The role of dietary cholesterol in people with diabetes has been little studied. We investigated the effect of a hypoenergetic high-protein high-cholesterol (HPHchol) diet compared to a similar amount of animal protein (high-protein low-cholesterol, HPLchol) on plasma lipids, glycaemic control and cardiovascular risk markers in individuals with type 2 diabetes. A total of sixty-five participants with type 2 diabetes or impaired glucose tolerance (age 54·4 (SD 8·2) years; BMI 34·1 (SD 4·8) kg/m 2 ; LDL-cholesterol (LDL-C) 2·67 (SD 0·10) mmol/l) were randomised to either HPHchol or HPLchol. Both hypoenergetic dietary interventions (6-7 MJ; 1·4-1·7 Mcal) and total carbohydrate:protein:fat ratio of 40:30:30 % were similar but differed in cholesterol content (HPHchol, 590 mg cholesterol; HPLchol, 213 mg cholesterol). HPHchol participants consumed two eggs per d, whereas HPHchol participants replaced the eggs with 100 g of lean animal protein. After 12 weeks, weight loss was 6·0 (SD 0·4) kg (P,0·001). LDL-C and homocysteine remained unchanged. All the subjects reduced total cholesterol (2 0·3 (SD 0·1) mmol/l, P, 0·001), TAG (2 0·4 (SD 0·1) mmol/l, P,0·001), non-HDL-cholesterol (HDL-C, 20·4 (SD 0·1) mmol/l, P,0·001), apo-B (2 0·04 (SD 0·02) mmol/l, P, 0·01), HbA1c (2 0·6 (SD 0·1) %, P, 0·001), fasting blood glucose (2 0·5 (SD 0·2) mmol/l, P,0·01), fasting insulin (21·7 (SD 0·7) mIU/l, P,0·01), systolic blood pressure (2 7·6 (SD 1·7) mmHg, P,0·001) and diastolic blood pressure (24·6 (SD 1·0) mmHg; P,0·001). Significance was not altered by diet, sex, medication or amount of weight loss. HDL-C increased on HPHchol (þ 0·02 (SD 0·02) mmol/l) and decreased on HPLchol (20·07 (SD 0·03) mmol/l, P, 0·05). Plasma folate and lutein increased more on HPHchol (P,0·05). These results suggest that a high-protein energy-restricted diet high in cholesterol from eggs improved glycaemic and lipid profiles, blood pressure and apo-B in individuals with type 2 diabetes.