Dietary advice for treatment of type 2 diabetes mellitus in adults.

BACKGROUND While initial dietary management immediately after formal diagnosis is an 'accepted' cornerstone of treatment of type 2 diabetes mellitus, a formal and systematic overview of its efficacy and method of delivery is not currently available. OBJECTIVES To assess the effect of type and frequency of different types of dietary advice to all adults with type 2 diabetes on weight, measures of diabetic control, morbidity, total mortality and quality of life. SEARCH STRATEGY We carried out a comprehensive search of The Cochrane Library (The Cochrane Library Issue 3, 2003), MEDLINE (1966 to October Week 1, 2003), EMBASE (1980 to Week 40, 2003), CINAHL (1982 to October Week 1, 2003), AMED (1985 to October 2003), bibliographies and contacted relevant experts. SELECTION CRITERIA All randomised controlled trials, of six months or longer, in which dietary advice was the main intervention in adults with type 2 diabetes mellitus. DATA COLLECTION AND ANALYSIS The lead investigator performed all data extraction and quality scoring with duplication being carried out by one of the other six investigators independently with discrepancies resolved by discussion and consensus. Authors were contacted for missing data. For continuous outcomes, endpoint data were preferred to change data. MAIN RESULTS Thirty-six articles reporting a total of eighteen trials following 1467 participants were included. Dietary approaches assessed in this review were low-fat/high-carbohydrate diets, high-fat/low-carbohydrate diets, low-calorie (1000 kcal per day) and very-low-calorie (500 kcal per day) diets and modified fat diets. Two trials compared the American Diabetes Association exchange diet with a standard reduced fat diet and five studies assessed low-fat diets versus moderate fat or low-carbohydrate diets. Two studies assessed the effect of a very-low-calorie diet versus a low-calorie diet. Six studies compared dietary advice with dietary advice plus exercise and three other studies assessed dietary advice versus dietary advice plus behavioural approaches. The studies all measured weight and measures of glycaemic control although not all studies reported these in the articles published. Other outcomes which were measured in these studies included mortality, blood pressure, serum cholesterol (including LDL and HDL cholesterol), serum triglycerides, maximal exercise capacity and compliance. The results suggest that adoption of regular exercise is a good way to promote better glycaemic control in type 2 diabetic patients, however all of these studies were at high risk of bias. REVIEWERS' CONCLUSIONS There are no high quality data on the efficacy of the dietary treatment of type 2 diabetes, however the data available indicate that the adoption of exercise appears to improve glycated haemoglobin at six and twelve months in people with type 2 diabetes. There is an urgent need for well-designed studies which examine a range of interventions, at various points during follow-up, although there is a promising study currently underway.