|
MULTICENTRE
STUDIES
QUANTITATIVE
LA TESTS AND THEIR APPLICATION ON PLASMA SAMPLES FROM THE WAPS STUDY
COORDINATORS
Eva M. Jacobsen1, Monica Galli2 , Giovanna
Borrelli3 and Finn Wisløff1
Ulleval University Hospital, Oslo, Norway1, Ospedali
Riuniti, Bergamo2 and
Istituto “Mario Negri Sud”, S. Maria Imbaro3, Italy
PROJECT
Any phospholipid (PL) dependent clotting
assay may be used for the detection of lupus anticoagulants (LA). Basically, with these tests four
clotting times are obtained: for the test plasma, clotting time with a low
(screening) and a high (confirmatory) PL concentration, termed “a” and “b”, respectively.
A pooled reference plasma is tested in the same run, giving clotting times
termed “c” with a low PL concentration and “d” with a high PL concentration.
Test results may be calculated in various ways, using two or all four of these
clotting times. The choice of calculation methods influences the apparent
sensitivity of the test, and there is an advantage to using all four parameters
(Jacobsen et al, 2001). The Lupus Ratio (LR) is a normalized test plasma ratio:
a/b divided by c/d. The LR is thus a continuous variable, with a cut-off (upper
reference limit) arbitrarily set at the 97.5th percentile of a
reference population (1.10 at the Haematological Research Laboratory, Ulleval).
However, as of yet there is no evidence that LA quantification gives clinically
important information beyond a simple positive/negative test result.
At
Ulleval University Hospital, LR tests have been developed on the basis of the
activated partial thromboplastin time (APTT), the Russell’s Viper Venom time
(RVVT) and the prothrombin time (PT). Test plasma are routinely mixed 1:1 with
a pooled reference plasma prior to testing. LRs obtained with these three tests
on large number of patient samples are correlated, with correlation
coefficients ranging from 0.57 to 0.81, which suggests considerable
discrepancy. On the basis of low CVs (1-4%), these diverging results indicate
that the three LR tests to a large extent detect different antibody
specificities.
The WAPS study was a
randomized clinical trial of high-intensity warfarin vs. conventional
antithrombotic therapy for the prevention of recurrent thrombosis in patients
with the antiphospholipid syndrome (APS) (Finazzi et al, 2005). 109 patients were randomized to
high-intensity or standard antithrombotic therapy. Additional APS patients were
followed without intervention. 108 plasma samples drawn at baseline, some from
the clinical trial and some from the observational study, were available for
the three quantitative LR tests. During four years of follow up, fifteen of
these patients experienced an arterial or venous thrombotic event. Logistic
regression was performed to determine the ability of the three LR tests to
predict these events. The dependent (outcome) variable was event/no event; the
independent (predictor) variable was the LR obtained with the APTT, RVVT or
PT-based LA test, respectively. The median LR was higher for patients with
events in all three tests. However, only the PT-based test was a significant
predictor in the logistic regression model. The odds ratio was 2.6 for each
unit increase in the LR (p=0.03).
Independent statistical analysis at the Mario Negri Sud Institute
confirmed that the LA-dPT test was the only assay associated with the clinical
endpoints. The number of patients and events was small in this study, and
independent confirmation is needed. A suggestion for a Forum project to
investigate this possible relationship will be presented at the meeting.
For additional
information and applications, please e-mail: f.g.b.wisloff@medisin.uio.no
|
Web contens: EWPSLE Info Desk. |
|
|
Last updated: 17 November 2005 |
|