THROMBOSIS RESEARCH 67; 355-365,1992 0049-3646/92 $5.00 + .OO Printed in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.

SILICA CLOTTIRG TIME (SCT) AS A SCREERIRG FOR DETECTIOR OF TRE LUPUS ARTICOAGULARTS

ARD CORFIRMATORY

TEST

Veena Chantarangkul, Annando Tripodi, Arnaldo Arbini, Pier Mannuccio Mannucci A. Bianchi Bonomi Hemophilia and Thrombosis Center, Institute of Internal Medicine, University and IRCCS Maggiore Hospital, Milano, Italy.

(Received

651992; accepted in revised form 4.6.1992 by Editor G.G. Neri Serneri) (Received by Executive Editorial Office 13.7.1992)

Abstract

Silica Clotting Time (SCT) test We have developed a suitable to screen patients with lupus anticoagulants (LA) and compatible with photo-optical instruments. The SCT results were considered to be positive for LA whenever the clotting times were longer than the upper normal limit at low phospholipid concentration and to be confirmed when the prolonged clotting times were corrected to normal by high phospholipid concentration. We studied plasmas from healthy subjects, patient$ with with acquired known diagnoses of LA, patients deficiencies of blood coagulation and hemophiliacs with anti-factor VIII antibodies. The test was positive for all LA patients, and negative for all non-LA patients except 7 hemophiliacs with anti-factor VIII antibodies. Our data indicate that the SCT is a sensitive test, suitable for screening patients suspected of having LA. Its compatibility with photo-optical instruments makes it a suitable candidate to replace the kaolin clotting time. The contemporaneous performance of SCT at low and high phospholipid concentrations provides screening and confirmation in a single procedure.

Key words: Screening test, LUPUS anticoagulant

355

SILICA CLOTTING TIME AND LAC

Vol. 67, No. 4

INTRODUCTION

The lupus anticoagulants (LA) are inhibitors directed against negatively charged phospholipids (1) that interfere in vitro in phospholipid-dependent clotting tests, such as prothrombin time time (APTT), kaolin (PT), activated partial thromboplastin Russell's viper venom time clotting time (KCT), and dilute (DRWT) (2). The interference causes prolongation of these times not correctable by mixing patient and normal plasma. Despite this apparently simple picture the laboratory diagnosis and characterization of LA is difficult because of the varied specificity and sensitivity of the tests and reagents used, the heterogeneity of LA and the lack of precise diagnostic criteria. The latter problem has recently been tackled by the Subcommittee on Standardization of Lupus Anticoagulant of the Scientific and Standardization Committee (SSC) of the International Society on Thrombosis and Haemostasis (ISTH), which proposed the criteria for diagnosing LA (3). Primary criteria are the abnormality of a phospholipid-dependent clotting test, the demonstration that such abnormality is due to an inhibitor and that the inhibitor is directed against phospholipids. Secondary criteria are the demonstration that the inhibitor is not directed against any single clotting factor and the rapid loss of inhibitory activity upon dilution with saline. The first primary criterion requires a screening test which depends on phospholipids, easy to perform and whose results are easily interpreted. Such requirements are met by the APTT. However, the sensitivity of APTT to LA detection is variable and depends largely on the amount and the conformation of the phospholipids used, the type and concentration of the activator and the buffering capacity (4-7). The KCT has been proposed for detection of LA (8) and is performed by mixing platelet-free plasma with kaolin as activator. The fact that no phospholipids are added in this test should make it more sensitive than APTT to detection of LA. However, due to the high turbidity and the tendency of kaolin to sediment, KCT cannot be measured with photo-optical coagulometers. In addition, KCT is not specific for LA because it does not meet the primary criterion for phospholipids, i.e., demonstration that the inhibitor is directed at phospholipids. The aims of the present study were to develop a screening clotting test for LA detection which employs silica as activator, and thus makes the assay compatible with optical instruments, and to devise in parallel a confirmatory procedure for LA based on testing at two different concentrations of phospholipids. Preliminary results of this study have already been presented (9).

NATERIALS.

MRTRODS AND PATIENTS

Blood was collected in siliconized vacuum tubes (Be&on-Dickinson, Meylan, France) containing one-tenth volume of 129 mM trisodium citrate. The tubes were first centrifuged at 2,500 x g for 15 minutes and the platelet-poor plasma was harvested. Platelet-free plasma was then obtained by filtration

Materials.

Vol. 67, No. 4

SILICA CLOTTING TIME AND LAC

357

through 0.22 micron filters (3), snap frozen and stored at -80' C until testing. Imidazole buffered saline containing 0.05 M imidazole and 0.1 M NaCl, pH 7.35, was used as diluent. A kaolin (Light kaolin, BDH, Poole, UK) suspension (2g/dL in buffer) was used in the KCT. Colloidal Silica (Ortho Diagnostic Systems, Raritan, N.J.) suspension (17.5 mg/dL in buffer) was used in the silica clotting time (SCT) test. A single batch of human lysed washed platelets was prepared according to Triplett (lo), stored at - 80 C and used throughout the study. APTT was measured with the Automated APTT reagent (Organon Teknika, Durham, N.C.). Highly concentrated bovine phospholipids were obtained from Ortho Diagnostic Systems. The total phospholipid concentration, as declared by the manufacturer, was 16.1 mg/ml and the composition was as follows: phosphatidylinositol 0.5 m'g/ml, phosphatidylserine 3.9 mg/ml, phosphatidylethanolamine 6.3 mg/ml and phosphatidylcholine 5.3 mg/ml. DRVVT was performed with reagents provided by American Diagnostica, (New York, N.Y.) Methods. KCT was measured by a manual (tilt tube) technique mixing 0.1 ml plasma and 0.05 ml kaolin suspension. After 3 minutes incubation at 37O C the mixture was recalcified with 0 .l ml 25 mM CaC12 and the clotting time recorded. The SCT test was performed by an automated photo-optical centrifugal coagulometer (ACL 300 R, Instrumentation Laboratory, Milano, Italy), in the research mode to automatically pipette one volume of patient plasma and two volumes of silica suspension, either in buffer alone or in a solution containing the specified concentration of phospholipids. After 3 minutes incubation at 37' C, the mixture was recalcified with 1 volume of 25 mM CaC12 and the clotting time recorded. The results for patient plasmas were expressed either as ratios (patient clotting time/normal clotting time) or clotting times. Anti-cardiolipin antibodies were measured by immunoassay, as described by Loizou et al. (11). Patients. We collected and studied plasmas from healthy subjects, hemophiliacs with anti-factor VIII antibodies, patients on heparin therapy, patients on oral anticoagulant therapy, patients with liver cirrhosis, patients with congenital clotting factor deficiencies of the intrinsic coagulation pathway and patients with a diagnosis of LA according to the revised criteria issued by the SSC Subcommittee for Standardization of LA (3). Briefly, the latter patients had normal or near normal PT and prolonged KCT and APTT. The 1:l mixture of patient plasma and normal platelet free plasma was also longer than the upper limit of the normal range. The presence of LA was confirmed by the platelet neutralization procedure, as described by Triplett (10). The test was regarded as positive whenever the APTT with lysed platelets was at least 3 seconds shorter than the buffer control APTT. A secondary criterion was the lack of parallelism for intrinsic factor assays done at different dilutions. RESULTS of silica. Fig. 1 shows the pattern of changes of SCT with different silica concentrations for plasmas

Qlptinal concentration

Vol. 67, No. 4

SILICA CLOTTING TIME AND LAC

358

from healthy subjects, patients on heparin therapy, patients with LA, and patients with liver cirrhosis. The shortest SCT were obtained when the concentration of silica was in the range 17.535.0 mg/dL. Fiaure

70

35

17.5

1.

Chanqes in Silica Clotting Time at different silica concentrations for plasmas from 12 healthy subjects (closed 3 circles); with patients liver cirrhosis (open squares); with LA 10 (closed squares); 5 heparin on therapy (open triangles).

8.8

SILICA (mg/dL)

Table 1 shows the between-assay coefficients of variation for repeated measurement by SCT and KCT for plasma pools from normal subjects, patients with LA and patients on heparin therapy. The best reproducibility (mean CV = 3.1%, range = 2.0-4.4%) was obtained with a concentration of silica of 17.5 mg/dL. The reproducibility of SCT was better than that obtained with KCT (mean CV = 5.2%, range = 3.7-7.4%). On the basis of these results, a concentration of silica of 17.5 mg/dL was chosen as the most suitable. Table 1. Between-assay coefficient of variation (%) for replicate determination (n=lO) by Kaolin Clotting Time and Silica Clotting Time for plasma pools

Plasma Pools

Kaolin Clotting Time

Silica Clotting Time silica concentrations (mg/dL) 70 17.5 8.8 35

Healthy subjects

3.7

4.7

3.6

3.0

1.5

LA

7.4

8.3

7.8

4.4

4.2

Heparin 4.4 4.0 3.0 2.0 3.6 therapy ~-----~-~~-----~-~_-------____---_____---____----____Mean 5.2 5.6 4.8 3.1 3.1 IUflUenCe of DhosDholiDids. Table 2 shows the pattern of changes of SCT with phospholipid concentrations for plasmas from normal

Vol. 67, No. 4

SILICA CLO-I-I-ING TIME AND LAC

359

subjects, patients with LA, patients on heparin therapy, patients with liver cirrhosis and patients with clotting factor deficiencies of the intrinsic coagulation pathway. For each plasma, the SCT test was performed with addition to the system of increasing dilutions of the stock phospholipid solution (16.1 mg/mL),from 1:lO to l:lOO,OOO (1.61 mg/mL-0.161 ug/mL). A baseline SCT value was obtained by adding buffer instead of phospholipids. Before phospholipid addition, the mean SCT was longer than the upper limit of the normal range (mean + 2SD of healthy subjects) in all groups of patients except those with liver cirrhosis. After phospholipid addition, the pattern of changes was similar for non-LA patients, for whom there was a progressive shortening of SCT with increasing phospholipid concentration from 0 to 161 ug/mL. At higher phospholipid concentrations (1.61 mg/mL), SCT lengthened still more. The pattern of change was quite different for LA patients, in whom the SCT shortened progressively and at a phospho'lipid concentration of 1.61 mg/mL had a mean SCT value very close to that for healthy subjects (52 + 8.6 vs 49 + 3.2 seconds).

2. Changes in Silica Clotting Time expressed as clotting time and ratio (mean + SD) after addition of increasing phospholipid concentrations. Table

Patients

Phospholipid concentrations o:i6i...............i6i....... ilki. 16.1 mg/ml w/ml Wml w/ml

Buffer

Healthy sec. 104 k16.6 subjects ratio 1.0+0.16 (n=20) CV* (16%)

91i10.4 l.O+O.ll (11%)

46+3.0 1.0+0.06 (6%)

36+2.3 1.0+0.06 (6%)

4923.2 1.0+0.07 (7%)

(n=18)

sec. ratio CV

144k45.2 1.4kO.43 (31%)

134239.4 1.520.43 (29%)

88k24.7 1.950.54 (28%)

52k12.0 1.450.33 (23%)

52+8.6 1.120.18 (17%)

Heparin therapy (n=5)

sec. ratio CV

184285.0 1.820.81 (46%)

165k83.4 1.8kO.91 (50%)

7Ok11.8 1.5kO.26 (17%)

54+8.9 1.520.25 (17%)

103528.1 2.1kO.57 (27%)

Liver cirrh. (n=5)

sec. ratio CV

97221.6 0.9tO.21 (23%)

88217.1 1.0+0.19 (20%)

4624.8 1.0+0.11 (11%)

3922.6 l.liO.07 (6%)

73+9.0 1.520.18 (12%)

Factor defic. (n=7)

sec. ratio CV

282k39.7 2.7kO.38 (14%)

245536.5 2.7kO.40 (15%)

99+7.6 2.120.17 (8%)

75+3.9 2.1kO.11 (5%)

10726.8 2.220.14 (6%)

LA

*

cv,

Choice

basis

coefficient of variation expressed as percentage of the oDtima1

of the

results

concentrations

obtained

for

On the addition of

of DhosDholiDids.

SCT

after

360

SILICA

CLOTTINGTIMEANDLAC

Vol.67, No.4

phospholipids, we tried to choose the most suitable concentration of phospholipids to be added to the system to maximize the sensitivity and specificity of SCT for LA detection and confirmation. To do this, we calculated the sensitivity and specificity of SCT in detecting LA by running tests with a pair of phospholipid concentrations (low and high). The SCT results were considered to be positive for LA whenever the percentage corrections of clotting times were higher than the upper limit of the normal limit (95th percentile of the distribution for healthy subjects) that particular pair of phospholipid for concentrations. The percentage correction was calculated by the following equation: % corr. = SCT (low nhosnholinids)-SCT (hiuh nhosnholinids2 X 100 SCT (low phospholipids) The results are reported in Table 3 and show that the sensitivity and specificity were best (100%) when SCT was performed with different pairs of phospholipid two concentrations: 161 ug/mL and 1.61 mg/mL or 16.1 ug/mL and 1.61 mg/mL. On the basis of these results, we have chosen to perform SCT with the combination 16.1 ug/mL and 1.61 mg/dL phospholipid concentrations. Table 3. Sensitivity and specificity of Silica Clotting Time performed at paired phospholipid concentrations.

Buffer versus 1.61 mg/ml w positive/total* %

6/19

0.161 ug/ml 16.1 ug/ml 161 ug/ml versus versus versus 1.61 mg/ml 1.61 mg/ml 1.61 mg/ml

14/19

19/19

19/19

32

74

100

100

36/37

27/37

37/37

37/37

73

100

100

Snecificitv negative/total** %

97

* **Total number of patients with previous diagnoses Total number of non-LA patients.

of

LA.

To assess the ability of SCT to discriminate between LA and non-LA patients, we calculated the percentage correction for each patient. The higher the percentage correction, the more likely the diagnosis of LA. Fig. 2 shows the results for plasmas from healthy subjects, patients

PerfQEaan 1

Vol. 67, No. 4

SILICA CLOlTlNG

TIME AND LAC

361

on heparin therapy, hemophiliacs with anti-factor VIII antibodies, patients on oral anticoagulant therapy and patients with LA. The median percentage correction for healthy subjects was -5.6% (range, from -16.9% to 4.7%). The results were regarded as positive whenever the percentage correction was greater than the upper limit of normal, set at the 95th percentile of the distribution for healthy subjects (4.6%). All patients w:ith LA had percentage corrections greater than the upper limit of normal, suggesting a high degree of sensitivity (100%) and responsiveness. All non-LA patients except 7 of 8 hemophiliacs with anti-factor VIII antibodies were negative.

80 .

40-

g ‘Z g s s

*

20-

. . .

T ” . _______~_____________-----------___.*__-__-______________---------_-_______________.---

0 .

iE

-20 -

a f

i * :

.

-40

. ?? ’

5

-60

. :

.

-80

.

I -100



I Normal (n = 21)

I LUPUS anticoagulants (n = 40)

I

I

Anti - F VIII antibodies (n = 8)

Heparin (n = 10)

I Oral anticoagulants (n I 11)

I Liver cirrhosis (n = 7)

.

uure 2. Percentage correction (see text) of Silica Clotting test at low and high phospholipid concentrations. The horizontal broken line represents the upper limit of normal (95th percentile). Plasmas from these 7 patients were subjected to further investigation to rule out the coexistence of LA. Table 4 shows the results in the SCT, the platelet neutralization proaedure, the DRVVT and the search for anti-cardiolipin antibodies. The results in the platelet neutralization procedure were consistent with the presence of LA in 2 and the DRWT results in 3 patients. 2 of the 7 patients were positive for antiIn addition, cardiolipin antibodies.

SILICA CLOTTING TIME AND LAC

362

Vol. 67, No. 4

Results for patients with anti-factor VIII antibodies and positive Silica Clotting Time

Table 4.

platelet neutral. procedure

anticardiolipin (IgC)

antiSCT PVIII1 corr. % Patients

..DRWT (set).. patient plasma mixture 2

normal range

36-49

36-49

neg.

neg.

c 4.7

C.A.

13

35

74

58

pos.

pos.

L.R.

2

24

53

48

neg.

neg.

M.C.

23

24

47

45

neg.

neg.

M.A.

4

13

n.d.

n.d.

neg.

pos.

S.M.

400

47

57

50

neg.

neg.

P.A.

3

35

57

50

pos.

neg.

P.L

53

35

55

47

neg.

pos.

1 values are expressed in Bethesda units/ml 2 mixture 1:l normal plasma:patient plasma n.d., not done

DISCUSSION

Silica has been successfully used as an activator in the APTT test and demonstrated to be fully compatible with photo-optical instruments (12). In this study we developed screening and confirmatory procedures for LA in which silica is substituted for kaolin and two different concentrations of phospholipids are added. The most sensitive tests for LA are those in which there are no phospholipids or platelets (8). The greatest sensitivity of SCT to LA was not seen when phospholipids were completely omitted from the system (mean ratio = 1.38) (Table 3), but at a phospholipid concentration of 16.1 ug/ml (mean ratio = 1.92) (Table 3). This is probably because the SCT without phospholipids was highly variable within each patient group (median CV = 23%; range 14-49%). In contrast, the variability was smaller when the test was performed at a phospholipid concentration of 16.1 ug/ml (median CV = 11%; range 6-28%) (Table 3), allowing a better discrimination between patients and normal subjects. Hence, a low concentration of phospholipids is necessary to maximize the sensitivity of SCT to the clotting time prolongation induced by LA. Once an anticoagulant substance has been detected in plasma, specific tests are necessary to differentiate LA from inhibitors

Vol. 67, No. 4

SILICA CLOTTING TIME AND LAG

363

directed against single clotting factors. The most specific tests for LA are those for which it is possible to demonstrate correction of the prolonged clotting time upon addition of excess phospholipids to the test system (10, 13-16). The SCT test performed at two different phospholipid concentrations confirmed the presence of LA in all patients with previous diagnoses and differentiated LA from other conditions known to prolong clotting tests, such as oral anticoagulant therapy, liver cirrhosis and heparin therapy (Fig. 2). The specificity of SCT was less good for plasmas from patients with anti-factor VIII antibodies, 7 out of 8 patients being misclassified as having LA. However, 2 of these patients had positive DRVVT and platelet neutralization procedure results and one had positive DRVVT. Both tests are known to be specific for LA (10, 17, 18). In addition, two patients were also positive for anti-cardiolipin antibodies. It is clearly established that LA (19) and anti-cardiolipin antibodies (20, 21) can be present in hemophiliacs. Although it is still a matter of debate whether or not anti-cardiolipin antibodies and LA are the same things (22, 23), it is likely that some of our hemophiliacs with anti-factor VIII antibodies are also positive for LA. In conclusion, we have devised a simple clotting test employing silica as activator and provide evidence that it is sensitive and suitable for screening large numbers of patients suspected of having LA. Its compatibility with photo-optical instruments makes it a suitable candidate to replace the KCT. In addition, using low and high phospholipid concentrations in parallel provides screening and confirmation a single procedure. REFERENCES

1 THIAGARAJAN SHAPIRO, S. and DE MARCO, L. P Monoclonal inhibitor with immundglobulin M coagulation phospholipid specificity. J. Clin. Invest. 66. 397-405, 1980. 2 TRIPLETT D.A. Laboratory diagnosis of lupus anticoagulants. Sem. Thrombos. Haemostas. 16. 182-192, 1990. 3 EXNER, T., TRIPLETT, D.A., TABERNER, D. and MACHIN S.J. Guidelines for testing and revised criteria for lupus anticoagulants. Thrombos. Haemostas. 65, 320-322, 1991. 4 MANNUCCI, P.M., CANCIANI, M.T., MARI, D. and MEUCCI P. The varied sensitivity of partial thromboplastin and prothrombin time of lupus-like reagents in the demonstration anticoagulant. Stand. J. Haematol. 22, 423-432, 1979. 5 BRANDT, J.T., TRIPLETT, D.A., MUSGRAVE, K. and ORR T. The sensitivity of different coagulation reagents to the presence of lupus anticoagulants. Arch. Pathol. Lab. Med. 111. 120-124, 1987. 6 LAZARCHICK, J. and KIZER, J. The laboratory diagnosis of lupus anticoagulants. Arch. Pathol. Lab. Med. 113, 177-180, 1989.

364

SILICA CLOTTING TIME AND LAC

Vol. 67, No. 4

7 KELSEY, I.R., STEVENSON, K.J. and POLLER L. The diagnosis of lupus anticoagulants by the activated partial thromboplastin time. The central role of phosphatidyl serine. Thrombos. Haemostas. 52, 172-175, 1984. 8 EXNER, T., RICKARD, K.A. and KRONEMBERG, H.A. A sensitive test demonstrating lupus anticoagulants and its behavioural patterns. Brit. J. Haematol. 40, 143-151, 1978. 9 CHANTARANGKUL, V., ARBINI, A., CRIPPA, L., TRIPODI, A. and MANNUCCI, P.M. Colloidal silica clotting time (CSCT) as a screening and confirmatory test for detection of lupus anticoagulant. Thrombos. Haemostas. 65. 1259, 1991. 10

TRIPLETT, D.A., BRANDT, J.T. KACZOR, D., and SCHAEFFER J. Laboratory diagnosis of lupus inhibitors: A comparison of the tissue thromboplastin inhibition procedure with a new platelet neutralization procedure. Am. J. Clin. Path. 79, 678-682, 1983.

11

LOIZOU, S., McCREA, J.D., RUDGE, A.C., REYNOLDS, R., BOYLE, C.C. and HARRIS, E.N. Measurement of anti-cardiolipin antibodies by an enzyme-linked immunosorbent assay (ELISA): standardization and quantitation of results. Clin Exn. Immunol 62, 738-745, 1985.

12

CHANTARANGKUL, V., TRIPODI, A. and MANNUCCI, P.M. Evaluation of a fully automated centrifugal analyzer for performance of haemostasis tests. Clin. Chem. 33, 1888-1890, 1987.

13

ALVIN, B.M., BALDWIN, P.E., RICHARDS, R.L. and JACKSON, B.J. The dilute phospholipid APTT: A sensitive assay for verification of lupus anticoagulants. Thrombos. Haemostas. 54, 709-714, 1985.

14

RAUCH, J. and JANOFF, A.S. Distinguishing plasma lupus anticoagulants from antifactor antibodies using hexagonal (II) phase phospholipids. Thrombos. Haemostas. 62, 892-896, 1989.

15

EXNER, T. Comparison of two simple tests for the lupus anticoagulant. Am. J. Clin. Pathol. 83, 215-218, 1985.

16

ROSOVE, M.H., ISMAIL, M., KOZIOL, B.J., RUNGE, A.and KASPER C.K. Lupus anticoagulants: improved diagnosis with a kaolin clotting time using rabbit brain phospholipids in standard and high concentrations. Blood, 472-478, 1986.

17

THIAGARAJAN, P., PENGO, V. and SHAPIRO, S.S. The use of the dilute Russell viper venom time for the diagnosis of lupus anticoagulants. Blood, 869-874, 1986.

18

EXNER, T., PAPADOPOULOS, G. and KOUTTS, J. Use of a simplified dilute Russell's viper venom time (DRVVT) confirms heterogeneity among lupus anticoagulants. Blood Coag. Fibrinol. 1, 259-266, 1990.

Vol. 67, No. 4

SILICA CLOTTING TIME AND LAC

365

19 HAIRE, W.D. The acquired immunodeficiency syndrome and lupus anticoagulant. Ann. Int. Med. 105, 301-302, 1986. 20

BRIEN, W., INWOOD, M. and DENOME, G. Antiphospholipid antibodies in hemophiliacs. Brit. J. Haematol. 67, 270, 1988

21 PANZER, S., STAIN, DUDCZAK, R. and C., HARTL, H., LECHNER, K. Anticardiolipin antibodies are elevated in HIV-l infected haemophiliacs but do not predict for disease progression. Thrombos. Haemostas. 61, 81-85, 1989. 22 HARRIS, E.N., LOIZOU, S., ENGLERT, H., DERUE, G., CHAN, JK, GHARAVI, A.F. and HUGHES, G.R.V. Anticardiolipin antibodies and lupus anticoagulants. Lancet 2, 1099, 1984. 23 TRIPLETT, D.A., BRANDT, J.T., MUSGRAVE, K.A. and ORR, C. Relationship between lupus anticoagulants and antibodies to phospholipids. JAMA 259, 550-554, 1988.

Silica clotting time (SCT) as a screening and confirmatory test for detection of the lupus anticoagulants.

We have developed a Silica Clotting Time (SCT) test suitable to screen patients with lupus anticoagulants (LA) and compatible with photo-optical instr...
755KB Sizes 0 Downloads 0 Views