Fifteen non-insulin-dependent diabetic patients with persistently elevated blood glucoses despite high doses of oral hypoglycaemic agents, were randomly allocated to a high carbohydrate-high fibre diet (HC) or a reinforced low carbohydrate diet (LC). After six weeks the diets were reversed for a similar period. Immediately preceding the study and at the end of each dietary period 24-h biochemical profiles were performed. In the 11 patients who completed the study, fasting and preprandial glucose, percentage glycosylated haemoglobin, VLDL cholesterol and mean 24-h triglycerides were significantly lower on HC than on LC or during the initial profile on their usual diet. There was no significant difference in any of the measurements on LC compared with the usual diet. Previous studies of high carbohydrate-high fibre diets in diabetes have been carried out in relatively well-controlled patients. These data show that poorly controlled non-insulin-dependent patients have an even more striking response.

Twenty diabetic outpatients (12 non-insulin-treated and 8 insulin-treated) were given guar granulate in a dose of 10 g daily for two months in order to study the effect on glycaemic control and lipid levels. Mean glycosylated haemoglobin levels (HbAlc%) fell from 11.1±2.0% pre-guar to 10.5 ± 2.2% (P< 0.001) after one month on guar and to 10.1±2.3% ( P<0.0001) after two months. Following discontinuation of guar, HbAlc% rose to 11.1±2.5% ( P<0.002). However, there were no significant changes in fasting blood glucose, 1 h postprandial blood glucose following a test meal, 24 h urinary glucose excretion or in lipid levels. Gastrointestinal side effects occurred in 4 patients during treatment with guar. Four patients reduced their dose of insulin and 2 patients reduced their dose of sulphonylurea therapy during this time because of symptoms suggestive of hypoglycaemia. We suggest that the low dose of guar used in this study may help improve glycaemic control in diabetic patients and that this may be achieved with a low incidence of gastrointestinal side effects.

No abstract

1233 endotoxic properties may be related to a lipopolysaccharide fraction which has not been found in other Gram-negative bacteria.5 Our patient had an initial encephalopathy with focal disturbance of the corpus striatum. Chorea is a previously unreported complication but, like other neurological dysfunctions, appears to resolve spontaneously after a protracted course.