- Email: [email protected]
Postprandial plasma glucose: a good index of glycemic control in type 2 diabetic patients having near-normal fasting glucose levels
Diabetes Research and Clinical Practice 46 (1999) 23 – 27 www.elsevier.com/locate/diabres
Postprandial plasma glucose: a good index of glycemic control in type 2 diabetic patients having near-normal fasting glucose levels Supamai Soonthornpun, Chatchalit Rattarasarn *, Rattana Leelawattana, Worawong Setasuban Di6ision of Endocrinology and Metabolism, Department of Medicine, Faculty of Medicine, Prince of Songkla Uni6ersity, Hat-Yai Songkhla 90110, Thailand Received 14 January 1999; received in revised form 23 February 1999; accepted 16 April 1999
Abstract To investigate the effect of postprandial plasma glucose (PG) concentrations on HbA1c levels in type 2 diabetic patients, we evaluated the relationship between HbA1c levels and postprandial PG concentrations after a meal tolerance test in 35 type 2 diabetic patients who had fasting PG concentrations persistently B7.8 mmol/l and stable HbA1c levels. Two-hour postprandial PG concentrations were found to be more strongly correlated (r= 0.51) with HbA1c levels than 1-h postprandial PG (r= 0.35) and fasting PG (r =0.46) concentrations. Patients whose HbA1c levels were high (HbA1c ]7%) had significantly higher 2-h postprandial PG concentrations and areas under the glucose curve than those whose HbA1c levels were lower (8.1291.10 (SD) vs 6.709 2.22 mmol l − 1, P= 0.004 and 17.43 9 1.92 vs 15.58 93.26 mmol h − 1 l − 1, P=0.02, respectively). Although fasting PG concentrations of patients with higher HbA1c levels were slightly higher, they did not differ significantly from those with lower HbA1c levels (6.21 90.89 vs 5.7390.68 mmol l − 1). Age, duration of diabetes, body mass index, serum C-peptide, both fasting and postprandial, did not differ between these two groups. This study suggests that postprandial hyperglycemia, particularly 2-h postprandial PG concentrations, is associated with high HbA1c levels in type 2 diabetic patients whose fasting PG levels were within normal or near-normal levels. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Postprandial plasma glucose; Type 2 diabetes; Glycosylated hemoglobin
1. Introduction
* Corresponding author. Tel.: +66-74-212070 extn. 1463; fax: +66-74-212900/212903. E-mail address: [email protected] (C. Rattarasarn)
The relationship between glycemic control and development of diabetic complications is widely accepted. Since diabetes mellitus is characterized by both fasting hyperglycemia and exaggerated
0168-8227/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 9 9 ) 0 0 0 6 1 - 3
S. Soonthornpun et al. / Diabetes Research and Clinical Practice 46 (1999) 23–27
24
postprandial increases in plasma glucose levels, the goals of therapy should be to normalize not only fasting but also postprandial plasma glucose (PG) concentrations. Although HbA1c is recommended for use as a target of glycemic control, fasting PG levels may be used more routinely in some clinics in the world than HbA1c, considering the cost of the latter [1]. Many studies have shown a correlation between fasting PG and HbA1c levels. However, the best correlations were in studies with fasting PG levels spanning wide ranges [2 – 4]. In our experience, we have observed significant numbers of type 2 diabetic patients with inappropriately high HbA1c levels despite fasting PG persistently B 7.8 mmol l − 1. From a theoretical standpoint, two possibilities could explain this discrepancy between fasting PG and HbA1c. First, it is possible that poor glycemic control had occurred at home and patients had better control before attending clinics. However since the rate of formation of HbA1c is faster than its rate of disappearance [5], short periods of caloric deprivation before a clinic visit resulting in a dramatic fall in fasting PG levels should not improve HbA1c levels. Second, there might be an excessive postprandial glycemic excursion among these patients resulting in an increase in mean blood glucose and HbA1c levels. Since one of the major defects in b-cell function in type 2 diabetes is a decrease in glucose responsiveness, this cannot be entirely corrected even though the fasting or preprandial glucose is lowered to within or close to normal levels. Therefore the likelihood that these patients would have postprandial hyperglycemia was high. Furthermore, most sulfonylureas, although shown to improve fasting PG levels, do not correct postprandial hyperglycemia in most circumstances [6]. This study was undertaken to test the hypothesis that postprandial hyperglycemia is associated with raised HbA1c levels in type 2 diabetic patients who have normal or near-normal fasting PG concentrations. 2. Patients and methods
2.1. Study design Type 2 diabetic patients who were regularly
followed at the diabetic clinic of Songklanagarind hospital and had fasting PGB 7.8 mmol l − 1 regardless of HbA1c levels were invited to participate in the study. They were asked to continue the same diet and oral hypoglycemic agents and returned to clinic twice, 2-months apart, when fasting PG and HbA1c were measured. Those who could maintain fasting PGB 7.8 mmol l − 1 and stable HbA1c indicated by changes in HbA1c levels of no more than 0.5% in these two consecutive visits were scheduled for meal tolerance test within the next 4 weeks. Patients were excluded if the previous criteria of fasting PG and HbA1c were exceeded on the day of meal tolerance test. The meal tolerance test was performed after an overnight fast. The 300 kcal-rice soup prepared by hospital dietician (50% carbohydrate, 30% fat and 20% protein) was given in the morning and consumed within 30 min. Blood samples were drawn before and at 1 and 2 h after meal for measurement of plasma glucose and serum C-peptide. Time 0 was at the first mouthful of food. Patients continued their usual oral hypoglycemic agents. The study was approved by the ethics committee of Faculty of Medicine, Prince of Songkla University and all patients gave written informed consent before entering the study.
2.2. Assays Plasma glucose was determined by the glucose oxidase method using automated Hitachi model 704. HbA1c was determined by the latex immunoagglutination inhibition method using DCA 2000 analyzer (Bayer, Germany) with a normal range of 4.3–5.7%. Intra- and inter-assay CV of the test were 2.2–3.7 and 0.9–4.3%, respectively. Serum C-peptide was determined by double antibody radioimmunoassay using a commercial kit (Diagnostic Products Corporation, USA).
2.3. Statistical analysis All results were given as mean 9 SD. Statistical analysis were carried out using SPSS for MS windows Release 6.0. Simple linear regression analysis was used to examine the relationship of plasma glucose and HbA1c. Mann–Whitney test
S. Soonthornpun et al. / Diabetes Research and Clinical Practice 46 (1999) 23–27
Fig. 1. This figure demonstrates the correlation between fasting (A), 1-h postprandial (B) and 2-h postprandial (C) plasma glucose (PG) and HbA1c levels.
was used for group comparisons. The area under the curve values were calculated by the trapezoidal method. For all analyses, a P value of B 0.05 was considered significant.
3. Results Of 64 patients initially eligible for the study, only 35 patients (13 men and 22 women) fulfilled all inclusion criteria. Their mean age and mean duration of diabetes were 60.319 10.08 (range
25
38–84) and 6.089 4.6 (range 1–18) years, respectively. Four patients were treated with diet alone, 20 were treated with sulfonylureas, four were treated with metformin and seven were treated with combined sulfonylureas and metformin. None were treated with insulin. The fasting, 1and 2-h postprandial PG values were all significantly and linearly correlated with HbA1c levels (Fig. 1). The 1-h postprandial PG had the weakest (r= 0.35) whereas 2-h postprandial PG showed the strongest correlation (r=0.51) with HbA1c. Based on glycemic targets classified by European NIDDM Policy Group [7], patients who have fasting PG levels B7.8 mmol l − 1 should have HbA1c levelsBmean+ 5SD (or 7% by our method). Of 35 patients who all had fasting PG persistently B7.8 mmol l − 1, 24 (68.6%) had HbA1cB7% and 11 (31.4%) had HbA1c] 7%. The clinical characteristics between these two groups included age, duration of diabetes and body mass index were not different (Table 1). Concerning mode of therapy; of 11 patients with high HbA1c levels ( ] 7%), one patient was treated with diet only, four were on sulfonylureas, two on metformin and four used combined sulfonylureas and metformin. The respective modes of therapy in the 24 patients with lower HbA1c levels (B 7%) were three, 16, two and three pa-
Table 1 Characteristics of patients grouped by levels of HbA1c HbA1cB7%
HbA1c]7%
P value
Number (male/female) Age (years) Duration of diabetes (years) Body mass index (kg m−2)
24 (9/15) 60.67 910.00 5.37 93.48 26.4493.67
11 (4/7) 59.55 910.69 7.64 96.34 25.81 92.23
0.741 0.241 0.917
Serum C-peptide (nmol l−1) Fasting 1-h Postprandial 2-h Postprandial
0.66 90.32 1.52 90.62 1.57 9 0.57
0.73 9 0.49 1.53 9 0.63 1.53 90.60
0.808 0.602 0.651
Plasma glucose (mmol l−1) Fasting 1-h Postprandial 2-h Postprandial
5.73 90.68 9.37 92.12 6.70 9 2.22
6.21 90.89 10.26 91.40 8.12 91.10
0.186 0.095 0.004
15.58 9 3.26
17.43 9 1.92
0.020
Areas under the glucose curve (mmol h−1 l−1)
26
S. Soonthornpun et al. / Diabetes Research and Clinical Practice 46 (1999) 23–27
tients. As shown in Table 1, plasma glucose levels of both groups did not differ in the fasting state but at 2-h postprandial, was significantly higher in those who had high HbA1c levels. The area under the glucose curve was also significantly higher in patients who had high HbA1c levels. Serum Cpeptide responses after meal in patients with high HbA1c levels were not different from those with lower HbA1c levels.
4. Discussion Hyperglycemia in type 2 diabetes develops as the result of inadequate insulin secretion together with peripheral resistance to the action of insulin. At the early stage of the disease, there is a delay and insufficiency of early phase insulin secretion in response to glucose resulting in hyperglycemia after glucose loading (impaired glucose tolerance). Thereafter, the defects of bcell function become progressive leading to deterioration of glucose homeostasis and both fasting and postprandial hyperglycemia ultimately develop [8]. Since it takes more than 2 – 3 h after a meal for postprandial PG to return to preprandial levels in type 2 diabetic subjects, these patients would be exposed to varying degrees of mealrelated hyperglycemia, yet the fasting PG on the next day may be normalized [9]. Therefore, treatment which corrects fasting but not postprandial hyperglycemia would not be able to normalize overall glucose control in diabetic patients. Our study confirmed this notion. Patients with postprandial hyperglycemia indicated by higher area under the glucose curve particularly 2 h post food had higher HbA1c levels despite normal or near-normal fasting PG levels. Furthermore, 2-h postprandial PG but not fasting PG, was correlated better with HbA1c levels indicating the importance of postprandial PG as an index of metabolic control under this circumstance. The results of our study are consistent with the study of Guillausseau [10] which showed better correlation of postprandial PG with HbA1c levels amongst type 2 diabetic patients who had mild fasting hyperglycemia. HbA1c was found to be strongly correlated with
fasting PG only in those with higher degree of hyperglycemia [2–4]. To simulate the actual practice of type 2 diabetic patients, oral hypoglycemic agents were not withdrawn at the day of the meal test in our study. Patients who were treated with diet or sulfonylureas and/or metformin in either group were similar. None was on a-glucosidase inhibitor, an agent known to decrease postprandial glucose levels. The administration of oral hypoglycemic agents prior to the test meal in this study would be unlikely to affect the study results. In conclusion, our study has shown that about one-third of type 2 diabetic patients who had fasting PG within normal or near-normal levels had inappropriately high HbA1c levels and elevated postprandial PG levels. These were associated with high HbA1c levels. Further studies with regard to the importance of postprandial or non-fasting hyperglycemia in diabetic complications of type 2 diabetes should be performed.
Acknowledgements We are grateful to Atchara Thamprasit for assistance in performing meal tolerance tests and Kalaya Leetanaporn for assisting in C-peptide assay. This study was supported by Faculty research grant from Faculty of Medicine, Prince of Songkla University.
References [1] A. Muir, S.A. Howe-Davies, R.C. Turner, General practice care of non-insulin-dependent diabetes with fasting blood glucose measurements, Am. J. Med. 73 (1982) 637 – 640. [2] B. Gonen, A.H. Rubenstein, H. Rochman, S.P. Tanega, D.L. Horwitz, Hemoglobin A1: an indicator of the metabolic control of diabetic patients, Lancet 2 (1977) 734 – 737. [3] R.J. Graf, J.B. Halter, D. Porte Jr., Glycosylated hemoglobin in normal subjects and subjects with maturity-onset diabetes: evidence for a saturable system in man, Diabetes 27 (1978) 834 – 839. [4] R.B. Paisey, P. Bradshaw, M. Hartog, Home blood glu-
S. Soonthornpun et al. / Diabetes Research and Clinical Practice 46 (1999) 23–27
27
top guide for the management of non-insulin-dependent diabetes mellitus (NIDDM): an update, Diabetic Med. 15 (1994) 899 – 909. [8] R.A. DeFronzo, R.C. Bonadonna, E. Ferrannini, Pathogenesis of NIDDM. A balanced overview, Diabetes Care 15 (1992) 318 – 368. [9] S. Dinneen, J. Gerich, R. Rizza, Carbohydrate metabolism in non-insulin-dependent diabetes mellitus, New Engl. J. Med. 327 (1992) 707 – 713. [10] P.J. Guillausseau, Monitoring of metabolic control in patients with non-insulin- dependent diabetes mellitus on oral hypoglycemic agents: value of evening blood glucose determination, Diabetic Med. 14 (1997) 798 – 802.
cose concentrations in maturity-onset diabetes, Br. Med. J. 280 (1980) 596 – 598. [5] G. Boden, R.W. Master, S.S. Gordon, C.R. Shuman, O.E. Owen, Monitoring metabolic control in diabetic outpatients with glycosylated hemoglobin, Ann. Intern. Med. 92 (1980) 357 –360. [6] E.T. Shapiro, E.V. Cauter, H. Tillil, et al., Glyburide enhances the responsiveness of the b-cell to glucose but does not correct the abnormal patterns of insulin secretion in noninsulin-dependent diabetes mellitus, J. Clin. Endocrinol. Metab. 69 (1989) 571–576. [7] K.G.M.M. Alberti, F.A. Gries, J. Jervell, et al., A desk-
.
Postprandial plasma glucose: a good index of glycemic control in type 2 diabetic patients having near-normal fasting glucose levels Supamai Soonthornpun, Chatchalit Rattarasarn *, Rattana Leelawattana, Worawong Setasuban Di6ision of Endocrinology and Metabolism, Department of Medicine, Faculty of Medicine, Prince of Songkla Uni6ersity, Hat-Yai Songkhla 90110, Thailand Received 14 January 1999; received in revised form 23 February 1999; accepted 16 April 1999
Abstract To investigate the effect of postprandial plasma glucose (PG) concentrations on HbA1c levels in type 2 diabetic patients, we evaluated the relationship between HbA1c levels and postprandial PG concentrations after a meal tolerance test in 35 type 2 diabetic patients who had fasting PG concentrations persistently B7.8 mmol/l and stable HbA1c levels. Two-hour postprandial PG concentrations were found to be more strongly correlated (r= 0.51) with HbA1c levels than 1-h postprandial PG (r= 0.35) and fasting PG (r =0.46) concentrations. Patients whose HbA1c levels were high (HbA1c ]7%) had significantly higher 2-h postprandial PG concentrations and areas under the glucose curve than those whose HbA1c levels were lower (8.1291.10 (SD) vs 6.709 2.22 mmol l − 1, P= 0.004 and 17.43 9 1.92 vs 15.58 93.26 mmol h − 1 l − 1, P=0.02, respectively). Although fasting PG concentrations of patients with higher HbA1c levels were slightly higher, they did not differ significantly from those with lower HbA1c levels (6.21 90.89 vs 5.7390.68 mmol l − 1). Age, duration of diabetes, body mass index, serum C-peptide, both fasting and postprandial, did not differ between these two groups. This study suggests that postprandial hyperglycemia, particularly 2-h postprandial PG concentrations, is associated with high HbA1c levels in type 2 diabetic patients whose fasting PG levels were within normal or near-normal levels. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Postprandial plasma glucose; Type 2 diabetes; Glycosylated hemoglobin
1. Introduction
* Corresponding author. Tel.: +66-74-212070 extn. 1463; fax: +66-74-212900/212903. E-mail address: [email protected] (C. Rattarasarn)
The relationship between glycemic control and development of diabetic complications is widely accepted. Since diabetes mellitus is characterized by both fasting hyperglycemia and exaggerated
0168-8227/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 9 9 ) 0 0 0 6 1 - 3
S. Soonthornpun et al. / Diabetes Research and Clinical Practice 46 (1999) 23–27
24
postprandial increases in plasma glucose levels, the goals of therapy should be to normalize not only fasting but also postprandial plasma glucose (PG) concentrations. Although HbA1c is recommended for use as a target of glycemic control, fasting PG levels may be used more routinely in some clinics in the world than HbA1c, considering the cost of the latter [1]. Many studies have shown a correlation between fasting PG and HbA1c levels. However, the best correlations were in studies with fasting PG levels spanning wide ranges [2 – 4]. In our experience, we have observed significant numbers of type 2 diabetic patients with inappropriately high HbA1c levels despite fasting PG persistently B 7.8 mmol l − 1. From a theoretical standpoint, two possibilities could explain this discrepancy between fasting PG and HbA1c. First, it is possible that poor glycemic control had occurred at home and patients had better control before attending clinics. However since the rate of formation of HbA1c is faster than its rate of disappearance [5], short periods of caloric deprivation before a clinic visit resulting in a dramatic fall in fasting PG levels should not improve HbA1c levels. Second, there might be an excessive postprandial glycemic excursion among these patients resulting in an increase in mean blood glucose and HbA1c levels. Since one of the major defects in b-cell function in type 2 diabetes is a decrease in glucose responsiveness, this cannot be entirely corrected even though the fasting or preprandial glucose is lowered to within or close to normal levels. Therefore the likelihood that these patients would have postprandial hyperglycemia was high. Furthermore, most sulfonylureas, although shown to improve fasting PG levels, do not correct postprandial hyperglycemia in most circumstances [6]. This study was undertaken to test the hypothesis that postprandial hyperglycemia is associated with raised HbA1c levels in type 2 diabetic patients who have normal or near-normal fasting PG concentrations. 2. Patients and methods
2.1. Study design Type 2 diabetic patients who were regularly
followed at the diabetic clinic of Songklanagarind hospital and had fasting PGB 7.8 mmol l − 1 regardless of HbA1c levels were invited to participate in the study. They were asked to continue the same diet and oral hypoglycemic agents and returned to clinic twice, 2-months apart, when fasting PG and HbA1c were measured. Those who could maintain fasting PGB 7.8 mmol l − 1 and stable HbA1c indicated by changes in HbA1c levels of no more than 0.5% in these two consecutive visits were scheduled for meal tolerance test within the next 4 weeks. Patients were excluded if the previous criteria of fasting PG and HbA1c were exceeded on the day of meal tolerance test. The meal tolerance test was performed after an overnight fast. The 300 kcal-rice soup prepared by hospital dietician (50% carbohydrate, 30% fat and 20% protein) was given in the morning and consumed within 30 min. Blood samples were drawn before and at 1 and 2 h after meal for measurement of plasma glucose and serum C-peptide. Time 0 was at the first mouthful of food. Patients continued their usual oral hypoglycemic agents. The study was approved by the ethics committee of Faculty of Medicine, Prince of Songkla University and all patients gave written informed consent before entering the study.
2.2. Assays Plasma glucose was determined by the glucose oxidase method using automated Hitachi model 704. HbA1c was determined by the latex immunoagglutination inhibition method using DCA 2000 analyzer (Bayer, Germany) with a normal range of 4.3–5.7%. Intra- and inter-assay CV of the test were 2.2–3.7 and 0.9–4.3%, respectively. Serum C-peptide was determined by double antibody radioimmunoassay using a commercial kit (Diagnostic Products Corporation, USA).
2.3. Statistical analysis All results were given as mean 9 SD. Statistical analysis were carried out using SPSS for MS windows Release 6.0. Simple linear regression analysis was used to examine the relationship of plasma glucose and HbA1c. Mann–Whitney test
S. Soonthornpun et al. / Diabetes Research and Clinical Practice 46 (1999) 23–27
Fig. 1. This figure demonstrates the correlation between fasting (A), 1-h postprandial (B) and 2-h postprandial (C) plasma glucose (PG) and HbA1c levels.
was used for group comparisons. The area under the curve values were calculated by the trapezoidal method. For all analyses, a P value of B 0.05 was considered significant.
3. Results Of 64 patients initially eligible for the study, only 35 patients (13 men and 22 women) fulfilled all inclusion criteria. Their mean age and mean duration of diabetes were 60.319 10.08 (range
25
38–84) and 6.089 4.6 (range 1–18) years, respectively. Four patients were treated with diet alone, 20 were treated with sulfonylureas, four were treated with metformin and seven were treated with combined sulfonylureas and metformin. None were treated with insulin. The fasting, 1and 2-h postprandial PG values were all significantly and linearly correlated with HbA1c levels (Fig. 1). The 1-h postprandial PG had the weakest (r= 0.35) whereas 2-h postprandial PG showed the strongest correlation (r=0.51) with HbA1c. Based on glycemic targets classified by European NIDDM Policy Group [7], patients who have fasting PG levels B7.8 mmol l − 1 should have HbA1c levelsBmean+ 5SD (or 7% by our method). Of 35 patients who all had fasting PG persistently B7.8 mmol l − 1, 24 (68.6%) had HbA1cB7% and 11 (31.4%) had HbA1c] 7%. The clinical characteristics between these two groups included age, duration of diabetes and body mass index were not different (Table 1). Concerning mode of therapy; of 11 patients with high HbA1c levels ( ] 7%), one patient was treated with diet only, four were on sulfonylureas, two on metformin and four used combined sulfonylureas and metformin. The respective modes of therapy in the 24 patients with lower HbA1c levels (B 7%) were three, 16, two and three pa-
Table 1 Characteristics of patients grouped by levels of HbA1c HbA1cB7%
HbA1c]7%
P value
Number (male/female) Age (years) Duration of diabetes (years) Body mass index (kg m−2)
24 (9/15) 60.67 910.00 5.37 93.48 26.4493.67
11 (4/7) 59.55 910.69 7.64 96.34 25.81 92.23
0.741 0.241 0.917
Serum C-peptide (nmol l−1) Fasting 1-h Postprandial 2-h Postprandial
0.66 90.32 1.52 90.62 1.57 9 0.57
0.73 9 0.49 1.53 9 0.63 1.53 90.60
0.808 0.602 0.651
Plasma glucose (mmol l−1) Fasting 1-h Postprandial 2-h Postprandial
5.73 90.68 9.37 92.12 6.70 9 2.22
6.21 90.89 10.26 91.40 8.12 91.10
0.186 0.095 0.004
15.58 9 3.26
17.43 9 1.92
0.020
Areas under the glucose curve (mmol h−1 l−1)
26
S. Soonthornpun et al. / Diabetes Research and Clinical Practice 46 (1999) 23–27
tients. As shown in Table 1, plasma glucose levels of both groups did not differ in the fasting state but at 2-h postprandial, was significantly higher in those who had high HbA1c levels. The area under the glucose curve was also significantly higher in patients who had high HbA1c levels. Serum Cpeptide responses after meal in patients with high HbA1c levels were not different from those with lower HbA1c levels.
4. Discussion Hyperglycemia in type 2 diabetes develops as the result of inadequate insulin secretion together with peripheral resistance to the action of insulin. At the early stage of the disease, there is a delay and insufficiency of early phase insulin secretion in response to glucose resulting in hyperglycemia after glucose loading (impaired glucose tolerance). Thereafter, the defects of bcell function become progressive leading to deterioration of glucose homeostasis and both fasting and postprandial hyperglycemia ultimately develop [8]. Since it takes more than 2 – 3 h after a meal for postprandial PG to return to preprandial levels in type 2 diabetic subjects, these patients would be exposed to varying degrees of mealrelated hyperglycemia, yet the fasting PG on the next day may be normalized [9]. Therefore, treatment which corrects fasting but not postprandial hyperglycemia would not be able to normalize overall glucose control in diabetic patients. Our study confirmed this notion. Patients with postprandial hyperglycemia indicated by higher area under the glucose curve particularly 2 h post food had higher HbA1c levels despite normal or near-normal fasting PG levels. Furthermore, 2-h postprandial PG but not fasting PG, was correlated better with HbA1c levels indicating the importance of postprandial PG as an index of metabolic control under this circumstance. The results of our study are consistent with the study of Guillausseau [10] which showed better correlation of postprandial PG with HbA1c levels amongst type 2 diabetic patients who had mild fasting hyperglycemia. HbA1c was found to be strongly correlated with
fasting PG only in those with higher degree of hyperglycemia [2–4]. To simulate the actual practice of type 2 diabetic patients, oral hypoglycemic agents were not withdrawn at the day of the meal test in our study. Patients who were treated with diet or sulfonylureas and/or metformin in either group were similar. None was on a-glucosidase inhibitor, an agent known to decrease postprandial glucose levels. The administration of oral hypoglycemic agents prior to the test meal in this study would be unlikely to affect the study results. In conclusion, our study has shown that about one-third of type 2 diabetic patients who had fasting PG within normal or near-normal levels had inappropriately high HbA1c levels and elevated postprandial PG levels. These were associated with high HbA1c levels. Further studies with regard to the importance of postprandial or non-fasting hyperglycemia in diabetic complications of type 2 diabetes should be performed.
Acknowledgements We are grateful to Atchara Thamprasit for assistance in performing meal tolerance tests and Kalaya Leetanaporn for assisting in C-peptide assay. This study was supported by Faculty research grant from Faculty of Medicine, Prince of Songkla University.
References [1] A. Muir, S.A. Howe-Davies, R.C. Turner, General practice care of non-insulin-dependent diabetes with fasting blood glucose measurements, Am. J. Med. 73 (1982) 637 – 640. [2] B. Gonen, A.H. Rubenstein, H. Rochman, S.P. Tanega, D.L. Horwitz, Hemoglobin A1: an indicator of the metabolic control of diabetic patients, Lancet 2 (1977) 734 – 737. [3] R.J. Graf, J.B. Halter, D. Porte Jr., Glycosylated hemoglobin in normal subjects and subjects with maturity-onset diabetes: evidence for a saturable system in man, Diabetes 27 (1978) 834 – 839. [4] R.B. Paisey, P. Bradshaw, M. Hartog, Home blood glu-
S. Soonthornpun et al. / Diabetes Research and Clinical Practice 46 (1999) 23–27
27
top guide for the management of non-insulin-dependent diabetes mellitus (NIDDM): an update, Diabetic Med. 15 (1994) 899 – 909. [8] R.A. DeFronzo, R.C. Bonadonna, E. Ferrannini, Pathogenesis of NIDDM. A balanced overview, Diabetes Care 15 (1992) 318 – 368. [9] S. Dinneen, J. Gerich, R. Rizza, Carbohydrate metabolism in non-insulin-dependent diabetes mellitus, New Engl. J. Med. 327 (1992) 707 – 713. [10] P.J. Guillausseau, Monitoring of metabolic control in patients with non-insulin- dependent diabetes mellitus on oral hypoglycemic agents: value of evening blood glucose determination, Diabetic Med. 14 (1997) 798 – 802.
cose concentrations in maturity-onset diabetes, Br. Med. J. 280 (1980) 596 – 598. [5] G. Boden, R.W. Master, S.S. Gordon, C.R. Shuman, O.E. Owen, Monitoring metabolic control in diabetic outpatients with glycosylated hemoglobin, Ann. Intern. Med. 92 (1980) 357 –360. [6] E.T. Shapiro, E.V. Cauter, H. Tillil, et al., Glyburide enhances the responsiveness of the b-cell to glucose but does not correct the abnormal patterns of insulin secretion in noninsulin-dependent diabetes mellitus, J. Clin. Endocrinol. Metab. 69 (1989) 571–576. [7] K.G.M.M. Alberti, F.A. Gries, J. Jervell, et al., A desk-
.