A wealth of evidence has substantiated that adults have much to gain from managing cardiovascular disease (CVD) risk factors such as hypertension and hypercholesterolemia. During the past 2 decades, studies also have made clear that atherosclerosis and other cardiovascular pathologies have life-long trajectories, and that reducing risks at an early age can reap rewards in later years. Given this background, an expert panel recently issued comprehensive guidelines on cardiovascular health and risk reduction in children and adolescents. Sponsored by the National Heart, Lung, and Blood Institute (NHLBI) and endorsed by the American Academy of Pediatrics (AAP), the recommendations will spell practice changes for pediatricians and the clinical labs that support them, albeit not without some controversy.
"One way of looking at the job of those who focus on pediatric healthcare is that we should try to deliver as many adolescents as possible to adulthood with optimum risk to give them the best opportunity to reduce their lifetime risk. We hope these guidelines will aid in that effort," said the panel's chair, Stephen Daniels, MD, PhD. "Our aim was not only to look comprehensively across risk factors and behaviors, but also developmentally so that primary care physicians would have a clear concept of what to do at each age, each stage of development from birth through young adulthood. The guideline was based on a comprehensive evidence review, and this is the first time these kinds of guidelines have been the subject of that kind of review." Daniels is professor and chair of pediatrics at the University of Colorado School of Medicine and pediatrician-in-chief at Children's Hospital Colorado in Aurora.
A Comprehensive View
The guidelines look at all major CVD risk factors, from diet and hypertension to smoking and diabetes, examine the latest evidence involving children, and offer age-specific recommendations for assessment and intervention (See below). The document, which was published in the December 2011 issue of Pediatrics, incorporates research that came to light after a series of less comprehensive guidelines were issued, including two from AAP: a 2008 publication on lipid screening and cardiovascular health in childhood, and one from 2004 involving the diagnosis, evaluation, and treatment of high blood pressure. The latest NHLBI guidelines of this sort were published in 1992.
Screening All Kids
Labs are most likely to be affected by the panel's recommendations involving lipid screening and management of hyperlipidemia, which also are among its most controversial. The guidelines call for universal screening of children ages 9–11 and again at ages 17–19. These proposals, rated as grade B evidence and strongly recommended, are the first to endorse cholesterol screening in children aside from those with strong family histories of CVD. Panelist Peter Kwiterovich, MD, who also served on the 1992 NHLBI-sponsored committee, explained how the group arrived at this recommendation and why he views it as a major step forward.
"The 1992 guidelines committee discussed possible approaches to screening to detect children who may be at high risk for CVD. After reviewing the literature, there wasn't a lot of information available at that time in regard to general screening and what its relation to early lesions of atherosclerosis might be, or its effect on treatments, so the panel felt the recommendations should be focused on family history of high cholesterol. We appreciated that family history is not the greatest tool since often parents are too young to have had their cholesterol measured and people can be pretty ignorant about their family medical histories, but nevertheless that's what we were left with," he recalled. Kwiterovich is a professor of pediatrics and medicine, chief of the lipid research and atherosclerosis division, and director of the lipid clinic at Johns Hopkins University School of Medicine in Baltimore.
"I was disappointed with this recommendation, and I think we lost the interest of a lot of pediatricians because a family history was what was being done anyway," he continued. "Then, over the ensuing 20 years we've seen this onslaught of obesity and a lot of dyslipidemia in obese children. There's no simple answer as to what the best approaches to this problem might be. But this time, our panel was broader, and we were given the charge to review the literature in a very specific way based on an exhaustive search of areas related to children and adolescents. In terms of lipids, I was struck by the additional amount of information that was in the literature related to the origins of atherosclerosis in children and adolescents, its strong relationship to lipoproteins and other cardiovascular disease risk factors, and the effect of treatment with statins on the early lesions of atherosclerosis, all of which gave us a firm foundation to recommend universal screening."
Kwiterovich went on to explain that starting with evidence of early atherosclerosis in young Korean and Vietnam War casualties, a series of studies have come out, including several significant ones that were not available to the 1992 NHLBI committee. "In addition to pathological studies evaluating atherosclerosis in young people who had died, there were several huge cohort studies, primarily in Europe, where 5,000–6,000 subjects were enrolled as adolescents and restudied as adults," he said. "These looked at the relationship between risk factors at baseline and degree of carotid intima-media thickness. They showed that low density lipoprotein cholesterol in the upper 25th percentile was a strong predictor of the extent of carotid atherosclerosis. Obesity also was predictor of carotid intima-media thickness. This was strong evidence that the early subclinical lesions of atherosclerosis in adults could be predicted from risk factors in childhood."
The New Face of Childhood Dyslipidemia
Back in 1992, risk reduction focused primarily on identifying children with elevated low density lipoprotein-cholesterol (LDL-C) levels. However, since then childhood dyslipidemia has changed, and the predominant pattern—driven by obesity—involves moderate to severely elevated triglycerides, normal to mildly raised LDL-C levels, and low high density lipoprotein-cholesterol (HDL-C) levels, according to the guidelines. Both dyslipidemia paradigms place children at risk for early atherosclerosis.
Daniels also pointed out that the proposed screening age range of 9–11 was based on both physiologic and practical considerations. "If you look at the trajectory of LDL-cholesterol in particular, it starts relatively low at birth, goes up for the first 9 to 10 years, and it's at age 9 to 10 that your values are most reflective of what they're going to be as an adult. So that's the major reason that age range was picked. It also happens to be the time when atherosclerosis starts in a more accelerated way," he observed. "In addition, we felt that kids at those ages are still in a family situation where parents can have an impact on their lifestyle."
The panel recommended lipid screening in children younger than age 9 only if they have a strong family history of early CVD events like myocardial infarction or stroke, parents with total cholesterol ≥240 mg/dL, or if the child himself has significant risk factors such as hypertension or diabetes. Reflecting the natural decline in cholesterol during puberty, the guidelines do not advise routine screening between ages 12–16 unless the child has a strong family history of CVD. The document recommends universal screening again in youth ages 17–21.
Screening with Nonfasting Tests
In addition to universal screening, another important change reflected in the guidelines is the panel's endorsement of nonfasting lipid testing. The guidelines note that non-HDL-C has proven to be as powerful a predictor of atherosclerosis as any other lipoprotein measurement, and seems to be more predictive than total cholesterol, LDL-C, or HDL-C levels alone. "The thing that changes the playing field more than anything is the acknowledgement that a non-fasting non-HDL-cholesterol measurement is a very useful screening tool," said Ronald Krauss, MD, senior scientist and director of atherosclerosis research at Children's Hospital Oakland Research Institute in Oakland, Calif. "That's not at all been a part of the current approach to lipid screening and it's obviously most relevant to kids in whom it might be difficult to achieve the fasting test."
Alan Remaley, MD, PhD, concurred with this assessment. "The figures are maybe 20–30 percent of fasting samples really aren't fasting, and that compromises the results. There's also pretty persuasive data that non-HDL-cholesterol maybe as good as or even better than LDL-cholesterol," he said.
Remaley, who is section chief of the Lipoprotein Metabolism Section of the Pulmonary and Vascular Medicine Branch at the National Institutes of Health, recently co-authored two reports that evaluated direct HDL-C and LDL-C assays and found some issues. In one study, several direct methods met the National Cholesterol Education Program's (NCEP) total error goals for nondiseased individuals, but none of the methods met the goals for diseased individuals (Clin Chem 2010;56:977–986). In the other, most direct LDL-C methods did not produce improved risk score classification compared with calculated LDL-C methods (Clin Chem 2011;57:490–501). The latter also demonstrated that non-HDL-C assays showed the best concordance with the reference measurement procedure for CVD risk score classification in both normal and hypertriglyceridemic subjects.
The panel also left the door open for measurement of apolipoprotein B (Apo B) and Apo A-1, although it noted that these lipid subclasses provide no advantage for screening over standard cholesterol measures. Both Krauss and Kwiterovich agreed that there are practical implications for most clinicians in focusing on more common cholesterol measurements. "Apo B is another way to approach screening, but we felt it was a more difficult concept to get across to practitioners in that many of them have not used it," said Kwiterovich. "On the other hand, they're familiar with total cholesterol and HDL-cholesterol so they just need to know you can measure non-HDL-cholesterol, which gives you all the atherogenic lipoproteins. That's a little easier to get across than number of LDL particles and size."
Kid-specific Cutpoints
The guidelines recommend two sets of age-based low, acceptable, borderline, and high cutpoints, one for children and adolescents up to age 19, and one for young adults ages 20–24 (See Table, below). Although other studies have evaluated various cutpoints in children, the committee felt that levels proposed by the 1992 NHLBI panel have stood the test of time when it comes to both sensitivity and specificity in detecting or ruling out dyslipidemia.
Both sets of cutpoints will require adjustments by some labs, which apparently report cholesterol levels for patients in these age groups using adult cutpoints, an issue Daniels deals with often in his pediatric cardiology practice. "I see a lot of kids who come to clinic with interpretations that are based on adult cutpoints. Since there's been a focus on adults with heart disease getting LDL-cholesterol lower and lower, I'm now seeing kids referred with LDL-cholesterol slightly above 100, which is a normal value for a child," he observed. "More lab directors understand the pediatric component of lipid testing and will need to if the guidelines are adopted in a broad way. They're going to see more and more pediatric lipid testing. It'll be important to get the interpretation right because if a child and family have to go through a referral process for an evaluation and then treatment, whether it's with diet or whatever, it's important to get the cutpoints right because that will save everybody a lot of trouble."
|
Recommended Lipid and Lipoprotein Cutpoints
The NHLBI panel on cardiovascular health and risk reduction in children and adolescents recommended age-specific lipid and lipoprotein cutpoints.*
|
|
Category
|
Low
|
Borderline
Low
|
Acceptable
|
Borderline
High
|
High
|
| Youth up to age 19 |
| Total C |
|
|
<170
|
170–199
|
≥200
|
| LDL-C |
|
|
<110
|
110–129
|
≥130
|
| Non-HDL-C |
|
|
<120
|
120–144
|
≥145
|
| HDL-C |
<40
|
|
>45
|
40–45
|
|
| Apolipoprotein A-1 |
<115
|
|
>120
|
115–120
|
|
| Apolipoprotein B |
<90
|
90–109
|
≥110
|
|
|
Triglycerides
(0–9 years) |
|
|
<75
|
75–99
|
≥100
|
Triglycerides
(10–19 years) |
|
|
<90
|
90–129
|
≥130
|
|
Young Adults Age 20–24
|
| Total C |
|
|
<190
|
190–224
|
≥225
|
| LDL-C |
|
|
<120
|
120–159
|
≥160
|
| Non-HDL-C |
|
|
<150
|
150–189
|
≥190
|
| HDL-C |
<40
|
40–44
|
>45
|
|
|
| Triglycerides |
|
|
<115
|
115–149
|
≥150
|
|
*All values in mg/dL
Abbreviations: Total C, total cholesterol; LDL-C, low-density lipoprotein cholesterol; Non-HDL-C, non-high-density lipoprotein; HDL-C, high-density lipoprotein.
Adapted from Pediatrics 2011;128:S6
|
The Statin Issue
The panel's most provocative recommendation involves its endorsement of lipid-lowering medications in children. The guidelines make clear that these drugs should be prescribed only under specific conditions, but drug safety concerns and the realities of prescription patterns have raised red flags for some experts.
The document's detailed dyslipidemia algorithms call for results from two fasting lipid panels obtained 2 weeks to 3 months apart, and when LDL-C and triglyceride levels are above specific cutpoints, to either refer the child directly to a lipid specialist or assess him or her further for secondary causes of these problems, followed by lifestyle modifications for at least 6 months. Further testing and reassessments would follow these interventions. The panel advised statin therapy in children with persistent LDL-C levels >160 mg/dL only when they also have two high-level risk factors, one high-level and at least two moderate risk factors, or clinical CVD. In contrast, the panel advised only continued follow-up in children with the same range of LDL-C values but no other risk factors. In fact, the guidelines recommend drug treatment in children with no additional risk factors only when their LDL-C is >190 mg/dL, suggesting a genetic cause of their dyslipidemia.
Therein lies the rub for some observers, including Steven Nissen, MD, chair of cardiovascular medicine at the Cleveland Clinic, who has been a vocal critic of the drug approval process. "I do not favor the guidelines as currently written. My concern is that the impact will be that more children will be treated with statin drugs when in fact there isn't any evidence that those drugs will benefit children of that age," he said. "We don't have trials that have tested whether or not statin administration in children reduces their likelihood of developing coronary artery disease as adults. So if we give these drugs, we're essentially practicing medicine by guessing that they'll work rather than proving they'll work."
Nissen was not alone in this concern. "There's a mismatch between the quality of evidence about the association between risk factors and events and the quality of evidence about the risk-benefit profile of lipid-lowering treatment in children," said Bruce Psaty, MD, PhD, a professor of medicine and epidemiology at the University of Washington in Seattle. "The evidence is strong that risk factors in children are important, tend to track into older adults, and that the duration of the risk factor is frequently associated with the level of cardiovascular risk. But the evidence that drug treatment in children is beneficial is less well documented." He added that although statins are known to be generally safe and effective in adults, other lipid-lowering drugs lack evidence that they prevent cardiovascular events in adults. For example, the U.S. Food and Drug Administration has warned that in studies involving adults with type 2 diabetes, fenofibrate did not reduce CVD-related morbidity or mortality. "There we've got a drug that's widely used in adults with very little evidence, and it would be sad to see it used in children," he said. Kwiterovich countered that the panel recommended fibrate or niacin therapy only in children whose triglycerides remain very elevated after a trial of lifestyle modification. In such cases, the medications would be aimed at reducing the risk of pancreatitis rather than CVD.
The panel referenced at least 20 studies that have evaluated lipid-lowering medications in children, all of relatively small size and with 2 years' follow-up at most. Nissen and Psaty both argued that that is simply too thin an evidence base to commit children to essentially life-long therapy.
Panel members and others contended that such long term safety and efficacy evidence in children probably would not be forthcoming, and that the relatively remote likelihood of known adverse events from statins would need to be weighed against the downside of not reducing CVD risk early-on. "There are studies of statins—albeit not life-long, so not about life-long risks—but we certainly know about risks of lifelong untreated persistent hypercholesterolemia. The question that's come up and probably will never be answered in any kind of proper scientific way, is how long can we wait before intervening when we're dealing with lifetime risk?" said Krauss. "So there's a good case for the life-long risk justifying what appears to be a minimal risk of adverse consequences with statins, but with the caveat that we don't have information about lifetime risk of adverse events. It's sort-of a free-floating anxiety because it doesn't have any foundation in any known effect."
An Epidemic of Statin Prescriptions?
Nissen also argued that physicians would reach for their prescription pads rather than follow the panel's thoughtful risk-related drug therapy recommendations. Both he and Psaty contended that statin use in children could skyrocket. Indeed, in a published commentary, Psaty suggested that the guideline's dyslipidemia-related recommendations "are likely to result in an epidemic of cholesterol screening and lipid-lowering drug therapy in children," and that as many as 200,000 children and adolescents might qualify for drug treatment.
Daniels was not persuaded by those arguments. "It's important to note that the cutpoints for starting medication have not changed since the early 1990s," he said. "The suggestion that our recommendations will lead to rampant statin use doesn't do justice to primary care pediatricians and family physicians, whom I think are very careful about using medications and thinking about when it's appropriate to use them. Furthermore, these concerns are not backed up by evidence. The question is more about how to identify the children who would most benefit from medication, and that's where the guidelines come in. The recommendations for the more aggressive type of treatment with medications are really built around dyslipidemia due to genetic causes."
The Education Equation
As experts bat the evidence ball around, labs should be prepared for an influx of pediatric lipid testing. Remaley urged laboratorians to take a look at their cholesterol assays. "The guidelines point out that non-HDL-cholesterol is an acceptable and maybe even preferred alternative, but if labs don't report it, people simply don't pay attention to it. So most labs, if they're not currently doing so, should start reporting non-HDL-cholesterol. They also should make sure they have a good HDL assay. Most of the total cholesterol assays on the market are pretty good, but the HDL assays can vary." The American Diabetes Association, American College of Cardiology, and the National Lipid Association all recommend that non-HDL-C should be reported on all lipid profiles. The AACC Lipoproteins and Vascular Diseases Division Working Group on Best Practices also agreed "that a greater emphasis on non-HDL-C rather than LDL-C will improve patient care" (Clin Chem 2009;55:407–419).
Labs also will need to step up efforts to educate clinicians about lipid cutpoints and associated risks. For instance, Krauss suggested that labs clean up the way they report lipid results. "Right now, there's a whole mess of ways that labs are providing risk assessments, and it's confusing for many people. Using the adult guidelines as an example, the way they're provided in the laboratory report for cholesterol, it looks like you're at high risk if your LDL level is less than 70 mg/dL. What it really means is, if you are high risk, then your LDL should be less than 70 mg/dL. So I think the pediatric values could be packaged in a way that's clearer," he said.
Given the sweeping nature of the document, and perhaps taking a cue from efforts to better disseminate NCEP's Adult Treatment Panel III recommendations, NHLBI is preparing a series of implementation tools designed to simplify the guideline's many algorithms and intervention points, according to Daniels. He also expects the panel's recommendations eventually to be incorporated into AAP's Bright Futures, a set of health supervision guidelines that many pediatricians rely on in providing well-child care.
Daniels added that perhaps the best thing laboratorians can do is to sit down with clinicians and review the guideline's lipid screening recommendations. "We've had a long discussion with our lab about how to incorporate these cutpoints and how to report out these values, and I think labs across the country, whether they're in a children's hospital lab or not, should be doing the same thing."