1. Forty-four young men with blood pressure elevation and 29 age-matched volunteer subjects were examined with invasive blood pressure and cardiac output measurements. The regional blood flow resistance of the hand during hyperaemia was calculated from blood pressure and flow data from venous occlusion plethysmography.
2. In a re-examination 5 years later the same protocol was applied and an oral glucose tolerance test was performed with glucose and insulin determinations. Body composition was calculated from total body potassium data (whole body 40K counter) and body weight.
3. Patients with blood pressure elevation were characterized by a significantly higher cardiac index at rest, and a significantly increased resistance during hyperaemia of the hand in comparison with that of the controls. The patients with blood pressure elevations were also divided according to cardiac output. The hyper-kinetic subgroup did not have an increased resistance during hyperaemia. Patients with blood pressure elevation had significantly increased body weight and body mass index in comparison with controls and there was a significant correlation between body wieght and regional resistance during hyperaemia.
4. In the re-examination it was found that the body weight difference between patients with blood pressure elevation and normotensive controls was explained by increased body fat. There was no difference in body weight between the normokinetic and hyperkinetic subgroups of patients with blood pressure elevation but the former group had significantly increased insulin concentrations on the glucose load.
5. Significant correlations were demonstrated between body weight, body fat and insulin concentration 30 min after glucose load against resistance during hyperaemia as the dependent variable. A multiple regression coefficient of 0.59 was found between the three independent variables and resistance during hyperaemia. Furthermore, the addition of the insulin concentration variable to the two independent variables body weight and body fat increased significantly the multiple regression coefficient.