AJCN EB Program 2010 Early Registration
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental data
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Morrison, J. A
Right arrow Articles by Wang, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Morrison, J. A
Right arrow Articles by Wang, P.
Agricola
Right arrow Articles by Morrison, J. A
Right arrow Articles by Wang, P.
American Journal of Clinical Nutrition, Vol. 88, No. 3, 778-788, September 2008
© 2008 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Pre-teen insulin resistance predicts weight gain, impaired fasting glucose, and type 2 diabetes at age 18–19 y: a 10-y prospective study of black and white girls1,2,3

John A Morrison, Charles J Glueck, Paul S Horn, George B Schreiber and Ping Wang

1 From the Division of Cardiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH (JAM); the Cholesterol Center, Jewish Hospital of Cincinnati, Cincinnati, OH (CJG and PW); the Department of Mathematics, University of Cincinnati, Cincinnati, OH (PSH); and Westat Inc, Rockville, MD (GBS)

2 Supported by NIH-NHLBI Contract HC-55023-26 and HL 48941 and by the Lipoprotein Research Fund of the Jewish Hospital of Cincinnati.

3 Reprints not available. Address correspondence to JA Morrison, OSB 4, Division of Cardiology, Children's Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229. E-mail: john.morrison{at}cchmc.org.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Identifying early pre-teen predictors of adolescent weight gain and the development of impaired fasting glucose (IFG) and type 2 diabetes (T2DM) at age 18–19 y could provide avenues for prevention.

Objective: We evaluated possible pre-teen predictors for development of IFG, T2DM, and changes in body mass index at age 18–19 y in black and white girls.

Design: In a prospective cohort study, body habitus and fasting insulin and glucose were measured at ages 9–10 and 18–19 y, and multiple 3-d diet records were collected. Factors predicting 10-y change in body mass index and development of IFG and T2DM together were assessed.

Results: In multivariate analyses, 10-y change in homeostatic model assessment of insulin resistance (HOMA-IR) and the age 9–10 y HOMA-IR x percentage of calories from fat interaction were positive predictors of 10-y changes in body mass index. At age 18–19 y, there were 5 incident cases of T2DM, 37 cases of IFG, and 597 noncases. Age 9–10 y IFG and HOMA-IR (or insulin), 10-y change in HOMA-IR (or insulin), and the age 9–10 y insulin x total caloric intake interaction predicted IFG and T2DM at age 18–19 y.

Conclusions: Pre-teen IFG, insulin resistance (and insulin), and rapidly increasing insulin resistance during adolescence identifies girls who are at greater risk of future IFG and T2DM. In addition, insulin resistance, interacting with high-fat diets, identifies girls who are at risk of greater weight gain. These findings could open avenues to primary prevention of obesity, IFG, and T2DM in children.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
During the past 25 y, the prevalences of obesity and type 2 diabetes mellitus (T2DM) have increased, and the age at onset of T2DM has dropped dramatically, especially in black females (1, 2). T2DM now presents in the teenage years with increasing frequency, which reflects the longer duration of the obese insulin-resistant state in late teens (2). Although impaired fasting glucose (IFG), obesity, insulin resistance (IR), and impaired insulin response to glucose are independent predictors of T2DM in population studies of adults (3), their association in childhood has not been established. Furthermore, the association may be dependent on a positive family history of T2DM (4). IR is robustly associated with the development of T2DM in adult offspring of 2 parents with T2DM, but not in adults whose parents did not have T2DM (4). Most obese, insulin-resistant persons do not develop T2DM, and their normal glucose tolerance is maintained by great enough increases in insulin secretion to compensate for IR (5). Gerich (6) suggested that the superimposition of obesity-related IR on a beta cell with genetically limited ability to compensate by increasing insulin secretion is "... the fundamental pathological sequence of events." In contrast (7), although this issue is hotly debated, IR may not necessarily be an essential component of T2DM (6), even though most patients with T2DM have some degree of IR. Models to predict which children are at greater risk of IFG and T2DM could provide diagnostic and therapeutic insights into the etiologic relations of IR to IFG and T2DM (8).

Numerous studies report that IR plays an important role in the development of obesity. There is, however, no consensus as to whether IR promotes weight gain or loss or whether its role may vary, given environmental factors such as total caloric and fat intake (9-16). Mosca et al (17) reported that IR in adults appears to interact with high-fat diets to increase weight gain.

Prospective analysis of the role of IR in the development of obesity, IFG, and T2DM in childhood and adolescence may provide therapeutic approaches to slow or even reverse the current trends of pandemic obesity and T2DM by young adulthood (18). An ancillary study of the National Heart, Lung, and Blood Institute (NHLBI) Growth and Health Study (NGHS), a 10-y prospective study of the development of obesity in black and white girls, offers such an opportunity (18). Two clinical centers of NGHS elected to collect fasting blood for the measurement of insulin and glucose in years 1 (age 9–10 y) and 10 (age 18–19 y).

In the current report, we evaluate the role of pre-teen IR resistance and insulin in adolescent weight gain and the development of IFG and T2DM. We hypothesized that pre-teen IR, interacting with dietary factors such as total calories and fat calories, and the 10-y change in IR would positively predict 10-y increases in body mass index (BMI) and the development of IFG and T2DM. We further hypothesized that age 9–10 y BMI would be a strong independent predictor of 10-y changes in BMI during adolescence, because the tracking of BMI is well documented (19-21).


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
The NGHS was described previously (18). It was conducted under contract with NHLBI and was motivated by the racial disparities in both obesity and cardiovascular disease (CVD) risk in women (18). The clinical centers in Cincinnati, OH, and Washington, DC, elected to conduct an ancillary project, collecting fasting blood samples for the measurement of glucose and insulin in years 1 and 10. Race was self-declared, and enrollment was restricted to racially concordant households—ie, to girls who said they were black or white and whose parents or guardians said that they were black or white, respectively. The Cincinnati clinic recruited girls from public and parochial schools in the inner city, within-city residential neighborhoods, and suburban areas; the Washington clinic recruited girls from a health maintenance organization.

Written informed consent was obtained from the girls’ parents or guardians, and written assent was obtained from the girls. The procedures followed were in accordance with the ethical standards of the institutional review boards of the 2 centers, which approved the study.

Clinical measures
In the NGHS, obesity was assessed annually according to a standard protocol (18) using BMI (in kg/m2) as recommended by several expert panels (22-24). In addition, beginning in year 2, waist circumference was measured at the minimum waist as an indicator of fat patterning. Pubertal maturation was visually assessed by trained, certified staff using a modification of Tanner staging to include areolar development instead of breast development (25). Insulin and glucose concentrations were measured after an overnight fast (≥8 h) using the Michigan Diabetes Research and Training Center (Ann Arbor) in year 1 (age 9–10 y) and the Endocrine Laboratory at the University of Cincinnati/Children's Medical Center in year 10 (age 18–19 y). All of the 9–10-y-old children were evaluated in one batch assay in Michigan, and all 18–19-y-old subjects were evaluated in a separate batch assay in Cincinnati. Both insulin assays used a competitive protein-binding radioimmunoassay. To assess whether possible differences in insulin assays may have affected interpretation of the data, z scores of each assay (separated by race) were used to calculate changes in insulin and IR values in analyses that incorporated changes in insulin and IR as explanatory variables. Glucose was measured at year 1 by using a hexokinase reagent (Boehringer-Mannheim Inc, Mannheim, Germany) and at year 10 by using the glucose oxidase method with the Hitachi 704 chemistry analyzer (Roche Diagnostics, Indianapolis, IN). CVs ranged from 5% to 11% for insulin and from 2% to 7% for glucose in year 1, and they were 9% and 4%, respectively, in year 10. Homeostasis model assessment of IR (HOMA-IR), which correlates HOMA with estimates of IR measured by the euglycemic clamp technique, was used as an index of IR (26). Although the HOMA-IR measure is less accurate than the euglycemic clamp method, in large epidemiologic studies, it is a reasonable alternative to the complicated clamp method that requires continuous intravenous administration of insulin and glucose for 3 h for calculation of insulin sensitivity (27). Huang et al (28) studied HOMA-IR in white and black children and concluded, "...a modified HOMA equation accurately predicted insulin sensitivity, but its utility is similar to [that of] fasting insulin alone."

In these analyses, we defined T2DM as a fasting glucose concentration of ≥126 mg/dL and IFG as a glucose concentration of ≥100 but <126 mg/dL to comport with the American Diabetes Association's 2004 changes in the definition of IFG (29). To increase the number of cases for analysis, we combined IFG and T2DM cases in some analyses. Parental history of diabetes was obtained for participants by investigator-directed interviews with each child's parents or guardians.

Dietary and parental socioeconomic data
In 8 of the 10 yearly follow-up visits (years 1, 2, 3, 4, 5, 7, 8, and 10), a 3-d dietary diary was completed by the girls and retrieved by Registered Dietitians. The data were entered and summarized for analyses by using the most current version of the NUTRITION DATA SYSTEM for RESEARCH software (30) developed for the Nutrition Coordinating Center (University of Minnesota, Minneapolis, MN) for the calculation of total calories and calories from protein, fat, and carbohydrate.

Socioeconomic data on the girls and their parents were collected by following NGHS protocols (18).

Statistical analysis
All analyses were performed with the use of SAS software (version 9.1; SAS Inc, Cary, NC). The analyses of pre-teen HOMA-IR and IFG, and the development of IFG and T2DM at age 18–19 y included subjects with complete data on age, race, and BMI, plus fasting HOMA-IR at age 9–10 and 18–19 y (n = 639). Including dietary factors as explanatory variables restricted the analysis sample to 521 girls. The addition of a parental history of diabetes to the regression models for development of IFG and T2DM required all previously mentioned data plus a family history, which further restricted the sample size to 437 (Table 4Go). The number of participants included in different analyses by virtue of the data required for those analyses and the data available are shown in Figure 1Go by data required and analyzed.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Significant predictors of 10-y change in BMI in black and white girls from age 9–10 y to age 18–19 y1

 

Figure 1
View larger version (30K):
[in this window]
[in a new window]

  		FIGURE 1.. Flow chart detailing the number of participants in the National Growth and Health Study by the table in which they are presented in the current report and detailing the data required and the analyses conducted.

 
Because we hypothesized that pre-teen HOMA-IR would interact with mean dietary fat during adolescence to explain changes in BMI, the sample size for these analyses were the same as that for the analyses of the development of IFG and T2DM using dietary factors (n = 521). The change in BMI was calculated as BMI at age 18–19 y minus BMI at age 9–10 y, without consideration of BMI fluctuations in the intervening 10 y.

Black-white differences in household income and maximal parental education were assessed by using the Mantel-Haenszel chi-square test. Summary statistics were used to describe the analysis sample of white and black girls at study entry (age 9–10 y) and 10 y later. Differences in age were compared by using Wilcoxon's tests, and differences in other study variables were compared by using analysis of variance (ANOVA) after adjustment for age. Mean ± SD insulin, glucose, and HOMA-IR also were calculated at each visit for girls categorized by obesity status (using the age- and sex-specific 95th percentile of BMI from the Centers for Disease Control and Prevention 2000 dataset as the cutoff for obesity). Differences between race in the percentage of girls who were pubertal at age 9–10 y, percentage of parents with T2DM, and percentage of subjects with IFG and T2DM at age 18–19 y were tested for significance by using chi-square tests.

Differences in nutrient intake between races were assessed by using ANOVA after adjustment for age. Within-race changes in nutrient intake over 10 y were assessed by using paired Wilcoxon's tests.

To evaluate the interaction of baseline HOMA-IR and the percentage of calories from fat during the 10-y follow-up as an explanation for the 10-y increase in BMI, we separated the girls in the top tertile and the bottom 2 tertiles of HOMA-IR at baseline and those in the top tertile and the bottom 2 tertiles of calories from fat during follow-up. This separation was done by race. Next we compared the mean (and median) changes in BMI during the 10-y follow-up. Differences in the distributions of glucose concentrations of <100, 100–126, and ≥126 mg/dL at age 18–19 y in subjects categorized by glucose concentrations at age 9–10 y were determined by using Fisher's exact test or the Mantel-Haenszel chi-square test.

To explain changes in BMI or waist circumference, regression analyses were carried out by using a backward elimination procedure. Explanatory variables included age, BMI (or waist), HOMA-IR, and maturation stage at age 9–10 y. Additional explanatory variables included the 10-y changes in HOMA-IR and dietary factors (mean total calories and percentage of calories from fat, protein, and carbohydrate during follow-up) and the age 9–10 y HOMA-IR x dietary factors interaction. For 10-y BMI changes, the 521 girls with complete data needed for regression analyses had 1–8 validated 3-d dietary records (mean: 7.3 ± 1.08 three-day dietary records). For 10-y waist changes, 512 girls had complete data. These analyses were conducted separately for black and white girls.

As an initial evaluation of pre-teen predictors of IFG and T2DM at age 18–19 y, a series of univariate logistic regression analyses were used to test predictive capabilities (area under receiver operating characteristic curve) of each of the conventional risk factors—ie, age 9–10 y BMI, waist, insulin, glucose, HOMA-IR, and IFG—for the outcomes. Next, a series of stepwise logistic regressions analyses were conducted to identify predictors of IFG and T2DM at age 18–19 y, beginning with race and age 9–10 y factors, including age, BMI, HOMA-IR (or insulin), and IFG status. In subsequent regression models, these same factors plus 10-y changes in BMI and HOMA-IR (or insulin), dietary intake, and the interactions of diet with baseline BMI, HOMA-IR, or insulin were added as candidate predictors. Finally, parental history of T2DM and the interactions of parental history of T2DM with baseline HOMA-IR or insulin, with BMI, and with IFG were added as candidate explanatory variables. In these logistic models, odds ratios were calculated for the increments in the predictors, scaled as follows: 5 uU/mL for insulin (representing 43% of SD for insulin in black girls and 57% of SD for white girls), 5 mg/dL for glucose (71% of SD for glucose for black girls and 63% for white girls), 100 cal/d for total caloric intake (25% of SD for total caloric intake in black girls and 29% of SD for white girls), and 1 unit for HOMA-IR (34% of SD of HOMA IR for black girls and 44% of SD for white girls).

Because the insulin assays for years 1 and 10 were performed in different laboratories, we repeated the analyses that involved changes in insulin and HOMA-IR as explanatory variables, by using the changes in z scores for these values. The results using z scores did not differ from those using the raw data; for the results using z scores, see Tables S1–S4 under "Supplemental data" in the current online issue. To assess the effects of pubertal stage and BMI at age 9–10 y on black-white comparisons, 108 black and 108 white 9–10-y-olds were matched by pubertal status and BMI, and subsequent comparisons were conducted by using Wilcoxon's tests.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
As shown in Table 1Go, there were significant socioeconomic differences between the black and white participants. Household income was higher in the white than in the black families, and the white parents had more education than did the black parents, but there was a wide distribution of income levels and educational achievement in each racial group.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Age, height, weight, BMI, waist circumference, percentage pubertal, and percentage with positive family history of type 2 diabetes mellitus (T2DM) in black and white girls at ages 9–10 and 18–19 y; household income; and parents’ education levels1

 
The black girls were significantly older than the white girls both at entry into NGHS and at year 10, but the mean difference in age was only 3.6 mo at entry and 2.4 mo at age 18–19 y (Table 1Go). A greater percentage of black girls than of white girls were pubertal at age 9–10 y. After adjustment for age, black girls had greater BMI, waist, insulin, and HOMA-IR at both age 9–10 and age 18–19 y. In addition, black girls had higher glucose at age 18–19 y. Given the strong positive association of serum insulin concentration with the onset of puberty and obesity and given the black-white differences in both pubertal status and BMI, we matched 108 black girls with 108 white girls by BMI and pubertal stage and then compared insulin, glucose, and HOMA-IR concentrations. The pattern of black-white differences in these 216 paired girls (data not shown) was essentially the same as that shown for the total cohort in Table 1Go.

When comparisons of mean insulin and HOMA-IR were examined within obesity categories, differences between obese black and white girls were not significant (P ≥ 0.3 at each visit), but differences between nonobese black and white girls were significant both at age 9–10 y and age 18–19 y (P ≤ 0.0001, Table 1Go). Signficantly more black girls than white girls developed IFG or T2DM (P = 0.047, Table 1Go).

The summary data for dietary intake by race at ages 9–10 and 18–19 y are shown in Table 2Go. The difference in total caloric intake was not significant at age 9–10 y but was significantly (P = 0.040) greater in black girls at 18–19 y. At age 9–10 y, black girls reported eating a greater percentage of calories from fat (P = 0.0005) and a lower percentage of calories from carbohydrate (P = 0.0022) than did white girls; the percentage of calories from protein did not differ by race. At age 18–19 y, black girls reported eating a greater percentage of calories from fat and a lower percentage of calories from carbohydrate (P < 0.0001 for both) than did white girls. The percentage of calories from protein did not differ significantly by race (Table 2Go). In 108 black-white pairs of 9–10-y-olds, matched by pubertal status and BMI, the pattern of black-white differences shown in Table 2Go was not changed (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Total caloric intake and percentage of calories from protein, fat, and carbohydrate by race at study entry and 10 y later

 
Pre-teen predictors of increases in body mass and waist circumference
As shown in Table 3Go, girls in the top tertiles of HOMA-IR and dietary fat at baseline had a greater 10-y increase in BMI than did girls with baseline HOMA-IR and dietary fat in the bottom 2 tertiles (white girls: 7.0 and 5.3, respectively; P = 0.002; black girls: 9.2 and 7.2, respectively; P = 0.054). Of white girls in the top tertile of age 9–10 y HOMA-IR, those also in the top tertile of percentage of calories from fat had a larger 10-y increase in BMI than did white girls in the bottom 2 tertiles of percentage of calories from fat (7.0 and 5.6, respectively; P = 0.014; Table 3Go). Of white girls in the top tertile for dietary fat, BMI increased more in those with top tertile HOMA-IR than in those in the bottom 2 tertiles (7.0 and 6.1, respectively; P = 0.062). Of black girls in the top tertile of percentage of calories from fat, those with top tertile HOMA-IR at age 9–10 y had a greater 10-y increment in BMI than did those in the bottom 2 tertiles of HOMA-IR (9.2 and 6.6, respectively; P = 0.023; Table 3Go).


View this table:
[in this window]
[in a new window]

  		TABLE 3. 10-y Change in BMI in black and white adolescent girls by homeostasis model assessment of insulin resistance (HOMA-IR) at age 9–10 y and mean percentage of calories from fat during follow-up1

 
For white girls, baseline BMI was not a significant predictor of 10-y change in BMI, but change in HOMA-IR (partial R2 = 12.4%) and the baseline HOMA-IR x fat calories interaction (27%) were significant (Table 4Go). In the model for black girls, baseline BMI, change in HOMA-IR, and the baseline HOMA-IR x percentage of calories from fat were significant; the highest partial R2 (13.7%) was for baseline BMI. In addition, in black girls, pubertal maturation stage at baseline was negatively associated with the change in BMI. The partial R2 associated with the baseline HOMA-IR x percentage of calories from fat interaction was 27.0% in the white girls and 3.6% in black girls. When dietary sugar, added sugar, and fiber were added to the regression models, they did not stay in the model as significant explanatory variables, and the resultant model did not change (data not shown).

In regression analyses to identify predictors of the changes in waist circumference from age 10–11 y to age 18–19 y, baseline waist, baseline IR interacting with fat calories, and HOMA-IR were significant in the model for both black and white girls (Table 5Go). In addition, in black girls, pubertal maturation stage was a significant (negative) predictor. The dietary fat x age 9–10 y HOMA-IR interaction explained 18.0% of the variability in the change in waist in white girls and 6.7% of that in black girls.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Significant predictors of 10-y change in waist circumference in black and white girls from age 9–10 y to age 18–19 y1

 
Pre-teen predictors of impaired fasting glucose and type 2 diabetes mellitus at age 18–19 y
By age 18–19 y, 5 (1.4%) of 359 black girls had developed T2DM, whereas none of the 280 white girls had done so; in addition, 25 (7%) of 359 black girls had developed IFG compared with 12 (4.3%) of 280 white girls (chi-square: 6.1, df: 2; P = 0.047; Table 1Go). The number and percentage of girls by race and by glucose concentration—normal or IFG—at entry (age 9–10) is shown in Table 6Go, which also shows glucose status at age 18–19 y: ie, normal, IFG, or T2DM. At entry, 37 (13%) of 280 white girls had glucose concentrations consistent with IFG, and 243 had glucose concentrations of <100 mg/dL (Table 6Go). None of the white girls had T2DM at follow-up (age 18–19 y). Of the 243 white girls with normal glucose at entry, 236 (97%) had normal glucose and 7 (3%) had IFG at follow-up (Table 6Go). Of the 37 white girls with IFG at age 9–10 y, 32 (86%) had normal glucose and 5 (14%) had IFG at age 18–19 y. Baseline IFG was significantly associated with IFG at age 18–19 y in white girls (P = 0.012, Fisher's exact test; Table 6Go). Of the 37 white girls with glucose concentrations consistent with IFG at entry, changes in BMI percentile and HOMA-IR did not differ between the 5 girls who had IFG at age 18–19 y and the 32 who did not (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 6. Impaired fasting glucose (≥100 < 126 mg/dL) at study entry at age 9–10 y and impaired fasting glucose and type 2 diabetes mellitus (glucose ≥ 126 mg/dL) at age 18–19 y

 
At age 9–10 y, 60 (17%) of 359 black girls had IFG and 299 (83%) of 359 had glucose concentrations of <100 mg/dL (Table 6Go). Five black girls had T2DM and 25 had IFG at age 18–19 y. Of the 299 black girls with normal glucose at entry, 284 (95%) had normal glucose, 12 (4%) had IFG, and 3 (1%) had T2DM at follow-up 10 y later. Of the 60 black girls with glucose concentrations consistent with IFG at age 9–10 y, 45 (75%) had normal glucose, 13 (22%) had IFG, and 2 (3%) had T2DM at age 18–19. The association of baseline glucose category with glucose category at age 18–19 y was significant (P = 0.0004; Table 6Go). Of the 60 black girls with IFG at age 9–10 y, the changes in BMI percentile or change in IR did not differ between the 15 girls who developed IFG or T2DM at age 18–19 y and the 45 who did not (data not shown).

BMI, IFG status, glucose, insulin, HOMA-IR at age 9–10 y, and waist at age10-11 y were evaluated as predictors of T2DM at age 18–19 y in univariate logistic regression analyses; only insulin and HOMA-IR were significant (Table 7Go). When IFG or IFG plus T2DM at age 18–19 y was the outcome in the univariate analyses, 9–10-y-old BMI, waist circumference, presence of IFG, and glucose, insulin, and IR concentrations were significant predictors (Table 7Go).


View this table:
[in this window]
[in a new window]

  		TABLE 7. Risk of developing type 2 diabetes mellitus (T2DM), impaired fasting glucose (IFG), and hyperglycemia (IFG+T2DM) at age 18–19 y in a biracial schoolgirl cohort1

 
Multivariate stepwise logistic regression analyses run to identify early predictors of elevated glucose at age 18–19 y consistently identified age 9–10 y HOMA-IR (or insulin) and IFG as significant (Tables 8Go, 9Go, and 10Go). When race and baseline age, BMI, HOMA-IR (or insulin), and IFG were the only candidate predictors, 639 girls (5 cases of T2DM and 37 cases of IFG) were used in the regression model (Table 8Go). Both age 9–10 y HOMA-IR (or insulin) and IFG were significant explanatory variables (Table 8Go). When changes in HOMA-IR and dietary data were added as candidates, 521 girls (2 incident cases of T2DM and 26 cases of IFG) were used in the regression model (Table 9Go). Baseline IFG and HOMA-IR (or insulin) again were significant, as was the 10-y increase in HOMA-IR (or insulin) (Table 3Go). When parental history of diabetes was added, 437 girls (2 incident cases of T2DM and 23 cases of IFG) were used in the regression model (Table 4Go). Here, baseline IFG and the 10-y increase in HOMA-IR (or insulin) were significant predictors for IFG+T2DM. The baseline insulin x total calories interaction was significantly associated with IFG + T2DM, but parental history of diabetes was not significant (Table 4Go). Results were consistent whether 1 category of elevated glucose (glucose ≥ 100 mg/dL) or 2 categories (100 ≤ glucose < 126 mg/dL and glucose ≥ 126 mg/dL) were used (Table 10Go).


View this table:
[in this window]
[in a new window]

  		TABLE 8. Significant explanatory variables for impaired fasting glucose (IFG) and type 2 diabetes mellitus (T2DM) by stepwise logistic regression1

 

View this table:
[in this window]
[in a new window]

  		TABLE 9. Significant explanatory variables for impaired fasting glucose (IFG) and type 2 diabetes mellitus (T2DM) by stepwise logistic regression1

 

View this table:
[in this window]
[in a new window]

  		TABLE 10. Significant explanatory variables for impaired fasting glucose (IFG) and type 2 diabetes mellitus (T2DM) by stepwise logistic regression1

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The central, novel finding in our 10-y prospective, longitudinal study of obesity and the development of IFG and T2DM in black and white girls starting at age 9–10 y was the key role of pre-teen HOMA-IR (or insulin) and its interactions with dietary fat or calories for the increases in BMI (and waist circumference) and the incidence of IFG and T2DM at age 18–19 y. Thus, documentation of hyperinsulinemia, HOMA-IR, and IFG at age 9–10 y identifies children at greater risk of developing IFG, T2DM, and obesity, as does rapidly increasing IR (and insulin) during adolescence. Within this frame of reference, Steinberger and Daniels (8) concluded, "...insulin resistance often is associated with T2DM[;] the first step in assessment is to identify children who would benefit from intervention." The concordance of insulin and HOMA-IR concentrations in obese black and white girls but not in nonobese black and white girls (Table 5Go) suggests that, when white girls become obese, they attain the same insulin resistance as do black girls and they lose the lower HOMA-IR present in nonobese white girls than in nonobese black girls. Moreover, after pairing black girls and white girls by BMI and pubertal status at age 9–10 y, group interracial differences in insulin, glucose, and HOMA-IR were maintained. Furthermore, although the interaction of HOMA-IR and fat calories was a highly significant independent predictor of 10-y change in BMI in both black and white girls, this interaction term accounted for 27.0% of the variability in the 10-y change in BMI in white girls and for 3.6% in black girls, which suggests that IR and dietary fat may play a larger role in the development of obesity in white girls than in black girls.

The finding of a positive role for the baseline HOMA-IR interaction with fat calories in the 10-y increases in BMI comports with the report by Mosca et al (17) in a 14-y longitudinal study in adults, in which weight gain over time was positively associated with fat intake in both insulin-resistant and insulin-sensitive subjects, but the slope of the fat intake–weight gain association was greater in the subjects with IR. Kahn and Flier (31) proposed that IR and hyperinsulinemia, in addition to being caused by obesity, can contribute to the development of obesity via an interaction with total and fat calories, as observed in the present longitudinal study. Lustig (15) theorized that the genesis of obesity may lie in hyperinsulinemia. In the current report, congruent with results from previous tracking studies (19-21), pre-teen BMI was a major independent predictor of change in BMI at age 18–19 y in black girls.

The prevalence of IFG among adolescents in 1999–2000, based on data from the National Health and Nutrition Examination Survey (NHANES), was 7%, higher in boys than in girls (10% compared with 4%) (32), and higher in obese than in normal-weight adolescents (17.8% compared with 2.8%) (32). Adolescents with IFG appeared to be at greater risk of T2DM and to have higher fasting insulin, total and LDL-cholesterol, and triglyceride concentrations; higher systolic blood pressure; and lower HDL-cholesterol concentrations than did adolescents with normal fasting glucose (<100 mg/dL) (32). Findings from the current study confirm that the presence of IFG at age 9–10 y is associated with greater risk of T2DM at age 18–19 y. Nichols et al (33) followed adult health maintenance organization patients over a 9-y period and reported that 8.1% of subjects who had glucose values of 100 to 109 mg/dL progressed to T2DM during follow-up and that those who did progress did so within a mean of 41.4 mo and at a rate of 1.34%/y. In contrast, 24.3% of subjects with original glucose values of 110 to 125 mg/dL progressed to T2DM during follow-up and did so within a mean of 29 mo and at a rate of 5.56%/y (33).

In the present study, the interaction of age 9–10 y HOMA-IR with the percentage of calories from fat was independently predictive of 10-y weight gain in childhood. Moreover, age 9–10 y HOMA IR and IFG, 10-y change in HOMA-IR (or insulin), and the interaction of age 9–10 y insulin with total caloric intake was independently predictive of the development of IFG and T2DM at age 18–19 y. These findings were congruent with reports by Mosca et al (17), Kahn and Flier (31), Odeleye et al (12), and Sigal et al (14). The consistent negative associations between maturation stage and the change in BMI seem to reflect the additional premenarcheal growth attained by these girls at age 9–10 y.

In contrast to the current study and to other reports that IR (12, 14) and interaction between IR and calories from fat (17) are central to the development of obesity in adolescence and young adulthood, studies by Travers et al (13), Maffeis et al (16), and Hoffman et al (34) reported that IR leads to less weight gain and less body fatness. Schwartz et al (35) prospectively studied Pima Indian adults over a 3-y period. The presence of IR relative to other children was associated with an independent reduction in the risk of weight gain in some regression analyses.

The present study has the following limitations. First, participants were not a random selection from the United States population, as in the National Health and Nutrition Examination Survey, but they came from a biracial schoolgirl population and from a health maintenance organization. Thus, the data, although suggestive, should be confirmed and cannot be extrapolated to all adolescent girls. Second, magnetic resonance imaging (36) or computed tomography visceral fat measurements to estimate intravisceral and extravisceral fat measurements were not done. Third, 3-d dietary diary data may reflect actual dietary intake less optimally than does a 7-d record (37). Fourth, the insulin concentrtions at baseline and follow-up were measured in 2 different laboratories using competitive protein-binding radioimmunoassays, which is a potential problem because the same insulin assay in different laboratories may give different results (38). However, in the regression models that included change in HOMA-IR or change in insulin, either of which may have been affected by systematic differences in insulin observed by virtue of different (albeit very similar) insulin assays, the z score transformation of change in HOMA-IR and insulin provided results nearly identical to the untransformed data. This finding suggested that the observations of the effects of changes in HOMA-IR or insulin were independent of putative assay differences.

If insulin or HOMA-IR is documented at age 9–10 y, and if these factors increase during adolescence, and if diet and exercise, metformin, or all (39-44) are then initiated, we speculate that the development of obesity as well as IFG and T2DM at age 18–19 y may be reduced, with an ultimate goal of primary prevention of both progressive obesity and T2DM. In the present study, the significant association between 10-y change in BMI and an interaction between HOMA-IR and calories from fat highlights IR as a major modifiable target in the prevention or reduction of obesity.


   ACKNOWLEDGMENTS
 
The authors’ responsibilities were as follows—JAM and GBS (Principal Investigators of the 2 clinical sites): conducte the ancillary project and designed and carried out the 10-y epidemiologic study, including data collection, verification that data were entered accurately, and editing of the manuscript; JAM, CJG, PSH, and PW: analyzed the data from the collaborative (anthropometry and diet, race, and family history) and ancillary (insulin and glucose) studies and wrote the draft of the manuscript; and all authors: revised the manuscript. None of the authors had a personal or financial conflict of interest.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Mokdad AH, Ford ES, Bowman BA, et al. Diabetes trends in the U.S.: 1990-1998. Diabetes Care 2000;23:1278–83.[Abstract/Free Full Text]
  2. Everhart JE, Pettitt DJ, Bennett PH, Knowler WC. Duration of obesity increases the incidence of NIDDM. Diabetes 1992;41:235–40.[Abstract]
  3. Bonora E, Kiechl S, Willeit J, et al. Population-based incidence rates and risk factors for type 2 diabetes in white individuals: the Bruneck study. Diabetes 2004;53:1782–9.[Abstract/Free Full Text]
  4. Goldfine AB, Bouche C, Parker RA, et al. Insulin resistance is a poor predictor of type 2 diabetes in individuals with no family history of disease. Proc Natl Acad Sci U S A 2003;100:2724–9.[Abstract/Free Full Text]
  5. Polonsky KS, Sturis J, Bell GI. Seminars in Medicine of the Beth Israel Hospital, Boston. Non-insulin-dependent diabetes mellitus—a genetically programmed failure of the beta cell to compensate for insulin resistance. N Engl J Med 1996;334:777–83.[Free Full Text]
  6. Gerich JE. Insulin resistance is not necessarily an essential component of type 2 diabetes. J Clin Endocrinol Metab 2000;85:2113–5.[Free Full Text]
  7. Cavaghan MK, Ehrmann DA, Polonsky KS. Interactions between insulin resistance and insulin secretion in the development of glucose intolerance. J Clin Invest 2000;106:329–33.[Medline]
  8. Steinberger J, Daniels SR. Obesity, insulin resistance, diabetes, and cardiovascular risk in children: an American Heart Association scientific statement from the Atherosclerosis, Hypertension, and Obesity in the Young Committee (Council on Cardiovascular Disease in the Young) and the Diabetes Committee (Council on Nutrition, Physical Activity, and Metabolism). Circulation 2003;107:1448–53.[Free Full Text]
  9. Gould AJ, Williams DE, Byrne CD, Hales CN, Wareham NJ. Prospective cohort study of the relationship of markers of insulin resistance and secretion with weight gain and changes in regional adiposity. Int J Obes Relat Metab Disord 1999;23:1256–61.[Medline]
  10. Swinburn BA, Nyomba BL, Saad MF, et al. Insulin resistance associated with lower rates of weight gain in Pima Indians. J Clin Invest 1991;88:168–73.[Medline]
  11. Hoag S, Marshall JA, Jones RH, Hamman RF. High fasting insulin levels associated with lower rates of weight gain in persons with normal glucose tolerance: the San Luis Valley Diabetes Study. Int J Obes Relat Metab Disord 1995;19:175–80.[Medline]
  12. 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.[Abstract]
  13. Travers SH, Jeffers BW, Eckel RH. Insulin resistance during puberty and future fat accumulation. J Clin Endocrinol Metab 2002;87:3814–8.[Abstract/Free Full Text]
  14. Sigal RJ, El-Hashimy M, Martin BC, Soeldner JS, Krolewski AS, Warram JH. Acute postchallenge hyperinsulinemia predicts weight gain: a prospective study. Diabetes 1997;46:1025–9.[Abstract]
  15. Lustig RH. Childhood obesity: behavioral aberration or biochemical drive? Reinterpreting the First Law of Thermodynamics. Nat Clin Pract Endocrinol Metab 2006;2:447–58.[Medline]
  16. Maffeis C, Moghetti P, Grezzani A, Clementi M, Gaudino R, Tato L. Insulin resistance and the persistence of obesity from childhood into adulthood. J Clin Endocrinol Metab 2002;87:71–6.[Abstract/Free Full Text]
  17. Mosca CL, Marshall JA, Grunwald GK, Cornier MA, Baxter J. Insulin resistance as a modifier of the relationship between dietary fat intake and weight gain. Int J Obes Relat Metab Disord 2004;28:803–12.[Medline]
  18. The NHLBI Growth and Health Study Research Group. Obesity and cardiovascular disease risk factors in 9- and 10-year-old black and white girls: the NHLBI Growth and Health Study. Am J Public Health 1992;82:1613–20.[Abstract/Free Full Text]
  19. Freedman DS, Khan LK, Serdula MK, Ogden CL, Dietz WH. Racial and ethnic differences in secular trends for childhood BMI, weight, and height. Obesity (Silver Spring) 2006;14:301–8.[Medline]
  20. Deshmukh-Taskar P, Nicklas TA, Morales M, Yang SJ, Zakeri I, Berenson GS. Tracking of overweight status from childhood to young adulthood: the Bogalusa Heart Study. Eur J Clin Nutr 2006;60:48–57.[Medline]
  21. Freedman DS, Khan LK, Serdula MK, Dietz WH, Srinivasan SR, Berenson GS. Racial differences in the tracking of childhood BMI to adulthood. Obes Res 2005;13:928–35.[Medline]
  22. Barlow SE, Dietz WH. Obesity evaluation and treatment: expert committee recommendations. The Maternal and Child Health Bureau, Health Resources and Services Administration and the Department of Health and Human Services. Pediatrics 1998;102:E29.[Medline]
  23. Cole T. Weight-stature indices to measure underweight, overweight, and obesity. In: Hines JH, ed. Anthropometric assessment of nutritional status. New York, NY: Wiley-Liss, 1991:83–111.
  24. Kuczmarski RJ, Ogden CL, Guo SS, et al. CDC growth charts for the United States: methods and development. Vital Health Stat 11 2002;(246):1–190.
  25. Tanner JM. Normal growth and techniques of growth assessment. Clin Endocrinol Metab 1986;15:411–51.[Medline]
  26. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–9.[Medline]
  27. Ferrannini E, Mari A. How to measure insulin sensitivity. J Hypertens 1998;16:895–906.[Medline]
  28. Huang TT, Johnson MS, Goran MI. Development of a prediction equation for insulin sensitivity from anthropometry and fasting insulin in prepubertal and early pubertal children. Diabetes Care 2002;25:1203–10.[Abstract/Free Full Text]
  29. Diagnosis and classification of diabetes mellitus. Diabetes Care 2004;27(suppl):S5–10.[Medline]
  30. Barton BA, Eldridge AL, Thompson D, et al. The relationship of breakfast and cereal consumption to nutrient intake and body mass index: the National Heart, Lung, and Blood Institute Growth and Health Study. J Am Diet Assoc 2005;105:1383–9.[Medline]
  31. Kahn BB, Flier JS. Obesity and insulin resistance. J Clin Invest 2000;106:473–81.[Medline]
  32. Williams DE, Cadwell BL, Cheng YJ, et al. Prevalence of impaired fasting glucose and its relationship with cardiovascular disease risk factors in US adolescents, 1999-2000. Pediatrics 2005;116:1122–6.[Abstract/Free Full Text]
  33. Nichols GA, Hillier TA, Brown JB. Progression from newly acquired impaired fasting glusose to type 2 diabetes. Diabetes Care 2007;30:228–33.[Abstract/Free Full Text]
  34. Hoffman RP, Stumbo PJ, Janz KF, Nielsen DH. Altered insulin resistance is associated with increased dietary weight loss in obese children. Horm Res 1995;44:17–22.[Medline]
  35. Schwartz MW, Boyko EJ, Kahn SE, Ravussin E, Bogardus C. Reduced insulin secretion: an independent predictor of body weight gain. J Clin Endocrinol Metab 1995;80:1571–6.[Abstract/Free Full Text]
  36. Siegel MJ, Hildebolt CF, Bae KT, Hong C, White NH. Total and intraabdominal fat distribution in preadolescents and adolescents: measurement with MR imaging. Radiology 2007;242:846–56.[Abstract/Free Full Text]
  37. Bingham SA, Gill C, Welch A, et al. Validation of dietary assessment methods in the UK arm of EPIC using weighed records, and 24-hour urinary nitrogen and potassium and serum vitamin C and carotenoids as biomarkers. Int J Epidemiol 1997;26(suppl):S137–51.[Abstract/Free Full Text]
  38. Robbins DC, Andersen L, Bowsher R, et al. Report of the American Diabetes Association's Task Force on standardization of the insulin assay. Diabetes 1996;45:242–56.[Abstract]
  39. Lustig RH, Mietus-Snyder ML, Bacchetti P, Lazar AA, Velasquez-Mieyer PA, Christensen ML. Insulin dynamics predict body mass index and z-score response to insulin suppression or sensitization pharmacotherapy in obese children. J Pediatr 2006;148:23–9.[Medline]
  40. Klein DJ, Cottingham EM, Sorter M, Barton BA, Morrison JA. A randomized, double-blind, placebo-controlled trial of metformin treatment of weight gain associated with initiation of atypical antipsychotic therapy in children and adolescents. Am J Psychiatry 2006;163:2072–9.[Abstract/Free Full Text]
  41. Morrison JA, Cottingham EM, Barton BA. Metformin for weight loss in pediatric patients taking psychotropic drugs. Am J Psychiatry 2002;159:655–7.[Abstract/Free Full Text]
  42. Freemark M, Bursey D. The effects of metformin on body mass index and glucose tolerance in obese adolescents with fasting hyperinsulinemia and a family history of type 2 diabetes. Pediatrics 2001;107:E55.[Medline]
  43. Glueck CJ, Aregawi D, Winiarska M, et al. Metformin-diet ameliorates coronary heart disease risk factors and facilitates resumption of regular menses in adolescents with polycystic ovary syndrome. J Pediatr Endocrinol Metab 2006;19:831–42.[Medline]
  44. Glueck CJ, Wang P, Fontaine R, Tracy T, Sieve-Smith L. Metformin to restore normal menses in oligo-amenorrheic teenage girls with polycystic ovary syndrome (PCOS). J Adolesc Health 2001;29:160–9.[Medline]
Received for publication January 7, 2008. Accepted for publication June 6, 2008.




This article has been cited by other articles:


Home page
 Diabetes CareHome page
T. C. Adam, C. Toledo-Corral, C. J. Lane, M. J. Weigensberg, D. Spruijt-Metz, J. N. Davies, and M. I. Goran
Insulin Sensitivity as an Independent Predictor of Fat Mass Gain in Hispanic Adolescents
Diabetes Care, November 1, 2009; 32(11): 2114 - 2115.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental data
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Morrison, J. A
Right arrow Articles by Wang, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Morrison, J. A
Right arrow Articles by Wang, P.
Agricola
Right arrow Articles by Morrison, J. A
Right arrow Articles by Wang, P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS