English (pdf)
Article in xml format
Curriculum ScienTI
Automatic translation
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
Permalink
INTRODUCTION
In the past years, it has become evident that atherosclerosis in adulthood begins precociously, since alterations in lipid metabolism during childhood and adolescence tend to persist in adult life and increase the risk for the development of atherosclerotic lesions1. The risk factors for cardiovascular diseases -in general-have increased in youth, with the main examples of dyslipidemias (DLP), hypertension and obesity, together with poor eating habits and a sedentary lifestyle2,3. Even though cardiovascular diseases are rare among children, DLP is associated with alterations in the elasticity of the carotid artery, average thickness of the intimate layer and the brachial-mediated dilatation from childhood to adulthood4. Risk factors related to metabolic imprinting such as gestational diabetes, maternal overweight and untreated DLP during pregnancy, parental overweight, low birth weight, preterm birth and short duration of exclusive breastfeeding also play a role in the development of cardiovascular diseases5.
The prevalence of dyslipidemia in children and adolescents varies across countries, with reports of 19.7% in Korean children5 and 22.9% in American adolescents6. Some studies report a prevalence variation from 2.9 to 33%, with an increase proportional with age7. Among biochemical markers of dyslipidemia, increased triglycerides is the most commonly reported (prevalence of around 14.2%), followed by increased total cholesterol levels (9.6%) in children6. In Brazil, the prevalence of dyslipidemia among children and adolescents varies from 10% to 23.5%, depending on the region of the country and type of DLP8.
Because it is a risk factor for the development of cardiovascular diseases that could be prevented, several public actions have been designed aiming at the early screening of DLP. In 2011, the National Heart, Lung, and Blood Institute (United States) published guidelines with recommendations for the screening, monitoring and treatment of children and adults under 21 years of age at risk of precocious cardiovascular events. Besides the screening process of children with positive family history, the group recommends a universal screening process -regardless of the presence of risk factors-for children over 9 years old9,10,11. According to an update from the Brazilian Directive on Dyslipidemias and Prevention of Atherosclerosis, carried out in 2017 by the Brazilian Society of Cardiology8, lipid profiles should be assessed in children starting at 2 years of age when there is a family history of early ischemic arterial disease; if parents have total cholesterol above 240 mg/dL (in addition to other diseases or risk factors for atherosclerosis or of diseases that occur with dyslipidemia); If parents use drugs that alter the lipid profile and/or in the presence of clinical manifestations of dyslipidemia. The updated material also establishes that the diagnosis of dyslipidemia is set through at least one altered measurement of markers (total cholesterol, LDL-c, HDL-c and/or triglycerides).
DLP can be both prevented and treated with the help of healthy eating habits, and current guidelines8 advise the reduction of LDL cholesterol through the decrease in intake of saturated and trans fatty acids, limited sugars and refined carbohydrates, as well as weight control for those with overweight and obesity, and adequate intake of fibers12. Sugar intake, especially the sugar added to processed foods, may be associated with adverse cardiovascular health factors in children, specifically elevated diastolic blood pressure and triglycerides13. A diet which is rich in antioxidant food sources (fruits, vegetables, oilseeds and fish) should also be standardized14; and both diets which are rich in polyunsaturated fats and low in overall fat have been proven efficient in the reduction of body mass index, blood pressure, increase of HDL-c, reduction of LDL-c and vasodilation15. Drug treatment such as statins, supplements, fish oil and bile acid sequestrants could be indicated depending on the type of dyslipidemia and sometimes, on the duration of unsuccessful clinical treatment8.
The prevalence of feeding difficulties (FD) -behaviors of food refusal and/or severe eating restrictions, together with a problematic relationship between child and caregiver-among children has been described to be as high as 50% worldwide. In general, children with FD do not present balanced eating habits. FD is usually categorized under the three main eating behaviors that concern parents: limited appetite, selective intake, and fear of feeding; with each category ranging from normal (misperceived) to severe (behavioral and organic) levels16. Although inadequate diets rarely hamper growth patterns and weight gain in these cases, selective children can present insufficient intake of specific food groups, which increases the concern among parents and clinical staff as to the possible adverse health effects. Children with FD usually avoid foods which are rich in protective factors against DLP (fruits and vegetables, whole foods and other sources of antioxidants), preferring snacks, milk and/or sugar sweetened beverages (SSB) in many cases16,17,18. The monotony of inadequate diet patterns may constitute a risk for development of dyslipidemia. Although this monotonous pattern could be much more relevant in picky eaters, it is also important to investigate the lipid profiles of children with other types of FD. Nevertheless, there is currently scarce evidence on dyslipidemia among children with FD. Hence, the objective of this study was to describe the lipid profile of children with different types of FD, as well as to explore associations between lipid profiles and type of FD, BMI, duration of exclusive breastfeeding and dietary intake characteristics.
MATERIAL AND METHODS
Study design and population
We conducted a cross-sectional study, carried out at the Centro de Dificuldades Alimentares (CDA), part of Instituto PENSI-Hospital Infantil Sabará-Fundação José Luiz Egydio Setúbal, located in São Paulo/Brazil. CDA is an outpatient service which follows children and adolescents between 0 and 18 years of age with complaints of FD (except for the diagnosis of eating disorders according to DSM-519). The population was assembled by convenience. All patients followed in the service until the end of data collection were invited to participate (n= 148, August, 2014 to November, 2017). Patients less than 24 months of age and 10 years or more were excluded to make the sample more homogeneous in age and to facilitate comparisons with biochemical data, as well as availability of full biochemical analysis of lipogram. Thus, our final sample size was 61 children. All patients presented written consent forms signed by their responsible caregiver after ethical approval of the project (CAAE 32939314.0.0000.5567; approval granted in 13/08/2014 under document n. 808.394).
Data collection
Data were collected from the interviews with patients’ mothers, as part of the service protocol, which consists of an appointment with a pediatrician, speech therapist and nutritionist together, followed by a multidisciplinary discussion. FD were diagnosed as “children with limited appetite”, “agitated children”, “phobic children”, “misperception of caregivers”, “picky eaters”, and “organic causes”, according to criteria suggested by Kerzner et al16. Families received feedback including a therapeutic plan designed by each specialty (such as diet plans and nutritional education activities, medications, stimulation and reestablishment of oral functions or even referral to other professionals). The guidelines used by each specialty were defined according to standards for age and are described by Maximino et al.20.
Data were collected at the initial appointment at CDA, and information was later extracted from records, of which the following variables were selected: age (months), gender, duration of exclusive breastfeeding (in months), biochemical levels of lipid profile (fasting total cholesterol, HDL-c, LDL-c, VLDL-c and triglycerides; classified according to official recommendations for age (Table 1), FD diagnosis conducted by the multidisciplinary team, BMI (classified according to WHO standards BMI/age (2007/2008)21, dietary intake of carbohydrates and total fat (both in % energetic value of the diet), and dietetic fibers (g/day classified according to guidelines from IOM)22; and daily intake of SSB (natural or artificial juices, refreshments or fizzy drinks, total daily intake in ml) and milk intake (ml/day). Dietary intake was obtained from 3-day food records analyzed by the service nutritionist according to the service protocol, previously described.
Table 1 Lipid profile reference. Brazilian Society of Cardiology8.
| Category | Low mg/dl | Acceptable mg/dl | Limit mg/dl | High mg/dl |
|---|---|---|---|---|
| Total Cholesterol (TC) | -- | < 170 | 170-199 | ≥ 200 |
| LDL | -- | < 110 | 110-129 | ≥ 130 |
| Non-HDL | -- | < 120 | 120-144 | ≥ 145 |
| Triglycerides: ages 0-9 | -- | < 75 | 75-99 | ≥ 100 |
| HDL | < 40 | > 45 | < 45 | -- |
Statistical analysis
After the evaluation of data consistency, statistical analysis was conducted using SPSS v21 software. Descriptive analysis was conducted through distribution of frequencies (%) for categorical variables and mean/median ± standard deviation (SD) and percentiles for continuous variables. After testing the normality and homogeneity of continuous variables comparisons were made using Student-t test, ANOVA-test (or equivalent non-parametric tests). A level of significance of 5% was considered.
RESULTS
General characteristics of sample are described in Table 2. The sample was composed of mostly male children (72.1%), with median of age around 48 months (Interquartile range= 35.5-68.0), and an average BMI of +0.04 ± 1.52. Around 22.1% of children presented excessive weight: 13.6% were overweight and 8.5% were obese. The most frequent FD diagnosis was picky eater (55.7%), followed by limited appetite (14.8%). Children were exclusively breastfed for average of 3.8 months.
Table 2 General characteristics of population. Instituto Pensi, 2017.
| Variables (n=61) | % (N) or mean values ± SD | |
|---|---|---|
| Gender | ||
| Female | 27.9% (17) | |
| Male | 72.1% (44) | |
| Age (months) | 53.9 ± 23.8 | |
| BMI/Age (z-score) | +0.04 ± 1.52 | |
| Duration of exclusive breastfeeding (months) | 3.8 ± 3.1 | |
| Lipid profile | ||
| Total Cholesterol mg/dl | 162.6 ± 32.3 | |
| HDL-c mg/dl | 56.2 ± 46.5 | |
| LDL-c mg/dl | 94.6 ± 25.2 | |
| Triglycerides mg/dl | 90.6 ± 55.2 | |
| Classification of dyslipidemia | 47.5% | |
| Feeding difficulties diagnosis | ||
| Agitated | 3% (3) | |
| Limited appetite | 14.8% (9) | |
| Phobia | 8.2% (5) | |
| Misinterpretation of caregivers | 11,5% (7) | |
| Organic | 4.9% (3) | |
| Picky eating | 55.7% (34) | |
| Other types | 4.9% (3) | |
Regarding the lipid profile of children, 47.5% of children presented at least one altered marker and were classified has having dyslipidemia. From these children, isolated low levels of HDL-c were identified in 27.6% (n= 8) of the sample, followed by isolated hypertriglyceridemia in 21.9% (n= 7), and isolated hypercholesterolemia in 17.2% (N= 5). One child (3.4%) presented alterations in all markers. The remaining children presented mixed hyperlipidemia, characterized by a combination of low HDL-c and hipertrigliceridemia (31%) or of hypercholesterolemia and hipertrigliceridemia (6.9%). Overall, the most common types of DLP were hypertriglyceridemia and low HDL levels (both around 30% prevalent) (Figure 1).
Figure 1 Classification of lipid profile according to recommendations by age. Instituto PENSI, 2017.
No relationship was found between presence of DLP and duration of exclusive breastfeeding (t= 0.081; IC95% −1.58-1.71; p= 0.93); average duration of breastfeeding was very similar in children with (3.75m ± 2.8) and without DLP diagnosis (3.8m ± 3.4). Lipid profiles were not associated with BMI (p> 0.40; Anova test) or with type of FD (p= 0.26; Kruskall Wallis test for LDL and triglycerides levels; p> 0.45, Anova test for HDL and total cholesterol).
The assessment of dietary intake showed that children with dyslipidemia tended to drink higher volumes of milk and SSB, ate slightly more carbohydrates and less fiber and fat than those with normal lipid profiles (Figures 2 and 3). However, values were very similar between groups, and T-Student tests showed no significant differences between these results (p> 0.12). Overall, dietary intake was within the recommended range (age-adjusted) for both carbohydrate and fat intake, and all children presented poor fiber intake patterns. According to the new recommendation for SSB intake in infancy (maximum 150 ml/day23), results showed that children classified as having dyslipidemia in this sample tended to drink higher volumes (around 90 ml/day) of SSB when compared to recommended values (t= 2.11; IC95% −2.8- 179.8; p= 0.044).
Figure 2 Macronutrient and fiber intakes according to presence of DLP. Instituto PENSI, 2017Footnote: Absolute intake values according to DLP classification (DLP vs normal values, respectively), after T-Student test: a) Fat intake (30.1% vs 31%, p= 0.49) b) Carbohydrate intake (57.7% vs 55.6%, p= 0.37) c) Fiber intake (7.7g vs 8.4g, p= 0.75).
DISCUSSION
Despite an overall healthy lipid profile found in the present population, high frequencies of inadequacies (values above or below recommendations) were identified -mostly alterations in HDL-c and triglycerides-and almost half of sample had dyslipidemia. Moreover, more than one third of children assessed presented mixed hyperlipidemia.
Prevalence of DLP in youth has been estimated as 23% worldwide6,24. In the northeast region of Brazil (Bahia), DLP in children between 7 and 14 years old was described as 25.5%22. In the Southern region of Brazil, much lower rates have been reported: 4.7% and 8.9% for hypercholesterolemia and hypertriglyceridemia, respectively, in teenagers between 10 and 12 years old25. Reuter et al.26 reported a prevalence of dyslipidemia in 1,243 children aged 7 to 17 years old -also in the South of Brazil-of 36.6% in girls and of 46.6% in boys and 30% overweight/obesity. Considering this context, the prevalence of dyslipidemia found in the present population is approximately double most of data described for children in the same age elsewhere; and similar to what has been described in Brazilian adults (59.74%)8. Hypertriglyceridemia and low HDL were the most frequent types of dyslipidemia in the present sample. Although the general DLP prevalence is different from the national average, individual markers are somewhat like other Brazilian studies performed, but with adolescents. In the ERICA study9, conducted with 38069 adolescents, results showed a similar prevalence of low HDL-c (46.8%), hypercholesterolemia (20.1%) and hypertriglyceridemia (7.8%). Faria et al.27, in a study with 1937 children between 2 and 19 years old, reported altered levels of LDL-c (44%) and triglycerides (56%) of the children, and 44% and 50% of adolescents, respectively.
When comparing the present results with the literature, two factors are noteworthy: 1) the prevalence of dyslipidemia in a population younger than that commonly described as having dyslipidemia; and 2) the high prevalence of DLP in a population with normal BMI. The fact that most studies on childhood DLP focus on populations at risk (such as overweight or obese children and adolescents, with family history of cardiovascular disease or dyslipidemia) impairs comparison and discussion of the present results. Moreover, the average age range of the present population (2 to 10 years old) is scarcely studied in this context, since mandatory screening is not applied at this life stage9. While searching for factors that could suggest reasons for these elevated DLP rates, the mean duration of exclusive breastfeeding was not associated with the changes described in the lipid profile, even though it has been described elsewhere that a short duration of exclusive breastfeeding increases the risk of childhood dyslipidemia28. The homogeneous results of this variable throughout the sample (there was no group with higher or lower duration of exclusive breastfeeding to compare) could be one of the reasons of the lack of association found. The sample analyzed presented a higher duration of exclusive breastfeeding compared to national data: in Brazil, the average duration of exclusive breastfeeding is low (around 2 months), and according to the latest national survey on breastfeeding29, only 36.6% of Brazilian children are exclusively breastfed for 6 months.
We found a relatively low prevalence of overweight compared to the prevalence of dyslipidemia. Similarly, in 414 healthy children 8.9 years of age in routine followup, Franca et al (2006)29 reported a prevalence of 30% of dyslipidemia and only 4% overweight. However, most studies on dyslipidemia during childhood address overweight populations. In a Brazilian study30 with a predominantly overweight population between 2 and 18 years of age (most of whom between 10 and 18 years old), dyslipidemia was reported in 85.3% of the population; 43.8% had only low HDL-c; 1.8% had high LDL-C and 0.5% high triglycerides. Distribution according to age showed that alterations in TC was more frequent in the range of six to nine years old, LDL-c and triglycerides between 10-18 years old, and HDL-c between two and five years old. Regarding other risk factors, level of physical activity was not assessed in this population, due to the absence of direct measurements to evaluate this profile; and the genetic risk of dyslipidemia was also not assessed in this analysis, constituting a possible bias.
Analysis of possible associations between dietary intake and lipid profiles have shown adequate intake of carbohydrates, fats and milk; and children presented inadequate intake of fibers and SSB. Children with dyslipidemia in this sample presented excessive SSB intake compared to recommendations. Accordingly, many studies report associations between inadequate dietary intake and altered lipid profiles. In a multi-ethnic sample of children, intake of SSB was positively associated with higher triglyceride concentrations among consumers31. In American children and adolescents, usual intake of added sugars was also significantly associated with several measures of lipid profiles32-33. The premise that higher SSB intakes could be one of the possible reasons of DLP in children with FD was first hypothesized by the CDA team due to a previous study carried out by Maximino et al (2016)20 with a population from the same research center as the present data. In that study, authors describe an excessive consumption of juices (carbohydrates with high glycemic index, which could impact biochemical levels), regardless of the type of FD, highlighting the need for nutritional counseling. Additionally, clinical observation also highlights frequent use of dairy compounds, often sugary and complemented with cereals and mucilage. In the present study, however, data on the composition of the dairy compounds used by this population was not available for comparison.
Unfortunately, no other studies with such comparisons (specifically children with FD) were found to discuss the present results. More controlled studies are necessary to determine whether there is risk for DLP related to dietary patterns in patients with FD or if the low intake of protective foods itself would determine such a risk. To our knowledge, this is the first report which describes a possible association between FD and dyslipidemia, possibly mediated by higher consumption of SSB. New studies are necessary to prove the need for screening or, minimally, to alert pediatricians about the risk for DLP in children with FD, regardless of age and BMI.
The study has limitations, such as the absence of a control group, superficial data on family history of dyslipidemia, as well as lack of information on children's physical activity. Moreover, the cross-sectional design does not allow for any causal inferences. On the other hand, the study contributes to the literature, given the scarceness of data regarding the prevalence of dyslipidemia in children with FD, and for the discussion about the cause/effect relationships that may exist between both.
CONCLUSIONS
The prevalence of dyslipidemia reported was higher than the average found for populations without diagnosis of feeding difficulties, even though the population had a normal BMI. The main types of dyslipidemia found were hypertriglyceridemia and low levels of HDL. The type of feeding difficulty and breastfeeding duration were not associated with dyslipidemia. There was excessive intake of SSB in children with dyslipidemia in the sample. More studies are needed to prove robust associations between FD and dyslipidemia.
