Why the study?
How do repeated measurements and device type affect the accuracy of blood pressure measurement in children?
How do repeated measurements and device type affect the accuracy of blood pressure measurement in children?
Answers are grounded in this paper and cite their sources.
Accurate diagnosis of childhood hypertension requires repeated blood pressure measurements and careful consideration of the measurement device, as oscillometric devices may yield inaccurate readings compared to aneroid devices.
The diagnosis of hypertension in children and adolescents is challenging for several reasons. Unlike in adults, where a single blood pressure (BP) cutpoint is utilized to designate any individual as hypertensive, normative BP values in children are age-, sex-, and even height-dependent.1 Additionally, since the hard cardiovascular endpoints used to define BP thresholds in adults—stroke, myocardial infarction and death—occur only rarely (if at all) in the pediatric age group, the definition of hypertension in children and adolescents is based on a statistical approach that has been refined over the past several decades.2 In addition to use of the correct normative values, repeated measurements of BP are also recommended before making a diagnosis of hypertension. Although the rationale for such a complex approach is to avoid overdiagnosis of hypertension in the young,1 an unfortunate consequence has actually been underdiagnosis of hypertension in children and adolescents, which has been demonstrated in several recent studies.3, 4 Even before a BP measurement can be recognized as normal or elevated, however, care must be taken that the BP is measured accurately. The study by Eliasdottir and colleagues5 in this issue of The Journal of Clinical Hypertension highlights two important aspects of accurately measuring BP in children: the accommodation effect, and choice of measurement device. BP in children is well-known to be labile and can be affected by many factors, including measurement technique, emotional state, and activity level, among others. The lability of BP in childhood was demonstrated as early as 1987, when Ogborn and Crocker6 found that 48% of children with secondary forms of hypertension had BP readings <95th percentile more than 50% of the time. This is remarkable, to say the least, and was one of the primary reasons why repeated BP measurements have long been recommended by the Working Group of the National High Blood Pressure Education Program (NHBPEP).1 More recently, a group of investigators in Bern has demonstrated that among children being followed up in a nephrology clinic, about one third of those with BP readings >90th percentile at the beginning of the clinic encounter had repeated BPs <90th percentile at the end of the encounter.7 They suggested that for children with initially elevated office BP readings, a repeat BP should be obtained at the end of the encounter, and that the repeat reading should be used for classification purposes. This is entirely consistent with the earlier data of Ogborn and Crocker6 and supports the recommendations of the NHBPEP Working Group.1 The data presented by Eliasdottir and colleagues5 are even more remarkable than these earlier studies. They demonstrated that even in a familiar nonclinical setting (school) with repeated BP measurements obtained over a short time interval, children accommodate to having their BP taken, and that this effect manifests as significantly lower BP readings with repeated measurement. The authors postulate that the initially high readings are a stress response to the measurement procedure. It is unfortunate that heart rate data were not reported in the study, as a reduction in heart rate as seen in the study by Tschumi and colleagues7 would have suggested decreased stress. However, when considered along with the prior study on the decrease in BP measurements during a clinic encounter,7 the present data add further weight to the recommendation to obtain repeated measurements when assessing a child with an initially elevated BP reading. The other notable finding of Eliasdottir and colleagues5 pertinent to BP measurement in children is the notable difference in measurements obtained by aneroid and oscillometric devices. Although there was a statistically significant correlation between BP readings obtained by different devices, the correlation was not that strong (r values of 0.62–0.64 for systolic BP and 0.25–0.29 for diastolic BP), and the averaged aneroid and oscillometric BP readings were within 5 mm Hg of each other in only about half of patients.5 Further, the agreement between the two devices (utilizing the aneroid device readings as the reference) was low for systolic BP and nearly nonexistent for diastolic BP. In sum, BP readings obtained by the two devices were not equivalent, which is a concern given the widespread use of oscillometric devices in pediatric patients.8 There are many known issues that can lead to inaccurate BP readings by oscillometric devices.9 Some problems—such as improper cuff size—can generally be avoided if the user follows proper technique, but others—such as cuff inflation rate—cannot be modified by the user because they are built into the devices by the manufacturers. Additionally, the internal algorithms used by the devices to calculate systolic and diastolic BP from the oscillations of the brachial artery vary from manufacturer to manufacturer; thus, two different oscillometric devices may give different BP values for the same patient even if used simultaneously. Seen in this context, the findings of Eliasdottir and colleagues5 should come as no surprise. Indeed, many other investigators have demonstrated similar findings in pediatric patients. Midgley and colleagues10 conducted a similar school-based study utilizing an unknown type of sphygmomanometer (presumably aneroid) and an oscillometric device, and demonstrated poor correlation between BP values obtained by the two techniques. Their study also demonstrated several other unique problems with oscillometric devices, namely a tendency to overinflate on the first reading, and failure to successfully obtain a diastolic BP value.10 This first problem in particular may actually account for the initially higher BP readings obtained in children utilizing oscillometric devices—pain from overinflation causing physiologic stress, which, in turn, elevates the BP. Poor correlation between oscillometric and aneroid BP values has also been clearly demonstrated in children with chronic kidney disease by Flynn and colleagues.11 In that study, both systolic and diastolic readings obtained using the oscillometric device were higher than those obtained using an aneroid sphygmomanometer, which differs from the studies of Eliasdottir and Midgley, in which the systolic readings were generally higher and the diastolic readings lower than the aneroid readings.5, 10 The reasons for this difference are not clear but are likely associated with the device-related issues mentioned earlier.9 The clinical implication of all of these studies is that BP values obtained using oscillometric devices should not be used in children and adolescents for the diagnosis of hypertension. How then should a clinician proceed? A suggested approach, which is supported by the main findings of Eliasdottir and colleagues5 and illustrated in the Figure, would be to repeat elevated readings obtained with an oscillometric device as recommended by the NHBPEP Working Group,1 and to also then consider repeating BP measurements later in the office encounter, as suggested by Tschumi and colleagues.7 Accurate diagnosis of childhood hypertension is possible—if one pays attention to the details of pediatric BP measurement.
No takes yet. Share an insight, caveat, or question.
Joseph T. Flynn (2013) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: