Nutrition Therapy in Non-insulin-dependent Diabetes ... - Diabetes Care

N

utrition Therapy in Non-insulin-dependent Diabetes Mellitus DAN STREJA, EVELYN BOYKO, AND SIMON W. RABKIN

The purpose of this study was to evaluate nutritional therapy for non-insulin-dependent diabetic subjects by examining the changes in, and interrelationship of, the outcome criteria of efficacy, namely, blood glucose, body weight and serum lipids, total cholesterol, and triglycerides. Eighty-two patients, who represented all those seen by one nutritionist at a diabetic education center, were contacted for longterm follow-up data, which was obtained on 80.5% of.the patients. They had short-term follow-up 66.0 ± 4.5 days after the initial visit and long-term follow-up 31.0 ± 1.1 mo later. Four treatment groups were defined: group 1, diet only at entry and follow-up; group 2, oral hypoglycemic agents at entry and follow-up; group 3, intensified treatment (diet or oral agents at entry and oral agents or insulin at follow-up); and group 4, decrease in intensity of treatment (oral agents at entry but not at long-term follow-up). Serum glucose fell significantly (P < 0.05) at short- and long-term follow-up in groups 1, 2, and 3. Body weight reductions were significant at short-term follow-up in all groups and were significant at long-term follow-up only in group 1. Serum cholesterol was significantly reduced at short-term followup in group 2. Serum triglyceride changes were significantly reduced at short-term follow-up in groups 1 and 2 and at long-term follow-up in group 1. Serum glucose changes correlated with changes in serum cholesterol and triglycerides but did not correlate with body weight changes. These findings demonstrate the effectiveness of nutritional counseling, show that lowering serum glucose did not correlate with weight reduction, and suggest that more attention should be given to factors reducing serum glucose and less attention to weight loss in the treatment of non-insulin-dependent diabetes, DIABETES CARE 4- 81-84, JANUARY-FEBRUARY 1981.

N

utritional counseling is a cornerstone in the treatment of non-insulin-dependent diabetes (NIDD)1"3 and has assumed even greater importance since the role of oral hypoglycemic agents has been questioned.2 Weight loss, change in blood glucose, and change in serum lipids (cholesterol and triglycerides) are the three major outcome criteria used to judge the effectiveness of dietary therapy. Although the effectiveness of various kinds of diets to modify favorably these outcomes has been well documented,4"6 there is comparatively less information on the interrelationship of these factors. Also, there is a great deal of justifiable concern that compliance with dietary advice is poor and that it seriously affects ability to achieve optimal results.7"11 It is the purpose of this investigation to ascertain the effectiveness of individual counseling in a Diabetic Education Center to achieve short- and long-term

changes in three outcome variables—blood glucose, body weight, and serum lipids (total cholesterol and triglycerides) —and to examine the interrelationship of changes in these outcomes. METHODS

Eighty-two patients represented all non-insulin-dependent diabetic persons who had been referred to the Diabetic Education Center (DEC) of the Health Sciences Centre between 1975 and 1977 for nutritional therapy. These patients were not receiving insulin, had been seen by one therapist, and had returned for short-term follow-up that included biochemical investigation. All patients had fasting serum glucose over 135 mg/dl at initial evaluation at the DEC. They were contacted by telephone and asked whether they would

DIABETES CARE, VOL. 4 NO. 1, JANUARY-FEBRUARY 1981

81

NUTRITION THERAPY IN NIDDM/D. STREJA, E. BOYKO, AND S. W. RABKIN

agree to return for follow-up examination. Seventy-six percent (62 of 82) came in for examination, 4.8% (4) requested that results of examinations from their personal physician be used, 3.7% (3) refused examination, 2.4% (2) were decreased, and 13.4% (11) could not be located. Thus, the study group consisted of 66 persons. Patient characteristics at the initial examination are as follows. Thirty-nine were men and 27 were women. The mean age of the group was 56.7 ± 1.1 yr (SEM). Most sub' jects were overweight as evidenced by a mean relative weight (actual weight/ideal weight) of 1.35 ± 0.03 (135% overweight). The mean serum glucose was 198 ± 7 mg/dl, mean serum cholesterol 246 ± 8 mg/dl, and mean serum triglycerides 305 ± 39 mg/dl. The methods for measuring serum glucose,12 serum cholesterol13 and triglycerides14 used enzymatic techniques. After the entry examination each patient received an individual 1-h counseling session that instructed them on nutrition for diabetes using a standard dietary approach (the exchange system for meal planning15'16). All persons received the St.Pauls' "Guide to Mean Planning,"16 which consists of a one-sheet foldout pamphlet displaying six food groups in colored modules (protein, milk, starch, fruit, fat, and vegetables). Portions are standardized to convenient units, for example, 1 ounce, Vz cup, or 1 teaspoon. A number of foods such as spices, clear broth, coffee, and tea are considered "free foods" and need not be measured. A number of foods do not appear on the exchange lists because of their high "sugar" content such as candy, preserves, and cakes. A section labeled daily food allowance is used to establish an individualized meal plan, specifying the amount of food from each list that is recommended for each meal and snack. Based on an estimate of the individual's usual daily food intake, a daily food allowance was recommended to initially reduce intake by 500-1000 calories per day. Patients were counseled about the necessity to lose weight as part of the dietary treatment of diabetes. At the short-term follow-up a brief discussion of their diet usually occurred. Short-term follow-up was 66.0 ± 4 . 5 days after the initial visit and long-term follow-up examination was 31.0 ± 1.1 mo later. At each examination blood samples were taken after an overnight fast and analyzed for serum glucose, cholesterol, and triglycerides. DATA ANALYSIS

T

he study design was to test the null hypothesis that there was no change in body weight or any biochemical measures between initial examination, short'term follow-up, and long-term follow-up. To adjust for the effect of pharmacologic therapy after the initial examination, the patients were divided into four groups. Group 1 (N = 3 1 ) consisted of those on dietary therapy only since entry. Group 2 (N = 21) consisted of persons who were receiving oral hypoglycemic agents at entry and on follow-up examinations. Group 3 (N = 8) consisted of six persons who were receiving only dietary management at entry

82

but were also receiving oral hypoglycemic agents at longterm follow-up and two persons who were receiving insulin at long-term follow-up, of whom one had been on diet and one on oral agents at entry. Group 4 (N = 6) consisted of those who were on oral agents at entry but not at the longterm follow-up examination. Within each group the outcome variables were compared between entry, short-term, and long-term examinations. Paired Student's t test17 was used to compare the different examinations. RESULTS

At short-term follow-up, statistically significant decreases in body weight and fasting serum glucose were found in all groups except for serum glucose in group 4 (Table 1). Increases in these variables were found between short- and long-term examination for all groups except fasting serum glucose for group 3. Nevertheless, group 1 showed significant (P < 0.01) decreases in body weight and serum glucose compared with entry. Serum triglycerides were significantly lower both at short- and long-term follow-up, compared with entry, for group 1. Regarding intercorrelations of the changes in the three TABLE 1 Changes in body weight, serum glucose, cholesterol, and triglycerides Follow-up Entry Body weight (kg) 84.8 ± 3.3 Group 1 2 89.3 ± 3.0 3 81.1 ± 8.0 79.1 ± 8 . 8 4 Serum glucose (mg/dl) 184 ± 10 Group 1 216 ± 17 2 220 ± 14 3 179 ± 15 4 Serum cholesterol (mg/dl) Group 1 250 ± 14 2 239 ± 14 234 ± 18 3 273 ± 39 4 Serum triglyceride;5 (mg/dl) Group 1 360 ± 76 2 233 ± 31 3 298 ± 77 291 ± 107 4

Short'term

80.0 86.0 76.6 76.3 125 152 165 166

± 3.1* ±3.1* ± 7.lt ± 8.0t ± ± ± ±

5* 11* 14* 50

Long-term

82.0 87.6 77.2 78.7

± ± ± ±

3.4* 3.3§ 8.5 8.6§

157 ± lit// 196 ± 16* 138 ± 16*§ 202 ± 41

236 ± 9

245 ± 11

216 ± 9t

249 ± 14t

217 ± 15 235 ± 32

241 ± 17 277 ± 51

186 ± 17*

276 257 263 330

185 ± 2lt 227 ± 53 187 ± 55

± ± ± ±

45t§ 35// 34 120

Change from entry: * P < 0.001, t P < 0.05, $ P < 0.01; change from short- to long-term follow-up: § P < 0.05, // P < 0.01. All Student's t tests are those for paired data but mean ± 1 SEM are shown for comparisons. Group 1 (N = 31) patients on diet therapy only; group 2 (N = 21) patients on oral hypoglycemic agents; group 3 (N = 8 ) patients in whom pharmacologic therapy was instituted or increased; group 4 (N = 6) patients in whom oral hypoglycemic agents were stopped.



outcome variables, groups 1,2, and 4 showed no significant (P > 0.10) correlation between changes in body weight and serum glucose either at short- or long-term follow-up. In group 3, a significant negative correlation was observed between change in body weight and change in serum glucose at short-term (r = -0.899; P < 0.001) and long-term (r = -0.710; P < 0.05) follow-up. At short-term follow-up the only correlation with change in serum cholesterol was change in body weight (r = 0.388; P < 0.05) and serum glucose (r = 0.494; P < 0.05) in group 2. Change in serum triglycerides showed a significant positive correlation with change in serum glucose in groups 1 (r = 0.362; P < 0.05) and 2 (r = 0.439; P < 0.05) but a negative correlation in group 4 (r = -0.882; P < 0.05). At long-term follow-up, serum cholesterol change correlated significantly with changes in body weight (r = —0.658; P < 0.05) and serum glucose (r = 0.811; P < 0.01) only in group 3. Change in serum triglycerides was significantly associated only with change in serum glucose in group 1 (r = 0.506; P < 0.01) and changes in body weight (r = —0.511; P < 0.05) in group 2. To examine the possibility that the lack of correlation between weight loss and the biochemical variables was due to little weight change in the minimally overweight, we further analyzed diabetic subsets of groups 1 and 2 who were 20% or more overweight (relative weight 120 lb or greater) at entry. In group 1, 74% (23) fulfilled this criteria and had a 5.5 ± 0.9-kg short-term and 3.2 ± 1.0kg long-term weight loss. There was no significant correlation between weight loss and serum glucose reduction in short-term (r = 0.075) and longterm (r = 0.163) follow-up. In group 2, 86% (18) were 20% or more overweight and had a 3.7 ± 0.6-kg short-term and 2.8 ± 1.0-kg long-term weight loss. There was no significant correlation between weight loss and serum glucose reduction at short-term (r = 0.064) and long-term (r = 0.046) followup.

DISCUSSION

T

he present study found significant decreases in fasting serum glucose and body weight, both at shortand long-term follow-up, in non-insulin-dependent diabetic subjects who had received nutritional counseling and whose pharmacologic treatment regimen had not changed (group 1 and 2). Skepticism11 about the ability of diet counseling to modify these outcome variables in the diabetic patient may be unwarranted. The problems of adherence to dietary advice are real and serious.7"10 Even greater positive results may be possible when better understanding of methods to improve compliance with medical regimens is obtained. The generalizability of any study, of course, depends on the study population and experimental design. Our study population consists of persons referred to a diabetes education center from medical outpatients departments as well as from practicing physicians in a metropolitan area. Those seen by a single dietitian were selected to eliminate between-

therapist effects.18 The study had retrospective elements but was not a pure retrospective study.19 It was also not a randomized control trial comparing nutritional therapy and a control group for each of the different treatment groups. Consequently, it is subject to some of the same sources of error and bias as all retrospective studies. One of these is the small number of subjects in groups 3 and 4, which may result in a type II statistical error. However, the similarity of some of our findings with those of other studies20"22 strengthens the validity and conclusions of this study. The findings of the present study suggest that improvement in glucose intolerance in diabetes mellitus may be unrelated to weight loss. Two observations support this suggestion. First, we found no correlation between changes in weight and changes in fasting serum glucose. Second, the marked fall in fasting serum glucose at short-term follow-up was out of proportion to the more modest weight loss seen over the same time. This is consistent with the conclusions of others.20*21 The interrelationship between changes in body weight and serum glucose in the diabetic patient are complex. We sought to examine the relationship between weight loss and improvement in diabetic "control." However, worsening diabetic control also may be associated with weight loss. It is unlikely, however, that the latter played a significant role in this study, as serum glucose levels indicated improvement in diabetic control in all groups except group 4. Dietary factors23"26 can improve glucose tolerance on isocaloric diets. They can account for reductions in serum glucose without changes in body weight. Our findings should not be misinterpreted to indicate that weight reduction has no role in management of maturityonset diabetes mellitus. This study did not show that weight reduction cannot lower serum glucose. Rather it found that with a current nutrition approach to diabetes mellitus, the major determinants of serum glucose reduction are not due to weight loss. Obesity is also a predictor of ischemic heart disease.27 Although the major problem with weight reduction is maintaining weight loss,28 attempts at weight reduction are justifiable in a group that is already at high risk of cardiovascular disease because of diabetes mellitus. Changes in serum lipids correlated more strongly and consistently with change in serum glucose than changes in body weight. This suggests that lipids are more strongly related to severity of metabolic disturbance in diabetes, as reflected by serum glucose, than is weight loss. The correlations between serum glucose and total serum cholesterol and triglycerides were not high and probably would have been higher if percent glycosylated hemoglobin had been used to assess degree of metabolic control29'30 (D. Streja, L. Koz, S. W. Rabkin, unpublished observations). It was not measured because glycosylated hemoglobin determinations were not available at the entry examinations and between-examination comparisons could not be possible. In summary, the present study demonstrates the effectiveness of nutritional therapy in non-insulin-dependent diabetes, shows that blood glucose reductions did not correlate with weight loss, and suggests that nutritional therapy might


83


be improved by concentrating more attention on dietary determinants of serum glucose rather than on weight reduction per se. ACKNOWLEDGMENTS: We thank J. Dubois for preparation of the manuscript and acknowledge the assistance of B. Marks. From the Departments of Medicine and Social and Preventive Medicine, University of Manitoba, Health Sciences Centre, Winnipeg, Manitoba, Canada. Address reprint requests to Dan Streja, 7301 Medical Centre Drive, Suite 300, Canoga Park, California 91307. REFERENCES 1 Krall, L. P., and Joslin, A. P.: Joslin's Diabetes Mellitus. Marble, A., White, P. Bradley, R. F., and Krall, L. P., Eds. Philadelphia, Lea & Febiger, 1971. 2 Ricketts, H. T.: Editorial statement on University Group Diabetes Program Results. Diabetes 19 (Suppl. 2) iii-v, 1970. 3 Bierman, E. L , Albrink, M. J., Arky, R. A., Connor, W. E., Dayton, S., Sputz, N., and Steinberg, D.: Special report: principle of nutrition and dietary recommendations for patients with diabetes mellitus: 1971. Diabetes 20: 633-34, 1971. 4 Newburgh, L. H.: Control of hyperglycemia of obese "diabetics" by weight reduction. Ann. Intern. Med. 17: 935, 1952. 5 Stone, D. B., and Connor, W. E.: The prolonged effects of a low cholesterol, high carbohydrate diet upon the serum lipids in diabetic patients. Diabetes 12: 127-32, 1963. 6 Salons, L. B., Hirsch, J., and Knittle, J. L : Obesity, carbohydrate metabolism and diabetes mellitus. In Diabetes Mellitus: Theory and Practice, 2d edit. Ellenberg, M., and Rifkin, H., Eds. New York, McGraw-Hill, 1970, pp. 424-35. 7 Wysocki, M., Czyzyk, A., Slonska, Z., Krolewski, A., and Saneczko, D.: Health behaviour and its determinants among insulin-dependent diabetics. Results of the Diabetes Warsaw Study. Diabetes Metab. 4: 117-22, 1978. 8 Holland, W. M.: The diabetes supplement of the National Health Survey J. Ann. Diet. Assoc. 52: 389-90, 1968. 9 Williams, R, Martin, D. A., Hogan, M. D., Watkins, J. D., and Ellis, E. V.: The clinical picture of diabetic control studied in four settings. Am. J. Public Health 57: 441-58, 1967. 10 Hulka, B. S., Cassel, J. C , Kupper, L. L , and Burdette, J. A.: Communication, compliance and concordance between physicians and patients with presented medication. Am. J. Public Health 66: 847-53, 1976. 11 West, K. M.: Diet therapy of diabetes: an analysis of failure. Ann. Intern. Med. 79: 425-34, 1973. 12 Stein, M. W., and Bergruiyer, H. P.: Methods of enzymatic analysis. New York, Academic Press, 1963, p. 117. 13 Allain, C. C , Poon, L. S., Chan, C. S. G., et al.: Enzymatic determination of total serum cholesterol. Clin. Chem. 20: 470-75, 1974.

84

14

Sampson, E. J., Demers, L. M., and Krieg, A. F.: Faster enzymatic procedure for serum triglycerdies. Clin. Chem. 21: 1938-85, 1975. 15 Canadian Diabetic Association: Exchange Lists for Meal Planning for Diabetics in Canada, 5th edit. Toronto, Canadian Diabetic Association 1971. 16 Guide to Meal Planning: St. Paul, Minn., Children's Hospital, 1975. 17 Snedecor, G. W., and Cochran, W. G.: Statistical Methods, 6th edit. Iowa, Iowa State University Press, 1967. 18 Sackett, D. C , and Haynes, R. B.: Compliance with Therapeutic Regimen. Baltimore, Johns Hopkins University Press, 1976. 19 Feinstein, A. P.: The epidemiologic trohoc, the ablatine risk ratio and "retrospective" research. Clin. Pharmacol. Ther. 14: 291, 1973. 20 H a d d e n , D . R., Montgomery, D . A . D . , Skelly, R. J., T r e m ble, E. R., Weaver, J. A . , Wilson, E. A . , and B u c h a n a n , K. D . : Maturity onset diabetes mellitus: response t o intensive dietary m a n agement. Br. Med. J. 3 ; 2 7 6 - 7 8 , 1975. 21 Wall, J. R., Pyke, D. A . , and Oakley, W . C : Effect of carbohydrate restriction in obese diabetic relationship of control to weight loss. Br. Med. J. 1: 5 7 7 - 7 8 , 1973. 22 Goldner, M. G., Knatterud, G. L , and Prout, T. E.: Effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes 111. Clinical implications of UGDP results. JAMA 218: 1400-10, 1971. 23 Grey, N. and Kipnis, D. W.: Effect of diet composition on the hyperinsulinemia to obesity. N. Engl. J. Med. 285: 827-31, 1971. 24 Jenkins, D . J. A . , Leeds, A . R . , Gassull, M . A . , Alberti, K. G. M. M., Woleno, T . M. S. and Hockday, T . D. R.: Unabsorbable carbohydrates and diabetes: decreased post prandial hyperglycemia. Lancet2: 111, 1976. 25 Gouldner, T . J., Alberti, K. G . M. M . , and Jenkins, D. A . : Effect of added fiber on the glucose and metabolic response to a mixed meal in normal and diabetic subjects. Diabetes Care 1: 335— 87, 1978. 26 Miranda, P. M., and Horwitz, D. L.: High fiber diets in the treatment of diabetes mellitus. A n n . Intern. Med. 88: 4 8 2 - 8 6 , 1978. 27 Rabkin, S. W . , Mathewson, F. A. L , and Hsu, P. H.: Relation of body weight to development of ischemic heart disease in a cohort of young North American men after a 26 year observation period. T h e Manitoba Study. A m . J. Cardiol. 39: 4 5 2 - 5 7 , 1977. 28 Johnson, D . , and Drenick, E. J.: Therapeutic fasting in morbid obesity: long term follow-up. Arch. Intern. Med. 137: 1 3 8 1 - 8 2 , 1977. 29 Koenig, R. J., Peterson, C. M., Kilo, C , Cerami, A . , and Williamson, J. R.: Hemoglobin A| C as an indicator of the degree of glucose intolerance in diabetes. Diabetes 25: 2 3 0 - 3 2 , 1976. 30 Peterson, C . M., Koenig, R. J., Jones, R. L , Saudek, C . D . , and Cerami, A . : Correlation of serum triglyceride levels and hemoglobin A t c concentrations in diabetes mellitus. Diabetes 26: 5 0 7 509, 1977.
