Plasma plasminogen activator inhibitor-I is associated with plasma leptin irrespective of body mass index, body fat mass, and plasma insulin and metabolic parameters in premenopausal women
V. De Mitrio et al., Plasma plasminogen activator inhibitor-I is associated with plasma leptin irrespective of body mass index, body fat mass, and plasma insulin and metabolic parameters in premenopausal women, METABOLISM, 48(8), 1999, pp. 960-964
Leptin, the satiety hormone expressed almost exclusively in adipose tissue,
is a marker of body fat accumulation in humans. Recent studies have shown
that plasminogen activator inhibitor-1 (PAI-1), a prothrombotic factor asso
ciated with atherosclerosis complications, is also produced in adipose tiss
ue. The objective of the present study was to determine whether PAI-1 antig
en plasma concentrations are associated with leptin plasma levels or the bo
dy fat mass (FM) independently of the variables known to influence PAI-1 pr
oduction. Sixty-one nondiabetic women aged 18 to 45 years with a wide range
of values for the body mass index ([BMI] 18.1 to 37.7 kg/m(2)) were evalua
ted for (1) body FM and fasting plasma levels of (2) PAI-1 antigen, (3) PAI
-1 activity, (4) leptin, (5) insulin, (6) blood glucose, and (7) lipids (ch
olesterol, high-density lipoprotein [HDL]-cholesterol, and triglycerides [T
G]). Body FM and fat-free mass (FFM) were estimated during fasting conditio
ns by the bioimpedance analysis (BIA) method using a tetrapolar device. Bod
y fat distribution was evaluated by the waist circumference and the waist t
o hip ratio (WHR). FM was directly associated with both PAI-I antigen (r =
.585, P < .001) and PAI-1 activity (r = .339, P < .001). Seemingly, leptin
was positively related to both PAI-1 antigen (r = .630, P < .001) and PAI-1
activity (r = .497, P < .001). Moreover, both PAI-I antigen and PAI-1 acti
vity were directly correlated with FFM (r = .285, P < .05, and r = .336, P
< .01, respectively), BMI (r = .594, P < .001, and r = .458, P < .001, resp
ectively), and WHR (r = .510, P < .001, and r = .391, P < .005, respectivel
y). Insulin was directly related to PAI-T antigen (r = .540, P < .001), PAI
-1 activity (r = .259, P < .05), leptin (r = .447, P < .001), and FM (r = .
435, P < .001). The association between PAI-1 antigen (dependent variable)
and leptin or FM was tested by a stepwise regression model simultaneously i
ncluding leptin, FM, BMI, WHR, age, FFM, and fasting insulin, blood glucose
, TG, cholesterol, and HDL-cholesterol as independent variables. PAI-1 anti
gen maintained a significant positive independent relationship only with le
ptin (t = 2.923, P < .01), insulin (t = 3.489, P < .001), and fasting blood
glucose (t = 2.092, P < .05), and a negative independent relationship with
HDL-cholesterol (t = -2.634, P < .05). In conclusion, the strong relations
hip between PAI-1 antigen and leptin irrespective of other variables known
to influence these factors seems to indicate that leptin per se may potenti
ally increase PAI-1 plasma concentrations in obese subjects. Copyright (C)
1999 by W.B. Saunders Company.