Background: Familial hypercholesterolemia (FH) is an inherited disorder of lipoprotein metabolism characterized by highly elevated total cholesterol (TC) concentrations early in life, independent of environmental influences. Around 1 in 200 to 1 in 500 persons in North America and Europe are estimated to have heterozygous FH. When untreated, FH is associated with a high incidence of premature clinical atherosclerotic cardiovascular disease.
Purpose: We conducted a systematic evidence review of the benefits and harms of screening children and adolescents for heterozygous FH. The purpose of this review is to assist the U.S. Preventive Services Task Force (USPSTF) in updating its previous recommendations on such screening.
Data Sources:
We searched MEDLINE, the Cochrane Central Register of Controlled Trials, and PubMed from 2006 through July 2014 to locate relevant trials for all key questions (KQs) published since the previous reviews in support of prior recommendations. We supplemented these searches with reference lists from relevant existing systematic reviews, cohort studies, suggestions from experts, and
Study Selection: Investigators independently reviewed 6,752 abstracts and 375 articles against a set of a priori inclusion criteria. Investigators also independently critically appraised each study using design-specific quality criteria based on USPSTF methods. We included fair- or good-quality studies that met the a priori criteria for each KQ. We resolved discrepancies by consensus.
Data Extraction and Analysis: One investigator abstracted data from the 27 included articles into evidence tables and a second reviewer verified the accuracy of the abstracted data. We qualitatively summarized the evidence for screening and the effects of treatments on health outcomes. Lipid concentrations and measures of atherosclerosis were expressed as percent change from baseline or as differences from baseline. For KQ6, the number of included studies was sufficient to permit meta-analysis. For the randomized trials of statins that reported means and standard deviations for percent change (k=6), we summarized the results using forest plots. We did not combine data across studies, given the variability in drug, dose, and intended duration in the included studies.
Results: We found no direct evidence for five KQs: the effectiveness of screening children and adolescents for FH in improving health outcomes (myocardial infarction [MI] or stroke) in adulthood (KQ1) or intermediate outcomes (lipid concentrations and atherosclerosis) in childhood (KQ2), the harms of screening for FH in children and adolescents (KQ4), the effectiveness of treating children and adolescents with FH on health outcomes (MI or stroke) in adulthood (KQ5), and the association between intermediate outcomes in childhood and adolescence and the future incidence or timing of MI and stroke in adulthood (KQ8). Studies met inclusion criteria for three KQs.
KQ3. What is the diagnostic yield of appropriate screening tests for FH in children and adolescents?
Two studies provided data allowing determination of the diagnostic yield of pediatric FH screening programs. A statewide universal screening program screened more than 80,000 10- to 11-year-olds in West Virginia schools and reported a diagnostic yield of about 1.3 cases per 1,000 screened. In this study, “probable FH” was defined as a low-density lipoprotein cholesterol (LDL-C) concentration greater than 155 mg/dL or TC concentration greater than 260 mg/dL plus DNA evidence of a low-density lipoprotein receptor (LDLR) mutation in a first- or second-degree relative. A Danish school-based study of more than 2,085 6- to 8-year-olds used the ApoB:ApoA-1 ratio and reported a diagnostic yield of 4.8 cases per 1,000 screened. We found no studies reporting diagnostic yield or effectiveness of selective screening for FH in youth (i.e., screening subjects with a family history or other targeting factor).
KQ6. Does treatment of FH with lifestyle modifications and/or lipid-lowering medications in children and adolescents improve intermediate outcomes (i.e., reduce lipid concentrations or reverse or slow the progression of atherosclerosis) in childhood and adolescence?
Eight good-quality randomized, controlled trials (RCTs) formed the evidence base for statin treatment of FH in youth. Studies of statins ranged from 6 weeks to 2 years long, with most shorter than 1 year. Treatment with statins lowered LDL-C and TC concentrations in the short term in children and adolescents with FH, with most studies reporting that statins lowered LDL-C by 20 to 40 percent compared to placebo. The greatest effect on LDL-C was in a trial of rosuvastatin. Participants who received the highest dose (20 mg/day) experienced a 50 percent decrease (least mean squares) in LDL-C from baseline compared to a 1 percent decrease among controls (p<0.001).
Eight studies reported the effect of statins on TC, all showing decreases of about 20 to 30 percent from baseline (compared to no change with placebo). The effect on high-density lipoprotein cholesterol (HDL-C) was minimal or null. A single study assessed the effect on a measure of atherosclerosis and found that pravastatin reduced carotid intima-media thickness (CIMT) by 2.01 percent (compared to a 1.02% increase in the control group; p=0.02). There were no consistent differences in treatment effect among different statins, but the number of studies for any one drug was limited. The two studies that compared different doses of statins reported a dose response with pravastatin and rosuvastatin. In the 2010 rosuvastatin trial, the only statin study reporting how many subjects attained the target LDL-C concentration, only 12 to 41 percent of participants reached a target LDL-C of less than 110 mg/dL, with greater effect at higher doses.
Six studies of statins provided the necessary data to create a forest plot of mean difference across statins between percent change from baseline of TC, LDL-C, and HDL-C. Treatment effects on TC and LDL-C were statistically significant for all five drugs in these six studies (atorvastatin, lovastatin, pravastatin, simvastatin, and rosuvastatin), with overlapping 95% confidence intervals across drugs.
Five fair- to good-quality RCTs evaluated nonstatin drugs in children and adolescents with FH. All trials reported decreases in LDL-C from baseline. Three RCTs studied bile-sequestering agents. A good-quality trial of colestipol found a mean reduction in LDL-C of 19.5 percent after 8 weeks of treatment compared to a 1 percent decrease in the control group. One fair-quality RCT of cholestyramine found an 18.6 percent reduction in LDL-C after 1 year compared to a 1.5 percent increase in the control group. One good-quality 8-week RCT of colesevelam published after the 2007 USPSTF review found a least squares mean decrease in LDL-C of 10 percent (standard error [SE], 2.1%) at the higher of two doses compared to a least squares mean increase of 2.5 percent (SE, 2.0%). A lower dose provided a smaller, nonsignificant reduction. Two good-quality RCTs of ezetimibe were published after the 2007 USPSTF review. One reported that LDL-C decreased by a mean of 54.0 percent (SE, 1.4%) in participants who received a combination of ezetimibe and simvastatin, whereas the mean decrease was 38.1 percent (SE, 1.4%) in the simvastatin-only group at 33 weeks. The second found that, at 12 weeks, ezetimibe monotherapy decreased LDL-C by 28 percent (95% CI, -31 to -25) from baseline compared to a negligible change in the placebo group.
KQ7. What are the harms of treatment of FH with medications in children and adolescents?
There is a fair- to good-quality body of evidence about the short-term harms of pharmacologic treatment of children and adolescents with FH. Most studies were conducted outside the United States but were applicable to U.S. primary care setting. Most studies were of short duration (6 weeks to 2 years); the longest was 10 years. Statins were generally well-tolerated, although reversible elevations of liver enzymes and/or creatine kinase concentrations were noted in some studies. One study found lower dehydroepiandrosterone sulfate concentrations in men with FH treated with statins compared to unaffected siblings. Bile-acid binding resins were commonly associated with adverse gastrointestinal symptoms and poor palatability. Long-term harms are unknown.
Limitations: Direct evidence for the impact of screening on intermediate or health outcomes is lacking. One of the two studies assessing the diagnostic yield of screening for FH may not be generalizable to a U.S. population, and the other provides few details as to the screening and confirmatory testing for FH. Evidence on the effectiveness of pharmacotherapy lacks long-term studies assessing the effect of lipid-lowering medications on intermediate outcomes in childhood and adolescence or on health outcomes in adults. Participants in the eight statin trials were patients at tertiary care centers; none of the studies were conducted in screen-detected populations. Few studies were conducted in nonwhite populations. Three statin trials included children as young as age 8 years; however, the age distribution of the statin studies as a whole is skewed to early adolescence. We found no studies comparing outcomes between groups of children or adolescents who initiated treatment at different ages. Long-term studies of harms of pharmacotherapy in youth are lacking. Finally, this review was limited to FH alone; other atherogenic dyslipidemias are addressed in a separate review.
Conclusions: We found no direct evidence of the effect of screening on intermediate or health outcomes. The evidence describing the diagnostic yield of screening for FH in children is minimal. There is good evidence of the effectiveness of statins in reducing LDL-C and TC concentrations in studies up to 2 years long and limited evidence of a statin effect on measures of atherosclerosis. Statins were generally well-tolerated in the short term, although reversible elevations of liver enzymes and/or creatine kinase concentrations were noted in some studies and a decrease in dehydroepiandrosterone sulfate was noted in one study. Bile-acid binding resins were commonly associated with adverse gastrointestinal symptoms and poor palatability. Long-term harms are unknown. Randomized trials of screening for FH in U.S. youth are needed, as are longer-term treatment trials evaluating the benefits and harms of medications in children and adolescents with FH.