Diabetes Contributes to Cholesterol Metabolism Regardless of Obesity
Total Page:16
File Type:pdf, Size:1020Kb
Clinical Care/Education/Nutrition ORIGINAL ARTICLE Diabetes Contributes to Cholesterol Metabolism Regardless of Obesity 1 PIIA P. SIMONEN, MD tes, is responsible for the alterations ob- 2 HELENA K. GYLLING, MD served in cholesterol metabolism in 1 TATU A. MIETTINEN, MD diabetes, or does diabetes have any inde- pendent role in regulating cholesterol me- tabolism. To this end, we studied cholesterol absorption efficiency and sterol balance OBJECTIVE — To investigate cholesterol metabolism in obesity with and without diabetes. in obese subjects with and without diabe- tes, the latter selected by BMI from a pop- RESEARCH DESIGN AND METHODS — We performed cross-sectional metabolic ulation-based cohort. studies in obese individuals with and without type 2 diabetes. The study population consisted of 16 obese (BMI Ͼ30 kg/m2) diabetic subjects with a mean age of 52 Ϯ 2 years (SE) and 16 nondiabetic control subjects of similar age and weight. Cholesterol absorption efficiency was RESEARCH DESIGN AND measured with peroral dual isotopes and cholesterol synthesis with sterol balance. METHODS RESULTS — Serum total cholesterol did not differ between the groups, but LDL and HDL Study population cholesterol were significantly lower and VLDL cholesterol and serum total and VLDL triglycer- The study population consisted of 16 ides were higher in the diabetic group than in the control group. Cholesterol absorption effi- obese diabetic patients (BMI Ͼ30 kg/m2), ciency was 29 Ϯ 1% in diabetic subjects vs. 42 Ϯ 2% in the control subjects (P Ͻ 0.01). Ϯ Ϯ ⅐ Ϫ1 ⅐ Ϫ1 Ͻ 13 men and 3 women, with a mean age of Cholesterol synthesis was higher (17 1 vs. 14 1mg kg day ; P 0.05) and neutral Ϯ sterol and bile acid excretion and cholesterol turnover tended to be higher in the diabetic group 52 2 (SE) years, recruited from the out- than in the control group. Blood glucose was positively related to cholesterol synthesis in the patient clinics of Helsinki University Hos- diabetic group (r ϭϩ0.663, P Ͻ 0.01) and in the control group (r ϭϩ0.590, P Ͻ 0.05), pital (diabetic group; Table 1). Diabetes suggesting that the higher blood glucose level, the higher the cholesterol synthesis. In addition, (fasting blood glucose Ͼ7.0 mmol/l) had blood glucose was significantly positively related to fecal neutral sterol excretion in both groups. been recently diagnosed (Ͻ2 years). None of the patients had insulin therapy, CONCLUSIONS — Cholesterol absorption efficiency was lower and cholesterol synthesis diabetic nephropathy, hepatic or thyroid was higher in obese subjects with diabetes than in those without diabetes, suggesting that disease, unstable angina pectoris, or myo- diabetes modulates cholesterol metabolism more than obesity alone. cardial infarction or invasive coronary Diabetes Care 25:1511–1515, 2002 treatment within a year. All women (in- cluding the control subjects) were post- menopausal without hormone replace- ment therapy. Diabetes was treated with n patients with type 2 diabetes, choles- high-to-normal blood glucose levels (11). diet in 10 patients, 3 had glibenclamide, terol metabolism differs from nondia- Accordingly, cholesterol metabolism and 3 had a combination therapy of glib- I betic patients because cholesterol mimics that observed in obesity (12–15). enclamide and biguanide. There was no synthesis is high (1–4) and is reduced by We have shown previously that cho- difference in variables of serum and li- insulin (5,6). Low cholesterol absorption lesterol absorption efficiency was in- poprotein lipids and variables of choles- efficiency has been reported earlier in a creased by weight reduction, and the terol metabolism between these treatment limited number of diabetic subjects with variables of glucose metabolism im- groups. mild hyperlipidemia (4) and in moder- proved in obese diabetic subjects, sug- From a random population-based ately overweight, markedly hypertriglyc- gesting that low cholesterol absorption is age-cohort of 50-year-old men (9) and eridemic subjects including both type 1 associated with insulin resistance and women (17), 16 healthy normoglycemic and type 2 diabetes (7). Also, serum plant metabolic syndrome (16), an association subjects (10 men and 6 women), with sterol levels, indicators of cholesterol ab- found earlier nondiabetic subjects (11). BMI and age similar to the diabetic group, sorption efficiency (8,9) are low in type 2 The question now is whether overweight, were recruited as control subjects. The ex- diabetes (10) and even in subjects with which is frequently associated with diabe- clusion criteria, except insulin therapy ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● and diabetic nephropathy, were the same as in the diabetic group.Their health sta- From the 1Department of Medicine, Division of Internal Medicine, University of Helsinki, Finland; and the 2Department of Clinical Nutrition, University of Kuopio and Kuopio University Hospital, Kuopio, Finland. tus was determined with medical exami- Address correspondence and reprint requests to Tatu A. Miettinen, Division of Internal Medicine, De- nation and laboratory tests. All subjects partment of Medicine, University of Helsinki, P.O. Box 340, FIN-00029 HUS, Finland. E-mail: volunteered to the study and gave in- tatu.a.miettinen@helsinki.fi. formed consent. The study protocol had Received for publication 12 February 2002 and accepted in revised form 16 May 2002. Abbreviations: apo, apoprotein. been accepted by the Ethics Committee of A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion the 2nd Department of Medicine, Univer- factors for many substances. sity of Helsinki. DIABETES CARE, VOLUME 25, NUMBER 9, SEPTEMBER 2002 1511 Diabetes and cholesterol metabolism Table 1—Study population formed using Student’s two-sided t test and the 2 test and by calculating Pear- Variables Diabetic group Control group son’s correlation coefficients. Logarithmic transformations were used with skewed n 16 16 distributions. A P value Ͻ0.05 was con- M/F 13/3 10/6 sidered statistically significant. Age (years) 52.2 Ϯ 1.8 50.8 Ϯ 0.5 Weight (kg) 93.2 Ϯ 3.7 96.4 Ϯ 3.3 RESULTS — No sex difference was Height (m) 1.70 Ϯ 0.02 1.70 Ϯ 0.01 found for different variables between the BMI (kg/m2) 32.2 Ϯ 1.0 33.3 Ϯ 0.9 diabetic and control groups (data not Blood glucose (mmol/l) 8.4 Ϯ 0.6* 4.6 Ϯ 0.2 shown). In the diabetic group, treatment Cholesterol intake (mg/day) 351 Ϯ 47 455 Ϯ 49 of diabetes had no effect on the different Fat intake (g/day) 92 Ϯ 9 101 Ϯ 9 variables. The two groups were similar ac- Plant sterol intake (mg/day) 356 Ϯ 31 352 Ϯ 34 cording to demographic variables (Table Data are means Ϯ SE, derived from Student’s t test. *Significantly different from control subjects. 1) and apoE phenotype distribution (data not shown), but blood glucose level was Study design ponifiable material on a 50-m capillary higher in the diabetic group than in the The diabetic group had been counseled to column (21–23), correcting fecal flow by control group. The average serum insulin level was 19.7 Ϯ 1.2 mU/l in the diabetic consume a low-fat, low-cholesterol diet the Cr2O3 measurement (24). with carbohydrates up to 55% energy, group and 8.0 Ϯ 0.8 mU/l in the control with preference of carbohydrates with Calculations group (measured randomly in six sub- low glycemic index and simple carbohy- Cholesterol absorption efficiency (frac- jects; P Ͻ0.001 from the diabetic group). drates comprising less than one-third of tional absorption of dietary cholesterol) The reference values of our hospital labo- the total amount, whereas the control was calculated by the altered 14C-to-3H ratory were 2–20 mU/l. The dietary vari- subjects consumed their normal ad libi- ratio in stools as compared with the fed ables, fat intake, dietary plant sterols, and tum home diet. The subjects visited the ratio (25). cholesterol intake did not differ between outpatient clinic twice, a week apart, Cholesterol synthesis was determined the study groups. when two blood samples were collected as difference between the fecal steroids Serum total cholesterol was similar after a 12-h fast.The subjects kept a food (neutral and acidic) of cholesterol origin between the groups, whereas LDL and record for 7 days, from which the dietary and dietary cholesterol. Total intestinal HDL cholesterol levels were lower and constituents were calculated (18). Also, in cholesterol pool was calculated as fecal VLDL cholesterol was higher in the dia- order to measure cholesterol absorption, neutral steroids/(1 Ϫ cholesterol absorp- betic group than in the control group (Ta- they were given a capsule containing tion efficiency). The absorbed mass of to- ble 2). Serum total and VLDL triglycerides 4-14C-cholesterol (4,500 Ϯ 19 dpm), tal, dietary, and biliary cholesterol was were higher and LDL triglycerides were 22,23-3H--sitostanol (11,588 Ϯ 42 calculated as respective fluxes multiplied lower in the diabetic group than in the control group. dpm), and Cr2O3 (200 mg) 3 times a day by cholesterol absorption efficiency. Cho- with their regular meals during the 7-day lesterol turnover equaled cholesterol syn- Percent cholesterol absorption and period. Stool was collected on the last 3 thesis plus dietary cholesterol absorbed. the absorbed mass of dietary, total, and days of the week. The samples were biliary cholesterol were ϳ30% (P Ͻ 0.01) pooled, and fecal elimination of choles- Statistics lower in the diabetic group than in the terol and bile acids and cholesterol ab- Statistical analyses of data were per- control group (Table 3). Cholesterol syn- sorption efficiency were measured from formed using the Biomedical Data Pro- thesis was significantly higher in the dia- these fecal samples. cessing Program and Microsoft Excel betic group than in the control group.