Broadly speaking diabetes may be divided into two categories. Insulin dependent diabetes (IDDM) affect mainly younger people who become acutely unwell and require immediate and life-long treatment with insulin. Non-insulin dependent diabetes (NIDDM) affects predominately people aged over 50, is often of gradual onset and can usually be managed with diet alone or supplementary tablets, often without the need for insulin. Both forms of diabetes are associated with potential complications. In particular, NIDDM patients have a 2 to 3-fold increased incidence of cardiovascular disease compared to the non-diabetic population. Some of this excess morbidity can be explained by changes in known cardiovascular risk factors. Thus diabetic patients, by definition, have increased blood glucose concentrations; However, such patients also characteristically have a raised serum triglyceride concentration and a low HDL cholesterol concentration. There is epidemiological evidence that these factors are also associated with increased cardiovascular morbidity in NIDDM patients.

In NIDDM patients management aimed at reducing cardiovascular risk by affecting serum lipids should therefore not only reduce the serum total and LDL cholesterol concentrations but also, and perhaps more importantly, should ideally increase the HDL cholesterol concentration and reduce serum triglycerides. A reduction in blood pressure and an improvement in blood glucose concentration should also reduce cardiovascular risk.

Several papers in the literature suggest that garlic can have a beneficial effect on serum lipids. Dried garlic tablets (Kwai) can now be formulated in a manner which preserves a standardised high level of biological activity with a reduced garlic odour, and hence may be more socially acceptable. It has been shown that such tablets can reduce the cardiovascular risk factors of serum total cholesterol and blood pressure in patients where these are moderately raised in non-diabetic subjects. There is also one suggestion in the literature that garlic can reduce the blood glucose concentration.

We therefore investigated the effect of dried garlic tablets on a detailed lipid profile, on blood pressure, and on diabetic control in a group of NIDDM patients. We also enquired whether patients or their families noted a garlic odour.

We recruited 60 (46 male) NIDDM patients, well controlled on diet or tablets, and aged 42 to 75 (median 63) years. On screening, the serum total cholesterol was between 6.0 and 8.0 mmol/l. After taking baseline blood samples and blood pressure, patients were randomised to take 3 x 300 mg dried garlic tablets (Kwai) with main meals, or a matching placebo. Patients were reassessed at 6 and 12 weeks.

Garlic tablets led to a marginal fall in serum total cholesterol from 6.51 (0.26) mmol/l (mean and standard error) to 6.21 (0.17) at 6 weeks (p<0.07 versus placebo). but this decrease was not sustained at 12 weeks. There was a fall in serum LDL cholesterol from 4.68 (0.24) to 4.27 (0.17) mmol/l at 6 weeks (p<0.05) , but again this was not sustained at 12 weeks, at 4.36 (0.17). The HDL cholesterol concentration rose from 0.94 (0.05) to 1.01 (0.04) mmol/l with garlic at 12 weeks, compared with a fall from 1.05 (0.05) to 1.01 (0.06) mmol/l with placebo (p<0.05). Thus the LDL/HDL cholesterol ratio fell from 5.38 (0.40) to 4.45 (0.28) at 6 weeks and 4.52 (0.23) at 12 weeks, compared to placebo (p<0.05 for both comparisons). There were no differences in VLDL cholesterol, HDL2 and HDL3 cholesterol, serum total or sub fractional triglycerides or in apolipoprotein A1 or B with garlic compared to placebo. Garlic tablets also had no significant effect on fasting blood glucose, HbA1c, serum insulin or c peptide. An odour was noted personally at least once weekly by 9 of 29 patients on garlic tablets compared with 1 of 31 on placebo, and by 10 of 29 patients' relatives and friends compared with 0 of 31 on placebo.