ABC | Volume 111, Nº6, December 2018

Original Article Radaelli et al Statin treatment in children: meta-analysis Arq Bras Cardiol. 2018; 111(6):810-821 Table 2 – Risk of bias of included studies Study, year Adequate sequence generation Allocation concealment a Blinding of investigator Blinding of participant Blinding of outcome assessors Intention - to-treat analysis b Description of losses and exclusions Knipscheer et al., 1996 Not reported Unclear Not reported Not reported Not reported Yes No Stein et al., 1999 Not reported Unclear Not reported Not reported Not reported Yes Yes de Jongh et al., 2002 Not reported Unclear Not reported Not reported Not reported Yes Yes McCrindle et al., 2003 Not reported Unclear Not reported Not reported Not reported Yes Yes Wiegman et al., 2004 Yes Unclear Not reported Not reported Not reported No Yes Clauss et al., 2005 Yes Adequate Not reported Not reported Not reported Yes Yes Rodenburg et al., 2006 Not reported Unclear Not reported Not reported Not reported Yes No Van der Graaf et al. 2008 Not reported Unclear Not reported Not reported Not reported Yes Yes Avis et al., 2010 Not reported Unclear Not reported Not reported Not reported Yes Yes Braamskamp et al., 2015 Not reported Unclear Not reported Not reported Not reported Yes Yes a Allocation concealment: Adequate (randomization method described that prevents caregivers or investigators from interfering or identifying before randomization; Unclear (randomization stated but no further information provided). b Intention-to-treat analysis: Intention-to-treat and completeness of follow-up are assessed by results available at the end of trial. Yes (specified by authors and confirmed by our analysis), No (specified or not specified by authors but no evidence of intention-to-treat confirmed by our analysis). In this analysis, all subgroups maintained significant reductions in cholesterol levels (-104.61 mg/dl, -67.60 mg/dl, -56.96 mg/dl), and intragroup heterogeneity was lower (18%, 99.7%, 95.4%). This analysis explained 99.4% of the original heterogeneity found in the main analysis (Figure 2). Change in LDL-cholesterol level Ten included studies evaluated the effect of statin therapy on the LDL-cholesterol level. 10,11,16-23 All subgroup analysis demonstrated significant reduction in this level: [-105.03 mg/ dl (95% CI -115.76, -94.30), I 2 19.2%], [-67.85 mg/dl (95% CI -83.36, -52.35), I 2 99.8%], [-58.97 mg/dl (95% CI -67.83, -50.11), I 2 93.8%], (Figure 3). The detailed analyzes are in Appendices II, III, IV, and V. Discussion We quantitatively analyzed ten randomized placebo- controlled trials in children with FH. Studies showed a clinically significant reduction in LDL-cholesterol levels in children treated with statin, compared to those treated with placebo. In addition, therapy with statins slightly increased HDL-cholesterol. The reduction in LDL-cholesterol levels varied between studies, probably due to different statins and dosages, and, possibly due to different settings of HeFH. In our meta-analysis, the results of all studies using statins were combined. All statins included present a commonmechanismof action, i.e., inhibition of hydroxy-methyl-glutary-Coa. All statins have shown beneficial effects in lowering lipid levels and have been approved for use in adult patients with dyslipidemia. When comparing some results: the study using lovastatin to evaluate efficacy and safety in children, focusing on female population, concluded that the lovastatin group showed a reduction in LDL-cholesterol levels of 23% to 27% against an increase of 5% in the placebo group (p < 0.001), TC of 17% to 22%, and APOB of 20% to 23%. 10 Whereas another study with young male patients, 21 lasting 24 weeks, lovastatin significantly reduced LDL-cholesterol levels at all dosages compared with placebo (17%, 24%, 27%with dosage of 10, 20, and 40mg/day, respectively; p < 0.001). Further treatment with the dose of lovastatin at 40 mg/day (from 24 to 48 weeks) reduced LDL‑cholesterol by 25% compared to placebo (p < 0.001). In a study with pravastatin, the assessed primary efficacy outcome was the IMT, showing a significant difference between pravastatin versus placebo (p = 0.02). 18 Also, pravastatin reduced LDL-cholesterol levels (-24.1%) versus placebo (+0.3%) and p < 0.001. The authors suggest that IMT findings and efficacy of treatment with pravastatin in this study should be limited to children with FH. The efficacy results of this study were similar to others. At the end of 48 weeks, patients treated with simvastatin showed statistically significant reductions in LDL- cholesterol levels (-41%), TC (31%), APOB (-34%), very low-density lipoprotein (VLDL) cholesterol (-21%) and triglycerides (TG) (-9%). 11 In the study of atorvastatin versus placebo, there was an average reduction in LDL-cholesterol (40%), TC (32%), TG (12%) and APOB (34%) in the atorvastatin group compared to the placebo group (p <0.001). The increase 814