Ps. Wells et al., PREVALENCE OF ANTITHROMBIN DEFICIENCY IN HEALTHY BLOOD-DONORS - A CROSS-SECTIONAL STUDY, American journal of hematology, 45(4), 1994, pp. 321-324
The prevalence of antithrombin (AT) deficiency in the general populati
on has been variously estimated to be between 0.05 and 5 per 1,000 in
the population; 2,491 blood donors were screened in an attempt to clar
ify this issue using plasma samples taken from the brood donor units.
From this initial population, 122 individuals were identified as havin
g plasma AT levers lower than 2 standard deviations below the normal m
ean. Twenty-two samples had evidence that thrombin had been generated
during blood collection and the remaining cohort of 100 blood donors w
ere asked to return but only 59 complied. The data obtained from these
59 were compared with that from 51 age- and sex-matched control blood
donors. Both groups of subjects were assessed for previous evidence,
or family history, of thrombotic events, as well as exposure to risk f
actors associated with the development of deep vein thrombosis (DVT).
All had venous brood samples taken from which the supernatant plasma w
as immediately removed and quick frozen for later assaying. Only 6 of
the 59 subjects with initial low AT levels had repeat AT-Xa levels bel
ow 0.80 units/ml (normal range 0.94 +/- 0.14). Upon repeating the AT-X
a determinations on new samples from these six individuals, only three
were found again to be low. One was found to have a type 3 AT deficie
ncy (an Arg47Cys substitution). The other two with a low AT level had
mean functional AT-Xa revels of 0.61 and 0.71 units/ml, respectively,
with correspondingly low AT:Ag revels consistent with a type 1 AT defi
ciency. Two of these three subjects has been in high risk situations w
ithout evidence of having developed DVT and none had evidence of venou
s reflux on Doppler venography. In addition, none had personal or fami
ly histories of previous thrombotic events. These present data indicat
e that the prevalence of AT deficiency in our blood donor population i
s 2 per 1,000 (95% confidence intervals: 0.7-6/1,000). (C) 1994 Wiley-
Liss, Inc.