A Systematic Review and Meta-Analysis of Statin Therapy in Children With Familial Hypercholesterolemia
Objective— Functional and morphological changes of the arterial wall already present in young children with heterozygous familial hypercholesterolemia (HeFH) suggest that treatment should be initiated early in life to prevent premature atherosclerotic cardiovascular disease. The purpose of this study was to assess the efficacy and particularly safety of statin therapy in children with HeFH.
Methods and Results— We performed a meta-analysis of randomized, double-blind, placebo-controlled trials evaluating statin therapy in children aged 8 to 18 years with HeFH. Six studies (n=798 children) with 12 to 104 weeks of treatment were included. Total cholesterol, LDL cholesterol, and apolipoprotein B were significantly reduced, whereas HDL cholesterol and apolipoprotein A1 were significantly increased by statin therapy. No statistically significant differences were found between statin- and placebo-treated children with respect to the occurrence of adverse events (RR 0.99; 95% CI: 0.79 to 1.25), sexual development (RR of advancing ≥1 stage in Tanner classification 0.96; 95% CI: 0.79 to 1.17), muscle toxicity (RR of CK ≥10 times the upper limit of normal [ULN] 1.38; 95% CI: 0.18 to 10.82), or liver toxicity (RR of ≥3 times the ULN for ASAT 0.98; 95% CI: 0.23 to 4.26 and for ALAT 2.03; 95% CI: 0.24 to 16.95). We found a minimal difference in growth in favor of the statin group (0.33 cm; 95% CI: 0.03 cm to 0.63 cm).
Conclusion— In addition to the fact that statin treatment is efficacious, our results support the notion that statin treatment in children with HeFH is safe. Thus, even though further studies are required to assess lifelong safety, statin treatment should be considered for all children aged 8 to 18 with HeFH.
Heterozygous familial hypercholesterolemia (HeFH) is a common monogenetic disorder with a prevalence of about 1 in 500 in Whites.1 It is characterized by defective low-density lipoprotein cholesterol (LDL-C) receptors on the surface of hepatocytes. This leads to severely elevated levels of plasma LDL-C from birth onwards,2 causing premature atherosclerosis and cardiovascular disease (CVD).3 If untreated, about 50% of males and 30% of females will develop CVD before the age of 60.4 In adults with HeFH, 3-hydroxy-3-methylglutaryl (HMG)-coenzyme A (CoA) reductase inhibitors (statins) are the core of pharmacological therapy. They have been proven safe and well-tolerated agents that reduce LDL-C levels as well as the incidence of coronary artery disease, stroke, and peripheral vascular disease.5
Even though cardiovascular events are rare in childhood, children with HeFH already have functional and morphological changes of the vessel wall. This is illustrated by an impaired flow-mediated dilatation of the brachial artery6 and an increased intima media thickness (IMT) of the carotid artery.7 Both are surrogate markers for atherosclerotic vascular disease8 and, thus, indicate that the atherosclerotic process has already been initiated early in childhood. Indeed, myocardial ischemia and coronary artery stenoses have been documented in young adults with HeFH.9,10 These findings strongly suggest that treatment should be started at young age. Until recently, only dietary interventions and bile acid binding resins were recommended but efficacy and compliance were mostly poor.11,12 Based on the efficacy and safety in adults, statins were expected to be beneficial for children as well. Hence, a number of clinical studies were performed to evaluate statin therapy in children with HeFH. All studies supported the efficacy of statin therapy during childhood. Consequently current scientific statement from the American Heart Association recommends initial treatment in childhood with statins at age ≥10 years in males and after the onset of menses in females.13 However, most studies that evaluated statin therapy in children with HeFH were powered on the primary efficacy outcome and other outcomes, such as safety, were likely to be underpowered. Therefore, we sought to perform a meta-analysis on safety outcomes of randomized placebo-controlled trials which evaluated statin treatment in children and adolescents with HeFH. In addition, efficacy outcomes were also considered.
We attempted to find all publications describing randomized controlled trials in which statin treatment was compared with placebo in patients with heterozygous familial hypercholesterolemia aged less than 18 years. Potentially eligible studies were retrieved via electronic databases (MEDLINE, EMBASE, CENTRAL). The following (combination of) medical subject headings and free text keywords were used: “familial hypercholesterolemia”, “hydroxymethylglutaryl-CoA reductase inhibitors”, “statin”, “atorvastatin”, “fluvastatin”, “lovastatin”, “pravastatin”, “simvastatin”, “rosuvastatin”, “randomized controlled trial”, “controlled clinical trial”, “random allocation”, “clinical trial”, “double blind”, “placebo”, and “placebo controlled”. To select a study population of less than 18 years of age we used a comprehensive search string that was standardized in our institution. Additional studies were sought by a manual search through reference lists of relevant publications, recent reviews, and editorials and through personal communication with experts in the field.
Study Selection and Data Extraction
Two reviewers determined the eligibility of retrieved studies independently, according to predetermined criteria. Difference in judgment by the reviewers was solved by discussion and consensus. Language restrictions were not imposed.
Inclusion was restricted to randomized placebo-controlled trials evaluating statin therapy in patients with HeFH ≤18 years of age. The following criteria were accepted for the diagnosis of HeFH: (1) genetic diagnosis, (2) fasting LDL-C ≥4.9 mmol/L (190 mg/dL), or (3) fasting LDL-C ≥4.1 mmol/L (160 mg/dL) with at least 1 biological parent with HeFH and a family history of premature atherosclerotic disease defined as onset of clinical atherosclerotic disease before age 55 in males or age 65 in females in first or second degree relatives. Studies were excluded if lipid lowering comedication was used, if treatment was unblinded, if they were duplicate reports or preliminary reports of data later presented in full, or if none of the following outcome measures were reported: lipid profile, IMT, or safety parameters.
Data of selected articles were extracted by 2 reviewers independently, following a predetermined form. In case a study had multiple treatment arms with different doses of a statin and only 1 placebo group, data of the treatment arm with the highest dose were extracted. If necessary, authors were approached for additional data. Difference in judgment by the reviewers was solved by discussion and consensus.
Efficacy and Safety Outcomes
To assess efficacy, we evaluated the differences in percentage change in total cholesterol (TC), LDL-C, high-density lipoprotein cholesterol (HDL-C), triglycerides, apolipoprotein B (ApoB), and apolipoprotein A1 (ApoA1) between statin- and placebo-treated children, and we evaluated the differences in absolute changes in intima media thickness (IMT). With respect to safety, we assessed the number of subjects with one or more adverse events and those with elevated aspartate aminotransferase (ASAT) or alanine aminotransferase (ALAT) more than 3 times the upper limit of normal (ULN) or creatinine kinase (CK) more than 10 times the ULN. Furthermore, we assessed the differences in growth and sexual maturation, expressed as the number of patients that proceeded one Tanner stage or more, during the study. Finally, we described the change in levels of testosterone, estradiol, follicle-stimulating hormone (FSH), luteinizing hormone (LH), and dehydroepiandrosterone (DHEAS).
For continuous outcomes, we calculated the differences in means between the statin and placebo arms and the combined variance for each study. If study results for a certain outcome were homogeneous, we used a fixed model that considers consistency in treatment effects over individual trials with natural variation around a constant effect. Then, the pooled mean difference was calculated by using the weighted sum of these differences, with the weight being the reciprocal of the combined variance for each study. In some cases, the mean changes for the treatment groups or standard deviation (SD) were not available in the source paper, nor after contacting the authors. If so, we estimated the difference by subtracting the mean at baseline from the mean after treatment. The SD was estimated by using formulas given by the Cochrane Handbook for Systematic Reviews of Interventions14 or Follmann et al.15
For dichotomous outcomes, we calculated a relative risk (RR) and 95% confidence interval (CI) for each trial separately. In case of homogeneous study results, the RRs were combined across studies using the Mantel-Haenszel procedure which assumes a fixed treatment effect.
Heterogeneity of study results was evaluated with the Chi2- and the I2- test, for each of the outcome measures separately. When the probability value of the Chi2- test was <0.10 and/or I2 was >50%, data were considered as heterogeneous. In case of significant heterogeneity, data from the studies were combined using a random effects model according to the method of DerSimonian and Laird.16 A Z test was performed to test the overall effect. All tests were performed using Review Manager 4.2.10 (The Cochrane Collaboration).
Description of Studies
Our search yielded 537 publications of which 10 were randomized placebo-controlled trials which evaluated statin treatment in patients with HeFH less than 18 years of age. Four of these had to be excluded because they met 1 or more exclusion criteria (Figure). The characteristics of the 6 remaining studies,11,17–21 published between 1996 and 2005, are shown in Table 1. The number of participants in the studies varied from 54 to 214. In total, 476 subjects received statin therapy and 322 subjects received placebo. The age of the participants ranged from 8 to 18. About 60% of the participants was male. The studies evaluated various types of statins for a period ranging from 12 to 104 weeks.
Two trials had adequate allocation concealment,18,19 whereas the other studies had unclear concealment.11,17,20,21 Discontinuations varied from 0%17 to 9%20 in the statin groups and 0%17 to 25%20 in the placebo groups (Table 1).
The study of de Jongh et al11 consisted of 2 periods. Twenty-one subjects did not continue to the second period of the study. Because the reason for discontinuation was unclear, we only analyzed data from the first study period (24 weeks). Furthermore, some safety data in this study were reported separately for male and female participants, whereas efficacy outcomes were presented for the total study population. Therefore, efficacy data are presented for males and females together and certain safety data are presented separately. In addition, the authors of this study provided additional safety data of the first period that were not previously published. Finally, because we had access to the original data set of the study of Wiegman et al,18 we were able to analyze outcomes for participants who received 20 or 40 mg pravastatin separately.
All included studies reported the effect of statin therapy on lipids and lipoproteins (Table 2). Compared with placebo, statin treatment reduced TC significantly with an overall mean of 23% (95% CI: −27% to −19%). The LDL-C reduction ranged from 21% (95% CI: −27% to −15%) for lovastatin 40 mg to 39% (95% CI: −44% to −35%) for atorvastatin 10 to 20 mg. Analysis of the pooled data of these studies showed a statistically significant reduction of 30% (95% CI: −36% to −24%). In general, the effect on TC and LDL-C increased with more potent drugs or higher doses. With respect to HDL-C, 2 of the 6 studies11,21 and the 40 mg group in the study of Wiegman et al18 showed a statistically significant increase after statin treatment as compared with placebo treatment. Analysis of the pooled data showed a mild but significant elevation of HDL-levels (3.64%; 95% CI: 1.33% to 5.94%). Data on triglycerides could not be pooled because of the naturally skewed distribution of triglyceride levels, but none of the individual studies showed a significant change.
The effects of statins on ApoB were slightly less but quite consistent with the effects on LDL-C. ApoB reductions ranged from 16% (95% CI: −23% to −9%) for lovastatin 40 mg to 35% (95% CI: −39% to −31%) for atorvastatin 10 to 20 mg, with an overall mean difference of 25% (95% CI: −31% to −19%). There was a small but significant increase in ApoA1 with simvastatin 40 mg. Analysis of pooled data revealed a significant increase of 2.43% (95% CI: 0.41% to 4.45%).
One study reported the effects of statin treatment on IMT.18 Results of this study showed a significant regression in carotid IMT (0.014±SD 0.046 mm; P=0.02) in the statin group as compared with the placebo group.
Four of the included studies reported on adverse events.11,17,19,21 One study showed less patients with adverse events in the statin-treated group,17 whereas the others did not find a significant difference in the occurrence of adverse events. Analysis of the pooled data revealed no increased risk of an adverse event when receiving statin therapy, as shown by a relative risk (RR) of 0.99 (95% CI: 0.79 to 1.25; Table 3).
With respect to growth, 4 studies reported data on height of the participants.11,18–20 One study showed a statistically significant mean difference in length for male participants that favored the treatment group,11 and analysis of the pooled data revealed a minimal but statistically significant change in height (0.33 cm; 95% CI: 0.03 cm to 0.63 cm) favoring the treatment group (Table 4).
None of the 3 studies that evaluated sexual development reported a difference between statin- and placebo-treated children.11,18,21 Analysis of the pooled data showed a RR of advancing to a next Tanner stage during the studies of 0.96 (95% CI: 0.79 to 1.17) for statin-treated subjects compared with those on placebo (Table 4).
In all studies laboratory data on ASAT, ALAT, and CK were collected. No significant differences in the number of children with marked elevations were reported between statin- and placebo-treated subjects. The data of Knipscheer et al17 were not included in the meta-analysis because they were not expressed in terms of more than 3 times the ULN (ASAT, ALAT) or more than 10 times the ULN (CK). Analysis of the number of children with elevated ASAT or ALAT more than 3 times the ULN of the 5 other studies11,18–21 showed a RR of 0.98 (95% CI: 0.23 to 4.26) for ASAT and 2.03 (95% CI: 0.24 to 16.95) for ALAT. Data on CK were also pooled for these 5 studies,11,18–21 and the RR for having an elevation of CK of more than 10 times the ULN amounted 1.38 (95% CI: 0.18 to 10.82; Table 5).
Four studies11,18–20 reported values of (some of) the hormones testosterone, estradiol, FSH, LH, DHEAS, cortisol, corticotrophin, and thyrotrophin. Two studies reported a small but significant difference in DHEAS levels between the statin- and placebo-treated group.11,20 De Jongh et al11 found a decrease whereas Stein et al20 found increased DHEAS levels in the statin group. Clauss et al19 found a slightly decreased LH in the placebo group (P<0.05) compared with the lovastatin group. Other hormone levels did not differ between statin-treated children and children on placebo.
In this systematic review, data from 6 randomized controlled clinical trials were analyzed to evaluate the efficacy and safety of statin therapy in children aged 8 to 18 years with HeFH. It could be determined that statins significantly reduced TC, LDL-C, and ApoB as well as significantly increased HDL-C and ApoA1 levels when compared with placebo. Furthermore, none of the safety parameters differed between the statin- and placebo-treated groups, except for a minimal difference in growth during the studies that favored the treatment group. These findings, in our opinion, underline both the efficacy and safety of statin therapy for children with HeFH.
In adult patients, efficacy and safety of statin therapy has already been studied extensively.22–24 In a systematic review and meta-analysis, Law et al23 quantified the effect of statins on LDL-C for the various types and doses of statins in adults with ischemic heart disease. In fact, the reductions of LDL-C in the studies included in our meta-analysis were quite similar compared with those presented in the meta-analysis of Law et al.23 For example, atorvastatin 10 to 20 mg reduced LDL-C levels by 39% (95% CI: −44% to −35%) in children and by 43% (95% CI: −47% to −40%) in adults (atorvastatin 20 mg). Also, pravastatin 20 mg reduced LDL-C by 30% (95% CI: −37 to −22) and 23% (95% CI: −29% to −8%) in 2 pediatric studies versus 24% (95% CI: −26% to −23%) in adults. Thus, disparity in age does not seem to influence the efficacy of statin therapy.
Arambepola et al recently published a review about the efficacy and safety of statin therapy in children with HeFH.25 Whereas they also included a randomized crossover trial, a nonrandomized parallel matched trial, time series, and several prospective case series, we explicitly restricted inclusion to double-blind randomized placebo-controlled trials. Although their meta-analysis for LDL-C, TC, and HDL-C was based on fewer subjects, the reported overall mean differences were comparable to our findings. However, Arambepola et al presented safety outcomes only descriptively. In contrast, we were able to perform a meta-analysis on safety outcomes. Therefore, our review provides additional crucial information for current pediatric practice.
Muscle and liver toxicity caused by statin therapy constitutes the main concern in clinical practice, both for adults and children. A recent systematic review on statin safety in adults estimated a very rare occurrence of rhabdomyolysis of 3 per 100 000 person-years for atorvastatin, simvastatin, lovastatin, pravastatin, and fluvastatin. Also, in that review, doubt was expressed toward the occurrence of liver disease attributable to statins.24 In our meta-analysis, the total number of subjects or person-years was far too small to estimate the risk of rhabdomyolysis. Five studies reported on elevations of ASAT, ALAT, and CK. No substantial elevations occurred in the treatment groups nor the placebo groups in 2 studies for ASAT and 3 studies for ALAT and CK. Because of the properties of Review Manager 4.2.10, which is the standard package of the Cochrane Collaboration for meta-analyses, these studies were not taken into account for calculation of the RR of these parameters. Consequently, the reported overall RR was only based on 2 or 3 studies. However, if we look at the absolute numbers of events, the total number of subjects with a CK elevation of more than 10 times the ULN was only 2 in the statin group (n=454) versus 1 in the placebo group (n=308). Similarly, 4 cases in the statin group (n=454) and 2 in the placebo group (n=308) had an elevation in ASAT of more than 3 times the ULN; for ALAT these figures were 3 cases in the statin group (n=454) and no cases in the placebo group. This implies that liver- and muscle-related adverse events are also rare in children. Nevertheless, because of the sample size studied and the relatively short duration of the studies, definite conclusions with respect to liver- and muscle-related adverse events cannot be drawn. Thus, even though statins seem safe, long-term muscle and liver safety in children and adolescents should still be monitored. However, based on current evidence, the risk of liver- and muscle-related adverse events does not appear to be a reason to withhold statin therapy in children and adolescents with HeFH.
The population in our study comprised children and adolescents, who are maturing to adulthood, and hence we particularly evaluated growth and sexual development. Meta-analysis showed that statin therapy did not impair growth or sexual development. For sexual development, we measured the relative risk of advancing to a next Tanner stage during the study. Ideally, sexual development should be monitored into full adulthood to assess whether full maturation (Tanner stage V) is reached. Long-term follow-up studies should evaluate sexual maturation into adulthood.
Cholesterol is a precursor for steroid and sex hormones, and as a consequence, possible effects of cholesterol-lowering agents on hormonal status are an important concern in children. Accordingly, 4 studies measured hormone levels.11,18–20 We did not perform a meta-analysis for these hormones, but the individual studies showed small but significant changes in both DHEAS and LH. Stein et al reported a small increase of DHEAS whereas de Jongh et al reported a small decrease11,20 in the statin-treated group. The 2 other studies did not find any differences.18,19 With respect to LH, Clauss et al showed slightly decreased levels of LH in the placebo group after 24 weeks of therapy, but levels in the lovastatin group were not affected. The difference between the statin- and placebo-treated groups was statistically significant.19 In other studies, LH changes did not differ between the statin- and placebo-treated children18,20 or LH levels were undetectable.11 Altogether, these findings suggest a large variability in hormone outcomes. This may be attributable to the fact that many of the hormones vary substantially within normal ranges, dependent on day- and nighttime (eg, cortisol) or menstrual cycle (eg, FH, FSH).26 It is unclear whether this was taken into account when data were collected. Furthermore, all differences between statin- and placebo-treated children were probably too small to be of any clinical relevance as evidenced by an absence of growth or pubertal development abnormalities. Thus, based on currently existing data, it is highly unlikely that statins affect hormonal regulation in children and adolescents with HeFH.
The findings discussed above support the safety of statin therapy in children with HeFH. However, a limitation of our study is the duration of statin therapy in the studies included. This ranged from 12 to 104 weeks, whereas in clinical practice patients with FH are likely to continue statin treatment for the rest of their lives, once therapy is initiated. Even though these studies do not indicate adverse effects, further long-term follow-up studies are required to assess lifelong safety.
All included studies, except 1 which evaluated IMT,18 measured the effect of therapy on lipids as the primary efficacy parameter. Although the relation between elevated LDL-C and CVD and the effect of lipid lowering treatment on CVD is evident,22 the effect of lipid lowering treatment in childhood on CVD later in life is not as clear. Furthermore, the optimal age to start treatment is unknown as yet.13 More randomized clinical trials on (surrogate) clinical end points such as IMT and, ideally, on cardiovascular events and death should be performed in order to further clarify these issues.
Based on the available evidence we conclude that statin therapy in children with HeFH is efficacious without untoward effects on safety, although further studies should assess lifelong safety. Because functional and morphological arterial wall changes are already present in these children, statin treatment should be considered for every child diagnosed with HeFH.
We thank Dr A. Hutten for kindly translating the Russian publication.
E.A. Stein is consultant, speaker, or performed funded research for Schering-Plough, Novartis, Takeda, Merck, ISIS, Wyeth, Reliant, DaiichiSankyo, and AstraZeneca, J.J.P. Kastelein has received consulting fees, lecture fees, and grant support from Pfizer, Merck, Schering-Plough, AstraZeneca, Bristol-Myers Squibb. and Sankyo.
Original received April 2, 2007; final version accepted May 22, 2007.
Hutter CM, Austin MA, Humphries SE. Familial Hypercholesterolemia, Peripheral Arterial Disease, and Stroke: A HuGE Minireview. Am J Epidemiol. 2004; 160: 430–435.
Stone NJ, Levy RI, Fredrickson DS, Verter J. Coronary artery disease in 116 kindred with familial type II hyperlipoproteinemia. Circulation. 1974; 49: 476–488.
Mabuchi H, Koizumi J, Shimizu M, Takeda R. Development of coronary heart disease in familial hypercholesterolemia. Circulation. 1989; 79: 225–232.
Mouratidis B, Vaughan-Neil ejectionfactor (EF), Gilday DL, Ash JM, Cullen-Dean G, McIntyre S, MacMillan JH, Rose V. Detection of silent coronary artery disease in adolescents and young adults with familial hypercholesterolemia by single-photon emission computed tomography thallium-201 scanning. The Am J Cardiol. 1992; 70: 1109–1112.
de Jongh S, Ose L, Szamosi T, Gagne C, Lambert M, Scott R, Perron P, Dobbelaere D, Saborio M, Tuohy MB, Stepanavage M, Sapre A, Gumbiner B, Mercuri M, van Trotsenburg AS, Bakker HD, Kastelein JJ. Efficacy and safety of statin therapy in children with familial hypercholesterolemia: a randomized, double-blind, placebo-controlled trial with simvastatin. Circulation. 2002; 106: 2231–2237.
Kavey RE, Allada V, Daniels SR, Hayman LL, McCrindle BW, Newburger JW, Parekh RS, Steinberger J. Cardiovascular risk reduction in high-risk pediatric patients: a scientific statement from the American Heart Association Expert Panel on Population and Prevention Science; the Councils on Cardiovascular Disease in the Young, Epidemiology and Prevention, Nutrition, Physical Activity and Metabolism, High Blood Pressure Research, Cardiovascular Nursing, and the Kidney in Heart Disease; and the Interdisciplinary Working Group on Quality of Care and Outcomes Research: endorsed by the American Academy of Pediatrics. Circulation. 2006; 114: 2710–2738.
Higgins JP, Green, S. The Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 2006. Internet Communication (http://www.cochrane.org/resources/handbook/Handbook4.2.6Sep2006.pdf).
Clauss SB, Holmes KW, Hopkins P, Stein E, Cho M, Tate A, Johnson-Levonas AO, Kwiterovich PO. Efficacy and safety of lovastatin therapy in adolescent girls with heterozygous familial hypercholesterolemia. Pediatrics. 2005; 116: 682–688.
Stein EA, Illingworth DR, Kwiterovich PO Jr, Liacouras CA, Siimes MA, Jacobson MS, Brewster TG, Hopkins P, Davidson M, Graham K, Arensman F, Knopp RH, DuJovne C, Williams CL, Isaacsohn JL, Jacobsen CA, Laskarzewski PM, Ames S, Gormley GJ. Efficacy and safety of lovastatin in adolescent males with heterozygous familial hypercholesterolemia. A randomized controlled trial. J Am Med Assoc. 1999; 281: 137–144.
Baigent C, Keech A, Kearney PM, Blackwell L, Buck G, Pollicino C, Kirby A, Sourjina T, Peto R, Collins R, Simes R. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet. 2005; 366: 1267–1278.
Law MR, Wald NJ, Rudnicka AR. Quantifying effect of statins on low density lipoprotein cholesterol, ischaemic heart disease, and stroke: systematic review and meta-analysis. BMJ. 2003; 326: 1423–1427.
Arambepola C, Farmer AJ, Perera R, Neil HA. Statin treatment for children and adolescents with heterozygous familial hypercholesterolaemia: A systematic review and meta-analysis. Atherosclerosis. In press.
Kasper DL, Fauci AS, Longo DL, Braunwald E, Hauser SL, Jameson JL. Harrison’s Principles of Internal Medicine. The McGraw-Hill Companies, Inc., New York 2001.