| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cincinnati Children’s Hospital Medical Center (D.A.E., L.M.D.), Department of Pediatrics, Division of Endocrinology, University of Cincinnati School of Medicine, Cincinnati, Ohio 45229; Geriatric Research, Education, and Clinical Center (R.L.P.), Baltimore Veterans Affairs Medical Center and Division of Gerontology, University of Maryland School of Medicine, Baltimore, Maryland 21201; Barbara Davis Center (R.P.W.), Denver, Colorado 80262; Department of Medicine (D.A.D.), Division of Endocrinology, University of Cincinnati, Cincinnati, Ohio 45267; and Cincinnati Veterans Administration Hospital (D.A.D.), Cincinnati, Ohio 45220
Address all correspondence and requests for reprints to: Deborah A. Elder, M.D., Cincinnati Children’s Hospital Medical Center Department of Pediatrics, Division of Endocrinology, 3333 Burnet Avenue, MLC 7012, Cincinnati, Ohio 45229. E-mail: deborah.elder{at}cchmc.org.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: Our objective was to characterize the metabolic defects underlying T2DM in adolescents and young adults.
Design, Setting, and Patients: We conducted a cross-sectional study of islet function and insulin sensitivity in 16 adolescents with T2DM and 13 obese (OB) and 13 lean (LN) age-matched nondiabetic subjects at a University Medical Center.
Intervention: We provided oral and iv glucose tolerance tests.
Main Outcome Measures: We measured insulin and glucagon levels, insulin sensitivity, acute insulin responses to iv glucose, and the ratio of proinsulin to immunoreactive insulin.
Results: The diabetic subjects had elevated fasting insulin levels and significantly reduced insulin sensitivity (P < 0.05). The acute insulin response to iv glucose was comparable in the T2DM and LN groups (P < 0.05 for the OB vs. LN and T2DM), but insulin secretion adjusted for insulin resistance, the disposition index, was severely impaired in the diabetic subjects (P < 0.05 for the T2DM vs. LN and OB). The ratio of proinsulin to immunoreactive insulin did not differ among the three groups in the basal or stimulated state. Plasma glucagon levels were comparable before and after ingestion of glucose.
Conclusions: These findings demonstrate that diabetic adolescents have significant insulin resistance, even compared with subjects of similar obesity and body fatness, and impaired insulin secretion relative to their degree of insulin resistance. However, the adolescent diabetic subjects retained a first-phase insulin response to glucose that was comparable to lean controls and did not have hyperproinsulinemia or hyperglucagonemia.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In adults, ß-cell dysfunction in the setting of insulin resistance, due in part to increased body weight, is the hallmark of T2DM. It has been generally assumed that this is also the case in adolescents. Despite the alarming demographics of T2DM, the pathogenesis of this condition in the pediatric age group has only recently been addressed (9, 10). Based on what is known about the characteristic abnormalities in diabetic adults, we hypothesized that adolescents with T2DM would have significant insulin resistance, due in great part to obesity, and islet cell dysfunction characterized by loss of first-phase insulin release (FPIR), hyperproinsulinemia and elevated glucagon secretion. In the present study, we characterized islet function and insulin sensitivity in 16 diabetic and 13 weight-matched control subjects representative of adolescents seen at our hospital.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Sixteen patients with T2DM were recruited from our diabetes clinic. Thirteen age- and weight-matched nondiabetic subjects (OB) and 13 lean nondiabetic subjects (LN) were recruited from the Cincinnati Children’s hospital campus. None of the control subjects had any ongoing medical conditions. All patients and control subjects gave informed consent for participation in the study, which was approved by Cincinnati Children’s Hospital Medical Center (CCHMC) Institutional Review Board. Demographic characteristics of the three study groups are given in Table 1
.
|
At their initial study visit, all subjects had oral glucose tolerance tests (OGTT), and dual-energy x-ray absorptiometry scans to determine body composition. They returned 1–2 wk later for an iv glucose tolerance test (IVGTT). Before the IVGTT, the diabetic subjects were asked to discontinue their oral medications (at least 7 d for metformin and 14 d for the one subject taking metformin and pioglitazone) before testing. This period of drug withdrawal was chosen to allow clearance of the active compounds from the circulation (12). The T2DM subjects received strict dietary recommendations and were encouraged to follow these guidelines during the drug withdrawal to maintain a target fasting blood glucose of less than 7.0 mmol/liter (126 mg/dl). Home glucose monitoring was reviewed to assess compliance and assist in reaching the glucose target. Four subjects required intermittent short-acting insulin to maintain a fasting blood glucose concentration of less than 7.0 mmol/liter (126 mg/dl) during the period of drug withdrawal. The two patients who used insulin as part of their usual regimen continued this therapy until the day before they were to have their IVGTT; one taking insulin Glargine received the last dose 36 h before testing, and another who used NPH insulin had the last dose 24 h before testing. Mean fasting glucose concentrations based on home glucose monitoring were 7.0 ± 2.8 mmol/liter (126 ± 50.4 mg/dl) in the week before the OGTT and 5.8 ± 1.2 mmol/liter (100.4 ± 21.6) in the week before the IVGTT.
Metabolic testing
OGTT. Subjects reported to the General Clinical Research Center (GCRC) at CCHMC in the morning after an overnight fast. After the collection of fasting blood samples, subjects ingested an oral glucose solution (1.75 g/kg, maximum 75 g) within 5 min. Blood samples for glucose, insulin, C-peptide, and glucagon were obtained before and 10, 20, 30, 45, 60, 75, 90, 120, 150, and 180 min thereafter.
IVGTT. Individuals were admitted to the GCRC in the evening, had a standard supper, and were then fasted overnight. Four subjects with diabetes who had elevated blood glucose levels after the meal were treated with short-acting insulin (Insulin Lispro; Lilly, Indianapolis, IN) from 2100–2400 h to decrease plasma glucose concentrations to the target of less than 8.3 mmol/liter (150.0 mg/dl); all other diabetic subjects had overnight glucose levels less than 8.3 mmol/liter without treatment. At 0800 h, three basal blood samples were taken, and iv glucose (250 mg/kg of a 50% solution of dextrose) was infused over 30 sec. Blood samples were drawn at 2, 3, 5, 7, 10, 12, 14, 16, and 19 min thereafter. At 20 min, insulin was administered as an iv infusion over 5 min. To accommodate for expected insulin resistance, the diabetic subjects received 0.06 U/kg of insulin during the IVGTT. LN and OB nondiabetic subjects received 0.015–0.03 U/kg. Additional blood samples were obtained at 22, 25, 30, 35, 40, 50, 60, 70, 80, 90, 100, 120, 140, 160, 180, and 240 min. Blood samples were placed on ice and centrifuged within 1 h. Plasma glucose was measured immediately using an automated glucose analyzer (YSI, Yellow Springs, OH). The remaining plasma was stored at –20 C until assayed.
Biochemical measurements
Glucose was measured using a glucose oxidase method. Insulin was measured by RIA using an antiinsulin serum raised in guinea pigs, 125I-labeled insulin as tracer, and a double-antibody method of separating bound from free peptide. The sensitivity of this assay is 2 pM, the intra- and interassay coefficients of variation are 5 and 7%, respectively, and proinsulin is recognized equally to insulin. C-peptide, glucagon, and proinsulin were measured using commercial RIA kits (Linco, Inc., St. Louis, MO). The antiserum used in the proinsulin assay reacts with intact proinsulin and the proinsulin split product des[31, 32]Proinsulin, but does not recognize insulin, C-peptide, or des[64, 65]Proinsulin.
Calculations
The glucose, insulin, C-peptide, and glucagon responses to oral glucose were computed as the incremental area under the curve (AUC) above fasting values. The insulin secretion ratio during the OGTT was calculated as the 30-min insulin or C-peptide increment divided by the plasma glucose at 30 min (13, 14). Proinsulin to insulin ratios were calculated for each subject in the basal state using the mean values of two fasting samples and in a stimulated state using the AUC in the first 10 min after iv glucose. Indices of insulin sensitivity (SI) and glucose effectiveness were determined from the glucose and insulin values during the IVGTT using the minimal model (15, 16, 17). The acute insulin response to glucose (AIRg) was computed as the average increment above basal for samples obtained in the first 10 min of the IVGTT. The disposition index (DI), was computed as the product of AIRg and SI (15, 18). The glucose disappearance constant (kg) was calculated as the slope of the natural logarithm of the glucose values from 10–19 min during the IVGTT.
Statistical analysis
Unless otherwise noted, values are presented as mean ± SEM. Parameters of glucose tolerance and hormone secretion were compared among the three groups using ANOVA with post hoc comparisons to determine specific differences among the groups; P values of <0.05 were taken to be significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Table 1
shows the characteristics of the three subject groups at the time of metabolic testing (mean ± SEM). Important clinical features of the 16 subjects with T2DM at the time of diagnosis and testing are shown in Table 2 (mean ± SD).
|
Plasma concentrations of glucose, insulin, and C-peptide before and after ingestion of oral glucose are shown in Fig. 1. Mean fasting glucose values were higher in the T2DM compared with OB and LN subjects (Table 3), and the postprandial glycemic excursion was also significantly greater (Fig. 1). Fasting insulin levels were higher in the T2DM group than the OB controls and lowest in the LN subjects, and fasting C-peptide showed a similar pattern, although the difference between the OB and LN group was not significantly different (Table 3). Over the course of the entire OGTT, the integrated insulin and C-peptide responses did not differ among the three groups (Fig. 1). However, the incremental C-peptide:glucose (T2DM, 0.16 ± 0.08; OB, 0.47 ± 0.08; LN, 0.37 ± 0.05; P < 0.001) and insulin:glucose (T2DM, 42 ± 9; OB, 82 ± 11; LN, 60 ± 6; P = 0.017) responses in the first 30 min of the test were significantly lower in the T2DM group than the controls. Fasting plasma glucagon was similar in the three groups of subjects (Table 3), and the degree of suppression after glucose ingestion also did not differ among the three groups (T2DM, –1016 ± 885; OB, –1264 ± 1062; LN, –723 ± 747 ng/liter; P = 0.65).
|
|
Correction of hyperglycemia induced by the iv glucose bolus occurred at a reduced rate in the T2DM subjects compared with the OB and LN groups (kg: T2DM, 1.2 ± 0.2; OB, 2.1 ± 0.2; LN, 1.8 ± 0.1%/min; P < 0.05 for T2DM vs. LN and OB). Indices of glucose effectiveness were similar in each adolescent group (T2DM, 0.021 ± 0.005; OB, 0.019 ± 0.001; LN, 0.020 ± 0.001%/min). The FPIR was present in the diabetic subjects and comparable to the LN group but lower than the OB subjects (AIRg: T2DM, 453 ± 214; OB, 1050 ± 152; LN, 543 ± 66 pmol/liter; P < 0.05 for OB vs. T2DM and LN) (Fig. 2A). The diabetic subjects were very insulin resistant with SI that were 3- to 4-fold less than age- and weight-matched controls (T2DM, 0.53 ± 0.2; OB, 1.97 ± 0.39; LN, 3.51 ± 0.47 min–1 pM–1 x 10–5; P < 0.05 for each pair-wise comparison) (Fig. 2B). Although the diabetic subjects secreted nearly as much insulin as the lean controls in response to the bolus of glucose (AIRg), this response was clearly insufficient for their degree of insulin resistance, as indicated by the markedly abnormal DI (T2DM, 217.6 ± 81.5; OB, 1803 ± 427.6; LN, 1939 ± 352.2; P < 0.05 for T2DM vs. LN and OB) (Fig. 2C). The proinsulin to immunoreactive insulin (PI/IRI) ratios were similar among the three groups in the fasting state (Table 3 and Fig. 3) and when integrated over the first 10 min after iv glucose (T2DM, 3.4 ± 1.1; OB, 1.9 ± 0.4; LN, 2.7 ± 0.7%; P = 0.35).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The diabetic participants in this study were representative of the approximately 300 adolescents with T2DM who are cared for at CCHMC. All of the participants in the present study met the diagnostic criteria for T2DM suggested by the American Diabetes Association (11, 19) and employed in other pediatric studies (2, 20): hyperglycemia, absent markers of islet autoimmunity, and features of insulin resistance including obesity, acanthosis nigricans, and elevated plasma insulin or C-peptide. Moreover, at diagnosis, the diabetic cohort had a degree of hyperglycemia that was comparable to participants in other studies of adolescents with new-onset T2DM (2, 3, 6, 21). The diabetic cohort also had a slight preponderance of African-Americans matching previous descriptions of a higher frequency of ethnic minorities in adolescents with T2DM (1, 2, 3, 4, 5). However, at our center, non-Hispanic whites comprise 40% of adolescents with T2DM documenting that all major ethnic groups in the United States are affected. True national prevalence rates and demographics of T2DM in the pediatric population have yet to be determined.
The OB group had similar body mass index (BMI) and body composition to the T2DM group, and both were made up of African-American and non-Hispanic white subjects, albeit in different proportions. Because other investigators have noted differences in insulin sensitivity and insulin clearance among African-American, Hispanic, and non-Hispanic white adolescents (22), we cannot exclude the possibility that some of the results in this study could be associated with race. However, the clear differences in insulin sensitivity and ß-cell function between the OB and the T2DM groups suggests that any such effects are small compared with the high degree of metabolic similarity within the diabetic and nondiabetic cohorts.
The diabetic subjects in this study had a marked reduction of insulin sensitivity, to approximately one fourth that of the control group of adolescents of similar weight and fat mass. This finding of insulin resistance significantly out of proportion to body weight is similar to what has been reported for adults with T2DM (23, 24, 25). However, our results differ from those of two recent reports that studied similar cohorts. Gungor and Arslanian (9), using euglycemic, hyperinsulinemic glucose clamps, demonstrated that SI in T2DM teenagers was half that of obese controls, whereas Weiss and colleagues (10) noted no differences in SI determined during hyperglycemic clamps among obese adolescents with normal, impaired, or diabetic glucose tolerance. The reason for the relative differences in insulin resistance among our T2DM subjects and those of Gungor and Weiss are unclear. Our diabetic patients were slightly older, having a mean age of 17 yr compared with 15 yr in the other studies, with a longer duration of diabetes, 28 vs. 18 (10) and 6 months (9). However, it seems unlikely that these differences account for the greater reduction in SI in our diabetic subjects. SI in our subjects was derived from the modified IVGTT with administration of iv insulin during the test to increase the precision of SI determined from the minimal model (15, 26). The diabetic subjects received a larger dose of iv insulin during the IVGTT than the controls to accommodate for their insulin resistance (17, 27), and they had higher peak plasma insulin values after insulin infusion (T2DM, 6590 ± 1492; OB, 3268 ± 659; LN, 3126 ± 928 pM; P < 0.01). In a previous study of nondiabetic subjects studied on two occasions with high and low doses of insulin given during a frequently sampled IVGTT, we noted that the estimate of SI was lower at insulin concentrations of approximately 6500 pM than it was at 3000 pM, suggesting that insulin action may be saturable (28). This raises the possibility that the estimates of SI in our diabetic patients could be artificially low and explain the relatively greater insulin resistance in our diabetic subjects compared with the other studies. However, in our previous study where a saturation effect was examined, SI was reduced only by 32% in the setting of higher insulin levels (28). Thus, even if this effect was present in our adolescent diabetic subjects, their SI would still be 2- to 3-fold lower than that of the OB group, and our major conclusions would not be altered.
Plasma insulin levels after oral and iv glucose were generally comparable in our diabetic subjects and the nondiabetic controls. Notably, FPIR in response to iv glucose was similar in the T2DM and LN subjects. However, the absolute magnitude of the insulin response in diabetic subjects can be misleading if not considered in the context of prevailing glycemia and insulin sensitivity (29, 30). Given their severe insulin resistance, and relative hyperglycemia during the OGTT, ß-cell responses in the diabetic teenagers were inadequate. The reduced levels of insulin and C-peptide 30 min after the glucose ingestion are one indication of the impaired ß-cell responses (13, 14). The failure of the AIRg to compensate for severe insulin resistance during the IVGTT is another. The severity of the ß-cell defect in our T2DM group is reflected in values of DI that were significantly lower than the values in weight-matched controls. These findings demonstrate that in adolescents with T2DM a critical determinant of glucose intolerance is the inability of the ß-cell to compensate for insulin resistance. In the study of Weiss et al. (10), ß-cell function in the diabetic subjects was comparable with our results in that FPIR was present but reduced to approximately 40% of the obese normoglycemic controls. This is in contrast to the observations of Gungor and coworkers (9) who noted a more severe defect in insulin secretion in their adolescent T2DM cohort, with a near absence of FPIR that is more comparable to what has been reported in adults with T2DM (29, 31). Interestingly, despite the apparent differences in ß-cell function between our T2DM subjects and those of this latter report, calculations of DI were approximately 8-fold less in the diabetic compared with control groups in both studies, suggesting comparable relative deficits of insulin secretion.
A second interesting feature of ß-cell function in our adolescent diabetic patients was the absence of excessive proinsulin secretion. Fasting and poststimulus hyperproinsulinemia is a common finding in adults with T2DM in whom the proinsulin/insulin ratio is generally elevated 3-fold (32, 33). Although this abnormality of proinsulin processing and secretion is most pronounced in patients with severe insulin secretory defects, it is present across the spectrum of clinically affected persons (29, 34), and both adult and adolescent subjects with impaired glucose tolerance have been reported to have increased poststimulus proinsulin/insulin ratios (35, 36). Weiss and colleagues (10) observed similar fasting proinsulin to insulin ratios in their diabetic and control subjects, consistent with our findings, although there was a transient, 2-fold increase of the PI/IRI ratio in the diabetic compared with control group during ß-cell stimulation. In keeping with the more severe ß-cell dysfunction demonstrated in the diabetic subjects studied by Gungor and Arslanian (9), the fasting PI/IRI was twice as high as obese nondiabetic controls. The compatibility of the PI/IRI with the measures of FPIR in the studies of Weiss and Gungor, as well as in our study, suggests that ß-cell function truly varied among the three diabetic cohorts.
Hyperglucagonemia is commonly found in adults with T2DM (37, 38), and excessive 
-cell secretion has been proposed to make an important contribution to fasting hyperglycemia (39). One previous study has reported abnormally elevated plasma glucagon in response to a mixed meal in hyperglycemic T2DM adolescents (40). Our T2DM subjects had fasting and post-OGTT glucagon levels that were similar to controls. Because the subjects were hyperglycemic, this finding could be interpreted as evidence for -cell dysfunction. However, the elevation of fasting glucose was only mild at the time of study, and glucose ingestion caused suppression of glucagon levels, suggesting that the -cell response to hyperglycemia was at least partially intact. Although we cannot rule out a mild defect in glucagon secretion, the maintenance of relatively normal circulating glucagon levels in our T2DM group may account for their relatively mild fasting hyperglycemia, even after medication withdrawal.
This paper is the third recent report on the metabolic characteristics of adolescents with type 2 diabetes (9, 10). All three studies involved small sample sizes of 10–16 T2DM subjects. The characteristics of the diabetic subjects in the three studies were very similar, including primarily obese, African-American and non-Hispanic white, male and female teenagers. The three groups of subjects had been treated for 6 months to several years before study and had mean hemoglobin A1c (HbA1c) values of near 7%. The majority of the diabetic subjects in the three studies were treated with metformin, with lesser numbers using insulin, other drugs, or diet alone. In each case, medications were withdrawn for several days, or in our study 1–2 wk, but glycemic control as reflected in the fasting blood glucose was only mildly elevated. Although the general results, that adolescents with type 2 diabetes are insulin resistant and have impaired insulin release, were in agreement among the studies, there were some apparent differences. The most notable were the greater relative insulin resistance and more modest ß-cell defect seen in our patients compared with those in the other studies. It is possible that methodological differences account for these discrepant results. However, previous work has demonstrated that results from minimal model analysis of IVGTT and glucose clamp techniques are highly correlated in individuals with T2DM (27). In addition, each of the studies included assessment of FPIR and fasting PI/IRI as indices of ß-cell function, and these are standard and reproducible measures. In the absence of clear explanations for the apparent variation in major metabolic parameters among the three cohorts, larger studies are needed to establish the full range of SI and ß-cell function in diabetic adolescents and the major factors affecting these measures.
In summary, the major findings of this study are that diabetic adolescents have significant insulin resistance, even compared with subjects of similar obesity and body fatness, and impaired insulin secretion relative to their degree of insulin resistance. However, the adolescent diabetic subjects in this study retained a FPIR to glucose that was comparable to lean controls and did not have hyperproinsulinemia or hyperglucagonemia. These results are generally consistent with other observations in teenagers with T2DM (9, 10), and taken together these studies implicate ß-cell dysfunction as central to adolescent T2DM. Detailed characterization of ß-cell function over time in adolescents with T2DM, to determine the stability or progression of impaired insulin secretion, will be critical for understanding the outcomes in these patients and for developing appropriate strategies to treat them.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
The authors have no conflict of interest to declare.
First Published Online November 1, 2005
Abbreviations: AIRg, Acute insulin response to glucose; AUC, area under the curve; BMI, body mass index; DI, disposition index; FPIR, first-phase insulin response; HbA1c, hemoglobin A1c; IVGTT, iv glucose tolerance test; LN, lean; OB, obese; OGTT, oral glucose tolerance test; PI/IRI, proinsulin to immunoreactive insulin; SI, insulin sensitivity; T2DM, type 2 diabetes mellitus.
Received April 19, 2005.
Accepted October 26, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-cell function in diabetes. Response to carbohydrate and protein ingestion. N Engl J Med 283:109–115[Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), OB (
), and LN adolescents (•). Incremental areas above fasting for glucose, insulin, and C-peptide are shown next to each line figure. Data are presented as mean ± SEM; P values are for comparisons of glucose AUC among the groups. *, Significantly greater than OB and LN.
, P < 0.05 vs. LN and DM; *, P < 0.05 vs. LN; **, P < 0.05 vs. OB; 