Thrombotic Risk Factors In Primary Antiphospholipid Syndrome
A 5-Year Prospective Study
Background and Purpose— Because thromboembolic events are frequently observed in primary antiphospholipid syndrome (PAPS), we assessed the risk factors for new thrombotic episodes.
Methods— Fifty-six PAPS patients (mean age, 37±10 years) were prospectively studied for 5 years. The preliminary Sapporo classification criteria for antiphospholipid syndrome (APS; a medium–high anticardiolipin antibody [aCL] titer and/or a positive lupus anticoagulant [LA] test in the presence of vascular thrombosis and/or pregnancy morbidity) were used to confirm the diagnosis. Thrombotic episodes or pregnancy losses before a diagnosis of PAPS were considered events, and any new disease manifestation other than thrombocytopenia was considered a recurrent event. Only patients with objectively verified thrombotic events were included in the study.
Results— Twenty-one new thrombotic events were observed in 15 subjects (26.8%), including 3 (5.4%) who died during the follow-up. The patients with IgG aCL levels of >40 IgG phospholipid unit (GPL-U) showed a higher incidence of new thrombotic events (43.3%) than those with levels of ≤40 GPL-U (7.7%). Univariate analysis identified a history of recurrent clinical events (P=0.004), a highly positive aCL titer (P=0.007), and the presence of cardiac abnormalities (P=0.036) as significant risk factors for new thrombotic events. A multivariate regression model confirmed that an IgG aCL titer of >40 GPL-U was an independent risk factor for thrombosis (odds ratio, 9.17; 95% confidence interval, 1.83 to 46.05).
Conclusions— A high IgG aCL titer is the strongest predictor of new thrombotic events in PAPS patients.
Antiphospholipid syndrome is a distinct clinical syndrome with a wide spectrum of clinical manifestations that is identified by the presence of antiphospholipid antibodies (aPL), anticardiolipin antibodies (aCL), and lupus anticoagulant (LA), and related to recurrent venous and/or arterial thrombosis and pregnancy morbidity.1,2⇓ It is called primary antiphospholipid syndrome (PAPS) in the absence of an underlying connective tissue disorder, whereas secondary syndrome is frequently observed in systemic lupus erythematosus.3
In the absence of clinical complications, the presence of aPL does not indicate antiphospholipid syndrome,4 and their role in predicting new thrombotic events (particularly recurrent ischemic stroke) is controversial. Recent studies5,6⇓ have found that aPL positivity does not increase the risk of recurrent thrombo-occlusive events or cause a differential treatment response; however, the results of a large-scale 4-year prospective study by Finazzi et al7 indicated that previous thrombosis and an IgG aCL titer of >40 IgG phospholipid unit (GPL-U) were independent predictors of new thrombotic episodes in a cohort of unselected patients with aPL, and we have recently found a correlation between potential embolic sources and aCL titers of >40 GPL-U in PAPS patients.8
Because no large-scale prospective studies have specifically addressed the prevalence of new thrombotic events in PAPS, we observed the clinical and serological changes occurring in PAPS patients over a 5-year follow-up period with the aim of determining the predictors of new thrombotic events.
Patients and Methods
Started in 1996, this 5-year prospective, nonrandomized cohort study enrolled 56 PAPS patients (8 men and 48 women) aged 37±10 years (range, 20 to 68) first seen at the Department of Internal Medicine and the Rheumatology Unit of L. Sacco University Hospital (Milan). The diagnosis of PAPS was established on the basis of accepted published criteria,9,10⇓ and no other diagnosis was possible. The patients were seen at our clinic every 6 to 12 months as outpatients: each visit included clinical and laboratory examinations and, if necessary, instrumental investigations. Any new disease manifestation other than thrombocytopenia was considered a recurrent clinical event.
The study was approved by our Institute’s Human Research Committee, and all of the subjects gave their informed consent.
IgG and IgM aCL levels were measured by means of an enzyme-linked immunosorbent assay and expressed in GPL or IgM phospholipid (MPL) units (GPL-U and MPL-U): IgG aCL positivity was defined as levels of >15 GPL-U and IgM aCL positivity as values >5 MPL-U, which are our laboratory standards.8 In line with other authors,7,11⇓ we considered titer ranges as being highly (>40 GPL-U) or not highly positive (between 15 and 40 GPL-U).
Anticoagulant therapy was monitored by means of prothrombin time tests using various thromboplastins. The results were expressed as international normalized ratios (INR).
The laboratory detection of LA is based on the initial use of phospholipid-depleted or platelet-depleted coagulation tests, such as kaolin clotting time, dilute Russell’s venom viper time, the tissue thromboplastin inhibition test, and dilute activated partial thromboplastin time. The LA tests were not performed while the patients were receiving anticoagulant therapy.
Thrombocytopenia was defined as platelet counts of <150×109/L on 2 different occasions.
Diagnosis of Thrombotic Events
Thrombotic episodes or pregnancy losses before the diagnosis of PAPS were considered events. All new disease manifestations other than thrombocytopenia during the study period were considered recurrent (new) events. Only patients with objectively verified thrombotic events were included in the study. Pregnancy losses from women with histopathological lesions of decidua or placental tissue typical of antiphospholipid syndrome (infarction, intravascular fibrin deposition, syncytial knot formation, and fibrosis) were considered aPL-related events, and the patients had PAPS diagnosed even in the absence of previous venous or arterial thrombotic events.12,13⇓
The diagnostic methods used to detect thrombosis during the follow-up were ultrasonography for deep vein thrombosis, radionuclide lung scanning or angiography for pulmonary embolism, arteriography for peripheral arterial occlusions, and computed tomography (CT) or angiography for cerebral thrombosis.
The diagnosis of a cerebral transient ischemic attack required a focal neurological deficit resolved within 24 hours and the CT exclusion of a cerebral hemorrhage. The clinical events occurring during the follow-up were classified as: (1) venous thrombosis (lower limbs, pulmonary embolism, retinal, cerebral and mesenteric thrombosis); (2) arterial thrombosis (retinal, renal, cerebral, mesenteric, coronary or peripheral arteries); (3) pregnancy events (spontaneous abortion before 20 weeks, miscarriage after 20 weeks, including abortion caused by an antiphospholipid syndrome-related indication); or (4) catastrophic events (clinical evidence of multiple organ involvement over a short period).5
All of the patients underwent transesophageal echocardiography (TEE) using a commercial ultrasound system (Philips, Sonos 5500) equipped with a multiplane transducer in our day hospital, except for 4 who were examined during hospitalization because of a thromboembolic event.
Mild sedation (diazepam 10 mg intravenous) was induced 5 to 10 minutes before the examination, followed by lidocaine spray local oropharynx anesthesia. The results were reviewed by 2 blinded expert echocardiographers. A mitral valve thickness of >5 mm and an aortic valve thickness of >3 mm were considered abnormal, as previously reported.14 All of the measurements were calculated on 2 planes (horizontal and longitudinal). The following were considered potential embolic sources: the presence of severe spontaneous echocontrast in the left atrium or appendage, intracardiac thrombi, and non-bacterial valvular masses (Libman-Sacks endocarditis).8
Forty-seven patients (83.9%) underwent repeat TEE, including 3 who died before the end of the study and 9 who were lost to follow-up.
The treatments used during the 5-year study period were: (1) venous thrombosis: anticoagulant therapy was prolonged for at least 6 months (12 months in the case of pulmonary embolism); (2) arterial thrombosis: long-term anticoagulant and/or antiplatelet therapy; and (3) patients with pregnancy events alone were treated with life-long antiplatelet therapy and life-long anticoagulant therapy in the case of a new thrombotic event.5
The antithrombotic drug was acenocoumarin (INR >3.0), and the antiplatelet drug was acetylsalicylic acid (100 mg).
The primary aim of the study was to determine the rate of recurrent thrombosis over 5 years. The secondary aim was to identify the clinical and serological risk factors for new thrombotic events.
Mean values and standard deviations were calculated for the continuous variables. The quantitative variables were compared using the Student t test for grouped data, and the qualitative variables by means of the χ2 test with Yates correction when indicated. A 2-tailed P<0.05 was considered statistically significant.
The follow-up of each patient was defined as the individual person-time at risk measured in patient-years (pt-yrs). The sum of the individual pt-yrs represented the total pt-yrs. The person-time incidence of thrombosis was calculated by dividing the number of new thrombotic episodes (ie, incidence events) by total pt-yrs, and expressed as episodes %pt-yrs, or episodes %pt-yrs.
The association of specific risk factors with the incidence of thrombosis was estimated using the standard χ2 test. The independent effects of selected risk factors were assessed by means of multivariate logistic regression analysis, with the odds ratios and the confidence intervals for each variable being adjusted for all of the other variables in the model.
Table 1 shows the main characteristics of the studied patients. At enrollment, 16 patients (28.6%) had experienced at least 1 venous or arterial thrombosis, 27 (48.2%) >1 thrombotic event (often associated with fetal loss), and 13 (23.2%) fetal losses.
During 5-year follow-up period, 15 patients (26.8%) experienced 21 new thrombotic events (Table 2).
Three patients (5.4%) with persistently high aCL titers (>40 GPL-U) died during the follow-up: one caused by hemorrhagic complications after valvuloplasty, one caused by acute myocardial infarction, and one caused by multiple organ involvement (catastrophic syndrome) with disseminated intravascular coagulation syndrome.
Of the 9 pregnancies observed in 6 women, 3 were unsuccessful in 2 patients: 2 because of major bleeding (including menometrorrhagia in one case), and 1 because of macrohematuria. Both patients were treated with acenocoumarin.
Cardiac involvement was found in 34 patients (60.7%) at baseline, with mitral valve thickening being the most frequent abnormality (88.2%). Embolic sources were demonstrated in 14 patients (25%): 9 showing spontaneous echocontrast and 5 with Libman-Sacks endocarditis. Follow-up TEE revealed unchanged cardiac involvement in 27/44 subjects (61.4%), a worsening in previous valve lesions in 11 (25%), and new lesions in the remaining 6 (13.5%) (Table 2).
Relationships Between Laboratory Test Results and the Occurrence of Thrombosis
The mean baseline IgG aCL titer was 57±51 GPL-U (range, 2.4 to 176), and the mean baseline IgM aCL titer was 12±14 MPL-U (range, 1.0 to 53.7).
On the basis of the IgG aCL titer proposed by Finazzi et al,7 the patients were divided into 2 groups: those with antibody titers of ≤40 GPL-U (n=26; mean IgG, 12.9±11 GPL-U) and those with antibody titers of >40 GPL-U (n=30; mean IgG, 94.9±39 GPL-U). The frequency of associated events (arterial and/or venous thrombosis or fetal loss) during the follow-up period was higher in the latter (57% versus 27%) who showed a particularly high incidence of thrombosis (44.8% versus 7.7%; χ2=7.296; P<0.001).
In line with previously published data, the IgM isotypes did not show any clinical correlations.
Risk Factors for Thrombosis
The patients’ clinical and laboratory characteristics at diagnosis were analyzed using univariate and multivariate models to evaluate their predictive value. Univariate analysis showed that the significant predictors of thrombosis were a high aCL titer (>40 GPL-U), multiple previous thrombotic events, and cardiac abnormalities (Table 3). The multivariate logistic regression analysis confirmed only an aCL titer of >40 GPL-U as an independent risk factor for thrombosis in our patients (OR, 9.17; 95% CI, 1.83 to 46.05).
To the best of our knowledge, this is the first study in which multivariate analysis has identified a persistently high aCL titer as an independent predictor of thrombosis in a cohort of PAPS patients, although other authors have reported a close relationship between thrombotic events and high aCL levels.8,11,15⇓⇓
In particular, Finazzi et al7 found that the main independent factors influencing the risk of vascular complications in aCL-positive patients were previous thrombosis and a high IgG aCL titer (>40 GPL-U), and previous studies have identified a high aCL titer as an independent risk factor for atherosclerotic vascular diseases, mainly myocardial infarction, stroke, and peripheral occlusion.11,16–18⇓⇓⇓
Univariate analysis of our prospective data showed that an aCL titer of >40 GPL-U is indicative of a significantly increased risk of thrombotic events in PAPS patients (time at risk, 8.7% pt-yrs versus 1.5% pt-yrs; P<0.01), although the results of the recent APASS study6 suggest that aPL positivity did not increase the risk for recurrent thrombotic events in young patients.
Another important finding of our study is that the patients who had experienced multiple clinical thrombotic events before enrolment had an increased 5-year risk of new events (9.0% pt-yrs versus 1.5% pt-yrs; P<0.01) despite optimal antithrombotic treatment.
Our patients showed a high incidence (41%) of thromboembolic sources (spontaneous echocontrast and Libman-Sacks endocarditis). In a recent similar 5-year prospective TEE follow-up study of PAPS patients Zavaleta et al14 found new valvular lesions in 25% and progressive lesions in 50%, thus suggesting that anticoagulant and/or antiplatelet treatment failed to resolve valvular vegetations completely. Our finding of new valvular abnormalities in 17 patients (38.6%) confirms that different mechanisms may play a role in their development.
Although mainly treated with acenocoumarin (INR >3.0), our patients with aCL titers of >40GPL-U showed a higher incidence of clinical thromboembolic events despite a significant progressive reduction in the titers themselves (from 97.3±42.7 to 51.6±36.8 GPL-U; P<0.01).
Nevertheless, some authors5,19⇓ have reported that high-dose acenocoumarin therapy may not only reduce the risk of recurrent vascular events but may also favor the disappearance of cardiac valvular lesions, although Crowther et al9 found no difference in the effectiveness of high-intensity and moderate-intensity anticoagulant therapy in preventing recurrent thrombosis in PAPS patients. The optimal management of anticoagulant or antiplatelet prophylaxis/treatment needs to be defined in long, prospective, randomized clinical trials as the risk of recurrent thrombosis is still high in such patients.19
In conclusion, our long-term follow-up study demonstrated that despite optimal antithrombotic treatment, multiple previous thrombotic events, aCL titers of >40 GPL-U, and cardiac abnormalities are all predictors of new thrombotic events. Multivariate logistic regression analysis confirmed that an IgG aCL titer of >40 GPL-U is an independent risk factor for thrombosis in PAPS patients. It is now important to establish whether the thromboembolic risk of PAPS patients is different from that of patients with antiphospholipid antibody syndrome associated with systemic lupus erythematosus.
- Received March 25, 2005.
- Revision received April 21, 2005.
- Accepted April 26, 2005.
Cervera R, Piette JC, Font J, Khamashta MA, Shoenfeld Y, Camps MT, Jacobsen S, Lakos G, Tincani A, Kontopoulou-Griva I, Galeazzi M, Meroni PL, Derksen RH, de Groot PG, Gromnica-Ihle E, Baleva M, Mosca M, Bombardieri S, Houssiau F, Gris JC, Quere I, Hachulla E, Vasconcelos C, Roch B, Fernandez-Nebro A, Boffa MC, Hughes GR, Ingelmo M. Euro-Phospholipid Project Group. Antiphospholipid syndrome: clinical and immunologic manifestations and patterns of disease expression in a cohort of 1,000 patients. Arthritis Rheum. 2002; 46: 1019–1027.
Asherson RA, Cervera R. Antiphospholipid antibodies and heart. Lessons and pitfalls for the cardiologist. Circulation. 1991; 84: 920–923.
Finazzi G, Brancaccio V, Moia M, Ciaverella N, Mazzucconi MG, Schinco PC, Ruggeri M, Pogliani EM, Gamba G, Rossi E, Baudo F, Manotti C, D’Angelo A, Palareti G, De Stefano V, Berrettini M, Barbui T. Natural history and risk factors for thrombosis in 360 patients with antiphospholipid antibodies: a four-year prospective study from the Italian registry. Am J Med. 1996; 100: 530–536.
Turiel M, Muzzupappa S, Gottardi B, Crema C, Sarzi-Puttini P, Rossi E. Evaluation of cardiac abnormalities and embolic sources in primary antiphospholipid syndrome by transesophageal echocardiography. Lupus. 2000; 9: 406–412.
Crowther MA, Ginsberg JS, Julian J, Denburg J, Hirsh J, Douketis J, Laskin C, Fortin P, Anderson D, Kearon C, Clarke A, Geerts W, Forgie M, Green D, Costantini L, Yacura W, Wilson S, Gent M, Kovacs MJ. A comparison of two intensities of warfarin for the prevention of recurrent thrombosis in patients with the antiphospholipid antibody syndrome. N Engl J Med. 2003; 349: 1133–1138.
Wilson WA, Gharavi AE, Koike T, Lockshin MD, Branch DW, Piette JC, Brey R, Derksen R, Harris EN, Hughes GR, Triplett DA, Khamashta MA. International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum. 1999; 42: 1309–1311.
Levine SR, Salowich-Palm L, Sawaya KL, Perry M, Spencer HJ, Winkler HJ, Alam Z, Carey JL. IgG anticardiolipin antibody titer >40 GPL and the risk of subsequent thrombo-occlusive events and death. A prospective cohort study. Stroke. 1997; 28: 1660–1665.
Zavaleta NE, Montes RM, Soto ME, Vanzzini NA, Amigo MC. Primary antiphospholipid syndrome: a 5-year transesophageal echocardiographic followup study. J Rheumatol. 2004; 31: 2402–2407.
Vaarala O, Manttari M, Manninen V, Tenkanen L, Puurunen M, Aho K, Palosuo T. Anti-cardiolipin antibodies and risk of myocardial infarction in a prospective cohort of middle-aged men. Circulation. 1995; 91: 23–27.
Janardhan V, Wolf PA, Kase CS, Massaro JM, D’Agostino RB, Franzblau C, Wilson PWF. Anticardiolipin antibodies and risk of ischemic stroke and transient ischemic attack: the Framingham Cohort and Offspring Study. Stroke. 2004; 35: 736–741.