Abnormal Glycemic Profiles by CGMS in Obese First-Degree Relatives of Type 2 Diabetes Mellitus Patients

Abstract

Introduction:

The conventionally used oral glucose tolerance test (OGTT) has been the mainstay for diagnosis of diabetes and prediabetes. However, recent studies have indicated that a continuous glucose monitoring system (CGMS) could detect impaired glycemia much earlier than OGTT, especially in certain groups. We aimed to study the 24-h glucose profile of high-risk obese first-degree relatives of type 2 diabetes patients by CGMS and ascertain if it was better than OGTT for early detection of type 2 diabetes.

Subjects and Methods:

CGMS data of 20 subjects each in normal glucose tolerance (NGT), impaired fasting glucose (IFG), impaired glucose tolerance (IGT), and newly detected diabetes mellitus (NDDM) groups were obtained. We considered minimum, maximum, mean, and range of glucose levels as well as number, duration, and area under the curve (AUC) for excursions.

Results:

We found three (15%) NGT, seven (35%) IFG, and eight (40%) IGT subjects showed excursions in the diabetes range, whereas 18 (90%) NGT and 17 (85%) pure IFG subjects showed excursions in the IGT range. The maximum glucose values for NGT and IFG subjects were 176.0±41.4 mg/dL and 186.5±39.3 mg/dL, respectively, which is much above the present 2-h OGTT cutoff limit of 140 mg/dL. However, the average number of excursions and AUC of excursions did not differ significantly among the NGT, IFG, and IGT groups. The differences in the duration of excursion between NGT subjects with IFG values and NGT subjects with IGT values were statistically significant for an excursion limit of 140 mg/dL. However, this did not differ significantly between the IFG and IGT groups.

Conclusions:

CGMS indicated the presence of significant dysglycemia in first-degree relatives of diabetes patients without diabetes who were centrally obese. Hence it could be useful for early identification of individuals at greater risk of diabetes. A deranged glycemic profile may precede onset of overt diabetes by a long time, which may partly explain why some patients with new-onset type 2 diabetes or even prediabetes present with vascular complications at the outset.

Introduction

T he diagnosis and classification of diabetes and prediabetes until now has been largely based on oral glucose tolerance test (OGTT) values¹ and recently on glycosylated hemoglobin (HbA1c) levels also.² The variations in glucose levels occurring throughout a day remains largely undetermined by the conventionally used OGTT in diabetes, prediabetes, and so-called normoglycemic subjects. Recent studies have shown that specific glucose parameters like postprandial hyperglycemia^3
–5 and acute glucose spikes^6,7 occurring in a patient throughout the day increase their cardiovascular risk independently. Detection of glucose values at 0 and 2 h by OGTT has the potential of missing these glucose peaks. Identification of these spikes by a continuous glucose monitoring system (CGMS) may enable us to detect a larger number of individuals who are not only at a higher cardiovascular risk but also at a greater risk of developing diabetes, much earlier than OGTT can detect, and hence help formulate appropriate preventive and therapeutic strategies.

Besides being widely used in diabetes patients to facilitate glycemic control of brittle diabetes states,⁸ CGMS has also been used in patients having type 1 diabetes, especially in patients having cystic fibrosis^9,10 and in obese children,¹¹ to detect early glucose abnormalities. In studies conducted in cystic fibrosis patients it has been observed that patients found to have impaired fasting glucose (IFG) or impaired glucose tolerance (IGT) on the conventionally used 75-g OGTT have glucose excursions in the diabetes range and that so-called normoglycemic subjects have glucose excursions in the prediabetes as well as the diabetes range.⁹ Many of the subjects having diabetes or IGT by CGMS but not by OGTT subsequently converted to diabetes or IGT after an average follow-up of 2.3 years.¹⁰ A very high prevalence of glucose derangements has been found by CGMS that were not detected on OGTT in obese children.¹¹ Nocturnal hypoglycemia was noted after overnight fasting in many of them.¹¹ There has also been a recent report of deranged glucose profiles in adults without diabetes.¹²

In view of the risk associated with postprandial glycemic peaks and availability of drugs to target them, it would be important to know the extent to which 24-h glucose profiles are abnormal in individuals without diabetes but at high risk of diabetes. We therefore studied the 24-h CGMS glucose profiles of obese first-degree relatives of type 2 diabetes mellitus patients who were classified as having normal glucose tolerance (NGT) or prediabetes based on OGTT. Our aim was to determine whether CGMS is useful in the early detection of dysglycemia in obese individuals at risk of developing type 2 diabetes mellitus that remains undetected after OGTT.

Subjects and Methods

First-degree relatives of type 2 diabetes mellitus patients also having central obesity (using World Health Organization criteria for obesity in Asians¹³) were recruited from the diabetes and endocrine clinic of our hospital from March 2009 to April 2010. Subjects recruited were more than 20 years old and were not suffering from any acute illness or systemic chronic illness. Those having a history of symptoms suggestive of diabetes or on drugs likely to affect blood glucose levels were also excluded from the study. A 75-g OGTT was done in all study subjects and they were categorized as having NGT (Group 1), isolated IFG (Group 2), IGT (Group 3), or newly detected diabetes mellitus (NDDM) (Group 4) using American Diabetes Association criteria: NGT as having fasting plasma glucose 100 mg/dL and postprandial glucose <140 mg/dL, isolated IFG as fasting plasma glucose ≥100 mg/dL and <126 mg/dL with postprandial glucose <140 mg/dL, IGT as fasting plasma glucose <126 mg/dL with postprandial glucose ≥140 mg/dL and <200 mg/dL, and NDDM as fasting plasma glucose ≥126 mg/dL or postprandial glucose ≥200 mg/dL. CGMS (CGMS^® System Gold™; Medtronic Minimed, Northridge, CA) studies were carried out in 80 subjects, 20 assigned to each group, after obtaining their informed consent. The study protocol was approved by our institutional Ethical Committee. Study subjects were admitted, and CGMS was used with subjects in the fasting state. Fasting samples for measurement of HbA1c, lipid profiles, and insulin were also taken. HbA1c was measured by the high-performance liquid chromatography method (Bio-Rad, Hercules, CA). Serum total cholesterol and triglyceride were estimated by commercially available kits (Labkit; Merck, Barcelona, Spain). High-density lipoprotein-cholesterol was estimated by the direct method (Accurex Biomedicals, Mumbai, India). The low-density lipoprotein-cholesterol was calculated by the equation of Friedwald et al.¹⁴ Insulin level was determined by an immunoradiometric assay (Immunotech; Beckman Coulter, Fullerton, CA). After 24 h the CGMS was taken out, and all data were downloaded to a PC as per the manufacturer's protocol.

Glucose calibration was done four times a day using a OneTouch^® glucometer (LifeScan, a Johnson & Johnson Company, Milpitas, CA). All study subjects were given a 1,600 kcal/day diet divided into two major meals and two snacks. The following parameters were specifically analyzed while studying the 24-h glucose profiles: CGMS mean glucose value, minimum glucose value, and maximum glucose values with range and glucose excursions for ≥200 mg/dL and ≥140 mg/dL limits including the number, duration, and area under the curve (AUC) of excursions. We considered a CGMS glucose value of ≥200 mg/dL twice in 24 h as CGMS diabetes and a CGMS glucose value of ≥140 mg/dL twice in 24 h as CGMS IGT.

Statistical analysis

The data in all the groups were compared using the one-way analysis of variance test and the repeated-measure analysis of variance test followed by Tukey's test to determine differences in the 24-h glucose profile and glucose excursions within study groups. Correlations between various CGMS glucose values and excursion parameters with HbA1c and OGTT 0-h and 2-h values were calculated.

Results

Most of our study subjects were in their fifth decade, with a mean age of 46.3±8.8 years. Thirty-five percent of them were females in all groups, and the remaining 65% were males. No significant difference was found in their age, body mass index, and waist circumference or any other baseline parameters between any of the four study groups as can be seen from Table 1.

Table 1.

Baseline Parameters of Study Subjects

Variable	NGT (n=20)	IFG (n=20)	IGT (n=20)	NDDM (n=20)	P value
Age (years)	45.7±10.6	47.6±8.6	45.5±8.4	48.2±7.7	<0.05
Sex (male/female)	7/13	7/13	7/13	7/13	—
BMI (kg/m²)	29.01±4.2	27.94±3.5	28.92±2.42	29.85±3.59	<0.05
Waist circumference (cm)	94.4±6.5	92.3±5.1	94.2±5.9	97.4±6.0	<0.05
OGTT 0 h (mg/dL)	86.05±8.7	106.5±6.3	103.8±13.6	175.6±50.5	<0.05
HbA1c (%)	5.52±0.5	5.94±0.64	5.65±0.47	8.98±2.13	<0.05
Insulin (fasting) (μg/dL)	9.25±5.48	6.67±4.73	9.29±10.56	7.67±6.3	<0.05

Data are mean±SD values except for sex distribution.

BMI, body mass index; HbA1c, glycosylated hemoglobin; IFG, impaired fasting glucose; IGT, impaired glucose tolerance; NDDM, newly diagnosed diabetes mellitus; NGT, normal glucose tolerant; OGTT 0 h, fasting glucose value in the oral glucose tolerance test.

Table 2 shows the mean glucose, mean SD, mean IQR (interquartile range), mean % CV of glucose fluctuation, maximum glucose, and the range of fluctuation as markers for their total glycemic burden in each of the study groups. This showed a progressive increase in derangement of glucose homeostasis from NGT through IFG to IGT, and many of the NGT subjects had maximum glucose values crossing the 2-h OGTT cutoff for diabetes or IFG/IGT criteria. IGT subjects included those with or without IFG, and these subgroups were not analyzed separately and compared, because of the small numbers. Prediabetes subjects also crossed the OGTT criteria for diabetes, namely, 200 mg/dL.

Table 2.

Continuous Glucose Monitoring System Glucose Values of the Study Subjects

Glucose parameter (mg/dL)	NGT (n = 20)	IFG (n = 20)	IGT (n = 20)	P value ^a	NDDM (n = 20) ^b
Mean glucose	124.3 ± 15.3	137.1 ± 21.5	140.8 ± 26.2	>0.05	238.8 ± 67.5
SD	18.0 ± 5.9	21.4 ± 5.6	23.4 ± 11.4	>0.05	42.1 ± 20.9
IQR	28.8 ± 19.1	29.0 ± 18.8	31.1 ± 24.8	>0.05	71.8 ± 60.9
%CV	14.4 ± 4.6	15.6 ± 3.3	16.8 ± 8.3	>0.05	17.8 ± 7.4^c
Minimum	89.1 ± 16.0	92.1 ± 23.8	97.3 ± 25.4	>0.05	151.0 ± 63.4
Maximum	176.0 ± 41.4	186.5 ± 39.3	194.9 ± 50.9	>0.05	328.8 ± 74.0
Range	86.9 ± 47.3	94.4 ± 38.6	97.6 ± 56.2	>0.05	177.4 ± 72.4

P value for groups with NGT, IFG and IGT.

P value for the NDDM group versus all other groups was significant for all parameters except % CV.

P value for %CV for NDDM >0.05 versus all other groups.

IFG, impaired fasting glucose; IGT, impaired glucose tolerance; NDDM, newly detected diabetes mellitus; NGT, normal glucose tolerance.

Table 3 shows the CGMS excursion parameters like number of excursions and AUC of excursions of the study subjects, which serve as indicators of their acute glucose fluctuations in the total 24-h study period. Eighteen of these 60 subjects without diabetes were found to have excursion into the diabetes range. The AUC of excursions in subjects who had excursions for both glucose limits ≥200 mg/dL and ≥140 mg/dL, although not found to be significantly different among NGT, IFG, and IGT groups, was much less in NGT subjects in comparison with IFG/IGT subjects. Similarly, the AUC of glucose excursions was significantly lower in IFG and IGT groups compared with the NDDM group, for either glucose excursion limits. When all groups were combined, the glucose AUC of excursions for both glucose limits correlated strongly with HbA1c levels and OGTT 0-h and 2-h glucose levels (P<0.001 for all three). However, the number of excursions did not correlate with either HbA1c or OGTT values.

Table 3.

Continuous Glucose Monitoring System Excursion Parameters for the Two Glucose Limits

Excursion limit, parameter^e	NGT (n = 20)	IFG (n = 20)	IGT (n = 20)	P value ^a	NDDM (n = 20) ^b
200 mg/dL
Number of patients having excursions	3 (15%)	7 (35%)	8 (40%)	—	15 (75%)
Number of excursions	2.0 ± 1.0	2.5 ± 2.2	2.0 ± 1.7	>0.05	2.7 ± 1.1
Area Under Curve (AUC) (mg/dL·day)	2.6 ± 1.1	5.0 ± 4.8	14.5 ± 11.7	>0.05	100.2 ± 94.0
Duration of excursions as fraction of day	0.05 ± 0.03	0.08 ± 0.10	0.11 ± 0.13	>0.05	0.61 ± 0.35
Average amplitude of excursion (mg/dL)	64.3 ± 36.6	17.8 ± 20.8	32.5 ± 49.6	>0.05	140.9 ± 125.7^c
140 mg/dL
Number of patients having excursions	18 (90%)	17 (85%)	19 (95%)	—	11 (55%)
Number of excursions	2.9 ± 1.8	4.0 ± 2.3	3.3 ± 2.1	>0.05	2.1 ± 1.6
Area Under Curve (AUC) (mg/dL.day)	9.8 ± 10.8	21.9 ± 25.9	22.4 ± 30.2	>0.05	122.1 ± 114.6
Duration of excursions as fraction of a day	0.13 ± 0.15	0.40 ± 0.32	0.41 ± 0.37	1vs 2 = 0.041	0.85 ± 0.27
				1vs 3 = 0.024
				2vs 3 = 0.998
Average amplitude of excursion (mg/dL)	116.1 ± 151.5	54.5 ± 61.7	35.8 ± 30.3	>0.05	158.5 ± 169.3^d

P value for groups with NGT, IFG, and IGT.

P value for the NDDM group versus all other groups was significant except for average amplitude of excursion.

P value for NDDM versus NGT and IGT not significant.

P value for NDDM versus NGT and IFT not significant.

All parameters viz. Number of Excursions, Area Under Curve, Duration of Excursions, and Average Amplitude of Excursions are given only for the subjects having one or more excursions above the limits of 200 mg/dL or 140 mg/dL respectively.

IFG, impaired fasting glucose; IGT, impaired glucose tolerance; NDDM, newly detected diabetes mellitus; NGT, normal glucose tolerance.

The study subjects were very different in respect to their duration of excursions for the two excursion limits. The IFG and IGT groups spent 9.66±7.8 h and 9.94±8.9 h, respectively, above the glucose cutoff of ≥140 mg/dL, compared with 3.16±3.7 h in the NGT group. Duration of excursions for the same two groups were 2.09±2.5 h and 2.85±3.1 h, respectively, above the 200 mg/dL cutoff, compared with 1.26±0.9 h for the NGT group. The NDDM patients had the highest time spent above both the glucose limits, as expected (i.e., mean duration of 14.7 and 20.6 h for ≥200 and ≥140 mg/dL limits, respectively). The duration of excursions was significantly higher for both IFG (P=0.04) and IGT (P=0.024) compared with NGT subjects for a cutoff limit of 140 mg/dL. There was no statistically significant difference between the IFG and IGT groups (P=0.999). Duration of excursions for both of the glucose excursion limits ≥140 mg/dL and ≥200 mg/dL was found to correlate strongly with the HbA1c (r=0.62 and 0.80, respectively), OGTT preprandial (r=0.58 and 0.74, respectively), and 2-h postprandial glucose (r=0.64 and 0.82, respectively) values with P<0.001 for all three parameters. Although CGMS diabetes mellitus was observed in 5% of NGT, 10% of IFG, and 10% of IGT subjects, CGMS IGT was found in 40% of NGT and 55% of IFG subjects.

Discussion

The present study found significant abnormalities in 24-h glucose profiles that reflected a far greater glycemic burden than indicated by OGTT alone. Both normoglycemic subjects and prediabetes subjects were having excursions into the higher OGTT dysglycemic range when studied by CGMS. Thirty percent of our study subjects who were classified as not having diabetes or having prediabetes by OGTT were having excursions into the diabetes range. Similarly, 90% of subjects classified as NGT by OGTT had excursions into the IGT range or higher, and 15% had excursions into the diabetes range. Their total burden of dysglycemia over a 24-h period as indicated by number, duration, and AUC of excursions was much higher than that expected from OGTT. NGT subjects remained above the IGT threshold for more than 3 h, during which they were also found crossing the diabetes threshold, whereas the prediabetes subjects spent over 9 h above the IGT threshold and 2.5 h above the diabetes threshold, respectively. The acute glucose fluctuations were found to occur as postprandial spikes as well as spikes not related to meals in these otherwise healthy normal glucose-tolerant subjects. These findings could have a major impact on our understanding of the overall glycemic burden in individuals and how this could influence complications related to dysglycemia.

Borg et al.¹² reported significant dysglycemic profiles by CGMS in subjects without diabetes. They found IGT in 93% and diabetes in 12.5% of cases, based on their 24-h glycemic profile. However, this study relied only on their fasting glucose and HbA1c values, which in the absence of OGTT may not have ruled out diabetes in many of them. We performed OGTT in all our study subjects, who were then correctly classified into NGT, IFG, IGT, and NDDM before CGMS studies were carried out. This allowed a more accurate interpretation of dysglycemia based on clearly defined categories of glucose tolerance. The previous study in obese children¹¹ reported that 11% of them had pathological glucose excursions by using CGMS.

Another interesting observation of this study is that over 85% of our subjects classified as IFG by OGTT crossed the IGT threshold by CGMS and remained in that range for more than 9 h. These isolated IFG subjects had a mean of four episodes of hyperglycemia above their conventional IFG limit. Their glycemic profile also peaked around three times into the OGTT diabetes limits during the same 24-h time period. This suggests that there is significant postprandial hyperglycemia even in subjects with IFG that remains undetected by OGTT. In fact, based on AUC excursion data it appears that the postprandial hyperglycemic burden over a 24-h period in IFG subjects is nearly as much as in that with IGT. The issue of postprandial dysglycemia in IFG subjects has remained unaddressed even by the study of Borg et al.,¹² and we report for the first time a similar degree of postprandial dysglycemia by CGMS in adult-onset IFG and IGT subjects.

A high degree of glucose variability was observed in this study in subjects with NGT and prediabetes. Glucose variability has been shown in recent studies to be an independent risk factor for diabetes complications and may further contribute to the cardiovascular risk in these individuals.

Our study has some limitations. CGMS was performed in hospitalized patients on a fixed diet schedule and may not have taken into account the effects of physical activity and other lifestyle factors. Also, the study was cross-sectional in nature, and the true significance of our observations can only be assessed through longitudinal outcome studies. Finally, our study population consisted of subjects at a higher risk for diabetes, and it is possible that those with average risk may not display similar degrees of dysglycemia.

Conclusions

High-risk individuals with NGT and prediabetes display significantly abnormal 24-h glucose profiles by CGMS that may contribute to their diabetes and cardiovascular risk. It would appear that OGTT underestimates this risk and that CGMS may serve as a better tool to detect early dysglycemia. If this finding is confirmed by long-term follow-up studies, CGMS may help identify, with greater sensitivity, subjects at risk for diabetes and cardiovascular disease and create early opportunities for interventions.

Footnotes

Acknowledgments

The authors are grateful to Mr. Aizaz Siddiqui, Mr. Mohammad Aslam, Mr. Aiman Abbas, and the staff of the hospital laboratory services for their unconditional support in this project. This work was supported by a grant from Research Society for the Study of Diabetes in India.

Author Disclosure Statement

No competing financial interests exist.

References

American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care, 2008; 31,Suppl 1:S55–S60.

American Diabetes Association. Standards of medical care in diabetes—2010. Diabetes Care, 2010; 33,Suppl 1:S11–S61.

Bonora

, Muggeo

. Postprandial blood glucose as a risk factor for cardiovascular disease in Type II diabetes: the epidemiological evidence. Diabetologia, 2001; 44:2107–2114.

Ceriello

, Hanefeld

, Leiter

, Monnier

, Moses

, Owens

, Tajima

, Tuomilehto

. Postprandial glucose regulation and diabetic complications. Arch Intern Med, 2004; 164:2090–2095.

Shaw

, Hodge

, de Courten

, Chitson

, Zimmet

. Isolated post-challenge hyperglycaemia confirmed as a risk factor for mortality. Diabetologia, 1999; 42:1050–1054.

Moore

. Glucose fluctuations and oxidative stress. JAMA, 2006; 296:1730–1731.

Sampson

, Gopaul

, Davies

, Hughes

, Carrier

. Plasma F2 isoprostanes: direct evidence of increased free radical damage during acute hyperglycemia in type 2 diabetes. Diabetes Care, 2002; 25:537–541.

Petrovski

, Dimitrovski

, Milenkovic

. Insulin pump therapy with continuous glucose monitoring improves metabolic control in brittle type 1 diabetes. Prilozi, 2007; 28:129–135.

Moreau

, Weiller

, Rosner

, Weiss

, Hasselmann

, Pinget

, Kessler

. Continuous glucose monitoring in cystic fibrosis patients according to the glucose tolerance. Horm Metab Res, 2008; 40:502–506.

10.

Schiaffini

, Brufani

, Russo

, Fintini

, Migliaccio

, Pecorelli

, Bizzarri

, Lucidi

, Cappa

. Abnormal glucose tolerance in children with cystic fibrosis: the predictive role of continuous glucose monitoring system. Eur J Endocrinol, 2010; 162:705–710.

11.

Zou

, Liang

, Hong

, Zhao

. Glucose metabolism disorder in obese children assessed by continuous glucose monitoring system. World J Pediatr, 2008; 4:26–30.

12.

Borg

, Kuenen

, Carstensen

, Zheng

, Nathan

, Heine

, Nerup

, Borch-Johnsen

, Witte

. Real-life glycaemic profiles in non-diabetic individuals with low fasting glucose and normal HbA1c: the A1C-Derived Average Glucose (ADAG) study. Diabetologia, 2010; 53:1608–1611.

13.

The World Health Organization Western Pacific Region, The International Association for the Study of Obesity, The International Obesity Task Force: The Asia-Pacific Perspective: Redefining Obesity and Its Treatment. Sydney: Health Communications Australia Pty Ltd., 2000.

14.

Friedewald

, Levy

, Fredrickson

. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 1972; 18:499–502.