| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Laboratory Analysis and2 Department of Haematology and Thrombotic Disorders, Ospedale Evangelico Valdese–ASL-1, Turin, Italy;
aaddress correspondence to this author at: Laboratorio Analisi, OspedaleEvangelico Valdese–ASL-1, Via Silvio Pellico, 28, 10125 Turin, Italy; fax 39-11-6688640, e-mail b.montaruli{at}tin.it
Lupus anticoagulants (LAs) are acquired circulating anticoagulants that interfere with phospholipid (PL)-dependent coagulation tests and are frequently associated with thromboembolic disorders and obstetric complications. Detection of LAs is of major importance in patients with these conditions (1)(2). LAs are diagnosed according to the criteria proposed by the Scientific and Standardization Committee on LAS of the International Society of Thrombosis and Hemostasis (3). According to these criteria, a diagnosis of LA should follow a 4-step procedure respecting the following principles: (a) prolongation of 1 (or more) PL-dependent clotting test (screening test); (b) evidence of inhibition demonstrated after mixing equal amounts of patient and normal plasma (mixing test); (c) evidence that the inhibitor is PL dependent, as demonstrated by correction of the clotting defect in the presence of excessively high PL concentrations (confirmatory test); and (d) lack of specific inhibition of any coagulation factor (distinction from other coagulopathies by specific factor assays).
Despite these criteria, diagnosis of LA remains a problem for the clinical laboratory. Contributing to these problems are the marked differences in sensitivity and specificity for the various LA screening assays that have been proposed, the lack of a universally accepted definition of a positive mixing test, technical variables affecting the various assays for LA, the difficulty with result interpretation, and the heterogeneous nature of LA itself (4)(5). At present, the most used screening tests for detecting LA are a dilute activated partial thromboplastin time, kaolin clotting time, and the diluted Russell venom time. Silica clotting time (SCT) has been described as a specific and sensitive alternative to kaolin clotting time for detecting LA (6)(7)(8)(9). We have evaluated a new automated "SCT screening and confirmatory" assay (not commercially available at this time) that has been proposed for the detection of LA.
The SCT Screen (HemosIL; IL) was run on an ACL 9000 automated coagulometer (IL). Prolongation of the SCT Screen tests was expressed as the ratio of patient coagulation time to the clotting time of the control (normal pool). Mixing studies were carried out on 1:1 and 4:1 mixtures of patient and normal plasmas on all samples that had prolonged SCT Screen times; failure to correct the clotting time was considered evidence of an inhibitor. For a confirmatory test of SCT, we used this new commercial assay (SCT Confirm, HemosIL; IL). LA was diagnosed when the confirmatory procedure was positive. Whereas the results of SCT Screen coagulation tests were expressed as the (sample clotting time/normal pool clotting time) ratio, the results of SCT Confirm tests were expressed as a "normalized LA ratio": the ratio result from the LA screen test divided by the ratio result from the LA confirmation test (patient confirm clotting time/normal pool confirm clotting time).
Mean (SD) and the parametric 95th percentile of clotting time ratios in 30 healthy controls (voluntary blood donors; 15 males and 15 females; age range, 16–81 years) were 1.04 (0.12) and 1.21 for SCT Screen and 1.02 (0.10) and 1.22 for the SCT Confirm test. Confirm ratios above the 95th percentile were regarded as positive.
For quality control, we used a normal pool and a LA-positive sample (LA+). Within-run imprecisions (CVs; n = 10) for the SCT Screen were 1.0% for the normal pool and 3.1% for the LA+ sample; for the SCT Confirm, the CVs were 2.0% for the normal pool and 2.6% for the LA+ sample. The CVs over 10 separate days for the SCT Screen were 3.9% for the normal pool and 5.2% for the LA+ sample; for the SCT confirm, the CVs were 4.8% for the normal pool and 5.7% for the LA+ sample.
We investigated the diagnostic specificity in 41 patients with known coagulation abnormalities: 12 patients on low–molecular-weight heparin therapy (LMWH), 23 patients on oral anticoagulant therapy [OAT LA+ (n = 7) and OAT LA– (n = 16)], 3 patients with factor deficiencies of the intrinsic coagulation system (1 with a factor VIII:C activity of 37%, 1 with a factor IX:C activity of 39%, and 1 with a factor XI:C activity of 41%), and 3 patients with type 1 von Willebrand disease (vWD; Table 1
). Six of 12 patients on LMHW, all of the patients with defects of intrinsic coagulation factors, and 1 of 3 patients with vWD had prolonged SCT Screen times, but all of them were identified as LA-negative by the SCT Confirm assay. Nineteen of 23 patients on OAT had prolonged SCT Screen times: 7 of these 19 previously identified as having a LA were confirmed as LA-positive by the SCT Confirm assay, whereas the other 12 were identified as LA-negative (see Fig. 1
).
|
|
To evaluate the screening performance (sensitivity and specificity) of the SCT Screen and Confirm assays, we collected and studied consecutive plasmas from 136 "anticoagulant-free" patients (54 males and 82 females; age range, 16–84 years) for whom a LA determination was requested by the Department of Thrombosis and Hemorrhagic Diseases. All 136 plasma samples were further analyzed for the presence of LA by our laboratory’s routine LA (SCT in-house method) screening and confirmation tests (Fig. 1
). Forty-six of 136 patients were identified as LA-positive by our routine LA tests. Of these samples, a prolonged SCT Screen test was found in 40. The inhibitor activity observed in SCT Screen-positive patients was confirmed to be of the LA type by the SCT Confirm assay. Six of 46 samples identified as having a LA were SCT-negative; in these patients, the only test positive was the diluted Russell venom time. Six of 91 LA-negative patients were positive by SCT Confirm assay. Of these patients, 2 were positive for anti-cardiolipin IgM (18.0 and 12.0 MPL, respectively; normal values <7.0 kMPL/L), 1 for anti-prothrombin IgG (12.5 kIU/L; normal values <9.0 kIU/L), 1 for anti-protein S IgM (21.0 kIU/L; normal values <15.0 kIU/L), and 1 for anti-protein C IgM (28.4 kIU/L; normal values <18 kIU/L) autoantibodies, and 2 were negative for all anti-phospholipid antibodies investigated by the ELISA method. Thus, the new SCT assay was positive in 87% of those who had a positive result by our LA test and was negative in 93% of those whose results were negative by our LA assay.
All of these patients had histories of thromboembolic disease (4 with venous thrombosis and 2 with arterial thrombosis). These findings can suggest, at least in some patients, a role of SCT as an independent risk factor for thrombosis. Furthermore, the presence of SCT positivity in patients with thromboembolic events reduces, at least in part, the number of patients who are otherwise seronegative for anti-phospholipid autoantibodies. More data and prospective studies are needed to confirm this hypothesis.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
