Screening Children for High Blood Pressure: Where the US Preventive Services Task Force Went Wrong

Thompson and colleagues1 recently published their conclusions based upon an evidence review prepared for the US Preventive Services Task Force (USPSTF) regarding blood pressure (BP) screening in children. Following established criteria from the USPTF, the report evaluates evidence supporting the cardiovascular (CV) impact of screening for hypertension (HTN) in children and adolescents. Although the report concludes that “there is no direct evidence that screening for hypertension in children and adolescents reduces CV outcomes in adults,” there are several flawed assumptions in their analysis that make their conclusions worrisome. The USPSTF methods are appropriately evidence-based and consist of a systematic search of pertinent literature and careful analysis of the available studies, but their primary question is essentially unanswerable. The Task Force attempted to find direct evidence that screening for essential HTN in children leads to decreased CV morbidity in adults. While direct evidence of correlation to “adverse health outcomes” can often be found for many screenable adult diseases where morbidity and mortality are relatively common, pediatric hypertension is likely a condition where direct evidence linking childhood BP to adult outcomes will never be available. Searching for direct evidence is therefore not an appropriate method of ascertaining future risk or benefit since children rarely develop severe CV outcomes. Any study attempting to directly answer the USPSTF question would necessarily have to follow both screened and unscreened children for many decades to determine the direct impact of early HTN detection. The report is also premature given the early state of evidence in pediatric hypertension. It was only 36 years ago that the first consensus report on BP in children was published.2 Pediatric norms for BP in their current state, including age, sex, and importantly height percentile, did not become stable until the Working Group update in 1996.3 The oldest someone could be who grew up with these guidelines in place is only 50 years old, much too young to accurately assess any meaningful impact of childhood BP on CV disease. Multiple longitudinal studies have shown, however, that elevated BP in childhood is predictive of elevated BP in adulthood after controlling for other demographic and metabolic factors.4-7 Although mentioned, this type of indirect evidence is not given enough emphasis in Thompson's analysis. While it might be both chronologically too soon and simply not feasible to discern a direct link between screening and CV mortality, an equally important question is whether there exists any difference in the pathophysiology of HTN in children compared with in adults. If the vascular biology of HTN in children is similar to that which leads to CV events in adults, then logic would indicate that children with HTN may be an especially at-risk group to develop hypertensive complications because of their long-term exposure to vascular injury. In fact, children with HTN are increasingly being found to have cardiac and vascular end-organ damage, including impaired vascular reactivity and left ventricular hypertrophy.8 Several reports from the Bogalusa Heart Study confirm that childhood BP abnormalities are associated with long-term sequelae.9-11 Similar data from the Young Finns study provide further indirect evidence that adolescent hypertension is associated with adverse vascular consequences in adults such as endothelial dysfunction and abnormal pulse wave velocity.12-14 These findings strongly implicate that HTN in childhood is not merely a risk factor for future CV damage, but a marker for current CV strain and damage. The approach of Thompson and colleages belies the underlying issue that early-onset HTN likely represents a more severe phenotype of CV disease. In addition to end-organ effects on the vascular system, data also implicate pediatric hypertension in disturbances of executive brain functioning.15, 16 Lande and colleagues have demonstrated both diminished scores on cognitive testing and increased odds of learning disability in children with HTN. Additional ongoing studies will further elucidate the link between pediatric HTN and other effects on the central nervous system. Active screening for an asymptomatic disease should be performed when early detection of the condition can result in interventions that can slow onset of overt disease, reduce cost of therapy, or improve overall outcome. Given the long silent lead time and significant subsequent complications of hypertension, BP screening in adults has long been considered a “Grade A” recommendation.17 The 2007 reaffirmation statement concludes that “substantial indirect evidence supports the effectiveness of screening adults to detect hypertension and treating them to reduce CV disease.” Fortunately, a growing body of data indicates that, as in adults, treatment of childhood HTN can lead to immediate improvements in intermediate CV outcomes, including regression of left ventricular hypertrophy.18, 19 While there might be a lack of direct evidence linking screening for hypertension in children and eventual CV mortality, immediate complications of pediatric HTN can be identified and effectively ameliorated. The cost of measuring BP in children during routine medical evaluations is not great. With the advent of automated oscillometric devices with appropriately sized cuffs, less training and time is required to collect a measurement. From a noneconomic standpoint, there is no evidence for lasting trauma or sequelae from measuring BP in pediatric patients, nor is there evidence of excess adverse events in the treatment of hypertension. Thus, the risk of screening is nil compared with the risk of not identifying children who often have measurable evidence of early CV disease, a major cause of mortality in adults. A potential response to Thompson's conclusions might be for primary care providers to forgo BP assessment in children. This would be unfortunate and unadvisable. Measurement of BP during medical encounters not only screens for primary (essential) HTN but also for secondary HTN, which is more common in children than adults and often requires specific treatment. For years the NHBPEP has recommended BP assessment as part of routine care of children.20 BP problems in childhood are becoming more common,21 and the epidemic of obesity will likely lead to an even greater risk of essential hypertension in adolescents.22 The Thompson analysis focuses on predicting adult outcomes for a pediatric disease, an unanswerable question. In doing so, they undervalue the many potential benefits of early identification of HTN in children. Routine screening for BP abnormalities in children is low risk (maybe no risk?) and high reward.