http://informahealthcare.com/aut ISSN: 0891-6934 (print), 1607-842X (electronic) Autoimmunity, Early Online: 1–7 ! 2014 Informa UK Ltd. DOI: 10.3109/08916934.2014.988329
ORIGINAL ARTICLE
Antiphospholipid antibodies correlate with stroke severity and outcome in patients with antiphospholipid syndrome Ana Rodrı´guez-Sanz1, Patricia Martı´nez-Sa´nchez1, Daniel Prefasi1, Blanca Fuentes1, Dora Pascual-Salcedo2, Marı´a Jesu´s Blanco-Ban˜ares3, and Exuperio Dı´ez-Tejedor1
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
1
Department of Neurology and Stroke Center, 2Department of Immunology, and 3Department of Haematology, IdiPAZ Health Research Institute, La Paz University Hospital, Auto´noma University of Madrid, Madrid, Spain Abstract
Keywords
Background: Our goal was to analyze the association of the level of antiphospholipid antibodies (aPLs) with stroke severity and outcome in patients with antiphospholipid syndrome (APS). Methods: Observational study included consecutive patients with ischemic stroke younger than 55 years (2007–2012). We analyzed serum levels of aPLs, including anticardiolipin (aCL) antibodies, anti-b2-glycoprotein I antibodies (anti-b2GPI) and antiprothrombin antibodies (aPS/ PT) within the first 48 h after admission, and again, in the case of a positive result, at least 12 weeks after the first measurement. Stroke severity was measured by the National Institutes of Health Stroke Scale (NIHSS), and the three-month stroke outcome by the modified Rankin Scale (mRS). Multiple linear regression models were used to analyze the correlation between the aPLs and stroke severity and outcome. Results: Overall 255 stroke patients were included, 22 (8.6%) with APS. Among them, a positive correlation was found between immunoglobulin M (IgM) aCL levels within 48 h and NIHSS (rho ¼ 0.471; p ¼ 0.027), as well as a tendency toward a positive correlation between immunoglobulin G (IgG) anti-b2GPI levels within 48 h and three-month mRS (rho ¼ 0.364; p ¼ 0.096). Multiple linear regression analyses showed a positive correlation between levels of IgM aCL548 h and the NIHSS ( -coefficient [standard error; SE] ¼ 0.127 [0.044]), as well as the levels of IgG anti-b2GPIwithin 48 h and the three-month mRS ( -coefficient [SE] ¼ 0.034 [0.011]). Conclusions: In young stroke patients with APS, serum levels of IgM aCL within 48 h are correlated with stroke severity and levels of IgG anti-b2GPI within 48 h are correlated with three-month outcomes.
Anticardiolipin antibodies, antiphospholipid antibodies, antiphospholipid syndrome, anti-b2-Glycoprotein I, brain infarction, stroke
Introduction Antiphospholipid syndrome (APS) is an autoimmune disease clinically characterized by recurrent venous and/or arterial thromboembolic events or pregnancy morbidity. The most common thrombotic manifestation (two-third of cases) is venous thromboembolism, usually presenting as a typical deep vein thrombosis of the lower extremities. Arterial thrombosis (one-third of cases) primarily affects cerebral arteries, although it may also appear in other beds such as those of the coronary or peripheral circulation [1,2]. In addition to clinical manifestations, there are laboratory criteria that consist of repeated high titers of antiphospholipid antibodies (aPLs) (lupus anticoagulant (LA) or anticardiolipin (aCL) antibodies or anti-b2-glycoprotein I (anti-b2GPI) at least 12 weeks apart). APS is present if at least one of the clinical criteria and one of the laboratory criteria are met [3]. Correspondence: Dr. Patricia Martı´nez-Sa´nchez, Department of Neurology and Stroke Center, Hospital Universitario La Paz, IdiPAZ Health Research Institute, Paseo de la Castellana 261, 28046 Madrid, Spain. Tel: +34 91 7277444. Fax: +34 913581403. E-mail: [email protected]
History Received 10 March 2014 Revised 22 October 2014 Accepted 9 November 2014 Published online 1 December 2014
The aPLs are acquired antibodies against anionic phospholipid containing moieties in cell membranes, and the prothrombotic tendency of these autoantibodies has been extensively investigated in patients with autoimmune diseases [4,5]. However, the role of aPLs in predicting new thrombotic events is debated [6]. Although some studies have found that aPL positivity does not increase the risk of recurrent thromboocclusive events [7,8], others [9–12] have demonstrated a relationship between the level of aPLs and a higher risk of thromboembolism. Although APS is a well-known cause of ischemic stroke (IS) [13,14], the relationship between aPLs and IS severity is controversial. Some studies have shown that, in IS patients, high titers of aCL of the serotype immunoglobulin G (IgG) had a poorer prognosis in terms of thrombosis occurrence and greater mortality [15], whereas other publications have shown that the aCL isotype IgG did not confer a worse prognosis for disability (measured by the three-months modified Rankin Scale [mRS] score) and recurrent ischemic events [16,17]. We hypothesized that the aPLs level could be associated with higher stroke severity and with a poorer outcome in patients with APS and acute IS. This study aims to investigate
2
A. Rodrı´guez-Sanz et al.
the correlation between aPLs level and stroke severity by the National Institutes of Health Stroke Scale (NIHSS), as well as the stroke outcome by the three-months mRS in APS patients with acute IS.
Methods
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
Study design We conducted an observational cross-sectional study that included consecutive patients younger than 55 years with an acute brain infarction, admitted to the Stroke Unit of a university hospital between 2007 and 2012. This hospital is the only stroke center for a population of about 750 000. The exclusion criteria were patients who were 55 years of age and older and having a transient ischemic attack (TIA) or a brain hemorrhage. The clinical data were prospectively collected from medical records and included in the Stroke Data Bank. This information was collected from all patients upon hospital arrival. A retrospective analysis was performed for this study. Patients with incomplete data were excluded from this analysis. In-hospital management and vascular risk factors All patients were first attended in an emergency room by a neurologist on duty who recorded data for the medical history and performed the physical examination. During the first 24 h of hospitalization, each patient was assessed according to a standard neurovascular protocol, which included an urgent brain computed tomography (CT) scan, a chest X-ray, an electrocardiogram, routine laboratory blood analyses and a carotid plus transcranial ultrasound examination. Further cranial- and angio-magnetic resonance imaging (MRI) were performed between 24 and 72 h after admission to examine the magnitude of the brain lesion and to detect arterial pathology. Additional studies, such as catheter angiography and lumbar puncture, were performed if the stroke etiology remained unknown despite previous studies. A brain infarction was diagnosed when the CT or MRI showed an ischemic brain lesion corresponding to the patient’s symptoms. A TIA was defined as a transient episode of neurological dysfunction caused by a focal brain ischemia without acute infarction. Demographic data, vascular risk factors, comorbidities, stroke severity and outcome The following parameters were analyzed: (a) demographic data (age and gender); (b) vascular risk factors: hypertension, diabetes mellitus, dyslipidemia, cardiac valvulopathy, atrial fibrillation, severe ventricle malfunction (left ventricular ejection fraction530%), left ventricle akinesis, previous APS, other thrombophilias, prior pregnancy morbidity, prior brain venous thrombosis, renal disease, headache, migraine, coronary arterial disease, smoking, other drug abuse, peripheral arterial disease, prior TIA and prior IS; information regarding prior vascular risk factors and comorbidities was collected from medical charts and confirmed by patient interviews by the physician in charge; (c) previous treatment with antiplatelets or anticoagulants; (d) stroke etiology: atherothrombotic, cardioembolic, lacunar, undetermined and
Autoimmunity, Early Online: 1–7
unusual [18,19]; (e) stroke severity according to the NIHSS [20] (a 15-item scale that rates the level of neurologic impairment; total NIHSS scores range from 0 to 42, with higher scores indicating more severe cerebral infarctions (scores of 8 indicate mild neurologic impairment, scores of 8–25 indicate moderate neurologic impairment and scores of 25 indicate severe impairment), evaluated by a certified rater; (f) three-month outcomes as measured by the mRS [21] (which ranges from 0 to 6, with 0 indicating no symptoms and 6 indicating death); and (g) systemic and neurological in-hospital complications. Immunologic study Serum aPLs were assessed as fasting values during the 48 h following admission. The aPLs included the IgG and immunoglobulin M (IgM) isotypes of aCL, the IgG and IgM isotypes of anti-b2GPI and the IgG and IgM isotypes of antiprothrombin (aPT). Both aCL and anti-b2GPI were assayed using commercial kits (INOVA Diagnostics, San Diego, CA for aCL; and Thermo Fisher Scientific, Waltham, MA, for anti-b2GPI). The aPT were detected by a homemade ELISA where the prothrombin is exposed to phosphatidylserine (PS)-coated plates (aPS/PT) [22]. Briefly, 96-well polystyrene EIA plates (Costar 3591) were coated with soy PS (Sigma, St Louis, MO), previously reconstituted at 10 mg/ml with chloroform:methanol (4:1) and diluted to 50 mg/ml in ethanol. Plates were dried in the dark and then blocked with 20 mM Tris, 1% bovine serum albumin and 5 mM CaCl2 (blocking solution). After washing with 20 mM Tris, 5 mM CaCl2 and 0.005% Tween-20 (washing solution), plates were incubated with human prothrombin (Diagnostica Stago, Marseille, France) at 10 mg/ml in phosphate-buffered solution for 1 h at 37 C. After washing, samples, calibrators and controls diluted in blocking solution were incubated for 1 h at room temperature. Then, a 1:4000 dilution of alkaline phosphatase-conjugated goat anti-human IgG and anti-human IgM antisera (Jackson, ImmunoResearch, Philadelphia, PA) in blocking solution was added. In the last step, p-nitrophenyl phosphate diluted in 1 M diethanolamine pH 9.8 was employed as substrate. Optical density (OD) at 405 nm was read after 15 min. Standard curves were constructed with two APS patient sera containing elevated levels of IgG or IgM aPS/PT, titrated in arbitrary units/ml (AU/ml). Positive and negative controls were run in every plate. In all assays, the samples were run in parallel on the PS/PT-coated plate and in a ‘‘blank’’ plastic plate, coated only with PS. For each sample, OD in the PS-coated wells (blank) was subtracted from the OD in the PS/ PT-coated wells. Cut-off was established in 10 AU/ml for IgG isotype and 15 AU/ml for IgM isotype, corresponding to mean OD + 10 SD of 150 blood donors. The LA was evaluated by screening and confirmatory procedures following the recommendations of the Scientific and Standardization Committee Subcommittee of the International Society on Thrombosis and Hemostasis [24]. Screening tests included the dilute Russell viper venom time (dRVVTS; Siemens LA1 screen) and kaolin clotting time, a variant which included the mixture assays. As a confirmatory test, the dilute Russell viper venom time (dRVVTC) and phospholipid-dependent test (Siemens LA2 confirmation) was
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
DOI: 10.3109/08916934.2014.988329
Antiphospholipid antibodies and stroke severity
3
used. The LA was considered positive if the ratio of dRVVTS/ dRVVTC41.3. The LA was obtained on at least two occasions 12 weeks apart. Antiphospholipid antibody syndrome (APS) was present if at least one clinical and one laboratory criteria was met. The laboratory criteria were as follows [23]: (1) LA present in plasma, on two or more occasions at least 12 weeks apart; (2) aCL antibody of IgG and/or IgM isotype in serum or plasma, present in medium or high titer or greater than the 99th percentile, on two or more occasions, at least 12 weeks apart, measured by a standardized ELISA; (3) anti-b2GPI of IgG and/or IgM isotype in serum or plasma (in titer greater than the 99th percentile), present on two or more occasions, at least 12 weeks apart, measured by a standardized ELISA, according to recommended procedures. Our positive cut-off values were 20 GPL/MPL (aCL) and 10 U/ml (ab2GPI) based upon the 99th percentile. This study was approved by the hospital’s Ethics Committee and was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. Data analysis First, patients with acute IS were classified into two groups according to the presence or absence of APS. The proportions between groups were compared using the 2 test or Fisher’s exact test. Continuous variables were expressed as means ± SD or medians (IQR) and compared using Student’s t-test (for dependent and independent samples) or the Mann–Whitney test when appropriate. Second, we analyzed only the APS patients. The serum levels of the aPLs were compared between the first measurement (within the first 48 h of admission) and the second measurement (at least 12 weeks after the first measurement). Furthermore, Spearman’s rho correlation was used to determine associations between the aPLs and stroke severity, as measured by the NIHSS on admission, as well as between the aPLs and stroke outcomes, as measured by the three-month mRS score. Finally, we used multiple linear regression analyses to test the independent correlation between the serum levels of aPLs, with the NIHSS as the dependent variable, and controlled for potential confounding factors (e.g. age and cardioembolic etiology); we also tested the independent association between the serum levels of aPLs, with a threemonth mRS as the dependent variable, and controlled for potential confounding factors (age, NIHSS). The results of multiple linear regression models are presented as an unstandardized -coefficient (standard error [SE]). p Values less than 0.05 were considered significant. The statistical analysis was performed using SPSS (Statistical Package for Social Science, SPSS Inc, version 21.0 for Windows, Chicago, IL).
Results A total of 255 stroke patients younger than 54 years were included in the study (Figure 1), of whom 161 (63.1%) were men. The mean age was 44.59 ± 8.6 years. Twenty-two patients (8.6%) had APS, 4 (18.2%) were diagnosed before and 18 (81.8%) after the IS. No patients in the non-APS group had elevated aPLs except one who had increased levels of IgG
Figure 1. Flow diagram demonstrating the sample selection. APS, antiphospholipid syndrome.
anti-b2GPI in the first measurement, although this was not confirmed in further analyses. Table 1 lists the demographic data, vascular risk factors, prior treatment, stroke data and outcomes in patients with and without APS. The demographic data were similar in both groups; however, the patients in the APS group more frequently had prior pregnancy morbidity, left ventricle akinesis and had been more frequently treated with antiplatelets before the stroke. The stroke severity, three-month outcomes and in-hospital complications did not differ between both groups (Table 1). The first and second measurements of serum levels of aPLs in patients with APS are shown in Table 2 and in Figure 2. The aCL antibodies decreased over time, although this was statistically significant only in the case of IgG aCL. In patients with APS, a positive correlation was found between the first measurement of IgM aCL serum levels and the NIHSS on admission (rho ¼ 0.471; p ¼ 0.027; Table 3). Furthermore, a tendency toward a positive correlation between the levels of IgG anti-b2GPI and the three-month mRS in the first (rho ¼ 0.364; p ¼ 0.096) and second (rho ¼ 0.401; p ¼ 0.079) measurement was detected (Table 3). The multiple linear regression analysis showed a positive correlation between the IgM aCL serum levels within 48 h of hospital admission and the NIHSS ( -coefficient [SE] ¼ 0.127 [0.044]), adjusted by age and cardioembolic etiology (Table 4). On the other hand, the IgG anti-b2GPI serum levels within 48 h of hospital admission, but not after 12 weeks, were positively correlated with the 3-month mRS ( -coefficient [SE] ¼ 0.034 [0.011]), adjusted for age and NIHSS on admission (Table 5).
Discussion To our knowledge, this is the first study that shows a positive correlation between serum levels of aPLs antibodies and
4
A. Rodrı´guez-Sanz et al.
Autoimmunity, Early Online: 1–7
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
Table 1. Demographic data, vascular risk factors, prior treatment, stroke data and outcomes in patients with and without APS. No APS (n ¼ 233)
APS (n ¼ 22)
p
42.8 (9) 149 (63.9) 26 (11.2)
44.8 (8.5) 12 (54.5) 1 (4.5)
0.312 0.382 0.485
Demographic data Age, mean (SD) Male, n (%) Prior mRS40, n (%) Vascular risk factors Arterial hypertension, n (%) Diabetes Mellitus, n (%) Dyslipidemia, n (%) Valvulopathy, n (%) Atrial fibrillation, n (%) Severe ventricle malfunction (LVEF530%), n (%) Left ventricle akinesis, n (%) Previous APS, n (%) Other thrombophilias, n (%) Prior pregnancy morbidity, n (%) Prior brain venous thrombosis, n (%) Renal disease, n (%) Headache, n (%) Migraine, n (%) Coronary arterial disease, n (%) Smoker, n (%) Other drug abuse, n (%) Peripheral arterial disease, n (%) Prior transient ischemic attack, n (%) Prior ischemic stroke, n (%) Prior treatments Antiplatelets, n (%) Anticoagulants, n (%) Stroke etiology Atherothrombotic, n (%) Cardioembolic, n (%) Lacunar, n (%) Undetermined, n (%) Unusual, n (%) Stroke severity NIHSS, median (IQR) NIHSS score 8 Three-month outcome mRS, median (IQR) mRS42, n (%) In-hospital complications Systemic complications, n (%) Neurological complications, n (%)
80 23 68 15 19 8 10 0 54 10 4 4 29 35 16 83 10 3 9 26
(34.3) (9.9) (29.2) (6.4) (8.2) (3.4) (4.3) (0) (23.8) (7.2) (2) (1.7) (12.4) (15.1) (6.9) (42.1) (4.3) (1.3) (3.9) (11.3)
6 1 6 0 1 1 4 4 2 4 1 0 4 9 2 7 1 0 0 2
31 (13.3) 14 (6) 20 46 56 63 48
(27.3) (4.5) (27.3) (0) (4.5) (4.5) (18.2) (18.2) (9.1) (23.5) (5) (0) (18.2) (40.9) (9.1) (31.8) (4.5) (0) (0) (9.1)
0.503 0.704 0.850 0.376 1 0.562 0.024 50.0001 0.179 0.05 0.379 1 0.502 0.005 0.660 0.351 1 1 1 1
7 (31.8) 2 (9.1)
(8.6) (19.7) (24) (27) (20.6)
0 7 0 0 15
0.020 0.636
(0) (31.8) (0) (0) (68.2)
0.233 0.065 0.005 0.003 50.0001
3 (5) 47 (20.1)
2 (6) 3 (13.6)
0.095 0.583
1 (1) 27 (11.6)
1 (1) 3 (13.6)
0.514 0.731
14 (6) 19 (8.2)
0 (0) 0 (0)
0.618 0.386
APS, antiphospholipid syndrome and LVEF: left ventricular ejection fraction. Significant values are in bold.
Table 2. Antiphospholipid antibodies in APS patients. Antiphospholipid syndrome (n ¼ 22) Variable Anticardiolipin antibodies Anticardiolipin IgG isotype, mean (SD), GPL Anticardiolipin IgM isotype, mean (SD), MPL Anti- 2-glycoprotein antibodies Anti-b2-glycoprotein IgG isotype, mean (SD), U/ml Anti-b2-glycoprotein IgM isotype, mean (SD), U/ml Antiprothrombin antibodies Antiprothrombin IgG isotype, mean (SD), AU/ml Antiprothrombin IgM isotype, mean (SD), AU/ml Lupus anticoagulant, n (%)
First measurement* 21.6 (31.9) 26.1 (28.3)
Prevalencey 50% 50%
Second measurementz
Prevalencey
px
14.8 (20.5) 23.8 (41.9)
36.4% 36.4%
0.032 0.754
13.6% 13.6%
0.410 0.329
9% 9%
0.347 0.178 1
4.9 (13.2) 14.6 (48.5)
22.7% 18.2%
4.2 (10.4) 14.7 (48.5)
0.1 (0.3) 0.2 (0.4) 4 (18.2)
9% 4.5%
4 (12) 6 (12.1) 4 (18.2)
*Within 48 h after admission; Significant values are in bold. yPercentage of antiphospholipid antibodies over the cut-off limit (except for lupus anticoagulant). zAt least 12 weeks after the first measurement. xSignificance for differences between means (SD). APS, antiphospholipid syndrome.
Antiphospholipid antibodies and stroke severity
DOI: 10.3109/08916934.2014.988329
5
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
Figure 2. Antiphospholipid antibody titers in APS patients. Data are expressed as mean ± standard error. The dark gray bars represent the first measurement (within 48 h after admission) and those in light gray the second measurement (at least 12 weeks after the first measurement). *p ¼ 0.032 for the differences between the first and second ACL IgG isotype measurements. p ¼ NS for other comparisons. APS, antiphospholipid syndrome; aPLs, antiphospholipid antibodies; ACL IgG, anticardiolipin IgG isotype; ACL IgM, anticardiolipin IgM isotype; anti-B2-GP IgG, anti-b2-glycoprotein IgG isotype; anti-B2-GP IgM, anti-b2-glycoprotein IgM isotype; anti-PT IgG, antiprothrombin IgG isotype; and anti-PT IgM, antiprothrombin IgM isotype.
Table 3. Correlation analysis between the first and second measurement of antibody levels with stroke severity and three-month mRS in stroke patients with APS. NIHSS on admission Antibodies levels First measurement* Anticardiolipin IgG isotype, GPL Anticardiolipin IgM isotype, MPL Anti-b2-glycoprotein IgG isotype, U/ml Anti-b2-glycoprotein IgM isotype, U/ml Antiprothrombin IgG isotype, AU/ml Antiprothrombin IgM isotype, AU/ml Second measurementy Anticardiolipin IgG isotype, GPL Anticardiolipin IgM isotype, MPL Anti-b2-glycoprotein IgG isotype, U/ml Anti-b2-glycoprotein IgM isotype, U/ml Antiprothrombin Ig G isotype, AU/ml Antiprothrombin IgM isotype, AU/ml
Three-month mRS
r Spearman
p
r Spearman
p
0.144 0.471 0.220 0.117 0.091 0.078
0.523 0.027 0.325 0.606 0.715 0.767
0.067 0.166 0.364 0.092 0.033 0.060
0.766 0.459 0.096 0.684 0.900 0.818
0.177 0.136 0.162 0.097 0.102 0.262
0.455 0.568 0.494 0.676 0.765 0.437
0.024 0.101 0.401 0.024 0.368 0.050
0.921 0.673 0.079 0.917 0.265 0.885
*Within 48 h after admission; Statistical trends are bold. yAt least 12 weeks after the first measurement. APS, antiphospholipid syndrome; NIHSS, National Institutes of Health Stroke Scale; and mRS, modified Rankin Scale.
Table 4. Multiple linear regression analysis of the NIHSS on admission in stroke patients with APS (n ¼ 22). Unstandardized coefficients* Variable Anticardiolipin IgM isotype levels, U/mly Cardioembolic etiology
B
Standard error
p
0.127 5.350
0.044 2.484
0.010 0.044
*Stepwise multiple linear regression model adjusted by male sex, age, cardioembolic etiology and anticardiolipin IgM isotype. Variables not in the final equation: male sex and age. R2 ¼ 0.612. yFirst measurement (within 48 h after admission). Statistical trends are bold.
stroke severity and outcome in patients with APS. The IgM aCL serum levels within 48 h of hospital admission were positively correlated with the NIHSS (stroke severity), whereas the levels of IgG anti-b2GPI were correlated with the three-month mRS (stroke outcome). Previous studies [25,26] have demonstrated that increased levels of various aPLs was associated with the severity of
APS. Detkov et al. [26] showed that the simultaneous presence of LA, aCL and anti-b2GPI was related to a more severe clinical course of APS and, as a consequence, was associated with a greater risk of arterial and venous thrombotic complications, including IS. However, the knowledge of the relationship between aPLs levels and stroke severity and outcomes is scarce. Verro et al. [17] followed 27 patients with
6
A. Rodrı´guez-Sanz et al.
Autoimmunity, Early Online: 1–7
Table 5. Multiple linear regression analysis of three-month mRS in stroke patients with APS (n ¼ 22). Unstandardized coefficients* Variable Anti-b2-glycoprotein IgG isotype, U/mlz Anti-b2-glycoprotein IgG isotype, U/mlx NIHSS on admission
Unstandardized coefficientsy
B
Standard error
p
0.034
0.011
0.004
0.102
0.026
0.001
B
Standard error
p
– 0.090
– 0.027
– 0.003
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
*Stepwise multiple linear regression model adjusted by age, anti-b2-glycoprotein IgG isotype (first measurement) and NIHSS on admission. Variables not in the final equation: age. R2 ¼ 0.726. yStepwise multiple linear regression model adjusted by age, anti-b2-glycoprotein IgG isotype (second measurement) and NIHSS on admission. Variables not in the final equation: age and anti-b2-glycoprotein IgG isotype (second measurement). R2 ¼ 0.730. zFirst measurement (within 48 h after admission). xSecond measurement (at least 12 weeks after first measurement). Statistical trends are bold.
cerebrovascular ischemic symptoms (cerebral infarct, TIA or ocular ischemia) and who were found to have high positive aCL levels of the IgG isotype on one or more occasions. Cerebral infarcts occurred in 74% of patients and were recurrent in 37%. The IgG aCL titers did not correlated with neurological impairment, poorer prognosis or recurrent ischemic events. Accordingly, our study did not show a correlation between the IgG aCL titers and the stroke severity and outcome. The main limitation of the study by Verro et al. [17] is that they did not analyze other antibodies such as IgM aCL or (IgG/IgM) anti-b2GPI. The mechanisms of the prothrombotic action of aPLs are poorly understood. The most recent hypotheses include the interference of aPLs with the protein C axis and the complement system, and the activation of many cells including platelets, monocytes and endothelial cells [27]. Cardiolipin is known to block the anticoagulant effect observed in plasma and is associated with thrombosis in both the venous and arterial circulations. The venous thrombosis mechanism is different from that of arterial thrombo-occlusion. It appears that aCL is associated with disrupted endothelium, platelet aggregation, complement activation and disruption of coagulation. Some studies have shown that levels of IgG aCL at medium to high titer are possible risk factors for arterial thrombosis [9,17]. Another study [10] has demonstrated that persistent levels of IgM aCL are predictors of the risk of thromboembolic events involving the central nervous system in pediatric patients with systemic lupus erythematosus. This association could occur also in adults with APS. In fact, in our study, the aCL decreased over time suggesting that persistent aCL could be related to a high risk of thromboembolism, besides that it could be associated with stroke severity. This could be the reason why the second IgM aCL measurement did not correlate with either the severity or the prognosis of stroke. On the other hand, anti-b2GPI is associated with the risk of venous thrombosis [28–30], and several studies in young women showed an association with a two-fold increased risk of IS [31,32]. Overall, the aCL levels generated in APS require the presence of b2GPI to bind to cardiolipin [33]. It has been demonstrated that b2GPI-dependent aCL also react with b2GPI on irradiated platelets constituting the so-called anti-b2GPI [34,35]. Thus, it is commonly accepted that among these aPLs the most clinically relevant are anti-b2GPI [36]. There are no previous studies analyzing the role of antib2GPI in stroke severity and outcome, being this study the
first in showing a positive relationship between IgG antib2GPI titer and a worse stroke outcome. Levels of aPS/PT are also associated with an increased risk of prothrombotic events, although it appears to be less significant than other antibodies [9,37]. In fact, aPS/PT is not included in the diagnostic criteria for APS [23]. In this study, aPS/PT was not related to either stroke severity or outcome. Our study has some limitations. First, it is based on a retrospective analysis of our stroke database. However, patients were consecutively added and the database was prospectively updated. Another limitation is the small sample size, particularly of the APS group. However, we have found a statistically significant correlation between some antibodies and stroke severity and outcome, which could be even greater if the number of patients increases. Finally, aCL levels could be influenced by the ischemic event, increasing both IgM and IgG isotypes of aCL titers and not only IgM isotype. This study suggests that high titers of IgM aCL could be correlated with stroke severity, whereas high IgG anti-b2GPI titers could be correlated with poorer stroke outcomes. Future prospective studies are warranted to better understand these results and to investigate whether specific treatments that reduce the antibody levels could prevent severe strokes and post-stroke disability. In conclusion, in young stroke patients with APS, serum levels of IgM aCL within 48 h of admission are correlated with stroke severity, as measured by the NIHSS, and levels of IgG anti-b2GPI within 48 h of admission are correlated with three-month outcomes, according to the mRS.
Acknowledgements The authors thank Juliette Siegfried for editing the language of the manuscript.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article. This project is part of the Red de Enfermedades Vasculares Cerebrales (Cerebrovascular Diseases Network) INVICTUS (RD12/0014/0006).
References 1. Cervera, R., J. C. Piette, J. Font, et al.; Euro-Phospholipid Project Group. 2002. Antiphospholipid syndrome: clinical and
DOI: 10.3109/08916934.2014.988329
2.
3. 4. 5.
6.
7.
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
8.
9. 10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
20.
immunologic manifestations and patterns of disease expression in a cohort of 1000 patients. Arthritis. Rheum. 46: 1019–1027. Muscal, E., and R. L. Brey. 2008. Neurologic manifestations of antiphospholipid syndrome: integrating molecular and clinical lessons. Curr. Rheumatol. Rep. 10: 67–73. Ruiz-Irastorza, G., M. Crowther, W. Branch, and M. A. Khamashta. 2010. Antiphospholipid syndrome. Lancet 376: 1498–1509. Musial, J. 2012. Antiphospholipid antibodies and thrombosis. Thromb. Res. 129: 345–347. Dafer, R. M., J. Biller. 2008. Antiphospholipid syndrome: role of antiphospholipid antibodies in neurology. Hematol. Oncol. Clin. North. Am. 22: 95–105. Turiel, M., P. Sarzi-Puttini, R. Peretti, et al. 2005. Thrombotic risk factors in primary antiphospholipid syndrome: a 5-year prospective study. Stroke. 36: 1490–1494. Krnic-Barrie, S., C. R. O’Connor, S. W. Looney, et al. 1997. A retrospective review of 61 patients with antiphospholipid syndrome. Analysis of factors influencing recurrent thrombosis. Arch. Intern. Med. 157: 2101–2108. The APASS investigators. 2004. Antiphhospholipid antibodies and subsequent thrombo-occlusive events in patients with ischemic stroke. J. Am. Med. Assoc. 291: 576–584. Ortel, T. L. 2012. Antiphospholipid syndrome: laboratory testing and diagnostic strategies. Am. J. Hematol. 87: S75–S781. Male, C., D. Foulon, H. Hoogendoorn, et al. 2005. Predictive value of persistent versus transient antiphospholipid antibody subtypes for the risk of thrombotic events in pediatric patients with systemic lupus erythematosus. Blood. 106: 4152–4158. Ludia, C., P. Domenico, C. Monia, et al. 1998. Antiphospholipid antibodies: a new risk factor for restenosis after percutaneous transluminal coronary angioplasty? Autoimmunity. 27: 141–148. Szodoray, P., T. Tarr, J. Tumpek, et al. 2009. Identification of rare anti-phospholipid/protein co-factor autoantibodies in patients with systemic lupus erythematosus. Autoimmunity. 42: 497–506. Ruiz-Iraztorza, G., M. Crowther, W. Branch, and M. A. Khamashta. 2010. Antiphospholipid syndrome. Lancet 376: 1498–1509. Ruiz-Irastorza, G., and M. A. Khamashta. 2005. Stroke and antiphospholipid syndrome: the treatment debate. Rheumatology. 44: 971–974. Levine, S. R., L. Salowich-Palm, K. L. Sawaya, et al. 1997. IgG anticardiolipin antibody titers440 GPL and the risk of subsequent thrombo-occlusive events and death. Stroke. 28: 1660–1665. The Antiphospholipid Antibodies and Stroke Study Group (APASS). 1997. Anticardiolipin antibodies and the risk of recurrent thrombo-occlusive events and death. Neurology. 48: 91–94. Verro, P., S. R. Levine, and G. E. Tietjen. 1998. Cerebrovascular ischemic events with highpositive anticardiolipin antibodies. Stroke. 29: 2245–2253. Bogousslavsky, J., G. van Melle, F. Regli. 1988. The Lausanne Stroke Registry: analysis of 1,000 consecutive patients with first stroke. Stroke. 19: 1083–1092. Arboix, A., J. Alvarez-Sabin, and L. Soler. 1998. Ad hoc Editorial Committee of the Task Force on Cerebrovascular Diseases of SEN. Stroke: classification and diagnostic criteria. Neurologia. 13: 3–10. Brott, T., H. P. Adams Jr , C. P. Olinger, et al. 1989. Measurements of acute cerebral infarction: aclinical examination scale. Stroke. 20: 864–870.
Antiphospholipid antibodies and stroke severity
7
21. Van Swieten, J. C., P. J. Koudstaal., M. C. Visser, et al. 1988. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 19: 604–607. 22. Vlagea, A., A. Gil, M. V. Cuesta, et al. 2013. Antiphosphatidylserine/prothrombin antibodies (aPS/PT) as potential markers of antiphospholipid sindrome. Clin. Appl. Thromb. Hemost. 19: 289–296. 23. Miyakis, S., M. D. Lockshin, T. Atsumi, et al. 2006. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J. Thromb. Haemost. 4: 295–306. 24. Pengo, V., A. Tripodi, G. Reber, et al. 2009. 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. 7: 1737–1740. 25. Reber, G., and P. Moerloose. 2004. Anti-b2-glycoprotein I antibodies – when and how should they be measured? Thromb. Res. 114: 527–531. 26. Detkov, D., A. Gil-Aguado, P. Lavilla, et al. 1999. Do antibodies to b2-glycoprotein 1 contribute to the better characterization of the antiphospholipid syndrome? Lupus. 8: 430–438. 27. De Groot, P. G. 2011. Mechanisms of anti-phospholipid antibody formation and action. Thromb. Res. 127: S40–S42. 28. Galli, M., G. Borrelli, E. M. Jacobsen, et al. 2007. Clinical significance of different antiphospholipid antibodies in the WAPS (warfarin in the antiphospholipid syndrome) study. Blood. 110: 1178–1183. 29. Atsumi, T., O. Amengual, S. Yasuda, et al. 2005. Research around beta 2-glycoprotein I: a major target for antiphospholipid antibodies. Autoimmunity. 38: 377–381. 30. Andreoli, L., M. Fredi, C. Nalli, et al. 2013. Antiphospholipid antibodies mediate autoimmunity against dying cells. Autoimmunity. 46: 302–306. 31. Urbanus, R.T., B. Siegerink, M. Roest, et al. 2009. Antiphospholipid antibodies and risk of myocardial infarction and ischaemic stroke in young women in the RATIO study: a casecontrol study. Lancet. Neurol. 8: 998–1005. 32. Brey, R. L., C. L. Stallworth, D. L. McGlasson, et al. 2002. Antiphospholipid antibodies and stroke in young women. Stroke. 33: 2396–2400. 33. Matsuura, E., Y. Igarashi, M. Fujimoto, et al. 1990. Anticardiolipin cofactor(s) and differential diagnosis of autoimmune disease. Lancet. 336: 177–178. 34. Galli, M., P. Comfurius, C. Maassen, et al. 1990. Anticardiolipin antibodies (ACA) directed not to cardiolipin but to a plasma protein cofactor. Lancet. 335: 1544–1547. 35. Roubey, R. A., R. A. Eisenberg, M. F. Harper, and J. B. Winfield. 1995. Winfield. ‘‘Anticardiolipin’’ autoantibodies recognize beta 2glycoprotein I in the absence of phospholipid. Importance of Ag density and bivalent binding. J. Immunol. 154: 954–960. 36. Ali, H. Y., and Z. A. Abdullah. 2008. Anti-beta(2)-glycoprotein I autoantibody expression as a potential biomarker for strokes in patients with anti-phospholipid syndrome. J. Immunotoxicol. 5: 173–177. 37. Bevers, E. M., M. Galli, T. Barbui, et al. 1991. Lupus anticoagulant IgG’s (LA) are not directed to phospholipids only, but to a complex of lipidbound human prothrombin. Thromb. Haemost. 66: 629–632.
ORIGINAL ARTICLE
Antiphospholipid antibodies correlate with stroke severity and outcome in patients with antiphospholipid syndrome Ana Rodrı´guez-Sanz1, Patricia Martı´nez-Sa´nchez1, Daniel Prefasi1, Blanca Fuentes1, Dora Pascual-Salcedo2, Marı´a Jesu´s Blanco-Ban˜ares3, and Exuperio Dı´ez-Tejedor1
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
1
Department of Neurology and Stroke Center, 2Department of Immunology, and 3Department of Haematology, IdiPAZ Health Research Institute, La Paz University Hospital, Auto´noma University of Madrid, Madrid, Spain Abstract
Keywords
Background: Our goal was to analyze the association of the level of antiphospholipid antibodies (aPLs) with stroke severity and outcome in patients with antiphospholipid syndrome (APS). Methods: Observational study included consecutive patients with ischemic stroke younger than 55 years (2007–2012). We analyzed serum levels of aPLs, including anticardiolipin (aCL) antibodies, anti-b2-glycoprotein I antibodies (anti-b2GPI) and antiprothrombin antibodies (aPS/ PT) within the first 48 h after admission, and again, in the case of a positive result, at least 12 weeks after the first measurement. Stroke severity was measured by the National Institutes of Health Stroke Scale (NIHSS), and the three-month stroke outcome by the modified Rankin Scale (mRS). Multiple linear regression models were used to analyze the correlation between the aPLs and stroke severity and outcome. Results: Overall 255 stroke patients were included, 22 (8.6%) with APS. Among them, a positive correlation was found between immunoglobulin M (IgM) aCL levels within 48 h and NIHSS (rho ¼ 0.471; p ¼ 0.027), as well as a tendency toward a positive correlation between immunoglobulin G (IgG) anti-b2GPI levels within 48 h and three-month mRS (rho ¼ 0.364; p ¼ 0.096). Multiple linear regression analyses showed a positive correlation between levels of IgM aCL548 h and the NIHSS ( -coefficient [standard error; SE] ¼ 0.127 [0.044]), as well as the levels of IgG anti-b2GPIwithin 48 h and the three-month mRS ( -coefficient [SE] ¼ 0.034 [0.011]). Conclusions: In young stroke patients with APS, serum levels of IgM aCL within 48 h are correlated with stroke severity and levels of IgG anti-b2GPI within 48 h are correlated with three-month outcomes.
Anticardiolipin antibodies, antiphospholipid antibodies, antiphospholipid syndrome, anti-b2-Glycoprotein I, brain infarction, stroke
Introduction Antiphospholipid syndrome (APS) is an autoimmune disease clinically characterized by recurrent venous and/or arterial thromboembolic events or pregnancy morbidity. The most common thrombotic manifestation (two-third of cases) is venous thromboembolism, usually presenting as a typical deep vein thrombosis of the lower extremities. Arterial thrombosis (one-third of cases) primarily affects cerebral arteries, although it may also appear in other beds such as those of the coronary or peripheral circulation [1,2]. In addition to clinical manifestations, there are laboratory criteria that consist of repeated high titers of antiphospholipid antibodies (aPLs) (lupus anticoagulant (LA) or anticardiolipin (aCL) antibodies or anti-b2-glycoprotein I (anti-b2GPI) at least 12 weeks apart). APS is present if at least one of the clinical criteria and one of the laboratory criteria are met [3]. Correspondence: Dr. Patricia Martı´nez-Sa´nchez, Department of Neurology and Stroke Center, Hospital Universitario La Paz, IdiPAZ Health Research Institute, Paseo de la Castellana 261, 28046 Madrid, Spain. Tel: +34 91 7277444. Fax: +34 913581403. E-mail: [email protected]
History Received 10 March 2014 Revised 22 October 2014 Accepted 9 November 2014 Published online 1 December 2014
The aPLs are acquired antibodies against anionic phospholipid containing moieties in cell membranes, and the prothrombotic tendency of these autoantibodies has been extensively investigated in patients with autoimmune diseases [4,5]. However, the role of aPLs in predicting new thrombotic events is debated [6]. Although some studies have found that aPL positivity does not increase the risk of recurrent thromboocclusive events [7,8], others [9–12] have demonstrated a relationship between the level of aPLs and a higher risk of thromboembolism. Although APS is a well-known cause of ischemic stroke (IS) [13,14], the relationship between aPLs and IS severity is controversial. Some studies have shown that, in IS patients, high titers of aCL of the serotype immunoglobulin G (IgG) had a poorer prognosis in terms of thrombosis occurrence and greater mortality [15], whereas other publications have shown that the aCL isotype IgG did not confer a worse prognosis for disability (measured by the three-months modified Rankin Scale [mRS] score) and recurrent ischemic events [16,17]. We hypothesized that the aPLs level could be associated with higher stroke severity and with a poorer outcome in patients with APS and acute IS. This study aims to investigate
2
A. Rodrı´guez-Sanz et al.
the correlation between aPLs level and stroke severity by the National Institutes of Health Stroke Scale (NIHSS), as well as the stroke outcome by the three-months mRS in APS patients with acute IS.
Methods
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
Study design We conducted an observational cross-sectional study that included consecutive patients younger than 55 years with an acute brain infarction, admitted to the Stroke Unit of a university hospital between 2007 and 2012. This hospital is the only stroke center for a population of about 750 000. The exclusion criteria were patients who were 55 years of age and older and having a transient ischemic attack (TIA) or a brain hemorrhage. The clinical data were prospectively collected from medical records and included in the Stroke Data Bank. This information was collected from all patients upon hospital arrival. A retrospective analysis was performed for this study. Patients with incomplete data were excluded from this analysis. In-hospital management and vascular risk factors All patients were first attended in an emergency room by a neurologist on duty who recorded data for the medical history and performed the physical examination. During the first 24 h of hospitalization, each patient was assessed according to a standard neurovascular protocol, which included an urgent brain computed tomography (CT) scan, a chest X-ray, an electrocardiogram, routine laboratory blood analyses and a carotid plus transcranial ultrasound examination. Further cranial- and angio-magnetic resonance imaging (MRI) were performed between 24 and 72 h after admission to examine the magnitude of the brain lesion and to detect arterial pathology. Additional studies, such as catheter angiography and lumbar puncture, were performed if the stroke etiology remained unknown despite previous studies. A brain infarction was diagnosed when the CT or MRI showed an ischemic brain lesion corresponding to the patient’s symptoms. A TIA was defined as a transient episode of neurological dysfunction caused by a focal brain ischemia without acute infarction. Demographic data, vascular risk factors, comorbidities, stroke severity and outcome The following parameters were analyzed: (a) demographic data (age and gender); (b) vascular risk factors: hypertension, diabetes mellitus, dyslipidemia, cardiac valvulopathy, atrial fibrillation, severe ventricle malfunction (left ventricular ejection fraction530%), left ventricle akinesis, previous APS, other thrombophilias, prior pregnancy morbidity, prior brain venous thrombosis, renal disease, headache, migraine, coronary arterial disease, smoking, other drug abuse, peripheral arterial disease, prior TIA and prior IS; information regarding prior vascular risk factors and comorbidities was collected from medical charts and confirmed by patient interviews by the physician in charge; (c) previous treatment with antiplatelets or anticoagulants; (d) stroke etiology: atherothrombotic, cardioembolic, lacunar, undetermined and
Autoimmunity, Early Online: 1–7
unusual [18,19]; (e) stroke severity according to the NIHSS [20] (a 15-item scale that rates the level of neurologic impairment; total NIHSS scores range from 0 to 42, with higher scores indicating more severe cerebral infarctions (scores of 8 indicate mild neurologic impairment, scores of 8–25 indicate moderate neurologic impairment and scores of 25 indicate severe impairment), evaluated by a certified rater; (f) three-month outcomes as measured by the mRS [21] (which ranges from 0 to 6, with 0 indicating no symptoms and 6 indicating death); and (g) systemic and neurological in-hospital complications. Immunologic study Serum aPLs were assessed as fasting values during the 48 h following admission. The aPLs included the IgG and immunoglobulin M (IgM) isotypes of aCL, the IgG and IgM isotypes of anti-b2GPI and the IgG and IgM isotypes of antiprothrombin (aPT). Both aCL and anti-b2GPI were assayed using commercial kits (INOVA Diagnostics, San Diego, CA for aCL; and Thermo Fisher Scientific, Waltham, MA, for anti-b2GPI). The aPT were detected by a homemade ELISA where the prothrombin is exposed to phosphatidylserine (PS)-coated plates (aPS/PT) [22]. Briefly, 96-well polystyrene EIA plates (Costar 3591) were coated with soy PS (Sigma, St Louis, MO), previously reconstituted at 10 mg/ml with chloroform:methanol (4:1) and diluted to 50 mg/ml in ethanol. Plates were dried in the dark and then blocked with 20 mM Tris, 1% bovine serum albumin and 5 mM CaCl2 (blocking solution). After washing with 20 mM Tris, 5 mM CaCl2 and 0.005% Tween-20 (washing solution), plates were incubated with human prothrombin (Diagnostica Stago, Marseille, France) at 10 mg/ml in phosphate-buffered solution for 1 h at 37 C. After washing, samples, calibrators and controls diluted in blocking solution were incubated for 1 h at room temperature. Then, a 1:4000 dilution of alkaline phosphatase-conjugated goat anti-human IgG and anti-human IgM antisera (Jackson, ImmunoResearch, Philadelphia, PA) in blocking solution was added. In the last step, p-nitrophenyl phosphate diluted in 1 M diethanolamine pH 9.8 was employed as substrate. Optical density (OD) at 405 nm was read after 15 min. Standard curves were constructed with two APS patient sera containing elevated levels of IgG or IgM aPS/PT, titrated in arbitrary units/ml (AU/ml). Positive and negative controls were run in every plate. In all assays, the samples were run in parallel on the PS/PT-coated plate and in a ‘‘blank’’ plastic plate, coated only with PS. For each sample, OD in the PS-coated wells (blank) was subtracted from the OD in the PS/ PT-coated wells. Cut-off was established in 10 AU/ml for IgG isotype and 15 AU/ml for IgM isotype, corresponding to mean OD + 10 SD of 150 blood donors. The LA was evaluated by screening and confirmatory procedures following the recommendations of the Scientific and Standardization Committee Subcommittee of the International Society on Thrombosis and Hemostasis [24]. Screening tests included the dilute Russell viper venom time (dRVVTS; Siemens LA1 screen) and kaolin clotting time, a variant which included the mixture assays. As a confirmatory test, the dilute Russell viper venom time (dRVVTC) and phospholipid-dependent test (Siemens LA2 confirmation) was
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
DOI: 10.3109/08916934.2014.988329
Antiphospholipid antibodies and stroke severity
3
used. The LA was considered positive if the ratio of dRVVTS/ dRVVTC41.3. The LA was obtained on at least two occasions 12 weeks apart. Antiphospholipid antibody syndrome (APS) was present if at least one clinical and one laboratory criteria was met. The laboratory criteria were as follows [23]: (1) LA present in plasma, on two or more occasions at least 12 weeks apart; (2) aCL antibody of IgG and/or IgM isotype in serum or plasma, present in medium or high titer or greater than the 99th percentile, on two or more occasions, at least 12 weeks apart, measured by a standardized ELISA; (3) anti-b2GPI of IgG and/or IgM isotype in serum or plasma (in titer greater than the 99th percentile), present on two or more occasions, at least 12 weeks apart, measured by a standardized ELISA, according to recommended procedures. Our positive cut-off values were 20 GPL/MPL (aCL) and 10 U/ml (ab2GPI) based upon the 99th percentile. This study was approved by the hospital’s Ethics Committee and was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. Data analysis First, patients with acute IS were classified into two groups according to the presence or absence of APS. The proportions between groups were compared using the 2 test or Fisher’s exact test. Continuous variables were expressed as means ± SD or medians (IQR) and compared using Student’s t-test (for dependent and independent samples) or the Mann–Whitney test when appropriate. Second, we analyzed only the APS patients. The serum levels of the aPLs were compared between the first measurement (within the first 48 h of admission) and the second measurement (at least 12 weeks after the first measurement). Furthermore, Spearman’s rho correlation was used to determine associations between the aPLs and stroke severity, as measured by the NIHSS on admission, as well as between the aPLs and stroke outcomes, as measured by the three-month mRS score. Finally, we used multiple linear regression analyses to test the independent correlation between the serum levels of aPLs, with the NIHSS as the dependent variable, and controlled for potential confounding factors (e.g. age and cardioembolic etiology); we also tested the independent association between the serum levels of aPLs, with a threemonth mRS as the dependent variable, and controlled for potential confounding factors (age, NIHSS). The results of multiple linear regression models are presented as an unstandardized -coefficient (standard error [SE]). p Values less than 0.05 were considered significant. The statistical analysis was performed using SPSS (Statistical Package for Social Science, SPSS Inc, version 21.0 for Windows, Chicago, IL).
Results A total of 255 stroke patients younger than 54 years were included in the study (Figure 1), of whom 161 (63.1%) were men. The mean age was 44.59 ± 8.6 years. Twenty-two patients (8.6%) had APS, 4 (18.2%) were diagnosed before and 18 (81.8%) after the IS. No patients in the non-APS group had elevated aPLs except one who had increased levels of IgG
Figure 1. Flow diagram demonstrating the sample selection. APS, antiphospholipid syndrome.
anti-b2GPI in the first measurement, although this was not confirmed in further analyses. Table 1 lists the demographic data, vascular risk factors, prior treatment, stroke data and outcomes in patients with and without APS. The demographic data were similar in both groups; however, the patients in the APS group more frequently had prior pregnancy morbidity, left ventricle akinesis and had been more frequently treated with antiplatelets before the stroke. The stroke severity, three-month outcomes and in-hospital complications did not differ between both groups (Table 1). The first and second measurements of serum levels of aPLs in patients with APS are shown in Table 2 and in Figure 2. The aCL antibodies decreased over time, although this was statistically significant only in the case of IgG aCL. In patients with APS, a positive correlation was found between the first measurement of IgM aCL serum levels and the NIHSS on admission (rho ¼ 0.471; p ¼ 0.027; Table 3). Furthermore, a tendency toward a positive correlation between the levels of IgG anti-b2GPI and the three-month mRS in the first (rho ¼ 0.364; p ¼ 0.096) and second (rho ¼ 0.401; p ¼ 0.079) measurement was detected (Table 3). The multiple linear regression analysis showed a positive correlation between the IgM aCL serum levels within 48 h of hospital admission and the NIHSS ( -coefficient [SE] ¼ 0.127 [0.044]), adjusted by age and cardioembolic etiology (Table 4). On the other hand, the IgG anti-b2GPI serum levels within 48 h of hospital admission, but not after 12 weeks, were positively correlated with the 3-month mRS ( -coefficient [SE] ¼ 0.034 [0.011]), adjusted for age and NIHSS on admission (Table 5).
Discussion To our knowledge, this is the first study that shows a positive correlation between serum levels of aPLs antibodies and
4
A. Rodrı´guez-Sanz et al.
Autoimmunity, Early Online: 1–7
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
Table 1. Demographic data, vascular risk factors, prior treatment, stroke data and outcomes in patients with and without APS. No APS (n ¼ 233)
APS (n ¼ 22)
p
42.8 (9) 149 (63.9) 26 (11.2)
44.8 (8.5) 12 (54.5) 1 (4.5)
0.312 0.382 0.485
Demographic data Age, mean (SD) Male, n (%) Prior mRS40, n (%) Vascular risk factors Arterial hypertension, n (%) Diabetes Mellitus, n (%) Dyslipidemia, n (%) Valvulopathy, n (%) Atrial fibrillation, n (%) Severe ventricle malfunction (LVEF530%), n (%) Left ventricle akinesis, n (%) Previous APS, n (%) Other thrombophilias, n (%) Prior pregnancy morbidity, n (%) Prior brain venous thrombosis, n (%) Renal disease, n (%) Headache, n (%) Migraine, n (%) Coronary arterial disease, n (%) Smoker, n (%) Other drug abuse, n (%) Peripheral arterial disease, n (%) Prior transient ischemic attack, n (%) Prior ischemic stroke, n (%) Prior treatments Antiplatelets, n (%) Anticoagulants, n (%) Stroke etiology Atherothrombotic, n (%) Cardioembolic, n (%) Lacunar, n (%) Undetermined, n (%) Unusual, n (%) Stroke severity NIHSS, median (IQR) NIHSS score 8 Three-month outcome mRS, median (IQR) mRS42, n (%) In-hospital complications Systemic complications, n (%) Neurological complications, n (%)
80 23 68 15 19 8 10 0 54 10 4 4 29 35 16 83 10 3 9 26
(34.3) (9.9) (29.2) (6.4) (8.2) (3.4) (4.3) (0) (23.8) (7.2) (2) (1.7) (12.4) (15.1) (6.9) (42.1) (4.3) (1.3) (3.9) (11.3)
6 1 6 0 1 1 4 4 2 4 1 0 4 9 2 7 1 0 0 2
31 (13.3) 14 (6) 20 46 56 63 48
(27.3) (4.5) (27.3) (0) (4.5) (4.5) (18.2) (18.2) (9.1) (23.5) (5) (0) (18.2) (40.9) (9.1) (31.8) (4.5) (0) (0) (9.1)
0.503 0.704 0.850 0.376 1 0.562 0.024 50.0001 0.179 0.05 0.379 1 0.502 0.005 0.660 0.351 1 1 1 1
7 (31.8) 2 (9.1)
(8.6) (19.7) (24) (27) (20.6)
0 7 0 0 15
0.020 0.636
(0) (31.8) (0) (0) (68.2)
0.233 0.065 0.005 0.003 50.0001
3 (5) 47 (20.1)
2 (6) 3 (13.6)
0.095 0.583
1 (1) 27 (11.6)
1 (1) 3 (13.6)
0.514 0.731
14 (6) 19 (8.2)
0 (0) 0 (0)
0.618 0.386
APS, antiphospholipid syndrome and LVEF: left ventricular ejection fraction. Significant values are in bold.
Table 2. Antiphospholipid antibodies in APS patients. Antiphospholipid syndrome (n ¼ 22) Variable Anticardiolipin antibodies Anticardiolipin IgG isotype, mean (SD), GPL Anticardiolipin IgM isotype, mean (SD), MPL Anti- 2-glycoprotein antibodies Anti-b2-glycoprotein IgG isotype, mean (SD), U/ml Anti-b2-glycoprotein IgM isotype, mean (SD), U/ml Antiprothrombin antibodies Antiprothrombin IgG isotype, mean (SD), AU/ml Antiprothrombin IgM isotype, mean (SD), AU/ml Lupus anticoagulant, n (%)
First measurement* 21.6 (31.9) 26.1 (28.3)
Prevalencey 50% 50%
Second measurementz
Prevalencey
px
14.8 (20.5) 23.8 (41.9)
36.4% 36.4%
0.032 0.754
13.6% 13.6%
0.410 0.329
9% 9%
0.347 0.178 1
4.9 (13.2) 14.6 (48.5)
22.7% 18.2%
4.2 (10.4) 14.7 (48.5)
0.1 (0.3) 0.2 (0.4) 4 (18.2)
9% 4.5%
4 (12) 6 (12.1) 4 (18.2)
*Within 48 h after admission; Significant values are in bold. yPercentage of antiphospholipid antibodies over the cut-off limit (except for lupus anticoagulant). zAt least 12 weeks after the first measurement. xSignificance for differences between means (SD). APS, antiphospholipid syndrome.
Antiphospholipid antibodies and stroke severity
DOI: 10.3109/08916934.2014.988329
5
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
Figure 2. Antiphospholipid antibody titers in APS patients. Data are expressed as mean ± standard error. The dark gray bars represent the first measurement (within 48 h after admission) and those in light gray the second measurement (at least 12 weeks after the first measurement). *p ¼ 0.032 for the differences between the first and second ACL IgG isotype measurements. p ¼ NS for other comparisons. APS, antiphospholipid syndrome; aPLs, antiphospholipid antibodies; ACL IgG, anticardiolipin IgG isotype; ACL IgM, anticardiolipin IgM isotype; anti-B2-GP IgG, anti-b2-glycoprotein IgG isotype; anti-B2-GP IgM, anti-b2-glycoprotein IgM isotype; anti-PT IgG, antiprothrombin IgG isotype; and anti-PT IgM, antiprothrombin IgM isotype.
Table 3. Correlation analysis between the first and second measurement of antibody levels with stroke severity and three-month mRS in stroke patients with APS. NIHSS on admission Antibodies levels First measurement* Anticardiolipin IgG isotype, GPL Anticardiolipin IgM isotype, MPL Anti-b2-glycoprotein IgG isotype, U/ml Anti-b2-glycoprotein IgM isotype, U/ml Antiprothrombin IgG isotype, AU/ml Antiprothrombin IgM isotype, AU/ml Second measurementy Anticardiolipin IgG isotype, GPL Anticardiolipin IgM isotype, MPL Anti-b2-glycoprotein IgG isotype, U/ml Anti-b2-glycoprotein IgM isotype, U/ml Antiprothrombin Ig G isotype, AU/ml Antiprothrombin IgM isotype, AU/ml
Three-month mRS
r Spearman
p
r Spearman
p
0.144 0.471 0.220 0.117 0.091 0.078
0.523 0.027 0.325 0.606 0.715 0.767
0.067 0.166 0.364 0.092 0.033 0.060
0.766 0.459 0.096 0.684 0.900 0.818
0.177 0.136 0.162 0.097 0.102 0.262
0.455 0.568 0.494 0.676 0.765 0.437
0.024 0.101 0.401 0.024 0.368 0.050
0.921 0.673 0.079 0.917 0.265 0.885
*Within 48 h after admission; Statistical trends are bold. yAt least 12 weeks after the first measurement. APS, antiphospholipid syndrome; NIHSS, National Institutes of Health Stroke Scale; and mRS, modified Rankin Scale.
Table 4. Multiple linear regression analysis of the NIHSS on admission in stroke patients with APS (n ¼ 22). Unstandardized coefficients* Variable Anticardiolipin IgM isotype levels, U/mly Cardioembolic etiology
B
Standard error
p
0.127 5.350
0.044 2.484
0.010 0.044
*Stepwise multiple linear regression model adjusted by male sex, age, cardioembolic etiology and anticardiolipin IgM isotype. Variables not in the final equation: male sex and age. R2 ¼ 0.612. yFirst measurement (within 48 h after admission). Statistical trends are bold.
stroke severity and outcome in patients with APS. The IgM aCL serum levels within 48 h of hospital admission were positively correlated with the NIHSS (stroke severity), whereas the levels of IgG anti-b2GPI were correlated with the three-month mRS (stroke outcome). Previous studies [25,26] have demonstrated that increased levels of various aPLs was associated with the severity of
APS. Detkov et al. [26] showed that the simultaneous presence of LA, aCL and anti-b2GPI was related to a more severe clinical course of APS and, as a consequence, was associated with a greater risk of arterial and venous thrombotic complications, including IS. However, the knowledge of the relationship between aPLs levels and stroke severity and outcomes is scarce. Verro et al. [17] followed 27 patients with
6
A. Rodrı´guez-Sanz et al.
Autoimmunity, Early Online: 1–7
Table 5. Multiple linear regression analysis of three-month mRS in stroke patients with APS (n ¼ 22). Unstandardized coefficients* Variable Anti-b2-glycoprotein IgG isotype, U/mlz Anti-b2-glycoprotein IgG isotype, U/mlx NIHSS on admission
Unstandardized coefficientsy
B
Standard error
p
0.034
0.011
0.004
0.102
0.026
0.001
B
Standard error
p
– 0.090
– 0.027
– 0.003
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
*Stepwise multiple linear regression model adjusted by age, anti-b2-glycoprotein IgG isotype (first measurement) and NIHSS on admission. Variables not in the final equation: age. R2 ¼ 0.726. yStepwise multiple linear regression model adjusted by age, anti-b2-glycoprotein IgG isotype (second measurement) and NIHSS on admission. Variables not in the final equation: age and anti-b2-glycoprotein IgG isotype (second measurement). R2 ¼ 0.730. zFirst measurement (within 48 h after admission). xSecond measurement (at least 12 weeks after first measurement). Statistical trends are bold.
cerebrovascular ischemic symptoms (cerebral infarct, TIA or ocular ischemia) and who were found to have high positive aCL levels of the IgG isotype on one or more occasions. Cerebral infarcts occurred in 74% of patients and were recurrent in 37%. The IgG aCL titers did not correlated with neurological impairment, poorer prognosis or recurrent ischemic events. Accordingly, our study did not show a correlation between the IgG aCL titers and the stroke severity and outcome. The main limitation of the study by Verro et al. [17] is that they did not analyze other antibodies such as IgM aCL or (IgG/IgM) anti-b2GPI. The mechanisms of the prothrombotic action of aPLs are poorly understood. The most recent hypotheses include the interference of aPLs with the protein C axis and the complement system, and the activation of many cells including platelets, monocytes and endothelial cells [27]. Cardiolipin is known to block the anticoagulant effect observed in plasma and is associated with thrombosis in both the venous and arterial circulations. The venous thrombosis mechanism is different from that of arterial thrombo-occlusion. It appears that aCL is associated with disrupted endothelium, platelet aggregation, complement activation and disruption of coagulation. Some studies have shown that levels of IgG aCL at medium to high titer are possible risk factors for arterial thrombosis [9,17]. Another study [10] has demonstrated that persistent levels of IgM aCL are predictors of the risk of thromboembolic events involving the central nervous system in pediatric patients with systemic lupus erythematosus. This association could occur also in adults with APS. In fact, in our study, the aCL decreased over time suggesting that persistent aCL could be related to a high risk of thromboembolism, besides that it could be associated with stroke severity. This could be the reason why the second IgM aCL measurement did not correlate with either the severity or the prognosis of stroke. On the other hand, anti-b2GPI is associated with the risk of venous thrombosis [28–30], and several studies in young women showed an association with a two-fold increased risk of IS [31,32]. Overall, the aCL levels generated in APS require the presence of b2GPI to bind to cardiolipin [33]. It has been demonstrated that b2GPI-dependent aCL also react with b2GPI on irradiated platelets constituting the so-called anti-b2GPI [34,35]. Thus, it is commonly accepted that among these aPLs the most clinically relevant are anti-b2GPI [36]. There are no previous studies analyzing the role of antib2GPI in stroke severity and outcome, being this study the
first in showing a positive relationship between IgG antib2GPI titer and a worse stroke outcome. Levels of aPS/PT are also associated with an increased risk of prothrombotic events, although it appears to be less significant than other antibodies [9,37]. In fact, aPS/PT is not included in the diagnostic criteria for APS [23]. In this study, aPS/PT was not related to either stroke severity or outcome. Our study has some limitations. First, it is based on a retrospective analysis of our stroke database. However, patients were consecutively added and the database was prospectively updated. Another limitation is the small sample size, particularly of the APS group. However, we have found a statistically significant correlation between some antibodies and stroke severity and outcome, which could be even greater if the number of patients increases. Finally, aCL levels could be influenced by the ischemic event, increasing both IgM and IgG isotypes of aCL titers and not only IgM isotype. This study suggests that high titers of IgM aCL could be correlated with stroke severity, whereas high IgG anti-b2GPI titers could be correlated with poorer stroke outcomes. Future prospective studies are warranted to better understand these results and to investigate whether specific treatments that reduce the antibody levels could prevent severe strokes and post-stroke disability. In conclusion, in young stroke patients with APS, serum levels of IgM aCL within 48 h of admission are correlated with stroke severity, as measured by the NIHSS, and levels of IgG anti-b2GPI within 48 h of admission are correlated with three-month outcomes, according to the mRS.
Acknowledgements The authors thank Juliette Siegfried for editing the language of the manuscript.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article. This project is part of the Red de Enfermedades Vasculares Cerebrales (Cerebrovascular Diseases Network) INVICTUS (RD12/0014/0006).
References 1. Cervera, R., J. C. Piette, J. Font, et al.; Euro-Phospholipid Project Group. 2002. Antiphospholipid syndrome: clinical and
DOI: 10.3109/08916934.2014.988329
2.
3. 4. 5.
6.
7.
Autoimmunity Downloaded from informahealthcare.com by Yale Dermatologic Surgery on 12/25/14 For personal use only.
8.
9. 10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
20.
immunologic manifestations and patterns of disease expression in a cohort of 1000 patients. Arthritis. Rheum. 46: 1019–1027. Muscal, E., and R. L. Brey. 2008. Neurologic manifestations of antiphospholipid syndrome: integrating molecular and clinical lessons. Curr. Rheumatol. Rep. 10: 67–73. Ruiz-Irastorza, G., M. Crowther, W. Branch, and M. A. Khamashta. 2010. Antiphospholipid syndrome. Lancet 376: 1498–1509. Musial, J. 2012. Antiphospholipid antibodies and thrombosis. Thromb. Res. 129: 345–347. Dafer, R. M., J. Biller. 2008. Antiphospholipid syndrome: role of antiphospholipid antibodies in neurology. Hematol. Oncol. Clin. North. Am. 22: 95–105. Turiel, M., P. Sarzi-Puttini, R. Peretti, et al. 2005. Thrombotic risk factors in primary antiphospholipid syndrome: a 5-year prospective study. Stroke. 36: 1490–1494. Krnic-Barrie, S., C. R. O’Connor, S. W. Looney, et al. 1997. A retrospective review of 61 patients with antiphospholipid syndrome. Analysis of factors influencing recurrent thrombosis. Arch. Intern. Med. 157: 2101–2108. The APASS investigators. 2004. Antiphhospholipid antibodies and subsequent thrombo-occlusive events in patients with ischemic stroke. J. Am. Med. Assoc. 291: 576–584. Ortel, T. L. 2012. Antiphospholipid syndrome: laboratory testing and diagnostic strategies. Am. J. Hematol. 87: S75–S781. Male, C., D. Foulon, H. Hoogendoorn, et al. 2005. Predictive value of persistent versus transient antiphospholipid antibody subtypes for the risk of thrombotic events in pediatric patients with systemic lupus erythematosus. Blood. 106: 4152–4158. Ludia, C., P. Domenico, C. Monia, et al. 1998. Antiphospholipid antibodies: a new risk factor for restenosis after percutaneous transluminal coronary angioplasty? Autoimmunity. 27: 141–148. Szodoray, P., T. Tarr, J. Tumpek, et al. 2009. Identification of rare anti-phospholipid/protein co-factor autoantibodies in patients with systemic lupus erythematosus. Autoimmunity. 42: 497–506. Ruiz-Iraztorza, G., M. Crowther, W. Branch, and M. A. Khamashta. 2010. Antiphospholipid syndrome. Lancet 376: 1498–1509. Ruiz-Irastorza, G., and M. A. Khamashta. 2005. Stroke and antiphospholipid syndrome: the treatment debate. Rheumatology. 44: 971–974. Levine, S. R., L. Salowich-Palm, K. L. Sawaya, et al. 1997. IgG anticardiolipin antibody titers440 GPL and the risk of subsequent thrombo-occlusive events and death. Stroke. 28: 1660–1665. The Antiphospholipid Antibodies and Stroke Study Group (APASS). 1997. Anticardiolipin antibodies and the risk of recurrent thrombo-occlusive events and death. Neurology. 48: 91–94. Verro, P., S. R. Levine, and G. E. Tietjen. 1998. Cerebrovascular ischemic events with highpositive anticardiolipin antibodies. Stroke. 29: 2245–2253. Bogousslavsky, J., G. van Melle, F. Regli. 1988. The Lausanne Stroke Registry: analysis of 1,000 consecutive patients with first stroke. Stroke. 19: 1083–1092. Arboix, A., J. Alvarez-Sabin, and L. Soler. 1998. Ad hoc Editorial Committee of the Task Force on Cerebrovascular Diseases of SEN. Stroke: classification and diagnostic criteria. Neurologia. 13: 3–10. Brott, T., H. P. Adams Jr , C. P. Olinger, et al. 1989. Measurements of acute cerebral infarction: aclinical examination scale. Stroke. 20: 864–870.
Antiphospholipid antibodies and stroke severity
7
21. Van Swieten, J. C., P. J. Koudstaal., M. C. Visser, et al. 1988. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 19: 604–607. 22. Vlagea, A., A. Gil, M. V. Cuesta, et al. 2013. Antiphosphatidylserine/prothrombin antibodies (aPS/PT) as potential markers of antiphospholipid sindrome. Clin. Appl. Thromb. Hemost. 19: 289–296. 23. Miyakis, S., M. D. Lockshin, T. Atsumi, et al. 2006. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J. Thromb. Haemost. 4: 295–306. 24. Pengo, V., A. Tripodi, G. Reber, et al. 2009. 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. 7: 1737–1740. 25. Reber, G., and P. Moerloose. 2004. Anti-b2-glycoprotein I antibodies – when and how should they be measured? Thromb. Res. 114: 527–531. 26. Detkov, D., A. Gil-Aguado, P. Lavilla, et al. 1999. Do antibodies to b2-glycoprotein 1 contribute to the better characterization of the antiphospholipid syndrome? Lupus. 8: 430–438. 27. De Groot, P. G. 2011. Mechanisms of anti-phospholipid antibody formation and action. Thromb. Res. 127: S40–S42. 28. Galli, M., G. Borrelli, E. M. Jacobsen, et al. 2007. Clinical significance of different antiphospholipid antibodies in the WAPS (warfarin in the antiphospholipid syndrome) study. Blood. 110: 1178–1183. 29. Atsumi, T., O. Amengual, S. Yasuda, et al. 2005. Research around beta 2-glycoprotein I: a major target for antiphospholipid antibodies. Autoimmunity. 38: 377–381. 30. Andreoli, L., M. Fredi, C. Nalli, et al. 2013. Antiphospholipid antibodies mediate autoimmunity against dying cells. Autoimmunity. 46: 302–306. 31. Urbanus, R.T., B. Siegerink, M. Roest, et al. 2009. Antiphospholipid antibodies and risk of myocardial infarction and ischaemic stroke in young women in the RATIO study: a casecontrol study. Lancet. Neurol. 8: 998–1005. 32. Brey, R. L., C. L. Stallworth, D. L. McGlasson, et al. 2002. Antiphospholipid antibodies and stroke in young women. Stroke. 33: 2396–2400. 33. Matsuura, E., Y. Igarashi, M. Fujimoto, et al. 1990. Anticardiolipin cofactor(s) and differential diagnosis of autoimmune disease. Lancet. 336: 177–178. 34. Galli, M., P. Comfurius, C. Maassen, et al. 1990. Anticardiolipin antibodies (ACA) directed not to cardiolipin but to a plasma protein cofactor. Lancet. 335: 1544–1547. 35. Roubey, R. A., R. A. Eisenberg, M. F. Harper, and J. B. Winfield. 1995. Winfield. ‘‘Anticardiolipin’’ autoantibodies recognize beta 2glycoprotein I in the absence of phospholipid. Importance of Ag density and bivalent binding. J. Immunol. 154: 954–960. 36. Ali, H. Y., and Z. A. Abdullah. 2008. Anti-beta(2)-glycoprotein I autoantibody expression as a potential biomarker for strokes in patients with anti-phospholipid syndrome. J. Immunotoxicol. 5: 173–177. 37. Bevers, E. M., M. Galli, T. Barbui, et al. 1991. Lupus anticoagulant IgG’s (LA) are not directed to phospholipids only, but to a complex of lipidbound human prothrombin. Thromb. Haemost. 66: 629–632.
