- Email: [email protected]
Insulin resistance syndrome in childhood: Associations of the euglycemic insulin clamp and fasting insulin with fatness and other risk factors
Insulin resistance syndrome in childhood: Associations of the euglycemic insulin clamp and fasting insulin with fatness and other risk factors Alan R. Sinaiko, MD, David R. Jacobs, Jr, PhD, Julia Steinberger, MD, Antoinette Moran, MD, Russell Luepker, MD, Albert P. Rocchini, MD, and Ronald J. Prineas, MD, PhD Objective: Our objective was to describe in children the relation of fatness and insulin resistance to the risk factors associated with the insulin resistance syndrome and to compare fasting insulin with the euglycemic insulin clamp as a measure of insulin resistance in children. Study design: This was a random selection of participants after blood pressure screening of 12,043 students in the fifth through eighth grades. Euglycemic insulin clamp studies with an insulin infusion rate of 1 mU/kg/min and a variable infusion of 20% glucose to maintain euglycemia, that is, plasma glucose at 5.6 mmol/L. Insulin sensitivity (Mlbm) is defined as the amount of glucose required to maintain euglycemia (milligrams of glucose infused per kilogram lean body mass per minute). Results: Body mass index was significantly correlated with fasting insulin and significantly inversely correlated with Mlbm. Fasting insulin was significantly correlated with systolic blood pressure in both sexes, all lipids, except high-density lipoprotein-cholesterol in males and triglycerides and high-density lipoprotein-cholesterol in females, but after adjustment was done for body mass index, it was significantly related only to triglycerides. Mlbm was significantly correlated only with triglycerides and high-density lipoprotein-cholesterol, and this did not change after adjustment was done for body mass index. A clustering effect for the risk factors was seen in children in the lowest quartile of Mlbm (highest degree of insulin resistance) compared with children in the highest quartile of Mlbm (lowest degree of insulin resistance). Conclusions: As defined by Mlbm, there is an early association of insulin resistance, independent of body fat, with the risk factors. There is a significant relation between fasting insulin, as an estimate of insulin resistance, and the risk factors, but this is significantly influenced by body fatness. The clustering of risk factors according to level of Mlbm suggests that adult cardiovascular disease is more likely to develop in children with the greatest degree of insulin resistance. (J Pediatr 2001;139:700–7)
From the Department of Pediatrics, Division of Epidemiology, School of Public Health, University of Minnesota Medical School, Minneapolis, Minnesota, the Department of Public Health Sciences, Wake Forest School of Medicine, WinstonSalem, North Carolina, and the Department of Pediatrics, University of Michigan, Ann Arbor.
Supported by grants HL 52851 and M01 RR00400 from the National Institutes of Health. Submitted for publication Nov 15, 2000; revisions received March 1, 2001, June 21, 2001; accepted July 2, 2001. Reprint requests: Alan R. Sinaiko, MD, University of Minnesota Medical School, 420 Delaware St, S.E., Box 491 UMHC, Minneapolis, MN 55455. Copyright © 2001 by Mosby, Inc. 0022-3476/2001/$35.00 + 0 9/21/118535 doi:10.1067/mpd.2001.118535
700
The insulin resistance syndrome, consisting of hyperinsulinemia, dyslipidemia, hypertension, and obesity, is associated with non-insulin dependent (type 2) diabetes and atherosclerotic cardiovascular disease in the adult population.1-3 The relation among these factors is complex,4 and it is not clear whether insulin resistance occurs in response to the pathophysiologic changes associated with the diseases or is etiologically related to their development. Studies in children have shown a significant relation among fasting insulin and lipids,5-8 blood pressure,7-10 and weight.11 This relation appears to follow a clustering pattern12 that tracks from childhood to young adulthood.13 Cross-sectional studies in children have confirmed a strong positive relation among overweight and fasting inBMI EIC HDL-C LBM LDL-C SBP
Body mass index Euglycemic insulin clamp High-density lipoprotein-cholesterol Lean body mass Low-density lipoprotein-cholesterol Systolic blood pressure
sulin, lipids, and blood pressure14 and, more specifically, significant correlations have been shown between body fat distribution and cardiovascular risk factors.15-17 When these findings are considered in the context of the significant tracking effect of blood pressure,18 lipids,19 and obesity20 between childhood and adulthood and the correlation of blood pressure and lipids21 with the extent of aortic and coronary
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS
VOLUME 139, NUMBER 5 atherosclerosis at autopsy in children, it is reasonable to suggest that the interaction among insulin, insulin resistance, and other risk factors begins during the first 2 decades of life. Although fasting insulin has been widely used to characterize the relation between insulin and other cardiovascular risk factors, it represents an integrated response to glucose metabolism influenced by secretion, distribution, and clearance of both insulin and glucose. Therefore it has been suggested that the euglycemic insulin clamp, conducted in vivo under controlled experimental conditions, is a better estimate of insulin resistance.22 EIC studies were performed in 357 11- to 14-yearold children enrolled in a longitudinal study to define the relation of insulin resistance during childhood to the development of hyperinsulinemia, hypertension, dyslipidemia, and obesity in adulthood. This report describes the relation of fatness and insulin resistance to the risk factors associated with the insulin resistance syndrome and compares fasting insulin with the EIC as a measure of insulin resistance in children.
RESEARCH DESIGN AND METHODS This study was approved by the Institutional Review Board Human Subjects Committee of the University of Minnesota. Consent was obtained from all children and their parents or guardians. The children participating in this study were recruited after blood pressure screening was performed on 12,043 fifth- through eighth-grade Minneapolis Public School students (3819 black, 4216 white; 6035 male, 6008 female), representing 93% of all eligible students in those grades. Recruitment letters were mailed to randomly selected black and non-Hispanic white children with stratification according to sex, race, and systolic blood
pressure percentile (half from the upper 25 percentiles and half from the lower 75 percentiles to enrich the study population with potentially higher risk children). Of 2915 students who received a recruitment letter, 537 attended an information meeting (held in groups of 20 to 30 children and their parents), where the study was explained in detail and appointments made for a clinic visit. After the information meeting, informed consent was obtained from 401 children. The screening blood pressure of this group did not differ from that of the randomly selected children choosing not to participate. Of the 401, 25 subsequently refused to attend the clinic, 2 were found to be ineligible for participation because of chronic illness, and 17 were unable to complete the EIC study because of technical difficulties with venipuncture and catheter placement in the Clinical Research Center. The remaining 357 children completed the EIC studies and form the cohort for this report. At the initial clinic visit the children underwent a complete physical examination including Tanner staging by a pediatrician. Anthropometric measurements were made with participants in study gowns and without shoes. Height was measured with a wallmounted stadiometer. Weight was determined by a balance scale. Triceps and subscapular skinfold thickness were measured twice to the nearest millimeter with Lange calipers, and the mean of the two measurements was used in the analyses. Waist and hip circumferences were measured to the nearest 0.5 cm. Blood pressure was measured twice on the right arm with a random-zero sphygmomanometer with subjects in the seated position; the averages of the two measurements (systolic and 5th phase Korotkoff diastolic) were used in the analyses. The EIC studies were conducted in the University of Minnesota Clinical Research Center. Participants were admitted after a 10-hour overnight
fast. An intravenous catheter was inserted into an arm vein 1 hour before the clamp studies were performed, and a blood sample was obtained immediately for measurement of serum lipids. This catheter was used for infusion of potassium phosphate, insulin, and glucose. A contralateral vein was cannulated for blood sampling, and the hand was placed in a heated box (65°C) to arterialize venous blood for measurement of glucose levels. Blood samples for serum insulin levels were obtained at baseline (–10, –5, and 0 minutes before the start of the insulin infusion) and at steady state during the clamp (+140, +160, +180 minutes). Plasma glucose was also measured at baseline (–10, –5, and 0 minutes) and every 5 minutes during the clamp. The insulin infusion was started at time 0 and continued at a rate of 1 mU/kg/min for 3 hours. An infusion of 20% glucose was started at time 0 and was adjusted, based on plasma glucose levels, to maintain euglycemia, that is, plasma glucose at 5.6 mmol/L. Insulin sensitivity was determined from the amount of glucose required to maintain euglycemia over the final 40 minutes of the EIC study and was expressed as Mlbm (ie, glucose use per kilogram lean body mass per minute), with lean body mass, or fat-free mass, calculated by the method of Slaughter et al.23 Mlbm was compared with other previously published methods for assessing insulin resistance, that is, fasting insulin, the homeostasis model,24 and the fasting insulin/fasting glucose ratio.25 Data from all participants were combined based on the relation of insulin sensitivity to Tanner stage as previously reported in this cohort.26 We have confirmed that insulin resistance increases significantly between Tanner stages 1 and 2, remains stable through Tanner stages 2, 3, and 4, and decreases significantly at Tanner stage 5. Despite these changes, the inverse relation between insulin sensitivity and body mass index was significant in each of the Tanner stages and was not 701
SINAIKO ET AL
significantly different among any of the 5 Tanner stages.26 Blood samples were analyzed for glucose immediately at the bedside with a Beckman Glucose Analyzer II (Beckman Instruments Inc, Fullerton, CA). The insulin samples were collected on ice and centrifuged within 20 minutes. Insulin levels were determined with a radioimmunoassay kit (Equate RIA, Binax Corp, Portland, ME). The average of the three baseline measurements was used in the analyses. Blood samples for serum lipids were analyzed in the University of Minnesota laboratory with a Cobas FARA. Cholesterol was determined by a standard enzymatic-cholesterol oxidase-based method; high-density lipoprotein-cholesterol was determined after precipitation of non-HDL lipoproteins with magnesium/dextran precipitating reagent. Triglycerides were determined with a standard glycerol blanked, enzymatic triglyceride method. Low-density lipoprotein-cholesterol was calculated by the Friedewald equation. Data are expressed as mean ± SEM. Analyses were performed with analysis of variance and Pearson correlation analysis, with adjustments made for sex, race, and Tanner stage as indicated. A P value <.05 was considered to be statistically significant.
RESULTS Despite similar body weight, boys participating in the EIC studies had a significantly greater height, LBM, waist, and waist-hip ratio than girls, but percent body fat, triceps and subscapular skinfold, and hip circumference were greater in the girls (Table I). Systolic blood pressure was significantly higher in the boys, and diastolic blood pressure was significantly higher in the girls. Blacks had a slightly but significantly greater Tanner score (P < .004), but there were no significant differences in anthropometric measures 702
THE JOURNAL OF PEDIATRICS NOVEMBER 2001 or blood pressure between blacks and whites. Mlbm and glucose were significantly higher in the boys than in the girls. Fasting insulin was higher in the girls, but the difference was not significant. Mlbm and fasting insulin were significantly correlated (r = –0.42, P = .0001), and this correlation was virtually identical for boys (r = –0.42, P = .0001) and girls (r = –0.41, P = .0001). There were no significant differences in lipids between boys and girls. Blacks had significantly higher Mlbm (P < .01), primarily because the black boys had the highest values of the 4 race-sex groups. Triglycerides were significantly lower (P = .0001) and HDL-C significantly higher (P = .0001) in blacks, but there were no significant differences between blacks and whites for the other laboratory data. BMI, LBM, and percent body fat were significantly correlated with SBP in boys and girls, although the relation with BMI and percent body fat was stronger (Table II). The correlation of BMI and percent body fat with all lipids was significant in the boys, but only HDL-C had a significant correlation with BMI in the girls. The correlation with LBM was not significant for any lipids in the boys, but LBM was significantly, although inversely, related to cholesterol and LDL-C in the girls. Because BMI is a measure of body size relative to height squared, analyses were done in which LBM and percent body fat were divided by height squared, and the results were not different (data not shown). The correlations of BMI, LBM, and body fat with SBP and lipids in the girls were virtually identical to those of the boys. The correlations in the blacks followed the same pattern and value for r as in the whites, but the statistical significance varied, probably because of the lower statistical power. There was a significant direct correlation between BMI and fasting insulin (r = 0.40, P = .0001) and a significant inverse correlation between BMI and
Mlbm (r = –0.26, P = .0001). Similar relations were found for LBM (fasting insulin: r = 0.21, P = .0001; Mlbm: r = –0.22, P = .0001). Partial correlation coefficients, holding race, sex, and Tanner stage constant, were similar (data not shown). There was a broad range of insulin sensitivity and body size. To further assess the influence of body fat on insulin sensitivity, the correlation coefficient of Mlbm with percent body fat (ie, [bw-lbm]/bw) was computed, with a result of r = –0.12, P < .02. Correlations with fasting insulin were significant for SBP in boys and girls, for lipids (except HDL-C) in the boys, and for triglycerides and HDL-C in the girls (Table III). Only triglycerides remained significantly related to fasting insulin in both sexes after adjustment was done for BMI, and relations to fasting insulin did not change when log insulin was substituted (data not shown). In contrast, the correlations with Mlbm were not significant for SBP and were significant only for triglycerides in both sexes and for HDL-C in the girls, although HDL-C in the boys was of borderline significance (P = .07). The relation of Mlbm to the risk factors did not change after adjustment was done for BMI. Mlbm was significantly correlated with triglycerides and HDL-C in the whites; although the correlation coefficients were similar in the blacks, they were not statistically significant. SBP, cholesterol, triglycerides, and HDL-C were significantly correlated with fasting insulin in the blacks, but only HDL-C was significantly correlated with fasting insulin in the whites. The glucose/insulin and homeostasis model measurements were highly correlated with fasting insulin (r = 0.99, P = .0001 and r = 0.99, P = .0001, respectively). Their correlations with Mlbm, body size, lipids, and blood pressure were virtually identical to that of fasting insulin (data not shown). To study the clustering effect of insulin resistance, body size, lipids, and
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS
VOLUME 139, NUMBER 5 Table I. Clinical and laboratory data (mean ± SEM) Male White N 148 Age (y) 13.1 ± 0.1 Tanner 3.1 ± 0.1 Height (cm) 164.6 ± 0.9 Weight (kg) 59.4 ± 1.2 21.9 ± 0.4 Body mass index (kg/m2) Lean body mass (kg) 42.9 ± 0.8 Body fat (%) 26.5 ± 1.0 Waist (cm) 80.0 ± 1.0 Hip (cm) 92.7 ± 0.9 Waist/hip ratio 0.86 ± 0.01 Triceps skinfold (mm) 20.2 ± 0.8 Subscapular skinfold (mm) 11.7 ± 0.6 SBP (mm Hg) 109 ± 1 DBP (mm Hg) 55 ± 1 12.9 ± 0.4 Mlbm(mg/kg/min) Fasting insulin (pmol/L) 84.6 ± 4.8 Steady-state insulin (pmol/L) 436 ± 11.0 Glucose (mmol/L) 5.62 ± 0.02 Cholesterol (mmol/L) 3.95 ± 0.06 Triglycerides (mmol/L) 1.00 ± 0.05 LDL-C (mmol/L) 2.3 ± 0.05 HDL-C (mmol/L) 1.10 ± 0.02
Female
Black
All
White
50 12.9 ± 0.2 3.5 ±0.2 162.6 ± 1.2 55.7 ± 2.1 20.9 ± 0.6 41.8 ± 1.1 22.6 ± 1.7 75.1 ± 1.6 89.5 ± 1.3 0.83 ± 0.01 20.6 ± 1.8 9.9 ± 0.6 108 ± 1 55 ± 2 14.3 ± 0.6 75.6 ± 8.4 471 ± 19 5.44 ± 0.04 3.76 ± 0.11 0.75 ± 0.08 2.1 ± 0.09 1.30 ± 0.03
198 13.1 ± 0.1 3.2 ± 0.1 164.1 ± 0.5 58.9 ± 0.9 21.6 ± 0.3 42.7 ± 0.4 25.5 ± 0.9 78.7 ± 0.8 92.3 ± 0.7 0.85 ± 0.01 20.3 ± 0.7 11.2 ± 0.5 109.0 ± 1.0 55 ± 1 13.1 ± 0.3 80 ± 4 450 ± 12 5.6 ± 0.03 3.89 ± 0.08 1.00 ± 0.04 2.27 ± 0.05 1.15 ± 0.02
136 12.9 ± 0.1 3.3 ± 0.1 160.6 ± 0.7 56.7 ± 1.3 22.1 ± 0.4 39.0 ± 0.5 30.6 ± 0.7 79.7 ± 1.0 95.1 ± 0.9 0.80 ± 0.01 24.4 ± 0.8 14.5 ± 0.6 106 ± 1 57 ± 1 11.5 ± 0.3 96 ± 4.8 516 ± 17 5.43 ± 0.03 3.90 ± 0.06 1.10 ± 0.05 2.30 ± 0.06 1.20 ± 0.02
Black 23 12.6 ± 0.3 4.0 ± 0.2 158.1 ± 2.0 59.4 ± 3.1 23.6 ± 0.9 38.6 ± 1.1 32.6 ± 2.1 76.3 ± 2.3 97.5 ± 2.5 0.78 ± 0.01 25.0 ± 2.0 17.3 ± 2.0 106 ± 2 60 ± 2 12.2 ± 0.8 88.8 ± 12 578 ± 45 5.38 ± 0.11 4.20 ± 0.16 0.89 ± 0.12 2.60 ± 0.14 1.20 ± 0.05
All 159 12.8 ± 0.1* 3.4 ± 0.1 159.3 ± 0.6† 57.0 ± 1.1 22.3 ± 0.3 38.9 ± 0.5† 30.9 ± 0.6† 76.4 ± 0.9* 95.4 ± 0.8 0.79 ± 0.01† 24.7 ± 0.7† 14.9 ± 0.6† 106 ± 1 58 ± 2 11.9 ± 0.3* 93 ± 5 528 ± 18* 5.4 ± 0.01† 4.00 ± 0.06 1.05 ± 0.05 2.34 ± 0.05 1.20 ± 0.02
Data adjusted by linear regression for race and Tanner score. * P < .008 cf all boys. † P = .0001 cf all boys.
Table II. Correlation of body mass index, lean body mass, and percent body fat with lipids and systolic blood pressure Male (n = 198)
Variable SBP Cholesterol Triglycerides HDL-C LDL-C
BMI r P r P r P r P r P
0.37 0.0001 0.35 0.0001 0.32 0.0001 –0.22 0.0018 0.38 0.0001
LBM 0.15 0.03 0.01 0.99 0.05 0.42 –0.11 0.11 0.02 0.81
Female (n = 159)
% Body fat 0.33 0.0001 0.35 0.0001 0.33 0.0001 –0.20 0.007 0.36 0.0001
BMI
LBM
0.27 0.0005 –0.02 0.83 0.13 0.09 –0.19 0.02 0.02 0.84
0.16 0.04 –0.16 0.05 0.06 0.43 –0.04 0.54 –0.18 0.02
% Body fat 0.29 0.0003 0.02 0.79 0.16 0.049 –0.13 0.11 0.02 0.84
Data adjusted by linear regression for race and Tanner score.
blood pressure, comparisons were made between the children in the lowest quartile of Mlbm (ie, highest degree of insulin resistance) and the children
in the highest quartile of Mlbm (ie, lowest degree of insulin resistance) (Tables IV and V). Despite similar ages, the boys in the low Mlbm quartile were
taller (P = .06) and heavier, with more lean and fat mass, larger waist and hip circumferences, and larger subscapular skinfold than the boys in the high 703
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS NOVEMBER 2001
Table III. Correlation of Mlbm and fasting insulin with lipids and systolic blood pressure
Male (n = 198) Variable Cholesterol Triglycerides HDL-C LDL-C SBP
r P r P r P r P r P
Female (n = 159)
Mlbm
Fasting insulin
Mlbm
Fasting insulin
–0.08 0.23 –0.26 0.0003 0.13 0.07 –0.05 0.48 –0.10 0.18
0.26 0.0003 0.38 0.0001 –0.09 0.19 0.19 0.0078 0.19 0.0093
0.01 0.83 –0.24 0.0027 0.32 0.0001 –0.01 0.87 –0.07 0.38
–0.07 0.38 0.23 0.0048 –0.21 0.0099 –0.09 0.23 0.21 0.0089
Data adjusted by linear regression for race and Tanner score.
Mlbm quartile. Compared with girls in the high Mlbm quartile, the girls in the low Mlbm quartile were younger, with similar height and lean body mass; however, they were heavier and had more fat mass and higher levels of all other fatness measures. The girls and boys in the low Mlbm quartile had significantly higher serum insulin, glucose, and triglycerides. They also had a lower HDL-C (P = .07 in boys). The mean SBP was higher in the boys and girls in the low Mlbm quartile, but these differences were not statistically significant (P = .07 in girls). Differences between whites in the high and low Mlbm quartiles followed this same pattern. The numbers of blacks in the high and low quartiles were too small for meaningful comparisons.
DISCUSSION By using the EIC to define insulin sensitivity, we were able to more precisely show the significant interrelations among the risk factors associated with the insulin resistance syndrome in children and the important influence of body fatness on the other currently used methods to assess insulin resistance. The EIC provides a measure of insulin resistance under controlled, 704
standardized conditions of insulin and glucose infusion, whereas fasting insulin and other measures represent an integrated and uncontrolled response to insulin and glucose metabolism.22 Moreover, normalizing insulin sensitivity to lean body mass (Mlbm), thus removing fat mass from the estimate, should provide a more accurate assessment of insulin resistance,27 because approximately 75% of glucose use is localized to skeletal muscle and only 4% or less to fat tissue.28 The initial analyses showed that fasting insulin is significantly related to SBP in both sexes; all lipids, except HDL-C in boys; and triglycerides and HDL-C in girls. Previous studies in children have used plasma insulin levels to determine insulin resistance, with cross-sectional studies showing a significant correlation of fasting insulin with lipids and blood pressure.5-10 Fasting insulin remained significantly related to blood pressure7-10 and lipids5,7,8 in most of those studies after adjustment was done for BMI or skinfold thickness. However, in the Bogalusa study intraclass correlations (insulin resistance index, triglycerides, HDL-C, and mean arterial pressure) fell by approximately 50% across all age groups after adjustment was done for BMI,29 and in this study fasting in-
sulin was significantly related only to serum triglycerides, after adjustment was done for BMI. Thus the relation of fasting insulin to other risk factors appears to be strongly influenced by body fatness. Substituting the homeostasis model or glucose/insulin for fasting insulin gave virtually identical results, and neither provided a better estimate of insulin resistance than fasting insulin. In contrast, Mlbm in this study was significantly related only to triglycerides and HDL-C, and this relation remained significant after adjustment was done for BMI. Similar differences between fasting insulin and glucose use previously have been shown in adults with either the minimal model30 or the EIC.31 An independent effect of insulin resistance on cardiovascular risk in children has also been suggested from other studies. Fasting insulin levels in 6- to 9-year-old children in the Cardiovascular Risk in Young Finns study predicted the children’s level of blood pressure at age 9 to 15 years,10 and in 5- to 9-year-old Pima Indian children, fasting insulin predicted the level of weight gain during the subsequent 9 years of childhood.32 The Bogalusa Heart Study has shown a strong relation over an 8-year period of observation between persistently high fasting insulin levels and the development of risk factors in children and young adults.13 There were significant differences between the boys and girls in the degree of insulin resistance. Mlbm was significantly higher in the boys, suggesting a higher degree of insulin resistance in the girls. This result may be due partly to a greater degree of adiposity in the girls,33 as represented by a significantly greater triceps and subscapular skinfold thickness and percent body fat. It is also supported by the finding that the sex difference in insulin resistance was not present in the very obese (BMI >27 kg/m2) boys and girls in whom the degree of adiposity was similar.26 The higher fasting
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS
VOLUME 139, NUMBER 5 Table IV. Comparison of clinical data between children in the highest and lowest quartiles of Mlbm
Male Variable
Low Mlbm High Mlbm (High ins resist) (Low ins resist)
N Age (y) Height (cm) Weight (kg) BMI (kg/m2) LBM (kg) Body fat (%) Waist (cm) Hip (cm) WHR Triceps (mm) Subscapular (mm) SBP
49 13.0 ± 0.1 163.8 ± 1.2 62.9 ± 2.0 23.1 ± 0.7 44.1 ± 0.9 26.5 ± 1.8 83.4 ± 1.6 93.8 ± 1.5 0.89 ± 0.01 18.8 ± 1.4 13.5 ± 1.0 110.1 ± 1.5
49 12.8 ± 0.1 160.4 ± 1.2 53.9 ± 2.0 20.6 ± 0.7 39.3 ± 0.9 25.8 ± 1.8 74.4 ± 1.6 89.6 ± 1.5 0.83 ± 0.01 20.6 ± 1.4 10.1 ± 1.0 108.7 ± 1.5
Female P .17 .06 .0042 .0124 .0007 .79 .0003 .055 .0007 .39 .0265 .49
Low Mlbm High Mlbm (High ins resist) (Low ins resist) 38 12.6 ± 0.1 159.6 ± 1.0 64.6 ± 2.2 25.1 ± 0.8 40.0 ± 0.9 35.6 ± 1.2 81.6 ± 1.8 99.7 ± 1.7 0.82 ± 0.02 29.4 ± 1.2 18.5 ± 1.1 108.9 ± 1.6
40 13.0 ± 0.1 159.9 ± 1.0 54.5 ± 2.2 21.2 ± 0.8 38.2 ± 0.8 28.8 ± 1.2 72.9 ± 1.8 91.7 ± 1.7 0.80 ± 0.02 23.0 ± 1.2 12.5 ± 1.1 104.6 ± 1.6
P .0236 .87 .0022 .0008 .16 .0003 .0011 .0024 .47 .0007 .0005 .070
Data adjusted by linear regression for race and Tanner score. WHR, Waist-hip ratio.
Table V. Comparison of laboratory data between children in the highest and lowest quartiles of Mlbm
Male Variable
Low Mlbm High Mlbm (High ins resist) (Low ins resist)
N Mlbm (mg/kg/min) Fast insulin (pmol/L) Glucose (mmol/L) Cholesterol (mmol/L) Triglycerides (mmol/L) HDL-C (mmol/L) LDL-C (mmol/L)
49 7.9 ± 0.3 125 ± 10 5.7 ± 0.1 4.00 ± 0.14 1.24 ± 0.09 1.12 ± 0.04 2.34 ± 0.12
49 19.0 ± 0.3 54 ± 10 5.4 ± 0.1 3.86 ± 0.14 0.79 ± 0.09 1.22 ± 0.04 2.27 ± 0.12
Female P .0001 .0001 .0005 .40 .0013 .074 .70
Low Mlbm High Mlbm (High ins resist) (Low ins resist) 38 7.0 ± 0.4 135 ± 10 5.5 ± 0.1 3.94 ± 0.11 1.33 ± 0.11 1.06 ± 0.03 2.27 ± 0.10
40 16.7 ± 0.3 70 ± 10 5.3 ± 0.1 4.00 ± 0.11 0.92 ± 0.11 1.29 ± 0.03 2.29 ± 0.09
P .0001 .0001 .0193 .67 .0119 .0001 .90
Data adjusted by linear regression for race and Tanner score.
insulin level in the girls is also consistent with a higher degree of insulin resistance, although it is not clear why this difference did not reach statistical significance. As noted earlier, there are substantial differences between the insulin clamp and fasting insulin as estimates of insulin resistance, and the correlation coefficient between the two is only 0.42. Moreover, others have reported differences between the relations of fasting insulin and the EIC to risk factors.30,31 It is interesting that despite the significantly lower degree of insulin resistance in the boys, BMI had a highly significant positive corre-
lation with cholesterol, triglycerides, and LDL-C and a significant negative correlation with HDL-C, in contrast to the girls. Although an explanation for this result is not immediately apparent, it suggests that obesity has an early adverse effect on boys that may lead to the increased degree of cardiovascular risk noted in men compared with women. An important finding in this study is that children in the lowest quartile for Mlbm (ie, highest degree of insulin resistance) had anthropometric measurements and laboratory measurements that were consistent with the insulin
resistance syndrome compared with children in the upper quartile for Mlbm. A clustering effect of risk factors at this early age also has been suggested in studies with fasting insulin.12 A wider separation of the cluster effect between the children with high versus low levels of insulin resistance would be expected as they mature into adulthood, because of the well-recognized tracking effects18-20 for the risk factors between childhood and adulthood. These results may have important implications for public health policy. First, they confirm the important influence of obesity early in life. Obesity is 705
SINAIKO ET AL
not only recognized as a current major problem within pediatrics,34 but the prevalence of overweight in children also is increasing35,36 and is accompanied by an increasing prevalence of type 2 diabetes.37 Epidemiologic studies have shown that obesity in childhood predicts adult obesity38 and is associated with an increased risk of coronary heart disease,39 and the rate of weight gain during childhood and adolescence predicts levels of cardiovascular risk in young adults.40 Thus strategies linked to weight control and the reduction of premature cardiovascular morbidity and mortality might be most effective if introduced early in life. Second, the identification of an independent effect of insulin resistance early in life suggests that additional efforts should be made to understand basic and clinical interactions among risk factors before adulthood. Although long-term studies relating childhood insulin resistance to the incidence of risk factors, morbidity, and mortality in adults will help clarify these issues, there is sufficient evidence to propose that the etiologic mechanisms controlling insulin resistance are present in childhood and should be studied in this age group.
REFERENCES 1. DeFronzo RA, Ferrannini E. Insulin resistance. A multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia and atherosclerotic cardiovascular disease. Diabetes Care 1991;14:173-94. 2. Haffner SM, Miettinen H. Insulin resistance implications for type II diabetes mellitus and coronary heart disease. Am J Med 1997;103:152-62. 3. Ginsberg HN. Insulin resistance and cardiovascular disease. J Clin Invest 2000;106:453-8. 4. Meigs JB, D’Agostino RB, Sr, Wilson PWF, Cupples A, Nathan DM, Singer DE. Risk variable clustering in the insulin resistance syndrome. The Framingham Offspring Study. Diabetes 1997;46:1594-600. 5. Jiang X, Srinivasan SR, Webber LS, Wattigney WA, Berenson GS. Associa706
THE JOURNAL OF PEDIATRICS NOVEMBER 2001
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
tion of fasting insulin level with serum lipid and lipoprotein levels in children, adolescents, and young adults: The Bogalusa Heart Study. Arch Intern Med 1995;155:190-6. Steinberger J, Moorehead C, Katch V, Rocchini AP. Relationship between insulin resistance and abnormal lipid profile in obese adolescents. J Pediatr 1995;126:690-5. Raitakari OT, Porkka KVK, Ronnemaa T, Knip M, Uhari M, Akerblom HK, et al. The role of insulin in clustering of serum lipids and blood pressure in children and adolescents. Diabetologia 1995;38:1042-50. Sinaiko AR, Gomez-Marin O, Prineas RJ. Relation of fasting insulin to blood pressure and lipids in adolescents and parents. Hypertension 1997;30:1554-9. Jiang X, Srinivasan S, Bao W, Berenson G. Association of fasting insulin with blood pressure in young individuals: The Bogalusa Heart Study. Arch Intern Med 1993;153:323-8. Taittonen L, Uhari M, Turtinen J, Pokka T, Akerblom HK. Insulin and blood pressure among healthy children. Cardiovascular risk in young Finns. Am J Hypertens 1996;9:193-9. Rocchini AP, Katch V, Schork A, Kelch RP. Insulin and blood pressure during weight loss in obese adolescents. Hypertension 1987;10:267-73. Chen W, Srinivasan AR, Elkasabany A, Berenson GS. Cardiovascular risk factors clustering features of insulin resistance syndrome (Syndrome X) in a biracial (black-white) population of children, adolescents, and young adults: The Bogalusa Heart Study. Am J Epidemiol 1999;150:667-74. Bao W, Srinivasan SR, Berenson GS. Persistent elevation of plasma insulin levels is associated with increased cardiovascular risk in children and young adults: The Bogalusa Heart Study. Circulation 1996;93:54-9. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: The Bogalusa Heart Study. Pediatrics 1999;103:1175-82. Daniels SR, Morrison JA, Sprecher DL, Khoury P, Kimball TR. Association of body fat distribution and cardiovascular risk foctors in children and adolescents. Circulation 1999;99:541-5. Morrison JA, Barton BA, Biro FM, Daniels SR, Sprecher DL. Overweight, fat patterning, and cardiovascular disease risk factors in black and white boys. J Pediatr 1999;135:451-7.
17. Morrison JA, Sprecher DL, Barton BA, Waclawiw MA, Daniels SR. Overweight, fat patterning, and cardiovascular disease risk factors in black and white girls: The National Heart, Lung, and Blood Institute Growth and Health Study. J Pediatr 1999;135:458-64. 18. Bao W, Threefoot SA, Srinivasan SR, Berenson GS. Essential hypertension predicted by tracking of elevated blood pressure from childhood to adulthood: The Bogalusa Heart Study. Am J Hypertens 1995;8:657-65. 19. National Cholesterol Education Program. Report of the expert panel on blood cholesterol levels in children and adolescents. Pediatrics 1992;89:525-83. 20. Clarke WR, Lauer RM. Does childhood obesity track into adulthood? Crit Rev Food Sci Nutr 1993;33:423-30. 21. Berenson GS, Srinivasan SR, Bao W, Newman WP, Tracy RE, Wattigney WA. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. N Engl J Med 1998;338:1650-6. 22. Ferrannini E, Mari A. How to measure insulin sensitivity. J Hypertens 1998; 16:895-906. 23. Slaughter MH, Lohman TG, Baileau RA, Horswill CA, Stillman RJ, Van Loan MD, et al. Skinfold equations for estimation of body fatness in children and youth. Hum Biol 1988;60:709-23. 24. Lansang MC, Williams GH, Carroll JS. Correlation between the glucose clamp technique and the homeostasis model assessment in hypertension. Am J Hypertens 2001;14:51-3. 25. Caro JF. Insulin resistance in obese and nonobese man. J Clin Endocrinol Metab 1991;73:691-5. 26. Moran A, Jacobs DR, Steinberger J, Hong C-P, Prineas R, Luepker R, et al. Insulin resistance during puberty: results from clamp studies in 357 children. Diabetes 1999;48:2039-44. 27. Ferrannini E, Natali A, Bell P, CavalloPerin P, Lalic N, Mingrone G. Insulin resistance and hypersecretion in obesity. J Clin Invest 1997;100:1166-73. 28. DeFronzo RA, Bonadonna RC, Ferrannini E. Pathogenesis of NIDDM: a balanced overview. Diabetes Care 1992;15:318-68. 29. Chen W, Bao W, Begum S, Elkasabany A, Srinivasan SR, Berenson GS. Agerelated patterns of the clustering of cardiovascular risk variables of syndrome X from childhood to young adulthood in a population made up of black and white subjects: The Bogalusa Heart Study. Diabetes 2000;49:1042-8.
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS
VOLUME 139, NUMBER 5 30. Istfam NW, Plaisted CS, Bistrian BR, Blackburn GL. Insulin resistance versus insulin secretion in the hypertension of obesity. Hypertension 1992; 19:385-92. 31. Ferrannini E, Natali A, Capaldo B, Lehtovirta M, Jacob S, Yki-Jahrvinen H. Insulin resistance, hyperinsulinemia, and blood pressure: role of age and obesity. Hypertension 1997; 30:1144-9. 32. Odeleye OE, de Courten M, Pettitt DJ, Ravussin E. Fasting hyperinsulinemia is a predictor of increased body weight gain and obesity in Pima Indian children. Diabetes 1997;46:1341-5. 33. Daniels SR, Khoury PR, Morrison JA. The utility of body mass index as a measure of body fatness in chil-
dren and adolescents: differences by race and gender. Pediatrics 1997; 99:804-7. 34. Christoffel KK, Aroza A. The epidemiology of overweight in children: relevance for clinical care. Pediatrics 1998; 101:103-4. 35. Troiano RP, Flegal KM, Kuczmarski RJ, Campbell SM, Johnson CL. Overweight prevalance in trends for children and adolescents. The National Health and Nutrition Examination Surveys, 1963-1991. Arch Pediatr Adolesc Med 1995;149:1085-91. 36. Luepker R, Prineas R, Jacobs D, Sinaiko A. Secular trends of blood pressure and body size in a multiethnic adolescent population: 19861996. J Pediatr 1999;134:668-74.
37. Rosenbloom AL, Roe J, Young RS, Winter WE. Emerging epidemic of type 2 diabetes in youth. Diabetes Care 1999;22:345-54. 38. Whitaker RD, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med 1997;337:869-73. 39. Must A, Jacques PF, Dallal GE, Bajema CJ, Dietz WH. Long-term morbidity and mortality of overweight adolescents. N Engl J Med 1992;327:1350-5. 40. Sinaiko AR, Donahue RP, Jacobs, DR, Prineas RJ. Relation of rate of growth during childhood and adolescence to fasting insulin, lipids, and systolic blood pressure in young adults. Circulation 1999;99:1471-6.
50 Years Ago in The Journal of Pediatrics GROWTH AND DEVELOPMENT OF INFANTS RECEIVING A PROPRIETARY PREPARATION OF D Frost LH, Jackson RL. J Pediatr 1951;39:585-92
EVAPORATED MILK WITH DEXTRI-MALTOSE AND VITAMIN
To a large extent, the history of pediatrics as a medical specialty is derived from the history of infant feeding. The devastating mortality rate of children who received artificial formulas (ie, nonhuman milk) attracted the attention of the founders of our profession. The response to this was the development of recipes and schedules of nearly incomprehensible complexity. A brief perusal of the feeding section of any early 20th century pediatric text will glaze most contemporary eyes. Infant feeding became easier in the 1950s. A simple recipe for diluting evaporated milk, with added carbohydrate and supplemental vitamins, became popular. In this study, Frost and Jackson provide the practitioner with simple “how to” advice on the preparation of this kind of formula. Among the many differences from practice of the previous decades was the realization that opened cans of evaporated milk could be kept safely in the refrigerator and bottles prepared as needed; complex sterilization procedures were not necessary. What makes this article particularly noteworthy is that it goes beyond a mere description to become a clinical study. Fifty-seven infants were fed according to the method described and subjected to careful serial follow-up for an average of 6 months. Gastrointestinal function (stooling, primarily) was recorded, and serial assessment of growth was reported to compare favorably with the contemporary Iowa growth standards. Vitamin D status was assessed (indirectly) by bone films in most infants, and hemoglobin concentrations were also measured. Frost and Jackson nicely provided the pediatrician of a half-century ago with practical, safe, and sound advice in formula feeding. Unfortunately, they did not consider the more basic question: why go through the trouble of preparing any formula when a free, safe, and physiologically optimal preparation (human milk) is available to virtually any mother? Thomas R. Welch, MD Chair, Department of Pediatrics University Hospital Syracuse, NY 13210 9/37/120092 doi:10.1067/mpd.2001.120092
707
From the Department of Pediatrics, Division of Epidemiology, School of Public Health, University of Minnesota Medical School, Minneapolis, Minnesota, the Department of Public Health Sciences, Wake Forest School of Medicine, WinstonSalem, North Carolina, and the Department of Pediatrics, University of Michigan, Ann Arbor.
Supported by grants HL 52851 and M01 RR00400 from the National Institutes of Health. Submitted for publication Nov 15, 2000; revisions received March 1, 2001, June 21, 2001; accepted July 2, 2001. Reprint requests: Alan R. Sinaiko, MD, University of Minnesota Medical School, 420 Delaware St, S.E., Box 491 UMHC, Minneapolis, MN 55455. Copyright © 2001 by Mosby, Inc. 0022-3476/2001/$35.00 + 0 9/21/118535 doi:10.1067/mpd.2001.118535
700
The insulin resistance syndrome, consisting of hyperinsulinemia, dyslipidemia, hypertension, and obesity, is associated with non-insulin dependent (type 2) diabetes and atherosclerotic cardiovascular disease in the adult population.1-3 The relation among these factors is complex,4 and it is not clear whether insulin resistance occurs in response to the pathophysiologic changes associated with the diseases or is etiologically related to their development. Studies in children have shown a significant relation among fasting insulin and lipids,5-8 blood pressure,7-10 and weight.11 This relation appears to follow a clustering pattern12 that tracks from childhood to young adulthood.13 Cross-sectional studies in children have confirmed a strong positive relation among overweight and fasting inBMI EIC HDL-C LBM LDL-C SBP
Body mass index Euglycemic insulin clamp High-density lipoprotein-cholesterol Lean body mass Low-density lipoprotein-cholesterol Systolic blood pressure
sulin, lipids, and blood pressure14 and, more specifically, significant correlations have been shown between body fat distribution and cardiovascular risk factors.15-17 When these findings are considered in the context of the significant tracking effect of blood pressure,18 lipids,19 and obesity20 between childhood and adulthood and the correlation of blood pressure and lipids21 with the extent of aortic and coronary
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS
VOLUME 139, NUMBER 5 atherosclerosis at autopsy in children, it is reasonable to suggest that the interaction among insulin, insulin resistance, and other risk factors begins during the first 2 decades of life. Although fasting insulin has been widely used to characterize the relation between insulin and other cardiovascular risk factors, it represents an integrated response to glucose metabolism influenced by secretion, distribution, and clearance of both insulin and glucose. Therefore it has been suggested that the euglycemic insulin clamp, conducted in vivo under controlled experimental conditions, is a better estimate of insulin resistance.22 EIC studies were performed in 357 11- to 14-yearold children enrolled in a longitudinal study to define the relation of insulin resistance during childhood to the development of hyperinsulinemia, hypertension, dyslipidemia, and obesity in adulthood. This report describes the relation of fatness and insulin resistance to the risk factors associated with the insulin resistance syndrome and compares fasting insulin with the EIC as a measure of insulin resistance in children.
RESEARCH DESIGN AND METHODS This study was approved by the Institutional Review Board Human Subjects Committee of the University of Minnesota. Consent was obtained from all children and their parents or guardians. The children participating in this study were recruited after blood pressure screening was performed on 12,043 fifth- through eighth-grade Minneapolis Public School students (3819 black, 4216 white; 6035 male, 6008 female), representing 93% of all eligible students in those grades. Recruitment letters were mailed to randomly selected black and non-Hispanic white children with stratification according to sex, race, and systolic blood
pressure percentile (half from the upper 25 percentiles and half from the lower 75 percentiles to enrich the study population with potentially higher risk children). Of 2915 students who received a recruitment letter, 537 attended an information meeting (held in groups of 20 to 30 children and their parents), where the study was explained in detail and appointments made for a clinic visit. After the information meeting, informed consent was obtained from 401 children. The screening blood pressure of this group did not differ from that of the randomly selected children choosing not to participate. Of the 401, 25 subsequently refused to attend the clinic, 2 were found to be ineligible for participation because of chronic illness, and 17 were unable to complete the EIC study because of technical difficulties with venipuncture and catheter placement in the Clinical Research Center. The remaining 357 children completed the EIC studies and form the cohort for this report. At the initial clinic visit the children underwent a complete physical examination including Tanner staging by a pediatrician. Anthropometric measurements were made with participants in study gowns and without shoes. Height was measured with a wallmounted stadiometer. Weight was determined by a balance scale. Triceps and subscapular skinfold thickness were measured twice to the nearest millimeter with Lange calipers, and the mean of the two measurements was used in the analyses. Waist and hip circumferences were measured to the nearest 0.5 cm. Blood pressure was measured twice on the right arm with a random-zero sphygmomanometer with subjects in the seated position; the averages of the two measurements (systolic and 5th phase Korotkoff diastolic) were used in the analyses. The EIC studies were conducted in the University of Minnesota Clinical Research Center. Participants were admitted after a 10-hour overnight
fast. An intravenous catheter was inserted into an arm vein 1 hour before the clamp studies were performed, and a blood sample was obtained immediately for measurement of serum lipids. This catheter was used for infusion of potassium phosphate, insulin, and glucose. A contralateral vein was cannulated for blood sampling, and the hand was placed in a heated box (65°C) to arterialize venous blood for measurement of glucose levels. Blood samples for serum insulin levels were obtained at baseline (–10, –5, and 0 minutes before the start of the insulin infusion) and at steady state during the clamp (+140, +160, +180 minutes). Plasma glucose was also measured at baseline (–10, –5, and 0 minutes) and every 5 minutes during the clamp. The insulin infusion was started at time 0 and continued at a rate of 1 mU/kg/min for 3 hours. An infusion of 20% glucose was started at time 0 and was adjusted, based on plasma glucose levels, to maintain euglycemia, that is, plasma glucose at 5.6 mmol/L. Insulin sensitivity was determined from the amount of glucose required to maintain euglycemia over the final 40 minutes of the EIC study and was expressed as Mlbm (ie, glucose use per kilogram lean body mass per minute), with lean body mass, or fat-free mass, calculated by the method of Slaughter et al.23 Mlbm was compared with other previously published methods for assessing insulin resistance, that is, fasting insulin, the homeostasis model,24 and the fasting insulin/fasting glucose ratio.25 Data from all participants were combined based on the relation of insulin sensitivity to Tanner stage as previously reported in this cohort.26 We have confirmed that insulin resistance increases significantly between Tanner stages 1 and 2, remains stable through Tanner stages 2, 3, and 4, and decreases significantly at Tanner stage 5. Despite these changes, the inverse relation between insulin sensitivity and body mass index was significant in each of the Tanner stages and was not 701
SINAIKO ET AL
significantly different among any of the 5 Tanner stages.26 Blood samples were analyzed for glucose immediately at the bedside with a Beckman Glucose Analyzer II (Beckman Instruments Inc, Fullerton, CA). The insulin samples were collected on ice and centrifuged within 20 minutes. Insulin levels were determined with a radioimmunoassay kit (Equate RIA, Binax Corp, Portland, ME). The average of the three baseline measurements was used in the analyses. Blood samples for serum lipids were analyzed in the University of Minnesota laboratory with a Cobas FARA. Cholesterol was determined by a standard enzymatic-cholesterol oxidase-based method; high-density lipoprotein-cholesterol was determined after precipitation of non-HDL lipoproteins with magnesium/dextran precipitating reagent. Triglycerides were determined with a standard glycerol blanked, enzymatic triglyceride method. Low-density lipoprotein-cholesterol was calculated by the Friedewald equation. Data are expressed as mean ± SEM. Analyses were performed with analysis of variance and Pearson correlation analysis, with adjustments made for sex, race, and Tanner stage as indicated. A P value <.05 was considered to be statistically significant.
RESULTS Despite similar body weight, boys participating in the EIC studies had a significantly greater height, LBM, waist, and waist-hip ratio than girls, but percent body fat, triceps and subscapular skinfold, and hip circumference were greater in the girls (Table I). Systolic blood pressure was significantly higher in the boys, and diastolic blood pressure was significantly higher in the girls. Blacks had a slightly but significantly greater Tanner score (P < .004), but there were no significant differences in anthropometric measures 702
THE JOURNAL OF PEDIATRICS NOVEMBER 2001 or blood pressure between blacks and whites. Mlbm and glucose were significantly higher in the boys than in the girls. Fasting insulin was higher in the girls, but the difference was not significant. Mlbm and fasting insulin were significantly correlated (r = –0.42, P = .0001), and this correlation was virtually identical for boys (r = –0.42, P = .0001) and girls (r = –0.41, P = .0001). There were no significant differences in lipids between boys and girls. Blacks had significantly higher Mlbm (P < .01), primarily because the black boys had the highest values of the 4 race-sex groups. Triglycerides were significantly lower (P = .0001) and HDL-C significantly higher (P = .0001) in blacks, but there were no significant differences between blacks and whites for the other laboratory data. BMI, LBM, and percent body fat were significantly correlated with SBP in boys and girls, although the relation with BMI and percent body fat was stronger (Table II). The correlation of BMI and percent body fat with all lipids was significant in the boys, but only HDL-C had a significant correlation with BMI in the girls. The correlation with LBM was not significant for any lipids in the boys, but LBM was significantly, although inversely, related to cholesterol and LDL-C in the girls. Because BMI is a measure of body size relative to height squared, analyses were done in which LBM and percent body fat were divided by height squared, and the results were not different (data not shown). The correlations of BMI, LBM, and body fat with SBP and lipids in the girls were virtually identical to those of the boys. The correlations in the blacks followed the same pattern and value for r as in the whites, but the statistical significance varied, probably because of the lower statistical power. There was a significant direct correlation between BMI and fasting insulin (r = 0.40, P = .0001) and a significant inverse correlation between BMI and
Mlbm (r = –0.26, P = .0001). Similar relations were found for LBM (fasting insulin: r = 0.21, P = .0001; Mlbm: r = –0.22, P = .0001). Partial correlation coefficients, holding race, sex, and Tanner stage constant, were similar (data not shown). There was a broad range of insulin sensitivity and body size. To further assess the influence of body fat on insulin sensitivity, the correlation coefficient of Mlbm with percent body fat (ie, [bw-lbm]/bw) was computed, with a result of r = –0.12, P < .02. Correlations with fasting insulin were significant for SBP in boys and girls, for lipids (except HDL-C) in the boys, and for triglycerides and HDL-C in the girls (Table III). Only triglycerides remained significantly related to fasting insulin in both sexes after adjustment was done for BMI, and relations to fasting insulin did not change when log insulin was substituted (data not shown). In contrast, the correlations with Mlbm were not significant for SBP and were significant only for triglycerides in both sexes and for HDL-C in the girls, although HDL-C in the boys was of borderline significance (P = .07). The relation of Mlbm to the risk factors did not change after adjustment was done for BMI. Mlbm was significantly correlated with triglycerides and HDL-C in the whites; although the correlation coefficients were similar in the blacks, they were not statistically significant. SBP, cholesterol, triglycerides, and HDL-C were significantly correlated with fasting insulin in the blacks, but only HDL-C was significantly correlated with fasting insulin in the whites. The glucose/insulin and homeostasis model measurements were highly correlated with fasting insulin (r = 0.99, P = .0001 and r = 0.99, P = .0001, respectively). Their correlations with Mlbm, body size, lipids, and blood pressure were virtually identical to that of fasting insulin (data not shown). To study the clustering effect of insulin resistance, body size, lipids, and
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS
VOLUME 139, NUMBER 5 Table I. Clinical and laboratory data (mean ± SEM) Male White N 148 Age (y) 13.1 ± 0.1 Tanner 3.1 ± 0.1 Height (cm) 164.6 ± 0.9 Weight (kg) 59.4 ± 1.2 21.9 ± 0.4 Body mass index (kg/m2) Lean body mass (kg) 42.9 ± 0.8 Body fat (%) 26.5 ± 1.0 Waist (cm) 80.0 ± 1.0 Hip (cm) 92.7 ± 0.9 Waist/hip ratio 0.86 ± 0.01 Triceps skinfold (mm) 20.2 ± 0.8 Subscapular skinfold (mm) 11.7 ± 0.6 SBP (mm Hg) 109 ± 1 DBP (mm Hg) 55 ± 1 12.9 ± 0.4 Mlbm(mg/kg/min) Fasting insulin (pmol/L) 84.6 ± 4.8 Steady-state insulin (pmol/L) 436 ± 11.0 Glucose (mmol/L) 5.62 ± 0.02 Cholesterol (mmol/L) 3.95 ± 0.06 Triglycerides (mmol/L) 1.00 ± 0.05 LDL-C (mmol/L) 2.3 ± 0.05 HDL-C (mmol/L) 1.10 ± 0.02
Female
Black
All
White
50 12.9 ± 0.2 3.5 ±0.2 162.6 ± 1.2 55.7 ± 2.1 20.9 ± 0.6 41.8 ± 1.1 22.6 ± 1.7 75.1 ± 1.6 89.5 ± 1.3 0.83 ± 0.01 20.6 ± 1.8 9.9 ± 0.6 108 ± 1 55 ± 2 14.3 ± 0.6 75.6 ± 8.4 471 ± 19 5.44 ± 0.04 3.76 ± 0.11 0.75 ± 0.08 2.1 ± 0.09 1.30 ± 0.03
198 13.1 ± 0.1 3.2 ± 0.1 164.1 ± 0.5 58.9 ± 0.9 21.6 ± 0.3 42.7 ± 0.4 25.5 ± 0.9 78.7 ± 0.8 92.3 ± 0.7 0.85 ± 0.01 20.3 ± 0.7 11.2 ± 0.5 109.0 ± 1.0 55 ± 1 13.1 ± 0.3 80 ± 4 450 ± 12 5.6 ± 0.03 3.89 ± 0.08 1.00 ± 0.04 2.27 ± 0.05 1.15 ± 0.02
136 12.9 ± 0.1 3.3 ± 0.1 160.6 ± 0.7 56.7 ± 1.3 22.1 ± 0.4 39.0 ± 0.5 30.6 ± 0.7 79.7 ± 1.0 95.1 ± 0.9 0.80 ± 0.01 24.4 ± 0.8 14.5 ± 0.6 106 ± 1 57 ± 1 11.5 ± 0.3 96 ± 4.8 516 ± 17 5.43 ± 0.03 3.90 ± 0.06 1.10 ± 0.05 2.30 ± 0.06 1.20 ± 0.02
Black 23 12.6 ± 0.3 4.0 ± 0.2 158.1 ± 2.0 59.4 ± 3.1 23.6 ± 0.9 38.6 ± 1.1 32.6 ± 2.1 76.3 ± 2.3 97.5 ± 2.5 0.78 ± 0.01 25.0 ± 2.0 17.3 ± 2.0 106 ± 2 60 ± 2 12.2 ± 0.8 88.8 ± 12 578 ± 45 5.38 ± 0.11 4.20 ± 0.16 0.89 ± 0.12 2.60 ± 0.14 1.20 ± 0.05
All 159 12.8 ± 0.1* 3.4 ± 0.1 159.3 ± 0.6† 57.0 ± 1.1 22.3 ± 0.3 38.9 ± 0.5† 30.9 ± 0.6† 76.4 ± 0.9* 95.4 ± 0.8 0.79 ± 0.01† 24.7 ± 0.7† 14.9 ± 0.6† 106 ± 1 58 ± 2 11.9 ± 0.3* 93 ± 5 528 ± 18* 5.4 ± 0.01† 4.00 ± 0.06 1.05 ± 0.05 2.34 ± 0.05 1.20 ± 0.02
Data adjusted by linear regression for race and Tanner score. * P < .008 cf all boys. † P = .0001 cf all boys.
Table II. Correlation of body mass index, lean body mass, and percent body fat with lipids and systolic blood pressure Male (n = 198)
Variable SBP Cholesterol Triglycerides HDL-C LDL-C
BMI r P r P r P r P r P
0.37 0.0001 0.35 0.0001 0.32 0.0001 –0.22 0.0018 0.38 0.0001
LBM 0.15 0.03 0.01 0.99 0.05 0.42 –0.11 0.11 0.02 0.81
Female (n = 159)
% Body fat 0.33 0.0001 0.35 0.0001 0.33 0.0001 –0.20 0.007 0.36 0.0001
BMI
LBM
0.27 0.0005 –0.02 0.83 0.13 0.09 –0.19 0.02 0.02 0.84
0.16 0.04 –0.16 0.05 0.06 0.43 –0.04 0.54 –0.18 0.02
% Body fat 0.29 0.0003 0.02 0.79 0.16 0.049 –0.13 0.11 0.02 0.84
Data adjusted by linear regression for race and Tanner score.
blood pressure, comparisons were made between the children in the lowest quartile of Mlbm (ie, highest degree of insulin resistance) and the children
in the highest quartile of Mlbm (ie, lowest degree of insulin resistance) (Tables IV and V). Despite similar ages, the boys in the low Mlbm quartile were
taller (P = .06) and heavier, with more lean and fat mass, larger waist and hip circumferences, and larger subscapular skinfold than the boys in the high 703
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS NOVEMBER 2001
Table III. Correlation of Mlbm and fasting insulin with lipids and systolic blood pressure
Male (n = 198) Variable Cholesterol Triglycerides HDL-C LDL-C SBP
r P r P r P r P r P
Female (n = 159)
Mlbm
Fasting insulin
Mlbm
Fasting insulin
–0.08 0.23 –0.26 0.0003 0.13 0.07 –0.05 0.48 –0.10 0.18
0.26 0.0003 0.38 0.0001 –0.09 0.19 0.19 0.0078 0.19 0.0093
0.01 0.83 –0.24 0.0027 0.32 0.0001 –0.01 0.87 –0.07 0.38
–0.07 0.38 0.23 0.0048 –0.21 0.0099 –0.09 0.23 0.21 0.0089
Data adjusted by linear regression for race and Tanner score.
Mlbm quartile. Compared with girls in the high Mlbm quartile, the girls in the low Mlbm quartile were younger, with similar height and lean body mass; however, they were heavier and had more fat mass and higher levels of all other fatness measures. The girls and boys in the low Mlbm quartile had significantly higher serum insulin, glucose, and triglycerides. They also had a lower HDL-C (P = .07 in boys). The mean SBP was higher in the boys and girls in the low Mlbm quartile, but these differences were not statistically significant (P = .07 in girls). Differences between whites in the high and low Mlbm quartiles followed this same pattern. The numbers of blacks in the high and low quartiles were too small for meaningful comparisons.
DISCUSSION By using the EIC to define insulin sensitivity, we were able to more precisely show the significant interrelations among the risk factors associated with the insulin resistance syndrome in children and the important influence of body fatness on the other currently used methods to assess insulin resistance. The EIC provides a measure of insulin resistance under controlled, 704
standardized conditions of insulin and glucose infusion, whereas fasting insulin and other measures represent an integrated and uncontrolled response to insulin and glucose metabolism.22 Moreover, normalizing insulin sensitivity to lean body mass (Mlbm), thus removing fat mass from the estimate, should provide a more accurate assessment of insulin resistance,27 because approximately 75% of glucose use is localized to skeletal muscle and only 4% or less to fat tissue.28 The initial analyses showed that fasting insulin is significantly related to SBP in both sexes; all lipids, except HDL-C in boys; and triglycerides and HDL-C in girls. Previous studies in children have used plasma insulin levels to determine insulin resistance, with cross-sectional studies showing a significant correlation of fasting insulin with lipids and blood pressure.5-10 Fasting insulin remained significantly related to blood pressure7-10 and lipids5,7,8 in most of those studies after adjustment was done for BMI or skinfold thickness. However, in the Bogalusa study intraclass correlations (insulin resistance index, triglycerides, HDL-C, and mean arterial pressure) fell by approximately 50% across all age groups after adjustment was done for BMI,29 and in this study fasting in-
sulin was significantly related only to serum triglycerides, after adjustment was done for BMI. Thus the relation of fasting insulin to other risk factors appears to be strongly influenced by body fatness. Substituting the homeostasis model or glucose/insulin for fasting insulin gave virtually identical results, and neither provided a better estimate of insulin resistance than fasting insulin. In contrast, Mlbm in this study was significantly related only to triglycerides and HDL-C, and this relation remained significant after adjustment was done for BMI. Similar differences between fasting insulin and glucose use previously have been shown in adults with either the minimal model30 or the EIC.31 An independent effect of insulin resistance on cardiovascular risk in children has also been suggested from other studies. Fasting insulin levels in 6- to 9-year-old children in the Cardiovascular Risk in Young Finns study predicted the children’s level of blood pressure at age 9 to 15 years,10 and in 5- to 9-year-old Pima Indian children, fasting insulin predicted the level of weight gain during the subsequent 9 years of childhood.32 The Bogalusa Heart Study has shown a strong relation over an 8-year period of observation between persistently high fasting insulin levels and the development of risk factors in children and young adults.13 There were significant differences between the boys and girls in the degree of insulin resistance. Mlbm was significantly higher in the boys, suggesting a higher degree of insulin resistance in the girls. This result may be due partly to a greater degree of adiposity in the girls,33 as represented by a significantly greater triceps and subscapular skinfold thickness and percent body fat. It is also supported by the finding that the sex difference in insulin resistance was not present in the very obese (BMI >27 kg/m2) boys and girls in whom the degree of adiposity was similar.26 The higher fasting
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS
VOLUME 139, NUMBER 5 Table IV. Comparison of clinical data between children in the highest and lowest quartiles of Mlbm
Male Variable
Low Mlbm High Mlbm (High ins resist) (Low ins resist)
N Age (y) Height (cm) Weight (kg) BMI (kg/m2) LBM (kg) Body fat (%) Waist (cm) Hip (cm) WHR Triceps (mm) Subscapular (mm) SBP
49 13.0 ± 0.1 163.8 ± 1.2 62.9 ± 2.0 23.1 ± 0.7 44.1 ± 0.9 26.5 ± 1.8 83.4 ± 1.6 93.8 ± 1.5 0.89 ± 0.01 18.8 ± 1.4 13.5 ± 1.0 110.1 ± 1.5
49 12.8 ± 0.1 160.4 ± 1.2 53.9 ± 2.0 20.6 ± 0.7 39.3 ± 0.9 25.8 ± 1.8 74.4 ± 1.6 89.6 ± 1.5 0.83 ± 0.01 20.6 ± 1.4 10.1 ± 1.0 108.7 ± 1.5
Female P .17 .06 .0042 .0124 .0007 .79 .0003 .055 .0007 .39 .0265 .49
Low Mlbm High Mlbm (High ins resist) (Low ins resist) 38 12.6 ± 0.1 159.6 ± 1.0 64.6 ± 2.2 25.1 ± 0.8 40.0 ± 0.9 35.6 ± 1.2 81.6 ± 1.8 99.7 ± 1.7 0.82 ± 0.02 29.4 ± 1.2 18.5 ± 1.1 108.9 ± 1.6
40 13.0 ± 0.1 159.9 ± 1.0 54.5 ± 2.2 21.2 ± 0.8 38.2 ± 0.8 28.8 ± 1.2 72.9 ± 1.8 91.7 ± 1.7 0.80 ± 0.02 23.0 ± 1.2 12.5 ± 1.1 104.6 ± 1.6
P .0236 .87 .0022 .0008 .16 .0003 .0011 .0024 .47 .0007 .0005 .070
Data adjusted by linear regression for race and Tanner score. WHR, Waist-hip ratio.
Table V. Comparison of laboratory data between children in the highest and lowest quartiles of Mlbm
Male Variable
Low Mlbm High Mlbm (High ins resist) (Low ins resist)
N Mlbm (mg/kg/min) Fast insulin (pmol/L) Glucose (mmol/L) Cholesterol (mmol/L) Triglycerides (mmol/L) HDL-C (mmol/L) LDL-C (mmol/L)
49 7.9 ± 0.3 125 ± 10 5.7 ± 0.1 4.00 ± 0.14 1.24 ± 0.09 1.12 ± 0.04 2.34 ± 0.12
49 19.0 ± 0.3 54 ± 10 5.4 ± 0.1 3.86 ± 0.14 0.79 ± 0.09 1.22 ± 0.04 2.27 ± 0.12
Female P .0001 .0001 .0005 .40 .0013 .074 .70
Low Mlbm High Mlbm (High ins resist) (Low ins resist) 38 7.0 ± 0.4 135 ± 10 5.5 ± 0.1 3.94 ± 0.11 1.33 ± 0.11 1.06 ± 0.03 2.27 ± 0.10
40 16.7 ± 0.3 70 ± 10 5.3 ± 0.1 4.00 ± 0.11 0.92 ± 0.11 1.29 ± 0.03 2.29 ± 0.09
P .0001 .0001 .0193 .67 .0119 .0001 .90
Data adjusted by linear regression for race and Tanner score.
insulin level in the girls is also consistent with a higher degree of insulin resistance, although it is not clear why this difference did not reach statistical significance. As noted earlier, there are substantial differences between the insulin clamp and fasting insulin as estimates of insulin resistance, and the correlation coefficient between the two is only 0.42. Moreover, others have reported differences between the relations of fasting insulin and the EIC to risk factors.30,31 It is interesting that despite the significantly lower degree of insulin resistance in the boys, BMI had a highly significant positive corre-
lation with cholesterol, triglycerides, and LDL-C and a significant negative correlation with HDL-C, in contrast to the girls. Although an explanation for this result is not immediately apparent, it suggests that obesity has an early adverse effect on boys that may lead to the increased degree of cardiovascular risk noted in men compared with women. An important finding in this study is that children in the lowest quartile for Mlbm (ie, highest degree of insulin resistance) had anthropometric measurements and laboratory measurements that were consistent with the insulin
resistance syndrome compared with children in the upper quartile for Mlbm. A clustering effect of risk factors at this early age also has been suggested in studies with fasting insulin.12 A wider separation of the cluster effect between the children with high versus low levels of insulin resistance would be expected as they mature into adulthood, because of the well-recognized tracking effects18-20 for the risk factors between childhood and adulthood. These results may have important implications for public health policy. First, they confirm the important influence of obesity early in life. Obesity is 705
SINAIKO ET AL
not only recognized as a current major problem within pediatrics,34 but the prevalence of overweight in children also is increasing35,36 and is accompanied by an increasing prevalence of type 2 diabetes.37 Epidemiologic studies have shown that obesity in childhood predicts adult obesity38 and is associated with an increased risk of coronary heart disease,39 and the rate of weight gain during childhood and adolescence predicts levels of cardiovascular risk in young adults.40 Thus strategies linked to weight control and the reduction of premature cardiovascular morbidity and mortality might be most effective if introduced early in life. Second, the identification of an independent effect of insulin resistance early in life suggests that additional efforts should be made to understand basic and clinical interactions among risk factors before adulthood. Although long-term studies relating childhood insulin resistance to the incidence of risk factors, morbidity, and mortality in adults will help clarify these issues, there is sufficient evidence to propose that the etiologic mechanisms controlling insulin resistance are present in childhood and should be studied in this age group.
REFERENCES 1. DeFronzo RA, Ferrannini E. Insulin resistance. A multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia and atherosclerotic cardiovascular disease. Diabetes Care 1991;14:173-94. 2. Haffner SM, Miettinen H. Insulin resistance implications for type II diabetes mellitus and coronary heart disease. Am J Med 1997;103:152-62. 3. Ginsberg HN. Insulin resistance and cardiovascular disease. J Clin Invest 2000;106:453-8. 4. Meigs JB, D’Agostino RB, Sr, Wilson PWF, Cupples A, Nathan DM, Singer DE. Risk variable clustering in the insulin resistance syndrome. The Framingham Offspring Study. Diabetes 1997;46:1594-600. 5. Jiang X, Srinivasan SR, Webber LS, Wattigney WA, Berenson GS. Associa706
THE JOURNAL OF PEDIATRICS NOVEMBER 2001
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
tion of fasting insulin level with serum lipid and lipoprotein levels in children, adolescents, and young adults: The Bogalusa Heart Study. Arch Intern Med 1995;155:190-6. Steinberger J, Moorehead C, Katch V, Rocchini AP. Relationship between insulin resistance and abnormal lipid profile in obese adolescents. J Pediatr 1995;126:690-5. Raitakari OT, Porkka KVK, Ronnemaa T, Knip M, Uhari M, Akerblom HK, et al. The role of insulin in clustering of serum lipids and blood pressure in children and adolescents. Diabetologia 1995;38:1042-50. Sinaiko AR, Gomez-Marin O, Prineas RJ. Relation of fasting insulin to blood pressure and lipids in adolescents and parents. Hypertension 1997;30:1554-9. Jiang X, Srinivasan S, Bao W, Berenson G. Association of fasting insulin with blood pressure in young individuals: The Bogalusa Heart Study. Arch Intern Med 1993;153:323-8. Taittonen L, Uhari M, Turtinen J, Pokka T, Akerblom HK. Insulin and blood pressure among healthy children. Cardiovascular risk in young Finns. Am J Hypertens 1996;9:193-9. Rocchini AP, Katch V, Schork A, Kelch RP. Insulin and blood pressure during weight loss in obese adolescents. Hypertension 1987;10:267-73. Chen W, Srinivasan AR, Elkasabany A, Berenson GS. Cardiovascular risk factors clustering features of insulin resistance syndrome (Syndrome X) in a biracial (black-white) population of children, adolescents, and young adults: The Bogalusa Heart Study. Am J Epidemiol 1999;150:667-74. Bao W, Srinivasan SR, Berenson GS. Persistent elevation of plasma insulin levels is associated with increased cardiovascular risk in children and young adults: The Bogalusa Heart Study. Circulation 1996;93:54-9. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: The Bogalusa Heart Study. Pediatrics 1999;103:1175-82. Daniels SR, Morrison JA, Sprecher DL, Khoury P, Kimball TR. Association of body fat distribution and cardiovascular risk foctors in children and adolescents. Circulation 1999;99:541-5. Morrison JA, Barton BA, Biro FM, Daniels SR, Sprecher DL. Overweight, fat patterning, and cardiovascular disease risk factors in black and white boys. J Pediatr 1999;135:451-7.
17. Morrison JA, Sprecher DL, Barton BA, Waclawiw MA, Daniels SR. Overweight, fat patterning, and cardiovascular disease risk factors in black and white girls: The National Heart, Lung, and Blood Institute Growth and Health Study. J Pediatr 1999;135:458-64. 18. Bao W, Threefoot SA, Srinivasan SR, Berenson GS. Essential hypertension predicted by tracking of elevated blood pressure from childhood to adulthood: The Bogalusa Heart Study. Am J Hypertens 1995;8:657-65. 19. National Cholesterol Education Program. Report of the expert panel on blood cholesterol levels in children and adolescents. Pediatrics 1992;89:525-83. 20. Clarke WR, Lauer RM. Does childhood obesity track into adulthood? Crit Rev Food Sci Nutr 1993;33:423-30. 21. Berenson GS, Srinivasan SR, Bao W, Newman WP, Tracy RE, Wattigney WA. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. N Engl J Med 1998;338:1650-6. 22. Ferrannini E, Mari A. How to measure insulin sensitivity. J Hypertens 1998; 16:895-906. 23. Slaughter MH, Lohman TG, Baileau RA, Horswill CA, Stillman RJ, Van Loan MD, et al. Skinfold equations for estimation of body fatness in children and youth. Hum Biol 1988;60:709-23. 24. Lansang MC, Williams GH, Carroll JS. Correlation between the glucose clamp technique and the homeostasis model assessment in hypertension. Am J Hypertens 2001;14:51-3. 25. Caro JF. Insulin resistance in obese and nonobese man. J Clin Endocrinol Metab 1991;73:691-5. 26. Moran A, Jacobs DR, Steinberger J, Hong C-P, Prineas R, Luepker R, et al. Insulin resistance during puberty: results from clamp studies in 357 children. Diabetes 1999;48:2039-44. 27. Ferrannini E, Natali A, Bell P, CavalloPerin P, Lalic N, Mingrone G. Insulin resistance and hypersecretion in obesity. J Clin Invest 1997;100:1166-73. 28. DeFronzo RA, Bonadonna RC, Ferrannini E. Pathogenesis of NIDDM: a balanced overview. Diabetes Care 1992;15:318-68. 29. Chen W, Bao W, Begum S, Elkasabany A, Srinivasan SR, Berenson GS. Agerelated patterns of the clustering of cardiovascular risk variables of syndrome X from childhood to young adulthood in a population made up of black and white subjects: The Bogalusa Heart Study. Diabetes 2000;49:1042-8.
SINAIKO ET AL
THE JOURNAL OF PEDIATRICS
VOLUME 139, NUMBER 5 30. Istfam NW, Plaisted CS, Bistrian BR, Blackburn GL. Insulin resistance versus insulin secretion in the hypertension of obesity. Hypertension 1992; 19:385-92. 31. Ferrannini E, Natali A, Capaldo B, Lehtovirta M, Jacob S, Yki-Jahrvinen H. Insulin resistance, hyperinsulinemia, and blood pressure: role of age and obesity. Hypertension 1997; 30:1144-9. 32. Odeleye OE, de Courten M, Pettitt DJ, Ravussin E. Fasting hyperinsulinemia is a predictor of increased body weight gain and obesity in Pima Indian children. Diabetes 1997;46:1341-5. 33. Daniels SR, Khoury PR, Morrison JA. The utility of body mass index as a measure of body fatness in chil-
dren and adolescents: differences by race and gender. Pediatrics 1997; 99:804-7. 34. Christoffel KK, Aroza A. The epidemiology of overweight in children: relevance for clinical care. Pediatrics 1998; 101:103-4. 35. Troiano RP, Flegal KM, Kuczmarski RJ, Campbell SM, Johnson CL. Overweight prevalance in trends for children and adolescents. The National Health and Nutrition Examination Surveys, 1963-1991. Arch Pediatr Adolesc Med 1995;149:1085-91. 36. Luepker R, Prineas R, Jacobs D, Sinaiko A. Secular trends of blood pressure and body size in a multiethnic adolescent population: 19861996. J Pediatr 1999;134:668-74.
37. Rosenbloom AL, Roe J, Young RS, Winter WE. Emerging epidemic of type 2 diabetes in youth. Diabetes Care 1999;22:345-54. 38. Whitaker RD, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med 1997;337:869-73. 39. Must A, Jacques PF, Dallal GE, Bajema CJ, Dietz WH. Long-term morbidity and mortality of overweight adolescents. N Engl J Med 1992;327:1350-5. 40. Sinaiko AR, Donahue RP, Jacobs, DR, Prineas RJ. Relation of rate of growth during childhood and adolescence to fasting insulin, lipids, and systolic blood pressure in young adults. Circulation 1999;99:1471-6.
50 Years Ago in The Journal of Pediatrics GROWTH AND DEVELOPMENT OF INFANTS RECEIVING A PROPRIETARY PREPARATION OF D Frost LH, Jackson RL. J Pediatr 1951;39:585-92
EVAPORATED MILK WITH DEXTRI-MALTOSE AND VITAMIN
To a large extent, the history of pediatrics as a medical specialty is derived from the history of infant feeding. The devastating mortality rate of children who received artificial formulas (ie, nonhuman milk) attracted the attention of the founders of our profession. The response to this was the development of recipes and schedules of nearly incomprehensible complexity. A brief perusal of the feeding section of any early 20th century pediatric text will glaze most contemporary eyes. Infant feeding became easier in the 1950s. A simple recipe for diluting evaporated milk, with added carbohydrate and supplemental vitamins, became popular. In this study, Frost and Jackson provide the practitioner with simple “how to” advice on the preparation of this kind of formula. Among the many differences from practice of the previous decades was the realization that opened cans of evaporated milk could be kept safely in the refrigerator and bottles prepared as needed; complex sterilization procedures were not necessary. What makes this article particularly noteworthy is that it goes beyond a mere description to become a clinical study. Fifty-seven infants were fed according to the method described and subjected to careful serial follow-up for an average of 6 months. Gastrointestinal function (stooling, primarily) was recorded, and serial assessment of growth was reported to compare favorably with the contemporary Iowa growth standards. Vitamin D status was assessed (indirectly) by bone films in most infants, and hemoglobin concentrations were also measured. Frost and Jackson nicely provided the pediatrician of a half-century ago with practical, safe, and sound advice in formula feeding. Unfortunately, they did not consider the more basic question: why go through the trouble of preparing any formula when a free, safe, and physiologically optimal preparation (human milk) is available to virtually any mother? Thomas R. Welch, MD Chair, Department of Pediatrics University Hospital Syracuse, NY 13210 9/37/120092 doi:10.1067/mpd.2001.120092
707