O R I G I N A L A R T I C L E Short-Term Effects of Alterations in Dietary Fat on Metabolic Control in IDDM PAUL IRELAND, MS, RD KERIN O'DEA, PHD ALISON NANKERVIS, MD, FRACP OBJECTIVE — Two experimental diets were evaluated to investigate the hypothesis that dietary fat has an independent influence on metabolic control in IDDM. RESEARCH DESIGN AND METHODS— The diets had similar CHO contents (26 and 22% of energy intake) but differed markedly in fat (53 vs. 16% energy) and protein (20 vs. 62% energy). We had 10 subjects follow the low-CHO, high-fat diet, and 8 subjects follow the low-CHO, low-fat, high-protein diet. In each case, markers of glycemic and lipid control obtained after adherence to the experimental diet for 2 wk were compared with corresponding data from a preceding control period during which subjects had followed their usual diet (protein 18-19%, CHO 41-46%, fat 33-37%). RESULTS — Despite the low CHO content of the high-fat diet, insulin requirements were unchanged relative to the control diet. Moreover, the glycemic response to a standard breakfast was elevated significantly (P < 0.001), suggesting that insulin resistance had either been induced or exacerbated. The small rise in total cholesterol concentration in response to the high-fat diet was accounted for by a rise in HDL cholesterol. Glycemic control and lipid metabolism were unchanged after the low- CHO, low-fat diet, although insulin requirements fell by an average of 6 U/day (P < 0.05) relative to those recorded during the 2-wk control period. CONCLUSIONS— Diets high in fat are deleterious to glycemic control in IDDM, but general applicability is limited by the small sample size and short duration of this study. FROM THE DEPARTMENT OF MEDICINE, UNIVERSITY OF MELBOURNE, ROYAL MELBOURNE HOSPITAL, PARKVILLE, VICTORIA; AND THE DEPARTMENT OF HUMAN NUTRITION, DEAKIN UNIVERSITY, VICTORIA, AUSTRALIA. ADDRESS CORRESPONDENCE AND REPRINT REQUESTS TO KERIN O'DEA, DEPT. OF HUMAN NUTRITION, DEAKIN UNIVERSITY, VICTORIA 3217, AUSTRALIA. RECEIVED FOR PUBLICATION 19 NOVEMBER 1991 AND ACCEPTED IN REVISED FORM 14 MAY 1992. IDDM, INSULIN-DEPENDENT DIABETES MELLITUS; H D L , HIGH-DENSITY L1POPROTEIN; N 1 D D M , NON- INSULIN-DEPENDENT DIABETES MELLITUS; C H O , CARBOHYDRATE; B M 1 , BODY MASS INDEX; S M B G , SELF- MONITORING OF BLOOD GLUCOSE; FFA, FREE FATTY ACID; LDL, LOW-DENSITY LIPOPROTEIN; CI, CONFI- DENCE INTERVAL. A key objective in the management of IDDM is avoidance of hyper- glycemia. Individual insulin regi- mens are designed to suppress acute postprandial hyperglycemia and chronic elevations of blood glucose resulting from unsuppressed endogenous glucose production. The risk of hyperglycemia and/or hypoglycemia associated with ex- ogenous insulin therapy has ensured that the focus of dietary instruction for IDDM has been on dietary CHO and its regular distribution throughout the day (1). Not- withstanding the importance of CHO re- striction in reducing postprandial excur- sions, little data are available relating to its effect on glycemic control in the fast- ing state. Recent data from this labora- tory (2) in NIDDM show deterioration in metabolic control on a low-CHO, high- fat diet, but improved metabolic control on both a low-CHO, low-fat and a high- CHO, low-fat diet suggests that it is the fat, rather than the CHO, content of the diet that might be critical in determining long-term glycemic control in diabetes. The aim of this study was to determine whether IDDM has a similar metabolic response to NIDDM when following two low-CHO diets that differ in fat content. RESEARCH DESIGN AND METHODS— Thirteen patients (8 women, 5 men, age 30.3 ± 1.6 yr, BMI 23.1 ± 0.6) were recruited from the outpa- tient clinic of the Royal Melbourne Hospital. The protocol had been approved by the Hospital's Ethics Committee, and all sub- jects gave their voluntary written consent before participation. All subjects met the composite clinical criteria of IDDM, namely, being <40 yr of age and being < 120% of desirable body weight at the time of diag- nosis, and having been placed on insulin therapy within 2 yr of diagnosis (3). Five subjects took part in both of the experiments described, another 5 fol- lowed the high-fat diet only, and the remaining 3 followed the low-fat, low- CHO diet only (Table 1). All subjects followed conventional twice-daily insu- DIABETES CARE, VOLUME 15, NUMBER 11, NOVEMBER 1992 1499