Multicenter randomized trial of a comprehensive prepared meal program in type 2 diabetes

Clinical

Care/Education/Nutrition

O R I G I N A L

A R T I C L E

Multicenter Randomized Trial of a Comprehensive Prepared Meal Program in Type 2 Diabetes F. XAVIER PI-SUNYER, MD CAROL A. MAGGIO, PHD DAVID A. MCCARRON, MD MOLLY E. REUSSER, BA JUDITH S. STERN, SCD R. BRIAN HAYNES, MD, PHD SUZANNE OPARIL, MD PENNY KRIS-ETHERTON, PHD

LAWRENCE M. RESNICK, MD ALAN CHAIT, MD CYNTHIA D. MORRIS, PHD DANIEL C. HATTON, PHD JILL A. METZ, PHD GEOFFREY W. SNYDER, MS SHARON CLARK, PHD MARGARET MCMAHON, MN

OBJECTIVE — To evaluate the clinical effects of a comprehensive prepackaged meal plan, incorporating the overall dietary guidelines of the American Diabetes Association and other national health organizations, relative to those of a self-selected diet based on exchange lists in free-living individuals with type 2 diabetes. RESEARCH DESIGN AND METHODS — A total of 202 women and men (BMI 42 kg/m2) whose diabetes was treated with diet alone or an oral hypoglycemic agent were enrolled at 10 medical centers. After a 4-week baseline period, participants were randomized to a nutrient-fortified prepared meal plan or a self-selected exchange-list diet for 10 weeks. On a caloric basis, both interventions were designed to provide 55–60% carbohydrate, 20–30% fat, and 15–20% protein. At intervals, 3-day food records were completed, and body weight, glycemic control, plasma lipids, and blood pressure were assessed. RESULTS — Food records showed that multiple nutritional improvements were achieved with both diet plans. There were significant overall reductions in body weight and BMI, fasting plasma glucose and serum insulin, fructosamine, HbA1c, total and LDL cholesterol, and blood pressure (P 0.001 or better for all). In general, differences in major end points between the diet plans were not statistically significant. CONCLUSIONS — Glycemic control and cardiovascular risk factors improve in individuals with type 2 diabetes who consume diets in accordance with the American Diabetes Association guidelines. The prepared meal program was as clinically effective as the exchange-list

From the Division of Endocrinology, Diabetes, and Nutrition (F.X.P.-S., C.A.M.), St. Luke’s-Roosevelt Hospital Center, Columbia University, College of Physicians and Surgeons, New York, New York; the Division of Nephrology, Hypertension and Clinical Pharmacology (D.A.M., M.E.R., C.D.M., D.C.H., J.A.M., G.W.S.), Department of Medicine, Oregon Health Sciences University; the Clinical Research Group of Oregon (S.C., M.M.), Portland, Oregon; the Department of Nutrition ( J.S.S.), Division of Clinical Nutrition and Metabolism, University of California, Davis, California; the Hypertension Program (S.O.), University of Alabama at Birmingham, Birmingham, Alabama; the Nutrition Department (P.K.-E.), College of Health and Human Development, Pennsylvania State University, University Park, Pennsylvania; the Division of Endocrinology/Hypertension (L.M.R.), Wayne State University Medical Center, Detroit, Michigan; the Division of Metabolism, Endocrinology, and Nutrition (A.C.), Department of Medicine, University of Washington, Seattle, Washington; and the Department of Clinical Epidemiology and Biostatistics (R.B.H.), McMaster University, Hamilton, Ontario, Canada. Address correspondence and reprint requests to F. Xavier Pi-Sunyer, MD, Professor of Medicine, Chief, Division of Endocrinology, Diabetes, and Nutrition, St. Luke’s -Roosevelt Hospital Center, 1111 Amsterdam Ave., New York, NY 10025. E-mail: [email protected]. Received for publication 31 March 1998 and accepted in revised form 19 October 1998. F.X.P.-S., D.A.M., J.S.S., R.B.H., S.O., P.K.-E., L.M.R., and A.C. serve on the medical advisory board for the Campbell Soup Company and have received honoraria for this service. Abbreviations: AHA, American Heart Association; CCNW, Campbell’s Center for Nutrition and Wellness; NCEP, National Cholesterol Education Program; SSD, self-selected diet. A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.

DIABETES CARE, VOLUME 22, NUMBER 2, FEBRUARY 1999

diet. The prepared meal plan has the additional advantages of being easily prescribed and eliminating the complexities of meeting the multiple dietary recommendations for type 2 diabetes management. Diabetes Care 22:191–197, 1999

ardiovascular disease is the leading cause of death in individuals with type 2 diabetes (1), the great majority of whom are obese (2) and many of whom have the additional cardiovascular risk factors of dyslipidemia (3,4) and/or hypertension (4). Diet, either alone or as an adjunct to medication, has long been considered essential to the therapeutic management of type 2 diabetes (5). There is much evidence to show that in individuals with type 2 diabetes, nutrition therapy and weight loss can improve glycemic control and ameliorate concomitant conditions such as dyslipidemia and hypertension (2,5,6). Nevertheless, compliance with dietary recommendations is often poor (7–9). Effectively incorporating dietary recommendations into the typical patient’s lifestyle requires creating an appropriate individualized meal plan, teaching appropriate food choices and preparation methods, and sustaining motivation and long-term compliance. The comprehensive meal plan assessed in the present study was designed to simplify these requirements by incorporating into a simple and convenient total meal program the overall macro- and micronutrient recommendations of the American Diabetes Association for the nutritional management of type 2 diabetes (10), as well as those of the American Heart Association (AHA) (11), the National Cholesterol Education Program (NCEP) (12), and the Food and Nutrition Board of the National Academy of Sciences (13) for cardiovascular risk reduction. This meal program eliminates the complexities, concerns, and time associated with the selection, preparation, and consistent consumption of a health-promoting diet and could thus facilitate better compliance. The Cardiovascular Risk Reduction Dietary Intervention Trial is a series of randomized multicenter studies evaluating the

C

191

Prepared meal plan in type 2 diabetes

clinical effects of this comprehensive meal program on type 2 diabetes, dyslipidemia, and hypertension alone or in combination. Overall results of the first study in this series, which involved 560 participants with one or more of the above disorders, have been reported previously (14). The purpose of that study was to evaluate whether the comprehensive meal plan, which offers greater convenience and could potentially promote better compliance, was as clinically effective and safe as a selfselected diet based on the exchange lists of the American Dietetic Association and the American Diabetes Association (15). The present paper describes results for the 202 participants with type 2 diabetes. This paper thus also provides an evaluation of the effectiveness of diets that follow American Diabetes Association guidelines in improving glycemic control and cardiovascular risk factors in individuals with type 2 diabetes. RESEARCH DESIGN AND METHODS — This phase of the Cardiovascular Risk Reduction Dietary Intervention Trial was conducted at 10 medical/research centers in the U.S. and Canada. The protocol was approved by each center’s institutional review board, and potential subjects provided written informed consent before screening. Study subjects Women and men aged 25–70 years who had type 2 diabetes treated with diet or an oral hypoglycemic agent were recruited through clinics and advertisements. Eligible subjects had a BMI 42 kg/m2. At screening, diet-treated subjects were required to have a fasting glucose level 7.77 mmol/l and an HbA1c level 200% of the median for assay (15.4%). Individuals with chronic life-threatening diseases or a history of alcohol or drug abuse, and women who were pregnant, lactating, or not practicing a medically approved method of birth control were excluded from participation. Study design During the 4-week baseline period (weeks 4 through 0), subjects were seen weekly for weight and blood pressure measurements. They maintained their usual diets, were instructed by the study nutritionist in the preparation of 3-day food records, and completed two sets of food records. Food records were completed for two nonconsecutive weekdays and one weekend day. 192

Subjects were instructed to quantify portion sizes by the use of diagrams, measuring cups and spoons, and rulers and to record food preparation methods and recipe ingredients. At weeks 2 and 1, study nutritionists reviewed food records with the subjects for completeness and detail. Blood samples were collected at week 4 for fasting glucose and HbA1c measurements, and at week 2, for lipoprotein determinations. At week 2, a nutrition prescription was calculated for each subject using the Harris-Benedict equation adjusted for the individual’s activity level (16). A caloric intake goal with a 200-kcal range was targeted, and the minimum prescription was 1,200–1,399 kcal/day. Weight loss was limited to 1 kg/wk; subjects not desiring weight loss were prescribed a weight-maintenance diet. Subjects were then randomized by the coordinating center at Oregon Health Sciences University to one of the two diet plans. During the 10-week diet intervention period, subjects consumed either the Campbell’s Center for Nutrition and Wellness (CCNW) meal program or a selfselected diet (SSD). At biweekly clinic visits, weight and blood pressure were monitored and 3-day food records were collected and reviewed with the nutritionist. HbA1c, fructosamine, and insulin were measured at weeks 0 and 10; fasting plasma glucose was measured at weeks 0, 4, and 10. Lipids were measured at weeks 0, 4, 8, and 10. On a caloric basis, both the CCNW and the SSD plans targeted approximately 55–60% carbohydrate, 15–20% protein, and 20–30% fat. Because the CCNW program was designed as a comprehensive diet for the treatment of generalized cardiovascular and metabolic disease, this diet prescription differs from the current American Diabetes Association recommendation of an individualized treatment plan based on a diabetes patient’s coexisting risk factors (10). The CCNW diet plan, described in detail in earlier publications (14,17), consisted of 6 prepackaged breakfast, 8 lunch, and 10 dinner entrées, with 6 snack selections. These foods were provided free of charge and ordered at clinic visits for home delivery. Subjects on the CCNW plan were instructed to consume one breakfast, lunch, and dinner entrée, as well as one serving each of fruit, vegetables, and low-fat dairy products daily. Calories were adjusted through the inclusion of snacks or addi-

tional entrées. One bonus selection of an alcoholic beverage or the isocaloric equivalent in fruits, vegetables, or dairy products was allowed. The CCNW diet plan satisfied sodium, total and saturated fat, cholesterol, fiber, and carbohydrate recommendations and was supplemented with vitamin and minerals to satisfy at least 100% of the recommended adult daily allowances for all nutrients except vitamin D. As also described earlier (14,17), subjects on the SSD plan were prescribed a fixed number of servings from each of the American Dietetic Association and American Diabetes Association exchange lists (15). One daily bonus selection was allowed as above. Subjects on the SSD plan received a food allowance of $40 per week. For all study participants, nutrition counseling was limited to a 30-min session at week 0 and a 15-min session at week 2. This counseling protocol was modeled after that typically used in routine clinical practice, but was less than that recommended by current guidelines of medical nutrition therapy for type 2 diabetes (18). At week 0, subjects were given verbal instructions and educational material, including an instruction manual on food record collection, and their diet prescription. Subjects on the SSD plan also received the American Dietetic Association and American Diabetes Association exchange lists (15) and sample menus and recipes. The number of servings from each exchange list and the suggested distribution of the food exchanges throughout the day were provided in the diet prescription. At week 2, subjects received further explanation of and advice on adhering to their diets. After week 2, they were referred to their manuals for further clarification. The only exception was when the prescribed calorie level appeared inappropriate according to the subject’s in-clinic body weight, and thus required adjustment by the nutritionist and revised instruction to the subject. Clinical measures Food records were analyzed at the coordinating center using a licensed copy of the Nutrition Coordinating Center database of the University of Minnesota (NDS version 2.8, 1995), supplemented with product content information provided by the Campbell Soup Company. Nutritionist training was standardized across centers in a pre-study session at which all study nutritionists received instruction based on the University of Minnesota database on the

DIABETES CARE, VOLUME 22, NUMBER 2, FEBRUARY 1999

Pi-Sunyer and Associates

Table 1—Average daily nutrient intakes CCNW diet Baseline Treatment n Energy (kcal) Percent carbohydrate Percent protein Percent fat Percent saturated fat Cholesterol (mg) Fiber (g) Sodium (mg)

100 1,917 ± 565 49 ± 6.9 18 ± 2.8 34 ± 6.8 11 ± 3.0 280 ± 141 21 ± 7.3 3,553 ± 1,058

94 1,655 ± 357* 61 ± 4.2 20 ± 1.6 18 ± 3.6† 6 ± 1.3‡ 114 ± 50 34 ± 6.0† 2,500 ± 511‡

Baseline

SSD diet Treatment

102 2,160 ± 706 46 ± 6.5 18 ± 3.2 36 ± 6.3 12 ± 2.8 331 ± 161 22 ±7.3 3,921 ± 1,429

101 1,679 ± 442 55 ± 7.0 19 ± 2.5 27 ± 6.9 8 ± 2.8 202 ± 86 24 ± 8.0 3,291 ± 1175

Data are means ± SD. P 0.0001 within diet groups for treatment vs. baseline for all variables listed. *P 0.05, †P 0.0001, ‡P 0.01 between diet groups for change from baseline.

completion of food records, calculation of diet prescriptions, and implementation of the protocol. Food records were further reviewed at the coordinating center before entry into the database and, when necessary, study nutritionists were queried to clarify incomplete or inaccurate records. Glucose, HbA1c, and fructosamine were measured by SmithKline Beecham Clinical Laboratories, Seattle. Spectrophotometric methods were used to determine glucose (Olympus Model 5200; Olympus America, Melville, NY) and fructosamine (Olympus AU800; Olympus America, Dallas, TX) levels; HbA1c was determined by high-performance liquid chromatography (Biorad Variant, San Francisco, CA). Insulin was measured with a solid-phase radioimmunoassay (Coat-A-Count; Diagnostic Products, Los Angeles, CA) at the Hormone and Mineral Laboratory, Oregon Health Sciences University. Plasma lipids were determined at the Northwest Lipid Research Laboratories, University of Washington. Triglycerides and total cholesterol were assayed by colorimetric enzymatic analysis (Abbott Spectrum Analyzer, Abbott Laboratories, Abbott Park, IL) (19). HDL cholesterol was analyzed by the heparin-manganese method (20), and LDL cholesterol was calculated using the Friedewald algorithm (21). Blood pressure measurement (22) was standardized across centers in a pre-study training session.

study were identical to those used in the present study, and, thus, the pilot sample was similar in demographic and clinical characteristics. The present study was designed with a power of 80% and a two-sided of 0.05 to detect, with a sample of 80 in each group, between-diet differences of 0.6 mmol/l in glucose, 1.0% in HbA1c, 0.26 mmol/l in total cholesterol, 0.23 mmol/l in triglycerides, and 3 mmHg in blood pressure. All data except weight were averaged over the baseline period when multiple measures were available. Since changes in weight could have occurred as a result of study participation, weight at the final baseline visit (week 0) was considered baseline weight. For blood pressure and serum lipids, values at weeks 8 and 10 were averaged. Using an intention-to-treat analysis, the last available measurement in the treatment period was used for subjects who did not complete the study. Statistical analyses were performed on an a priori basis. A repeated measures analysis of variance model was used to compare differences between baseline and treatment periods. An analysis of covariance model was used to adjust for the effect of weight loss. Categorical variables were analyzed using a 2 or Fisher’s exact test. A two-sided probability value 0.05 was considered statistically significant. Data are presented as means ± SD.

Statistical analysis Power calculations were based on data obtained in a pilot study (23), which involved a total of 77 subjects (39 randomized to the prepared meal plan and 38 to self-selected diet). The design, recruitment methods, and analysis of the pilot

RESULTS — A total of 202 subjects were randomized: 100 to the CCNW meal plan and 102 to the self-selected diet. The two groups did not differ significantly in age, sex, or ethnic distribution or in the use of oral diabetic medications. The overall cohort was 55.5 ± 8.9 years of age and con-

DIABETES CARE, VOLUME 22, NUMBER 2, FEBRUARY 1999

sisted of 102 women and 100 men; 83% (n = 168) of the cohort was white and 11% (n = 22) was black. The remaining 6% were Asian/Pacific Islander (n = 5), Hispanic (n= 5), or Native American (n = 2). Oral diabetic medication was used by 75% of the cohort; the remaining 25% were treated by diet alone. Antihypertensive medication was used by 46% of the cohort, and lipidlowering medication was used by 7%. Use of antihypertensive or lipid-lowering medication did not differ in the two diet groups. Four randomized subjects in the CCNW group and one in the SSD group withdrew from the study before beginning diet treatment. During the 10-week treatment period, one CCNW subject withdrew from the study for an unspecified reason; no SSD subjects withdrew. Table 1 shows self-reported daily nutrient intakes before and at the end of the intervention period. Total caloric intakes decreased significantly in both diet groups (P 0.0001). This decrease was greater (P 0.05) in the SSD ( 468 ± 509 kcal) than in the CCNW group ( 257 ± 558 kcal). Both diet groups reported significant increases in the percentages of calories consumed as carbohydrate and protein (P 0.0001 for both). The increase in carbohydrate intake was 11 ± 7.3% in the CCNW group and 9 ± 7.7% in the SSD group; the increase in protein intake was 2% in both groups. Significant decreases were observed in the percentages of calories consumed as total and saturated fat (P 0.0001 for both). These decreases were significantly greater in the CCNW than in the SSD group (total fat: 15 ± 6.5 vs. 10 ± 7.3%, respectively, P 0.0001; saturated fat: 5 ± 2.9 vs. 4 ± 2.9%, P 0.01). The diet interventions resulted in significant improvements in cholesterol, fiber, and sodium intakes (P 0.0001 for all) (Table 1). The observed decrease in cholesterol intake tended to be greater in the CCNW group ( 161 ± 133 mg) than in the SSD ( 128 ± 137 mg) group. The CCNW group, compared with the SSD group, showed a significantly greater increase in fiber intake (12 ± 6.9 vs. 3 ± 6.6 g, respectively, P 0.0001) and a significantly greater decrease in sodium intake ( 1,016 ± 927 vs. 600 ± 1,064 mg, P 0.01). At week 0, body weight was 92.2 ± 17.0 kg in the CCNW group and 95.0 ± 17.3 kg in the SSD group; BMI was 31.8 ± 4.8 and 32.1 ± 4.6, respectively. Although not an objective of this study, weight loss was desired by 93% of subjects; average 193

Prepared meal plan in type 2 diabetes

Table 2—Glycemic control parameters CCNW diet Baseline Treatment Plasma glucose 10.6 ± 3.3 (97) (mmol/l) HbA1c (%) 8.8 ± 1.6 (96) Fructosamine (µmol/l) 316 ± 56 (93) Insulin (pmol/l) 112 ± 77 (94)

Baseline

SSD diet Treatment

8.9 ± 3.2 (92)* 11.4 ± 3.2 (101) 9.7 ± 3.0 (100)* 7.8 ± 1.5 (92)* 8.9 ± 1.6 (101) 8.2 ± 1.6 (100)* 289 ± 59 (87)* 327 ± 64 (97) 306 ± 72 (94)* 91 ± 53 (88)† 127 ± 122 (97) 103 ± 73 (97)†

Data are means ± SD (n). Sample sizes differ because of variations in specimen quality or quantity. *P 0.0001, †P 0.05 within diet groups for treatment vs. baseline.

weight loss goals were 6.2 ± 0.3 kg in the CCNW and 7.2 ± 0.3 kg in the SSD group. With treatment, body weight and BMI were significantly reduced in both diet groups (P 0.0001 for both). Average weight losses were 3.4 ± 3.1 kg in the CCNW group and 2.9 ± 2.8 kg in the SSD group. The percent of goal weight loss achieved was significantly higher in the CCNW group (68 ± 89%) than in the SSD group (44 ± 46%) (P = 0.02). Reductions in BMI were 1.2 ± 1.1 and 1.0 ± 1.0 in the CCNW and SSD groups, respectively. The effects of the diets on parameters of glycemic control are shown in Table 2. With the diets, significant overall reductions in fasting plasma glucose, HbA1c, and fructosamine (P 0.0001), as well as in fasting serum insulin (P 0.001), were observed. In the CCNW and SSD groups, respectively, these reductions amounted to: 1.4 ± 2.9 and 1.8 ± 3.0 mmol/l for glucose, 0.9 ± 1.0 and 0.7 ± 1.0% for HbA1c, 28 ± 49 and 22 ± 53 µmol/l for fructosamine, and 18 ± 7 and 27 ± 10 pmol/l for insulin. Improvements in glycemic control parameters did not differ significantly in the two diet groups. Plasma lipid parameters are shown in Table 3. Plasma triglycerides were reduced in both diet groups, although this reduction was statistically significant only in the SSD group (P 0.01). The diet interventions were associated with significant overall reductions in total (P 0.0001), as well as in LDL and HDL (P 0.001), cholesterol levels. Total cholesterol levels were reduced 0.19 ± 0.65 mmol/l in the CCNW group and 0.26 ± 0.58 mmol/l in the SSD group. LDL cholesterol levels were reduced by 0.12 ± 0.50 and 0.14 ± 0.45 in the CCNW and SSD groups, respectively; HDL cholesterol decreased by 0.02 ± 0.09 in both groups. Changes in lipid parameters were not significantly different between diet groups. 194

At baseline, sitting systolic and diastolic blood pressures were 132.9 ± 15.0 and 82.1 ± 8.3 mmHg in the CCNW group and 135.3 ± 11.8 and 84.0 ± 7.8 mmHg in the SSD group. There were significant overall reductions in systolic and diastolic blood pressures (P 0.0001 for both); changes in blood pressures were similar in the two diet groups. Systolic blood pressure was reduced 5.7 ± 8.6 mmHg in the CCNW group and 4.5 ± 8.7 mmHg in the SSD group; for diastolic blood pressure, these reductions were 4.2 ± 5.2 and 3.1 ± 4.6 mmHg, respectively. CONCLUSIONS — Nutrition therapy is an essential component of diabetes management, but is often difficult to implement. The present report compares the clinical effects in subjects with type 2 diabetes of a 10-week nutritional intervention with a comprehensive prepackaged meal plan, incorporating current overall recommendations of the American Diabetes Association and other national health organizations, with those of a self-selected diet based on exchange lists. Subjects on both nutrition plans showed weight loss and improvements in diet composition, glycemic control, plasma lipid parameters, and blood

pressure. These results thus reaffirm the utility of nutrition therapy as recommended by the American Diabetes Association in improving glycemic control and cardiovascular risk factors in individuals with type 2 diabetes. Clinical improvements in subjects consuming the prepackaged meal plan, which eliminates the complexities of meal planning and preparation, were generally similar to those observed in subjects in the self-selected diet group. Analysis of food records indicated that subjects on both diet plans significantly reduced daily caloric intake and made multiple nutritional adjustments consistent with current health-promoting recommendations. The decrease in reported caloric intake was significantly greater in the SSD group than in the CCNW group. While increases in the percentage of calories consumed as carbohydrates were equivalent in the two groups, decreases in the percentages of calories consumed as total and saturated fat were significantly greater, and the decrease in cholesterol intake tended to be greater in the CCNW group. In fact, the reported intakes of significantly more subjects on the CCNW than on the selfselected diet met both the step 1 (97 vs. 62%, respectively; P 0.0001) and the more stringent step 2 (76 vs. 31%; P 0.0001) recommendations of the NCEP and the AHA for dietary fat reduction. Finally, the CCNW group reported significantly greater improvements in fiber and sodium intakes. On average, study participants were obese (BMI of 32), as are the majority of individuals with type 2 diabetes. Significant weight loss, amounting to 3 kg, was observed in both diet groups. Despite the greater reduction in caloric intake reported by the SSD group, average weight losses in the groups were equivalent, and the CCNW group achieved a higher percentage

Table 3—Plasma lipids CCNW diet Baseline Treatment n Triglycerides Cholesterol VLDL cholesterol HDL cholesterol LDL cholesterol

100 2.65 ± 1.73 5.64 ± 0.85 1.22 ± 0.79 1.05 ± 0.29 3.36 ± 0.94

Baseline

95 2.59 ± 2.29 5.44 ± 0.88* 1.19 ± 1.05 1.02 ± 0.29‡ 3.26 ± 0.85‡

Data are means ± SD and are given in millimoles per liter. *P groups for treatment vs. baseline.

SSD diet Treatment

102 2.39 ± 1.54 5.37 ± 0.81 1.10 ± 0.71 1.04 ± 0.27 3.22 ± 0.74

0.01, †P

0.001, ‡P

100 2.17 ± 1.44* 5.11 ± 0.77† 0.99 ± 0.66* 1.01 ± 0.24‡ 3.09 ± 0.72* 0.05 within diet

DIABETES CARE, VOLUME 22, NUMBER 2, FEBRUARY 1999

Pi-Sunyer and Associates

of its weight loss goal. The basis of this discrepancy between weight loss and reported reduction in caloric intake is unclear. However, as suggested in an earlier publication in this series (17), a differential bias in reporting nutrient intakes may have existed between the two diet groups. In the CCNW group, the provision of foods in predetermined quantities may have facilitated more accurate recording of actual intake. The SSD group, in contrast, may have underreported intake, as often occurs in free-living subjects who select their own foods (24). Hyperglycemia improved in both diet groups, as evidenced by significant reductions in the short-term parameter of fasting plasma glucose and the longer-term parameters of fructosamine and HbA1c. In concert with these glycemic improvements, both groups showed significant reductions in fasting serum insulin. It was also noted that six subjects in the CCNW group and three in the SSD group discontinued oral hypoglycemic medication, while eight subjects in the CCNW group and five in the SSD group had the dosage of their oral hypoglycemic medication reduced. In contrast, only two subjects in the CCNW group and four in the SSD group had the dosage of their oral hypoglycemic medication increased. Average baseline lipid values in study subjects were generally in the acceptable or borderline risk categories (25). The importance of improving lipid abnormalities in patients with type 2 diabetes is underscored by their increased risk of and increased mortality and morbidity from cardiovascular disease (1,26,27). While cholesterol levels in type 2 diabetic patients do not differ from those of the general population (3), type 2 diabetic patients have a threefold greater risk of coronary heart disease at every cholesterol level than do individuals without diabetes (27). In the present study, both diet groups showed significant reductions in total serum and LDL cholesterol levels. Two subjects in the CCNW group discontinued lipid-lowering medication; no changes in lipid-lowering medication occurred in the SSD group. Although both groups also showed significant reductions in HDL cholesterol, this result might be expected on the basis of the observed reduction in total cholesterol (17). Moreover, other studies indicate that effects on HDL cholesterol levels are influenced by the amount of weight loss and the duration of weight maintenance (28,29). The present intervention may have been of

too short a duration to effect an increase in HDL cholesterol levels. On average, baseline systolic blood pressures were in the high-normal range and diastolic blood pressures were normal (30). Since hypertension compounds morbidity and mortality in type 2 diabetes (30,31), its early detection and treatment are recommended (30). With the diets, significant reductions in blood pressures were observed and average systolic blood pressures approximated the value ( 130 mmHg) recommended for patients with diabetes (30). Three subjects in the CCNW and one in the SSD group discontinued an antihypertensive medication. Reductions in the dosage of antihypertensive medication were observed in two CCNW subjects. Increases in dosage were observed in two CCNW and three SSD subjects. The optimal diet composition for the management of type 2 diabetes has been a subject of controversy (5). Until recently, dietary recommendations have emphasized a reduction of calories from total and saturated fat and their replacement with calories from complex carbohydrates (32). The rationale underlying this approach has been that lower-fat diets will reduce cardiovascular risk by reducing total and LDL cholesterol levels. However, there has been concern that such diets may impair glucose tolerance, increase triglyceride levels, and decrease HDL cholesterol levels (33,34). In the present study, consumption of a high-carbohydrate low-fat diet was associated with improved glycemic control. As noted above, lipid levels of study participants were in the acceptable or borderline range. Nevertheless, total and LDL cholesterol levels were significantly reduced. There was no evidence of elevation of triglycerides on the prescribed high-carbohydrate low-fat diets; instead, triglyceride levels tended to improve. As explained above, the small but significant decrease of HDL cholesterol levels observed may be attributed both to the overall reduction of cholesterol levels and the relatively short duration of the diet intervention. Although diet remains the first-line treatment for type 2 diabetes and its associated disorders of dyslipidemia and hypertension, its effectiveness is often limited by difficulties in maintaining compliance. Compliance with diet may be improved by the provision of foods (35), as implemented by the prepared meal plan. The results of the present report demonstrate that in a 10-week trial, the

DIABETES CARE, VOLUME 22, NUMBER 2, FEBRUARY 1999

prepared meal plan was generally as effective as a self-selected exchange-list–based diet for the management of type 2 diabetes and its associated cardiovascular risk factors. The similar clinical effectiveness of the diet interventions was undoubtedly related to the fact that the two interventions were based on the same nutritional prescription and, thus, the finding that the dietary changes observed in the two treatment groups were generally similar in direction. The present results support the conclusion that the comprehensive meal plan could represent an equally effective but simpler and more convenient alternative to meeting the multiple nutritional demands of type 2 diabetes management. In addition, these results support the utility of diets based on the overall recommendations of the American Diabetes Association in improving glycemic control and cardiovascular risk factors in individuals with type 2 diabetes.

Acknowledgments — This study was supported by the Campbell’s Center for Nutrition and Wellness, Campbell Soup Company, Camden, New Jersey. These data have previously been published in abstract form as Pi-Sunyer FX, McCarron DA: Dietary management of NIDDM: Vanguard Study. Diabetes47 (Suppl. 1):A312, 1998. Participating institutions and individuals include the following: Oregon Health Sciences University, Portland: David McCarron, MD, Cynthia Morris, PhD, Jill Metz, PhD, Daniel Hatton, PhD, Jean-Baptiste Roullet, PhD, Grace McMillan, MN, Patricia Rufolo, RD, Peggy Cook, Geoffrey Snyder, MS, Scott Holcomb, MS, and Patricia Bartholomew; University of Alabama, Birmingham: Suzanne Oparil, MD, Alena Hammond, RN, and Sara Hood, MS; University of California, Davis: Judith Stern, ScD, Mary Haan, MD, Sharon Wesley, and Shannan Young, RD; Wayne State University, Detroit, Michigan: Lawrence Resnick, MD, M. Cathy Krempasky, and Janice Roszler, MS; St. Luke’s-Roosevelt Hospital Center, Columbia University, New York: F. Xavier Pi-Sunyer, MD, Nicholas H.E. Mezitis, MD, Carol A. Maggio, PhD, Eleanor Bauza, and Nancy Aronoff, RD; McMaster University and Hamilton Hospital, Hamilton, Ontario, Canada: R. Brian Haynes, MD, PhD, Ann McKibbon, MLS, Mary Ann Pim, RN, Monica Gray, RD, George Sweeney, MD, Matthew McQueen, MD, and Theresa Mahoney, RD. Pennsylvania State University, University Park: Penny Kris-Etherton, PhD, RD, William Evans, PhD, Jan Ulbrecht, MD, Nina Pletcher, RD, Uma Srinath, MS, Kristin Moriarty, MS, RD, and Satya Jonnalagadda, PhD; 195

Prepared meal plan in type 2 diabetes University of Washington, Seattle: Alan Chait, MD, David Nevin, MD, and Rebecca Eastgard, MS; and the Clinical Research Group of Oregon, Portland: Sharon Clark, PhD, Margaret McMahon, MN, Cynthia Gaboury, MD, Patricia de Garmo, MPH, Leslie Chester, and Shirley Hackney.

References 1. Geiss LS, Herman WH, Smith PJ: Mortality in non-insulin-dependent diabetes. In Dia betes in America.2nd ed. Harns MI, Cowie CC, Stern MP, Boyko EJ, Reiber GE, Bennett PH, Eds. Washington, DC, U.S. Govt Printing Office, 1995, p. 233–257 (NIH publ. no. 95-1468) 2. Maggio CA, Pi-Sunyer FX: The prevention and treatment of obesity: application to type 2 diabetes (Technical Review). Diabetes Care 20:1744–1766, 1997 3. Stern MP, Patterson JK, Haffner SM, Hazuda HP, Mitchell BD: Lack of awareness and treatment of hyperlipidemia in type II diabetes in a community survey. JAMA 262:360–364, 1989 4. National Center for Health Statistics: Plan and Operation of the Second National Health Interview Survey 1976–80. In Vital and Health Statistics.Washington, DC, U.S. Govt Printing Office, 1981 (DHHS publ. no. PHS 81–1317) 5. Franz MJ, Horton ES, Bantle JP, Beege CA, Brunzell JD, Coulston AM, Henry RR, Hoogwerf BJ, Stacpoole PW: Nutrition principles for the management of diabetes and related complications (Technical Review). Diabetes Care17:490–518, 1994 6. Albu J, Konnarides C, Pi-Sunyer FX: Weight control: metabolic and cardiovascular effects. Diabetes Rev3:335–347, 1995 7. West KM: Diet therapy of diabetes: an analysis of failure. Ann Intern Med 79:425–434, 1973 8. Close EJ, Wiles PG, Lockton JA, Walmsley D, Oldham J, Wales JK: Diabetic diets and nutritional recommendations: what happens in real life? Diabet Med 9:181–188, 1992 9. Milne RM, Mann JI, Chisolm AW, Williams SM: Long-term comparison of three dietary prescriptions in the treatment of NIDDM. Diabetes Care17:74–80, 1994 10. American Diabetes Association: Nutritional recommendations and principles for individuals with diabetes mellitus (Position Statement). Diabetes Care 17:519–522, 1994 11. Krauss RM, Deckelbaum RJ, Ernst N, Fisher E, Howard BV, Knopp RH, Kotchen T, Lichtenstein AH, McGill HC, Pearson 196

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

TA, Prewitt TE, Stone NJ, Van Horn L, Weinberg R: Dietary guidelines for healthy American adults: a statement for health professionals from the Nutrition Committee, American Heart Association. Circulation 94:1795–1800, 1996 National Cholesterol Education Program: Report of the Expert Panel on Population Strategies for Blood Cholesterol Reduction: executive summary. Arch Intern Med 151:1071–1084, 1991 National Research Council (US) Subcommittee on the Tenth Edition of the RDAs: Recommended Dietary Allowances . Washington, DC, National Academy Press, 1989 McCarron DA, Oparil S, Chait A, Haynes B, Kris-Etherton P, Stern JS, Resnick LM, Clark S, Morris CD, Hatton DC, Metz JA, McMahon M, Holcomb S, Synder GW, Pi-Sunyer FX: Nutritional management of cardiovascular risk factors: a randomized clinical trial. Arch Intern Med157:169–177, 1997 American Dietetic Association and American Diabetes Association: Exchange Lists for Weight Management.Chicago, American Dietetic Association, 1989 Harris HA, Benedict FG: A Biometric Study of Basal Metabolism. Washington, DC, Carnegie Institute, 1919 (Publication 279) Metz JA, Kris-Etherton PM, Morris CD, Mustad VA, Stern JS, Oparil S, Chait A, Haynes RB, Resnick LM, Clark S, Hatton DC, McMahon M, Holcomb S, Synder GW, Pi-Sunyer FX, McCarron DA: Dietary compliance and cardiovascular risk reduction with a prepared meal plan compared to a self-selected diet. Am J Clin Nutr 66:373–385, 1997 Monk A, Barry B, McClain K, Weaver T, Cooper N, Franz MH: Practice guidelines for medical nutrition therapy provided by dietitians for persons with non-insulindependent diabetes mellitus. J Am Diet Assoc95:999–1006, 1995 Albers JJ, Segrest JP, Eds.: Plasma lipoproteins, part B: characterization, cell biology, and metabolism. Methods in Enzymology . New York, Academic Press, 1986, p. 101–123 Warnick GR, Benderson J, Albers JJ: Dextran sulfate-Mg2 precipitation procedure for quantification of high-density lipoprotein cholesterol. Clin Chem28:1379–1388, 1982 Friedewald WT, Levy RI, Fredrickson DS: Estimation of the concentration of lowdensity lipoprotein cholesterol in plasma, without the use of the preparative ultracentrifuge. Clin Chem18:499–502, 1972 Perloff D, Grim C, Flack J, Frohlich ED, Hill M, McDonald M, Morgenstern BZ: Human

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

blood pressure determination by sphygmomanometry. Circulation88:2460–2470, 1993 Hatton DC, Haynes RB, Oparil S, KrisEtherton P, Pi-Sunyer FX, Resnick LM, Stern JS, Clark S, McMahon M, Morris C, Metz J, Ward A, Holcomb S, McCarron DA: Improved quality of life in patients with generalized cardiovascular metabolic disease on a prepared diet. Am J Clin Nutr 64:935–943, 1996 Mertz W, Tsui JC, Judd JT, Reiser S, Hallfrisch J, Morris ER, Steele PD, Lashley E: What are people really eating? The relation between energy intake derived from estimated diet records and intake determined to maintain body weight. Am J Clin Nutr 54:291–295, 1991 American Diabetes Association: Detection and management of lipid disorders in diabetes (Consensus Statement). Diabetes Care 16:828–834, 1993 Garcia MJ, McNamara PM, Gordon T, Kannell WB: Morbidity and mortality in diabetics in the Framingham population: sixteen year follow-up study. Diabetes 23:105–111, 1974 Stamler J, Vaccaro O, Neaton JD, Wentworth D for the Multiple Risk Factor Intervention Trial Research Group: Diabetes, other risk factors and 12-yr cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care 16:434–444, 1993 Uusitupa MIJ, Laakso M, Sarlund H, Majander H, Takala J, Penttila I: Effects of a very-low calorie diet on metabolic control and cardiovascular risk factors in the treatment of obese non-insulin-dependent diabetics. Am J Clin Nutr51:768–773, 1990 Friedman CI, Falko JM, Patet ST, Kim MH, Newman HAI, Barrows H: Serum lipoprotein response during active and stable weight reduction in reproductive obese females. J Clin Endocrinol Metab 55:258–262, 1982 American Diabetes Association: Treatment of hypertension in diabetes (Consensus Statement). Diabetes Care16:1394–1401, 1993 National High Blood Pressure Education Program Working Group: Report on hypertension in diabetes. Hypertension23:145–158, 1994 American Diabetes Association: Nutritional recommendations and principles for individuals with diabetes mellitus 1986. Dia betes Care10:126–132, 1987 Coulston AM, Hollenbeck CB, Swislock ALM, Reaven GM: Persistence of hypertriglyceridemic effect of low fat, high carbohydrate diets in NIDDM patients. Diabetes Care12:94–101, 1989 Garg A: High-monounsaturated fat diet for

DIABETES CARE, VOLUME 22, NUMBER 2, FEBRUARY 1999

Pi-Sunyer and Associates diabetic patients: is it time to change the current dietary recommendations? Diabetes Care 17:242–246, 1994

35. Jeffery RW, Wing RR, Thorson C, Burton LR, Raether C, Harvey J, Mullen M: Strengthening behavioral interventions for

DIABETES CARE, VOLUME 22, NUMBER 2, FEBRUARY 1999

weight loss: a randomized trial of food provision and monetary incentives. J Consult Clin Psychol61:1038–1045, 1993

197