Author Affiliations: Department of Cardiology, Boston Children's Hospital, Harvard Medical School (Dr de Ferranti), and Obesity Prevention Program, Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute and Department of Nutrition, Harvard School of Public Health (Dr Gillman), Boston, Massachusetts.
In 2011, the National Heart, Lung, and Blood Institute Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents reiterated the recommendations in the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents published in 2004.1,2 These guidelines recommend universal blood pressure screening in pediatric practice starting at age 3 years. If an initial blood pressure (BP) reading exceeds the 90th percentile for age, sex, and height, the practitioner is to repeat the measurement, preferably at least twice on separate occasions. Then he or she can categorize the patient into prehypertensive (90th to 95th percentile), stage 1 hypertension (95th percentile to 5 mm Hg above the 99thpercentile), or stage 2 hypertension (above that high cutpoint). For stage 1 and stage 2 hypertension, the guidelines recommend evaluation with renal and cardiac ultrasound, along with additional laboratory tests.
Renal ultrasound and echocardiography have dual purposes in hypertensive youth. They can both identify secondary causes of hypertension, such as coarctation of the aorta or polycystic kidney disease, and assess end-organ damage, such as left ventricular hypertrophy (LVH) or renal scarring, which can lead to starting or intensifying pharmacotherapy. The guidelines recommend against using electrocardiography (EKG) because it is a poor diagnostic tool for secondary causes or end-organ damage.1,2 They recommend BP-lowering pharmacotherapy for patients with stage 2 hypertension with secondary causes or stage 1 hypertension refractory to lifestyle intervention.
Against this background, Yoon et al3 assessed the use of renal ultrasound and EKG among Michigan adolescents aged 11 to 19 years with Medicaid coverage who had both a diagnosis of hypertension and were prescribed BP-lowering medications. About a quarter of the patients underwent an echocardiogram and one fifth underwent renal ultrasound; only 8% (77 of 951) had both. On the other hand, around 50% received an EKG. Their findings imply that ultrasonography is underused and EKG is overused.
It is instructive to examine the extent to which the study's limitations, many of which the authors recognized, could weaken their conclusions about inappropriate use of testing. It is possible that most of the patients in the study had stage 2 hypertension, which requires pharmacotherapy whether end-organ damage appears on renal ultrasound or cardiac echocardiogram. As just noted, renal and cardiac ultrasound are also useful for identifying secondary causes, which most believe are more likely in stage 2 rather than stage 1 hypertension,2,4 so one might have expected high use in this group already taking medication. However, one exclusion criterion for the study by Yoon et al was an already identified cause of secondary hypertension.3 Thus, if the study sample was highly enriched with adolescents with stage 2 hypertension and no secondary cause, health care providers could have consciously forgone ultrasonography with good reason. Another possibility is that health care providers could have decided not to order renal ultrasonography for those patients with stage 1 hypertension who were also obese or had a family history of hypertension. In this case, health care providers could have inferred that these risk factors sufficiently reduced the probability of secondary hypertension and assumed that renal ultrasound will rarely uncover kidney damage due to elevated BP, which is usually at a relatively early stage in adolescence.5 However, one might have expected greater use of cardiac echocardiogram to detect LVH, whereas observed use was low for that test as well. Without actual BP levels, height and weight measurements, and family history information, which are unavailable in claims data, one cannot assess the degree to which these scenarios led to exaggerating the extent of underuse of ultrasonography.
Is it also possible that some of the ultrasounds were not indicated (ie, there was overuse as well as underuse)? Overuse of testing can lead to anxiety, peripheral asymptomatic findings, and unnecessary expenditure. Without actual BP levels, the authors could not assess the number of patients for whom practitioners coded the diagnosis of hypertension when BP levels did not actually reach threshold for testing (for example, after a single elevated BP reading). While we find it implausible that many physicians would prescribe without testing in this situation, the study design of Yoon et al does not allow examination of ultrasound use in all adolescents with elevated BP levels because they included only the small proportion that were prescribed antihypertensive medication.
Although it is critical to think through these methodologic issues, together they probably account for only a modest proportion of underuse of ultrasonography implied by the study's findings. Why then would adolescents with medication-treated hypertension get less testing than is recommended? Perhaps health care providers do not agree with the published guidelines. Indeed, the evidence base for the guidelines contains a large dose of expert opinion rather than evidence-based conclusions.6 A recently published cost-effectiveness analysis of BP screening in adolescents suggested that population approaches were more effective and less costly than all screening approaches.7 This finding was based in part on the lack of evidence of treatment effectiveness. No long-term randomized controlled trials exist to address how well reducing BP in adolescents translates into reduced risk for adult cardiovascular endpoints, although some smaller trials suggest that antihypertensive pharmacotherapy reduces LVH,8 which itself is related to contemporaneous measures of atherosclerosis and in adults, predicts subsequent cardiac and cerebrovascular events.9- 12 More broadly, the perceived lack of efficacy of the treatment for most hypertensive children (ie, lifestyle modification) may limit diagnostic efforts. A large proportion of adolescents identified with prehypertension or stage 1 hypertension will never be eligible for medications, but the recommended lifestyle interventions are difficult to achieve, leaving clinicians frustrated. Without effective treatment, making a diagnosis can be fruitless.
Regarding the testing itself, health care providers may not realize that the same test can evaluate both secondary causes and end-organ damage. The guidelines' algorithms for screening, diagnosis, and treatment are complex, and subtleties like this can get lost. Alternatively, perhaps health care providers believe LVH is too uncommon to warrant ordering an echocardiogram on all adolescents with hypertension. Some small studies of subspecialist referral populations have shown relatively high rates (30%-40%) of LVH in hypertensive youth.9,13 In the cost-effectiveness analysis previously cited, when we limited the analysis to screening approaches only (ie, no population approaches) and assumed a 30% prevalence of LVH, identifying adolescents with LVH appeared useful.7 However, the prevalence of LVH may be much lower in the general population of adolescents with hypertension. It is also possible that barriers to implementation of guidelines exist within patients or families. Health care providers may make recommendations, but adolescents and their families may not be interested in getting tested. In addition, the patient population in the study by Yoon et al was composed of Medicaid recipients, for whom access to ultrasonographic testing may be more limited than among those with commercial insurance (and perhaps less limited than among the underinsured and uninsured). Socioeconomic and racial/ethnic disparities in access to cardiovascular testing and procedures are widespread among adult populations.14,15
The answers to the many questions raised here are unclear, pointing to the need for more information about the extent to which the BP guidelines overall, and recommended diagnostic testing in particular, are implemented. It is important to get these answers. Hypertension as defined in the guidelines is under-recognized in the United States,16 and BP levels in most groups of children and adolescents have stopped decreasing during the past decade after some progress from the 1960s to 1980s.17 Future revision of the guidelines will require not only updating the evidence base for what should be done in ideal circumstances, but also what can be done in the real world given the range of possible health care provider, patient, and payor facilitators and barriers to implementation.
Correspondence: Sarah D. de Ferranti, MD, MPH, Department of Cardiology, FA607, Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115 (email@example.com).
Published Online: July 23, 2012. doi:10.1001/archpediatrics.2012.1503
Financial Disclosure: Dr de Ferranti has received royalties from UpToDate for chapters on hyperlipidemia and risk for early atherosclerosis in childhood. She serves as a member of the American Academy of Pediatrics Committee on Nutrition and is liaison to the American Heart Association Atherosclerosis, Hypertension, and Obesity in Youth Committee. She was also part of an advisory panel and writing group on familial hyperlipidemias for the National Lipid Association. Dr Gillman has given invited talks in meetings sponsored by the International Life Sciences Institute, Nestle Nutrition Institute, and Danone. He has received royalties from UpToDate for a chapter on dietary fat and provided external reviews for the US Preventive Services Task Force. He was also a member of the Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents convened by the National Heart, Lung, and Blood Institute.
Funding/Support: Dr de Ferranti's work is supported by grant K23 HL 085308-03 from the National Institutes of Health and funding from the Boston Children's Heart Foundation. Dr Gillman's work is funded in part by grant K24 HL 068041 from the National Heart, Lung, and Blood Institute.
de Ferranti SD, Gillman MW. Hypertension in Children and Adolescents—More Questions Than Answers. Arch Pediatr Adolesc Med. 2012;166(9):872-874. doi:10.1001/archpediatrics.2012.1503