Obesity Research & Clinical Practice (2010) 4, e231—e237
ORIGINAL ARTICLE
Effect of modest changes in BMI on cardiovascular disease risk markers in severely obese, minority adolescents Unab I. Khan a,∗, Jessica Rieder a, Hillel W. Cohen b, Susan M. Coupey a, Rachel P. Wildman b a
Department of Pediatrics, Division of Adolescent Medicine, Children’s Hospital at Montefiore, Albert Einstein College of Medicine, 111 East 210th Street, Bronx, NY 10467, United States b Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, United States Received 5 May 2009 ; received in revised form 2 March 2010; accepted 4 March 2010
KEYWORDS Adolescent obesity; C reactive protein; Cardiovascular disease risk markers
Summary Background: African American and Hispanic adolescents have disproportionately higher rates of obesity compared to white adolescents. In adults, modest weight loss of five percent improves CVD risk marker levels. Less is known about the effects of modest changes in BMI on CVD risk markers in adolescents, particularly newer markers such as C reactive protein (CRP), lipoprotein (a) and homocysteine. Objective: To examine the effect of modest BMI change on CVD risk marker levels in a group of severely obese, African American and Hispanic adolescents. Study design: A six-month longitudinal analysis. Subjects: Eighty-three African American and Hispanic adolescents were recruited (mean age ± sd: 15.1 ± 2.0 years); 50 (60%) were reevaluated at 6 ± 2 months. Results: At baseline, mean BMI was 42.3 ± 7.8 kg/m2 . BMI directly correlated with CRP (p = < 0.001); homocysteine (p = 0.02); insulin (p = 0.05); and systolic and diastolic blood pressures (both p = 95th percentile for age and sex) African American and Hispanic adolescents (aged 12—20 years) at their initial visit to
a children’s hospital weight loss clinic. One hundred and forty-three consecutive adolescents were approached; 96 consented; 83 had baseline fasting blood tests and were included in this study. Fifty of the 83 subjects (60%) completed the reevaluation after a mean of 6 ± 2 months. Table 1 provides the baseline demographics and levels of CVD risk markers in our study sample. We excluded adolescents with acute infections or chronic inflammatory conditions, as that may invalidate the inflammatory marker assessment. The Montefiore Medical Center Institutional Review Board approved the study protocol. All subjects and their parents provided written informed consent.
Measures We measured height and weight in light clothing on a Scaletronix® stadiometer and calculated the BMI. BMI z-score, which is the BMI standardized for age and sex, was calculated using the Epi info software from the Centers for Disease Control and Prevention (CDC) [24]. Severe obesity was defined as BMI of >99th percentile for age and sex [20]. An adolescent medicine physician performed Tanner staging for sexual maturity rating on all subjects. Blood samples were obtained after an overnight fast of 8—12 h. Samples were centrifuged within 1 h of collection. CRP levels were measured using latex enhanced immunoturbidometry (CRP ultra wide reagent kit, Equal Diagnostics, Pennsylvania, USA) on the Olympus AU400/400 analyzer (Olympus Life and Material Science Europa GmbH, Lismeehan, Ireland). The analyzer has a lower limit of detection of 0.1 mg/dL. Lp(a) concentration was measured by immunoturbidometry with the Olympus analyzer AU400/400. Hcy levels were measured using immunoreaction using the Immulite 2000 kit and analyzer. Glucose, total cholesterol, HDL and triglycerides were measured using calorimetric methods (Olympus reagents, Olympus Life and Material Science Europa GmbH, Lismeehan, Ireland) using the Olympus AU400/400 analyzer. Insulin levels were measured by radioimmunoassay using
Effect of modest changes in BMI on cardiovascular disease risk markers Table 1 Baseline anthropometrics and levels of cardiovascular disease markers. Characteristics
N = 83
Mean age, years ± sd Females, N (%) Ethnicity, N (%) Hispanic African American Mean weight, kg ± sd Mean BMI, kg/m2 ± sd Mean BMI z-score ± sd Median C-reactive protein, mg/dl(IQR) Median lipoprotein (a), mg/dl (IQR) Median homocysteine, mol/L (IQR) Mean cholesterol, mg/dl ± sd Mean LDL, mg/dl ± sd Mean HDL, mg/dl ± sd Mean triglycerides, mg/dl ± sd Mean fasting glucose, mg/dl ± sd Median fasting insulin, units/l (IQR) Median HOMA-IR (IQR) Mean systolic blood pressure, mmHg ± sd Mean diastolic blood pressure, mmHg ± sd
15.1 ± 2.0 50 (60) 47 (57) 36 (43) 117.2 ± 26.2 42.3 ± 7.8 2.58 ± 0.3 0.5 (0.1, 1.1) 21 (9, 64) 6.3 (5.5, 7.7) 167 ± 33 99 ± 29 45 ± 10 110 ± 52 88 ± 7 88 (83, 93) 5.7 (4.2, 7.5) 118 ± 11 66 ± 7
human insulin RIA kits (Millipore, Massachusetts, USA). Intra-assay coefficients of variation were 100 mg/dL. Examining the association between severity of obesity and CVD risk markers, we found a significant direct correlation between BMI and both CRP levels (Spearman’s rho = 0.50, p = < 0.001) and Hcy levels (Spearman’s rho = 0.25, p = 0.03) (Table 2). We also found a significant direct correlation between BMI and fasting insulin levels (Spearman’s rho = 0.22, p = 0.05); mean systolic blood pressure (Pearson’s r = 0.43, p =
ORIGINAL ARTICLE
Effect of modest changes in BMI on cardiovascular disease risk markers in severely obese, minority adolescents Unab I. Khan a,∗, Jessica Rieder a, Hillel W. Cohen b, Susan M. Coupey a, Rachel P. Wildman b a
Department of Pediatrics, Division of Adolescent Medicine, Children’s Hospital at Montefiore, Albert Einstein College of Medicine, 111 East 210th Street, Bronx, NY 10467, United States b Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, United States Received 5 May 2009 ; received in revised form 2 March 2010; accepted 4 March 2010
KEYWORDS Adolescent obesity; C reactive protein; Cardiovascular disease risk markers
Summary Background: African American and Hispanic adolescents have disproportionately higher rates of obesity compared to white adolescents. In adults, modest weight loss of five percent improves CVD risk marker levels. Less is known about the effects of modest changes in BMI on CVD risk markers in adolescents, particularly newer markers such as C reactive protein (CRP), lipoprotein (a) and homocysteine. Objective: To examine the effect of modest BMI change on CVD risk marker levels in a group of severely obese, African American and Hispanic adolescents. Study design: A six-month longitudinal analysis. Subjects: Eighty-three African American and Hispanic adolescents were recruited (mean age ± sd: 15.1 ± 2.0 years); 50 (60%) were reevaluated at 6 ± 2 months. Results: At baseline, mean BMI was 42.3 ± 7.8 kg/m2 . BMI directly correlated with CRP (p = < 0.001); homocysteine (p = 0.02); insulin (p = 0.05); and systolic and diastolic blood pressures (both p = 95th percentile for age and sex) African American and Hispanic adolescents (aged 12—20 years) at their initial visit to
a children’s hospital weight loss clinic. One hundred and forty-three consecutive adolescents were approached; 96 consented; 83 had baseline fasting blood tests and were included in this study. Fifty of the 83 subjects (60%) completed the reevaluation after a mean of 6 ± 2 months. Table 1 provides the baseline demographics and levels of CVD risk markers in our study sample. We excluded adolescents with acute infections or chronic inflammatory conditions, as that may invalidate the inflammatory marker assessment. The Montefiore Medical Center Institutional Review Board approved the study protocol. All subjects and their parents provided written informed consent.
Measures We measured height and weight in light clothing on a Scaletronix® stadiometer and calculated the BMI. BMI z-score, which is the BMI standardized for age and sex, was calculated using the Epi info software from the Centers for Disease Control and Prevention (CDC) [24]. Severe obesity was defined as BMI of >99th percentile for age and sex [20]. An adolescent medicine physician performed Tanner staging for sexual maturity rating on all subjects. Blood samples were obtained after an overnight fast of 8—12 h. Samples were centrifuged within 1 h of collection. CRP levels were measured using latex enhanced immunoturbidometry (CRP ultra wide reagent kit, Equal Diagnostics, Pennsylvania, USA) on the Olympus AU400/400 analyzer (Olympus Life and Material Science Europa GmbH, Lismeehan, Ireland). The analyzer has a lower limit of detection of 0.1 mg/dL. Lp(a) concentration was measured by immunoturbidometry with the Olympus analyzer AU400/400. Hcy levels were measured using immunoreaction using the Immulite 2000 kit and analyzer. Glucose, total cholesterol, HDL and triglycerides were measured using calorimetric methods (Olympus reagents, Olympus Life and Material Science Europa GmbH, Lismeehan, Ireland) using the Olympus AU400/400 analyzer. Insulin levels were measured by radioimmunoassay using
Effect of modest changes in BMI on cardiovascular disease risk markers Table 1 Baseline anthropometrics and levels of cardiovascular disease markers. Characteristics
N = 83
Mean age, years ± sd Females, N (%) Ethnicity, N (%) Hispanic African American Mean weight, kg ± sd Mean BMI, kg/m2 ± sd Mean BMI z-score ± sd Median C-reactive protein, mg/dl(IQR) Median lipoprotein (a), mg/dl (IQR) Median homocysteine, mol/L (IQR) Mean cholesterol, mg/dl ± sd Mean LDL, mg/dl ± sd Mean HDL, mg/dl ± sd Mean triglycerides, mg/dl ± sd Mean fasting glucose, mg/dl ± sd Median fasting insulin, units/l (IQR) Median HOMA-IR (IQR) Mean systolic blood pressure, mmHg ± sd Mean diastolic blood pressure, mmHg ± sd
15.1 ± 2.0 50 (60) 47 (57) 36 (43) 117.2 ± 26.2 42.3 ± 7.8 2.58 ± 0.3 0.5 (0.1, 1.1) 21 (9, 64) 6.3 (5.5, 7.7) 167 ± 33 99 ± 29 45 ± 10 110 ± 52 88 ± 7 88 (83, 93) 5.7 (4.2, 7.5) 118 ± 11 66 ± 7
human insulin RIA kits (Millipore, Massachusetts, USA). Intra-assay coefficients of variation were 100 mg/dL. Examining the association between severity of obesity and CVD risk markers, we found a significant direct correlation between BMI and both CRP levels (Spearman’s rho = 0.50, p = < 0.001) and Hcy levels (Spearman’s rho = 0.25, p = 0.03) (Table 2). We also found a significant direct correlation between BMI and fasting insulin levels (Spearman’s rho = 0.22, p = 0.05); mean systolic blood pressure (Pearson’s r = 0.43, p =
