Diabetes Research and Clinical Practice - Volume 120 - Supplement 1

also more in those with CKD (17.6% vs 3.6%, p = 0.001). We

found mean ABI was positively correlated with eGFR (

= 0.184,

P = 0.007), and negatively correlated with age (

−

0.326,

P < 0.001), duration of DM (

−

0.386, P < 0.001), and systolic

BP (

−

0.200, P = 0.003). Multivariate regression analysis

revealed that age and duration of DM were correlated with

mean ABI.

Conclusions:

Type 2 diabetic patients with CKD are associated

with higher prevalence of PAD than those without. Moreover,

older age and longer duration of DM account for the highest

risks. Routine performance of ABI for early detection of PAD is

indicated in these patients.

PA-13

The relationship between hyperglycemic clamp and low dose

graded glucose infusion in quantifying second phase insulin

secretion

Dee PEI

*, Te-Lin HSIA

, Jiunn-Diann LIN

, Chung-Ze WU

Yen-Lin CHEN

Department of Internal Medicine, Cardinal Tien

Hospital, School of Medicine, Fu-Jen Catholic University,

Division of

Endocrinology and Metabolism, Department of Medicine, Shuang-Ho

Hospital, Taipei Medical University,

Department of Pathology,

Cardinal Tien Hospital, School of Medicine, Fu-Jen Catholic

University, New Taipei, Taiwan

Aim:

First phase insulin secretion usually disappears even in

pre-diabetic stage. Thus, maintaining glucose control after

occurrence of diabetes must rely on the existence of second

phase insulin secretion (SPIS). There are several methods to

quantify SPIS, such as oral glucose tolerance, low dose graded

glucose infusion (LDGGI) and hyperglycemic clamp (HC). In the

same time, whether to use c-peptide and deconvolution to

estimate

“

true

”

insulin secretion was also suggested to have

some influence on the measurement. If we take the HC as the

gold standard, it is interesting to note that LDGGI has never

been validated against HC. In the same time, whether to use

deconvolution is also interesting.

Methods:

Fourteen subjects (3 with normal glucose tolerance,

8 with pre-diabetes and 3 with diabetes) were enrolled. They

received both 120 min HC and 200 min LDGGI proposed by

Polonsky et al. Four different SPIS were measured. Two from

the HC, i.e., plasma insulin level (PIclamp) and insulin

secretion rate (ISRclamp, by using deconvolution) between 80

and 120 min. Similarly, the other two were from the LDDGI,

plasma insulin level (PILDDGI) and ISR (ISRLDDGI) by decon-

volution during the whole test. Simple correlation was used to

evaluate their concordance.

Results:

Both PILDDGI and ISRLDDGI were highly correlated

with PIclamp (r = 0.851, 0.748, p = 0.000). It should be noted that

the r value is higher for the PILDDGI than that of ISRLDDGI. In

the same time, the PIclamp and ISRclamp is also highly

correlated (r = 0.886, p = 0.000).

Conclusion:

Our results showed that if we take clamp as the

gold standard, PILDDGI could be used as another accurate

method to measure SPIS. In the same time, to use deconvo-

lution is not mandatory in order to have precise estimation

PA-14

Clinical pictures associated with borderline heteroplasmy rate

of 3243 mitochondrial tRNA Leu (UUR) mutation in type 2

diabetes

Yoshihiko SUZUKI

*, Junichiro IRIE

, Shigeo OHTA

Motoaki SANO

, Toshihide KAWAI

, Shu MEGURO

Nobuhiro IKEMURA

HDC Atlas Clinic,

Department of Internal

Medicine, Keio University School of Medicine,

Department of

Biochemistry and Cell Biology, Institute of Development and Aging

Sciences, Graduate School of Medicine, Nippon Medical School,

Department of Cardiology, Keio University School of Medicine,

Department of Internal Medicine, Saiseikai Central Hospital,

Tokyo, Japan

Approximately 0.5

–

2.8% of Japanese diabetic patients popula-

tion is estimated to have mitochondrial DNA (mtDNA)

mutation at position 3243 in the blood. However, the cut-off

limitation for detecting the mutation in blood differs depend-

ing upon the judgements of researchers. In general, using the

ordinary polymerase chain reaction (PCR) method, the thresh-

old is around 1% of the heteroplasmic rate. A new technology

using real-time PCR with a TaqMan Probe was introduced,

which can quantify as little as 0.001% of the 3243 mtDNA

mutation in blood cells. By using this method, we examined

the association of low borderline heteroplasmic rate with

clinical pictures in cross sectional study of type 2 diabetes.

189 patients with type 2 diabetic were subjected. The profile of

the 189 subjects was 142males and 47 females. Their ages were

59.4 ± 10.2 years and diabetes durationwas 12.1 ± 7.9 years. BMI

22.9 ± 4.0 and HbA1c 8.2 ± 1.3% (mean ± SD).

In result, the heteroplasmy rates were about six times higher

than the healthy controls (aged 20

–

60 years, n = 186). In all the

type 2 diabetes subjects, the mathematical mean and SD of

heteroplasmy rate was 0.041 ± 0.018%. When logarithmically

converted, the data were seemingly distributed normally. By

the data, the mean was

–

1.408, and SD was 0.142.

“

Mean + 2SD

”

was

−

1.124, which corresponds to 0.075% of the heteroplasmy

rate. Therefore, we regarded the patients with over 0.075%

heteroplasmic rate as having a high level among type 2

diabetes patients.

As the result, five patients were positioned over 0.075%.

Interestingly, all the five had clinical pictures associated with

mitochondrial dysfunction, such as juvenile cataract (Subject-

1: 0.090% of heteroplasmy rate), ophthalmoplegia and lipoma

(Subject-2: 0.102%) and high lactate levels in the blood

(Subject-3: 0.172% and Subject-4:0.115%). Subject 5 (hetero-

plasmy rate: 0.127%) had a maternal inheritance of diabetes

and cerebellar atrophy. The details of Subjects 1 and 2 have

been described in Diabetologia 2004 and Diab.Res.Clin, Prac.

2004 63(3): 225

–

Thus, when we define the upper-borderline level of hetero-

plasmy rate to be 0.075% among type 2 diabetes patients, five

patients were found to have a high heteroplasmy rate.

Interestingly, the five patients had one or two mitochondria-

associated characteristics. They did not have hearing loss.

Therefore, we propose that clinicians should not overlook the

important features of mitochondrial diabetes, even when the

result of examination for 3243 mtDNA mutation is negative by

ordinary PCR method and even when hearing loss is absent.

PA-15

The correlation between lifestyle factors and subclinical

neuropathy in patients with type 2 diabetes

Chieh-Hsiang LU

–

*, Hsiang-Hsun CHUANG

Ditmanson

Medical Foundation Chia-Yi Christian Hospital,

Department of

Business Administration, College of Management, National Chung

Cheng University,

Department of Nursing, DAYEH University,

Taiwan

Background:

Little is known regarding correlation between the

lifestyle factors and asymptomatic neuropathy in patients

with type 2 diabetic (T2D) in Taiwan. The aim of the study was

to evaluate which lifestyle factors were associated with

increased risk of subclinical diabetic neuropathy.

Method:

We conducted a prospective cross-sectional study

at Chia-Yi Christian Hospital, Taiwan. 153 T2D patients

with subclinical neuropathy, which was diagnosed by using

neuro-electrophysiology, 2-point discrimination, and pressure

threshold measurement were enrolled. Questionnaires

including food frequency questionnaire (FFQ) and Douleur

Neuropathique 4 (DN4) questionnaire were obtained.

Results:

Of 153 participants, 38 (24.8%) patients were diag-

nosed to have subclinical neuropathy. Comparing the lifestyle

factors between the two groups, T2D patients with neuropathy

had significant lower amounts of daily vegetable intake

Poster Presentations / Diabetes Research and Clinical Practice 120S1 (2016) S65

–

S211

S69