Ambulatory blood pressure profiles and plasminogen activator inhibitor (PAI-1) activity in lean women with and without the polycystic ovary syndrome

OBJECTIVE Hyperinsulinaemic women with the polycystic ovary syndrome (PCOS) may be at increased risk of vascular disease later in life, mediated by blood pressure or lipid abnormalities or by elevated plasma levels of plasminogen activator inhibitor-1 (PAI-1) activity. PAI-1 may also be involved in ovarian follicle development and ovarian connective tissue remodelling. We measured plasma PAI-1 activity and 24-hour ambulatory blood pressure records in women with and without PCOS. DESIGN Cross-sectional study of three groups. PATIENTS Twenty-four non-obese women with a classic ovarian ultrasound appearance of PCO and extreme menstrual disturbance (Group 1), 26 matched controls with a normal menstrual cycle and an ultrasound appearance of PCO (Group 2) and 10 matched controls with a normal menstrual cycle and normal ovarian ultrasound (Group 3), MEASUREMENTS Twenty-four hour ambulatory blood pressure recordings (Spacelabs 90207), ovarian ultrasonography, fasting plasma insulin and glucose, plasma PAI-1 activity, HDL and total cholesterol, triglycerides, gonadotrophins and testosterone. Family history of premature vascular disease. RESULTS Median fasting plasma insulin was significantly higher in Group 1 (45.8 pmol/l, range 12.9-161.9) than in Group 2 (28.1 pmol/l; range 13.6-91; P < 0.05) or Group 3 (26.0 pmol/l; range 13.5-63.3; P < 0.05). There were no differences between groups in 24-hour, daytime or night-time ambulatory blood pressure measurements, and no relation between plasma insulin and any blood pressure variable. Mean plasma PAI-1 activity was higher in Group 1 (10.0 +/- 7.1 AU/I) than in Group 2 (6.0 +/- 4.6 AU/I; P < 0.05) or Group 3 (5.1 +/- 3.5 AU/I; P = 0.06). There was a significant independent direct relation between tasting plasma insulin and PAI-activity (r = 0.41, R(2) = 0.154; F-1,F-59 = 11.38; P = 0.001). Groups did not differ in parental history of premature vascular disease, or in mean HDL or fasting triglyceride levels. CONCLUSIONS The only measurable vascular risk factor associated with hyperinsulinaemia and menstrual disturbance in non-obese women with PCOS is an elevated plasma PAI-1 activity. These women did not differ from controls in ambulatory blood pressure profiles, lipid measurements or in a parental history of premature vascular disease. PAI-1 and plasminogen are involved in ovarian follicle maturation and the present finding suggests a biologically plausible link between hyperinsulinaemia, anovulation and vascular risk in PCOS.