Glycosuria

ABSTRACT. Background: Type 2 diabetic patients may need enteral nutrition support as part of their treatment. The objective was to compare glycemie and lipid control in hospitalized patients with type 2 diabetes requiring feeding via nasogastric tube using enteral feedings with either a highcarbohydrate or a high-monounsaturated-fat content. Methods: This trial included type 2 diabetes patients admitted to the hospital for neurologic disorders or head and neck cancer surgery who received either a low-carbohydrate-high-monounsaturated-fat (Glucerna) or a high-carbohydrate diet (Precitene Diabet). Glycemic and lipid control was determined weekly. Safety and gastrointestinal tolerance were also assessed. Results: A total of 104 patients were randomized and 63 were evaluable according to preestablished protocol criteria. Median duration of therapy was 13 days in both groups. Mean glucose was significantly increased at 7 days of treatment (p = .006) in the Precitene arm, with no significant variations in the Glucerna arm. Mean weekly blood triglycerides levels in the Precitene arm were increased without reaching statistical significance, whereas patients in the Glucerna arm showed a stable trend. Patients in the Precitene arm showed a significantly higher incidence of diarrhea than patients in Glucerna arm (p = .008), whereas the incidence of nausea was smaller in the Precitene arm than in the Glucerna arm (p = .03). Conclusions: An enteral formula with lower carbohydrate and higher monounsaturated fat (Glucerna) has a neutral effect on glycemic control and lipid metabolism in type 2 diabetic patients compared with a highcarbohydrate and a lower-fat formula (Precitene Diabet). (Journal of Parenteral and Enteral Nutrition 29:21-29, 2005)

During the course of their disease, diabetic patients may need artificial nutrition support via enteral or parenteral routes to maintain an adequate nutritional status. When these patients receive enteral nutrition (EN), glycemic control may be difficult and, if inadequate, may result in hyperglycemia, glycosuria with osmotic diuresis, and even may lead to the hyperosmolar syndrome.1 In addition to hyperglycemia, diabetes mellitus type 2 patients may have lipid, immune system, and endothelial abnormalities.

The glucose response of diabetic patients receiving EN can be affected by a series of factors, of which the most important are total caloric intake, composition of the nutritional formula administered, amount given, route and method of administration (bolus or continuous, gravity fed or infusion pump) and, in the case of bolus administration, the number of daily boluses.2 Among factors relating to the nutritional formula, the presence of fiber in the diet has not shown a significant effect on short-term glucose response in diabetic patients3; however, 1 study in patients with type 2 diabetes has demonstrated that enterai formulas enriched with fiber and fructose result in lower postprandial glucose response and decreased area under the insulin curve after a trial breakfast.4

Low-carbohydrate (CH;

Diabetic patients may have abnormalities in plasma lipids and an increased risk of cardiovascular disease.11 Scientific societies have published nutritional recommendations for lipid control, because macronutrient distribution in the diet has a clear impact on lipid profiles.12-14 High-monounsaturated-fat diets tend to produce lower triglycerides and very-low-density lipoprotein (VLDL) cholesterol levels, and higher concentrations of high-density lipoprotein (HDL) cholesterol, when compared with high-CH diets. Total cholesterol and low-density lipoprotein (LDL) cholesterol are no different between the two types of diets.15,16 However, in a similar study comparing both types of diets, no differences in lipid profiles were found, but 24-hour systolic and diastolic blood pressures were found to be lower with a high-fat diet.

All these trials tested the effects of solid kitchen diets. Nevertheless there have been several studies in diabetic patients treated with EN, and those that have been conducted were usually short-term studies where patients were given a test breakfast.3,9,17-20 Most of these studies rely on a breakfast test, and information on more prolonged nutrition is very scarce.21,22 Therefore, a study on the medium- and long-term effects of both types of enterai feedings on glycemic control and on plasma lipid levels is needed.

The primary objective of this study was to compare control of glycemia and plasma lipid levels, and units of insulin necessary to achieve it, in hospitalized type 2 diabetic patients fed 2 enterai feeding formulas that are specifically indicated for diabetic patients but differ in their composition: a high-fat diet (50% fat and 33.3% CH enriched with MUFAs) and a high-CH diet (31% fat and 54% CH based on starch and fructose instead of maltodextrin and sucrose as in the standard enterai formula for nondiabetic patients). Gastrointestinal tolerance of the 2 feeding formulas was assessed, and safety profiles were based on laboratory tests and monitoring of adverse reactions in each treatment group.

MATERIALS AND METHODS

Study Design

A phase IV, multicenter, comparative, randomized, open-label study in 2 parallel groups was designed to evaluate the effect of the 2 enterai feeding formulas on metabolic control (glycemia and plasma lipid levels). The study included patients admitted to the hospital for a neurologic disorder or head or neck cancer who had a documented diagnosis of type 2 diabetes mellitus requiring at the time of recruitment in the study at least 14 consecutive days of EN (continuous feeding by infusion pump) as their sole nutritional intake.

Patients were considered evaluable if they received at least 75% of their requirements during 8 consecutive days. To be eligible for the study, patients had to (a) be ≥18 years but 2 mg/dL; (c) had clinical contraindications for EN support; (d) had a triglyceride value of 400 mg/dL or greater on the baseline laboratory test of the study; (e) had received steroids within 24 hours of enrollment in the study; and (f) had received parenteral nutrition with fat or > 150 g/d of glucose within 1 week of enrollment in the study.

Conduct of the Study

After explaining the nature of the study, the patient or, in case of incapacity, the patient's legal representative, was asked to give consent to participate in the study and to sign the informed consent form. Approval by the ethics committee of each participating center was obtained before starting with the study in any center. This study was conducted in accordance with the recommendations of the Declaration of Helsinki for clinical trials in human subjects.23 Subject selection could be started from the time when participating subjects were admitted to the hospital. Subjects were enrolled in the study after verifying that they met eligibility criteria and signed informed consent had been obtained. Then, patients were randomized to one or the other study arms. Randomization was performed per center in blocks of 4 patients to guarantee proper balance of both treatment arms per study site. From study day 1 to study completion (2 weeks), patients were fed the assigned study formula as their sole nutritional intake: a high-fat, high-monounsaturated-fat diet (16.7% protein, 50% fat, and 33.3% CH, Glucerna, Abbott Laboratories, SA) or a high-CH diet (15% protein, 31% fat, and 54% CH; Precitene Diabet, Laboratorios Novartis, SA). The quantitative composition of these formulas is shown in Table I.

The patient started to receive nutrition support within 24 hours after signing the informed consent form. EN was given via a nasoenteral feeding tube and continuous infusion pump at a rate of 50 mL/h, which was gradually increased in 25 mL every 8-12 hours until the caloric needs of the patient were met. In all patients, at least 75% of the target caloric intake was reached within 72 hours of starting enterai feeding. The infusion rate and volume of the study formula and water requirements were adjusted according to the patient's status and tolerance of the study treatment. On the first visit, the patient's caloric requirements (kcal/day) were calculated using the Harris-Benedict formula adjusted by a single stress factor estimated to be 1.3 (target caloric intake = REE X 1.3).

Patients participating in the study were not allowed to receive vitamin or mineral supplementation during the study period, and administration of hypolipidemic drugs, steroids or oral hypoglycemic agents was considered a reason for immediate removal of the patient from the study. However, patients were allowed to receive any other medication that the physician deemed indispensable for their health and well-being. Patients received either subcutaneous or IV insulin doses with the goal of maintaining blood glucose levels between 60 and 200 mg/dL. This study was designed before a more stringent glycemic control was found to be beneficial in critically ill patients24; besides, our patients did not need treatment in an ICU.

Because existing studies similar to the present study differed in design or in the doses of the formulas used, sample size predetermination was performed according to Cohen's model,25 which substitutes the mean and SD of a population for an effect (dt). Because large differences were not expected in blood glucose levels between the 2 nutritional interventions, a relatively small dt effect of 0.30 was chosen with a significance level of 5% (α error

Outcome Variables

The primary variables for the study were mean weekly blood glucose level (weekly average of mean daily blood glucose values obtained from 3 daily capillary glucose measurements), mean weekly triglyceride level, and mean total daily units of insulin administered per patient and averaged weekly. Also, mean daily insulin dose per kilogram and daily calories per unit of insulin are calculated.

The secondary variables for the study were the following: mean urinary glucose and ketone levels obtained using a reagent strip and averaged weekly, fructosamine and glycosylated hemoglobin levels, and levels of cholesterol and its fractions (HDL, LDL, and the cholesterol: HDL ratio). All measurements were performed in the laboratory at each study site. The gastrointestinal tolerance and safety profile of both feeding formulas were also evaluated. The symptoms used to assess gastrointestinal tolerance were diarrhea, nausea, dyspepsia, constipation, and vomiting. The number of days the patient experienced the symptom was recorded along with its severity, graded as mild, moderate, or severe.

Patient data were collected weekly on the 3 study visits on days 1 (visit 1), 8 (visit 2), and 15 (visit 3). However, if a patient was withdrawn from the study, an end-of-treatment (EOT) visit was performed at that time. Therefore, there are some patients for whom visit 2 was the EOT visit (if they remained in the study for only 8 days) and others for whom visit 3 was the EOT visit (if they remained in the study from 8 to 15 days). Because of the nature of the study, no imputation method was used for missing data. Body mass index (BMI) was calculated using the following formula: BMI = weight (kg)Theight (m2).26

Statistical Analysis

Statistical analysis included a baseline descriptive statistic for the overall study population and by groups (patients grouped by the formula they received). Goodness of fit to a Gaussian distribution was analyzed using the Kolmogorov-Smirnov test for all continuous variables and symmetry statistics (skewness and kurtosis). For qualitative variables, absolute frequencies and percentages for the categories of each variable were computed. Baseline quantitative variables were compared between groups (baseline homogeneity) using Student's t test or Mann-Whitney U test, depending on whether the variables are distributed normally or non-normally, respectively. Qualitative variables were compared using Fisher exact test or Pearson χ2 test.

Both primary and secondary variables were analyzed using an analysis of variance with 2 factors (type of diet and duration of treatment), with center as covariable, one of them with repeated measures with point comparison between baseline levels and 8 or 15 days of treatment. Student's test corrected using the Bonferroni method was used for post hoc comparisons between groups and between treatment visits. For nonnormally distributed variables, the Mann-Whitney and Wilcoxon tests were applied, in both cases using Bonferroni's correction for the number of comparisons performed. Naturally ordered qualitative variables (results of the analysis of reagent strips) were analyzed using the Mann-Whitney U test for differences between study diets on each visit, with Bonferroni's correction and Friedman's test for differences between study visits within the same treatment group. Unordered qualitative variables (incidence of adverse events, tolerance of study diets, etc) were tabulated and analyzed using general contingency tests (Fisher's exact test for 2 X 2 tables and Pearson χ2 test for 2 X 2 tables). A 95% confidence interval was calculated for all comparisons.

All tests used in the analysis of study data were performed considering a maximum α error of 5% (p

RESULTS

Patient Evaluability

A total of 108 hospitalized type 2 diabetic patients were enrolled in the study; 104 were randomized and are evaluable for safety; 51 in the Precitene arm (PA) and 53 in the Glucerna arm (GA). Three out of 108 withdrew the written consent before randomization, and the other was lost to follow-up. Twenty-six (49%) in the PA and 15 (29%) in the GA were excluded from the efficacy evaluation because of criteria not fulfilled or caloric intake

Patients who completed 2 weeks of treatment were 14 (26%) in the PA and 18 (35%) in the GA (see Fig. 1). No significant differences between study groups were observed in premature discontinuation (p = .279). Reasons for not completing the study were the following: 13 serious adverse events (included 12 deaths; 5 in the PA and 7 in the GA), 1 lost to follow-up (in the PA), 6 protocol violations (1 in the PA and 5 in the GA), and 11 patients left the study because of healing (deglutition recovery) or investigator decision (6 in the PA and 5 in the GA).

The median duration of treatment in evaluable patients was similar in both groups: 13 days (range, 8-33) in the PA and 13.5 days (range, 8-27) in the GA, p = .518. Mean hospital stay, although not directly related to the study diets, showed a trend toward less hospital days in the GA group (45.5 ± 29.9 days [range, 18-109] in the PA and 32.3 ± 12.3 days [range, 15-58] in the GA; p = .084).

Characteristics of the Patients

Baseline demographic and metabolic characteristics of the evaluable patients in each enteral feeding group were not significantly different (Table II). Nutritional parameters during 2 weeks of therapy were similar from a statistically standpoint in the 2 study groups; mean daily caloric intake and dose of insulin showed no differences during the first week or the second (Table III). Also, mean daily dose of insulin per kilogram and daily calories per unit of insulin administered were similar in both groups. Nutritional parameters were slightly higher during the second week in both treatment arms, except for daily calories per unit of insulin, but differences were not statistically significant.

Primary and Secondary Variables

Table IV shows changes and evolution in the primary and secondary study variables according to the formula received. Mean blood glucose was significantly increased in the PA group at visit 2 (7 days of treatment, p = .006) by 21% (11% to 30%, 95% confident interval), but not in the GA group (there were no differences between groups). No statistically significant differences were observed at visit 3 (14 days), nor interor intragroup comparisons, for this primary endpoint. Mean weekly blood triglyceride levels in patients in the PA changed from a baseline value of 138 mg/dL (109165) to 150 mg/dL (107-192) on the last visit, showing a trend toward statistical significance (p = .075), whereas patients in the GA showed a stable trend in triglyceride levels during the study from a baseline value of 126 mg/dL (111-141) to 137 mg/dL (102-171) on the final visit (no significant differences; see Table IV).

Neither formula evaluated in this study showed statistically significant changes for other lipid parameters (total cholesterol, HDL-c or LDL-c), except for total cholesterol/HDL cholesterol ratio, which showed an increase close to a significant level (p = .051) at visit 3 in patients in the GA (Table IV).

Changes in urinary glucose and ketone levels, although they were lower at the end of the study, did not reach statistical significance because of the high variability in patient response to treatment. Patients in the PA showed a trend toward more days with urinary glucose levels than patients in the GA, particularly during the second week of study. On average, patients showed a glycosuria with a positive reagent strip of 0.22 days in the PA compared with 0.06 days in the GA (p = .088).

Glycosylated hemoglobin and fructosamine levels did not show significant changes with either of the study formulas. Likewise, anthropometric variables did not show any statistical differences from baseline to the end of the study.

Adverse Events

Adverse events with an incidence ≥ 2% are listed in Table V, whether or not they were considered related to the formula received by the patient. Patients in the PA showed a significantly higher incidence of diarrhea (all causes) than patients in the GA (19 [11.1%] vs 6 [3.3%] events, respectively, p = .008). Diarrhea was defined as 3 or more stools per day for at least 2 consecutive days.27 The incidence of nausea was smaller in the PA than in the GA (0 vs 6 [3.3%] events, respectively, p = .03).

The incidence of other adverse events was similar in both groups. Table V also shows the distribution of adverse events by severity and seriousness in each study group; patients in the PA showed a smaller incidence of moderate adverse events than patients in the GA, but incidence of severe adverse events was higher in the PA than in the GA, showing a trend toward statistical significance (p = .054). Incidence of serious adverse events was the same in both study groups, 23% of all events. In 28% of adverse events in the PA and in 24% of those in the GA, the investigator established a probable or possible causal relationship of the adverse event to the formula received (no significant differences, p = .449).

DISCUSSION

Until 15 years ago, most guidelines for nutrition therapy of diabetes mellitus recommended a limited fat intake (30% of total calories) combined with a high supply of complex carbohydrates (55%-60% of total calories). Current nutritional treatment of the diabetic patient also includes the partial replacement of carbohydrates by monounsaturated fat.12

Metabolic control in patients with diabetes and other forms of hyperglycemia is difficult to achieve with standard enteral formulas.28,29 Landmark studies in EN have demonstrated that liquid formulas produce faster gastric emptying and absorption, higher serum glucose levels, and greater insulin responses than do solid-food diets.29 Currently available enteral formulas differ significantly with respect to caloric content, osmolarity, nutrient composition, relative concentrations of macronutrients, vitamins, etc. Although standard enteral formulas are suitable for patients who have no metabolic or physiologic abnormalities, standard formulas can increase the complexity of feeding and potential for complications in patients who have altered metabolic requirements or tolerances.30 Current international dietary guidelines recommend modifying the relative contribution of fat and CH in the diet to adjust for metabolic control, to improve cardiovascular risk factors, and because of personal preference. Guidelines also emphasize the utility of MUFA-enriched diets as an alternative to a CH-rich diet for type 1 and type 2 diabetes.29,31

Our study compared 2 liquid formulas that have been designed for diabetic patients and that were given as full EN through a nasogastric tube for a minimum of 8 days in hospitalized neurologic or head and neck cancer patients. The composition of both formulas currently available in the market has changed since the study was performed. However, our study focused on the comparison of 2 different concepts in the design of enterai formulas for diabetic patients more than the comparison of 2 particular commercial products. We consider that the results of the present study may be clinically relevant in our setting both in terms of the results obtained and the duration of patient follow-up, which was longer than in previous studies.32,33

We tested the hypothesis that a low-CH with a highMUFA-content enteral formula for type 2 diabetic patients would have some metabolic advantages in diabetic control with respect to a high-carbohydrate and low-fat enteral formula in which starch and fructose have replaced maltodextrins and sucrose. This hypothesis was based on the fact that low-CH solid diets lower plasma glucose levels and requirements of insulin and oral hypoglycemic in type 2 diabetic patients.5 In contrast, a very-high-CH diet may increase postprandial glycemia and 24-hour urinary glucose excretion in poorly controlled diabetic patients.7

We partially confirmed our hypothesis according to the results of this study, as patients fed with Glucerna (a diet with a low CH but high fat content compared with Precitene, which has a lower fat and higher CH content) showed a neutral effect on weekly blood glucose levels during the study, whereas patients fed with Precitene showed a significant temporary increment in plasma blood glucose. Also, patients fed with Precitene showed a trend toward an increment of mean days with urinary glucose. At the same time, no significant differences in mean insulin consumption between groups were observed during the trial. Also, mean daily insulin dose per kilogram and insulin units per calorie were the same in both groups. Mean insulin dose and caloric intake increased slightly during the second week of treatment compared with the previous week, in both treatment arms, although differences were not of statistical significance.

Although our study has been performed in hospitalized patients and during a long period of time, the results are partially in agreement with those observed by Sanz-Paris et al10 in ambulatory type 2 diabetic patients who received a unique dose of 250 mL of 2 formulas, similar to those we used in our study. These authors concluded that a high-monounsaturated-fat diet may improve glycemic control in type 2 diabetes in comparison with a high-complex-carbohydrate, low-fat formulation. Moreover, Mesejo et al,34 studied during 14 days a group of critically ill patients receiving 2 high-protein enteral formulas that were different in the type and percentage of CH and lipids. These authors suggest that the overall composition of the formula with starch, fructose, MUFA, and soluble fiber contributes to a better glycemic control rather than the total amount of carbohydrates alone, as it has already been demonstrated by other investigators.16,22,35 Both studies are in agreement with results observed in our trial, showing that high-monounsaturated-fat diet does not have a deterioration effect on glycemic control. High-complex-carbohydrate formulas do interfere with glycemic control.

In addition, diabetic patients may have abnormalities in plasma lipids and an increased risk of cardiovascular disease11; therefore, several groups13 recommend decreasing fat intake to

Therefore, another primary objective of our study was to identify changes in lipid parameters. Patients in the GA showed stable triglyceride levels with no significant changes during the study, despite receiving a high-fat diet. However, patients receiving the high-CH and lower fat diet showed a trend toward statistically significant increase in triglyceride levels at the end of the study, which was generally higher than in patients in the GA (although it did not reach a statistical significance). Neither of the 2 formulas evaluated caused significant changes in the other lipid parameters, because the reduction in HDL values observed in patients in the GA did not reach statistical significance.

These results confirm the idea, in agreement with other studies,9,10,15,36 that a high-CH, low-fat diet is accompanied by an elevation of triglycerides, which may increase atherogenic risk.32 Conversely, a highfat, low-CH diet not only appears to improve glycemic control but also is not associated with imbalances in the lipid profile in type 2 diabetic patients. Therefore, it is recommended to add MUFA to replace some of the CH calories, thereby limiting the postprandial rise in serum glucose without adversely affecting other metabolic control parameters.33,37

It should also be pointed out that the change in blood glucose levels experienced by patients in both treatment groups did not translate into a change in plasma glycosylated hemoglobin or fructosamine levels (although the average duration of use of the formulas was too short for differences to be observed). Similarly, the 2 treatments did not appreciably modify the nutritional status of the patients, as measured by anthropometric variables (triceps skinfold thickness and arm muscle circumference). Both formulas had a similar and expected tolerance profile, with similar percentages of severe and serious adverse events, even though patients in the PA showed a significantly higher incidence of diarrhea than those in the GA, which is very interesting, given its lower fat content. On the other hand, nausea was significantly less frequent in patients in the PA than in the GA. Also, incidence of severe adverse events was higher in the PA than in the GA group, showing a trend toward statistical significance.

In our study, the administration of a high-fat high-monounsaturated-fat diet showed a trend toward significantly lower hospital length of stay than feeding with high-complex-carbohydrate, low-fat formulation (32.3 days vs 45.5 days, respectively). Although the study was not powered to demonstrate differences in this variable, nor was an economic evaluation intended to be performed, cost of stay should be analyzed by health decision makers and clinicians when choosing an enteral formula for nutrition of type 2 diabetic patients.

Although the sample size of this study was small and a number of patients discontinued the study before the end of 2 weeks, it can be concluded that an enteral feeding formula with a lower CH content and higher fat content, predominantly of MUFAs, is associated with a neutral effect on control of glycemia and triglycerides in patients with type 2 diabetes mellitus with neurologic disorders or in the postoperative period of head or neck cancer. Moreover it has no adverse effects on other lipid or metabolic variables, and shows better gastrointestinal tolerance.

ACKNOWLEDGMENTS

The authors thank the members of the Cooperative group who are coauthors of the trial: Dr José Chamorro, Hospital Ciudad de Jaén (Jaén); Dr José L. Pereira-Cunill, Hospital Virgen del Rocío (Sevilla); Dr MS Ángeles Martín-Palmero, Hospital Ntra. Sra. de la Candelaria (Tenerife); Dr Roser Trallero, Hospital Parc Tauli (Barcelona); Dr MS José Martínez, Hospital Princesa de España (Jaén); Dr Francisco Javier Ordóñez, Hospital Marqués de Valdecilla (Santander); Dr Pilar García-Peris, Hospital Gregorio Marañón (Madrid); Dr Emma Camarero, Hospital General de Galicia (Santiago de Compostela); Dr Pilar Gómez-Enterría, Hospital Central de Asturias (Oviedo); Dr Lucio Cabrerizo, Hospital Clínico de San Carlos (Madrid); Dr Antonio Pérez-de-la-Cruz, Hospital Virgen de las Nieves (Granada); Dr Carmen Sánchez, Hospital Gral. Universitario (Murcia); Dr Abelardo García-de-Lorenzo, Hospital La Paz (Madrid); Dr MS Nelly Rodríguez and Dr Luis Usán, Abbott Laboratories, Spain.

The study was supported in part by a grant from Abbott Laboratories.

The authors would also like to thank their statistician Ignacio Álvarez MD (Clinic- Data S.L.).

REFERENCES

1. Peters AL, Davidson MB. The challenge of blood glucose management in diabetic patients receiving enteral or parenteral nutrition. JPEN J Parenter Enteral Nutr. 1993;17:1-2.

2. Schafer RC, Bohannon B, Franz M, et al. Translation of the diabetes nutrition recommendations for health care institutions. Diabetes Care. 1997;20:96-105.

3. Peters AL, Davidson MB. Effects of various enteral feeding products on postprandial blood glucose response in patients with type I diabetes. JPEN J Parenter Enteral Nutr. 1992;16:69-74.

4. Golay A, Schnedier H, Bloise D, Vadas L, Assal J. The effect of liquid supplement containing guar gum and fructose on glucose tolerance in non-insulin-dependent diabetic patients. Nutr Metab Cardiovasc Dis. 1995;5:141-148.

5. Simpson HCR, Simpson RW, Lousley S, et al. A high carbohydrate leguminous fibre diet improves all aspects of diabetic control. Lancet. 1981;1:2-5.

6. Campbell LV, Marmot PE, Dyer JA, Borkman M, Storlien LH. The high-monounsaturated fat diet as a practical alternative for NIDDM. Diabetes Care. 1994;17:177-182.

7. Coulston AM, Hollenbeck CB, Swisloci ALM, Chen Y, Reaven GM. Deleterious metabolic effects of high-carbohydrate, sucrose-containing diets in patients with non-insulin-dependent diabetes mellitus. Am J Med. 1987;82:220-231.

8. Rasmussen OW, Thomson C, Hansen KW. Effects of blood pressure, glucose and lipid levels of a high-monounsaturated fat diet compared with a high-carbohydrate diet in NIDDM subjects. Diabetes Care. 1993;16:1565-1571.

9. Garg A, Grundy M, Unger RH. Comparison of effects of high and low carbohydrate diets on plasma lipoproteins and insulin sensitivity in patients with mild NIDDM. Diabetes. 1992;41:1278-1285.

10. Sanz-París A, Calvo L, Guallard A, Salazar I, Albero R. High-fat versus high-carbohydrate enteral formulae: effect on blood glucose, C-peptide, and ketones in patients with type 2 diabetes treated with insulin or sulphonylurea. Nutrition. 1998;14:840-845.

11. Fontbonne A, Eschwege E, Cambien F, et al. Hypertriglyceridemia as a risk factor of coronary heart disease mortality in subjects with impaired glucose tolerance or diabetes. Diabetologia. 1989;32:300-304.

12. Franz MJ, Bantle JP, Beebe CA, et al. Evidence-based nutrition principles and recommendations for the treatment and prevention of diabetes and related complications. Diabetes Care. 2002; 25:148-198.

13. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001;285:2486-2497.

14. Kennedy E, Meyers L, Layden W. The 1995 dietary guidelines for Americans: an overview. J Am Diet Assoc. 1996;96:234-237.

15. Garg A, Bonanome A, Grundy SM, Zhang ZJ, Unger RH. Comparison of a high-carbohydrate diet with a high-monounsaturated-fat diet in patients with non-insulin-dependent diabetes mellitus. N Engl J Med. 1988;319:829-834.

16. Garg A, Bantle J, Henry R, et al. Effects of varying carbohydrate content of diet in patients with non-insulin-dependent diabetes mellitus. JAMA. 1994;271:1421-1428.

17. Thomas BL, Laine DC, Goetz FC. Glucose and insulin response in diabetic subjects: acute effect of carbohydrate level and the addition of soy polysaccharide in defmed-formula diets. Am J Clin Nutr. 1988;48:1048-1052.

18. Harley JR, Pohl SL, Isaac RL. Low carbohydrate with fiber vs high-carbohydrate without fiber enteral formulas: effect on blood glucose excursion in patients with type II diabetes [abstract]. Clin Res. 1989;37:141A.

19. Peters A, Davidson M, Isaac RM. Lack of glucose elevation after simulated tube feeding with a low carbohydrate, high-fat enteral formula in patients with type I diabetes. Am J Med. 1989;87: 178-182.

20. Sturmer W, Kramer H, Schrezenmeir J. Favourable glycaemic effects of a new balance liquid diet for enteral nutrition: results of a short term study in 30 type II diabetic patients. Clin Nutr. 1994;13:221-227.

21. Abbruzzese B, D'Amico A, Green AA, et al. Effect of a low carbohydrate formula on circadian glycemic control and lipids in elderly patients with non-insulin dependent diabetes mellitus (NIDDM) receiving total enteral nutrition support [abstract]. FASEB J. 1993;7:A847.

22. Craig LD, Nicholson S, Silverstone PA, Kennedy RD. Use of a reduced-carbohydrate modified fat enteral formula for improving metabolic control and clinical outcomes in long-term care residents with type II diabetes: results of a pilot trial. Nutrition. 1998;14:529-534.

23. The World Medical Association Ethics Unit. Declaration of Helsinki. Available at: http://www.wma.net/e/policy/pdf717c.pdf. Accessed October 10, 2004.

24. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med. 2001;345: 1359-1367.

25. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.

26. Smalley KJ, Knerr AN, Kendrick ZV, Colliver JA, Owen OE. Reassessment of body mass indices. Am J Clin Nutr. 1990;52: 405-408.

27. Morbarhan S, DeMeo M. Diarrhea induced by enteral feeding. Nutr Rev. 1995;53:67-70.

28. Coulston AM. Clinical experience with modified enteral formula with patients with diabetes. Clin Nutr. 1998;17(Suppl 2):S46S56.

29. Schrezenmeir J. Rationale for specialized nutrition support for hyperglycemic patients. Clin Nutr. 1998;17(Suppl 2):S26-S34.

30. Campbell S, Schiller M. Considerations for enteral nutrition support of patients with diabetes. Top Clin Nutr. 1991;7:23-32.

31. Borges VC. Specialized enteral formulae for diabetic patients. Nutrition. 2003;19:196-198.

32. Storlien L, Tapsell L, Calvert G. Diabetic diets: whither goest? Nutrition. 1998;14:865-867.

33. Wright J. Effect of high-carbohydrate vs high-MUFA diet on metabolic control in diabetics and hyperglycemic patients. Clin Nutr. 1998;17(Suppl 2):S35-S45.

34. Mesejo A, Acosta JA, Ortega C, et al. Comparison of a high-protein disease-specific enteral formula with a high-protein enteral formula in hyperglycemic critical ill patients. Clin Nutr. 2003;22:295-305.

35. Coulston AM, Hollenbeck CB, Swislocki LA, Reaven GM. Persistence of hypertriglyceridemic effect of low-fat high-carbohydrate diets in NIDDM patients. Diabet Care. 1989;12:94-101.

36. Garg A. High-MUFA diets for patients with DM: a meta-analysis. Am J Clin Nutr. 1998;67(Suppl 3):577S-582S.

37. Wright J. Total parenteral nutrition and enteral nutrition in diabetes. Curr Opin Clin Nutr Metab Care. 2000;3:5-10.

Miguel León-Sanz, MD, PhD*; Pedro P. García-Luna, MD, PhD[dagger]; Mercé Planas, MD, PhD[double dagger]; Alejandro Sanz-París, MD, PhD§; Carmen Gómez-Candela, MD, PhD||; César Casimiro, MD¶; and the Abbott SPAI-97-004 Study Cooperative Group

From the * Hospital 12 de Octubre, Madrid, Spain; [dagger] Hospital Virgen del Rocío, Sevilla, Spain; [double dagger] Hospital Vall D' Hebroón, Barcelona, Spain; § Hospital Miguel Servet, Zaragoza, Spain; || Hospital La Paz, Madrid, Spain; and ¶ Abbott Laboratories, Madrid, Spain

Received for publication May 21, 2004.

Accepted for publication October 14, 2004.

Correspondence: M. Leén-Sanz, MD, PhD, Unidad de Nutrición Clínica, Hospital Universitario 12 de Octubre, Av de Córdoba s/n, 28041 Madrid, Spain. Electronic mail may be sent to mleon@h12o.es.

Copyright American Society for Parenteral and Enteral Nutrition Jan/Feb 2005
Provided by ProQuest Information and Learning Company. All rights Reserved