Diabetes Research and Clinical Pructice, 13 (1991) 181-188
0 1991 Elsevier Science Publishers
181
B.V. 0168-8227/91/$03.50
DIABET 00525
Urinary C-peptide as an index of unstable glycemic control in insulin-dependent diabetes mellitus (IDDM) Tepyon Cha’, Yasuhiro Tahara’, Hiroshi Ikegami’, Masahiro Fukuda’, Hiroko Yoneda’, Eiji Yamato’, Yoshihiro Yamamoto’, Yoshihiko Noma2, Kenji Shima and Toshio Ogihara’ ‘Department
of Geriatric Medicine,
Osaka University Medical School, Fukushimu-ku,
Laboratory Medicine.
Tokushimu University School of Medicine,
Osaka, Japan and ‘Department
of
Tokushimu, Japan
(Received 29 January 1991) (Revision accepted 15 April 1991)
Summary
In order to investigate whether urinary C-peptide (UCP) excretion can be a useful index of insulindependent diabetes mellitus (IDDM) with unstable glycemic control, UCP was measured in nine IDDM patients with unstable glycemic control, nine IDDM patients with stable glycemic control, and 12 non-insulin-dependent diabetic (NIDDM) patients treated with insulin. The UCPs in overnight urine (U 1) and fasting single void urine (U2) in IDDM patients with unstable glycemic control were significantly lower than those in IDDM patients with stable glycemic control (Ul: 0.03 + 0.03 vs 0.24 _+0.20 nmol/mmol-Creatinine, U2: 0.02 k 0.01 vs 0.20 2 0.20 nmol/mmol-Cr, mean + SD, both P < 0.01). The UCPs in Ul and U2 in both groups of IDDM were significantly lower than those in NIDDM (Ul: 0.97 + 0.52, U2: 0.73 k 0.41 nmol/mmol-Cr, both P < 0.01). The UCPs in Ul and U2 significantly correlated with incremental C-peptide response to intravenous glucagon injection and with glycemic stability assessed by the standard deviation of 10 previous fasting plasma glucose levels. These results suggest that UCP reflects their residual insulin secretory capacity and that UCP can be a useful index which distinguishes patients with unstable IDDM from those with stable diabetes mellitus. Key words: Urinary C-peptide
excretion;
Brittle diabetes;
Introduction
Unstable glycemic control is associated with many factors such as insulin resistance, an incorrect therapeutic regimen, various recurrent complications. psychological problems, lack of residCorrespondence
to: Tepyon Cha, Dept. of Geriatric
Osaka University 553, Japan.
Medical
School Fukushima-ku,
Medicine, Osaka,
Residual p-cell function;
Glycemic
stability
ual p-cell function, and abnormalities of counterregulatory hormone secretion [ l-41. Of these factors, residual p-cell function is especially important in glycemic stability because patients without any residual P-cell function cannot restore unexpected hyper- or hypoglycemia through regulating endogenous insulin secretion. Many reports have demonstrated the relationship between residual /?-cell function and glycemic stability in insulin-dependent diabetes mellitus (IDDM)
182
[2,3,5]. Although some papers showed that patients without residual /?-cell function did not always show unstable glycemic control [ 1,6,7], our previous study with a highly sensitive assay for C-peptide [4] revealed the existence of very little residual P-cell function in IDDM with relatively stable glycemic control, and a total lack of insulinogenic reserve in patients with unstable glycemic control. Thus, complete absence of an insulinogenic reserve appears to be the fundamental cause of glycemic instability, yet even if there is a very small amount of residual P-cell function, below the detection limit of the conventional assay, this function is still very important in glycemic stability. Although a highly sensitive assay for C-peptide is very useful in the evaluation of minimal residual /?-cell function in IDDM patients, a simpler method is more suitable for routine clinical use. Urinary C-peptide (UCP) excretion is considered to reflect integrated insulin secretion because C-peptide is secreted equimolarly with insulin, and part of the secreted C-peptide is excreted in the urine [ 8- 131. Although the amount of C-peptide excreted in the urine is affected by various conditions [ 14,151, many authors showed that UCP can provide useful information in the evaluation of insulin secretory capacity, especially in IDDM [ 8,16-211. However, there is little information about whether UCP can also provide information in the evaluation of very low residual P-cell function and glycemic stability in IDDM. In the present study we examined the relationships between UCP and glycemic stability in IDDM, and we investigated whether UCP can be a useful index of residual b-cell function for distinguishing IDDM patients with unstable glycemic control from those with stable glycemic control.
Subjects and Methods The subjects studied were 18 patients with IDDM and 12 patients with NIDDM who were attending the Department of Geriatric Medicine at Osaka University Hospital. The diagnosis of
diabetes was made according to the criteria of the World Health Organization. None of them had chronic renal failure as defined by either endogenous creatinine clearance of less than 0.5 ml/s or a plasma creatinine level greater than 130 PM. Patients with urinary tract infections were excluded. IDDM was determined when patients lacked endogenous insulin secretion, as assessed by the incremental area of C-peptide after intravenous administration of glucagon at less than 1.5 nM min (see below). All patients with IDDM experienced an episode of ketonuria or diabetic ketoacidosis. Glycemic stability was assessed by the standard deviation of the 10 values of fasting plasma glucose (SDFPG), which was measured during their visits once a month PI. IDDM patients were divided into 2 groups according to their glycemic stability. Patients with SDFPG larger than the maximal value of SDFPG for NIDDM were regarded as having unstable glycemic control, while those with SDFPG less than that value were regarded as having stable glycemic control. Clinical characteristics of the patients are shown in Table 1. There were no significant differences in age, known duration of diabetes, body mass index, or glycemic control level between the IDDM patients with unstable and stable glycemic control. The therapeutic mode for each patient was determined by each attending physician in order to achieve a permissible glycemic control level. IDDM patients with unstable glycemic control had been treated with more intensive therapy than those with stable glycemic control. Patients were asked to void urine just before sleeping, and this urine was discarded. Overnight urine voided just after getting up in the morning was collected and designated as U 1. A fasting urine voided just after arriving at the outpatient clinic was also collected and designated as U2. UCP was measured on three days and the mean of the 3 determinations was used in this study. The mean coefficients of day-to-day variation of UCP in Ul and U2 were 27.0 and 17.6% in IDDM patients with unstable glycemic control, 26.1 and 23.1 Y0 in patients with stable glycemic
183 TABLE 1 Clinical characteristics
of the patients
Number (M:F)
studied
Age (yrs)
Duration of DM
HbA,, (%)
BMI
(mM)
(kg/m’ )
Therapy CSII:Mix:II:I
10.21 + 2.61 8.38 + 0.55 9.58 f 1.78
9.3 + 0.9 8.9 f 1.7 8.3 k 1.5
19.6 + 3.0 19.2 + 1.5;’ 22.1 + 3.0
5:4:0:0 2:6:1:1 0: 1:7:4
FPG
(yrs) IDDM unstable stable NIDDM
4:5 4:5 9:3
45 + 14” 39 + 13b 57* 7
10 * 4 12 + 8 14 f 7
Values are mean + SD. Unstable: IDDM patients with unstable glycemic control. Stable: IDDM patients with stable glycemic control. FPG: fasting plasma glucose. HbA,,: hemoglobin A,, (normal range: 4.4-6.4%). BMI: body mass index. CSII: continuous subcutaneous insulin infusion. Mix: two or three daily injections of a mixture ofintermediateand short-acting insulin. II: two daily injections of intermediate-acting insulin. I: single daily injection of intermediate-acting insulin. “P < 0.05; ‘P < 0.01 vs type 2.
control, and 28.0 and 23.2% in patients with NIDDM, respectively. The UCP was expressed as nmol per mmol-urinary creatinine. A glucagon stimulation test was done according to the following method. After an overnight fast, 1 mg of glucagon was injected intravenously. The plasma C-peptide level was measured 0, 6 and 10 min after glucagon injection. Insulin responsiveness to glucagon was assessed by an incremental area of C-peptide response following the glucagon injection. Plasma and urinary C-peptide concentrations were measured by radioimmunoassay using a double antibody method (SN-30, Shionogi, Osaka, Japan). Assay for urinary C-peptide concentration was done with urine diluted IO-fold with 0.1 M phosphate buffer. When the obtained value of diluted urine sample was below 0.33 nM, urine was assayed without dilution of urine sample again. When the plasma C-peptide concentration was below 0.33 nM, it was assayed again by the highly sensitive method for C-peptide assay described previously [4]. Intra- and interassay coefficients of variation were 5.4 and 9.2% in the conventional C-peptide assay, and 4.9 and 9.5 ::, in the highly sensitive C-peptide assay, respectively. Plasma glucose was measured by the glucose oxidase method using a Beckman Autoanalyzer (Beckman, Fullerton, CA, U.S.A.). Hemoglobin A,, (HbA,,) was measured by the
HPLC method. Urinary creatinine was measured by an enzymatic method. Statistical analysis Data is given as mean + SD. Statistical analysis was done with the Student’s two-tailed t-test for unpaired data.
I
r---=-l
**
1
A
6-
.. t : :
.
Unstable IDDM
Stable
IDDM
NIDDM
Fig. 1. The distribution of the standard deviation of 10 previous fasting plasma glucose levels (SDFPG) in patients with IDDM and NIDDM. The IDDM were divided into two groups using the value of 3.45 nM, which was the maximal value of SDFPG of NIDDM. Dashed line represents the value of 3.45 nM. **P-c0.01.
184
: . . : Y
Unstable I DDM
Stable IDDM
NIDDM
Fig. 2. Incremental C-peptide responses to intravenous glucagon injection in the 3 groups of diabetic patients. *P < 0.05, **P < 0.01.
Results Fig. 1 shows the results of SDFPG of the three groups of patients. The value of SDFPG in NIDDM patients ranged from 0.46 to 3.45 mM, **
and the mean k SD of SDFPG was 1.53 if 0.77 mM. IDDM patients were divided into two groups according to their glycemic stability: patients with SDFPG larger than the maximal value in NIDDM patients, 3.45 mM, were regarded as having unstable glycemic control, and those with SDFPG less than 3.45 mM were regarded as having stable glycemic control. The values of SDFPG in IDDM with unstable and stable glycemic control were 4.64 + 0.57 and 2.52 k 0.60 mM, respectively. The SDFPG in IDDM with unstable glycemic control was significantly higher than that in IDDM patients with stable glycemic control (P < 0.05), and the SDFPG in IDDM patients with stable glycemic control was significantly higher than in NIDDM patients (P < 0.01). Fig. 2 shows plasma C-peptide responses to intravenous glucagon injection in three groups of patients. The incremental C-peptide response in IDDM patients with unstable glycemic control was significantly lower than in IDDM patients with stable glycemic control (0.02 +_0.02 vs 0.57 k 0.43 nM - min, P < 0.01). The incremental C-peptide response in IDDM patients with stable glycemic control was significantly lower
I
I
I
A*
**
1
AT--l
.5l
.o .5-
..
.
. 8
.
.0 -
. .
1
.
. .5-
. .. 8 _ . .
oI
i :.
c
Unstable I DDM
Stable IDDM
NIDDM
Unstable I DDM
Stable I DDM
NIDDM
Fig. 3. Urinary C-peptide excretion in overnight urine (Ul) and fasting single void urine (U2) in the 3 groups of diabetic patients. **P < 0.01.
185
0.4
0.8
1.2 1.4 1.8 2.0 UCP In U 1 (nmol/mmol creat~n~ne)
Fig. 4. Correlation between urinary C-peptide excretions in overnight urine (Ul) and fasting single void urine (U2) in the 3 groups of diabetic patients. r = 0.81; P < 0.01.
6 5c
i
0.5 UCP
1.0
in U
1
(nmol/mmol
than in NIDDM patients (3.20 k 1.12 nM . min, P < 0.01). Fig. 3 shows the results of UCP. The UCPs in Ul and U2 in NIDDM patients with unstable glycemic control (0.03 k 0.03 and 0.02 & 0.01 nmol/mmol-Cr) were significantly lower than in IDDM patients with stable glycemic control (0.24 t 0.21 and 0.20 k 0.20 nmol/mmol-Cr, both P c 0.05). The UCPs in Ul and U2 in IDDM patients with stable glycemic control were significantly lower than in NIDDM patients (0.97 k 0.51 and 0.73 + 0.41 nmol/ mmol-Cr, both P c 0.01). The maximum UCPs in Ul and U2 in IDDM patients with unstable glycemic control were as small as 0.11 and 0.03 nmol/mmol-Cr, respectively. Fig. 4 shows the correlation between UCPs in Ul and U2. The correlation was significant (Y = 0.81, P < 0.01). Fig. 5 shows the correlation between UCP and SDFPG. SDFPG correlated inversely with the UCPs in both U 1 and U2 (U 1: Y= -0.70;U2:r= -0.72,bothP
0 1991 Elsevier Science Publishers
181
B.V. 0168-8227/91/$03.50
DIABET 00525
Urinary C-peptide as an index of unstable glycemic control in insulin-dependent diabetes mellitus (IDDM) Tepyon Cha’, Yasuhiro Tahara’, Hiroshi Ikegami’, Masahiro Fukuda’, Hiroko Yoneda’, Eiji Yamato’, Yoshihiro Yamamoto’, Yoshihiko Noma2, Kenji Shima and Toshio Ogihara’ ‘Department
of Geriatric Medicine,
Osaka University Medical School, Fukushimu-ku,
Laboratory Medicine.
Tokushimu University School of Medicine,
Osaka, Japan and ‘Department
of
Tokushimu, Japan
(Received 29 January 1991) (Revision accepted 15 April 1991)
Summary
In order to investigate whether urinary C-peptide (UCP) excretion can be a useful index of insulindependent diabetes mellitus (IDDM) with unstable glycemic control, UCP was measured in nine IDDM patients with unstable glycemic control, nine IDDM patients with stable glycemic control, and 12 non-insulin-dependent diabetic (NIDDM) patients treated with insulin. The UCPs in overnight urine (U 1) and fasting single void urine (U2) in IDDM patients with unstable glycemic control were significantly lower than those in IDDM patients with stable glycemic control (Ul: 0.03 + 0.03 vs 0.24 _+0.20 nmol/mmol-Creatinine, U2: 0.02 k 0.01 vs 0.20 2 0.20 nmol/mmol-Cr, mean + SD, both P < 0.01). The UCPs in Ul and U2 in both groups of IDDM were significantly lower than those in NIDDM (Ul: 0.97 + 0.52, U2: 0.73 k 0.41 nmol/mmol-Cr, both P < 0.01). The UCPs in Ul and U2 significantly correlated with incremental C-peptide response to intravenous glucagon injection and with glycemic stability assessed by the standard deviation of 10 previous fasting plasma glucose levels. These results suggest that UCP reflects their residual insulin secretory capacity and that UCP can be a useful index which distinguishes patients with unstable IDDM from those with stable diabetes mellitus. Key words: Urinary C-peptide
excretion;
Brittle diabetes;
Introduction
Unstable glycemic control is associated with many factors such as insulin resistance, an incorrect therapeutic regimen, various recurrent complications. psychological problems, lack of residCorrespondence
to: Tepyon Cha, Dept. of Geriatric
Osaka University 553, Japan.
Medical
School Fukushima-ku,
Medicine, Osaka,
Residual p-cell function;
Glycemic
stability
ual p-cell function, and abnormalities of counterregulatory hormone secretion [ l-41. Of these factors, residual p-cell function is especially important in glycemic stability because patients without any residual P-cell function cannot restore unexpected hyper- or hypoglycemia through regulating endogenous insulin secretion. Many reports have demonstrated the relationship between residual /?-cell function and glycemic stability in insulin-dependent diabetes mellitus (IDDM)
182
[2,3,5]. Although some papers showed that patients without residual /?-cell function did not always show unstable glycemic control [ 1,6,7], our previous study with a highly sensitive assay for C-peptide [4] revealed the existence of very little residual P-cell function in IDDM with relatively stable glycemic control, and a total lack of insulinogenic reserve in patients with unstable glycemic control. Thus, complete absence of an insulinogenic reserve appears to be the fundamental cause of glycemic instability, yet even if there is a very small amount of residual P-cell function, below the detection limit of the conventional assay, this function is still very important in glycemic stability. Although a highly sensitive assay for C-peptide is very useful in the evaluation of minimal residual /?-cell function in IDDM patients, a simpler method is more suitable for routine clinical use. Urinary C-peptide (UCP) excretion is considered to reflect integrated insulin secretion because C-peptide is secreted equimolarly with insulin, and part of the secreted C-peptide is excreted in the urine [ 8- 131. Although the amount of C-peptide excreted in the urine is affected by various conditions [ 14,151, many authors showed that UCP can provide useful information in the evaluation of insulin secretory capacity, especially in IDDM [ 8,16-211. However, there is little information about whether UCP can also provide information in the evaluation of very low residual P-cell function and glycemic stability in IDDM. In the present study we examined the relationships between UCP and glycemic stability in IDDM, and we investigated whether UCP can be a useful index of residual b-cell function for distinguishing IDDM patients with unstable glycemic control from those with stable glycemic control.
Subjects and Methods The subjects studied were 18 patients with IDDM and 12 patients with NIDDM who were attending the Department of Geriatric Medicine at Osaka University Hospital. The diagnosis of
diabetes was made according to the criteria of the World Health Organization. None of them had chronic renal failure as defined by either endogenous creatinine clearance of less than 0.5 ml/s or a plasma creatinine level greater than 130 PM. Patients with urinary tract infections were excluded. IDDM was determined when patients lacked endogenous insulin secretion, as assessed by the incremental area of C-peptide after intravenous administration of glucagon at less than 1.5 nM min (see below). All patients with IDDM experienced an episode of ketonuria or diabetic ketoacidosis. Glycemic stability was assessed by the standard deviation of the 10 values of fasting plasma glucose (SDFPG), which was measured during their visits once a month PI. IDDM patients were divided into 2 groups according to their glycemic stability. Patients with SDFPG larger than the maximal value of SDFPG for NIDDM were regarded as having unstable glycemic control, while those with SDFPG less than that value were regarded as having stable glycemic control. Clinical characteristics of the patients are shown in Table 1. There were no significant differences in age, known duration of diabetes, body mass index, or glycemic control level between the IDDM patients with unstable and stable glycemic control. The therapeutic mode for each patient was determined by each attending physician in order to achieve a permissible glycemic control level. IDDM patients with unstable glycemic control had been treated with more intensive therapy than those with stable glycemic control. Patients were asked to void urine just before sleeping, and this urine was discarded. Overnight urine voided just after getting up in the morning was collected and designated as U 1. A fasting urine voided just after arriving at the outpatient clinic was also collected and designated as U2. UCP was measured on three days and the mean of the 3 determinations was used in this study. The mean coefficients of day-to-day variation of UCP in Ul and U2 were 27.0 and 17.6% in IDDM patients with unstable glycemic control, 26.1 and 23.1 Y0 in patients with stable glycemic
183 TABLE 1 Clinical characteristics
of the patients
Number (M:F)
studied
Age (yrs)
Duration of DM
HbA,, (%)
BMI
(mM)
(kg/m’ )
Therapy CSII:Mix:II:I
10.21 + 2.61 8.38 + 0.55 9.58 f 1.78
9.3 + 0.9 8.9 f 1.7 8.3 k 1.5
19.6 + 3.0 19.2 + 1.5;’ 22.1 + 3.0
5:4:0:0 2:6:1:1 0: 1:7:4
FPG
(yrs) IDDM unstable stable NIDDM
4:5 4:5 9:3
45 + 14” 39 + 13b 57* 7
10 * 4 12 + 8 14 f 7
Values are mean + SD. Unstable: IDDM patients with unstable glycemic control. Stable: IDDM patients with stable glycemic control. FPG: fasting plasma glucose. HbA,,: hemoglobin A,, (normal range: 4.4-6.4%). BMI: body mass index. CSII: continuous subcutaneous insulin infusion. Mix: two or three daily injections of a mixture ofintermediateand short-acting insulin. II: two daily injections of intermediate-acting insulin. I: single daily injection of intermediate-acting insulin. “P < 0.05; ‘P < 0.01 vs type 2.
control, and 28.0 and 23.2% in patients with NIDDM, respectively. The UCP was expressed as nmol per mmol-urinary creatinine. A glucagon stimulation test was done according to the following method. After an overnight fast, 1 mg of glucagon was injected intravenously. The plasma C-peptide level was measured 0, 6 and 10 min after glucagon injection. Insulin responsiveness to glucagon was assessed by an incremental area of C-peptide response following the glucagon injection. Plasma and urinary C-peptide concentrations were measured by radioimmunoassay using a double antibody method (SN-30, Shionogi, Osaka, Japan). Assay for urinary C-peptide concentration was done with urine diluted IO-fold with 0.1 M phosphate buffer. When the obtained value of diluted urine sample was below 0.33 nM, urine was assayed without dilution of urine sample again. When the plasma C-peptide concentration was below 0.33 nM, it was assayed again by the highly sensitive method for C-peptide assay described previously [4]. Intra- and interassay coefficients of variation were 5.4 and 9.2% in the conventional C-peptide assay, and 4.9 and 9.5 ::, in the highly sensitive C-peptide assay, respectively. Plasma glucose was measured by the glucose oxidase method using a Beckman Autoanalyzer (Beckman, Fullerton, CA, U.S.A.). Hemoglobin A,, (HbA,,) was measured by the
HPLC method. Urinary creatinine was measured by an enzymatic method. Statistical analysis Data is given as mean + SD. Statistical analysis was done with the Student’s two-tailed t-test for unpaired data.
I
r---=-l
**
1
A
6-
.. t : :
.
Unstable IDDM
Stable
IDDM
NIDDM
Fig. 1. The distribution of the standard deviation of 10 previous fasting plasma glucose levels (SDFPG) in patients with IDDM and NIDDM. The IDDM were divided into two groups using the value of 3.45 nM, which was the maximal value of SDFPG of NIDDM. Dashed line represents the value of 3.45 nM. **P-c0.01.
184
: . . : Y
Unstable I DDM
Stable IDDM
NIDDM
Fig. 2. Incremental C-peptide responses to intravenous glucagon injection in the 3 groups of diabetic patients. *P < 0.05, **P < 0.01.
Results Fig. 1 shows the results of SDFPG of the three groups of patients. The value of SDFPG in NIDDM patients ranged from 0.46 to 3.45 mM, **
and the mean k SD of SDFPG was 1.53 if 0.77 mM. IDDM patients were divided into two groups according to their glycemic stability: patients with SDFPG larger than the maximal value in NIDDM patients, 3.45 mM, were regarded as having unstable glycemic control, and those with SDFPG less than 3.45 mM were regarded as having stable glycemic control. The values of SDFPG in IDDM with unstable and stable glycemic control were 4.64 + 0.57 and 2.52 k 0.60 mM, respectively. The SDFPG in IDDM with unstable glycemic control was significantly higher than that in IDDM patients with stable glycemic control (P < 0.05), and the SDFPG in IDDM patients with stable glycemic control was significantly higher than in NIDDM patients (P < 0.01). Fig. 2 shows plasma C-peptide responses to intravenous glucagon injection in three groups of patients. The incremental C-peptide response in IDDM patients with unstable glycemic control was significantly lower than in IDDM patients with stable glycemic control (0.02 +_0.02 vs 0.57 k 0.43 nM - min, P < 0.01). The incremental C-peptide response in IDDM patients with stable glycemic control was significantly lower
I
I
I
A*
**
1
AT--l
.5l
.o .5-
..
.
. 8
.
.0 -
. .
1
.
. .5-
. .. 8 _ . .
oI
i :.
c
Unstable I DDM
Stable IDDM
NIDDM
Unstable I DDM
Stable I DDM
NIDDM
Fig. 3. Urinary C-peptide excretion in overnight urine (Ul) and fasting single void urine (U2) in the 3 groups of diabetic patients. **P < 0.01.
185
0.4
0.8
1.2 1.4 1.8 2.0 UCP In U 1 (nmol/mmol creat~n~ne)
Fig. 4. Correlation between urinary C-peptide excretions in overnight urine (Ul) and fasting single void urine (U2) in the 3 groups of diabetic patients. r = 0.81; P < 0.01.
6 5c
i
0.5 UCP
1.0
in U
1
(nmol/mmol
than in NIDDM patients (3.20 k 1.12 nM . min, P < 0.01). Fig. 3 shows the results of UCP. The UCPs in Ul and U2 in NIDDM patients with unstable glycemic control (0.03 k 0.03 and 0.02 & 0.01 nmol/mmol-Cr) were significantly lower than in IDDM patients with stable glycemic control (0.24 t 0.21 and 0.20 k 0.20 nmol/mmol-Cr, both P c 0.05). The UCPs in Ul and U2 in IDDM patients with stable glycemic control were significantly lower than in NIDDM patients (0.97 k 0.51 and 0.73 + 0.41 nmol/ mmol-Cr, both P c 0.01). The maximum UCPs in Ul and U2 in IDDM patients with unstable glycemic control were as small as 0.11 and 0.03 nmol/mmol-Cr, respectively. Fig. 4 shows the correlation between UCPs in Ul and U2. The correlation was significant (Y = 0.81, P < 0.01). Fig. 5 shows the correlation between UCP and SDFPG. SDFPG correlated inversely with the UCPs in both U 1 and U2 (U 1: Y= -0.70;U2:r= -0.72,bothP
