Learn more: PMC Disclaimer | PMC Copyright Notice
Obesity, Hypertension and Metabolic Syndrome in Children in West Virginia
Background
Obesity is a major public health concern in the United States. The prevalence of pediatric obesity has increased more than four-fold from the 1971–1974 National Health and Nutrition Examination Survey (NHANES) to the 2009–2010 data. The rate of obesity in children 2 to 19 years is 16.9%, according to the 2009–2010 NHANES report.1–2 West Virginia ranks second in the US for obesity prevalence in adults and children, with 20.9% of children ages 10–17 being obese. This is much higher than the national average of 14.8%.3
With the increase in obesity, an increase in hypertension in children has been noted and is believed to be frequently underdiagnosed.4 Hypertension in children is defined as having either a systolic or diastolic blood pressure (SBP, DBP) that is ≥95th percentile for height, gender and age.5 Various studies involving children have shown a definite link between obesity and hypertension. Obese children have a higher likelihood of hypertension compared to children with a normal BMI. In fact, obesity has been shown to increase the risk of developing hypertension threefold in children and adolescents.6
An underlying association, which links hypertension and obesity, is insulin resistance.7–9 Insulin resistance, in pediatrics, is also one of the components of metabolic syndrome (metS).14 It is associated with glucose intolerance—an indicator of metabolic syndrome. Viner et al defined metabolic syndrome being present when any 3 of the 5 following criteria are met: Body Mass Index (BMI) ≥ 95th percentile, SBP or DBP ≥ 95th percentile, triglycerides ≥ 155 mg/dL, HDL < 35 mg/dL, and an impaired glucose tolerance defined by either a fasting plasma glucose (FPG) ≥110 mg/dL or an abnormal glucose tolerance test with a 2-hour glucose >140 mg/dL.8 Obese hypertensive children already manifest 2 out of the 3 criteria required to diagnose metabolic syndrome.
The importance of recognizing metabolic syndrome in children is highlighted by Morrison et al, who found that metabolic syndrome in childhood is predictive of adult metabolic syndrome.10 Consequently these children are at increased risk of atherosclerotic disease, type 2 diabetes and steatorrheic hepatitis as adults.11
Objectives of the Study
To document the prevalence of metabolic syndrome in obese children with hypertension at a pediatric clinic in Charleston, WV.
To determine if an appropriate laboratory workup was initiated in children with obesity and hypertension and to assess other cardiovascular risk factors associated with Metabolic Syndrome.
Methods
IRB approval was obtained from the Charleston Area Medical Center (CAMC) Institutional Review Board.
Design
A retrospective study, which involves a review of records from patients that received care at the Children’s Medical Center (CMC). an urban pediatric primary care practice in Charleston, West Virginia.
Sample
The patient sample was derived from the primary care patients of the CMC.
Patients were identified from a computer generated list of medical record numbers with ICD-9 diagnosis codes pertaining to pediatric obesity and hypertension. Codes used pertained to: metabolic syndrome, obesity unspecified, morbid obesity, overweight, abnormal weight gain, hypertension unspecified, elevated blood pressure without the diagnosis of hypertension, acquired acanthosis nigricans, hyperinsulinism, impaired fasting glucose, impaired glucose tolerance test, pure hypercholesterolemia, pure hypertriglyceridemia, mixed hyperlipidemia and other unspecified hyperlipidemia. From this set of codes, 184 patients were identified and a chart review was completed. Of the 184 patients, 78 patients met inclusion criteria of obesity and hypertension.
These 78 charts were further reviewed for other criteria characteristic of metabolic syndrome. These criteria were defined by Viner and are as follows: triglycerides ≥ 155 mg/dL, High Density Lipoprotein (HDL) < 35 mg/dL, and an impaired glucose tolerance defined by either a Fasting Plasma Glucose (FPG) ≥110 mg/dL or an abnormal glucose tolerance test with a 2-hour glucose >140 mg/dL.8
Data Analysis
Statistical analysis was done by the Center for Health Services and Outcomes Research. A two-sample t-test was used to compare the values for triglyceride, HDL, Low Density Lipoprotein (LDL), total cholesterol, fasting blood sugar and BMI between the patients with and without metabolic syndrome. All analysis was done at level of significance of 0.05. Analysis was done on SAS 9.3 (SAS Institute, Cary, NC).
Results
A total of 78 patients met the inclusion criteria of either obesity and or hypertension. Fifty-two or (67%) were male and 26 (34%) female with ages ranging from 3 to 18 years. Age group distribution of the study population is detailed in Table 1.
Table 1
Demographic Characteristics of Obese-Hypertensive Children
| Mean+/−std.dev. (Min-Max) | |
|---|---|
| Age | 11.27+/−3.27(3–18) |
| Height (cm) | 150.62+/−17.72(100.5–179) |
| Weight (Kg) | 78.51+/−29.33(23.9–183.3) |
| BMI | 33.46+/−7.51(21.47–62.68) |
| N (%) | |
| Gender | |
| Male | 52(66.67%) |
| Female | 26(33.33%) |
| Age by categories | |
| Preschool (3–5) | 5(6.41%) |
| Middle Childhood (6–9) | 16(20.51%) |
| Early adolescence (10–13) | 32(41.03%) |
| Middle adolescence (14–16) | 23(29.49%) |
| Late adolescence (17–18) | 2(2.56%) |
The first inclusion criteria for the study was obesity as defined by a BMI ≥ 95 percentile on the CDC BMI growth charts. The BMI ranged from 21.5 to 62.7 kg/m2 with a mean BMI of 33.5 kg/m2 as illustrated in Table 1.
The second inclusion criteria for the study was hypertension (either SBP or DBP ≥ 95th percentile). Mean systolic blood pressure was 134 mmHg and mean diastolic blood pressure was 70 mmHg. Figure 1 shows the percentage distribution of our patients according to BP percentiles.
Of the 78 obese and or hypertensive patients, 56 patients (72%) met the inclusion criteria of obesity and hypertension, and 22 patients (28%) met at least one additional criterion for metabolic syndrome as illustrated in Figure 2. There were 18 patients out of 78 (23%) who met obesity and hypertension inclusion criteria but could not be evaluated for metabolic syndrome since they did not have any laboratory evaluation performed.
When comparing the data by groups, patients with metabolic syndrome (N=22, 28%) and without metabolic syndrome (N= 56, 72%) using the two-sample t-test, there was a statistically significant difference found in TG (p=0.0003) and HDL (p=0.0002) as illustrated in Table 2. There were 38 patients out of 56 (68%) patients in the group without metabolic syndrome who had a laboratory evaluation for metabolic syndrome risk factors (i.e. fasting blood sugar, oral glucose tolerance test, fasting lipid profile) but those values were not high enough to meet criteria for metabolic syndrome. Despite the diagnosis of obesity and hypertension, 18 of 56 patients (32%) patients, did not have any laboratory tests performed.
Table 2
Comparison of Metabolic Syndrome and no Metabolic Syndrome.
| Do Not Meet Metabolic Syndrome (N=56) | Meet Metabolic Syndrome (N=22) | P-value | |
|---|---|---|---|
| Age | 11.35+/−3.30(3–18) | 11.04+/−3.24(3–15) | 0.7074 |
| Triglyceride | 102.9+/−30.08(44–147) | 186.8+/−89.89(51–369) | 0.0003* |
| HDL | 48.13+/−10.73(35–82) | 37.40+/−8.12(28–64) | 0.0002* |
| LDL | 91.89+/−23.73(56–144) | 98.40+/−35.41(49–196) | 0.4477 |
| Total cholesterol | 160.3+/−26.00(108–248) | 173.2+/−46.02(104–267) | 0.2381 |
| Fasting blood sugar | 89.35+/−9.12(66–106) | 104.4+/−44.17(83–298) | 0.1289 |
| BMI | 33.37+/−8.33(21.4–62.7) | 33.66+/−5.02(23.4–42.6) | 0.8541 |
We compared age by categories as shown in Table 1 for the study population (78 patients) with and without metabolic syndrome but we did not find any statistical difference (p= 0.7627).
Discussion
The increasing prevalence of obesity has been consistently reported by several studies.1–3,6,16,19 The National Health and Nutrition Examination Surveys (NHANES) data support that not only are American children heavier on average, but the heaviest are getting heavier. These data are supported by our sample, as most of our patients are considered morbidly obese even by adult standards. The Bogalusa Heart Study found that pediatric obesity is predictive of adult obesity.15 Furthermore, obesity that begins before age 8 and persists into adulthood is associated with an average final BMI of 41 kg/m2 as compared to a BMI of 35 kg/m2 for adult-onset obesity.15
Childhood obesity is associated with hypertension, which is usually isolated systolic hypertension.6 The blood pressure results as seen in Figure 1 are consistent with these reports. Sun et al reported that children with a single elevated blood pressure in childhood are at increased risk of hypertension and metabolic syndrome later in life. Also, a difference in blood pressure measurements between adults with and without the metabolic syndrome was found as early as age 5 for boys and age 8 for girls.25
As BMI increases, so does the risk of metabolic syndrome due to increasing insulin resistance.16 This study utilized Viner’s8 criteria for metabolic syndrome: BMI ≥ 95th percentile, BP ≥ 95th percentile, FPG ≥ 100 mg/do, HDL < 35 mg/do, TG ≥ 155 mg/do. These criteria are stricter than those recently released by the International Diabetes Federation, which defines metabolic syndrome in children and adolescents as an HDL < 40 mg/dL (< 50 in females), FPG ≥ 100 mg/dL and TG ≥ 150 mg/dL.17 All comparative statistical analyses done in this study support the diagnostic criteria of metabolic syndrome as found in other studies.8, 17
Despite using stricter criteria, the prevalence of metabolic syndrome in our study was 28%. This is lower than the prevalence of 35% reported by Mir,14 which could be due to difference in diagnostic criteria used, patient selection, ethnicity/demographics, and age range. The Mir study had a small sample size of 20 obese (defined as BMI >97th percentile) hypertensive patients aged 7–20 from a nephrology referral center. These subjects were likely referred specifically for hypertension; average SBP was 144 in the Mir study and 134 in ours. Our study population was from a primary care setting and included younger patients down to the age of 3 who are at lower risk of metabolic syndrome and had only 2% above the age of 16 (Refer to Table 1). In addition, the Mir study found glucose intolerance to be predictive of metabolic syndrome whereas our study found dyslipidemia (as defined by an elevated TG or low HDL) to be more predictive (Refer to Table 2).
Our results show a prevalence of metabolic syndrome (28%) slightly lower than the 31.2% prevalence of metabolic syndrome reported by NHANES III (2004), which reported on obese adolescents aged 12–19. Our rates are likely lower because of the younger population sampled, as we reviewed patients down to the age of three. Also, we have a low ethnic minority population, and some ethnic minorities, particularly Hispanics, have a higher rate of metabolic syndrome than the general population.
The second aspect of our study was to determine if an appropriate laboratory workup was initiated in obese hypertensive children. At the time of obesity diagnosis, 77% of our study population had laboratory investigations initiated by their pediatrician. Conversely, the potential diagnosis of metabolic syndrome was missed in 23%. Expert committee guidelines and AAP clinical practice guidelines clearly state that for the diagnosis of obesity alone, a fasting lipid profile should be obtained to assess co-morbidities as early as age 2.12–13
Increased BMI in childhood has been shown to directly correlate with increased risk of a cardiovascular event in adulthood.19 Pediatric metabolic syndrome and changes in age-specific BMI percentile from childhood to adulthood were significant predictors of adult metabolic syndrome.11 Furthermore, the CHD risk has been postulated to depend on the duration and lifetime effects of childhood obesity into adulthood15 since fibrous plaques and coronary artery calcifications have been shown to begin in childhood.24 The association between metabolic syndrome, Type 2 Diabetes and cardiovascular disease is well-documented in the adult literature. These associations have not been ascertained in prospective pediatric trials.20 This is an area where more longitudinal research should focus since lifestyle changes resulting in weight loss could prevent development of type 2 diabetes21 and alter other cardiovascular risk factors.
Conclusion
The diagnosis of obesity is the first step to preventing metabolic syndrome and its subsequent cardiovascular complications. At well child care visits, Benson reported that physicians failed to diagnose obesity in 46% of the cases.22 The only clinical tools needed in diagnosing obesity are BMI measurement and a growth chart. In our study, once obesity was diagnosed, 23% did not have the recommended workup ordered. Obesity in childhood should be considered a chronic medical condition.6 If medical providers intend to impact the current epidemic, then the diagnosis and recommended workup of obesity have to be implemented.
Limitations
Several study limitations were identified. The first is the nature of a retrospective study and a homogenous study population. Additionally, some patients may have been missed by our method of patient identification by computer-generated list from ICD-9 codes. The small sample size, especially in older adolescents, that met criteria could be reflective of missing patients who were obese or hypertensive but were not officially diagnosed as such.
Contributor Information
Jeannie Co, Children’s Medical Center, Charleston, WV.
Jamie Jeffrey, Children’s Medical Center, Charleston, WV.
Mary Emmett, Center for Health Services and Outcomes Research, Charleston Area Medical Center.
Asmita Modak, Center for Health Services and Outcomes Research, Charleston Area Medical Center.
Stephen B. Sondike, Medical Director, Disordered Eating Center of Charleston (DECC), Section Head, Adolescent Medicine, Professor of Pediatrics, West Virginia University, Charleston Division.