Background and aims: In recent years it has been proposed that hypertension is part of a cluster of metabolic risk factors (syndrome X) involving hyperlipidaemia and hyperglycaemia, with hyperinsulinaemia as the common link. This study has investigated: (1) the prevalence of the metabolic syndrome and its component variables and their relationship to body mass index (BMI) and non-fasting insulin levels in a general population; and (2) the distribution and clustering of metabolic variables in normotensives and hypertensives. Methods: Cross-sectional study of 5222 men aged 40-59 years with no history of coronary heart disease (CHD), diabetes mellitus or stroke drawn from general practices in 18 British towns. The men were a subgroup of the 7735 men in the British Regional Heart Study (BRHS) cohort whose baseline non-fasting serum was analysed for insulin, using a specific ELISA method. Main outcome measures: Hyperinsulinaemia, hyperglycaemia, high serum total cholesterol, high triglyceride and hyperuricaemia were defined as the top 20% of the distribution in the 5222 men. Low HDL-cholesterol was defined as the bottom 20%. Results: BMI and non-fasting insulin were both significantly and strongly associated with non-diabetic hyperglycaemia, lipid abnormalities (HDL-cholesterol, triglyceride and total cholesterol) and hyperuricaemia. BMI was strongly associated with hypertension whereas non-fasting insulin showed a much weaker relationship which was abolished after adjustment for BMI. However, only 2.9% of men showed the 'full metabolic syndrome' (hypertension, hyperglycaemia and dyslipidaemia) and a large proportion of these men were hyperinsulinaemic (65%) or obese (47%). Dyslipidaemia (any one of low-HDL-cholesterol, high triglyceride or high cholesterol) was common in both normotensives and hypertensives (40.5% vs 46.4%). Hypertensives showed significantly higher levels of total cholesterol, triglyceride, blood glucose, urate and more clustering of hyperglycaemia and dyslipidaemia than normotensives even after adjustment for BMI. Conclusion: Hypertensives were more likely to have lipid abnormalities and clustering of risk factors than normotensives even after adjustment for BMI. The metabolic syndrome is more strongly associated with hyperinsulinaemia than with obesity but it is relatively uncommon in men with no history of cardiovascular disease or diabetes. Given the weak relationship between hypertension and hyperinsulinaemia, the latter is unlikely to explain the higher levels of lipid abnormalities and clustering seen in hypertensives. Overweight/obesity may be primarily involved in the pathways to hypertension and lipid abnormalities but the unravelling of these relationships require more specific measures of adipose tissue distribution, composition and function.