- Email: [email protected]
The effect of improved post-prandial blood glucose control on post-prandial metabolism and markers of vascular risk in people with Type 2 diabetes
Diabetes Research and Clinical Practice 67 (2005) 196–203 www.elsevier.com/locate/diabres
The effect of improved post-prandial blood glucose control on post-prandial metabolism and markers of vascular risk in people with Type 2 diabetes$ A. Gallagher*, P.D. Home School of Clinical Medical Sciences, Floor 4, William Leech Building, The Medical School, Framlington Place, Newcastle upon Tyne NE2 4HH, UK Received 21 January 2004; received in revised form 30 June 2004; accepted 7 July 2004
Abstract A variety of abnormalities of metabolic, haemostatic and endothelial markers are associated with Type 2 diabetes. Evidence suggests that poor post-prandial blood glucose control may contribute to vascular risk. We aimed to examine whether the restoration of a more physiological insulin profile post-prandially would improve these abnormalities. Twenty-one patients with insulin-treated Type 2 diabetes were recruited into a single centre, crossover, double-blind study. Patients were randomized to unmodified human insulin or insulin aspart before main meals for 6-week study periods, both together with NPH insulin. At the end of each study period, pre-breakfast levels of markers of vascular risk were assessed and a test meal performed. There was no significant difference in HbA1c (7.04 0.13% (S.E.) versus 7.15 0.11%, P = 0.060) with insulin aspart compared to human insulin at the end of each study period. The mean post-prandial blood glucose concentration at 90 min from self-monitored results was lower with insulin aspart than with human insulin (7.9 0.4 mmol/l versus 9.3 0.4 mmol/l, P = 0.011) as was study day post-prandial blood glucose at 90 min (8.4 0.5 mmol/l versus 9.2 0.6 mmol/l, P = 0.046). No significant differences were found in fasting lipid profile, apolipoproteins, fibrinogen, plasminogen activator inhibitor-1, E-selectin, or homocysteine between the two study periods. Insulin aspart resulted in improved post-prandial glycaemic control when compared to human insulin in Type 2 diabetic patients, but this was not associated with changes in markers of vascular risk. # 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Type 2 diabetes; Insulin aspart; Post-prandial hyperglycaemia; Lipids
$ Data from this manuscript has been presented at the following meetings: the 36th Annual Meeting of the EASD, Jerusalem, Israel, September 2000; the 37th Annual Meeting of the EASD, Glasgow, UK, September 2001; the 38th Annual Meeting of the EASD, Budapest, Hungary, September 2002. Abbreviations: PAI-1, plasminogen activator inhibitor-1; NPH, neutral protamine hagedorn * Corresponding author. Tel.: +44 191 222 7019; fax: +44 191 222 0723. E-mail address: [email protected] (A. Gallagher).
0168-8227/$ – see front matter # 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2004.07.010
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
197
1. Introduction
Table 1 Clinical characteristics of the Type 2 diabetic patients studied
Type 2 diabetes is associated with a wide variety of metabolic and haemostatic abnormalities. These include dyslipidaemia [1], increased fibrinogen [2,3], plasminogen activator inhibitor-1 (PAI-1) [4,5], homocysteine [6] and E-selectin levels [7,8]. These abnormalities may contribute to the adverse risk of arterial disease in people with Type 2 diabetes. Recently attention has focused on the importance of post-prandial hyperglycaemia since the toxic effect of high glucose concentrations will be greatest at this time. Epidemiological studies suggest that postprandial hyperglycaemia [9] and post-glucose challenge hyperglycaemia [10,11] are more predictive of adverse cardiovascular outcome than fasting glucose levels in Type 2 diabetes. Rapid-acting insulin analogues reduce post-prandial blood glucose levels in Type 1 [12–15] and Type 2 diabetes [16–20]. In insulin aspart, the amino acid aspartate has been substituted for proline at position B28 of the insulin molecule. This promotes dissolution from hexamers in the insulin vial to monomers, at the insulin concentrations found in the subcutaneous site after injection, giving more rapid absorption [21–23]. The aim of the present study was to examine whether the restoration of a more physiological post-prandial insulin profile with insulin aspart would have any secondary effect on the metabolic and haemostatic abnormalities observed in people with Type 2 diabetes.
Patients (n) Sex (M/F) Age (years) BMI (kg/m2) Serum C-peptide (nmol/l) HbA1c (%) Random serum cholesterol (mmol/l) Random serum triglycerides (mmol/l) Duration of diabetes (years) Hypertension (treated) Dyslipidaemia (treated) Ischaemic heart disease
2. Materials and methods 2.1. Subjects Twenty-one insulin-treated people with Type 2 diabetes were recruited. Patients’ clinical characteristics are given in Table 1. All patients gave written informed consent to participate in the study, which was carried out according to the principles of the Declaration of Helsinki and was approved by the local Ethics Committee. 2.2. Methods A single centre, randomized, double-blind, crossover design was used. Patients were converted from
21 16/5 66 30 1.19 7.8 4.8 2.4 11 13 (62) 12 (57) 12 (57)
5 2 0.46 0.6 0.7 0.7 4
Data are mean S.D., or n (%); HbA1c DCCT aligned, normal < 6.1%.
their usual twice-daily insulin regimen to a pre-meal plus basal regimen using unmodified human insulin (Human Actrapid, Novo Nordisk, Bagsvaerd, Denmark) before each main meal and human NPH insulin (Insulatard, Novo Nordisk) before bed for a 4–6 week run-in period. Patients were then randomized by third party contact at a remote site to either human soluble insulin or insulin aspart (NovoRapid, Novo Nordisk) before main meals, both with bedtime NPH, for a 6week treatment period. At the end of each treatment period, blood was taken for biochemical markers and a test meal study was performed. Patients were then crossed-over to the second treatment arm, with the other pre-meal insulin when an identical study day took place. Patients maintained a minimum of weekly contact with the investigator, when insulin doses were adjusted, and were asked to carry out one seven-point glucose profile per week. Target blood glucose levels were set as 4.0–6.0 mmol/l pre-prandially and 5.0–7.5 mmol/l post-prandially in the absence of hypoglycaemia. Patients were instructed to inject pre-meal insulin subcutaneously 5 min before meals. This timing is optimal for insulin analogue administration and also reflects the injection timing used by the majority of people injecting unmodified human insulin [24–26]. An extra daytime dose of NPH was advised at the discretion of the investigator. Concomitant medication was continued throughout the study. At randomization and at the end of each 6-week treatment period, pre-breakfast serum or plasma levels of the following metabolic and biochemical variables were measured: HbA1c, total cholesterol,
198
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
HDL cholesterol, triglycerides, apolipoprotein A1, apolipoprotein B, fibrinogen, plasminogen activator inhibitor-1 (PAI-1), E-selectin and homocysteine. 2.3. Study day protocol Patients arrived at the Investigation Unit at 08:00 h having fasted overnight. At 08:30 h patients were given breakfast after taking their usual dose of premeal insulin. Only patients with a blood glucose concentration between 4.0–12.0 mmol/l at 11:00 h and whose blood glucose concentration at 11:30 h was less than 2.5 mmol/l different from the first study day, completed the study day. At 11:55 h, patients were given their usual lunchtime insulin dose and lunch was given at 12:00 h. The choice of food and quantity consumed was recorded and identical meals for breakfast and lunch were provided on the second study day. At 11:30 h, 12:00 h and at 30 min intervals up to and including 18:00 h, blood samples were obtained for blood glucose, plasma-free insulin, blood 3hydroxybutyrate, plasma total cholesterol, HDL cholesterol and triglyceride levels. In the event of blood glucose concentrations falling to <3.5 mmol/l, blood sampling frequency was increased. Treatment for hypoglycaemia was initiated if patients developed symptoms of hypoglycaemia or if blood glucose concentration fell below 3.0 mmol/l. Hypoglycaemia was managed with a 20 g carbohydrate snack, which was repeated if symptoms did not abate within 10 min.
Basingstoke, UK). Fibrinogen concentration was determined using an automated analyser (ACL 3000 Coagulometer, Instrumentation Laboratory, Lexington, MA, USA). PAI-1 and E-selectin levels were measured by enzyme-linked immunoassay kits (Innogenetics, Ghent, Belgium and R & D Systems, Minneapolis, MN, USA, respectively). Plasma homocysteine was measured by a fluorescence polarization immunoassay (Abbott, Maidenhead, UK). DCCT-aligned HbA1c (non-diabetic < 6.1%) was determined by HPLC (Eurogenetics, Hampton, UK). 2.5. Statistical analysis Statistical analysis was by GraphPad Instat (San Diego, CA, USA). Skewed data were normalized by logarithmic transformation. Statistical comparison was by paired Student’s t-test. Results are expressed as mean S.E. for normally distributed data.
3. Results 3.1. Patient disposition Twenty-four patients with insulin-treated Type 2 diabetes were recruited. Three patients were withdrawn from the study and thus 21 patients were studied. One patient was withdrawn at their own request and a further two patients were withdrawn by the investigator, since they were elderly, and infirm and it was felt that they would have difficulty completing the study.
2.4. Biochemical analysis
3.2. Insulin doses
Blood glucose was measured using a glucose oxidase method (Yellow Springs Instruments Glucose Analyzer, Yellow Springs, OH, USA). Plasma free insulin was measured as previously described using a double antibody radioimmunoassay (Pharmacia, Uppsala, Sweden) [27]. Serum C-peptide was measured by an enzyme-linked immunosorbent assay (Dako, Ely, UK). 3-Hydroxybutyrate was measured as previously described [28]. Standard laboratory methods were used to measure serum total cholesterol, HDL cholesterol, and triglyceride concentrations. Serum apolipoprotein A1 and B concentrations were estimated using a Technicon DPA-1 analyser (Technicon,
The total dose of pre-meal insulin taken per day was lower with insulin aspart than human insulin (45 4.5 U/day versus 50 5 U/day, P = 0.030). The dose of NPH insulin given did not change between the two periods (21 2 U/day versus 21 2 U/day). Two patients required a second dose of NPH at lunchtime, one in the insulin aspart arm of the study and the other in both treatment periods. 3.3. Overall blood glucose control Glycated haemoglobin fell during the study (randomization versus insulin aspart versus human
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
insulin: 7.60 0.12% versus 7.04 0.13% versus 7.15 0.11%, both P < 0.001). The difference in HbA1c between insulin aspart and human insulin at the end of the 6-week treatment periods did not reach statistical significance (P = 0.060). There was no order effect comparing the end of the first and second treatment periods. 3.4. Self-monitored blood glucose concentrations Self-monitored pre-breakfast blood glucose concentrations did not differ between the two pre-meal insulins (Fig. 1). Mean post-prandial blood glucose concentration after the three main meals of the day was lower with insulin aspart than with human insulin (7.9 0.4 mmol/l versus 9.3 0.4 mmol/l, P = 0.011). Although post-prandial concentrations 90 min after breakfast were not significantly lower with insulin aspart (9.2 0.6 mmol/l versus 10.2 0.6 mmol/l, NS), glucose levels before and after lunch were lower with insulin aspart compared to human insulin (5.2 0.3 mmol/l versus 7.0 0.6 mmol/l, P = 0.007; 7.1 0.5 mmol/l versus 9.0 0.6 mmol/l, P = 0.006). Blood glucose concentrations around dinner were not statistically significantly different. 3.5. Study day results 3.5.1. Test meal blood glucose concentrations Following breakfast at 08:30 h concentrations were lower before lunch at 12:00 h with insulin aspart
Fig. 1. Self-monitored blood glucose profiles in Type 2 diabetic patients when using insulin aspart (&) or unmodified human insulin (*). Mean S.E.
199
compared to human soluble insulin (6.7 0.4 mmol/l versus 7.4 0.4 mmol/l, P = 0.033) (Fig. 2). Mean blood glucose concentrations 90 min after lunch were lower with insulin aspart than with human soluble insulin (8.4 0.5 mmol/l versus 9.2 0.6 mmol/l, P = 0.046) although there was no difference in prandial blood glucose increment. The area under the blood glucose concentration curve in the early post-prandial period (12:00–15:00 h) was lower with pre-prandial insulin aspart than human insulin (22.2 1.1 mmol/ l h versus 24.0 1.3 mmol/l h, P = 0.038). Blood glucose concentrations fell more rapidly with insulin aspart to equal levels below 6.0 mmol/l at 15:30 h. By 18:00 h, blood glucose concentrations were higher with insulin aspart than with human soluble insulin (5.4 0.2 mmol/l versus 4.8 0.2 mmol/l, P = 0.012). 3.5.2. Test meal plasma free insulin concentrations Before lunch at 11:30 h plasma free insulin concentrations were comparable between insulin aspart and human insulin (Fig. 2). Following insulin aspart, plasma free insulin concentrations peaked earlier (60 min versus 90 min) and declined more rapidly than after human insulin. At 60 min, insulin aspart levels were significantly higher than the levels of human insulin (75 7 mU/l versus 61 6 mU/l, P = 0.008). 3.5.3. Test meal 3-hydroxybutyrate concentrations Concentrations of blood 3-hydroxybutyrate were significantly lower with insulin aspart than with unmodified human insulin in the early post-prandial period (at 60 min, 25 3 mmol/l versus 31 3 mmol/ l, P = 0.037) (Fig. 2). However, between 16:30 and 18:00 h, concentrations were lower on unmodified human insulin (at 18:00 h, 83 18 mmol/l versus 58 11 mmol/l, P = 0.030). 3.5.4. Test meal plasma lipid concentrations Plasma total cholesterol levels and HDL cholesterol levels did not change significantly from baseline following consumption of the test meal, and there was no difference in levels with insulin aspart compared to unmodified human insulin (Fig. 2). Triglyceride levels increased significantly over the post-prandial period with insulin aspart (basal 2.2 0.2 mmol/l to peak 2.9 0.2 mmol/l at 15:30 h, P < 0.0001) (Fig. 2). With
200
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
Fig. 2. Blood glucose, plasma free insulin and plasma 3-hydroxybutyrate, plasma total cholesterol, HDL cholesterol and triglyceride concentrations following a midday test meal in Type 2 diabetic patients after a subcutaneous injection of insulin aspart (&) and unmodified human insulin (*). Mean S.E.
unmodified human insulin, changes were identical in extent but to an earlier peak time (14:30 h). 3.6. Biochemical outcome measures All measurements of biochemical outcome are shown in Table 2. There were no significant differ-
ences in pre-breakfast serum lipids, apolipoproteins, fibrinogen or PAI-1 following treatment with human insulin or insulin aspart. Concentrations of lipids, apolipoproteins, fibrinogen and PAI-1 did not change significantly between randomization and at the end of each treatment period on either insulin, in spite of the significant improvement in HbA1c.
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
201
Table 2 A comparison of markers of vascular risk following treatment with meal-time insulin aspart or human insulin, both with basal NPH insulin, in people with Type 2 diabetes Parameter (fasting)
Randomization
Total cholesterol (mmol/l) HDL cholesterol (mmol/l) Triglycerides (mmol/l) Fibrinogen (g/l) Apolipoprotein A1 (g/l) Apolipoprotein B (g/l) PAI-1 (ng/ml) Homocysteine (mmol/l) E-selectin (mg/l)
4.8 1.1 1.8 3.8 1.4 1.0 83.5 14.0 75.8
0.2 0.1 0.1 0.3 0.1 0.1 4.0 1.0 5.1
Insulin aspart 4.9 1.1 2.0 4.2 1.4 1.1 82.9 9.5 77.7
0.2 0.1 0.2 0.3 0.1 0.1 5.0 0.4 7.8
Human insulin 4.9 1.2 2.4 4.4 1.4 1.1 82.9 9.7 80.0
0.2 0.1 0.5 0.3 0.1 0.1 4.3 0.4 7.3
Mean S.E.; n = 21, except at randomization for PAI-1 n = 20, homocysteine n = 18, E-selectin n = 19. Normal ranges: HDL cholesterol 1.0– 1.5 mmol/l in males, 1.2–1.8 mmol/l in females, triglycerides 0.5–1.8 mmol/l, fibrinogen 2.0–4.4 g/l, apolipoprotein A1 1.0–1.9 g/l, apolipoprotein B 0.8–1.6 g/l, PAI-1 11–69 ng/ml, homocysteine 5–20 mmol/l, E-selectin 29–63 mg/l.
There was no difference in levels of homocysteine (9.5 0.4 mmol/l versus 9.7 0.4 mmol/l, NS) or Eselectin (77.7 7.8 mg/l versus 80.0 7.3 mg/l, NS) between insulin aspart and human insulin.
4. Discussion The aim of this study was to assess the effect of amelioration of post-prandial blood glucose concentrations, normally the highest and thus most toxic levels of glucose during the day, on a wide variety of markers of the vascular risk that causes most of the morbidity and mortality of Type 2 diabetes. To do this, we attempted to improve control of post-prandial blood glucose with the rapid-acting insulin analogue, insulin aspart. This might also be expected to have some impact on overall blood glucose control as measured by HbA1c, but evidence from earlier studies is that without attention to basal insulin replacement, higher basal blood glucose levels to some extent balance out the change in HbA1c expected from reduced post-prandial glucose excursions [12,16,17]. While previous studies with rapid-acting insulin analogues in people with Type 2 diabetes have tended to use twice daily injections, in the present study, the intention of controlling all post-prandial glucose excursions was better met by using a pre-meal plus basal multiple insulin regimen, to ensure control of lunchtime blood glucose excursions. This was con-
firmed in the test meal study, which suggests reductions of around 0.7 mmol/l averaged across the immediate post-prandial period (Fig. 2). One of the factors limiting the achievement of optimal post-prandial blood glucose levels in the target range was the double-blind nature of this study. There are quite marked differences in the pharmacokinetics of human insulin and insulin aspart [21] and the achievement of optimal post-prandial glycaemic control was not always possible because of the risk of hypoglycaemia, particularly late post-prandially with unmodified human insulin. A possible criticism of this study is that the range of baseline blood glucose concentrations on each study day was too wide. Ideally, an intravenous insulin infusion would have been used to control blood glucose concentrations overnight prior to the test meal. Despite the improvements in post-prandial blood glucose control, the study did not demonstrate any changes in fasting lipid or apolipoprotein levels (Table 2). While aspects of blood lipid metabolism are known to improve with improved blood glucose levels, it is now generally accepted that this occurs only where more marked hyperglycaemia is ameliorated, and not from the relatively good HbA1c levels reported in the present study [29]. No changes in haemostatic markers or measures of endothelial dysfunction in these Type 2 diabetic patients with improved post-prandial blood glucose control were demonstrated. Although this intensive
202
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
study was statistically relatively small, the range of markers studied, and the lack of any suggestion of trend, mitigates against the argument that the lack of demonstrated differences is simply due to inadequate statistical power. It does, however, remain possible that there may have been a type 2 statistical error or that there was not sufficient improvement in postprandial hyperglycaemia to cause significant change. An increased sample size and longer duration of study would reduce the chance of obtaining a type 2 error. The use of surrogate markers, however, does not necessarily lead to unequivocal answers to clinical questions. A large, long-term study looking at clinically useful patient orientated endpoints such as myocardial infarction may be better at answering the question posed by this study but this is beyond the scope of this research. It remains possible that greater normalization of blood glucose levels, and in particular of the higher post-prandial levels, would exert a more marked and thus more detectable influence on these vascular markers, and that that might be particularly important for a group of patients with overall metabolic control poorer than the typical values found with modern management approaches as demonstrated in our study cohort. However, it will be noted that the typical abnormalities of HDL cholesterol and triglyceride concentrations are found even in people whose HbA1c has been normalized [30], and that part of the increased vascular risk associated with diabetes is found with sub-diabetic levels of glucose intolerance, notably in those with IGT [11]. Furthermore, small improvements in blood glucose control achieved with insulin lispro, in a patient cohort with very much worse overall control, also failed to improve the serum lipid profile [17]. In summary, improving post-prandial but not overall blood glucose control with insulin aspart failed to provide any evidence of improvement in the basic features of the lipoprotein profile or of levels of vascular risk markers in a well-controlled group of people with Type 2 diabetes already taking insulin therapy. This suggests that the abnormalities found in many of these variables are more related to the basic underlying pathological disturbance than to the effects of hyperglycaemia.
Acknowledgements This study was supported by a research grant to PDH from Novo Nordisk. PDH has advised Novo Nordisk on behalf of the University of Newcastle upon Tyne during the development programme of insulin aspart, and the University has been paid for teaching activities performed by PDH for the company.
References [1] S.M. Haffner, R.A. Valdez, H.P. Hazuda, B.D. Mitchell, P.A. Morales, M.P. Stern, Prospective analysis of the insulin-resistance syndrome (syndrome X), Diabetes 41 (1992) 715– 722. [2] W.B. Kannel, R.B. D’Agostino, P.W. Wilson, A.J. Belanger, D.R. Gagnon, Diabetes, fibrinogen, and risk of cardiovascular disease: the Framingham experience, Am. Heart J. 120 (1990) 672–676. [3] O.P. Ganda, C.F. Arkin, Hyperfibrinogenemia. An important risk factor for vascular complications in diabetes, Diabetes Care 15 (1992) 1245–1250. [4] J. Auwerx, R. Bouillon, D. Collen, J. Geboers, Tissue-type plasminogen activator antigen and plasminogen activator inhibitor in diabetes mellitus, Arteriosclerosis 8 (1988) 68–72. [5] I. Juhan-Vague, C. Roul, M.C. Alessi, J.P. Ardissone, M. Heim, P. Vague, Increased plasminogen activator inhibitor activity in non insulin dependent diabetic patients—relationship with plasma insulin, Thromb Haemost 61 (1989) 370–373. [6] E.K. Hoogeveen, P.J. Kostense, P.J. Jeks, et al. Hyperhomocysteinemia is associated with an increased risk of cardiovascular disease, especially in non-insulin-dependent diabetes mellitus: a population-based study, Arterioscler. Thromb. Vasc. Biol. 18 (1998) 133–138. [7] M. Steiner, K.M. Reinhardt, B. Krammer, B. Ernst, A.D. Blann, Increased levels of soluble adhesion molecules in Type 2 (non-insulin dependent) diabetes mellitus are independent of glycaemic control, Thromb Haemost 72 (1994) 979–984. [8] L. Cominacini, A. Fratta Pasini, U. Garbin, et al. Elevated levels of soluble E-selectin in patients with IDDM and NIDDM: relation to metabolic control, Diabetologia 38 (1995) 1122–1124. [9] M. Hanefeld, S. Fischer, U. Julius, et al. Risk factors for myocardial infarction and death in newly detected NIDDM: the Diabetes Intervention Study, 11-year follow-up, Diabetologia 39 (1996) 1577–1583. [10] B. Balkau, E. Eschwege, L. Papoz, et al. Risk factors for early death in non insulin dependent diabetes and men with known glucose tolerance status, Br. Med. J. 307 (1993) 295–299. [11] Anonymous, Glucose tolerance and mortality: comparison of WHO and American Diabetes Association diagnostic criteria. The DECODE study group, Lancet 354 (1999) 617–621. [12] J.H. Anderson, R.L. Brunelle, V.A. Koivisto, et al. Reduction of post prandial hyperglycemia and frequency of
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
hypoglycemia in IDDM patients on insulin-analog treatment, Diabetes 46 (1997) 265–270. P.D. Home, A. Lindholm, B. Hylleberg, P. Round, Improved glycemic control with insulin aspart: a multicenter randomized double-blind crossover trial in Type 1 diabetic patients, Diabetes Care 21 (1998) 1904–1909. P. Raskin, R.A. Guthrie, L. Leiter, A. Riis, L. Jovanovic, Use of insulin aspart, a fast-acting insulin analog, as the mealtime insulin in the management of patients with Type 1 diabetes, Diabetes Care 23 (2000) 583–588. A. Lindholm, J. McEwen, A.P. Riis, Improved post-prandial glycaemic control with insulin aspart. A randomized doubleblind cross-over trial in Type 1 diabetes, Diabetes Care 22 (1999) 801–805. J.H. Anderson, R.L. Brunelle, V.A. Koivisto, M.E. Trautmann, L. Vignati, R. DiMarchi, Improved mealtime treatment of diabetes mellitus using an insulin analogue, Clin. Ther. 19 (1997) 62–72. J.H. Anderson, R.L. Brunelle, P. Keohane, V.A. Koivisto, M.E. Trautmann, L. Vignati, Mealtime treatment with insulin analog improves postprandial hyperglycemia and hypoglycemia in patients with non-insulin-dependent diabetes mellitus, Arch. Intern. Med. 157 (1997) 1249–1255. M.N. Feinglos, C.H. Thacker, J. English, M.A. Bethel, J.D. Lane, Modification of postprandial hyperglycemia with insulin lispro improves glucose control in patients with Type 2 diabetes, Diabetes Care 20 (1997) 1539–1542. A.M. Rosenfalck, P. Thorsby, L. Kjems, et al. Improved postprandial glycaemic control with insulin aspart in Type 2 diabetic patients treated with insulin, Acta Diabetol. 37 (2000) 41–46. B.O. Boehm, P.D. Home, C. Behrend, N.M. Kamp, A. Lindholm, Premixed insulin aspart 30 versus premixed human insulin 30/70 twice daily: a randomized trial in Type 1 and Type 2 diabetic patients, Diabet. Med. 19 (2002) 393–399. P.D. Home, L. Barriocanal, A. Lindholm, Comparative pharmacokinetics and pharmacodynamics of the novel rapid-acting insulin analogue, insulin aspart, in healthy volunteers, Eur. J. Clin. Pharmacol. 55 (1999) 199–203. S.R. Mudaliar, F.A. Lindberg, M. Joyce, et al. Insulin aspart (B28 asp-insulin): a fast-acting analog of human insulin: absorption kinetics and action profile compared with regular human insulin in healthy non diabetic subjects, Diabetes Care 22 (1999) 1501–1506.
203
[23] L. Heinemann, T. Heise, L.N. Jorgensen, A.A. Starke, Action profile of the rapid acting insulin analogue: human insulin B28Asp, Diabetic Med. 10 (1993) 535–539. [24] M. Berger, L. Heinemann, Are presently available insulin analogues clinically beneficial? Diabetologia 40 (Suppl. 2) (1997) S91–S96. [25] L. Heinemann, A.A.R. Starke, A. Hohmann, M. Berger, Timing between the subcutaneous administration of insulin and consumption of a carbohydrate rich meal, Horm. Metab. Res. 10 (Suppl. 26) (1992) 137–139. [26] L. Heinemann, Do insulin-treated diabetic patients use an injection-meal-interval in daily life? Diabetic Med. 12 (1995) 449–450. [27] I. Hanning, P.D. Home, K.G. Alberti, Measurement of free insulin concentrations: the influence of the timing of extraction of insulin antibodies, Diabetologia 28 (1985) 831–835. [28] J. Harrison, A.W. Hodson, A.W. Skillen, R. Stappenbeck, L. Agius, K.G.M.M. Alberti, Blood glucose, lactate, pyruvate, glycerol, 3-hydroxybutyrate, and acetoacetate measurements in man using a centrifugal analyser with fluorimetric attachment, J. Chem. Clin. Biochem. 26 (1988) 141–146. [29] M.R. Taskinen, Hyperlipidaemia in diabetes, in: D.J. Betteridge (Ed.), Clinical Endocrinology and Metabolism, vol. 3, Balliere Tindall, London, 1990, pp. 743–775. [30] J. Joven, E. Vilella, B. Costa, P.R. Turner, C. Richart, L. Masana, Concentrations of lipids and apolipoproteins in patients with clinically well-controlled insulin-dependent and non-insulin-dependent diabetes, Clin. Chem. 35 (1989) 813–816.
Further reading [1] A. Gallagher, P.D. Home, The effect of insulin aspart on metabolic outcome and albumin excretion in Type 2 diabetes, Diabetologia 43 (Suppl. 1) (2000) A200. [2] A. Gallagher, P.D. Home, The effect of insulin aspart on postprandial metabolism and metabolic outcome in Type 2 diabetes, Diabetologia 44 (Suppl. 1) (2001) A209. [3] A. Gallagher, P.D. Home, The effect of insulin aspart on postprandial metabolism and lipid profile in Type 2 diabetes, Diabetologia 45 (Suppl. 2) (2002) A253.
The effect of improved post-prandial blood glucose control on post-prandial metabolism and markers of vascular risk in people with Type 2 diabetes$ A. Gallagher*, P.D. Home School of Clinical Medical Sciences, Floor 4, William Leech Building, The Medical School, Framlington Place, Newcastle upon Tyne NE2 4HH, UK Received 21 January 2004; received in revised form 30 June 2004; accepted 7 July 2004
Abstract A variety of abnormalities of metabolic, haemostatic and endothelial markers are associated with Type 2 diabetes. Evidence suggests that poor post-prandial blood glucose control may contribute to vascular risk. We aimed to examine whether the restoration of a more physiological insulin profile post-prandially would improve these abnormalities. Twenty-one patients with insulin-treated Type 2 diabetes were recruited into a single centre, crossover, double-blind study. Patients were randomized to unmodified human insulin or insulin aspart before main meals for 6-week study periods, both together with NPH insulin. At the end of each study period, pre-breakfast levels of markers of vascular risk were assessed and a test meal performed. There was no significant difference in HbA1c (7.04 0.13% (S.E.) versus 7.15 0.11%, P = 0.060) with insulin aspart compared to human insulin at the end of each study period. The mean post-prandial blood glucose concentration at 90 min from self-monitored results was lower with insulin aspart than with human insulin (7.9 0.4 mmol/l versus 9.3 0.4 mmol/l, P = 0.011) as was study day post-prandial blood glucose at 90 min (8.4 0.5 mmol/l versus 9.2 0.6 mmol/l, P = 0.046). No significant differences were found in fasting lipid profile, apolipoproteins, fibrinogen, plasminogen activator inhibitor-1, E-selectin, or homocysteine between the two study periods. Insulin aspart resulted in improved post-prandial glycaemic control when compared to human insulin in Type 2 diabetic patients, but this was not associated with changes in markers of vascular risk. # 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Type 2 diabetes; Insulin aspart; Post-prandial hyperglycaemia; Lipids
$ Data from this manuscript has been presented at the following meetings: the 36th Annual Meeting of the EASD, Jerusalem, Israel, September 2000; the 37th Annual Meeting of the EASD, Glasgow, UK, September 2001; the 38th Annual Meeting of the EASD, Budapest, Hungary, September 2002. Abbreviations: PAI-1, plasminogen activator inhibitor-1; NPH, neutral protamine hagedorn * Corresponding author. Tel.: +44 191 222 7019; fax: +44 191 222 0723. E-mail address: [email protected] (A. Gallagher).
0168-8227/$ – see front matter # 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2004.07.010
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
197
1. Introduction
Table 1 Clinical characteristics of the Type 2 diabetic patients studied
Type 2 diabetes is associated with a wide variety of metabolic and haemostatic abnormalities. These include dyslipidaemia [1], increased fibrinogen [2,3], plasminogen activator inhibitor-1 (PAI-1) [4,5], homocysteine [6] and E-selectin levels [7,8]. These abnormalities may contribute to the adverse risk of arterial disease in people with Type 2 diabetes. Recently attention has focused on the importance of post-prandial hyperglycaemia since the toxic effect of high glucose concentrations will be greatest at this time. Epidemiological studies suggest that postprandial hyperglycaemia [9] and post-glucose challenge hyperglycaemia [10,11] are more predictive of adverse cardiovascular outcome than fasting glucose levels in Type 2 diabetes. Rapid-acting insulin analogues reduce post-prandial blood glucose levels in Type 1 [12–15] and Type 2 diabetes [16–20]. In insulin aspart, the amino acid aspartate has been substituted for proline at position B28 of the insulin molecule. This promotes dissolution from hexamers in the insulin vial to monomers, at the insulin concentrations found in the subcutaneous site after injection, giving more rapid absorption [21–23]. The aim of the present study was to examine whether the restoration of a more physiological post-prandial insulin profile with insulin aspart would have any secondary effect on the metabolic and haemostatic abnormalities observed in people with Type 2 diabetes.
Patients (n) Sex (M/F) Age (years) BMI (kg/m2) Serum C-peptide (nmol/l) HbA1c (%) Random serum cholesterol (mmol/l) Random serum triglycerides (mmol/l) Duration of diabetes (years) Hypertension (treated) Dyslipidaemia (treated) Ischaemic heart disease
2. Materials and methods 2.1. Subjects Twenty-one insulin-treated people with Type 2 diabetes were recruited. Patients’ clinical characteristics are given in Table 1. All patients gave written informed consent to participate in the study, which was carried out according to the principles of the Declaration of Helsinki and was approved by the local Ethics Committee. 2.2. Methods A single centre, randomized, double-blind, crossover design was used. Patients were converted from
21 16/5 66 30 1.19 7.8 4.8 2.4 11 13 (62) 12 (57) 12 (57)
5 2 0.46 0.6 0.7 0.7 4
Data are mean S.D., or n (%); HbA1c DCCT aligned, normal < 6.1%.
their usual twice-daily insulin regimen to a pre-meal plus basal regimen using unmodified human insulin (Human Actrapid, Novo Nordisk, Bagsvaerd, Denmark) before each main meal and human NPH insulin (Insulatard, Novo Nordisk) before bed for a 4–6 week run-in period. Patients were then randomized by third party contact at a remote site to either human soluble insulin or insulin aspart (NovoRapid, Novo Nordisk) before main meals, both with bedtime NPH, for a 6week treatment period. At the end of each treatment period, blood was taken for biochemical markers and a test meal study was performed. Patients were then crossed-over to the second treatment arm, with the other pre-meal insulin when an identical study day took place. Patients maintained a minimum of weekly contact with the investigator, when insulin doses were adjusted, and were asked to carry out one seven-point glucose profile per week. Target blood glucose levels were set as 4.0–6.0 mmol/l pre-prandially and 5.0–7.5 mmol/l post-prandially in the absence of hypoglycaemia. Patients were instructed to inject pre-meal insulin subcutaneously 5 min before meals. This timing is optimal for insulin analogue administration and also reflects the injection timing used by the majority of people injecting unmodified human insulin [24–26]. An extra daytime dose of NPH was advised at the discretion of the investigator. Concomitant medication was continued throughout the study. At randomization and at the end of each 6-week treatment period, pre-breakfast serum or plasma levels of the following metabolic and biochemical variables were measured: HbA1c, total cholesterol,
198
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
HDL cholesterol, triglycerides, apolipoprotein A1, apolipoprotein B, fibrinogen, plasminogen activator inhibitor-1 (PAI-1), E-selectin and homocysteine. 2.3. Study day protocol Patients arrived at the Investigation Unit at 08:00 h having fasted overnight. At 08:30 h patients were given breakfast after taking their usual dose of premeal insulin. Only patients with a blood glucose concentration between 4.0–12.0 mmol/l at 11:00 h and whose blood glucose concentration at 11:30 h was less than 2.5 mmol/l different from the first study day, completed the study day. At 11:55 h, patients were given their usual lunchtime insulin dose and lunch was given at 12:00 h. The choice of food and quantity consumed was recorded and identical meals for breakfast and lunch were provided on the second study day. At 11:30 h, 12:00 h and at 30 min intervals up to and including 18:00 h, blood samples were obtained for blood glucose, plasma-free insulin, blood 3hydroxybutyrate, plasma total cholesterol, HDL cholesterol and triglyceride levels. In the event of blood glucose concentrations falling to <3.5 mmol/l, blood sampling frequency was increased. Treatment for hypoglycaemia was initiated if patients developed symptoms of hypoglycaemia or if blood glucose concentration fell below 3.0 mmol/l. Hypoglycaemia was managed with a 20 g carbohydrate snack, which was repeated if symptoms did not abate within 10 min.
Basingstoke, UK). Fibrinogen concentration was determined using an automated analyser (ACL 3000 Coagulometer, Instrumentation Laboratory, Lexington, MA, USA). PAI-1 and E-selectin levels were measured by enzyme-linked immunoassay kits (Innogenetics, Ghent, Belgium and R & D Systems, Minneapolis, MN, USA, respectively). Plasma homocysteine was measured by a fluorescence polarization immunoassay (Abbott, Maidenhead, UK). DCCT-aligned HbA1c (non-diabetic < 6.1%) was determined by HPLC (Eurogenetics, Hampton, UK). 2.5. Statistical analysis Statistical analysis was by GraphPad Instat (San Diego, CA, USA). Skewed data were normalized by logarithmic transformation. Statistical comparison was by paired Student’s t-test. Results are expressed as mean S.E. for normally distributed data.
3. Results 3.1. Patient disposition Twenty-four patients with insulin-treated Type 2 diabetes were recruited. Three patients were withdrawn from the study and thus 21 patients were studied. One patient was withdrawn at their own request and a further two patients were withdrawn by the investigator, since they were elderly, and infirm and it was felt that they would have difficulty completing the study.
2.4. Biochemical analysis
3.2. Insulin doses
Blood glucose was measured using a glucose oxidase method (Yellow Springs Instruments Glucose Analyzer, Yellow Springs, OH, USA). Plasma free insulin was measured as previously described using a double antibody radioimmunoassay (Pharmacia, Uppsala, Sweden) [27]. Serum C-peptide was measured by an enzyme-linked immunosorbent assay (Dako, Ely, UK). 3-Hydroxybutyrate was measured as previously described [28]. Standard laboratory methods were used to measure serum total cholesterol, HDL cholesterol, and triglyceride concentrations. Serum apolipoprotein A1 and B concentrations were estimated using a Technicon DPA-1 analyser (Technicon,
The total dose of pre-meal insulin taken per day was lower with insulin aspart than human insulin (45 4.5 U/day versus 50 5 U/day, P = 0.030). The dose of NPH insulin given did not change between the two periods (21 2 U/day versus 21 2 U/day). Two patients required a second dose of NPH at lunchtime, one in the insulin aspart arm of the study and the other in both treatment periods. 3.3. Overall blood glucose control Glycated haemoglobin fell during the study (randomization versus insulin aspart versus human
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
insulin: 7.60 0.12% versus 7.04 0.13% versus 7.15 0.11%, both P < 0.001). The difference in HbA1c between insulin aspart and human insulin at the end of the 6-week treatment periods did not reach statistical significance (P = 0.060). There was no order effect comparing the end of the first and second treatment periods. 3.4. Self-monitored blood glucose concentrations Self-monitored pre-breakfast blood glucose concentrations did not differ between the two pre-meal insulins (Fig. 1). Mean post-prandial blood glucose concentration after the three main meals of the day was lower with insulin aspart than with human insulin (7.9 0.4 mmol/l versus 9.3 0.4 mmol/l, P = 0.011). Although post-prandial concentrations 90 min after breakfast were not significantly lower with insulin aspart (9.2 0.6 mmol/l versus 10.2 0.6 mmol/l, NS), glucose levels before and after lunch were lower with insulin aspart compared to human insulin (5.2 0.3 mmol/l versus 7.0 0.6 mmol/l, P = 0.007; 7.1 0.5 mmol/l versus 9.0 0.6 mmol/l, P = 0.006). Blood glucose concentrations around dinner were not statistically significantly different. 3.5. Study day results 3.5.1. Test meal blood glucose concentrations Following breakfast at 08:30 h concentrations were lower before lunch at 12:00 h with insulin aspart
Fig. 1. Self-monitored blood glucose profiles in Type 2 diabetic patients when using insulin aspart (&) or unmodified human insulin (*). Mean S.E.
199
compared to human soluble insulin (6.7 0.4 mmol/l versus 7.4 0.4 mmol/l, P = 0.033) (Fig. 2). Mean blood glucose concentrations 90 min after lunch were lower with insulin aspart than with human soluble insulin (8.4 0.5 mmol/l versus 9.2 0.6 mmol/l, P = 0.046) although there was no difference in prandial blood glucose increment. The area under the blood glucose concentration curve in the early post-prandial period (12:00–15:00 h) was lower with pre-prandial insulin aspart than human insulin (22.2 1.1 mmol/ l h versus 24.0 1.3 mmol/l h, P = 0.038). Blood glucose concentrations fell more rapidly with insulin aspart to equal levels below 6.0 mmol/l at 15:30 h. By 18:00 h, blood glucose concentrations were higher with insulin aspart than with human soluble insulin (5.4 0.2 mmol/l versus 4.8 0.2 mmol/l, P = 0.012). 3.5.2. Test meal plasma free insulin concentrations Before lunch at 11:30 h plasma free insulin concentrations were comparable between insulin aspart and human insulin (Fig. 2). Following insulin aspart, plasma free insulin concentrations peaked earlier (60 min versus 90 min) and declined more rapidly than after human insulin. At 60 min, insulin aspart levels were significantly higher than the levels of human insulin (75 7 mU/l versus 61 6 mU/l, P = 0.008). 3.5.3. Test meal 3-hydroxybutyrate concentrations Concentrations of blood 3-hydroxybutyrate were significantly lower with insulin aspart than with unmodified human insulin in the early post-prandial period (at 60 min, 25 3 mmol/l versus 31 3 mmol/ l, P = 0.037) (Fig. 2). However, between 16:30 and 18:00 h, concentrations were lower on unmodified human insulin (at 18:00 h, 83 18 mmol/l versus 58 11 mmol/l, P = 0.030). 3.5.4. Test meal plasma lipid concentrations Plasma total cholesterol levels and HDL cholesterol levels did not change significantly from baseline following consumption of the test meal, and there was no difference in levels with insulin aspart compared to unmodified human insulin (Fig. 2). Triglyceride levels increased significantly over the post-prandial period with insulin aspart (basal 2.2 0.2 mmol/l to peak 2.9 0.2 mmol/l at 15:30 h, P < 0.0001) (Fig. 2). With
200
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
Fig. 2. Blood glucose, plasma free insulin and plasma 3-hydroxybutyrate, plasma total cholesterol, HDL cholesterol and triglyceride concentrations following a midday test meal in Type 2 diabetic patients after a subcutaneous injection of insulin aspart (&) and unmodified human insulin (*). Mean S.E.
unmodified human insulin, changes were identical in extent but to an earlier peak time (14:30 h). 3.6. Biochemical outcome measures All measurements of biochemical outcome are shown in Table 2. There were no significant differ-
ences in pre-breakfast serum lipids, apolipoproteins, fibrinogen or PAI-1 following treatment with human insulin or insulin aspart. Concentrations of lipids, apolipoproteins, fibrinogen and PAI-1 did not change significantly between randomization and at the end of each treatment period on either insulin, in spite of the significant improvement in HbA1c.
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
201
Table 2 A comparison of markers of vascular risk following treatment with meal-time insulin aspart or human insulin, both with basal NPH insulin, in people with Type 2 diabetes Parameter (fasting)
Randomization
Total cholesterol (mmol/l) HDL cholesterol (mmol/l) Triglycerides (mmol/l) Fibrinogen (g/l) Apolipoprotein A1 (g/l) Apolipoprotein B (g/l) PAI-1 (ng/ml) Homocysteine (mmol/l) E-selectin (mg/l)
4.8 1.1 1.8 3.8 1.4 1.0 83.5 14.0 75.8
0.2 0.1 0.1 0.3 0.1 0.1 4.0 1.0 5.1
Insulin aspart 4.9 1.1 2.0 4.2 1.4 1.1 82.9 9.5 77.7
0.2 0.1 0.2 0.3 0.1 0.1 5.0 0.4 7.8
Human insulin 4.9 1.2 2.4 4.4 1.4 1.1 82.9 9.7 80.0
0.2 0.1 0.5 0.3 0.1 0.1 4.3 0.4 7.3
Mean S.E.; n = 21, except at randomization for PAI-1 n = 20, homocysteine n = 18, E-selectin n = 19. Normal ranges: HDL cholesterol 1.0– 1.5 mmol/l in males, 1.2–1.8 mmol/l in females, triglycerides 0.5–1.8 mmol/l, fibrinogen 2.0–4.4 g/l, apolipoprotein A1 1.0–1.9 g/l, apolipoprotein B 0.8–1.6 g/l, PAI-1 11–69 ng/ml, homocysteine 5–20 mmol/l, E-selectin 29–63 mg/l.
There was no difference in levels of homocysteine (9.5 0.4 mmol/l versus 9.7 0.4 mmol/l, NS) or Eselectin (77.7 7.8 mg/l versus 80.0 7.3 mg/l, NS) between insulin aspart and human insulin.
4. Discussion The aim of this study was to assess the effect of amelioration of post-prandial blood glucose concentrations, normally the highest and thus most toxic levels of glucose during the day, on a wide variety of markers of the vascular risk that causes most of the morbidity and mortality of Type 2 diabetes. To do this, we attempted to improve control of post-prandial blood glucose with the rapid-acting insulin analogue, insulin aspart. This might also be expected to have some impact on overall blood glucose control as measured by HbA1c, but evidence from earlier studies is that without attention to basal insulin replacement, higher basal blood glucose levels to some extent balance out the change in HbA1c expected from reduced post-prandial glucose excursions [12,16,17]. While previous studies with rapid-acting insulin analogues in people with Type 2 diabetes have tended to use twice daily injections, in the present study, the intention of controlling all post-prandial glucose excursions was better met by using a pre-meal plus basal multiple insulin regimen, to ensure control of lunchtime blood glucose excursions. This was con-
firmed in the test meal study, which suggests reductions of around 0.7 mmol/l averaged across the immediate post-prandial period (Fig. 2). One of the factors limiting the achievement of optimal post-prandial blood glucose levels in the target range was the double-blind nature of this study. There are quite marked differences in the pharmacokinetics of human insulin and insulin aspart [21] and the achievement of optimal post-prandial glycaemic control was not always possible because of the risk of hypoglycaemia, particularly late post-prandially with unmodified human insulin. A possible criticism of this study is that the range of baseline blood glucose concentrations on each study day was too wide. Ideally, an intravenous insulin infusion would have been used to control blood glucose concentrations overnight prior to the test meal. Despite the improvements in post-prandial blood glucose control, the study did not demonstrate any changes in fasting lipid or apolipoprotein levels (Table 2). While aspects of blood lipid metabolism are known to improve with improved blood glucose levels, it is now generally accepted that this occurs only where more marked hyperglycaemia is ameliorated, and not from the relatively good HbA1c levels reported in the present study [29]. No changes in haemostatic markers or measures of endothelial dysfunction in these Type 2 diabetic patients with improved post-prandial blood glucose control were demonstrated. Although this intensive
202
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
study was statistically relatively small, the range of markers studied, and the lack of any suggestion of trend, mitigates against the argument that the lack of demonstrated differences is simply due to inadequate statistical power. It does, however, remain possible that there may have been a type 2 statistical error or that there was not sufficient improvement in postprandial hyperglycaemia to cause significant change. An increased sample size and longer duration of study would reduce the chance of obtaining a type 2 error. The use of surrogate markers, however, does not necessarily lead to unequivocal answers to clinical questions. A large, long-term study looking at clinically useful patient orientated endpoints such as myocardial infarction may be better at answering the question posed by this study but this is beyond the scope of this research. It remains possible that greater normalization of blood glucose levels, and in particular of the higher post-prandial levels, would exert a more marked and thus more detectable influence on these vascular markers, and that that might be particularly important for a group of patients with overall metabolic control poorer than the typical values found with modern management approaches as demonstrated in our study cohort. However, it will be noted that the typical abnormalities of HDL cholesterol and triglyceride concentrations are found even in people whose HbA1c has been normalized [30], and that part of the increased vascular risk associated with diabetes is found with sub-diabetic levels of glucose intolerance, notably in those with IGT [11]. Furthermore, small improvements in blood glucose control achieved with insulin lispro, in a patient cohort with very much worse overall control, also failed to improve the serum lipid profile [17]. In summary, improving post-prandial but not overall blood glucose control with insulin aspart failed to provide any evidence of improvement in the basic features of the lipoprotein profile or of levels of vascular risk markers in a well-controlled group of people with Type 2 diabetes already taking insulin therapy. This suggests that the abnormalities found in many of these variables are more related to the basic underlying pathological disturbance than to the effects of hyperglycaemia.
Acknowledgements This study was supported by a research grant to PDH from Novo Nordisk. PDH has advised Novo Nordisk on behalf of the University of Newcastle upon Tyne during the development programme of insulin aspart, and the University has been paid for teaching activities performed by PDH for the company.
References [1] S.M. Haffner, R.A. Valdez, H.P. Hazuda, B.D. Mitchell, P.A. Morales, M.P. Stern, Prospective analysis of the insulin-resistance syndrome (syndrome X), Diabetes 41 (1992) 715– 722. [2] W.B. Kannel, R.B. D’Agostino, P.W. Wilson, A.J. Belanger, D.R. Gagnon, Diabetes, fibrinogen, and risk of cardiovascular disease: the Framingham experience, Am. Heart J. 120 (1990) 672–676. [3] O.P. Ganda, C.F. Arkin, Hyperfibrinogenemia. An important risk factor for vascular complications in diabetes, Diabetes Care 15 (1992) 1245–1250. [4] J. Auwerx, R. Bouillon, D. Collen, J. Geboers, Tissue-type plasminogen activator antigen and plasminogen activator inhibitor in diabetes mellitus, Arteriosclerosis 8 (1988) 68–72. [5] I. Juhan-Vague, C. Roul, M.C. Alessi, J.P. Ardissone, M. Heim, P. Vague, Increased plasminogen activator inhibitor activity in non insulin dependent diabetic patients—relationship with plasma insulin, Thromb Haemost 61 (1989) 370–373. [6] E.K. Hoogeveen, P.J. Kostense, P.J. Jeks, et al. Hyperhomocysteinemia is associated with an increased risk of cardiovascular disease, especially in non-insulin-dependent diabetes mellitus: a population-based study, Arterioscler. Thromb. Vasc. Biol. 18 (1998) 133–138. [7] M. Steiner, K.M. Reinhardt, B. Krammer, B. Ernst, A.D. Blann, Increased levels of soluble adhesion molecules in Type 2 (non-insulin dependent) diabetes mellitus are independent of glycaemic control, Thromb Haemost 72 (1994) 979–984. [8] L. Cominacini, A. Fratta Pasini, U. Garbin, et al. Elevated levels of soluble E-selectin in patients with IDDM and NIDDM: relation to metabolic control, Diabetologia 38 (1995) 1122–1124. [9] M. Hanefeld, S. Fischer, U. Julius, et al. Risk factors for myocardial infarction and death in newly detected NIDDM: the Diabetes Intervention Study, 11-year follow-up, Diabetologia 39 (1996) 1577–1583. [10] B. Balkau, E. Eschwege, L. Papoz, et al. Risk factors for early death in non insulin dependent diabetes and men with known glucose tolerance status, Br. Med. J. 307 (1993) 295–299. [11] Anonymous, Glucose tolerance and mortality: comparison of WHO and American Diabetes Association diagnostic criteria. The DECODE study group, Lancet 354 (1999) 617–621. [12] J.H. Anderson, R.L. Brunelle, V.A. Koivisto, et al. Reduction of post prandial hyperglycemia and frequency of
A. Gallagher, P.D. Home / Diabetes Research and Clinical Practice 67 (2005) 196–203
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
hypoglycemia in IDDM patients on insulin-analog treatment, Diabetes 46 (1997) 265–270. P.D. Home, A. Lindholm, B. Hylleberg, P. Round, Improved glycemic control with insulin aspart: a multicenter randomized double-blind crossover trial in Type 1 diabetic patients, Diabetes Care 21 (1998) 1904–1909. P. Raskin, R.A. Guthrie, L. Leiter, A. Riis, L. Jovanovic, Use of insulin aspart, a fast-acting insulin analog, as the mealtime insulin in the management of patients with Type 1 diabetes, Diabetes Care 23 (2000) 583–588. A. Lindholm, J. McEwen, A.P. Riis, Improved post-prandial glycaemic control with insulin aspart. A randomized doubleblind cross-over trial in Type 1 diabetes, Diabetes Care 22 (1999) 801–805. J.H. Anderson, R.L. Brunelle, V.A. Koivisto, M.E. Trautmann, L. Vignati, R. DiMarchi, Improved mealtime treatment of diabetes mellitus using an insulin analogue, Clin. Ther. 19 (1997) 62–72. J.H. Anderson, R.L. Brunelle, P. Keohane, V.A. Koivisto, M.E. Trautmann, L. Vignati, Mealtime treatment with insulin analog improves postprandial hyperglycemia and hypoglycemia in patients with non-insulin-dependent diabetes mellitus, Arch. Intern. Med. 157 (1997) 1249–1255. M.N. Feinglos, C.H. Thacker, J. English, M.A. Bethel, J.D. Lane, Modification of postprandial hyperglycemia with insulin lispro improves glucose control in patients with Type 2 diabetes, Diabetes Care 20 (1997) 1539–1542. A.M. Rosenfalck, P. Thorsby, L. Kjems, et al. Improved postprandial glycaemic control with insulin aspart in Type 2 diabetic patients treated with insulin, Acta Diabetol. 37 (2000) 41–46. B.O. Boehm, P.D. Home, C. Behrend, N.M. Kamp, A. Lindholm, Premixed insulin aspart 30 versus premixed human insulin 30/70 twice daily: a randomized trial in Type 1 and Type 2 diabetic patients, Diabet. Med. 19 (2002) 393–399. P.D. Home, L. Barriocanal, A. Lindholm, Comparative pharmacokinetics and pharmacodynamics of the novel rapid-acting insulin analogue, insulin aspart, in healthy volunteers, Eur. J. Clin. Pharmacol. 55 (1999) 199–203. S.R. Mudaliar, F.A. Lindberg, M. Joyce, et al. Insulin aspart (B28 asp-insulin): a fast-acting analog of human insulin: absorption kinetics and action profile compared with regular human insulin in healthy non diabetic subjects, Diabetes Care 22 (1999) 1501–1506.
203
[23] L. Heinemann, T. Heise, L.N. Jorgensen, A.A. Starke, Action profile of the rapid acting insulin analogue: human insulin B28Asp, Diabetic Med. 10 (1993) 535–539. [24] M. Berger, L. Heinemann, Are presently available insulin analogues clinically beneficial? Diabetologia 40 (Suppl. 2) (1997) S91–S96. [25] L. Heinemann, A.A.R. Starke, A. Hohmann, M. Berger, Timing between the subcutaneous administration of insulin and consumption of a carbohydrate rich meal, Horm. Metab. Res. 10 (Suppl. 26) (1992) 137–139. [26] L. Heinemann, Do insulin-treated diabetic patients use an injection-meal-interval in daily life? Diabetic Med. 12 (1995) 449–450. [27] I. Hanning, P.D. Home, K.G. Alberti, Measurement of free insulin concentrations: the influence of the timing of extraction of insulin antibodies, Diabetologia 28 (1985) 831–835. [28] J. Harrison, A.W. Hodson, A.W. Skillen, R. Stappenbeck, L. Agius, K.G.M.M. Alberti, Blood glucose, lactate, pyruvate, glycerol, 3-hydroxybutyrate, and acetoacetate measurements in man using a centrifugal analyser with fluorimetric attachment, J. Chem. Clin. Biochem. 26 (1988) 141–146. [29] M.R. Taskinen, Hyperlipidaemia in diabetes, in: D.J. Betteridge (Ed.), Clinical Endocrinology and Metabolism, vol. 3, Balliere Tindall, London, 1990, pp. 743–775. [30] J. Joven, E. Vilella, B. Costa, P.R. Turner, C. Richart, L. Masana, Concentrations of lipids and apolipoproteins in patients with clinically well-controlled insulin-dependent and non-insulin-dependent diabetes, Clin. Chem. 35 (1989) 813–816.
Further reading [1] A. Gallagher, P.D. Home, The effect of insulin aspart on metabolic outcome and albumin excretion in Type 2 diabetes, Diabetologia 43 (Suppl. 1) (2000) A200. [2] A. Gallagher, P.D. Home, The effect of insulin aspart on postprandial metabolism and metabolic outcome in Type 2 diabetes, Diabetologia 44 (Suppl. 1) (2001) A209. [3] A. Gallagher, P.D. Home, The effect of insulin aspart on postprandial metabolism and lipid profile in Type 2 diabetes, Diabetologia 45 (Suppl. 2) (2002) A253.