also adversely affect ABPM levels in children.63 Similarly, use of stimulant medications, which also increase CV reactivity, result in higher heart rate (HR) and ambulatory BP values in children.64 In a double-blind, randomized, cross-over trial, a significantly higher HR×BP product or rate pressure product was found in children receiving active treatment for attention deficit hyperactivity disorder.64 Elevated rate pressure product is an index of myocardial oxygen demand and is believed to be a proxy for silent myocardial ischemia in adults,65 suggesting that stimulant medications may significantly increase metabolic demands on the CV systems of children being treated for attention deficit hyperactivity disorder. Caffeine is another widely consumed vasoactive drug, with more than 75% of US adults and adolescents consuming caffeine at least daily. For adolescents, the primary source of caffeine is soft drinks.66 Caffeine consumption increases the BP of adolescents (measured by ABPM) with the greatest effect during the daytime, when sympathetic nervous system responses dominate BP control.67 Clinicians should also probe patients for use of other substances that may affect BP, such as tobacco68 and recreational drugs.Methods for Collecting and Interpreting ABPM DataEquipmentThe most recent recommendations for BP measurement in adults published by the American Heart Association Council for High Blood Pressure Research include the use of ABPM and summarize findings published in previous national and international guidelines.69 Although many of the recommendations for adults are applicable in children, substantial differences exist. First, careful selection of equipment for use in pediatric patients is essential for accurate recording. The ideal pediatric monitor is light, with the weight of available monitors ranging from 168 to 457 g. Monitors should be able to tolerate some subject movement without giving excessive error readings. Several of the devices are sold with cuffs designated for pediatric use. One device offers neonatal cuffs, but there are no validation studies available for their use. As with the measurement of BP at rest, cuff size is a critical variable in the accuracy of BP data. The width of the cuff used should be at least 40% of the mid-arm circumference.4There are 2 different BP detection techniques in use, oscillometry70 and auscultatory detection with microphone detection of Korotkoff sounds.71 Of the 23 validated monitors currently available in the United States, 3 offer auscultatory detection in addition to oscillometry. One monitor offers ECG gating of the Korotkoff sounds to improve accuracy. There is still controversy relating to the Korotkoff sound (K4 or K5) that more accurately estimates DBP in children younger than 13 years of age,72,73 so potential buyers/users should consult the manufacturer's specifications to determine which was used for validation. Furthermore, although published normal values for CBP4 were obtained with an auscultatory technique, normative cut points for auscultatory ABPM data in children are lacking. The largest cross-sectional study of ABPM in pediatrics to date used an oscillometric technique.70 However, oscillometric devices are subject to the same potential errors as oscillometric devices used for casual BP measurement and, accordingly, have received lower ratings than auscultatory devices when evaluated according to the British Hypertension Society (BHS) protocol.74 However, oscillometric devices usually have a lower percentage of erroneous readings than auscultatory devices and are easier to use than auscultatory devices. For these reasons, most centers performing ABPM in children and adolescents use oscillometric monitors.The software offered with ambulatory BP monitors varies. At a minimum, monitors should be programmable to record every 15 to 20 minutes throughout the 24 hours. A report that can be customized to include pediatric reference data is ideal. Many laboratories have adapted their software to enter the 95th percentile ABPM cutoffs specified by Soergel et al70 so that variables such as BP load can be calculated automatically for each child. Alternatively, cut points from the LMS-transformed data (based on age- and gender-specific estimates of the distribution median [M], coefficient of variation [S], and degree of skewness [L]) of Wühl et al can be used (see Appendix).75Although dozens of monitors are available for purchase in the United States, few have been validated in children. A Web site (www.dableducational.org) has been created to try to provide a list of monitors that have undergone independent testing and have been shown to perform well enough to pass a national standard, such as the Association for the Advancement of Medical Instrumentation (AAMI) US National Standard76 or the BHS Standard.77 These standards, as well as others from Europe and Japan, are written to ensure that the monitors are accurate and durable. Currently, the International Standards Organization is developing a worldwide standard for automated BP monitors. Unfortunately, monitors that have not undergone validation testing and US Food and Drug Administration clearance can be sold in the United States.Depending on the age of the subject, there are 2 reference standards against which device recordings may be compared, auscultation and intra-arterial catheter measurements. There is no consensus regarding the age at which Korotkoff sounds are audible and/or accurate. The AAMI Standard requires intra-arterial comparison data for children younger than 3 years of age. Either intra-arterial catheter measurement or auscultation may be used for subjects 3 years of age and older. Studies are under way to determine whether Korotkoff sounds could be reliably used in children younger than 3 years of age.Ages Studied With ABPMAlthough patients as young as 2 months have been studied using 24-hour ABPM, routine use is usually limited to children 5 to 6 years of age or older. Varda et al studied 97 healthy infants and toddlers aged 2 to 30 months using an oscillometric device and found usable recordings in 87% of subjects. One limitation noted was that the smallest available cuff was too large for some infants.78 Gellermann et al obtained useable recordings in 77% of 101 children 3 to 6 years of age with and without renal disease and/or hypertension, with the ability to obtain useable recordings improving with age.79 One half of the children diagnosed with high BP in the clinic setting were actually found to have normal ABPM,80 emphasizing the use of ABPM in the diagnosis of hypertension. Children as young as 5 years of age were successfully included in a large school-based study in Germany that is widely quoted as a reference for normal oscillometric ABPM levels in children.70 In England, O'Sullivan et al conducted a similar school study in 1121 children aged 6 to 16 years using a device that gave both oscillometric and auscultatory readings. Only 3 studies had to be excluded because fewer than 41 successful readings were obtained.71 Finally, in a community sample of 300 healthy 10- to 18-year-olds using a similar auscultatory and oscillometric device, Harshfield et al reported that 84% of subjects had useable data.81Frequency of ABPM MeasurementIn published studies of ABPM, recording frequency varies from every 15 to 30 minutes for daytime or waking measures and from every 20 to 60 minutes for sleep or nighttime measures. Regardless of the frequency selected, most authorities on pediatric ABPM require at least 1 valid reading per hour, including during sleep, as a primary criterion for an interpretable study.Accounting for ActivityAn important concern in interpreting ABPM data in pediatric patients is how to divide the recording into sleep and wake times and how to account for variations in levels of physical activity. Daytime or awake time has been defined by different authors as beginning at 6 am to 9 am and ending at 9 pm to midnight. Sleep or nighttime has been defined as beginning at 9 pm until midnight and ending at anywhere from 6 am to 9 am. With this approach, readings obtained during transition times (ie, 6 am to 8 am and 10 pm to midnight) are discarded in the analysis.70 Alternatively, self-reported sleep-wake times recorded in a diary have been used to divide an ABPM study into awake and asleep periods.35,82 Finally, limited data suggest that actual sleep and wake times determined from an actigraph (a wrist device that senses motion in 3 dimensions) may be superior even to patient-initiated diary entry.83Subject activity clearly influences both the success of ABPM studies and the BP readings themselves. Portman et al assessed ABPM in 99 healthy 5th graders, each of whom simultaneously recorded their activity and emotional state in a log.84 The analysis showed that reliable and reproducible ABPM was feasible and that both ambulatory SBP and DBP varied by 10 mm Hg from lowest to highest level of activity. Jacoby et al studied 22 healthy children 4 to 17 years of age using an oscillometric ABPM unit for 24 hours of normal activity and during treadmill testing and stair-climbing activities.85 Although all measures could be obtained at rest (19/19), only 68% of the values obtained during the treadmill testing (13/19) were valid at 300 kilopond meters and 36% (5/14) at 600 kilopond meters. Therefore, most hypertension specialists recommend that children undergoing ABPM should continue their normal activities but refrain from contact sports and vigorous exercise. Many recommend that patients hold their arm still during measurements.Editing ABPM DataExtreme outlier BP readings during ABPM are unlikely to be valid and are most likely artifact. However, it is sometimes difficult to decide reliably which BP values to discard, making editing a labor-intensive process that is prone to error and possibly also to observer bias. Given this, various automated approaches have been developed in an attempt to prevent these problems.86 Winnicki et al investigated a number of automated editing methods using oscillometric ABPM in a cohort of 584 older adolescents and adults with mild hypertension.87 Of the 6 methods studied, a modification of the Casadei method (see below) was found to have the most favorable variability, reproducibility, and validity and was, therefore, considered the method of choice.87 Briefly, the method calls for a visual inspection for grossly inconsistent ABPM readings before interpretation. In this method, only measurements with SBP 70 mm Hg, DBP 40 mm Hg, HR 40 but <100 mm Hg with a DBP<SBP are accepted as valid. These settings can be programmed into the analysis software of most ABPM devices, thereby avoiding manual editing, which is not recommended. However, these settings may not be appropriate for younger children whose normal resting values for HR and BP may differ greatly from adults (see Adult ABPM Normals in Table 1). Table 1. American Heart Association Recommendation for the Upper Limit of Normal Ambulatory Blood Pressure in AdultsOptimalNormalAbnormalReprinted from Pickering et al,69 with permission from Lippincott Williams & Wilkins. Copyright 2005, American Heart Association.Daytime<130/80 140/90Nighttime<115/65 125/7524-Hour<125/75 135/85Calculations Used in InterpretationInterpretation of ABPM studies is usually based on a combination of criteria, including mean BP and BP loads. Mean SBP and DBP are calculated by the analysis software, which allows the user to define wake and sleep times, for calculation of average values for the entire 24-hour period, daytime and nighttime.88 Mean BP levels are then compared with normative values to determine whether a subject's BP is normal or elevated. Either the seated resting BP values published in the Fourth Report on BP in Children4 or ambulatory BP values such as those reported by the Heidelberg group (see Appendix)70,75 theoretically can be used for this analysis. It is important to recognize that ambulatory BP measured with an oscillometric device tends to be higher than resting BP obtained with auscultation. In fact, Sorof et al found that in a population of 71 children with elevated office BP, hypertension was diagnosed in only 41% of patients using the higher ambulatory criteria, whereas the Fourth Report cut points would have led to 69% being diagnosed with hypertension (P<0.001).74 Thus, although each of these criteria is useful and has its adherents, outcome studies will be necessary to resolve which is best in assessing risk or effect of treatment.42BP load is defined as the percentage of valid ambulatory BP measures above a set threshold value, such as the 95th percentile of BP for age, gender, and height.89 As for mean ABPM values, this can be assessed for the entire 24-hour period or for the awake and asleep periods separately. Loads in excess of 25% to 30% are typically considered elevated.90 Loads in excess of 50% were demonstrated to be predictive of LVH in one pediatric study.16 Most experts in pediatric ABPM use a combination of mean BP and BP load to categorize ABPM results as normal or abnormal. Usually this involves an elevated mean BP plus an elevated BP load. However, some patients with normal mean BP levels may have elevated BP loads. These patients may be truly hypertensive and at risk for target organ damage even if they do not fit into proposed criteria for analyzing ABPM studies (Table 2).74 Furthermore, it is important to note that although no ABPM classification has ever been validated in outcome studies, criteria similar to the scheme presented in Table 2 are receiving increasing recognition by experts in pediatric hypertension. Table 2. Suggested Schema for Staging of Ambulatory BP Levels in ChildrenClassificationClinic BP*Mean Ambulatory SBP†SBP Load, %70,75Modified from Lurbe et al,74 with permission.BP indicates blood pressure; SBP, systolic blood pressure.*Based on the National High Blood Pressure Education Program Task Force Standards.†Based on ABPM values of Soergel et al or the smoothed values of Wühl.Normal BP<95th percentile<95th percentile 95th percentile<95th percentile<25Masked hypertension 95th percentile>25Prehypertension>95th percentile 95th percentile>95th percentile25–50Severe ambulatory hypertension (at risk for end-organ damage)>95th percentile>95th percentile>50Nocturnal DippingAbnormalities of circadian variation of BP and of BP variability have both been examined for their prognostic significance. Dipping refers to the physiological decline in SBP and DBP seen at night. Normal dipping is generally defined as a ≥10% decline in mean systolic and diastolic ambulatory BP levels from day to night ([mean daytime ABPM−mean nighttime ABPM]/mean day ABPM×100).61 Blunted nocturnal dipping has been associated with nephropathy in patients with types 191 and 2 diabetes mellitus82 and may be an early marker for renal deterioration. Racial differences also have been demonstrated in nocturnal dipping, with a difference in the relationship between body size and BP contributing to the elevated nighttime pressures seen in African American as compared with white youth.92BP VariabilityAnother area where ABPM is useful is in the evaluation of BP variability. The activity of both short-term and long-term BP regulatory systems are needed to meet the changing physical and psychological demands of a normal day. ABPM provides an index of the regulation of these systems.93,94 BP variability, which is most easily assessed by calculating the standard deviation of BP during a defined time period, may also have prognostic value. Increased BP variability has been demonstrated in obese children and is most likely related to increased sympathetic nervous system activation in obesity-related hypertension.95 In adults, greater BP variability has been correlated with the development of hypertensive LVH.52 Similar data are not available in children. Therefore, evaluat
Referência(s)