Avaliação dos testes de mistura e do índice de anticoagulante circulante na detecção de anticoagulantes lúpicos
Diésica Suiane Ferreira; Samuel Ricardo Comar
Universidade Federal do Paraná, Curitiba, Paraná, Brazil.
Samuel Ricardo Comar
Email id: firstname.lastname@example.org
First Submission on 02/12/21;
Last Submission on 02/18/21;
Accepted for publication on 05/16/21;
Published on 12/20/21
Introduction: For the diagnosis of lupus anticoagulants (LAC), the mixing tests (MT) and the index of circulating anticoagulant (ICA) are considered useful to differentiate factor deficiency from inhibitor. However, the clinical usefulness of the percentage of correction of MT (%C-MT) and ICA still needs to be investigated. Objectives: To evaluate the clinical usefulness of %C-TM and ICA of diluted Russells viper venom (dRVVT) and silica clotting time (SCT) screening tests in identifying LAC, as well as to verify their responses to warfarin, enoxaparin, and direct oral anticoagulants (DOACs). Methods: Analysis of 605 patient samples tested for dRVVT and SCT (103 positives for LAC and 502 negatives), 28 using warfarin, 14 using enoxaparin, and 22 using DOACs. Results: The parameters showed the following values of sensitivity, specificity, positive predictive value, negative predictive value, and efficiency, respectively, in the diagnosis of LAC: %C-dRVVTSCREEN- MIX (83.1%; 65.6%; 77%; 73.6%; 79.5%); ICA-dRVVT-SCREEN-MIX (75%; 89.9%; 84.2%; 83.3%; 86.6%); %C-dRVVT-CONFIRM (76.4%; 94.3%; 95%; 72%; 90.3%); %C-SCT-SCREEN-MIX (45.9%; 86.9%; 82.6%; 54.1%; 69.4%); ICA-SCT SCREEN-MIX (78.8%; 90.4%; 85.7%; 85.3%; 90.9%), and %C-SCT-CONFIRM (82.5%; 92.6%; 95%; 75%; 93.9%). The ICA and %C-MT of LAC-positive samples were in most cases significantly higher and lower, respectively, compared to normal, warfarin, enoxaparin, and DOACs. Conclusion: The ICA and %C-MT of dRVVT and SCT can be considered useful and reliable tools in the interpretation of dRVVT and SCT tests in the identification of LAC.
Key words: antiphospholipid syndrome; lupus coagulation inhibitor; thrombosis.
Introdução: Para o diagnóstico de anticoagulantes lúpicos (ALs), os testes de mistura (TM) e o índice de anticoagulante circulante (ICA) são considerados úteis para diferenciar um inibidor de deficiência de fator. No entanto, a utilidade clínica do percentual de correção dos TM (%C-TM) e do ICA ainda necessita de investigações. Objetivos: Avaliar a utilidade clínica dos %C-TM e do ICA dos testes de triagem de veneno da víbora de Russell diluído (dRVVT) e do tempo de coagulação da sílica (SCT) na identificação de ALs, assim como verificar suas respostas frente a varfarina, enoxaparina e anticoagulantes orais diretos (DOACs). Métodos: Análise de 605 amostras de pacientes testados para dRVVT e SCT (103 positivos para ALs e 502 negativos), 28 em uso de varfarina, 14 de enoxaparina e 22 de DOACs. Resultados: Os parâmetros mostraram os seguintes valores de sensibilidade, especificidade, valor preditivo positivo, valor preditivo negativo e eficiência, respectivamente, no diagnóstico de ALs: %C-dRVVT TRIAGEM-MIST (83,1%; 65,6%; 77%; 73,6%; 79,5%); ICA-dRVVT-TRIAGEM-MIST (75%; 89,9%; 84,2%; 83,3%; 86,6%); %C-dRVVT-CONFIRM (76,4%; 94,3%; 95%; 72%; 90,3%); %C-SCT-TRIAGEM-MIST (45,9%; 86,9%; 82,6%;54,1%; 69,4%); ICA-SCT-TRIAGEM-MIST (78,8%; 90,4%; 85,7%; 85,3%; 90,9%) e %C-SCTCONFIRM (82,5%; 92,6%; 95%; 75%; 93,9%). ICA e %C-TM das amostras AL-positivas foram, na maioria dos casos, significativamente maiores e menores, respectivamente, em comparação com normal, varfarina, enoxaparina e DOACs. Conclusão: ICA e %C-TMs do dRVVT e do SCT podem ser considerados ferramentas úteis e confiáveis na interpretação dos testes dRVVT e SCT na identificação de ALs.
Unitermos: síndrome antifosfolipídica; inibidor de coagulação do lúpus; trombose.
Introducción: Para el diagnóstico de anticoagulantes lúpicos (ACL), las pruebas de mezcla [mixing tests (MT)] y el índice de anticoagulante circulante (ICA) se consideran útiles para diferenciar un inhibidor de la deficiencia de factor. Sin embargo, aún debe investigarse la utilidad clínica del porcentaje de corrección del MT (%C-MT) e ICA. Objetivos: Evaluar la utilidad clínica de %C-TM e ICA de las pruebas de detección de veneno de víbora de Russell diluido (dRVVT) y tiempo de coagulación con sílica (SCT) para identificar ACL, así como verificar sus respuestas a warfarina, enoxaparina y anticoagulantes orales directos (ACOD). Métodos: Análisis de 605 muestras de pacientes ensayados para dRVVT y SCT (103 positivos para ACL y 502 negativos), 28 en uso de warfarina, 14 en uso de enoxaparina y 22 en uso de ACOD. Resultados: Los parámetros mostraron los siguientes valores de sensibilidad, especificidad, valor predictivo positivo, valor predictivo negativo y eficiencia, respectivamente, en el diagnóstico de ACL:%C-dRVVT-SCREEN-MIX (83,1%; 65,6%; 77%; 73,6%; 79,5%); ICA-dRVVT-SCREEN-MIX (75%; 89,9%; 84,2%; 83,3%; 86,6%); %C-dRVVT-CONFIRM (76,4%; 94,3%; 95%; 72%; 90,3%); %C-SCT-SCREEN-MIX (45,9%; 86,9%; 82,6%; 54,1%; 69,4%); ICA-SCTSCREEN- MIX (78,8%; 90,4%; 85,7%; 85,3%; 90,9%) y %C-SCT-CONFIRM (82,5%; 92,6%; 95%; 75%; 93,9%). El ICA y %C-MT de las muestras positivas para ACL fueron en la mayoría de los casos significativamente más altos y más bajos, respectivamente, en comparación con los normales, warfarina, enoxaparina y ACOD. Conclusión: El ICA y %C-MT de dRVVT y SCT pueden considerarse herramientas útiles y confiables en la interpretación de las pruebas dRVVT y SCT en la identificación de ACL.
Palabras clave: síndrome antifosfolípido; inhibidor de coagulación del lupus; trombosis.
Lupus anticoagulants (LAC) are acquired autoantibodies, heterogeneous in nature (IgG, IgM, or IgA) and directed against phospholipid (PL) complexes, coagulation factors, plasma and membrane proteins associated with PL(1). They were so named because they were first found in patients with systemic lupus erythematosus (SLE)(2). The term anticoagulant is part of the name because LACs prolong, in vitro, the clotting time of laboratory clotting tests that are dependent on interactions with PL-proteins, such as activated partial thromboplastin time (APTT), silica clotting time (SCT), and dilute Russell’s viper venom time (dRVVT), behaving as acquired inhibitors of coagulation(2). This represents a paradox, as this prolongation of the in vitro clotting times, as is evident in the presence of LACs, is indicative of a bleeding propensity. However, patients with antiphospholipid syndrome (APS) suffer from thrombotic complications. Thus, we can assume that the manifestation of LACs in vitro does not necessarily reflect the in vivo mechanism that causes thrombosis or pregnancy complications. Briefly, LACs prolong in vitro clotting time, which is not reversed by mixing with normal plasma but is reversed when extra phospholipid is added(2-5).
LACs are one of the three primary anti-PL antibodies that are associated with APS, the other two are anticardiolipin (aCL) antibody and anti-β2-glycoprotein-1 (anti-β2GPI) antibody(6-9).
In an attempt to standardize recommended practices for the diagnosis of APS, some studies have agreed that the presence of at least one clinical feature (arterial or venous thrombosis) and one laboratory abnormality (LAC and/or aCL and/or anti-β2GPI) must be observed. Furthermore, positive laboratory tests must be present on two or more occasions after 12 weeks, to exclude transient anti-PL antibodies, which often appear secondary to infectious processes(10-12).
The detection o aCL and anti-β2GPI can be performed using solid-phase enzyme-linked immunosorbent assay (Elisa), as these negatively charged PL are used as antigens, while the search for LACs is performed using PL-dependent coagulation tests(3, 12). The guidelines developed by the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis (SSC-ISTH), by the British Committee for Standards in Hematology (BCSH), and by the Clinical & Laboratory Standards Institute (CLSI) differ, mainly, as regards the way to obtain the reference intervals and cutoff values and the order in which the screening, mixing, and confirmatory tests are carried out(13-15).
Since the laboratory demonstration of LACs has a strong link with thromboembolic events, the search for greater standardization and harmonization of laboratory tests, to circumvent issues related to antibody heterogeneity, poor assay reproducibility, generation of appropriate cutoff values, sensitivity, specificity, and variability of reagents, remains of great importance(16).
Evaluate the results of the dRVVT and SCT tests, in particular, the indexes of circulating anticoagulant (ICA) and the percentages of correction of the mixing tests (%C-MT) and the confirmatory tests (%C-CONFIRM) in the identification of LAC in normal samples and the presence of warfarin, enoxaparin, DOAC, and LAC, to contribute to the choice of appropriate parameters for the laboratory investigation of LAC, according to the main international guidelines.
The work was carried out at the Hematology Laboratory of the Unidade de Laboratório de Análises Clínicas do Complexo Hospital de Clínicas da Universidade Federal do Paraná (ULAC-CHC-UFPR) after approval by the local Research Ethics Committee, recognized by the Brazilian National Research Ethics Committee [Comitê Nacional de Ética em Pesquisa (CONEP)].
The results of the dRVVT and SCT tests from 605 patients (493 women and 112 men), with a mean age of 40.8 ±
15.1 years (range: 20 days to 82 years), were retrospectively accessed. The most prevalent clinical indications for ordering the investigation of lupus anticoagulants were: APS, SLE, thrombophilia, spontaneous abortions, investigation of prolonged APTT, arthritis, cerebrovascular accident, encephalic vascular accident, collagenosis, dermatomycosis, epilepsy, sclerosis, fibrosis, nephritis, thrombocytopenia, and bone marrow transplantation. Among the CHC-UFPR Service Units that ordered the laboratory tests the most prevalent was the rheumatology unit with 28.4% of the laboratory tests orders; followed by neurovascular unit (14%); neurology (7.6%); coagulation (4.8%); hematology (4.6%); stroke unit (4.5%); infertility (3.8%); thrombocytopenia outpatient clinic (3.6%); nephrology (3.6%); internal medicine unit (2.3%); sclerosis (1.8%); rearguard beds unit (1.5%); arthritis (1.3%); adult intensive care unit (1.2%) and other outpatient clinics (16.9 %).
Samples, handling and storage
The analyzed plasmas were collected in Vacuette tubes (Greiner Bio-one Ltd, Stonehouse, UK) containing 0.109 mol/l (3.2%) of sodium citrate. The samples were double centrifuged at 4,000 rpm for 15 minutes, at room temperature, to obtain a plasma poor in platelets (< 10 × 103/µl), since the platelet membranes contain PL that contribute to generating false-positives in LAC tests(15). In cases where the tests were not performed on the same day, the plasmas were frozen at -40°C for a maximum of one week. Frozen samples were quickly thawed (five minutes) in a water bath (37°C) and gently shaken just before analysis. The pooled normal plasmas for the mixing tests was obtained from at least 40 healthy volunteers.
Laboratory determination of lupus anticoagulants
For the dRVVT test, the integrated HemosIL dRVVT-SCREEN test (LA1 LACs-sensitive)/dRVVT-CONFIRM (LA2 with a high concentration of PL, LACs-insensitive) (IL Company, Bedford, MA, USA) was used. For the SCT test, which is an APTT-based assay and which was first described by Chantarangkul (1992)(17), the integrated HemosIL SCT-SCREEN test (LACs-sensitive)/SCT- CONFIRM (LACs-insensitive) was used (IL Company, Bedford, MA, USA), which uses colloidal silica as an activator in the presence of synthetic PL that make up the cephalin component of the reagent. The clotting times of the screening, screening mix, and confirmatory tests were converted to normalized ratios (patient clotting time/pooled normal plasma clotting time), and the reference intervals (IR) and the cutoff limits of the normalized ratios were defined locally, the latter was defined as > 1.20 for dRVVT (screening, screening mix, and confirmatory), and > 1.16 for SCT (screening, screening mix, and confirmatory), respectively. Coincidentally, these values were the same as those recommended by the reagent manufacturer. All clotting times, from which the cutoff limits were generated presented a Gaussian distribution and were locally defined using plasma from normal donors following the recommendations of the SSC-ISTH(13), using the 99th percentile, which is equivalent to + 2.3 standard deviations (SD). The samples were analyzed in an ACL TOP 300 (IL Company, Bedford, MA, USA) automated coagulation analyzer by coagulometric method with optical detection of clot formation. Our hematology laboratory followed the recommendations of the SSC-ISTH(13) to detect LAC, as the following order, according to the cases: screening, screening mix, and confirmatory. The collected and evaluated data, as well as their attainment calculations, are shown in Table 1. Before all the analytical runs, imprecision was verified using normal and positive controls for LAC.
Samples with altered screening test results (ratio > 1.20 for dRVVT-SCREEN and > 1.16 for SCT- dRVVT, as specifications of the HemosIL, IL dRVVT, and HemosIL SCT kits) were further analyzed by the respective mixing test, which consisted of adding pooled normal plasmas to the patients’ plasma, in a 1:1 ratio. If there was no correction of the respective mixing test (ratio > 1.20 for dRVVT-SCREEN-MIX and > 1.16 for SCT-SCREEN-MIX), the respective confirmatory test was performed. In the confirmatory step, reagents with a high concentration of PL, capable of proving the phospholipid dependence of the possible antibodies involved in the previous steps, were used. Finally, samples were considered positive for the presence of LACs when any of the screening/ confirmatory ratios of the test systems used presented results above the cutoff values (> 1.20 for dRVVT-SCREEN/CONFIRM and > 1.16 for SCT-SCREEN/CONFIRM). Therefore, samples were considered positive for LACs when they were positive in the screening, mixing and confirmatory step in at least one of the testing systems. Based on this gold standard for classifying samples between positive and negative for LACs, the following parameters were evaluated in relation to their measures of accuracy and agreement as diagnostic tests: patient/normal control ratio dRVVT and SCT screening (dRVVT-SCREEN and SCT-SCREEN); ICA of dRVVT and SCT mixing tests (ICA-dRVVT-MIX and ICA- SCT-MIX)(19); the percentages of correction of the dRVVT and SCT screening tests (%C-dRVVT-SCREEN-MIX and %C-SCT-SCREEN- MIX); and the percentages of correction of the dRVVT and SCT confirmatory tests (%C-dRVVT-CONFIRM and %C-SCT-CONFIRM).
Measures of diagnostic accuracy and agreement
Sensitivity, specificity, efficiency, negative predictive value (NPV), positive predictive value (PPV), negative likelihood ratio (LR-), positive likelihood ratio (LR+), and the Youden index were obtained using the Statistical Analysis software for Windows V.1.8(20).
Data were analyzed using Microsoft Excel 2016 software and R v.4.0.0 software (R Foundation for Statistical Computing). Receiver operating characteristic (ROC) curves were performed using the plotROC v.2.2.1 package(21), to obtain the best cutoff limits for diagnostic discrimination. Box and whisker plots to show the distributions and centers of the data sets were performed in the MEDCALC® software version 22.214.171.124 (MedCalc Software, Mariakerke, Belgium). Comparison of data for various parameters was performed using the Mann-Whitney U test. Values with p ≤0.05 were considered statistically significant.
From the 605 patients included in the study, 476 (78.7%) did not take anticoagulants, 78 (12.9%) took them, and 51 (8.4%) did not have information about the use of anticoagulants. Among the 78 patients who took anticoagulants, there is the following division: warfarin (28), rivaroxaban (16), enoxaparin (14), dabigatran (4), apixaban (2) and other drugs (24), such as acid acetylsalicylic, clopidogrel, etc.
Regarding the investigation of LACs, from the 605 patients included in the study, 262 (43.3% of the total) demonstrated prolongation in the screening stages: 94 (15.5% of the total) with both prolonged dRVVT-SCREEN and SCT-SCREEN; 43 (7.1% of the total) only with prolonged dRVVT-SCREEN, and 125 (20.7% of the total) only with prolonged SCT-SCREEN. The remaining 343 (56.7% do total), of the total) did not demonstrate a prolongation in the screening stage in any of the tests. From the 137 (22.6% of the total) samples with prolonged dRVVT-SCREEN, 91 (15% of the total) resulted in correction in the mixing step, and 46 (7.6% of the total) did not correct in the mixing step. Regarding the SCT, from the 219 (36.2% of the total) samples with prolonged SCT- SCREEN, 117 (19.3% of the total) resulted in correction in the mixing step, and 102 (16.9% of the total) did not correct in the mixing step. The corrections in the mixing step demonstrated that the initial prolongation in the screening step were not due to the presence of coagulation inhibitors with characteristics compatible with LACs, while the non-corrections in the mixing step, in a total of 175 patients, were submitted to the confirmation step, which resulted in 103 (17% of the total) cases compatible with the presence of LACs (41 by dRVVT and SCT; 52 by SCT only; and 10 by dRVVT only).
The diagnostic test evaluation results of the parameters derived from dRVVT and SCT are illustrated in Table 2. In this evaluation, specificity indicates the rate of truly negative tests for LACs that were correctly classified by the parameter concerned, according to the test system used. Sensitivity correlated with the rate of truly positive tests for LACs that were correctly classified by the parameter concerned. Efficiency demonstrates the total percentage of correct ratings, whether positive or negative for LACs. PPVs indicate the rate of truly positive tests for LACs among all tests that were positive (true positives and false positives). NPVs indicate the proportion of truly negative tests for LACs among all tests that were negative (true negatives and false negatives). Positive likelihood ratios express how many times it is more likely to find a positive result for LACs in diseased patients compared to non-diseased patients. Negative likelihood ratios express how many times it is more likely to find a negative result for LACs in diseased patients when compared to non- diseased patients. Finally, the Youden index, which is a function of both sensitivity and specificity, has values ranging from -1 to 1, and has a value of zero when a diagnostic test provides the same rate of positive results for groups with and without presence of LACs, that is, the test is useless. A value closer to 1 indicates that there are no false- positive or false-negative results, that is, the test is perfect. Table 2 also shows the best cutoff values for each parameter, as shown by the ROC curves analysis, which were used as the basis for the calculations of the performance characteristics presented.
From the 51 positive cases for LACs through dRVVT, it was observed that: in 48 (94.1%), the %C-dRVVT-SCREEN-MIX was ≤ 70%; and in 42 (82.4%) was ≤ 61.1%. The ICA-dRVVT-MIX showed a value of ≥ 15.6% in 32 (62.7%) cases and ≥ 10.9% in 44 (86.3%) cases. The %C-dRVVT-CONFIRM was ≥ 32.8% in 41 (80.4%) cases. As for SCT, which investigates the intrinsic coagulation pathway, it was shown that from 93 positive cases for LACs: in 87 (93.5%), the %C-SCT-SCREEN-MIX was ≤ 70%; and in 82 (88.2%) was ≤ 64.6%. The ICA-SCT-MIX showed a value of ≥ 15.6% in 62 (66.7%) cases and ≥ 13.4% in 75 (80.6%) cases among the positives. Finally, the %C-SCT-CONFIRM was ≥ 29% in 79 (84.9%) cases among the positives. ROC curve analyzes for %C-dRVVT-SCREEN-MIX, ICA-dRVVT- MIX, %C-dRVVT-CONFIRM, %C-SCT-SCREEN-MIX, ICA-SCT-MIX, and %C-SCT-CONFIRM, as well as the demonstration of the best cutoff limits for each parameter, are represented in the Figure 1.
The %C-MT, ICA and %C-CONFIRM among normal patients, using warfarin, enoxaparin, DOACs and samples positive for LACs were compared for dRVVT and SCT (Figure 2), and the means in the %C-dRVVT-SCREEN-MIX test were 72.9%; 77.6%; 77.6%; 71.1%; and 44.9%; in ICA-dRVVT-MIX they were 5.9%; 6.9%; 6.7%; 9.5%; and 20.5%; in %C-dRVVT-CONFIRM they were 22.6%; 44.3%; 36.7%; 17.7%; and 41.8%; in %C-SCT-SCREEN-MIX they were 58.7%; 90.7%; 67.6%; 51.4%; and 39.7%; in ICA-SCT-MIX they were 8.1%; 12.4%; 8.1%; 11.7%; and 24.6%; and in %C-SCT- CONFIRM they were 18.9%; 35.3%; 14.3%; 18.9% ; and 43%, respectively, for normal patients using warfarin, enoxaparin, DOACs, and positive samples for LACs. In general, %C-MT were significantly lower in LAC-positive samples compared to normal using warfarin, enoxaparin, and DOACs. ICA values were significantly higher in LAC-positive samples compared to normal, warfarin, enoxaparin, and DOACs; and C-CONFIRM values were significantly higher in LAC-positive samples compared to normal, warfarin, enoxaparin, and DOACs.
The presence of LACs is of great importance in vascular medicine, since their detection on two occasions, within a 12-week interval, in patients with venous or arterial thromboembolism, constitutes the first step in the diagnosis of APS. In these diseases, thrombotic events are common, making the detection of LACs crucial in identifying these patients. On the other hand, despite the evolution of laboratory medicine in terms of standardization of tests to determine LACs, there is still a low performance in the efficiency of tests to detect LACs, which makes the accurate diagnosis of APS more challenging. This is largely due to pre-analytical and analytical factors (characteristics of the reagents and principle of detection of clot formation), in addition to the heterogeneity of anti-lupus antibodies and their forms of interaction with proteins associated with phospholipids, which makes no LAC test able to detect all LACs. As a result, it is recommended to perform at least two tests with different principles to exclude/include the Diésica Suiane Ferreira; Samuel Ricardo Comar presence of LACs. The research of LACs in our laboratory involves the use of two testing systems: the dRVVT and the SCT. The dRVVT is more sensitive to β2GPI, -dependent LACs, while tests that use the intrinsic coagulation pathway, such as SCT and APTT, seem to be more sensitive to the LACs that react with prothrombin(22-24).
The diagnostic criteria for LACs have not changed much since their introduction by Brandt et al. (1995)(25): (1) screening test, demonstrating the prolongation of a phospholipid-dependent clotting time; (2) mixing test, confirming the presence of an inhibitor and excluding a clotting factor deficiency; and (3) confirmatory test, confirming that the inhibitor is phospholipid-dependent and not directed against a specific clotting factor. In this study, we used the SSC-ISTH recommendations(13) (screen-mix-confirmatory). However, more recent international guidelines state that integrated LAC detection tests do not require the mixing tests execution(15), a fact that led to a further study to verify the clinical importance of mixing tests in a thrombotic risk context, in which it has been shown that even negative mixing tests, that is, which were corrected, were also associated with significant thrombotic risk(26).
A issue that arises is whether we need more data and parameters, which are more sensitive in the screening or more specific in the confirmatory, to help interpret the results. Some studies claim that one of the greatest issues in diagnosing APS is not the lack of sensitivity, but the lack of specificity(27-30). In screening tests for LACs, sensitivity is more important than specificity, as high sensitivity prevents false-negative results. However, with high sensitivity, there will always be false-positive results in screening and screening mixture tests, which is the main reason why confirmatory tests are performed, for which a high specificity is desired, that is, exclude with confidence the negative LACs.
In this study, the parameters %C-dRVVT-SCREEN-MIX, ICA- dRVVT-MIX, and ICA-SCT-MIX showed high sensitivity (83.1%, 75%, and 78.8%, respectively) in detecting cutoff LACs used (61.1%, 10.9%, and 13.4%, respectively), which were determined through the analysis of ROC curves. The %C-SCT-SCREEN-MIX showed low sensitivity (45.9%) for the cutoff limit of 64.6%, therefore, less suitable as a screening parameter for samples containing LACs. A possible explanation for this result would be the fact that, in this study, the RI of the clotting times that generated the normalized ratios used the 99th percentile, as recommended by the SSC-ISTH which may have generated more false-negative results (low sensitivity) that would be generated from the 97.5th percentile, as recommended by the CLSI(15) and by the BCSH(14). Hong et al. (2012)(26), using the ACL TOP coagulation analyzer and the same reagents used in this study, they obtained sensitivity values of 89.4%; 55.3%; 71.8%, and 13.3% for %C-dRVVT-SCREEN- MIX, %C-SCT-SCREEN-MIX, ICA-dRVVT-MIX, and ICA-SCT-MIX, respectively, considering the cutoff limits of 60, 1%; 67%; 15.7% and 42.8%, respectively. Except for the cutoff limit of the ICA-SCT- MIX, the other cutoff limits of this study were similar to those by Hong et al. (2012)(26). In this regard, it is recommended that each laboratory determine their cutoff limits.
Confirmatory tests for LACs are based on comparing the clotting times of a reagent insensitive to LACs, which contains extra amount of PL in the test system, which binds to the LACs, neutralizing them and correcting the clotting times to next to normality(13). In this study, the confirmatory parameters %C-dRVVT-CONFIRM and %C-SCT-CONFIRM showed high specificity, as expected and desirable in confirmatory tests, to reliably exclude truly negative samples. They also showed efficiency and the high Youden index, confirming the fact that they are useful in distinguishing positive and negative samples for LACs, providing few false-positives and false-negative results. In general, the parameters derived from dRVVT e SCT, screening and confirmatory, evaluated in this study, showed acceptable sensitivities and specificities, making them able to assist in the diagnosis of LACs and, therefore, compose the final reports to clinicians. To improve the relevant information that aids in the interpretation of the results, in the Appendix, a way to report the results is suggested, including the parameters presented in Table 2.
In this study, five dRVVT-positive patients and ten SCT- positive patients were observed for LACs with strong evidence of the presence of the lupus cofactor phenomenon, in which there were negative values of %C-dRVVT-SCREEN-MIX and %C-SCT- SCREEN-MIX (Figure 2). As described by Loeliger (1959)(31), this phenomenon occurs when a moderately prolonged clotting time is initially observed in screening tests, but when mixed with pooled normal plasmas, they show clotting times paradoxically longer than in pure plasma, also called plasma index. This phenomenon is not observed in many cases of positive LACs, and its explanation may be due to the fact that some patients with LACs would need the presence of a cofactor to then express their full anticoagulant activity. This cofactor, absent in the patient and present in the pool, may be prothrombin (factor II) and β2GPI(31, 32).
We compared the %C-TM, ICA, and %C-CONFIRM values of dRVVT and SCT among normal patients using warfarin, enoxaparin, DOACs, and LAC-positive samples (Figure 2). The ICA was proposed by Rosner et al. (1987)(18) as able to distinguish between LAC- positive and normal samples or factor-deficient samples by mixing with combined normal plasma, even considering the dilution factor of the mixing test that can decrease the reactivity of the LACs. As a result, the clotting times of dRVVT and SCT in mixed samples remain longer in LAC-positive samples than in normal ones, and thus the ICA has a high value and the %C-MT have a low value obtained with the respective formulas when testing a LAC-positive sample. On the other hand, when testing samples deficient in some clotting factor, the clotting factors are added causing the clotting times of dRVVT and SCT to be shortened, generating a low ICA and high %C-MT.
In the present study, we demonstrate that %C-MT, ICA and %C-CONFIRM of dRVVT and SCT were significantly and respectively lower, higher, and higher in LAC-positive samples compared to normal samples and using warfarin, enoxaparin, and DOACs. Kumano et al. (2014)(19) investigated the usefulness of the ICA calculated from the clotting times of various brands of APTT reagents and demonstrated the occurrence of significantly higher ICA values in LAC-positive samples compared to normal, warfarin, and hemophiliac patients, concluding that the ICA was able to distinguish the LAC-positive samples from the others, except for samples with heparin. In contrast, Moore et al. (2016)(33), in a large cohort of samples positive for LACs, demonstrated that the cutoff limits for clotting times specifically determined for the dRVVT and APTT mixing tests were superior to the ICA in detecting the in vitro inhibition of LACs.
The SSC-ISTH guidelines can be interpreted ambiguously concerning the fact that some may consider the use of clotting times or ICA for evaluating the mixing test, while others would use the normalized ratio of the screening mixture step (clotting time of the patient/clotting time of pooled normal plasmas) as the correct interpretation(34). In this study, the normalized ratios in the screening, mixing, and confirmatory stages were considered as the reference in the interpretation of the results, and the evaluation of the other parameters was carried out by comparing them with the normalized ratios. The %C-DRVVT-SCREEN-MIX and ICA-DRVVT- MIX showed sensitivities close to those found for the normalized ratios of dRVVT screening and dRVVT screening mixture, whereas the sensitivities of %C-SCT-SCREEN-MIX and ICA-SCT-MIX were lower and higher, respectively than those observed in the normalized ratios of SCT screening and SCT screening mixture. Although there is no gold standard test to detect the presence of LACs, the results presented in this study showed satisfactory sensitivities and specificities for the parameters evaluated and, based on these observations, we would recommend its use in the routine analysis of LACs.
In the present study, %C-dRVVT-SCREEN-MIX, ICA-dRVVT- MIX, ICA-SCT-MIX, and %C-SCT-CONFIRM in the enoxaparin group were significantly different from those observed in the LAC- positive group, and one possible explanation would be due to the presence of polybrene in the dRVVT and SCT reagents used, which has a neutralizing effect of heparin in coagulation tests, since the test sample does not contain a concentration of heparin, whether unfractionated or of low molecular weight, above 1 U/ml(35). The %C-dRVVT-SCREEN-MIX, %C-SCT-SCREEN-MIX, ICA-dRVVT-MIX and ICA-SCT-MIX satisfactorily distinguish the samples from the group using warfarin from the LAC-positive group, showing the diagnostic efficiency of these parameters, although studies have already shown that warfarin can change the results of the dRVVT and SCT screening tests due to the reduction in the levels of vitamin K-dependent factors (factors II, VII, IX, and X)(35). The group using DOACs was satisfactorily distinguished by %C-dRVVT- SCREEN-MIX, ICA-dRVVT-MIX, ICA-SCT-MIX, and %C-SCT-CONFIRM, although studies have already shown that dabigatran and rivaroxaban prolong the clotting times of dRVVT and SCT(35).
We cannot fail to emphasize that this study has some limitations, such as the low number of cases that were taking anticoagulants, in addition to the fact that the retrospective nature of data collection is not complete, as it depends on the sending of information about the use of medication in orderings. It is unlikely that all anticoagulant drugs were captured in the database; likewise patients in the normal group may also have used medication. Blood concentrations of DOACs were not routinely measured, and data on drug administration time were not available. Therefore, it was not possible to make correlations between drug levels and test positivity rates. Please be aware that if the search for LACs is performed while using anticoagulants, an observation must be included after the test, stating that these agents may interfere with routinely used coagulation assays, rendering the interpretation invalid and, therefore, suggest repeating the test after at least one week after the end or interruption of treatment.
The accurate diagnosis of LACs is of great importance, as it is considered a risk factor for thrombosis. Positive results from LACs have a direct impact on the duration of anticoagulant therapy in patients who develop thrombosis. Unfortunately, due to the heterogeneity of LACs, there is still no diagnostic test that can qualitatively and quantitatively identify all types of LACs, which does not allow the establishment of a gold standard diagnostic test. As a result, several types of tests are used; each one with different parameters and their variables, a fact that makes it crucial to have appropriate knowledge of their clinical and laboratory interferences and limitations.
We recommend the use of the parameters evaluated in this study, as they proved to be useful tools in the diagnosis of LACs. However, we also recommend caution in some situations where it may be difficult to separate LAC-positive samples from patients who are on anticoagulant therapy. %C-MT, ICA, and %C-CONFIRM of dRVVT and SCT were useful for the diagnosis of LACs, as they could efficiently distinguish LAC-positive samples from normal samples.
The authors would like to thank the team at the CHC-UFPR hematology laboratory for their assistance in conducting this study.
- Rosborough TK, Jacobsen JM, Shepherd MF. Factor X and factor II activity levels do not always agree in warfarin-treated lupus anticoagulant patients. Blood Coagul Fibrinolysis. 2010; 21(3): 242-4.
- Pengo V, Bison E, Banzato A, Zoppellaro G, Jose SP, Denas G. Lupus anticoagulant testing: diluted Russell viper venom time (dRVVT). Methods Mol Biol. 2017; 1646: 169-76.
- Sangle NA, Smock KJ. Antiphospholipid antibody syndrome. Arch Pathol Lab Med. 2011; 135(9): 1092-6.
- Levine JS, Branch DW, Rauch J. The antiphospholipid syndrome. N Engl J Med. 2002; 346(10): 752-63.
- Conley C, Hartmann RC. A hemorrhagic disorder caused by circulating anticoagulant in patients with disseminated lupus erythematosus. J Clin Invest. 1952; 31: 621-2.
- Hughes GR. Thrombosis, abortion, cerebral disease, and the lupus anticoagulant. Br Med J (Clin Res Ed). 1983; 287(6399): 1088-9.
- Abreu JSF, Santos AO, Medeiros JR. N, Gouvêa CP. Models for releasing the lupus anticoagulant test. J Bras Patol Med Lab. 2018; 54(3): 153-7.
- Fujieda Y, Atsumi T. Has lupus anticoagulant testing had its day? Nat Rev Rheumatol. 2019; 15(6): 324-5.
- Schreiber K, Sciascia S, de Groot PG, et al. Antiphospholipid syndrome. Nat Rev Dis Primers. 2018; 11(4): 17103.
- Bertolaccini ML, Amengual O, Andreoli L, et al. 14th International congress on antiphospholipid antibodies task force. Report on antiphospholipid syndrome laboratory diagnostics and trends. Autoimmun Rev. 2014; 13(9): 917-30.
- Miyakis S, Lockshin M, Atsumi T, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost. 2006; 4(2): 295-306.
- Rand JH. The antiphospholipid syndrome. Hematology Am Soc Hematol Educ Program. 2007; 136-42.
- Pengo V, Tripodi A, Reber G, et al. Update of the guidelines for lupus anticoagulant detection. Subcommittee on lupus anticoagulant/antiphospholipid antibody of the Scientific and Standardisation Committee of the International Society on Thrombosis and Haemostasis. J Thromb Haemost. 2009; 7(10): 1737-40.
- Keeling D, Mackie I, Moore GW, Greer IA, Greaves M. Bristish Committee for Standards in Haematology. Guidelines on the investigation and management of antiphospholipid syndrome. Br J Haematol. 2012; 157(1): 47-58.
- Clinical and Laboratory Standards Institute (CLSI). Laboratory testing for the lupus anticoagulant: approved guideline. Wayne, PA: CLSI Document H60-A; 2014.
- Kumano O, Moore GW. Ruling out lupus anticoagulants with mixing test-specific cutoff assessment and the index of circulating anticoagulant. Res Pract Thromb Haemost. 2019; 3(4): 695-703.
- Chantarangkul V, Tripodi A, Arbini A, Mannucci PM. Silica clotting time (SCT) as a screening and confirmatory test for lupus anticoagulants. Thromb Res. 1992; 67(4): 355-65.
- Rosner E, Pauzner R, Lusky A,Modan M, Many A. Detection and quantitative evaluation of lupus circulating anticoagulant activity. Thromb Haemost. 1987; 57(2): 144-7.
- Kumano O, Ieko M, Naito S, Yoshida M, Takahashi N, Suzuki T, Aoki T. Verification of the guidelines for lupus anticoagulant detection: usefulness of index for circulating anticoagulant in APTT mixing test. Thromb Res. 2014; 134(2): 503-9.
- Braile DM, Godoy MF. Programa cálculos estatísticos for Windows V.1.8. 1999.
- Sachs MC. plotROC: a tool for plotting ROC Curves. J Stat Softw. 2017; 79(2): 1-19.
- Tripodi A, Cohen H, Devreese KMJ. Lupus anticoagulant detection in anticoagulated patients. Guidance from the Scientific and Standardization Committee for lupus anticoagulant/antiphospholipid antibodies of the International Society on Thrombosis and Haemostasis. J Thromb Haemost. 2020; 18(7): 1569-75.
- Pengo V, Biasiolo A, Rampazzo P, Brocco T. dRVVT is more sensitive than KCT or TTI for detecting lupus anticoagulant activity of anti-beta2- glycoprotein I autoantibodies. Thromb Haemost. 1999; 81(2): 256-8.
- Pengo V, Bison E, Denas G, Jose SP, Bracco A, Banzato A. The paradox of the lupus anticoagulant: history and perspectives. Semin Thromb Hemost. 2014; 40(8): 860-5.
- Brandt JT, Triplett DA, Alving B, Scharrer I. Criteria for the diagnosis of lupus anticoagulants: an update. On behalf of the Subcommittee on Lupus Anticoagulant/Antiphospholipid Antibody of the Scientific and Standardisation Committee of the ISTH. Thromb Haemost. 1995; 74(4): 1185-90.
- Hong SK, Hwang SM, Kim JE, Kim HK. Clinical significance of the mixing test in laboratory diagnoses of lupus anticoagulant: The fate of the mixing test in integrated lupus anticoagulant test systems. Blood Coagul Fibrinolysis. 2012; 23(8): 739-44.
- Devreese K. Antiphospholipid antibody testing and standardization. Int J Lab Hematol. 2014; 36(3): 352-63.
- Favaloro EJ, Wong RC. Antiphospholipid antibody testing for the antiphospholipid syndrome: a comprehensive practical review including a synopsis of challenges and recent guidelines. Pathology. 2014; 46(6): 481-95.
- Tay Za K, Jayaranee S, Shanmugam H. Practice, and performance of lupus anticoagulant tests: a single centre experience. Malays J Pathol. 2020; 42(1): 51-7.
- Devreese KMJ. Evaluation of a new silica clotting time in the diagnosis of lupus anticoagulats. Thromb Res, 2007; 120(3): 427-38.
- Loeliger A. Prothrombin as cofactor for circulating anticoagulant in systemic lupus erythematosus? Throm Diath Haemorrh. 1959; 3: 237-56.
- Tripodi A. To mix or not to mix in lupus anticoagulant testing? That is the question. Semin Thromb Hemost. 2012; 38: 385-9.
- Moore GW, Culhane AP, Daw CR, Noronha CP, Kumano O. Mixing test specific cut-off is more sensitive at detecting lupus anticoagulants than index of circulating anticoagulant. Thromb Res, 2016; 139(3): 98-101.
- Depreter B, Devreese KMJ. Differences in lupus anticoagulant final conclusion through clotting time or Rosner index for mixing test interpretation. Clin Chem Lab Med. 2016; 54(9): 1511-6.
- Seheult JN, Meyer MP, Bontempo FA, Chibisov I. The effects of indirect- and direct-acting anticoagulants on lupus anticoagulant assays: a large, retrospective study at a coagulation reference laboratory. Am J Clin Pathol. 2017; 147(6): 632-40.
APPENDIX – Report template for lupus anticoagulant research (with explanatory observations and comments)
dRVVT – SCREENING (patient/control pooled normal plasma ratio): ……………
SCT or APTT LAC-SENSITIVE – SCREENING (patient/control pooled normal plasma ratio): …….
SCREENING TEST RESULTS ARE POTENTIALLY SUGGESTIVE OF LUPUS ANTICOAGULANT (LAC) WHEN PATIENT/NORMAL RATINGS ARE ABOVE THE REFERENCE INTER- VAL (RI).
MIXING TEST (dRVVT)
dRVVT – MIXTURE (patient/control pooled normal plasma ratio): ……………
% CORRECTION dRVVT – MIXTURE: ……………
ICA dRVVT – MIXTURE: ……………
MIXING TEST (SCT or APTT LAC-SENSITIVE)
SCT/APTT LAC-INSENSITIVE – MIXTURE (patient/control pooled normal plasma ratio): ………….
SCT % CORRECTION or APTT LAC-SENSITIVE – MIX: ……………
ICA SCT or APTT LAC-SENSITIVE – MIXTURE: ……………
MIXING TEST RESULTS ARE SUGGESTIVE OF LAC WHEN THE % CORRECTION IS BELOW 70% ON MIXING TESTS PROVIDEDPOSITIVE SCREENING FOR AT LEAST ONE TEST (dRVVT, SCT, APTT) OR WHEN ICA IS INCREASED OR PATIENT/NORMAL RATIO ARE ABOVE RI.
dRVVT – CONFIRMATORY (patient/control pooled normal plasma ratio): ……………
dRVVT % CORRECTION – CONFIRMATORY: ……………
dRVVT SCREENING/CONFIRMATORY ratio: ……………
CONFIRMATORY (SCT or APTT HIGH CONCENTRATION of PL)
SCT or APTT HIGH CONCENTRATION PL – CONFIRMATORY (patient/control pooled normal plasma ratio): ……………
SCT % CORRECTION or APTT HIGH CONCENTRATION PL – CONFIRMATORY: ……………
SCT RATIO or APTT HIGH CONCENTRATION of PL SCREENING/CONFIRMATORY: ……………
CONFIRMATORY TEST RESULTS ARE COMPATIBLE WITH THE PRESENCE OF LAC WHEN THE SCREENING/CONFIRMATORY RATIO ARE ABOVE RI, SINCE THEY HAVE PROVIDED POSITIVE SCREENING AND CONFIRMATORY OF CIRCULANT ANTICOAGULANT THROUGH MIXING TESTS.
OPTIONALLY CONFIRMATORY MIXTURE (dRVVT)
dRVVT – CONFIRMATORY MIXTURE (patient/control pooled normal plasma ratio): ……………
% CORRECTION dRVVT – CONFIRMATORY MIXTURE: ……………
dRVVT SCREENING MIXTURE/CONFIRMATORY MIXTURE RATIO: ……………
Diésica Suiane Ferreira; Samuel Ricardo Comar
CONFIRMATORY MIXTURE (SCT or APTT HIGH CONCENTRATION PL)
SCT or APTT HIGH CONCENTRATION PL – CONFIRMATORY MIXTURE (patient/control pooled normal plasma ratio): ………….
% CORRECTION SCT or APTT HIGH CONCENTRATION PL – CONFIRMATORY MIXTURE: ……………
SCT RATIO or APTT HIGH CONCENTRATION of PL SCREENING MIXTURE/ CONFIRMATORY MIXTURE:…………………
CONFIRMATORY MIXTURE TEST RESULTS ARE COMPATIBLE WITH THE PRESENCE OF LAC WHEN THE SCREENING/CONFIRMATORY RATIO ARE ABOVE RI, SINCE THEY HAVE PRESENTED POSITIVE SCREENING AND CONFIRMATORY OF CIRCULANT ANTICOAGULANT THROUGH MIXING TESTS. CONFIRMATORY MIXTURE MAY OPTIONALLY BE PERFORMED IN AN ATTEMPT TO CONFIRM LAC IN A PATIENT WITH COAGULATION FACTOR DEFICIENCY DUE TO ANTICOAGULANT THERAPY BUT MAY RESULT IN FALSE NEGATIVES IN THE PRESENCE OF WEAK LAC.
CONCLUSION POSSIBLE CONCLUSIONS:
Note 1: Results compatible with the absence of lupus anticoagulant. Note 2: Results compatible with the presence of lupus anticoagulant.
Note 3: Undetermined result, repeat after one week of completion or interruption of anticoagulant treatment.
If the result is positive for the presence of lupus anticoagulant, a new test should be performed after the 12-week interval.