Previous studies have suggested that hyperinsulinaemia and other components of metabolic syndrome are risk factors for clinical prostate cancer. This prospective study tested the hypothesis that hyperinsulinaemia and other components of metabolic syndrome are risk factors for lethal clinical prostate cancer. The clinical, haemodynamic, anthropometric, metabolic and insulin profile at baseline in men who had died from clinical prostate cancer during follow-up was compared with the profile of men who were still alive at follow-up. If the hypothesis is true, men with an unfavourable prognosis would have a higher profile at baseline than those with a favourable prognosis. A total of 320 patients in whom clinical prostate cancer, stages T2-3, had been diagnosed were consecutively included in the study during 1995-2003. Height, body weight, waist measurement, hip measurement and blood pressure were determined. Body mass index and waist/hip ratio (WHR) were calculated. Blood samples were collected to determine triglycerides, total cholesterol, high-density lipoprotein (HDL)-cholesterol, low-density lipoprotein (LDL)-cholesterol, uric acid, alanine aminotransferase and fasting plasma insulin level. The prostate gland volume was measured using transrectal ultrasound. The annual benign prostatic hyperplasia (BPH) growth rate was calculated. The diagnosis of prostate cancer was established using transrectal ultrasound-guided automatic needle biopsy of the prostate gland. All patients with clinical prostate cancer were followed up until their death or until the study was terminated on 31 December 2003. At follow-up, 54 patients had died from prostate cancer and 219 were still alive. The results showed that the men who died of clinical prostate cancer during the follow-up period were older (P < 0.001), had a larger prostate gland volume (P < 0.001), a faster BPH growth rate (P < 0.001), a higher prevalence of type 2 diabetes (P < 0.035) and treated hypertension (P < 0.023), a higher stage (P < 0.001) and grade (P = 0.028) of clinical prostate cancer, a higher prostate-specific antigen (PSA) level (P < 0.001) and a higher PSA density (P < 0.001) at baseline than men still alive with clinical prostate cancer at follow-up. These men also had a lower HDL-cholesterol level (P = 0.027), a higher fasting plasma insulin level (P = 0.004), a higher WHR (P = 0.097) of borderline significance and a higher uric acid level (P = 0.079) of borderline significance. Eliminating the effect on mortality of higher stage and grade of the clinical prostate cancer and PSA at baseline, the following statistically significant correlations remained: a higher fasting plasma insulin level (P = 0.010) and a lower HDL-cholesterol level of borderline significance (P = 0.065). In conclusion, hyperinsulinaemia and five other previously established components of metabolic syndrome are shown to be prospective risk factors for deaths that can be ascribed to prostate cancer. These findings confirm previous study, which indicate that prostate cancer is a component of metabolic syndrome. Moreover, these data indicate that hyperinsulinaemia and other metabolic disorders precede deaths caused by prostate cancer. Thus, our data support the hypothesis that hyperinsulinaemia is a promoter of clinical prostate cancer. Furthermore, our data suggest that the insulin level could be used as a marker of prostate cancer prognosis and tumour aggressiveness, regardless of the patient's prostate cancer stage, cancer grade and PSA level.