Anda di halaman 1dari 6

p r i m a r y c a r e d i a b e t e s 1 1 ( 2 0 1 7 ) 555–560

Contents lists available at ScienceDirect

Primary Care Diabetes

journal homepage: http://www.elsevier.com/locate/pcd

Original research

Rethinking the accuracy of 75 g glucose used in the


oral glucose tolerance test in the diagnosis and
management of diabetes

Sun Hee Kim, Tae Sun Park, Heung Yong Jin ∗


Division of Endocrinology and Metabolism, Department of Internal Medicine, Chonbuk National University Medical
School, Research Institute of Clinical Medicine of Chonbuk National University—Biomedical Research Institute of
Chonbuk National University Hospital, Jeonju, South Korea

a r t i c l e i n f o a b s t r a c t

Article history: Aims: We investigated the specificity of the 75 g oral glucose tolerance test (OGTT) and the
Received 31 March 2017 clinical usefulness of the 2 h post loading glucose (2 h PLG) value in the clinical care of
Received in revised form diabetes patients.
31 May 2017 Methods: The 75 g OGTT data of 1755 subjects were analyzed. The relationships and degrees of
Accepted 3 June 2017 consistency among 2 h PLG, fasting plasma glucose (FPG), and HbA1c values were assessed.
We also investigated the degree of contribution of 2 h PLG in the prescription of glucose-
lowering agents and in the pitfall group for use of 75 g OGTT.
Keywords: Results: Among 595 subjects with normal FPG, only 329 (55.3%) showed normal 2 h PLG level,
Diabetes and 66 (11.1%) patients could be considered as having diabetes. Among 454 diabetes patients
Oral glucose tolerance test (based on FPG and HbA1c), 409(90.1%) showed 2 h PLG values in the range of diabetes, 45
2 h post loading glucose (PLG) (9.9%) subjects did not exhibit diabetes. Pitfall group who used 75 g OGTT for diagnosis
diabetes (lower 2 h PLG value compared to fasting and upper 2 h PLG value compared to
fasting) showed differences in body weight and height.
Conclusions: Based on OGTT results, around 10% patients cannot be diagnosed with diabetes
based solely on the 2 h PLG value. Further studies on differences in glucose loading according
to body weight, individual life pattern, and calorie requirement are needed for improvement
of the specificity of the OGTT in the clinical management of diabetes.
© 2017 Primary Care Diabetes Europe. Published by Elsevier Ltd. All rights reserved.

toms, HbA1c value, and the 75 g oral glucose tolerance test


1. Introduction (OGTT) [1]. Among these diagnostic tools, the 75 g OGTT is
very useful in under diagnosed patients with the pre-diabetic
The diagnosis of diabetes is based on the American Diabetes
condition of impaired fasting glucose (IFG) [2]. However, some
Association (ADA) criteria using fasting plasma glucose (FPG),
patients previously diagnosed with diabetes based on the 75 g
random glucose higher than 11.1 mmol/L with typical symp-
OGTT in hospital have shown glucose values within normal


Corresponding author at: Division of Endocrinology and Metabolism, Department of Internal Medicine, Chonbuk National University
Medical School, Gungiro 20 (634-18, Keum-Am Dong), Deok Jin Gu, Jeonju 561-712, South Korea. Fax: +82 63 254 1609.
E-mail address: mdjinhy@jbnu.ac.kr (H.Y. Jin).
http://dx.doi.org/10.1016/j.pcd.2017.06.003
1751-9918/© 2017 Primary Care Diabetes Europe. Published by Elsevier Ltd. All rights reserved.
556 p r i m a r y c a r e d i a b e t e s 1 1 ( 2 0 1 7 ) 555–560

or pre-diabetes range in self-monitoring blood glucose (SMBG) 05-030-001). 270 patients were excluded due to malignancy,
records in routine daily life. Such patients do not need anti- anemia, chronic kidney disease or other disease. Therefore,
diabetes medication, and life style modification is the initial OGTT results of total 1755 patients were analyzed in this study.
management strategy of choice. However, the possibility that We performed a retrospective analysis of 75 g OGTT data in
these patients might be over-diagnosed with diabetes due to 1755 patients. We evaluated history of disease, family his-
the relatively high glucose dose of 75 g in the OGTT should tory, and medications using medical records. We also assessed
not be overlooked. This situation might be common in clini- body weight, height, BMI, and all laboratory findings, including
cal practice. The 75 g OGTT is well known to be helpful for the HbA1c, 75 g OGTT glucose level, fasting insulin, and c-peptide
diagnosis of diabetes, especially when patients show an IFG level.
state; however, it is not clear why a 75 g dose of glucose is used Diagnoses of diabetes, pre-diabetes, and normal glu-
and whether this amount is adequate for every IFG patient. In cose state were defined based on the ADA guidelines
addition, inter-or intra-individual variability and low repro- follows [4] normal, FPG <5.6 mmol/L; impaired fasting
ducibility can be pitfalls in the 75 g OGTT. The amount of glucose, 5.6 mmol/L ≤ FPG <7.0 mmol/L; and diabetes, FPG
glucose loading might need to be individualized according to ≥7.0 mmol/L. 2 h post loading glucose (PLG) after 75 g OGTT
body weight, height, daily caloric intake, and race, if neces- <7.8 mmol/L indicated a normal state, 7.8 mmol/L ≤ 2 h PLG
sary, although 75 g is well accepted as a standard glucose dose <11.1 mmol/L indicated impaired glucose tolerance, and 2 h
in the OGTT. PLG ≥11.1 mmol/L indicated the presence of diabetes. Among
Therefore, we need to reconsider whether 75 g glucose 1755 patients, 595 patients (33.9%) showed normal FPG (below
load is appropriate irrespective of patient clinical charac- 5.6 mmol/L), however, they had high risk factors of diabetes
teristics. The relationship between glucose value and 75 g such as 40 years old, family history of diabetes, overweight,
OGTT and its clinical significance should also be considered hypertension, dyslipidemia, or cardiovascular disease [4] or
in the clinical care of high-risk patients with diabetes with showed 7.8–11.1 mmol/L range of random glucose value. FPG
regard to medication initiation and avoidance of unnecessary of 706 patients (40.2%) were between 5.6 and 7.0 mmol/L,
medication. Micro-vascular complications such as diabetic hence OGTT was performed to diagnosis prediabetes or dia-
autonomic neuropathies can also impact the glucose level, betes exactly. 454 patients (25.9%) already showed FPG over
resulting in a lack of response despite the loaded glucose 7.0 mmol/L, however, OGTT also was performed to assess the
amount [3]. However, the glucose response from a 75 g glu- degree of glucose tolerance. The 75 g OGTT data of 1755 sub-
cose load is expected to differ according to variable patient jects were analyzed to compare the 2 h PLG value after 75 g
factors such as beta cell function, enteric hormones, and neu- glucose loading with HbA1c and FPG in the coincidence of dia-
ral responses to nutrient ingestion. For example, the results betes diagnosis to reaffirm the accuracy of the 2 h PLG value
might differ between a 50 kg patient and a 100 kg patient, in the diagnosis of diabetes. We also reviewed the role of the
even if they have similar pancreatic beta cell function or 2 h glucose value in the decision to initiate glucose-lowering
insulin resistance. Therefore, low-body weight patients might agents in diabetes patients who were diagnosed by 75 g OGTT.
be diagnosed with diabetes, even though routine daily glu- Furthermore, we evaluated the clinical characteristics of the
cose values do not exceed the diabetes threshold. On the pitfall group in the use of 2 h glucose value after 75 g OGTT for
contrary, high-body weight patients might not show glucose the diagnosis of diabetes.
values indicating diabetes, even though routine daily glucose All statistical analyses were performed using the Statis-
values are in the diabetic range, because 75 g is too small of a tical Package for the Social Sciences 18 (SPSS Inc., Chicago,
dose to increase the glucose level to 11.1 mmol/L compared to IL). Data are expressed as mean ± standard deviation (SD)
the routine daily caloric intake needed to maintain their body or median (interquartile range [IQR]), and categorical vari-
mass index (BMI). ables are presented as percentage. For comparisons between
In this study, we investigated the rate of diagnosis of dia- groups, Student’s t test was used for normally distributed vari-
betes based on the 75 g OGTT compared to that based on the ables and the Mann–Whitney U test for variables with skewed
FPG and HbA1c. Furthermore, we analyzed the usefulness of distribution. The Spearman correlation method was used for
75 g OGTT results in the decision to prescribe anti-diabetes evaluation of significant correlation between variables. We
medication in clinical practice. We also assessed the charac- assessed the sensitivity and specificity of each glucose value
teristics of the pitfall group in the use 75 g OGTT, which could determined in the 75 g OGTT for diabetes diagnosis and ana-
be under-diagnosed or over-diagnosed with diabetes using a lyzed receiver operating characteristic curves for the diagnosis
75 g glucose load compared to FBS, HbA1c, and SMBG results of diabetes according to 75 g OGTT fasting, 1 h, and 2 h glucose
based on routine daily caloric intake. levels, and HbA1c. P values <0.05 were considered statistically
significant for all calculations.

2. Methods
3. Results
Total 2810 patients had 75 g OGTT in Chonbuk National Univer-
sity Hospital, Korea from 1995 to 2013. And there was already 3.1. Subject characteristics
an informed consent of permission for using patient’s OGTT
data in the future for study at that time. Based on these Baseline subject characteristics are presented in Table 1. The
consents, we received the approval of Institutional Review mean age of 1755 subjects was 52.2 ± 12.9 years. Among 1755
Board of Choubuk national university hospital (Reg. No. 2016- patients, the number of subjects less than 30 years old was
p r i m a r y c a r e d i a b e t e s 1 1 ( 2 0 1 7 ) 555–560 557

Table 1 – Clinical characteristics of patients by glucose tolerance test results.


Normal Overt diabetes Prediabetes (N = 1017) P value
(N = 329) (N = 409)

IFG or IGT IFG only IGT only


(N = 612) (N = 205) (N = 200)
Age (year) 46.7 ± 16.2 53.0 ± 12.3 52.8 ± 12.9 52.2 ± 10.9 50.0 ± 14.3 <0.01
Sex (female/male) 183/146 201/208 262/350 102/103 110/90
HbA1c (%) 5.5 ± 0.4 7.4 ± 1.6 6.0 ± 0.5 5.9 ± 0.5 5.8 ± 0.6 <0.0.1
OGTT fasting (mmol/L) 5.1± 0.4 7.9 ± 2.3 6.2 ± 0.5 6.0 ± 0.4 5.1 ± 0.4 <0.01
OGTT 1 h (mmol/L) 8.7 ± 2.4 16.1 ± 3.6 11.8 ± 2.2 10.7 ± 2.8 10.7 ± 1.9 <0.01
OGTT 2 h (mmol/L) 6.3 ± 1.0 16.2 ± 4.3 8.9 ± 1.5 6.5 ± 1.6 9.0 ± 1.4 <0.01
Insulin (pmol/L) 0.12 0.07 0.07 0.05 0.08 0.18
[0.08–0.16] [0.06–0.07] [0.06–0.07] [0.04–0.06] [0.06–0.10]
C-peptide (nmol/L) 1.4 0.9 1.0 0.8 0.8 <0.01
[1.0–1.8] [0.9–1.0] [0.8–1.2] [0.7–0.8] [0.7–0.9]
Height (cm) 162.5 ± 8.1 162.2 ± 8.7 163.4 ± 8.7 164.7 ± 7.1 161.6 ± 9.3 0.04
Body weight (kg) 64.1 ± 11.3 66.1 ± 11.8 67.3 ± 8.7 66.7 ± 11.4 69.6 ± 19.7 0.08
BMI (kg/m2 ) 24.3 ± 4.2 25.1 ± 3.8 25.2 ± 2.9 24.6 ± 3.5 26.7 ± 7.6 0.34
Normal group: fasting plasma glucose <5.6 mmol/L and 2 h PLG <7.8 mmol/L
Overt diabetes group: fasting plasma glucose ≥7.0 mmol/L and 2 h PLG ≥11.1 mmol/L
BMI: body mass index, IFG: impaired fasting glucose, IGT: impaired glucose tolerance, OGTT: oral glucose tolerance test, 2 h PLG: 2 h post loading
glucose

100 (5.7%), subjects from 30 to 39 years old were 177 (10.1%).


Table 2 – The relation of 2 h glucose value after 75 g
The other distribution of subjects’s age was as follow; 40–49 glucose loading with fasting plasma glucose.
years old: 366 (20.9%), 50–59 years old: 550 (31.3%), 60–69
FPG (number) 2 h PLG (number) Proportion (%)
years old: 392 (22.3%), and over 70 years old: 170 (9.7%). The
subjects consisted of 858 (48.9%) females and 897 (51.1%) Normal Normal (329) 55.3
(595) IGT (200) 33.6
males. All subjects were Asian. According to the analyzed 75 g
Diabetes (66 11.1
OGTT results, 309 subjects showed a normal glucose toler-
[16,HbA1c ≥ 6.5%])
ance state, 409 patients were in an overt diabetic state, and IFG (706) Normal (205) 29.0
1017 patients were in a pre-diabetic state. There were sig- IGT (217) 30.7
nificant differences in HbA1c, fasting glucose, 1 h glucose, Diabetes (284 40.2
and 2 h glucose values after 75 g glucose loading in each [122,HbA1c ≥ 6.5%])
group (P < 0.01). Subjects that showed normal glucose toler- Diabetes Normal (7) 1.5
(454) IGT (38) 8.4
ance were younger than those with pre-diabetes or diabetes.
Diabetes (409 90.1
There were no significant differences in body weight or BMI [347,HbA1c ≥ 6.5%])
in any group. The male/female ratio was 858/897 (48.9/51.1%):
183/146 (55.6/44.3%) in normal group, 201/208 (49.1/50.9%) in
overt diabetes group, and 474/543 (46.6/53.4%) in prediabetes 2 h PLG (number) FPG (number) Proportion (%)
group. Overt diabetes group did not show sex differences in Normal Normal (329) 60.8
2 h PLG. However, prediabetes group included more female, (541) IFG (205) 37.9
and more males were included in the normal group. Further Diabetes (7) 1.3
studies need to have consideration of these sex differences. IGT (455) Normal (200) 36.3
IFG (217) 47.7
Diabetes (38) 8.4
3.2. Coincidence of 2 h PLG value with FPG and HbA1c Diabetes Normal (66 8.7
for use in the diagnosis of diabetes (759) [18,HbA1c < 5.7%])
IFG (284) 37.4
We assessed the degree of accordance of each 75 g OGTT glu- Diabetes (409) 53.9

cose value according to glucose tolerant state (Table 2). Among FPG: fasting plasma glucose, IFG: impaired fasting glucose, IGT:
595 patients with normal FPG, only 329 (55.3%) were in a nor- impaired glucose tolerance, PLG: post loading glucose.
mal glucose tolerance state according to 2 h PLG (<7.8 mmol/L)
and HbA1c ( < 5.7%). Two hundred patients (33.6%) with a
normal FPG level showed an IGT state, 66 patients (11.1%) PLG of 75 g OGTT based on daily caloric intake. Among 454
showed 2 h PLG greater than 11.1 mmol/L, and 16 patients patients with diabetes based on FPG, 409 (90.1%) also showed
showed HbA1c higher than 6.5%, compatible with a diagno- 2 h PLG higher than 11.1 mmol/L, and 347 (76.4%) showed
sis of diabetes according to the ADA criteria. Among these HbA1c higher than 6.5%. However, 45 patients (9.9%) showed
66 patients with diabetes, only 28 (42.0%) had been treated 2 h PLG below 11.1 mmol/L. The accordance rate of 2 h PLG
with glucose-lowering agents; the others underwent glucose level with FPG was 55.1% in the normal group and 90.1% in
monitoring due to a random glucose value lower than 2 h the diabetes group to confirm a normal state and diabetic
558 p r i m a r y c a r e d i a b e t e s 1 1 ( 2 0 1 7 ) 555–560

Table 3 – Clinical characteristics of patients with diabetes medication (n = 592) by glucose tolerance test results (2 h PLG).
2h 2h 7.8 mmol/L ≤ 2 h P value
PLG < 7.8 mmol/L PLG ≥ 11.1 mmol/L PLG < 11.1 mmol/L
(n = 23) (n = 450) (n = 119)
Age (year) 43.5 ± 17.5 51.5 ± 12.4 51.2 ± 14.4 <0.01
Sex (female/male) 15/8 201/249 58/61
HbA1c (%) 5.3 ± 0.5 7.8 ± 1.7 6.3 ± 0.7 <0.01
OGTT fasting (mmol/L) 5.2 ± 0.3 8.4 ± 2.5 6.0 ± 0.8 <0.01
OGTT 1 h (mmol/L) 8.8 ± 2.1 16.9 ± 3.8 11.8 ± 2.5 <0.01
OGTT 2 h (mmol/L) 6.8 ± 0.4 17.0 ± 4.6 8.8 ± 1.6 <0.01
Insulin (pmol/L) 0.05 [0.04–0.07] 0.07 [0.07–0.08] 0.07 [0.06–0.08] 0.70
C-peptide (nmol/L) 1.7 [0.8–2.6] 0.9 [0.8–0.9] 0.8 [0.7–0.9] 0.47
Height (cm) 163.3 ± 4.6 162.6 ± 8.5 162.0 ± 9.8 0.83
Body weight (kg) 68.2 ± 7.3 67.3 ± 12.0 66.7 ± 10.8 0.53
BMI (kg/m2 ) 25.7 ± 3.4 25.5 ± 3.9 25.4 ± 3.4 0.68
BMI: body mass index, OGTT: oral glucose tolerance test

state, respectively. We assessed the correlation between 2 h


Table 4 – The relationship with OGTT result and
PLG value and each clinical and laboratory metabolic variable. anti-diabetic medication status.
HbA1c (r = 0.7, P < 0.01), FPG (r = 0.7, P < 0.01), and 75 g OGTT
Group I Group II P value
1 h values (r = 0.8, P < 0.01) showed a significant positive cor-
(n = 66) (n = 103)
relation with 2 h PLG value, as did age (r = 0.1, P < 0.01) and
Age (year) 54.5 ± 12.6 49.8 ± 15.2 <0.01
C-peptide level (r = 0.2, P < 0.01).
Sex (female/male) 31/35 55/48
HbA1c (%) 5.9 ± 0.4 5.8 ± 0.5 0.08
3.3. The practical use of 75 g OGTT results for the OGTT fasting (mmol/L) 6.2 ± 0.9 5.8 ± 0.7 <0.01
initiation of glucose-lowering agents OGTT 1 h (mmol/L) 13.2 ± 2.0 11.2 ± 2.8 <0.01
OGTT 2 h (mmol/L) 13.2 ± 2.3 8.4 ± 1.8 <0.01
In the present study, 592 subjects were treated with anti- Insulin (pmol/L) 0.07 [0.06–0.09] 0.07 [0.06–0.07] 0.09
diabetes medication and were categorized into either the C-peptide (nmol/L) 1.2 [0.9–1.5] 1.0 [0.8–1.2] 0.85
Height (cm) 162.1 ± 9.8 162.3 ± 9.2 0.65
normal group (n = 23), diabetes group (n = 450),or pre-diabetes
Body weight (kg) 63.9 ± 12.1 66.6 ± 11.5 0.22
group (n = 119) by 2 h PLG level (Table 3).There were signifi- BMI (kg/m2 ) 24.2 ± 3.3 25.2 ± 3.4 0.09
cant differences between HbA1c, FPG, PLG 1 h, PLG 2 h, and
BMI: body mass index, DM: diabetes mellitus, OGTT: oral glucose
age among the groups (P < 0.01). The normal glucose toler-
tolerance test
ance group was taller and heavier than the diabetes and Group I: Participants with DM diagnosed by OGTT result, but did not
pre-diabetes groups, but these differences were not signif- treated any anti-diabetic agent
icant. We also found significant differences between group Group II: Participants without DM diagnosed by OGTT result, but
I: subjects with diabetes (diagnosed by 75 g OGTT, but not who treated with anti-diabetic agent.
treated with any anti-diabetes agent) and group II: subjects
without diabetes (confirmed by 75 g OGTT result but treated
the groups. Group A showed significantly lower 2 h PLG value
with glucose-lowering agents)(Table 4). Group I was signifi-
and higher HbA1c compared to group B (P < 0.01). Furthermore,
cantly younger than group II (P < 0.01) and had lower BMI and
we performed receiver operating characteristic curves for the
body weight.
diagnosis of diabetes according to 75 g OGTT fasting, 1 h and
2 h PLG levels and HbA1c. In this analysis, 75 g OGTT 2 h PLG
3.4. Pitfall group in the use of 75 g OGTT for diabetes
(91.0–93.6%) showed higher sensitivity than the other variables
diagnosis
[OGTT 1 h PLG (86.6–89.8%), HbA1c (86.4–89.6%), and 75 g OGTT
fasting glucose level (83.4–87.0%)] for the diagnosis of diabetes
We assessed the clinical characteristics of a pitfall group
(Fig. 1).
in the use of 75 g OGTT for diabetes diagnosis (Table 5).
The pitfall group who showed inconsistent glucose status
between FBS and 75 g OGTT consisted of patients with a 4. Discussion
lower 2 h PLG value compared to fasting (n = 250) and upper
2 h PLG value compared to fasting (n = 359). Group A (lower The prevalence of diabetes continues to increase worldwide;
2 h PLG value compared to FPG) consisted of patients with therefore, determining high risk groups and early detection
(1) FPG ≥ 7.0 mmol/L and 7.8 mmol/L <2 h PLG < 11.1 mmol/L are of utmost importance. Currently, the 75 g OGTT is a very
or (2) 5.6 mmol/L ≤ FPG < 7.0 mmol/L and 2 h PLG < 7.8 mmol/L. useful tool for assessing glucose tolerance status, in addi-
Group B (upper 2 h PLG value compared to FPG) consisted tion to the diagnosis of diabetes [1]. The National Diabetes
of patients with (1) FPG < 5.6 mmol/L and 7.8 mmol/L ≤2 h Data Group first defined impaired glucose tolerance, a state
PLG < 11.1 mmol/L or (2) 5.6 mmol/L ≤ FPG <7.0 mmol/L and 2 h between normal and diabetes, in 1979 [5]. Identification of
PLG > 11.1 mmol/L. Group A was significantly taller and heav- this state requires OGTT results and is endorsed by the World
ier than group B (P <0.01); however, BMI did not differ between Health Organization (WHO) and the ADA [1]. The cut-off val-
p r i m a r y c a r e d i a b e t e s 1 1 ( 2 0 1 7 ) 555–560 559

is a major drawback [9–11], and a second OGTT is needed


Table 5 – Clinical characteristics in the pitfall group to
use 2 h PLG value for diabetes diagnosis. to confirm the diagnosis of diabetes. In addition, the clinical
utility of the 75 g OGTT in the management of patients with
Group A: Group B: P value
pre-diabetes has also been an issue [12]. Nonetheless, patients
lower PLG upper PLG
result result with IGT according to the 75 g OGTT show high cardiovascular
compared to compared to risk compared to those with normal glucose tolerance [8,13].
fasting fasting A considerable number of patients with normal fasting glu-
(n = 250) (n = 359) cose level show IGT and might be diagnosed with diabetes
Age (year) 52.0 ± 11.2 52.5 ± 14.5 0.05 using 75 g OGTT. Therefore, 75 g OGTT is still an important
Sex (female/male) 104/146 219/140 diagnostic tool for diagnosing diabetes [2]; however, we should
HbA1c (%) 6.2 ± 0.8 5.8 ± 0.7 <0.01 consider whether patients might be over-diagnosed with dia-
OGTT fasting (mmol/L) 6.5 ± 0.8 5.1 ± 0.4 <0.01 betes by 75 g OGTT or whether 75 g OGTT is limited in its ability
OGTT 1 h (mmol/L) 11.5 ± 2.9 11.1 ± 2.5 0.19
to detect hidden diabetes due to insufficient loading dose of
OGTT 2 h (mmol/L) 7.1 ± 1.9 10.2 ± 2.6 <0.01
glucose.
Insulin (pmol/L) 0.06 [0.05–0.06] 0.09 [0.07–0.12] 0.33
C-peptide (nmol/L) 0.7 [0.7–0.8] 1.0 [0.8–1.1] 0.81 The present study compared the clinical usefulness of 2 h
Height (cm) 164.9 ± 7.9 161.5 ± 8.8 <0.01 PLG value with FPG and HbA1c with respect to the diagnosis of
Body weight (kg) 68.0 ± 10.3 66.9 ± 17.0 <0.01 diabetes and the decision to initiate anti-diabetic medication.
BMI (kg/m2 ) 25.0 ± 3.2 25.7 ± 6.7 0.72 In addition, we investigated the characteristics of the pitfall
BMI: body mass index, OGTT: oral glucose tolerance test group in the use of 75 g OGTT in clinical practice. According
Group A: Lower PLG result compared to fasting: (1) fast- to our observations, the 2 h PLG value from 75 g OGTT showed
ing ≥ 7.0 mmol/L and 7.8 mmol/L < PLG < 11.1 mmol/L, (2) high sensitivity for the diagnosis of diabetes compared to the
5.6 mmol/L ≤ fasting < 7.0 mmol/L and PLG < 7.8 mmol/L
other glycemic indexes HbA1c and FPG. This result is simi-
Group B: Upper PLG result compared to fasting: (1) fast-
lar to those found in previous studies [8,14]. Glucose value
ing < 5.6 mmol/L and 7.8 mmol/L ≤ PLG < 11.1 mmol/L, (2)
5.6 mmol/L ≤ fasting < 7.0 mmol/L and PLG > 11.1 mmol/L. after 75 g glucose loading can be influenced by many vari-
ables, including insulin sensitivity with pancreatic beta cell
function, enteric hormones, and neural responses to nutri-
ent ingestion, such as gastrointestinal motility and gastric
emptying [15,16]. All of these factors vary widely and influ-
ence post-loading glucose metabolism. Previous studies have
indicated a problem with the reproducibility of the 75 g OGTT
according to intra-individual variation and glucose load[9,17].
However, it is unclear whether the 75 g glucose dose can be
used to diagnose diabetes irrespective of race, weight, height,
and individual metabolic characteristics. In addition, we do
not have any information on the rate of use of 2 h PLG value
of the 75 g OGTT by clinical physicians in the decision to initi-
ate glucose-lowering agents. Our study confirmed that the 2 h
PLG value from 75 g OGTT is consistent with the results of FPG
and HbA1c in the diagnosis of diabetes in about 90% of dia-
betic subjects. However, we should not overlook the fact that
only 55.3% of subjects in a normal population showed a nor-
mal value of 2 h PLG. Therefore, adjustment of the 75 g glucose
Fig. 1 – Selectivity of diabetes index for diagnosis. dose might be needed in order to increase the specificity of the
diagnosis of diabetes based on the 75 g OGTT.
In our study, 8.7% of patients who were diagnosed with
ues and glucose loading dose have been debated for more than diabetes by the 75 g OGTT showed normal SMBG value, and
a century [6]. In 1997, an expert committee established the 37.4% of patients who were compatible with the diagnosis of
fasting and 2 h PLG criteria for diabetes [1]. The FPG level was diabetes according to 2 h PLG did not show an SMBG value
lowered and the ADA expected that the lowered cut point for in the diabetic range. These patients had not been treated
FPG would include most people with IGT. However, IFG is not with any anti-diabetic medication and showed significantly
equivalent to IGT, and several studies have reported that 2 h lower body weight compared to patients who had been treated
PLG more effectively predicts the risk of heart disease risk than with glucose-lowering medications, although their 2 h PLG
FPG [7,8]. In addition, the 75 g OGTT is necessary to identify value was not in the diabetic range at the time of diagnosis.
undiagnosed diabetic patients through the screening of sub- Therefore, we also assessed the characteristics of the pitfall
jects at high risk of diabetes and is useful for the prevention group in the use of 75 g OGTT, as it is possible for diabetes
of diabetes progression through the application of behavioral to be under-diagnosed or over-diagnosed using only 2 h PLG
or pharmacological interventions. However, there are several compared to FBS, HbA1c, and SMBG results based on routine
limitations to the 75 g OGTT with respect to time consumption, daily caloric intake. Patients with low body weight and height
cost, inconvenience, the 75 g dose of glucose requirement, and showed high 2 h PLG in consideration of FBG, HbA1c, and SMBG
reproducibility [9,10]. In particular, the reproducibility of OGTT data. Therefore, 2 h PLG by 75 g OGTT can be a useful tool in the
560 p r i m a r y c a r e d i a b e t e s 1 1 ( 2 0 1 7 ) 555–560

diagnosis of diabetes and initiation of anti-diabetic manage- [2] L. Perreault, S.E. Kahn, C.A. Christophi, et al., Diabetes
ment. However, clinicians should consider clinical variables Prevention Program Research, G: regression from
such as individual life style and physical factors such as body pre-diabetes to normal glucose regulation in the diabetes
prevention program, Diabetes Care 32 (9) (2009) 1583–1588.
weight and height, in addition to metabolic parameters related
[3] WHO, World Health Organization technical report series,
to insulin action. According to our study, the interpretation 19801–80.
of 75 g OGTT results should consider that, in around 10% of [4] American Diabetes, A standards of medical care in
patients, a diagnosis of diabetes was not able to be made based diabetes—2014, Diabetes Care 37 (1) (2014) S14–S80.
solely on 2 h PLG value. It must also be considered whether [5] National Diabetes Data Group, Classification and diagnosis
the 75 g glucose loading dose is appropriate in these patients, of diabetes mellitus and other categories of glucose
especially for the clinical decision to initiate anti-diabetes intolerance, Diabetes 28 (12) (1979) 1039–1057.
[6] W.H. Herman, Diabetes epidemiology guiding clinical and
medication. Moreover, the loading dose of glucose in the tol-
public health practice: the Kelly West Award Lecture, 2006,
erance test needs to be subdivided or categorized according to Diabetes Care 30 (7) (2007) 1912–1919.
body weight, individual life style, or necessary caloric intake [7] DECODE Study Group on behalf of the European Diabetes
in order to increase its usefulness in clinical diabetes manage- Epidemiology Study Group, Will new diagnostic criteria for
ment, although our results did not show a detailed numeric diabetes mellitus change phenotype of patients with
cut off value for these diverse variables. In short, the 75 g glu- diabetes? Reanalysis of European epidemiological data, BMJ
317 (7155) (1998) 371–375.
cose dose might not be suitable for a diagnosis of diabetes in
[8] R.C. Perry, A.D. Baron, Impaired glucose tolerance. Why is it
all sizes and body types.
not a disease? Diabetes Care 22 (6) (1999) 883–885.
There are several limitations to the present study. First, the [9] M. Toeller, R. Knussmann, Reproducibility of oral glucose
study took place at a single center and included only one Asian tolerance tests with three different loads, Diabetologia 9 (2)
population, limiting our ability to generalize the study results. (1973) 102–107.
Second, prospective long-term follow-up including diabetic [10] I.M. Libman, E. Barinas-Mitchell, A. Bartucci, et al.,
complications and mortality of the 75 g OGTT data should be Reproducibility of the oral glucose tolerance test in
overweight children, J. Clin. Endocrinol. Metab. 93 (11) (2008)
investigated in addition to the present comparison to FBG,
4231–4237.
SBMG, and HbA1c. Third, more detailed categorization of the [11] G. Brohall, C.J. Behre, J. Hulthe, et al., Prevalence of diabetes
glucose loading dose is necessary to determine the necessity and impaired glucose tolerance in 64-year-old Swedish
of reconsidering the 75 g glucose loading dose. Additionally, women: experiences of using repeated oral glucose
the study data did not examine any insulin secretion or sensi- tolerance tests, Diabetes Care 29 (2) (2006) 363–367.
tivity factors. Further studies addressing these limitations are [12] M.P. Stern, K. Williams, S.M. Haffner, Identification of
needed. persons at high risk for type 2 diabetes mellitus: do we need
the oral glucose tolerance test? Ann. Intern. Med. 136 (8)
In conclusion, we re-affirmed the usefulness of 2 h PLG for
(2002) 575–581.
the diagnosis of diabetes with FPG and HbA1c. However, the [13] Q. Qiao, K. Pyorala, M. Pyorala, et al., Two-hour glucose is a
pitfall group must be considered in the use of the 75 g OGTT for better risk predictor for incident coronary heart disease and
diabetes confirmation according to height or body weight, in cardiovascular mortality than fasting glucose, Eur. Heart J.
addition to diverse individual variables. Therefore, physicians 23 (16) (2002) 1267–1275.
should use caution when interpreting the 75 g OGTT result, [14] M.I. Harris, R.C. Eastman, C.C. Cowie, et al., Comparison of
diabetes diagnostic categories in the U.S. population
and the 75 g glucose loading dose should be reconsidered to
according to the 1997 American Diabetes Association and
determine whether it is adequate in the pitfall group. Further
1980–1985 World Health Organization diagnostic criteria,
studies are needed to examine these issues according to body Diabetes Care 20 (12) (1997) 1859–1862.
mass index and ethnic differences in large populations. [15] M. Stumvoll, A. Fritsche, H. Haring, The OGTT as test for
beta cell function? Eur. J. Clin. Invest. 31 (5) (2001) 380–381.
[16] C.C. Jensen, M. Cnop, R.L. Hull, American Diabetes
Conflict of interest Association, G.S.G, et al., Beta-cell function is a major
contributor to oral glucose tolerance in high-risk relatives of
None four ethnic groups in the U.S, Diabetes 51 (7) (2002)
2170–2178.
[17] J.M. Mooy, P.A. Grootenhuis, H. de Vries, et al.,
Acknowledgements Intra-individual variation of glucose, specific insulin and
proinsulin concentrations measured by two oral glucose
We would like to thank the Research Institute of Clinical tolerance tests in a general Caucasian population: the Hoorn
Medicine of Chonbuk National University and the Biomedical Study, Diabetologia 39 (3) (1996) 298–305.
Research Institute of Chonbuk National University Hospital for
partly supporting our research with a grant.

references

[1] Report of the Expert Committee on the diagnosis and


classification of diabetes mellitus, Diabetes Care 20 (7) (1997)
1183–1197.

Anda mungkin juga menyukai