Antiphospholipid antibodies in young Indian patients with strokeMN Mishra1, S Rohatgi2
1 Department of Pathology, Command Hospital, Southern Command, Pune - 411 040, India
2 Military Hospital, Cardiothoracic Centre, Pune - 411 040, India
Background : Stroke may be caused by antiphospholipid antibodies (APL), especially in young persons without other risk factors. Aim : The aim of this study was to determine the prevalence of two clinically significant APL-anticardiolipin antibody (ACL) and lupus anticoagulants (LA) in young patients presenting with sudden neurological deficit. Settings and Design : A case-control study performed in a large tertiary care government hospital. Materials and Methods : Ten milliliters of blood was collected from 51 consecutive young patients (age less than 45 years) diagnosed as ischemic stroke and 50 healthy age- and sex-matched controls. Statistical Analysis : Statistical analysis was done using Epi Info TM 6 software. Results : Overall, the risk factor profile was: Smoking (23.5%), positive family history (13.7%), hyperlipidemia (8%), and diabetes mellitus (3.6%). APL (LA and ACL) were present in 29.4% of the samples and in 4% of controls. The 'P' value for ACL and LA was 0.03 and 0.02 respectively, but the maximum level of ACL was 25 GPL units only. One-fourth of the patients were smokers and one-sixth had a family history of thrombosis ( P = 0.048 and 0.036, respectively). Conclusions : APL, positive family history and smoking were significantly associated with stroke in the young. We advocate screening for APL in all young patients with stroke.
Keywords: Antiphospholipid antibodies, ischemic stroke, risk factors
Antiphospholipid antibodies (APL) are a heterogeneous spectrum of IgG, and/or IgM, or less frequently also IgA immunoglobulins, which despite their name, do not recognize phospholipids, but plasma proteins bound to suitable anionic (not necessarily phospholipid) surfaces.  Lupus anticoagulants (LA) and anti-cardiolipin antibodies (ACL) were the first two such antibodies to be described. They have the greatest clinical relevance and LA is associated with stronger risk for thrombosis. The "Sapporo" laboratory criteria for definite anti-phospholipid antibody syndrome require LA and/or ACL to be present on two or more occasions at least six weeks apart.  Presence of LA can be tested by LA-sensitive activated partial thromboplastin time (aPTT) or the dilute Russell's viper venom time (dRVVT) and one test is adequate.  ACL must be measured by a "standardized" ELISA for b2-glycoprotein I-dependent antibodies; IgG and/or IgM anticardiolipin antibodies have to be present at medium or high titers. Definition of antiphospholipid antibody syndrome (APAS) requires that at least one clinical and one laboratory criterion are met and there are revised Sapporo criteria for the diagnosis of APS.  Primary LA thrombosis syndrome is much more common than secondary type. In the former, the patients have no underlying disease.  A recent article on pediatric stroke has not mentioned APL as a likely cause, which shows that they may be having no role in this age group.  Cojocaru et al. in a study on 38 young women with stroke, but without common risk factors showed that 35% of women had anti-cardiolipin antibodies of the IgG isotype.  Another study on 160 cases and 340 controls enrolled in the Stroke prevention in young women by Brey et al. also showed that APL are independent risk factors for stroke in this population.  This study was undertaken to examine the prevalence of APL in young patients with ischemic stroke who had no other risk factors. Neurological manifestations such as dementia, epilepsy, migraine, chorea, Guillain-Barre` Syndrome and diabetic peripheral neuropathy have been anecdotally associated with APL, but none except ischemic stroke, can be definitely associated with APL. 
The study was done at a tertiary care hospital from October 2001 to January 2004 in which 28 consecutive cases of ischemic stroke presenting with primary thrombotic episode and 50 controls were analyzed. Twenty-eight cases of stroke were worked up at our center while retrospective data of 23 cases was obtained from patients who had visited the laboratory for monitoring of oral anticoagulant therapy. All these 23 cases were diagnosed as ischemic stroke. Inclusion criteria were presence of one or more of the following:-
Blood samples for LA were collected in blue-topped vacutainers from Becton Dickinson, USA containing 3.2% (0.109 M) Trisodium citrate, centrifuged at 2500 x g for 15 min within 30min of collection, re-centrifuged at 4°C to obtain platelet poor plasma (less than 10x10 9 /L) and then frozen in aliquots of 200 μl in 500-μl Eppendroff tubes. Samples for ACL were collected in sterile vacutainers with gel from Becton Dickinson, USA. Sera were separated and aliquots made as for LA. Testing for LA and ACL assay was carried out in batches of 10-15 samples. Plasma collected from normal healthy individuals and commercial controls both normal and abnormal were also tested along with each batch of patient samples for APL.
Sera were analyzed for β2- glycoprotein I- dependent ACL using commercial ELISA kit (OrgenTek, GmbH, Germany). ACL was done on a semi-automatic ELISA reader (Labsystems , USA). Briefly, sera were diluted 1:100, added to pre-coated plates along with reference standards provided by the manufacturer. The plates were washed, anti-human Horse Radish Peroxidase (HRP) conjugate added, followed by substrate addition and the subsequent color developed was measured using ELISA reader at 450 nm. ACL was expressed in GPL and MPL units, 1 unit being equivalent to one microgram of IgG (for GPL) or IgM (for MPL) anticardiolipin antibody purified from 1 ml of serum, as quoted by Gali.  A reading of IgG ACL > 10 IgG phospholipid (GPL units)/L and IgM ACL > 7 IgM phospholipid (MPL units)/L was considered positive (according to product insert and further corroborated as mean + 2 SD from values derived after testing controls).
Clotting-based tests for LA were done using St art 4 (Stago Diagnostica, France), a four-channel semiautomatic coagulometer which was calibrated with the help of commercial calibrators, prior to using each new batch of kits. The procedure of testing and criteria for LA positivity were as previously mentioned.  All patients' samples were tested in duplicate at least on two occasions more than eight weeks apart and a positive result was given only if two or more samples were abnormal. Samples from all 51 patients were screened for LA, but only 28 patients were screened for ACL as the test was not available at the zonal service hospital where data from patients of stroke on follow-up was obtained.
History of important risk factors including smoking, hyperlipidemia, diabetes mellitus and positive family history was elicited from them and the controls. Biochemical tests carried out on patients and the controls included fasting and postprandial blood sugar, lipid profile and Lipoprotein(a). Diagnostic methods for detecting thrombosis were computerized tomography (CT) scan, magnetic resonance imaging or arteriography for cerebral thrombosis. Diagnosis of transient ischemic attack (TIA) required a focal neurological deficit resolving within 24 h whereas if it persisted beyond 24 h a diagnosis of stroke was made. Multiple episodes of thrombosis occurred in four cases of stroke of which one patient also had myocardial infarction. All patients had de novo (in situ) thrombosis. Detailed history was taken to include reversible precipitating causes like smoking, dyslipidemias, hypertension, prolonged bed rest, uncontrolled diabetes mellitus, prolonged use of oral contraceptives and atherosclerosis. Echocardiography done for 11 patients, including seven cases with positive family history and four cases with multiple episodes of stroke was normal in all.
Epi Info™ 6 software was used for statistical analysis of accrued data. Chi-square test was used for data analysis . The 'P' values were calculated and both Mantel-Haenzel's and Yates's correction was done where applicable. Fisher's exact test both one- and two-tailed was done for drawing conclusions. The odds ratio was calculated for both ACL and LA.
Fifty-one patients (M/F 44/ 7) and 50 (M/F 46/4) age-matched controls were enrolled in the study. The mean age and range for patients and controls was 34.5 years (3-45 years) and 34 years (5-45 years), respectively. In 51 cases of ischemic cerebrovascular accident (CVA) the clinical presentation was stroke (n = 40) with predominance of sensory symptoms in three patients, transient ischemic attacks (n = 5) and onset with seizures (n = 6). The peak value of ACL was 25 GPL units. No patient or control had elevated IgM. Fourteen patients who had stroke and tested positive for APL (LA and/or ACL) met the criteria for APAS and these are shown in [Table 1].
Risk factors included smoking alone (n = 8), positive family history alone in one patient, both (n = 3), one patient had in addition to both, controlled NIDDM and two had raised Lp(a). LA was detected in 10/51 (19.6%) patients while ACL was positive in 6/28 (21.5%) patients including two patients that tested positive for both. On analysis of 28 patients we found that 11 patients had a combination of risk factors with APL, 10 had only one risk factor and seven patients had neither. The details of patients with positive APL are shown in [Table 1]. The frequency of acquired thrombophilia markers and risk factors in controls and patients along with 'P' values are depicted in [Table 2]. The odds ratio for LA and ACL was 6.49 and 3.51 respectively. All patients were in good health during two years follow-up after the diagnosis.
In the present series, elevation of ACL IgG was seen in 6/28 (21.4%) patients and in 4% of controls, Similar results (prevalence of ACL antibodies: 23% in young stroke patients compared to 3.2% in controls) were obtained by Nagaraja et al.  Chandrasekhar et al. detected ACL antibodies in 25% of patients with thrombosis of central nervous system arteries. 
The association of APL with stroke is still controversial and there are studies which show association; ,,,,,,, while others have yielded contradictory findings. , The Framingham cohort and offspring study on an older population showed raised ACL to be significantly associated with an increased risk of TIA/stroke in women but not men. This variation may be due to racial and ethnic factors, accelerated atherosclerosis associated with greater prevalence of smoking in men and hormonal differences.  Kitagawa et al. also mention that APL, including phosphatidylserine-dependent anti-prothrombin antibody, are independent risk factors for first stroke, especially in young women. 
LA was detected in nearly 20% of patients. Levels and isotype of LA are not relevant for diagnostic workup. The titers of IgG ACL were only moderately high [Table1] and this was also seen in two previous studies. , The Antiphospholipid Antibodies in Stroke Study (APASS) reported that in patients with anti- phospholipid antibody, the risk of cerebral infarction was 2.31 times higher than in those negative for the antibody.  A study by Van Goor et al. on 128 consecutive patients with stroke or TIA showed that APL is not a strong risk factor for recurrent strokes, TIA or other thrombotic episode in young women with previous stroke. 
This raises the question whether monitoring for APL is required in such young patients with stroke. In this study both LA and APL showed significant correlation to the occurrence of stroke in the patients ( P < 0.05) which is different from the observation of Gali et al. who concluded that measuring LA is helpful to define a patient's risk for arterial and venous thrombosis, but the role of ACL is less convincing.  The study findings are limited by smaller sample size and post-event sampling. Strengths include case-control study, complete workup and adequate follow-up of patients, use of commercial kit with calibrators and reference controls as well as stringent criteria for their inclusion - namely absence of any coexistent morbidities.
APL is associated with ischemic stroke in young persons and modifiable risk factors like NIDDM, smoking and hyperlipidemia must be controlled to minimize the risk of stroke. Thromboprophylaxis may be commenced to prevent mortality and morbidity from stroke. Further epidemiological data of the Indian population is required for evaluating the strategy for further research on thrombosis in this condition.
Dr. (Mrs) Vani Suryam for statistical analysis.
[Table 1], [Table 2]