Childhood obesity in specialist care – searching for a healthy obese child

Abstract

Introduction One in three obese adults is classified as metabolically healthy, but there is less evidence in obese children. We studied the overall clinical presentation of Finnish obese children and the prevalence of cardiometabolic risk factors with child-specific cut-offs.

Material and methods This is a cross-sectional register-based study of 2–18-year-old children (n = 900) evaluated for obesity in three hospitals in 2005–2012. Clinical and metabolic data were related to sex, age, puberty, and obesity grade and analyzed using chi-square and non-parametric tests.

Results In 80% of cases at least one cardiovascular risk factor was present. Only 3% of subjects for whom complete metabolic data were available (n = 360) had no metabolic disorder. Systolic blood pressure was hypertensive in 50.2% and diastolic in 14.5% of the children. The youngest children had highest body mass index SD score. Obesity was more severe in boys than girls (p < 0.001). Hypertensive systolic blood pressure values (p = 0.012), prediabetes (p < 0.001), fatty liver (p < 0.001), and dyslipidemia (p = 0.025) were more prevalent in 15–18-year-old boys than girls.

Conclusion Most obese children in specialist care have cardiovascular risk factors; this indicates that earlier intervention is needed.

Key messages

• Most obese children evaluated in specialist care have one or more cardiovascular risk factors, and very few have no metabolic disturbances.

• In late adolescence these risk factors are more common in obese boys than girls.

• Primary care personnel are less likely to intervene in cases of obesity in young children than in adolescents, leading to a delay in treatment.

• Hypertensive blood pressure values require more attention in clinical work.

Keywords:

• Adolescent
• blood glucose
• blood pressure
• child
• dyslipidemias
• fatty liver
• hyperinsulinism
• metabolism
• obesity

Introduction

Childhood obesity has become a worldwide challenge in health care. The global prevalence of childhood obesity is high. In 2013, 23.8% of boys and 22.6% of girls in developed countries were overweight or obese; in developing countries the corresponding rates were 12.9% and 13.4%, respectively (1). In 2012, 16.9% of 2–19-year olds in the United States were obese (2). The global prevalence of childhood obesity is rising, and the rate may have reached a plateau only in a few countries in the last few years (3). Moreover, the prevalence of obesity in very young children and the proportion of seriously obese children are increasing (3).

The health concerns associated with childhood obesity are real (4), and the consequences of obesity are expensive (5). Childhood obesity causes physical symptoms and psychosocial disturbances and is associated with other illnesses such as asthma and attention deficit hyperactive disorder (ADHD) (6). Obese and overweight schoolchildren have significantly more and worse cardiovascular risk factors than their lean peers (4). Obesity and overweight in childhood are likely to carry over into adulthood (7), and they pose a threat to health, increasing morbidity and premature mortality especially in at-risk groups (8,9). The International Childhood Cardiovascular Cohort (i3C) Consortium and other groups have reported that childhood overweight and obesity, which track into adulthood, predispose to type 2 diabetes (T2DM), hypertension, dyslipidemia, and atherosclerosis (10–12). Atherosclerosis begins to develop in childhood (13–15). Furthermore, cardiovascular risk factors such as high blood pressure, low high-density lipoprotein cholesterol (HDL-C), high triglyceride (TG) concentrations, and impaired glucose metabolism have a tendency to cluster together with obesity even in childhood, and this cluster is associated with early exposure to diabetes and adult metabolic syndrome (16,17). This accelerates the progression of atherosclerosis to cardiovascular disease (CVD) (10) and may also induce liver enzyme changes indicative of fatty liver even in childhood (18,19). These comorbidities of obesity, once expected to arise in middle age, may now be present in childhood (3).

The only way to halt or slow the progression of childhood obesity into disease states is to prevent childhood overweight and ensure that it does not persist into adulthood. To do this effectively we need a thorough and up-to-date knowledge of the characteristics of obese children.

In this clinical study we examined all obese children admitted to specialist care from a well-defined population in Eastern Finland over the past 8 years. Our main hypothesis was that most obese children would meet criteria for metabolic or cardiovascular disturbances on the first clinic visit if relevant clinical measurements and childhood specific cut-offs were used.

Materials and methods

Sample and design

Our data consisted of the patient records of 900 children aged 2–18 years (girls n = 420, 47%; boys n = 480, 53%) who were examined and treated for obesity between 2005 and 2012 in the pediatric units of Kuopio University Hospital (n = 363, 40%), Mikkeli Central Hospital (n = 247, 27%), or North Karelia Central Hospital (n = 290, 33%). These three hospitals are responsible for pediatric secondary and tertiary care in their hospital districts. The units use similar obesity treatment programs which are based on the Finnish National Current Care Guidelines of Childhood Obesity (20). These guidelines state that services related to prevention and treatment of overweight and obesity should be provided at the primary care level in free child health clinics, as part of school and student health care and in health centers. Cases of severe childhood obesity and cases where there are signs of other diseases are referred to specialist units. This study focused on the status of patients on the first clinic visit to these pediatric units. Anthropometric, clinical, and metabolic data were analyzed according to sex, age, and pubertal and obesity status.

The study protocol was approved by the Research Ethics Committee of the Hospital District of Northern Savo (Kuopio, Finland). Permission for the use of patient registers was obtained from the National Institute for Health and Welfare (THL) and from participating hospitals.

Methods

Height was measured by an experienced nurse using a wall-mounted Harpenden stadiometer with an accuracy of 0.1 cm. The mean of the closest two out three measurements of barefoot height was used. Weight was measured in light underwear using a calibrated electronic scale with an accuracy of 0.1 kg as the mean of two measurements. Body height standard deviation score (SDS) and body mass index (BMI)-SDS were calculated according to recent Finnish gender- and age-specific population standards (21). BMI-SDS was used to assign children to four obesity categories: overweight (9%), obesity (47%), severe obesity (31%), and morbid obesity (13%). The BMI-SDS cut-off points for girls were 1.16, 2.11, 2.76, and 3.24 for overweight, obesity, severe obesity, and morbid obesity, respectively. The corresponding values for boys were 0.78, 1.70, 2.36, and 2.85. The values corresponded to BMIs of 25, 30, 35, and 40 kg/m², respectively, at the age of 18 years and thus represented childhood analogues of the commonly used adult obesity thresholds.

The children were classified into four age groups: 2–6.9 years (13%), 7–9.9 years (20%), 10–14.9 years (53%), and 15–18 years (14%). Children over the age of 10 years were also considered adolescents (66%) (22). Pubertal status was assessed by physicians using the Tanner staging method (23,24). For the purpose of this study children were classified as prepubertal (M/G1, 46%), early/midpubertal (M/G2–3, 22%), late pubertal (M/G4–5, 23%), or postpubertal (adult height achieved, 9%). In figures, puberty status was classified simply as prepubertal (46%) and pubertal (64%). Boys with testicular volume >3 mL were assigned to G2 (25). Data on pubertal stage were recorded in 846 cases (94%).

Blood pressure (BP) was measured two or three times using Criticon Dinamap Vital Signs monitor 1846SX with a suitable Duracuff (8–13 or 38–50 cm) or in a few cases with a standard sphygmomanometer from the right arm in the supine position after a 15-minute rest whilst seated, and the lowest recording was registered. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were classified as normal, high normal, and hypertensive (stages 1 and 2) on the basis of height-, age-, and gender-specific percentile cut-offs (<90th, 90th but less than 95th, and ≥95th) as recommended by the Fourth Report from the National High Blood Pressure Education Program (NHBPEP) Working Group on Children and Adolescents (26). For clinical purposes, the hypertensive category was divided into two subgroups: stage 1 hypertension (BP between the 95th percentile and the 99th percentile plus 5 mmHg) and stage 2 (BP above the 99th percentile plus 5 mmHg). BP values were recorded on the first clinic visit in 733 cases (81%).

Clinical diagnoses were recorded using the International Classification of Diseases ICD-10, 2010 version. Medication was recorded by generic names.

Laboratory analyses

All analyses done in the 6-month periods before and after the first clinic visit were included in this study. All samples were taken after a 12-h overnight fast. From 2008 all laboratory analyses were performed in one regional laboratory, the Eastern Finland Laboratory Centre (ISLAB); before that each hospital carried out the analyses in their own laboratories. We assessed possible differences between analyses carried out before and after the move to central laboratory analysis using separate general linear models for boys and girls, controlling for age, pubertal status, and obesity status.

Serum insulin was analyzed using an electrochemiluminescence immunoassay (Roche Diagnostics GmbH, Mannheim, Germany). Total plasma cholesterol (P-TC) and P-TG were analyzed with a colorimetric enzymatic assay, and plasma concentrations of low-density lipoprotein and high-density lipoprotein (P-LDL-C and P-HDL-C respectively) were determined with a homogeneous colorimetric enzymatic assay (both Roche Diagnostics GmbH, Mannheim, Germany). Plasma glucose (P-Gluc) was analyzed by the hexokinase method and glycosylated hemoglobin (HbA1c) with a turbidimetric inhibition immunoassay (both Roche Diagnostics GmbH, Mannheim, Germany). The IFCC kinetic method was used to quantify plasma alanine aminotransferase (P-ALT) (Roche Diagnostics GmbH, Mannheim, Germany).

Fasting (f) P-Gluc concentrations (n = 641) were categorized as normal (<5.6 mmol/L), impaired fasting glucose (IFG) (5.6–6.9 mmol/L), or diabetic (≥7 mmol/L). P-Gluc values (n = 190) following a 2-hour oral glucose tolerance test (OGTT; load of 1.75 g/kg anhydrous glucose up to a maximum of 75 g, dissolved in water) were classified as normal (<7.8 mmol/L), impaired glucose tolerance (IGT) (7.8–11.0 mmol/L), or diabetic (≥11.1 mmol/L). HbA1c% (n = 253) was classified as normal (<5.8%), prediabetic (5.8%–6.4%), or diabetic (≥6.5%). Diabetes mellitus (DM) was recognized in cases where there were symptoms of DM and where casual P-Gluc concentration ≥11.1 mmol/L or fP-Gluc ≥7.0 mmol/L or post-OGTT glucose ≥11.1 mmol/L or HbA1c% ≥6.5%. Prediabetes was recognized in cases where there was IFG or IGT or HbA1c% in the range 5.8%–6.4%. These definitions are in accordance with the International Society for Pediatric and Adolescent Diabetes (ISPAD) Clinical Practice Consensus Guidelines 2014 Compendium (27). Fasting serum insulin concentration (n = 480) was categorized as normal or hyperinsulinemic using pubertal stage-specific cut-offs for hyperinsulinemia, prepubertal: >15 mU/L, pubertal: >30 mU/L, and postpubertal: >20 mU/L (28).

Lipid concentrations were classified on the basis of the Summary Report (2011) of the Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents as acceptable, borderline, or high (or low for HDL-C) (26). The cut-off values were as follows: fP-TC (n = 644), <4.40, 4.40–5.17, and ≥5.18 mmol/L; fP-LDL-C (n = 617), <2.84, 2.84–3.35, and ≥3.36 mmol/L; fP-TG (n = 635), for children under 10 years old, <0.84, 0.84–1.12, and ≥1.13 mmol/L; for adolescents, <1.02, 1.02–1.46, and ≥1.47 mmol/L. The cut-offs of fP-HDL-C (n = 630) sub-groups were acceptable, borderline, and low, respectively, >1.17, 1.04–1.17, and <1.04 mmol/L. Fasting P-ALT data (n = 563) were categorized as elevated at ≥40 IU/L (29).

Children with normal fasting plasma glucose, 2-h OGTT glucose, B-HbA1c, serum insulin, plasma ALT, and normal plasma lipids were assigned metabolically healthy obese (MHO). CVD risk factors comprised hypertensive SBP or DBP, high TC, high LDL-C, high TG, low HDL-C, and diabetes and prediabetes.

Statistics

Statistical analyses were performed using the SPSS statistical analyses software, Version 19 (IBM Corp., Armonk, NY, USA). Because the majority of continuous data variables were not normally distributed, descriptive data are reported (median [Md] and interquartile range [IQR]), and Pearson’s chi-square test was used to explore the relationships between categorical variables. Mann–Whitney U and Kruskall–Wallis tests were used to analyze continuous variables within gender, age, pubertal status, and obesity status groups. A P value <0.05 was considered statistically significant.

Results

Most (55%) children were referred from public primary health care, and 35% were referred from other pediatric disciplines. Some referrals (6%) came from other specialties (mostly psychiatry), and 2% were from private outpatient clinics. Overweight or obesity was the main reason for the referral in 71% of cases. In a quarter of these cases, one or more metabolic disturbances had been diagnosed before the first clinic visit. In 29% of cases the excess weight was recorded as being an important contributor to another illness. The most frequent symptoms associated with excess weight were musculoskeletal disorders (32%), gastrointestinal symptoms (30%), and headache (26%). Nine percent of the patients complained of teasing. Menstrual irregularity was present in 31% and amenorrhea in 7% of 130 sexually mature girls. After obesity, the two most frequent diagnoses were asthma (in 22% of children) and mental, behavioral, or developmental disorders (in 25% of children) (Table I). ADHD was present in 5% of cases, and it was more frequent in boys (chi-square, p < 0.001), whereas a depressive disorder was present in 4% of cases and was more frequent in girls (chi-square, p = 0.009). The most common long-term medication was continuous or periodic asthma medication, which was used by 145 children (16%). Drugs affecting the central nervous system were used by 75 patients (8.3%). A total of 335 children (37%) used at least one pharmacotherapy, and 64 (7%) were taking three to six different daily medications on the first clinic visit.

Table I. Clinical characteristics of the study objects.

	Girls^a	Boys^a	p value
All	420 (46.7)	480 (53.3)
Age, years	11.4 (8.4, 13.9)	12.0 (9.3, 13.9)	0.073^b
Age  2–6.9	68 (16.2)	52 (10.8)	0.033^c
7–9.9	92 (21.9)	90 (18.7)
10–14.9	204 (48.6)	272 (56.7)
15–18	56 (13.3)	66 (13.8)
BMI-SDS^d	2.52 (2.28, 2.84)	2.42 (2.13, 2.73)		<0.001^b
Obesity stage, n^d	420	480
Overweight	62 (14.8)	21 (4.4)	<0.001^c
Obesity	231 (55.0)	190 (39.6)
Severe obesity	92 (21.9)	186 (38.7)
Morbid obesity	35 (8.3)	83 (17.3)
Height-SDS^d	0.6 (–0.1, 1.3)	0.4 (–0.3, 1.3)		0.116^b
Pubertal stage, n^e	401	445
Prepubertal	144 (35.9)	244 (54.8)	<0.001^c
Early/midpubertal	84 (20.9)	103 (23.1)
Late pubertal	112 (27.9)	81 (18.2)
Postpubertal	61 (15.3)	17 (3.9)
SBP, n^f	341	392
Normal	130 (38.1)	151 (38.5)	0.180^c
High normal	48 (14.1)	36 (9.2)
Stage 1	119 (34.9)	144 (36.7)
Stage 2	44 (12.9)	61 (15.6)
DBP, n^f	341	392
Normal	231 (67.7)	288 (73.7)	0.319^c
High normal	57 (16.7)	50 (12.8)
Stage 1	48 (14.1)	49 (12.5)
Stage 2	5 (1.5)	4 (1.0)
Asthma	75 (17.9)	120 (25.0)		0.009^c
Hypothyroidism	20 (4.5)	10 (2.1)	0.026^c
Familial hypercholesterolemia	4 (0.9)	6 (1.3)	0.671^c
Mental, behavioral, and  developmental disorders	99 (23.5)	125 (26.0)	0.393^c
ADHD	5 (1.2)	41 (8.5)	<0.001^c
Developmental disorders	30 (7.1)	64 (13.3)	0.002^c
Depressive disorders	25 (6.0)	12 (2.5)	0.009^c
Intellectual disabilities	13 (3.1)	15 (3.1)	0.980^c
Epilepsy	11 (2.6)	7 (1.5)	0.215^c

a Results for girls and boys are represented as medians and interquartile ranges (IQR 25, 75) for continuous variables and n(%) for categorical variables and prevalences.

b Mann–Whitney U test.

c Pearson’s chi-square test. p < 0.05 is statistically significant.

d Based on the Finnish growth references (21).

e Staged on the basis of the Tanner method (23,24).

f Categorized on the basis of cut-offs of the NHBPEP Working Group on Children and Adolescents (26); Stage 1 and Stage 2 are hypertensive.

ADHD = attention deficit hyperactivity disorder; BMI-SDS = body mass index standard deviation score; DBP = diastolic blood pressure; NHBPEP = The National High Blood Pressure Education Program; SBP = systolic blood pressure.

Age, pubertal stage, and anthropometrics on the first clinic visit

The median age on the first clinic visit was 11.7 years (IQR 25, 75: 8.9, 13.9 years) (Table I). The girls were younger than the boys. Most obese adolescents were boys (57%; chi-square, p = 0.007). The age of the morbidly obese children (Md age: girls = 5.5 years; boys = 7.0 years) was lower than that of less obese children (Table II). The girls were more mature than boys (Tables I and II), and girls at Tanner stage M2 (Md age = 10.4; 9.7, 11.8 years) were younger than boys at Tanner stage G2 (Md age 12.7; 12.0, 13.4 years). The median age of menarche recorded on the first clinic visit for 130 girls was 12.0 (11, 12.5) years, and 18 girls had reached menarche before the age of 10.5 years. Two girls were treated for precocious puberty. All girls over the age of 13.5 years were pubertal. None of the boys had early puberty (G2 <9 years), but 15 boys showed delayed pubertal development on the first clinic visit (had not reached G2 by 13.5 years).

Table II. Anthropometric, clinical, and metabolic measurements in girls and boys by age.

	Girls (n = 420)				Boys (n = 480)
	2.0–6.9	7–9.9	10–14.9	15–18	2.0–6.9	7–9.9	10–14.9	15–18
Age, years	n (%)^a	n (%)^a	n (%) ^a	n (%)^a	n (%)^a	n (%)^a	n (%)^a	n (%)^a
n, % within sex	68 (16.2)	92 (21.9)	204 (48.6)	56 (13.3)	52 (10.8)	90 (18.7)	272 (56.7)	66 (13.8)
Age, years	5.6 (4.7, 6.4)	8.8 (7.9, 9.4)	12.6 (11.4, 13.9)	15.5 (15.3, 16.0)	5.3 (4.5, 6.1)	8.6 (7.6, 9.3)	12.5 (11.5, 13.5)	15.6 (15.2, 15.9)
BMI-SDS ^b	3.10 (2.61, 3.36)	2.48 (2.29, 2.70)	2.45 (2.20, 2.70)	2.53 (2.30, 2.88)	3.32 (2.91, 3.76)	2.44 (2.25, 2.66)	2.31 (1.98, 2.55)	2.50 (2.19, 2.70)
Obesity stage, n^b	68	92	204	56	52	90	272	66
Overweight	3 (4.4)	14 (15.2)	39 (19.1)	6 (10.7)	0 (0.0)	1 (1.1)	17 (6.3)	3 (4.5)
Obesity	16 (23.5)	60 (65.2)	126 (61.8)	29 (51.8)	1 (1.9)	34 (37.8)	132 (48.5)	23 (34.9)
Severe obesity	24 (35.3)	17 (18.5)	33 (16.2)	18 (32.1)	10 (19.2)	43 (47.8)	102 (37.5)	31 (47.0)
Morbid obesity	25 (36.8)	1 (1.1)	6 (2.9)	3 (5.4)	41 (78.9)	12 (13.3)	21 (7.7)	9 (13.6)
Weight, kg	33.4 (28.7, 38.2)	51.1 (44.0, 56.8)	80.9 (67.5, 94.3)	94.8 (84.2, 101.5)	34.6 (31.1, 40.4)	51.7 (45.0, 58.5)	75.6 (65.1, 91.6)	107.8 (92.4, 121.0)
Height, cm	117.6 (113.4, 124.2)	138.5 (131.9, 143.3)	159.3 (151.1, 166.0)	165.2 (160.4, 169.1)	118.4 (112.0, 123.9)	138.2 (130.5, 143.9)	158.0 (150.5, 167.1)	175.0 (169.4, 181.7)
Height-SDS^b	1.0 (0.6, 1.7)	0.8 (0.1, 1.6)	0.5 (–0.1, 1.3)	0.0 (–0.8, 0.6)	1.1 (0.1, 1.8)	0.8 (–0.1, 1.6)	0.3 (–0.4, 1.2)	0.0 (–0.8, 0.9)
Pubertal stage, n^c	68	85	192	56	52	87	241	65
Prepubertal	67 (98.5)	63 (74.1)	15 (7.8)	0 (0.0)	52 (100.0)	87 (100.0)	103 (42.7)	2 (3.1)
Early/midpubertal	1 (1.5)	21 (24.7)	60 (31.3)	1 (1.8)	0 (0.0)	0 (0.0)	92 (38.2)	11 (16.9)
Late pubertal	0 (0.0)	1 (1.2)	88 (45.8)	23 (41.1)	0 (0.0)	0 (0.0)	41 (17.0)	40 (61.5)
Postpubertal	0 (0.0)	0 (0.0)	29 (15.1)	32 (57.1)	0 (0.0)	0 (0.0)	5 (2.1)	12 (18.5)
Blood pressure, n	53	71	170	47	29	78	222	63
SBP, mmHg^d	110 (105, 117)	113 (107, 122)	122 (113, 131)	124 (114, 134)	115 (104, 121)	114 (107, 123)	122 (114, 131)	136 (126, 148)
Normal	16 (30.2)	27 (38.0)	67 (39.4)	20 (42.6)	12 (41.5)	36 (46.2)	87 (39.2)	16 (25.4)
High normal	11 (20.8)	10 (14.1)	21 (12.4)	6 (12.8)	3 (10.3)	4 (5.1)	24 (10.8)	5 (7.9)
Stage 1	21 (39.6)	24 (33.8)	58 (34.1)	16 (34.0)	11 (37.9)	31 (39.7)	84 (37.8)	18 (28.6)
Stage 2	5 (9.4)	10 (14.1)	24 (14.1)	5 (10.6)	3 (10.3)	7 (9.0)	27 (12.2)	24 (38.1)
DBP, mmHg^d	66 (62, 71)	69 (61, 75)	70 (64, 77)	73 (65, 82)	66 (62, 70)	67 (63, 72)	71 (63, 76)	74 (67, 82)
Normal	33 (63.5)	47 (66.2)	119 (70.0)	32 (68.1)	18 (62.1)	63 (80.8)	171 (77.0)	37 (58.7)
High normal	9 (17.3)	15 (21.1)	27 (15.9)	6 (12.8)	7 (24.1)	6 (7.7)	27 (12.2)	10 (15.9)
Stage 1	10 (19.2)	7 (9.9)	23 (13.5)	7 (14.9)	3 (10.3)	9 (11.5)	22 (9.9)	15 (23.8)
Stage 2	0	2 (2.8)	1 (0.6)	2 (4.2)	1 (3.5)	0	2 (0.9)	1 (1.6)
fP-ALT, n	39	52	121	38	29	54	181	49
IU/L	21 (17, 28)	24 (18, 32)	24 (18, 35)	26 (15, 34)	24 (20, 39)	26 (19, 35)	27 (20, 44)	45 (30, 76)
<40	34 (87.2)	45 (86.5)	97 (80.2)	32 (84.2)	22 (75.9)	47 (87.0)	128 (70.7)	21 (42.9)
≥40	5 (12.8)	7 (13.5)	24 (19.8)	6 (15.8)	7 (24.1)	7 (13.0)	53 (29.3)	28 (57.1)
fP-Gluc, n^e	40	61	156	46	29	60	190	59
mmol/L	5.1 (4.9, 5.3)	5.2 (5.0, 5.4)	5.4 (5.2, 5.8)	5.2 (5.0, 5.5)	5.2 (5.0, 5.6)	5.3 (5.0, 5.6)	5.5 (5.2, 5.7)	5.6 (5.4, 6.1)
<5.6	35 (87.5)	50 (82.0)	95 (60.9)	36 (78.2)	21 (72.4)	44 (74.6)	114 (69.0)	24 (40.7)
5.6–6.9 (IFG)	5 (12.5)	11 (18.0)	58 (37.2)	9 (19.6)	7 (24.2)	15 (25.4)	75 (39.5)	33 (55.9)
≥7.0	0	0	3 (1.9)	1 (2.2)	1 (3.4)	0	1 (0.5)	2 (3.4)
2-h-OGTT-Gluc, n^e	1	8	58	20	0	6	70	27
mmol/L	–	5.5 (4.7, 6.8)	6.5 (5.7, 7.3)	5.7 (5.3, 6.3)	–	6.4 (5.3, 7.1)	6.2 (5.5, 7.1)	6.7 (5.7, 7.5)
<7.8	1	7 (87.5)	46 (79.3)	19 (95.0)	0	6	61 (87.2)	21 (77.8)
7.8–11.0 (IGT)	0	1 (12.5)	10 (17.3)	1 (5.0)	0	0	8 (11.4)	6 (22.2)
≥11.1	0	0	2 (3.4)	0	0	0	1 (1.4)	0
B-HbA1c, n^e	17	20	62	18	10	19	80	27
mmol/mol	32 (30, 36)	35 (32, 38)	35 (33, 38)	35 (33, 38)	37 (35, 39)	34 (33, 35)	35 (34, 38)	35 (34, 38)
%	5.1 (4.9, 5.4)	5.4 (5.1, 5.6)	5.4 (5.2, 5.6)	5.4 (5.2, 5.6)	5.5 (5.4, 5.7)	5.3 (5.2, 5.4)	5.4 (5.3, 5.0)	5.4 (5.3, 5.6)
≤5.7	17 (100.0)	19 (95.0)	55 (88.7)	17 (94.4)	8 (80.0)	17 (89.5)	70 (87.5)	22 (81.5)
5.8–6.4	0	1 (5.0)	4 (6.5)	0	1 (10.0)	2 (10.5)	8 (10.0)	5 (18.5)
≥6.5	0	0	3 (4.8)	1 (5.6)	1 (10.0)	0	2 (2.5)	0
fS-Insulin, n^f	26	44	122	39	18	43	150	45
mU/L	12.0 (8.9, 15.8)	16.4 (12.5, 22.3)	27.3 (19.4, 32.7)	26.8 (17.4, 36.7)	10.5 (7.4, 12.7)	13.7 (7.5, 17.0)	19.8 (11.4, 27.7)	27.3 (19.1, 39.4)
Normal	19 (73.1)	23 (54.8)	62 (51.7)	19 (48.7)	16 (88.9)	22 (51.2)	91 (61.9)	20 (44.4)
High	7 (26.9)	19 (45.2)	58 (48.3)	20 (51.3)	2 (11.1)	21 (48.8)	56 (38.1)	25 (55.6)
fP-TC, n^g	43	65	152	44	32	63	193	52
mmol/L	4.70 (4.10, 5.20)	4.80 (4.10, 5.30)	4.40 (3.80, 5.03)	4.30 (3.60, 4.90)	4.20 (3.63, 4.80)	4.30 (4.00, 5.00)	4.60 (3.80, 5.20)	4.45 (3.60, 5.18)
Acceptable	15 (34.9)	24 (36.9)	71 (46.7)	23 (52.3)	19 (59.4)	32 (50.8)	77 (39.9)	24 (46.2)
Borderline	16 (37.2)	19 (29.2)	50 (32.9)	14 (31.8)	7 (21.9)	22 (34.9)	64 (33.2)	15 (28.8)
High	12 (27.9)	22 (33.9)	31 (20.4)	7 (15.9)	6 (18.7)	9 (14.3)	52 (26.9)	13 (25.0)
fP-LDL-C, n^g	37	64	146	45	28	59	186	52
mmol/L	3.10 (2.50, 3.50)	3.00 (2.50, 3.50)	2.70 (2.28, 3.20)	2.50 (2.10, 3.05)	2.35 (1.91, 2.70)	2.70 (2.30, 3.10)	2.90 (2.33, 3.40)	2.90 (2.22, 3.40)
Acceptable	16 (43.2)	29 (45.3)	88 (60.3)	31 (68.9)	22 (78.6)	34 (57.6)	91 (48.9)	25 (48.1)
Borderline	8 (21.6)	12 (18.8)	28 (19.2)	5 (11.1)	1 (3.5)	16 (27.1)	46 (24.7)	13 (25.0)
High	13 (35.2)	23 (35.9)	30 (20.5)	9 (20.0)	5 (17.9)	9 (15.3)	49 (26.4)	14 (26.9)
fP-HDL-C, n^g	39	64	149	44	31	61	189	53
mmol/L	1.17 (1.01, 1.36)	1.27 (1.00, 1.47)	1.12 (0.96, 1.30)	1.10 (0.94, 1.36)	1.38 (1.04, 1.59)	1.24 (1.04, 1.47)	1.13 (0.97, 1.35)	0.94 (0.86, 1.14)
Acceptable	20 (51.3)	37 (57.8)	63 (42.3)	19 (43.2)	20 (64.5)	34 (55.7)	82 (43.4)	8 (15.1)
Borderline	8 (20.5)	8 (12.5)	38 (25.5)	10 (22.7)	4 (12.9)	12 (19.7)	45 (23.8)	10 (18.9)
Low	11 (28.2)	19 (29.7)	48 (32.2)	15 (34.1)	7 (22.6)	15 (24.6)	62 (32.8)	35 (66.0)
fP-TG, n^g	40	65	149	44	31	62	191	52
mmol/L	0.88 (0.69, 1.26)	1.08 (0.81, 1.64)	1.31 (1.00, 1.87)	1.19 (0.91, 1.74)	0.75 (0.50, 1.26)	0.99 (0.70, 1.48)	1.16 (0.82, 1.70)	1.27 (0.94, 2.03)
Acceptable	17 (42.5)	18 (27.7)	38 (25.5)	16 (36.4)	18 (58.1)	25 (40.3)	79 (41.4)	17 (32.7)
Borderline	11 (27.5)	17 (26.1)	47 (31.5)	13 (29.5)	5 (16.1)	11 (17.8)	45 (23.5)	12 (23.1)
High	12 (30.0)	30 (46.2)	64 (43.0)	15 (34.1)	8 (25.8)	26 (41.9)	67 (35.1)	23 (44.2)
MHO, n^h	21	37	84	27	12	35	111	33
Yes	1 (4.8)	0 (0)	1 (1.2)	1 (3.7)	1 (8.3)	3 (8.6)	5 (4.5)	0 (0)
No	20 (95.2)	37 (100)	83 (98.8)	26 (96.3)	11 (91.7)	32 (91.4)	106 (95.5)	33 (100)

a Medians and interquartile ranges (IQR 25, 75) for continuous and n (%) for categorical variables are reported unless otherwise stated.

b According to the Finnish growth references (21).

c Pubertal stage categorized on the basis of the Tanner method (23,24).

d Classified on the basis of the NHBPEP Working Group on Children and Adolescents (26).

e fP-Gluc, IFG, 2-h-OGTT-Gluc, IGT, and B-HbA1c classified on the basis of ISPAD (27).

f High: >15 mU/L in prepuberty, >30 mU/L in puberty, and >20 mU/L in postpuberty (28).

g Classified on the basis of NCEP (26).

h Defined as normal fasting plasma glucose (27), 2-h-OGTT-Gluc (27), B-HbA1c (27), serum insulin (28), plasma ALT (29), plasma TC, LDL-C, HDL-C, and TG (26).

2-h-OGTT-Gluc = 2-hour oral glucose tolerance test; B-HbA1c = glycosylated hemoglobin; BMI-SDS = body mass index standard deviation score; DBP = diastolic blood pressure; fP-ALT = fasting plasma alanine aminotransferase (29); fP-Gluc = fasting plasma glucose; fS-Insulin = fasting serum insulin; IFG = impaired fasting glucose; IGT = impaired glucose tolerance; ISPAD = International Society for Pediatric and Adolescent Diabetes; MHO = metabolically healthy obese; NCEP = National Cholesterol Education Program Expert Panel; NHBPEP = National High Blood Pressure Education Program; SBP = systolic blood pressure.

Of all children, 9% were overweight, 47% obese, 31% severely obese, and 13% morbidly obese. The youngest children had highest BMI-SDS (Kruskall–Wallis, p < 0.001). Of boys aged between 2 and 6.9 years, 98% were severely or morbidly obese, and of girls, 72% (Table II). The boys had more severe obesity than the girls (chi-square, p < 0.001) (Table I). Striae were present in half of the children, and they were more common in older age groups (chi-square, p < 0.001) and in those with more severe obesity (chi-square, p < 0.001).

There was no sex difference in statural growth (height-SDS) (Table I), but younger children had higher height-SDS scores than older children (Table II).

Blood pressure

SBP was higher in boys than girls (Mann–Whitney U, p = 0.002), but there was no sex difference in DBP (Mann–Whitney U, p = 0.818). SBP was in the hypertensive range in 50.2% and DBP in 14.4% of children. The highest proportions of hypertensive SBP and DBP values were found in boys aged 15–18 years (66.7% and 25.4%, respectively) (Table II). The prevalence of stage 2 hypertensive SBP and DBP values was higher in severely obese pubertal girls and boys than in their less obese peers (Figure 1).

Figure 1. The relationship between severity of obesity and blood pressure, glucose metabolism, and plasma alanine aminotransferase (ALT) levels in prepubertal and pubertal girls and boys. A: Systolic blood pressure (SBP), hypertensive levels Stage 1 and Stage 2 (≥95th percentile) (26). B: Diastolic blood pressure (DBP), hypertensive levels Stage 1 and Stage 2 (≥95th percentile) (26). C: Impaired fasting glucose (IFG), criterion was plasma glucose (Gluc) 5.6–6.9 mmol/L; diabetic, criterion was Gluc ≥7.0 mmol/L (27). D: Hyperinsulinism, criterion was serum insulin >15 mU/L in prepuberty, >30 mU/L in puberty, and >20 mU/L in postpuberty (28). E: Fatty liver, criterion was plasma ALT ≥40 IU/L (29). OW/OB = overweight and obesity; SOB/MOB = severe obesity and morbid obesity (21). Pubertal = Tanner ≥M2/G2 (breast bud/testicular volume >3 mL) (23,24).

Metabolic profile

Boys had significantly higher fasting glucose levels than girls, and this difference increased with age and sexual maturity (Tables II, III, and IV). IFG was present in 33.2% of cases, and half (50.0%) of the severely and morbidly obese pubertal boys had IFG (Figure 1). Although IFG was more prevalent in boys than girls, there was no sex difference in response to the OGTT in terms of plasma glucose concentrations or HbA1c values (Table III). On the basis of the OGTT, IGT was present in 13.7% of cases and diabetes in 1.6%. HbA1c values were prediabetic in 8.3% and diabetic in 2.8% of cases. Prediabetes was present in 34.7%—in 28.4% of girls (89 cases) and in 40.2% of boys (140 cases) (chi-square, p = 0.001). T2DM was recognized in six cases (3 boys).

Table III. Laboratory values in girls and boys.

	Girls (n = 420)	Boys (n = 480)	p value
	n (%)^a	n (%)^a
fP-ALT, n	250	313
IU/L	24 (17, 33)	28 (21, 45)	<0.001^b
<40	208 (83.2)	218 (69.6)	<0.001^c
≥40	42 (16.8)	95 (30.4)
fP-Gluc, n^d	303	338
mmol/L	5.3 (5.1, 5.6)	5.4 (5.2, 5.8)	<0.001^b
<5.6	216 (71.3)	204 (60.3)	0.012^c
5.6–6.9 (IFG)	83 (27.4)	130 (38.5)
≥7.0	4 (1.3)	4 (1.2)
2-h-OGTT-Gluc, n^d	87	103
mmol/L	6.1 (5.4, 7.3)	6.4 (5.5, 7.2)	0.392^b
<7.8	73 (83.9)	88 (85.4)	0.763^c
7.8–11.0 (IGT)	12 (13.8)	14 (13.6)
≥11.1	2 (2.3)	1 (1.0)
B-HbA1c, n^d	117	136
mmol/mol	35 (32, 38)	35 (34, 38)		0.103^b
%	5.4 (5.1, 5.6)	5.4 (5.3, 5.6)	0.097^b
≤5.7	108 (92.3)	117 (86.0)	0.088^c
5.8–6.4	5 (4.3)	16 (11.8)
≥6.5	4 (3.4)	3 (2.2)
fS-Insulin, n^e	231	256
mU/L	22.9 (15.4, 30.7)	17.8 (11.4, 27.2)	<0.001^b
Normal	123 (54.2)	149 (58.9)	0.299^c
High	104 (45.8)	104 (41.1)
fP-TC, n^f	304	340
mmol/L	4.50 (3.90, 5.10)	4.50 (3.80, 5.10)	0.655^b
Acceptable	133 (43.7)	152 (44.7)	0.967^c
Borderline	99 (32.6)	108 (31.8)
High	72 (23.7)	80 (23.5)
fP-LDL-C, n^f	292	325
mmol/L	2.79 (2.30, 3.40)	2.80 (2.30, 3.30)	0.807^b
Acceptable	164 (56.2)	172 (52.9)	0.278^c
Borderline	53 (18.1)	76 (23.4)
High	75 (25.7)	77 (23.7)
fP-HDL-C, n^f	296	334
mmol/L	1.14 (0.98, 1.34)	1.13 (0.97, 1.38)	0.660^b
Acceptable	139 (47.0)	144 (43.1)	0.508^c
Borderline	64 (21.6)	71 (21.3)
Low	93 (31.4)	119 (35.6)
fP-TG, n^f	298	336
mmol/L	1.20 (0.87, 1.67)	1.10 (0.78, 1.65)	0.075^b
Acceptable	89 (29.9)	139 (41.4)	0.006^c
Borderline	88 (29.5)	73 (21.7)
High	121 (40.6)	124 (36.9)
MHO, n^g	169	191
Yes	3 (1.8)	9 (4.7)	0.121^c
No	166 (98.2)	182 (95.3)

a Medians and interquartile ranges (IQR 25, 75) for continuous and n (%) for categorized variables are reported.

b Mann–Whitney U test.

c Pearson’s chi-square test.

d Classified on the basis of guidelines of ISPAD (27).

e High: >15 mU/L in prepuberty, >30 mU/L in puberty, and >20 mU/L in postpuberty (28).

f Categorized on the basis of the guidelines of NCEP (26).

g Defined as normal fasting plasma glucose (27), 2-h-OGTT-glucose (27), B-HbA1c (27), serum insulin (28), plasma ALT (29), plasma TC, LDL-C, HDL-C, and TG (26).

2-h-OGTT-Gluc = 2-hour oral glucose tolerance test; B-HbA1c = glycosylated hemoglobin; fP-ALT = fasting plasma alanine aminotransferase (29); fP-Gluc = fasting plasma glucose; fS-Insulin = fasting serum insulin; fP-TC = fasting plasma total cholesterol; HDL-C = high-density lipoprotein cholesterol; IFG = impaired fasting glucose; IGT = impaired glucose tolerance; ISPAD = The International Society for Pediatric and Adolescent Diabetes; LDL-C = low-density lipoprotein cholesterol; MHO = metabolically healthy obese; NCEP = The National Cholesterol Education Program Expert Panel; TG = triglycerides.

Table IV. Anthropometric, clinical, and metabolic measurements in girls and boys by pubertal stage.

	Girls (n = 401)				Boys (n = 445)
	Prepubertal	Early/midpubertal	Late pubertal	Postpubertal	Prepubertal	Early/midpubertal	Late pubertal	Postpubertal
Pubertal stage^a	n (%)^b	n (%)^b	n (%)^b	n (%)^b	n (%)^b	n (%)^b	n (%)^b	n (%)^b
n	144	84	112	61	244	103	81	17
Height, cm	128.8 (118, 139)	150.4 (143, 158)	164.1 (159, 169)	166.0 (160, 170)	143 (129, 152)	163.4 (157, 170)	174.4 (170, 180)	179.4 (173, 183)
Height-SDS^c	0.8 (0.0, 1.6)	0.6 (–0.2, 1.4)	0.6 (–0.1, 0.1)	0.2 (–0.8, 0.9)	0.5 (–0.3, 1.5)	0.3 (–0.3, 1.2)	0.6 (–0.2, 1.3)	0.5 (–0.3, 1.2)
BMI-SDS^c	2.68 (2.39, 3.10)	2.36 (2.11, 2.55)	2.55 (2.33, 2.74)	2.56 (2.24, 2.92)	2.44 (2.18, 2.81)	2.34 (2.08, 2.60)	2.54 (2.23, 2.81)	2.52 (2.12, 2.78)
Overweight	17 (11.8)	19 (22.6)	7 (6.3)	10 (16.4)	10 (4.1)	1 (1.0)	2 (2.5)	0
Obesity	63 (43.7)	55 (65.5)	79 (70.5)	27 (44.3)	89 (36.5)	52 (50.5)	24 (29.6)	6 (35.3)
Severe obesity	38 (26.4)	7 (8.3)	25 (22.3)	19 (31.1)	91 (37.3)	40 (38.8)	39 (48.1)	8 (47.1)
Morbid obesity	26 (18.1)	3 (3.6)	1 (0.9)	5 (8.2)	54 (22.1)	10 (9.7)	16 (19.8)	3 (17.6)
Blood pressure, n	112	66	99	51	195	84	73	16
SBP, mmHg^d	111 (106, 119)	119 (113, 127)	122 (113, 131)	130 (121, 138)	115 (108, 124)	125 (117, 131)	139 (124, 149)	141 (127, 152)
Normal	43 (38.4)	24 (36.4)	42 (42.4)	17 (33.3)	89 (45.7)	33 (39.3)	18 (24.7)	4 (25.0)
High normal	20 (17.9)	10 (15.2)	11 (11.1)	7 (13.8)	18 (9.2)	9 (10.7)	4 (5.5)	2 (12.5)
Stage 1	38 (33.9)	23 (34.8)	34 (34.4)	17 (33.3)	72 (36.9)	34 (40.5)	23 (31.5)	3 (18.7)
Stage 2	11 (9.8)	9 (13.6)	12 (12.1)	10 (19.6)	16 (8.2)	8 (9.5)	28 (38.3)	7 (43.8)
DBP, mmHg^d	67 (60, 72)	70 (65, 77)	73 (65, 79)	67 (62, 75)	67 (62, 73)	72 (63, 77)	74 (67, 81)	74 (69, 81)
Normal	75 (67.0)	43 (65.2)	62 (62.6)	44 (86.3)	151 (77.3)	65 (77.4)	44 (60.3)	9 (56.3)
High normal	18 (16.0)	14 (21.2)	20 (20.2)	3 (5.9)	19 (9.8)	10 (11.9)	15 (20.5)	3 (18.7)
Stage 1	17 (15.2)	9 (13.6)	14 (14.2)	4 (7.8)	23 (11.9)	9 (10.7)	13 (17.8)	3 (18.8)
Stage 2	2 (1.8)	0	3 (3.0)	0	2 (1.0)	0	1 (1.4)	1 (6.2)
fP-ALT, n	82	43	79	38	155	77	56	11
IU/L	23 (18, 33)	26 (18, 34)	22 (16, 32)	30 (18, 36)	26 (19, 35)	27 (20, 44)	46 (31, 64)	48 (26, 75)
<40	69 (84.1)	34 (79.1)	66 (83.5)	31 (81.6)	123 (79.4)	56 (72.7)	21 (37.5)	5 (45.5)
>40	13 (15.9)	9 (20.9)	13 (16.5)	7 (18.4)	32 (20.6)	21 (27.3)	35 (62.5)	6 (54.5)
fP-Gluc, n^e	95	59	92	51	163	79	68	13
mmol/L	5.2 (4.9, 5.5)	5.3 (5.1, 5.7)	5.4 (5.2, 5.7)	5.4 (5.1, 5.7)	5.4 (5.1, 5.6)	5.5 (5.3, 5.9)	5.5 (5.4, 6.0)	5.8 (5.2, 6.0)
<5.6	78 (82.1)	42 (71.2)	56 (60.9)	35 (68.6)	116 (71.2)	40 (50.6)	35 (51.5)	5 (38.5)
5.6–6.9	17 (17.9)	17 (28.8)	34 (36.9)	14 (27.5)	46 (28.2)	38 (48.1)	31 (45.6)	8 (61.5)
≥7.0	0	0	2 (2.2)	2 (3.9)	1 (0.6)	1 (1.3)	2 (2.9)	0
2-h-OGTT-Gluc, n^e	10	18	47	11	35	34	25	7
mmol/L	5.5 (4.5, 7.0)	6.4 (5.9, 7.2)	6.3 (5.3, 7.3)	6.0 (5.5, 6.9)	6.1 (5.4, 7.0)	6.1 (5.7, 6.9)	6.7 (5.8, 7.7)	7.4 (5.5, 8.0)
<7.8 mmol/L	9 (90.0)	16 (88.9)	38 (80.9)	9 (81.8)	32 (91.4)	31 (91.2)	19 (76.0)	4 (57.1)
7.8–10 mmol/L	1 (10.0)	2 (11.1)	8 (17.0)	1 (9.1)	3 (8.6)	3 (8.8.)	6 (24.0)	2 (28.6)
≥11.1	0	0	1 (2.1)	1 (9.1)	0	0	0	1 (14.3)
B-HbA1c, n^e	33	25	31	23	48	37	38	5
mmol/L	34 (31, 37)	35 (34, 38)	34 (32, 37)	35 (33, 38)	35 (33, 37)	35 (34, 38)	37 (34, 39)	38 (36, 43)
%	5.3 (5.0, 5.5)	5.4 (5.3, 5.6)	5.3 (5.1, 5.5)	5.4 (5.2, 5.6)	5.4 (5.2, 5.5)	5.4 (5.3, 5.6)	5.5 (5.3, 5.7)	5.6 (5.4, 6.1)
≤5.7	32 (97.0)	23 (92.0)	28 (90.3)	20 (87.0)	43 (89.6)	32 (86.5)	30 (79.0)	4 (80.0)
5.8–6.4	0 (0.0)	2 (8.0)	2 (6.5)	1 (4.3)	4 8.3)	5 (13.5)	7 (18.4)	0 (0.0)
fS-Insulin, n^f	63	44	77	41	118	67	50	11
mU/L	14.1 (9.8, 19.7)	23.8 (17.6, 31.7)	27.2 (19.1, 31.7)	28.5 (20.3, 38.6)	13.2 (9.6, 17.8)	22.0 (14.5, 31.3)	27.0 (18.6, 39.6)	28.0 (22.1, 48.1)
Normal	34 (54.0)	30 (68.2)	48 (62.3)	10 (24.4)	68 (57.6)	49 (73.1)	26 (52.0)	1 (9.1)
High	29 (46.0)	14 (31.8)	29 (37.7)	31 (75.6)	50 (42.4)	18 (26.9)	24 (48.0)	10 (90.9)
fP-TC, n^g	106	53	93	45	174	79	63	11
mmol/L	4.80 (4.18, 5.30)	4.40 (3.90, 4.95)	4.40 (3.80, 5.10)	4.20 (3.50, 4.63)	4.60 (4.00, 5.20)	4.00 (3.50, 4.80)	4.60 (3.80, 5.20)	5.09 (4.30, 5.80)
Acceptable	34 (32.1)	25 (47.2)	43 (46.2)	28 (62.3)	75 (43.1)	45 (57.0)	25 (39.7)	3 (27.3)
Borderline	38 (35.8)	21 (39.6)	28 (30.1)	11 (24.4)	54 (31.0)	20 (25.3)	22 (34.9)	5 (45.4)
High	34 (32.1)	7 (13.2)	22 (23.7)	6 (13.3)	45 (25.9)	14 (17.7)	16 (25.4)	3 (27.3)
fP-LDL-C, n^g	99	52	90	44	165	75	62	10
mmol/L	3.00 (2.50, 3.50)	2.80 (2.23, 3.30)	2.74 (2.30, 3.21)	2.31 (2.00, 3.03)	2.80 (2.30, 3.40)	2.60 (2.00, 3.10)	2.85 (2.40, 3.33)	3.20 (2.58, 3.53)
Acceptable	47 (47.5)	29 (55.8)	51 (56.7)	33 (75.0)	85 (51.5)	45 (60.0)	31 (50.0)	3 (30.0)
Borderline	16 (16.1)	11 (21.1)	21 (23.3)	4 (9.1)	38 (23.0)	15 (20.0)	16 (25.8)	3 (30.0)
High	36 (36.4)	12 (23.1)	18 (20.0)	7 (15.9)	42 (25.5)	15 (20.0)	15 (24.2)	4 (40.0)
fP-HDL-C, n^g	102	51	92	44	171	75	63	11
mmol/L	1.24 (1.02, 1.44)	1.11 (1.00, 1.44)	1.13 (0.92, 1.29)	1.10 (0.99, 1.35)	1.23 (1.04, 1.48)	1.09 (0.95, 1.24)	0.95 (0.87, 1.13)	0.95 (0.85, 1.13)
Acceptable	57 (56.5)	19 (36.5)	39 (42.4)	19 (43.2)	98 (57.3)	27 (36.0)	9 (14.3)	2 (18.2)
Borderline	17 (16.8)	15 (28.8)	18 (19.6)	12 (27.3)	32 (18.7)	17 (22.7)	16 (25.4)	2 (18.2)
Low	27 (26.7)	18 (34.7)	35 (38.0)	13 (29.5)	41 (24.0)	31 (41.3)	38 (60.3)	7 (63.6)
fP-TG, n^g	103	51	92	45	172	74	65	11
mmol/L	0.98 (0.74, 1.45)	1.42 (1.04, 1.73)	1.34 (1.03, 1.88)	1.16 (0.92, 1.58)	0.97 (0.70, 1.53)	1.02 (0.77, 1.51)	1.56 (1.04, 2.35)	1.56 (1.19, 2.68)
Acceptable	40 (38.8)	9 (17.6)	23 (25.0)	16 (35.6)	81 (47.1)	36 (48.6)	14 (21.5)	1 (9.1)
Borderline	25 (24.3)	14 (27.5)	30 (32.6)	15 (33.3)	33 (19.2)	18 (24.3)	15 (23.1)	3 (27.3)
High	38 (36.9)	28 (54.9)	39 (42.4)	14 (31.1)	58 (33.7)	20 (27.1)	36 (55.4)	7 (63.6)
MHO, n^h	53	29	62	24	94	49	34	8
Yes	1 (1.9)	0 (0)	2 (3.3)	0 (0)	8 (8.5)	0 (0)	0 (0)	0 (0)
No	52 (98.1)	29 (100)	60 (96.7)	24 (100)	86 (91.5)	49 (100)	34 (100)	8 (100)

a Pubertal stage categorized on the basis of the Tanner method (23,24), prepubertal (M/G1), early/midpubertal (M/G2–3), late pubertal (M/G4–5), postpubertal (adult height achieved).

b Medians and interquartile ranges (IQR 25, 75) are reported for continuous variables and n (%) for categorized variables unless otherwise stated.

c Categorized according to the Finnish growth reference (21).

d Stage 1 and 2 are hypertensive; classified on the basis of NHBPEP (26).

e Classified on the basis of ISPAD (27).

f High, hyperinsulinism >15 mU/L in prepuberty, >30 mU/L in puberty, and >20 mU/L in postpuberty (28).

g Categorized by NCEP (26).

h Defined as normal fasting plasma glucose (27), 2-h-OGTT-Gluc (27), B-HbA1c (27), serum insulin (28), plasma ALT (29), plasma TC, LDL-C, HDL-C, and TG (26).

2-h-OGTT-Gluc = 2-hour oral glucose tolerance test; B-HbA1c = glycosylated hemoglobin; BMI-SDS = body mass index SDS; DBP = diastolic blood pressure; fP-ALT = fasting plasma alanine aminotransferase; fP-Gluc = fasting plasma glucose; fP-HDL = high-density lipoprotein; fP-LDL = low-density lipoprotein; fP-TC = fasting plasma total cholesterol; fP-TG = triglycerides; fS-Insulin = fasting serum insulin; ISPAD = The International Society for Pediatric and Adolescent Diabetes; MHO = metabolically healthy obese; NCEP = The National Cholesterol Education Program Expert Panel; NHBPEP = The National High Blood Pressure Education Program; SBP = systolic blood pressure; SDS = SD score.

Hyperinsulinism (HI) was present in 43% of cases. Girls had significantly higher serum insulin levels than boys (Table III). However, as serum insulin values increased in both sexes with age and sexual maturity, this difference had disappeared by the end of puberty (Table II and IV). HI was present in 54% of 15–18-year-olds and in 79% of postpubertal children. Severe obesity increased the frequency of HI in boys and in pubertal girls (Figure 1). Acanthosis nigricans was present in 40% of severely and morbidly obese children and in 11% of overweight children (chi-square, p = 0.001). It was more prevalent in children with HI (49%) than in those with normal serum insulin concentrations (27%; chi-square, p < 0.001).

TC values were within the acceptable range in 44% of children and high in 24%, and there was no sex difference (Table III). LDL-C values were high in 25% of all cases, but higher proportions of prepubertal girls (36%) and postpubertal boys (40%) had high LDL-C (Table IV). In 15–18-year-olds who were late or postpubertal (n = 107), 34% of boys and 20% of girls had high LDL-C. Severe obesity was not associated with higher TC or LDL-C (Figure 2). The overall prevalence of low HDL-C was 34%. In boys HDL-C concentrations decreased with increasing age and sexual maturity such that at the end of puberty over 60% of boys had low HDL-C (Tables II and IV). Severe obesity was associated with low HDL-C regardless of sex or pubertal stage (Figure 2). The overall prevalence of high TG values was 39%. In boys TG values increased with age and sexual maturity (Tables II and IV). In 15–18-year-olds who were late or postpubertal, 52% of boys and 34% of girls had high TG. In severely obese prepubertal girls and pubertal boys, and in all pubertal girls, the proportion with unacceptable TG was as high as 73% (Figure 2).

Figure 2. The relationship between severity of obesity and lipid metabolism and metabolically healthy obesity in prepubertal and pubertal girls and boys. A: Plasma total cholesterol (TC) in mmol/L, acceptable: <4.40; high: ≥5.18 (26). B: Plasma low-density lipoprotein cholesterol (LDL-C) in mmol/L, acceptable: <2.84; high: ≥3.36 (26). C: Plasma high-density lipoprotein cholesterol (HDL-C) in mmol/L, acceptable: >1.17; low: <1.04 (26). D: Plasma triglyceride (TG) in mmol/L, for <10-year-olds, acceptable: <0.84; high: ≥1.13; and for ≥10-year-olds, acceptable: <1.02; high ≥1.47 (26). E: Metabolically healthy obese (MHO), defined as normal fasting plasma glucose, 2-h OGTT glucose, B-HbA1c, serum insulin, plasma ALT, TC, LDL-C, HDL-C, and TG. OW/OB = overweight and obesity, SOB/MOB = severe obesity and morbid obesity (21). Pubertal = Tanner ≥M2/G2 (breast bud/testicular volume >3 mL) (23,24).

Plasma ALT ≥40 IU/L was present in 24% of cases, of which 70% were boys. ALT levels were higher in boys than girls, and boys’ ALT levels increased with age and sexual maturity (Tables II, III, and IV). In the 15–18-year-old group 57% of boys and 16% of girls had high ALT (Table II). ALT was independent of the degree of obesity in prepuberty, but, during puberty, ALT levels in both sexes were highest in those with severe or morbid obesity (Figure 1).

Among subjects for whom complete metabolic data were available on the first clinic visit (n = 360), 3% (12 cases) were assigned MHO (Tables II, III, and IV), and only four of the MHO children had normal blood pressure. Most MHO children (8 cases) were prepubertal boys (Figure 2). At least one risk factor for CVD was present in 649 cases (80%). In the 15–18-year age group (n = 122), boys were more obese (Mann–Whitney U, p = 0.043) and more likely to have hypertensive SBP than girls (chi-square, p = 0.012); they were also more likely to have dyslipidemia (chi-square, p = 0.025), prediabetes (chi-square, p < 0.001), or signs of fatty liver (chi-square, p < 0.001) than girls of the same age. In this age group, having at least one CVD risk factor was more frequent in boys than in girls (chi-square, p = 0.005).

Discussion

This study showed an alarming quantity of cardiometabolic abnormalities in overweight and obese children referred to specialized care. The children had a high prevalence of impaired glucose metabolism (prediabetes 35%, HI 43%), dyslipidemia (low HDL 34%, high TG 39%), signs of hepatic steatosis (24%), and hypertensive blood pressure (high SBP 50%). Boys tended to have worse metabolic and cardiovascular risk profiles and more comorbidity than girls. This difference between the sexes became even more evident towards adulthood. Unexpectedly, even the youngest children were not free of cardiometabolic risk factors.

Though dysfunctional adipose tissue causes chronic inflammation, insulin resistance, and metabolic complications, there is a subset of obese subjects who have normal metabolic profiles and they have been identified as the metabolically healthy obese (MHO) (30,31). In our study, there were only few MHO children (3%), much fewer than in previous studies reporting prevalences of MHO children from 25% to 68% (30). Definitions of MHO are variable, and there are no standardized criteria for it; thus comparing different studies is difficult. Also the maintenance and long-term prognosis of MHO is unclear. A number of studies have shown that MHO are at lower risk for cardiovascular events and mortality compared to metabolically unhealthy obese individuals and are not at elevated risk compared to normal-weight individuals. Other studies show that MHO individuals are still at increased risk for diabetes and cardiovascular events compared to lean, healthy individuals (31). Likewise, though obese children with metabolic abnormalities are clearly at increased risk of meaningful adult outcomes, high childhood BMI alone predicts adult metabolic syndrome, T2DM, and cardiovascular changes (32).

Current evidence indicates that up to 82% of obese children are likely to be obese as adults (10) and their obesity is expected to get even worse (3). There is clear evidence that atherosclerotic changes begin in childhood (15), and in the presence of other cardiovascular risk factors these early changes may progress to frank CVD. Similarly, prediabetes may progress to T2DM (10,11). There is also some evidence that the association between childhood BMI and CVD in adulthood is stronger in boys (12). The consequences of excess weight manifest clearly even in childhood. In addition to the immediate adverse effects (6), there is evidence that overweight and obesity have long-term consequences that will affect the individuals throughout life (3,6,9,10,12,16,32).

Disturbances of glucose metabolism have become more common in children. Abnormal glucose metabolism is regarded as a risk factor for adult metabolic syndrome, T2DM, and CVD. In our study 33.2% of subjects had IFG, 13.7% IGT, and 0.9% met criteria for T2DM. These rates are higher than those found in other contemporaneous European studies of overweight and obese children. A study in Swedish and German children (n = 35,633) aged 2–18 years reported IFG rates of 17.1% and 5.7% in Sweden and in Germany, respectively (33); a study of children from central Europe (n = 11,681) reported that IFG was present in 5.5% and IGT in 8.9% (34); and an Italian study (n = 510) reported IFG in 2.0%, IGT in 11.2%, and T2DM in 0.4% (35).

Insulin resistance is an important factor in adult obesity, T2DM, metabolic syndrome, and CVD. Also in children insulin resistance is related to obesity and cardiometabolic risk. There are no agreed criteria for juvenile insulin resistance, and in 2010 there was an international consensus that screening for insulin resistance in children was not justified (36). One method of assessing insulin sensitivity in obese children is to report fasting insulin concentrations. A study of overweight and obese children in the UK (n = 103), which used the same criteria as this study, reported a similar rate of hyperinsulinism, 40%, compared to our 43% (37). A recent Swiss study (n = 774), which used a lower threshold for hyperinsulinism (>15 mU/L regardless of pubertal stage) reported that the condition was present in 33% of overweight and obese children (38). The fasting insulin levels reported in Norwegian obese children (n = 203), Md = 13 mU/L, IQR 25, 75: 9, 21 mU/L, are lower than those from our sample (39).

Both the connection between childhood obesity and dyslipidemia, and the relationship between lipid values in childhood and adulthood are well established (16,17). In this study, obesity was clearly associated with elevated TG levels and with low HDL-C levels but not with elevated TC or LDL-C. This is probably an instance of what is known as the ‘atherogenic lipid triad’ (low HDL-C, high TG, and the presence of small dense LDL-C particles). Standard analyses of plasma LDL-C do not recognize small dense LDL-C, which becomes a problem especially with high TG concentrations. In childhood, insulin resistance is associated with an increase in the level of small dense LDL-C particles that play a very active role in the development of atherosclerosis and CVD (40). A review of German and Swiss overweight and obese children (n = 260 000) concluded that, unlike the other lipids, levels of LDL-C were not correlated with excess weight (41). The proportions of children with abnormal lipid values in this central European study (high TC: 14.1%; high LDL-C: 15.8%; low HDL-C: 11.0%; high TG: 14.3%) were lower than in our study (using the same thresholds, the corresponding rates in our study were: 23.6%, 24.6%, 16.0%, and 25.3%). The prevalence of abnormal lipid profiles in our study was also higher than in comparable studies carried out in the UK (37) and USA (16). The introduction of new mass spectrometric methods for the analysis of lipid composition, such as ionization by electrospray (ESI), atmospheric pressure photo-ionization (APPI), and atmospheric pressure chemical ionization (APCI), raise the question of what method or methods should be used in future to identify lipid alterations in childhood (40).

Elevated ALT is considered a predictor of fatty liver, i.e. non-alcoholic liver disease, which is at present the most common chronic liver disease in pediatrics. The prevalence of fatty liver has been increasing and may reach 80% in obese children (42); as in our study, it generally appears to be more frequent in boys, older children, and the extremely obese (43–45). The American Academy of Pediatrics recommends using ALT levels as a screen for fatty liver in all obese children. In our study we set the upper boundary of the normal range at 40 IU/L, a commonly accepted limit of normality (29). However, new thresholds have recently been proposed: 26 IU/L in boys and 22 IU/L in girls (46). Using these new threshold values the overall prevalence of elevated fP-ALT in this study would be 60% (186 cases) in boys and 54% (136 cases) in girls; in the sub-sample of children with severe or morbid obesity the rates would be 65% and 59%, respectively.

Mean blood pressure in children has increased in parallel with obesity (47). The prevalence of hypertensive BP in this study was very high. This is partly due to our measurement technique; the automated oscillometric device Dinamap 1846SX provides somewhat higher SBP recordings than the auscultatory method using a mercury sphygmomanometer (48) which is regarded as the ‘gold standard’ for office blood pressure measurement. Unfortunately, the ban on the use of mercury sphygmomanometers continues to diminish the role of this technique, and there is a need for easily available reference values using other techniques. However, our results cannot be entirely explained by measurement difficulties or excitement. The high prevalence of hypertensive BP is alarming, as childhood BP has been shown to predict adult BP (11,49). Moreover, adolescent hypertensive BP is associated with atherosclerotic changes, and hence CVD, independent of adult BP (50). One of the latest International Childhood Cardiovascular Cohort (i3C) Consortium studies concluded that a combination of high BMI and elevated BP increased the risk of left ventricular hypertrophy already in childhood (51). It is questionable whether elevated BP is given adequate consideration in the clinical evaluation and management of childhood obesity, for example relating it to parental history of hypertension and to lifestyle (34). There is evidence that life-style management can decrease the negative consequences of obesity (32).

The estimated prevalence of asthma is 7%–9% in pediatric populations. In our study, 22% of children had been diagnosed with asthma, and the condition was more frequent in boys and in prepuberty. Asthma patients may be over-represented in our data as treating the child’s obesity may have been part of a strategy for clinical management of his or her asthma.

The relationship between asthma and overweight is controversial. A recent systematic review concluded that there is a weak relationship between asthma and body weight (52). A cross-sectional analysis of a large US cohort (n = 681 122) of 6–19-year-olds drawn from the general population concluded that asthma was more prevalent in obese children (53), but other studies have reached the opposite conclusion. An Australian study (n = 5993) (54) and a study of Italian children (n = 1179) (55) both found no relationship between asthma and body weight.

Many studies have reported that obese children exhibit more psychopathology than normal weight children, but it is not known if psychiatric comorbidity is a cause or a consequence of obesity or if there are common factors that predispose to overweight and psychiatric disturbance (56). In this study 224 cases (25%) had a current or historical diagnosis of a psychological developmental disorder or a behavioral or mental disorder. The prevalence of ADHD in our study was 5.1%, 8.5% in boys, which is higher than in normal-weight children (3.5%), as has been reported in other studies (56,57). The connection between psychotropic drugs and obesity is often a concern in clinical work. Use of second-generation antipsychotic medications such as olanzapine, quetiapine, risperidone, and aripiprazole has been associated with significant weight gain (58). This class of anti-psychotic medication was used by 25 of the 45 children being treated with anti-psychotic medication in this study.

Confusingly, there are different definitions of what constitutes childhood overweight or obesity. There are no cut-offs for childhood obesity which are based on the adverse effects of extra adiposity; all cut-offs are based on an agreement. In this study, we used the Finnish definition of childhood overweight and obesity, which is analogous to the definition used by the International Obesity Task Force (IOTF), but based on Finnish population reference data (21,59). Using the Finnish reference data 4.4% of boys in the sample were overweight, and 95.6% were obese; the corresponding figures using IOTF data were 5.6% and 94.4%. For girls the figures were as follows (IOTF-referenced values in parentheses), overweight: 14.8% (6.0%), obese: 85.2% (94.0%). In other words 37% (31 out of 83) overweight children in this study were obese according to the IOTF definition. Only six children (0.7%) in our obese group were overweight according to the IOTF definition. The youngest children in our study had the highest BMI-SDS and were classified most obese, but this does not necessarily mean that they had the highest proportion of extra adipose tissue and were physiologically most obese.

Inevitably this study has some limitations. Because the sample was limited to children in specialist care the findings may not be generalizable to all overweight and obese children. Because of differences in the way patient information was recorded by the participating centers we were not able to analyze the parental history. The change in analysis practices at the beginning of 2008 had a small effect on mean levels of HDL-C and fasting glucose in girls, and on LDL-C in boys; these effects are unlikely to have influenced the results, and the slight increases in levels since 2008 might be a reflection of reality rather than an artifact caused by laboratory factors. Additionally, blood pressure was recorded with an oscillometric technique, and it is likely that SBP was somewhat overestimated.

This study also has some notable strengths. First, the sample consisted of all children who received specialist treatment for overweight and obesity in the study region during the study period; the results are therefore representative of this group of patients. Second, clinical and laboratory data were available for a high proportion of the sample: height and weight data, 100%; pubertal stage data, 94%; BP data, 94%; glucose levels, 71%; insulin levels, 53%; all lipids, 67%; and ALT, 63%. The laboratory data also correlated well with contemporary obesity status. Third, the sample was an ethnically homogeneous group originating from the area of Eastern Finland; from this perspective the sample (n = 900) was a large one. Finally, this study is the first to describe the metabolic and cardiovascular profile of obese children in Finland and contributes Northern European data to the body of evidence on childhood obesity.

Declaration of interest

The authors state that there are no conflicts of interest and no payment was received for preparation of this manuscript.

Funding was received from the Päivikki and Sakari Sohlberg Foundation; The Foundation for Pediatric Research; and State Research Funding.