Health Issues With Being Overweight - Dr. Fried in Ocean New Jersey
With increasing overweight, the body becomes resistance to insulin which can eventually lead to diabetes. With resistance to insulin, blood sugar is also deposited in the liver as fat causing liver damage as evidenced by elevated liver tests which with repeated insult progresses to liver failure. Abnormal blood sugar, insulin levels and liver tests are easy to measure in routine blood tests. Dr Fried also specializes in measuring Body composition of muscle and fat so as you lose weight you may be monitored to preserve muscle and selectively lose fat. Insulin resistance, fatty liver, blood pressure, reflux and sleep problems are all reversible with weight loss. Call Dr Fried now to help you or your loved one begin a healthy, physician supervised weight loss program so that all your days may be healthier.
Elevated Fasting serum insulin occurs with advancing BMI in children with Obesity
Ryan Himes, MD* and Martin D. Fried, MD**
From the departments of General Pediatrics* and Pediatric Gastroenterology and Nutrition**,
The K. Hovnanian Children’s Hospital at Jersey Shore Medical University Center
1945 Route 33
Neptune, New Jersey 07712
Email to firstname.lastname@example.org
We conducted a retrospective chart review of 1400 patients 5-18 years old in order to determine if there was a correlation between blood pressure, fasting glucose, fasting insulin, hyperlipidemia and non alcoholic fatty infiltration of the liver and advancing Body Mass Index (BMI) in overweight and obese patients. Data was abstracted from medical records including height, weight, blood pressure, fasting glucose, fasting insulin, fasting total cholesterol, LDL, HDL, triglycerides, liver enzymes including AST, ALT, and the presence of fatty infiltration of the liver as assessed by ultrasound or CT scan. BM I was calculated dividing weight in kg by the heigh in meters, squared. A BMI greater than or equal to 95% for age was considered overweight.
Fasting insulin levels was the only comorbid condition associated with obesity in children that correlated positively with an increasing BMI.
Although the data lacks the statistical power to support the relationship between BMI and all of its comorbid conditions, this study did highlight a fasting insulin level as one of the associated conditions with Obesity. The relationship between elevated fasting insulin and BMI deserves further investigation
Obesity is widely recognized as an important risk to the health of children. Evidence suggests that overweight children experience many of the same comorbidities as do overweight adults. Conditions like insulin resistance, diabetes mellitus,1,2 hypertension, dyslipidemia,3 non-alcoholic steatohepatitis, gallstones,4 musculoskeletal disorders, obstructive sleep apnea, pseudotumor cerebri and depression5 have been described with increasing frequency in young people. It is estimated that up to 25% of obese adolescent Americans satisfy diagnostic criteria for the metabolic syndrome, which has been identified as an important risk factor for cardiovascular disease in adults.6 The morbidity and mortality associated with pediatric overweight as patients accrue comorbid conditions, is cause for concern. In the adult population, studies have demonstrated that increasing body mass index (BMI) is positively related toincreased morbidity and mortality from the risk of developing diabetes mellitus, cholelithiasis, hypertension and coronary heart disease.7
This study describes the prevalence of obesity and the associated comorbidities of hypertension, impaired fasting glucose, impaired fasting insulin, dyslipidemia, elevated aspartate aminotransferase (AST), elevated alanine aminotransferase (ALT) and the presence of fatty infiltration of the liver in a pediatric gastroenterology referral clinic at a tertiary care, academic medical center. We also investigated whether increasing BMI is associated with more severe presentations of the conditions under study.
We conducted a retrospective chart review of 1400 patients aged 5-18 years who were referred to a pediatric gastroenterology clinic at a tertiary care academic medical center, over a one year period from April 2003 thru March 2004.
Data abstracted from the medical record included demographic information, height, weight, blood pressure, fasting glucose, fasting insulin, fasting lipid panel (total cholesterol, low-density lipoprotein (LDL), high-density lipoprotein (HDL) and triglycerides), AST, ALT and presence of fatty infiltration of the liver as assessed by ultrasound or computed tomography (CT) scan.
Normative lab values were determined based on age and sex-specific norms as reported in the Harriet Lane Handbook, 16th edition.
Overweight was defined as a BMI in excess of the 95%ile according to CDC NCHS growth charts. BMI was calculated as weight in kilograms divided by height in meters squared (kg / m2).
Criteria for inclusion were: 1) height and weight recorded for at least one visit with a BMI in excess of the 95%ile for age based on CDC NCHS growth charts, 2) at least one documented clinical or laboratory finding as outlined above 3) age between 5-18 years.
Criteria for exclusion were: 1) charts for which BMI was plotted but corresponding height or weight information was missing, 2) charts which did not include objective evidence for any of the conditions under study as outlined above, 3) children with documented syndromic obesity (n = 1, Prader-Willi syndrome).
A review of 1400 medical records from137 (9.8%) patients with overweight or obesity revealed that girls accounted for 47% (65/137) and boys accounted for 53% of that group. Average BMI for boys and girls was similar, 31 ± 7.5 and 32 ± 9 (P = 0.83), respectively.
Fasting glucose was obtained for 94 patients (Table 1) and 6 (6.4%) were found to have impaired fasting glucose (>100 mg/dL). Boys were more likely to have impaired fasting glucose than girls (10% vs. 2%, P = 0.21). Among all participants there was a weakly positive correlation between BMI and fasting serum glucose (Table 2) that did not achieve statistical significance (r = 0.11, P = 0.31). Analysis by gender yielded similar results for boys (r = 0.18, P = 0.21) and for girls (r = 0.13, P = 0.39).
Fasting insulin level was documented for 41 patients and 21 (51%) were abnormally high (≥15 mµ/mL). Boys were more likely to have abnormal fasting insulin levels than girls (57% vs. 45%, P = 0.44). An increasing BMI was correlated positively with an increased fasting insulin level for all children (r = 0.52, P = 0.0005). Statistical significance persisted in a sub analysis for males (r = 0.65, P = 0.0015), but did not do so for females (r = 0.31, P = 0.18).
Total fasting cholesterol exceeded the upper limit of normal (>199 mg/dL) for 19 of 78 (24%) of patients. More boys than girls had high fasting cholesterol (26% vs. 23%, P = 0.69). Overall, increasing BMI demonstrated a weakly positive, statistically insignificant, correlation with total cholesterol (r = 0.10, P = 0.37). Results were similar for females (r = 0.23, P = 0.15), however, males demonstrated a weakly negative, insignificant, trend towards increasing cholesterol with increasing BMI (r = -0.03, P = 0.88).
Low fasting HDL cholesterol (< 45 mg/dL) was present in 45 of 73 (62%) children with obesity. The proportion of boys with low fasting HDL cholesterol exceeded that of girls (80% vs. 45%, P = 0.002). There was a weak positive correlation between BMI and low fasting HDL cholesterol (r = 0.03, P = 0.78) that persisted in subgroup analysis for boys (r = 0.03, P = 0.86) and girls (r = 0.03, P = 0.62).
Increased fasting LDL cholesterol (> 129 mg/dL) was found in 18 of 72 subjects (25%). Boys were more likely to have increased fasting LDL than girls (29% vs. 21%, P = 0.41). A weak negative correlation was observed between BMI and increased LDL (r = -0.09, P = 0.47) which was similar in analyses for both boys (r = -0.16, P = 0.38) and girls (r = -0.04, P = 0.82).
Elevated fasting triglycerides (for males and females, respectively, by age: >108 mg/dL, >114 mg/dL for 6-11 year-olds, >138mg/dL, >138 mg/dL for 12-15 year-olds, >163 mg/dL, >128 mg/dL for 16-19 year-olds) were found in 33 of 74 children (45%). Boys more frequently had increased fasting triglycerides than girls (53% vs. 37%, P = 0.17). Statistically insignificant negative trends between BMI and triglycerides were observed for all children (r = -0.09, P = 0.47) and in sub analysis for boys (r = -0.08, P = 0.65) and girls (r = -0.13, P = 0.45).
Increased serum AST (> 50 U/L for 4-6 year-olds, > 40 U/L for 7-9 year-olds, > 60 U/L for 10-11 year-olds and > 45 U/L for 12-19 year-olds) was found in 10 of 109 children (9%). More boys than girls had elevations in their AST (13% vs. 6%, P = 0.32). Across all analyses there were weakly negative correlations between BMI and elevated AST observed (r = -0.18, P = 0.06 for all participants, r = -0.20, P = 0.14 for boys and r = -0.16, P = 0.27 for girls).
Serum ALT levels were obtained for 109 patients and 20 (18%) were found to be abnormally elevated (> 40 U/L for males, > 35 U/L for females). Boys were more likely than girls to have increased serum ALT (30% vs. 6%, P = 0.0009). There was a weak negative relationship between BMI and ALT that did not reach statistical significance (r = -0.05, P = 0.62); results among boys were similar (r = -0.06, P = 0.64). For girls, however, there was a small positive correlation between BMI and elevated ALT (r = 0.04, P = 0.79).
Twelve of 35 (34%) patients examined radiographically for the presence of fatty liver had the condition. Boys had a higher rate of fatty liver changes than did girls (36% vs. 31%, P = 1.00). An increased BMI was significantly and positively correlated with having fatty infiltration of the liver (r = 0.33, P = 0.05). A similar relationship was observed among boys (r =0.52, P = 0.01) and girls (r = 0.12, P = 0.71).
Of 83 patients for whom blood pressure was recorded, 48 (58%) of those were hypertensive by age, sex and height-specific norms. More females than males had hypertension (59% vs. 56%, P = 0.81). A weak positive correlation was observed for BMI and hypertension (r = 0.08, P = 0.46), an association that persisted for girls (r = 0.17, P = 0.28) but not for boys (r = -0.02, P = 0.88).
In this cohort of overweight and obese children there were few strong, direct relationships between comorbidities of obesity and BMI. The condition for which there was the strongest correlation with higher BMI was impaired fasting insulin, which has been shown by other authors as well.8, 9 Freedman et. al. conclude in their analysis of The Bogalusa Heart Study that insulin levels were most strongly associated with increasing adiposity, an association that persisted into adulthood.8, 9 Furthermore, the precipitous rise in juvenile type 2 diabetes mellitus has paralleled that of pediatric obesity and has been shown to be directly related to increasing BMI.1, 3 Although our results for the other measures did not necessarily achieve statistical significance, there are trends which merit comment and beg further research.
Compared with results from the most recent NHANES data, the prevalence of obesity in our population of 5-18 year-olds referred to a gastroenterology clinic was about half (9.8%) of the 17.1% reported in this nationally representative sample.10 This apparent discrepancy may reflect two possibilities. First, selection bias played a role.since the most obese children or those with complications are referred to a gastroenterology practice. This practice may underestimate the actual prevalence of obesity in the community.. The racial profile of our sample is unknown based on the information available in the medical record. Minority populations, who are known to have greater rates of childhood obesity, may also have been underrepresented.
Our sample was comprised of slightly more boys than girls (53% vs. 47%), in contrast to a similar study undertaken by Quattrin et. al. of obese children referred to a pediatric endocrinology clinic over 20 years.11 More frequently, their patients were girls (57.6%), which runs contrary to the slight male predominance of obesity demonstrated in the NHANES data.10 These girls had lower BMI’s than the boys, leading the authors to conclude that there was a tendency to refer girls to specialists when they are less obese than boys. Mean BMI was similar for boys and girls in our sample, suggesting that other factors may influence referral in our community. One such factor may be the presence of or suspicion of a condition comorbid with obesity. For all conditions being evaluated, except for hypertension, the proportion of boys with an adverse result exceeded that of girls. Whether boys are more susceptible to these outcomes, or if they may manifest them earlier in the course of their disease than girls cannot be extrapolated from our data. If this finding were to be confirmed by further study, it has the propensity to substantially impact the methodology and framework by which clinicians screen and treat obesity.
Measures of dysregulated carbohydrate metabolism, i.e. impaired fasting glucose and insulin, were most strongly and consistently associated with increasing BMI. Although statistical significance was achieved in correlating insulin level to BMI, interpretation of this should be done cautiously, since only 41 of 137 (30%) patients were tested for the condition. More children underwent fasting glucose measurements (69%), but correlations were generally weaker as well The association between overweight and impaired glucose handling in children3 can be detected early in its course.
Assessments of lipid metabolism were more variably related to BMI than other comorbidities. Trends towards positive associations were generally seen for elevated fasting total cholesterol and depressed HDL cholesterol levels. Paradoxically, weak negative correlations were observed for elevated fasting LDL cholesterol and triglycerides with BMI, though they were not statistically meaningful. Although seemingly unrelated in this cohort, other authors have reported direct relationships between elevated total cholesterol, LDL, triglycerides and low HDL cholesterol and BMI.3, 8 The disparate nature of these findings may again be partially explained by the multifactorial nature of an individual’s cholesterol metabolism.
Hepatic transaminases were the most frequently obtained laboratory test in this sample (80%). Generally, elevated AST and ALT were negligibly but negatively correlated with BMI. Though there was, overall, a moderately positive correlation between BMI and fatty infiltration of the liver, the small number of patients investigated limits generalization.
Clinical hypertension correlated inconsistently with BMI on our cohort, demonstrating a weak positive association with BMI in our sample overall. Once again, other research has demonstrated a direct association between the two.3, 8, 9 For boys there was a small but unexpected negative correlation between BMI and hypertension. The apparent incongruity may be attributable to small sample size or the limitations of one-time measurements.
Our data are subject to several important limitations which influence the generalizability of the conclusions that may be drawn from them. First, inherent to the retrospective design of the study, we are not able to demonstrate causative relationships, only associations. In order to overcome this impediment, a properly designed prospective study with non-obese controls would be needed. Next, our small sample size precluded comparisons that achieved statistical significance for most measures. Participant recruitment at additional clinical sites and expansion of inclusion criteria are simple ways to address this shortcoming. Also, ours was a highly selected group of patients. As they were all seen in the subspecialist referral setting, they may have represented the severest cases of obesity in the community. The fact that our sample was referred to a gastroenterologist may have increased the likelihood that they did have or were suspected of having obesity complicated by a comorbid condition. The rates of these conditions found herein may be elevated relative to their rates in the obese community at large. It is also not known if pediatricians selectively refer certain overweight children to gastroenterologists rather than endocrinologists or other specialists in our community. Lastly, the subjects under analysis here were uniformly extremely overweight. Study parameters were intentionally set to capture the most overweight children who were felt to have the greatest likelihood of manifesting complications of their obesity. However, we may have unintentionally obscured the true relationship between BMI and its comorbidities by examining only patients at the most extreme end of the spectrum. To remedy this, future investigations should include children with lesser degrees of overweight and non-obese controls.
Although the data lacks the statistical power to make solid conclusions about the relationship between BMI and its comorbid conditions, this study provides a starting point from which to launch further investigations. Specifically, the relationship between disordered carbohydrate metabolism and BMI merits closer examination with the benefits of a larger sample size studied prospectively.
1 Sinha r, Fisch G, Teague B, et. al. Prevalence of Impaired Glucose Tolerance Among Children and Adolescents with Marked Obesity. New England Journal of Medicine. 2002: 346 (11) 802-855.
2 Goran M, Ball G, Cruz M. Obesity and Risk of Type 2 Diabetes and Cardiovascular Disease in Children and Adolescents. Journal of Clinical Endocrinology and Metabloism. 2003: 88 (4) 1417-1427.
3 Weiss R, Dziura J, Burgert T. Obesity and the Metabolic Syndrome in Children and Adolescents. New England Journal of Medicine. 2004: 350 (23) 2362-2374.
4 Schwimmer J, McGreal N, Deutch R, Finegold M, Lavine J. Influence of Gender, Race, and Ethnicity on Suspected Fatty Liver Disease in Obese Adolescents. Pediatrics (electronic pages). 2005: 115 (5) 1405. Available from: http://www.pediatrics.org/cgi/doi/10.1542/peds.2004-1832.
5 Schneider M, Brill S. Obesity in Children and Adolescents. Pediatrics in Review. 2005: 26 (5) 155-162.
6 Someshwar J, Someshwar S, Perkins K. The Obese Adolescent. Pediatric Annals. 2006; 35: 181-186.
7 Willett W, Dietz W, Colditz G. Guidelines for Healthy Weight. New England Journal of Medicine. 1999: 341 (6) 427-433.
8 Freedman D, Dietz W, Srinivansan S, Berenson G. The Relation of Overweight to Cardiovascular Risk Factors Among Children and Adolescents: The Bogalusa Heart Study. Pediatrics. 1999: 103 (6): 1175-1182.
9 Freedman D, Khan L, Dietz W, Srinivasan S, Berenson G. Relationship of Childhood Obesity to Coronary Heart Disease Risk Factors in Adulthood: The Bogalusa Heart Studay. Pediatrics. 2001: 108 (3): 712-718.
10 Ogden C, Carroll M, Curtin L, McDowell, M, Tabak C, Flegal K. Prevalence of Overweight and Obesity in the United States, 1999-2004. JAMA. 2006: 295 (13): 1549- 1555.
11 Quattrin T, Liu E, Shaw N, Shine B, Chiang E. Obese Children Who Are Referred to the Pediatric Endocrinologist: Characteristics and Outcome. Pediatrics. 2005: 115 (2): 348-351.
12 Kish W, Baker S, Cochran W, et. al. AAP Policy Statement: Cholesterol in Childhood. Pediatrics. 1998; 101: 141-147.
TABLE 1. Proportion of overweight children who have comorbid conditions.
† Fasting triglycerides for males and females, respectively, by age: >108 mg/dL, >114 mg/dL for 6-11 year-olds, >138 mg/dL, >138 mg/dL for 12-15 year-olds, >163 mg/dL, >128 mg/dL for 16-19 year-olds.
* Serum AST >50 U/L for 4-6 year-olds, >40 U/L for 7-9 year-olds, >60 U/L for 10-11 year-olds, >45 U/L for 12-19 year-olds.
‡ Serum ALT >40 U/L for males, >35 U/L for females.
§ Hypertension defined by sex, age and height specific norms from The Harriet Lane Handbook, 16th ed.
TABLE 2. Correlation of Biomarkers with BMI
Call 732-682-3425 or contact Dr Fried in New Jersey if you have any questions about how being overweight can effect your health.