Clinical Nutrition (2008) 27, 254e257

available at www.sciencedirect.com

http://intl.elsevierhealth.com/journals/clnu

ORIGINAL ARTICLE

Glycaemic index of modified cornstarch in solutions with different viscosity. A study in subjects with diabetes mellitus type 2* Eva A. Olausson*, Anders Kilander Institute of Medicine, The Sahlgrenska Academy, Go¨teborg University, SE-413 45 Go¨teborg, Sweden Received 17 August 2007; accepted 11 December 2007

KEYWORDS Diet; Diabetes mellitus; Glycaemic index; Swallowing abnormalities; Blood glucose; Cornstarch

Summary Background & aims: Modified cornstarch may be used to increase the viscosity of liquid food for patients with swallowing abnormalities. The aim of this study was to estimate glycaemic index (GI) of modified cornstarch in solutions with different viscosities. Methods: Eight subjects with diabetes mellitus type 2 participated in the study. Their mean  SD glycosylated haemoglobin was 7.1  0.6%. Twenty-five gram of carbohydrate from the modified cornstarch was studied in 150 and 300 g water, respectively, and compared with white bread. Results: GI for the pudding was found to be 77 (P < 0.05 compared to white bread) and 88 for the stew (N.S.). The peak time of blood glucose was earlier for stew and pudding, 45 min (P Z 0.009) and 75 min (P Z 0.01), respectively, compared with 90 min for white bread. Both stew and pudding produced lower glucose levels at 180 min than the white bread. Conclusions: The meal with the highest viscosity, pudding, had a GI of 77. The pudding has a more favourable postprandial blood glucose profile than thinner solutions. GI appears to be misleading in subjects with DM. The postprandial glucose profile should be determined in subjects with DM for selection of food products suitable for these patients. ª 2008 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

* Conference presentation: The results have been presented in an abstract at one national meeting in Sweden. The complete paper has not been published. * Corresponding author. Tel.: þ46 31 342 10 00; fax: þ46 31 82 38 20. E-mail addresses: [email protected] (E.A. Olausson), [email protected] (A. Kilander).

0261-5614/$ - see front matter ª 2008 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. doi:10.1016/j.clnu.2007.12.008

Glycaemic index of modified cornstarch

255

Introduction Swallowing abnormalities are common functional impairments of acute stroke, affecting as many as 50e71% of all patients.1,2 They are not only disabling to patients, but also have the potential to cause serious complications such as dehydration and aspiration pneumonia.2,3 Malnutrition is prevalent among these patients.4,5 These patients are often older and have diabetes mellitus type 2.6,7 It is important to get good metabolic control in subjects with diabetes mellitus to prevent macro- and micro-vascular complications.8e12 The quality of life may be influenced by the glycaemic control.13 Safe swallowing may be achieved in patients with neurogenic dysphagia by manipulation of the viscosity of ingested fluids.14 Modified cornstarch is often used to thicken liquid food. However, the impact of this starch on the blood glucose response has not been evaluated. In 1981, Jenkins et al. introduced the concept of the glycaemic index (GI).15 The GI allows a ranking of carbohydrate foods in terms of their blood glucose rising potential.16 Raben et al. demonstrated that resistant starch had a low glycaemic index.17 The influences on blood glucose for unprepared and heated cornstarch have been evaluated in a previous study.18 Modified cornstarch i.e. hydroxypropyldiamylosephosphate, E1442, is a commonly used product for thickening of liquids. This product is not the same as cooked or unprepared cornstarch. This is used for thickening liquids to a gel without heating. The aim of this study was to evaluate the influence of modified cornstarch with different viscosity on postprandial blood glucose concentrations in subjects with diabetes mellitus type 2 and to estimate the glycaemic index with white bread as the reference food.

The subjects were informed about the study and gave their consent to the study, which was approved by the Ethics Committee, Medical Faculty at the Go ¨teborg University, Sweden.

Procedure The subjects arrived to the laboratory at eight o’clock in the morning after an overnight fast. The antidiabetic medication was omitted during the study. The study was performed in randomized order with at least one week but not more than three weeks between each test occasion. Capillary blood samples were taken 5 min immediately before the start of the study (0 min) and 15, 30, 45, 60, 75, 90, 120, 150 and 180 min from the beginning of ingestion the test meal. The peak glucose concentration and the time to reach those concentrations were determined.

Test food Two solutions of modified cornstarch (‘‘Thick & Easy’’, Hormel health Labs, Inc., USA) with different amounts of water were studied. White bread was the reference food. The modified cornstarch thickens the liquid solution. In each test 25 g carbohydrates were studied. The pudding had higher viscosity like a compact pudding with a content of 150 g water than the stew, which was like jelly with 300 g water. Table 2. The energy content in 100 g modified cornstarch is 360 kcal (1550 kJ), protein < 0.1 g, carbohydrate 90 g and fat < 0.1 g. The carbohydrate content for bread was calculated from the National Food Composition Tables19 and for modified cornstarch from the product declaration. The test meal with white bread was studied twice and each solution of modified cornstarch once. No beverage was served at the meals.20

Patients and methods Laboratory analyses Subjects From the Outpatient Diabetes Clinic at the Sahlgrenska University Hospital of about 600 patients with diabetes mellitus type 2, eight patients with DM type 2 (three women and five men) participated in the study. None of the patients had insulin therapy. Characteristics of them are shown in Table 1. They had no clinical history of gastrointestinal autonomic neuropathy, stroke or any neurological disease.

Table 1

The blood glucose was measured with an automatic glucose analyzer (YSI, Yellow Springs Instrumentals, Yellow Springs OH, US). The incremental area under the curve above baseline (IAUC) for glucose was calculated according to the trapezium rule.21 IAUC for white bread was used as control. The glycaemic index for white bread was set to 100. In this study, all glycosylated haemoglobin (HbA1c) values were converted to the DCCT standard levels using the formula: HbA1c (DCCT) Z (0.923  HbA1c (MonoS)) þ 1.345; R2 Z 0.998.21

Characteristics of diabetic subjects

Subjects (n) Male/female (n) Age (year) BMI (kg/m2) HbA1c (%) Diet therapy (n) Oral antidiabetic drug: glimepiride (n) Metformin (n) Glibenclamide (n)

8 3/5 63  4.5 30.3  3.5 7.1  0.6 1 1 4 2

Data are presented as means  SD; BMI: Body mass index.

Table 2 The composition of the tests: white bread, stew and pudding Test food (g)

Carbohydrate (g)

White bread (50) Stew: modified starch (27.5); water (300) Pudding: modified starch (27.5); water (150)

25 25 25

256

Figure 1 Blood glucose response from the beginning of the ingestion of white bread (-:-), stew (-B-) and pudding (-C-) in subjects with diabetes type 2 (P < 0.05).

Statistical analyses Friedman’s test for three variables and Wilcoxon signed rank test were used. The level of significance was set to P  0.05.

Results The mean glucose response curves are shown in Fig. 1. After the stew the mean peak time was found to be at 45 min and the mean peak time for the pudding was 75 min. The latest mean peak time was observed after white bread at 90 min. Friedman’s analysis shows a significant difference, P Z 0.0004 in peak time for white bread, stew and pudding. There was a significant difference in peak time between white bread and stew (P Z 0.009 Wilcoxon sign rank test), and for white bread and pudding (P Z 0.01). After the stew the mean blood glucose increased 3.1 mmol/L, after the pudding 2.7 mmol/L and after the white bread the mean peak response was 3 mmol/L (no significant differences, P Z 0.42 and P Z 0.33, respectively). At 180 min the mean glucose concentration was 1 mmol/ L higher after white bread compared to stew and pudding. IAUC for blood glucose was largest for white bread, 357  29 (mean  SE), for stew 315  49 and less for pudding 275  45. Glycaemic index was found to be 77 for pudding and 88 for stew compared to the reference value (100) for white bread. It was a significant difference between white bread and pudding (P < 0.0499), but not between the stew and the white bread.

Discussion The modified cornstarch solution (pudding) that has a higher viscosity than the studied ‘‘stew’’ was found to have a GI significantly lower than white bread. The blood glucose

E.A. Olausson, A. Kilander peak time for stew and pudding appeared significantly earlier than that of these after white bread despite a lower GI. These important findings in subjects with DM have not been reported previously. A product may thus have a low GI, but a very early and high initial blood glucose rise. Blood glucose profiles may be quite different in subjects with DM from that in healthy controls, which have a normal release of insulin. Subjects with diabetic type 2 do not have the early phase insulin response as the healthy individuals have. Our results indicate that the best type of meal for diabetic subjects are the thickest, pudding, and do not support the use of thinner solutions. Collings et al., observed in a previous study that after raw cornstarch the mean blood glucose response was <1 mmol/L, and that cooked cornstarch gave a mean blood glucose raise of 3 mmol/L in healthy subjects.18 In this study the ingestion of modified cornstarch as stew was followed by a mean blood glucose raise that was similar to that after cooked cornstarch. It seems to be that modified cornstarch and cooked cornstarch influence the blood glucose in same way. The blood glucose concentration is measured after ingestion of 50 g carbohydrates for determination of glycaemic index. A meal with 25 g carbohydrates was used in present study, because it would have been hard to ingest such a big portion as one with 50 g of modified cornstarch. Wolever et al. estimated GI for cooked cornstarch to 97, when GI for white bread was 100 in subjects with diabetes.22 However, in present study GI for stew was 88 and for pudding 77. Thus, the present pudding formula seems to be the best for patients with diabetes. Previous studies have shown that the rate of gastric emptying influence the glucose response.23,24 The effect of gel fibre on gastric emptying has been evaluated in several studies and shown that gel fibre delays the gastric emptying.25,26 When different amounts of gel fibre were studied, the viscosity was changed and was found to correlate to the gastric emptying.26 We found that the modified starch makes a solution like a gel. The blood glucose response was decreased by a higher viscosity like a gel fibre. At a higher viscosity the rate of gastric emptying is decreased, which influences the blood glucose response. Thus, it is not unexpected that the stew gave an early blood glucose peak. High GI is often associated with an early and high peak in the glucose response. However, both the stew and the pudding had significantly (P Z 0.009 and P Z 0.01, respectively) earlier peak than white bread despite lower GI. This indicates that a high GI not always must include early high blood glucose peak. GI gives only information of the area under the incremental curve for blood glucose, but not its shape. The area can be the same, but the shape is different. One shape could be from a low glycaemia response, but during a long time, and another shape could be a high and early glycaemic response, but during a short time. In conclusion, it was found that a rather thick solution is needed to get a favourable postprandial glucose curve. Such a thick solution is also often easier to swallow for patients with neurological disorders affecting the swallowing function. In the choice of suitable products not only GI should be taken into account but also the early postprandial

Glycaemic index of modified cornstarch blood glucose rise.8 This may be important when such products are recommended to patients with diabetes.

Conflict of interest statement None declared.

Acknowledgements This work was supported by grants from the Local Diabetes Association of Gothenburg. The sponsor had no other involvement in the study.

References 1. Gordon C, Hewer RL, Wade DT. Dysphagia in acute stroke. Br Med J (Clin Res Ed) 1987 Aug 15;295(6595):411e4. 2. Wojner AW, Alexandrov AV. Predictors of tube feeding in acute stroke patients with dysphagia. AACN Clin Issues 2000 Nov; 11(4):531e40. 3. Mann JI, De Leeuw I, Hermansen K, et al. Evidence-based nutritional approaches to the treatment and prevention of diabetes mellitus. Nutr Metab Cardiovasc Dis 2004 Dec;14(6):373e94. 4. Keller HH. Malnutrition in institutionalized elderly: how and why? J Am Geriatr Soc 1993 Nov;41(11):1212e8. 5. Elmstahl S, Bulow M, Ekberg O, Petersson M, Tegner H. Treatment of dysphagia improves nutritional conditions in stroke patients. Dysphagia 1999 Spring;14(2):61e6. 6. Lugger KE. Dysphagia in the elderly stroke patient. J Neurosci Nurs 1994 Apr;26(2):78e84. 7. Meigs JB. Epidemiology of the metabolic syndrome. Am J Manag Care 2002 Sep;8(11 Suppl.):S283e92 [quiz: S293e6]. 8. Bonora E. Postprandial peaks as a risk factor for cardiovascular disease: epidemiological perspectives. Int J Clin Pract Suppl 2002 Jul;129:5e11. 9. Yokoyama I, Yonekura K, Ohtake T, et al. Coronary microangiopathy in type 2 diabetic patients: relation to glycemic control, sex, and microvascular angina rather than to coronary artery disease. J Nucl Med 2000 Jun;41(6):978e85. 10. Rizkalla SW, Bellisle F, Slama G. Health benefits of low glycaemic index foods, such as pulses, in diabetic patients and healthy individuals. Br J Nutr 2002 Dec;88(Suppl. 3):S255e62. 11. Laakso M. Hyperglycemia as a risk factor for cardiovascular disease in type 2 diabetes. Prim Care 1999 Dec;26(4):829e39. 12. Buysschaert M, Hermans MP. Glycaemic and blood pressure controls achieved in a cohort of 318 patients with type 2 diabetes. Acta Clin Belg 1999 Dec;54(6):328e33.

257 13. Marcus MD, Wing RR, Guare J, Blair EH, Jawad A. Lifetime prevalence of major depression and its effect on treatment outcome in obese type II diabetic patients. Diabetes Care 1992 Feb;15(2):253e5. 14. Goulding R, Bakheit AM. Evaluation of the benefits of monitoring fluid thickness in the dietary management of dysphagic stroke patients. Clin Rehabil 2000 Apr;14(2):119e24. 15. Jenkins DJ, Wolever TM, Taylor RH, et al. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981 Mar;34(3):362e6. 16. Wolever TM, Nuttall FQ, Lee R, et al. Prediction of the relative blood glucose response of mixed meals using the white bread glycemic index. Diabetes Care 1985 SepeOct;8(5):418e28. 17. Raben A, Tagliabue A, Christensen NJ, Madsen J, Holst JJ, Astrup A. Resistant starch: the effect on postprandial glycemia, hormonal response, and satiety. Am J Clin Nutr 1994 Oct;60(4):544e51. 18. Collings P, Williams C, MacDonald I. Effects of cooking on serum glucose and insulin responses to starch. Br Med J (Clin Res Ed) 1981 Mar 28;282(6269):1032. 19. Swedish National Food Administration. Food composition tables. Uppsala, Sweden: Norstedt Printing Office; 1993. 20. Torsdottir I, Andersson H. Effect on the postprandial glycaemic level of the addition of water to a meal ingested by healthy subjects and type 2 (non-insulin-dependent) diabetic patients. Diabetologia 1989 Apr;32(4):231e5. 21. Hoelzel W, Weykamp C, Jeppsson JO, et al. IFCC reference system for measurement of hemoglobin A1c in human blood and the national standardization schemes in the United States, Japan, and Sweden: a method-comparison study. Clin Chem 2004 Jan;50(1):166e74. 22. Wolever TMS, Katzman-Relle L, Jenkins AL, Vuksan V, Josse RG, Jenkins DJA. Glycaemic index of 102 complex foods in patients with diabetes. Nutr Res 1994;14(5):651e69. 23. Torsdottir I, Alpsten M, Andersson D, Brummer RJ, Andersson H. Effect of different starchy foods in composite meals on gastric emptying rate and glucose metabolism. I. Comparisons between potatoes, rice and white beans. Hum Nutr Clin Nutr 1984 Sep;38(5):329e38. 24. Mourot J, Thouvenot P, Couet C, Antoine JM, Krobicka A, Debry G. Relationship between the rate of gastric emptying and glucose and insulin responses to starchy foods in young healthy adults. Am J Clin Nutr 1988 Oct;48(4):1035e40. 25. Holt S, Heading RC, Carter DC, Prescott LF, Tothill P. Effect of gel fibre on gastric emptying and absorption of glucose and paracetamol. Lancet 1979 Mar 24;1(8117):636e9. 26. Torsdottir I, Alpsten M, Andersson H, Einarsson S. Dietary guar gum effects on postprandial blood glucose, insulin and hydroxyproline in humans. J Nutr 1989 Dec;119(12): 1925e31.