• β2GPI;
  • Lupus anticoagulant;
  • Anticardiolipin;
  • Antiphospholipid;
  • Pediatric


  1. Top of page
  2. Abstract
  7. Acknowledgements


To determine whether serum β2-glycoprotein I antibody (anti-β2GPI) detection improves identification of pediatric subjects at risk for antiphospholipid syndrome (APS).


Serum antiphospholipid antibodies (aPL) were identified by anticardiolipin enzyme-linked immunosorbent assay (ELISA), lupus anticoagulant assays, and syphilis screening in children with primary APS, systemic lupus erythematosus (SLE), or SLE plus APS. Anti-β2GPI level and isotype were determined by β2GPI ELISA and correlated with clinical manifestations and other aPL assays.


One hundred-ten subjects under 22 years of age and of mixed ethnicity were evaluated. Fifty-seven had SLE (including 14 with APS), 25 had primary APS, 16 had SLE-like APS, 6 were healthy children with aPL detected incidentally, 4 had other rheumatic diseases and 2 had other conditions. Anti-β2GPI were detected in 48% of SLE subjects and did not improve aPL detection over standard tests. Anti-β2GPI were associated with stroke (P = 0.014), but not with other APS manifestations, and were rarely detected in primary APS. Among subjects with APS manifesting as chronic thrombocytopenia, anti-β2GPI distinguished subjects with SLE from those with primary APS.


With the exception of stroke, anti-β2GPI detection does not improve identification of pediatric APS over that of traditional aPL assays. Anti-β2GPI are rare in pediatric primary APS, but may predict evolution of chronic thrombocytopenia to SLE.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Antiphospholipid antibodies (aPL) are a heterogeneous group of autoantibodies reactive to phospholipid or phospholipid-binding protein complexes that can be detected by lupus anticoagulant (LAC) assays, anticardiolipin (aCL) enzyme-linked immunosorbent assays (ELISA), or screening tests for syphilis. Recent efforts to better characterize the specificity of aPL have resulted in the identification of autoantibodies directed against β2-glycoprotein I (anti-β2GPI), a phospholipid-binding protein. Evidence suggests that β2GPI may be necessary for the binding of particular aPL to phospholipids and may actually represent the true target of aPL. The physiologic function of β2GPI is not known; however, in vitro studies have demonstrated a potential role in anti-coagulation (1, 2).

Antiphospholipid syndrome (APS) has been defined as the association of aPL with arterial and venous thrombosis, fetal loss, thrombocytopenia, and other hematologic and neurologic complications. Recent stringent criteria for APS have been developed in adult populations (3); however, similar criteria have not been developed for pediatric populations. As in adults, APS can be seen in children with systemic lupus erythematosus (SLE) (secondary APS) or in children without evidence of an underlying autoimmune disorder (primary APS). In general, both autoimmunity and thrombosis occur less frequently in children than adults, suggesting that children may exhibit a different relationship between aPL and clinical manifestations of APS.

Many adult and pediatric patients with SLE have detectable aPL in their serum, yet never develop the clinical complications characteristic of APS. Additionally, patients may develop clinical complications of APS without testing positive for aPL by standard assays. This inconsistency likely reflects the underlying heterogeneity of aPL, the indirect nature of many of the aPL assays, in particular LAC, and the currently unexplained role of aPL in pathophysiology of APS (4). The standard methods for detection of aPL include coagulation assays for LAC (partial thromboplastin time [PTT], and dilute Russell's viper venom time [dRVVT]), detection by ELISA of aCL, and screening tests for syphilis (VDRL and rapid plasma reagin [RPR]). In general, there is good correlation across assays for aPL detection. However, the discordant behavior of some patient sera suggests that there may be a number of aPL antibody subsets that recognize distinct phospholipid or phospholipid-binding protein antigens.

Previous observations in adult cohorts suggest that anti-β2GPI are frequently seen in SLE and APS, and may demonstrate greater specificity for APS than standard aPL assays (5–7). However, this has not been assessed in a pediatric cohort. Although earlier studies in children have evaluated the prevalence of aPL and clinical characteristics of APS in children, none have included detection of anti-β2GPI. This investigation was undertaken to determine if measurement of serum anti-β2GPI improves the clinical characterization of pediatric patients with aPL. Children and adolescents with primary APS, SLE alone, or secondary APS were studied. The isotype and level of anti-β2GPI were correlated with clinical manifestations of hypercoagulability and results of other aPL assays. The goals of this study were to determine whether a particular aPL assay correlated best with specific clinical manifestations and whether β2GPI screening improved identification of children at risk for APS.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Study design.

This is a cross-sectional cohort study of subjects presenting with either SLE (with or without aPL) or aPL positivity (either uncomplicated or associated with APS). Subjects were recruited consecutively from 1996 to 1999 through the pediatric rheumatology clinic at the University of California, San Francisco Medical Center (UCSF), which serves as a referral site for general pediatric, hematology, neurology, and nephrology clinics throughout northern California. Clinical evaluation included physical examination by a pediatric rheumatologist and medical record review. Clinical manifestations of APS were confirmed by medical record, and thrombotic events were confirmed by review of radiographic evidence, such as venography or ultrasound for deep venous thrombosis (DVT), ventilation/perfusion scan for pulmonary embolism (PE), and computerized tomography or magnetic resonance imaging for central nervous system (CNS) events. Laboratory evaluation included screening for the following autoantibodies: antinuclear antibody (ANA) by immunofluorescence; and anti-double-stranded DNA (dsDNA), anti-SMITH, anti-ribonucleoprotein (RNP), anti-Ro and anti-La by immunodiffusion with enzyme immunoassay quantification. The presence of aPL was determined utilizing the following assays: aCL by ELISA, LAC by either PTT or dRVVT, and either RPR or VDRL. Assays were performed either in the UCSF clinical laboratory or in standard reference laboratories. All sera demonstrating prolonged clotting in a LAC assay were evaluated for evidence of an inhibitor by 1:1 mixing with normal plasma and if present, a LAC was confirmed according to standard criteria (8). Positive results were repeated and demonstrated to be positive on 2 separate specimens obtained at least 8 weeks apart. Analyses for protein S and C deficiency and factor V Leiden mutation were performed. All subjects provided informed consent according to the UCSF Committee for Human Research.

Study population.

The study population consisted of 110 subjects < 22 years of age. Subjects were classified according to the following characteristics adapted from the literature (9–12): 1) aPL positivity, defined as a positive aCL, LAC, or false positive test for syphilis on 2 separate occasions; 2) APS, based on the presence of 1 major criteria (arterial or venous thrombosis, pulmonary hypertension, peripheral arterial gangrene, cutaneous necrosis, or isolated chronic thrombocytopenia [defined as <100,000 platelets and, for primary APS, the presence of antiplatelet antibodies or bone marrow aspirate and biopsy findings consistent with idiopathic thrombocytopenia or, for SLE or SLE-like APS, isolated chronic thrombocytopenia in absence of other organ involvement]); or 2 minor criteria (livedo reticularis, migraine headache, chorea, Bell's palsy, psychosis, or severe Raynaud's phenomenon) plus aPL positivity; and 3) the presence of an underlying disorder defined as primary APS (if APS existed as an isolated phenomenon in the absence of malar rash, palatal ulcer, nasal ulcer, arthritis, major organ involvement [e.g., hepatitis]), proteinuria [>0.5 gm/dl], lymphopenia [<1,000 cells/mm3], anti-dsDNA, anti-Sm, anti-RNP positivity, or ANA titer >1:320); or secondary APS (if the patient met American College of Rheumatology [ACR] criteria for SLE [13] or mixed connective tissue disease [MCTD]; and SLE-like APS (if patients met 1–3 ACR criteria for SLE).

Anti-β2GPI assay.

Serum anti-β2GPI levels were determined by standard ELISA utilizing QUANTA LiteTM β2GPI kits (INOVA Diagnostics, San Diego, CA). Serum samples were first analyzed by screening ELISA; all samples with optical density (OD) >0.46 were reassayed for anti-β2GPI isotype (IgG, IgA, IgM) and level by confirmatory ELISA, according to the manufacturer's instructions. Titers are expressed in Standard IgG β2GPI IgG Units (SGU), Standard IgA β2GPI IgA Units (SAU) and Standard IgM β2GPI IgM Units (SMU).

Statistical analysis.

Differences for unordered categorical (nominal) outcomes were assessed using Fischer's exact test. Ordered categorical (ordinal) outcomes were compared using a trend test (14).


  1. Top of page
  2. Abstract
  7. Acknowledgements

Clinical features of the cohort.

One hundred and ten subjects (79 female, 31 male) were studied, ranging from 1 to 22 years (mean 14 years). Subjects were of mixed ethnicity (47% Caucasian, 21% Asian, 18% Hispanic, 11% African American, 3% other), reflecting referral patterns to UCSF. Fifty-seven subjects had SLE, 25 had primary APS, 16 SLE-like APS, 6 were healthy children with aPL detected incidentally, 4 had other rheumatic diseases, such as Wegener's granulomatosis (WG) or MCTD, and 2 had other conditions (Table 1). All aPL negative subjects had SLE.

Table 1. Clinical classification of the cohort
  1. * All subjects were classified according to underlying condition, presence or absence of aPL, and criteria for APS. Values are absolute number (n) of subjects in each group and percentage of the entire cohort. Total subjects = 110. aPL = antiphospholipid antibody; APS = antiphospholipid syndrome; SLE = systemic lupus erythematosus.

Incidental finding of aPL65
Primary APS2523
 SLE/no aPL19 
 SLE/aPL-no APS24 
Other rheumatic disorder/APS44

Fifty-eight subjects met criteria for APS, of which 25 (43%) had primary APS, 14 (24%) had SLE, 15 (26%) had SLE-like APS, and 4 (7%) had another rheumatic disorder associated with APS. Subjects with APS were more likely to be Caucasian or Hispanic, and less likely to be Asian than subjects without APS (P < 0.0005). There were no differences in age or sex between subjects with and without APS. Clinical manifestations of APS were diverse (Figure 1) and included chronic thrombocytopenia (n = 17), DVT (n = 7), and stroke (n = 4). Raynaud's phenomenon, migraine headache, pulmonary hypertension, cutaneous necrosis, and gastrointestinal necrosis occurred exclusively in primary and SLE-like APS subjects. Eleven subjects (7 female, 4 male) had a history of thrombosis, including 4 with stroke, 7 with DVT, and 2 with an associated PE. Age at initial thrombotic event ranged from 7 to 17 years (mean 12.2 years; SD 3.4). Underlying disorders in subjects with thrombosis included primary APS (n = 3), SLE-like APS (n = 3), SLE (n = 3), and WG (n = 2). The cohort included 57 subjects who met ACR criteria for SLE, of which aPL were detected in 38 (67%). Fourteen of these 38 (37%) SLE subjects who tested positive for aPL also had associated clinical manifestations of APS. In total, 25% of the entire SLE cohort met criteria for APS.

thumbnail image

Figure 1. Clinical manifestation of antiphospholipid antibody for all 58 subjects who met criteria, separated according to underlying condition (systemic lupus erythematosus [SLE] or non-SLE). DVT = deep venous thrombosis; Pulm HTN = pulmonary hypertension; GI = gastrointestinal.

Download figure to PowerPoint

Screening for abnormalities of protein C level or activity, free protein S level, or anti-thrombin III activity was negative for all subjects. Genetic analysis revealed that 1 subject with DVT in association with SLE-like APS was heterozygous for the Factor V Leiden mutation. There were no deaths in the followup period included in this study.

aPL profile.

At aPL screening of the entire cohort it was demonstrated that, although the majority of subjects had detectable aCL, the same was not observed for LAC, false-positive RPR, or anti-β2GPI (Table 2). Moreover, aCL were found in most subjects regardless of diagnosis or clinical manifestation and there was no group difference for aCL titer or isotype (Table 3). Anti-β2GPI were identified in the sera of 34 subjects. Of these samples, 7 had detectable IgG anti-β2GPI, with levels ranging from 27 SGU to >150 SGU (including 4 >100 SGU), 24 had detectable IgA anti-β2GPI, with levels ranging from 21 SAU to >150 SAU (including 3 >100 SAU), and 16 had detectable IgM anti-β2GPI, with levels ranging from 23 SMU to >150 SMU (including 4 >100 SMU). Subjects with detectable anti-β2GPI tended to be older (P = 0.054), but did not differ with regard to sex or ethnicity from subjects without anti-β2GPI.

Table 2. Summary of positive aPL assays for the entire cohort*
Assayn tested% positive
  • *

    aPL = antiphospholipid antibody; aCL = anticardiolipin; ELISA = enzyme-linked immunosorbent assay; LAC = lupus anticoagulant; PTT = partial thromboplastin time; dRVVT = dilute Russel's viper venom time; RPR = rapid plasma reagin; Anti-β2GPI = anti–β2-glycoprotein I antibody.

aCL ELISA (any isotype)11061
LAC (any assay)10737
 LAC by PTT10719
 LAC by dRVVT10526
False-positive RPR9713
 Anti-β2GPI (any  isotype)9634
 Anti-β2GPI IgG967
 Anti-β2GPI IgA9624
 Anti-β2GPI IgM9616
Table 3. Summary of positive aPL assays for each major clinical classification of subjects
n% positiven% positiven% positiven% positive
  1. * aPL = antiphospholipid antibody; SLE = systemic lupus erythematosus; APS = antiphospholipid syndrome; aCL = anticardiolipin; LAC = lupus anticoagulant; Anti-β2GPI = anti–β2-glycoprotein I antibody.


Among the 83 subjects who completed testing for LAC, aCL, and anti-β2GPI, 71 demonstrated at least one of these antibodies. As expected, these individuals demonstrated overlapping positivity across these aPL assays (Figure 2). Although many subjects had a positive result by multiple aPL methods, 34 (49%) had detectable aPL by only 1 assay. This was most notable for aCL, which was the only positive test result for 18 subjects. Additionally, 11 subjects had positive results only for LAC, 1 for RPR, and 6 for anti-β2GPI. Of the 6 subjects who were positive for aPL only by the β2GPI ELISA, 5 had SLE without evidence of APS and 1 had SLE-like APS manifested as chronic thrombocytopenia and migraine headache. None of the 6 subjects had IgG anti-β2GPI; 5 had IgA anti-β2GPI, and 2 had IgM anti-β2GPI. Analysis of sensitivity and specificity of anti-β2GPI screening for aPL versus all other aPL assays (PTT, dRVVT, aCL and RPR) combined revealed a sensitivity for anti-β2GPI of 35% and a specificity of 65%.

thumbnail image

Figure 2. Venn diagram demonstrating overlapping positivity of antiphospholipid antibody assays. The number of subjects with lupus anticoagulant (LAC), anticardiolipin (aCL), and anti–β2-glycoprotein I (anti-β2GPI) antibodies is shown for the 71 subjects who completed testing with all three assays.

Download figure to PowerPoint

Association of aPL assay with clinical manifestation.

Anti-β2GPI were highly associated (P = 0.014) with stroke (4/4 stroke patients were anti-β2GPI positive with IgG and IgA detected in 1/4, IgM alone in 1/4, and IgA and IgM in 2/4). However, there was no significant association between anti-β2GPI positivity and other clinical manifestations of APS. Presence of aCL was neither significantly associated with stroke (P = 0.15), nor any other clinical manifestation including DVT, Raynaud's phenomenon, migraine headache, or thrombocytopenia. Anti-β2GPI isotype was not associated with any particular clinical manifestation. Similarly, anti-β2GPI level was not related to clinical manifestation and did not differ between subjects meeting criteria for APS and those who did not.

Surprisingly, the relationship between thrombosis (DVT, PE, or stroke) and any aPL assay (LAC, aCL, or β2GPI ELISA) was not significant for either the entire cohort or for separate analysis of SLE subjects (Table 4). In contrast, analysis for an association between aCL, LAC, or anti-β2GPI with APS (Table 5) revealed significance for aCL and LAC, both for the entire cohort and for separate analysis of SLE subjects. However, the presence of anti-β2GPI was not significantly associated with APS for either the entire cohort or for the subset with SLE.

Table 4. Association of antiphospholipid antibody with thrombosis (i.e., deep venous thrombosis, pulmonary embolism, or stroke) according to the assay performed*
 N+ thrombosis− thrombosisOR (95% CI)P
  • *

    The relative risk based on the presence of aCL, LAC and anti-β2 GPI is shown. aCL = anticardiolipin; LAC = lupus anticoagulant; anti-β2GPI = anti–β2-glycoprotein I antibody; OR = odds ratio; 95% CI = 95% confidence interval; SLE = systemic lupus erythematosus.

All subjects     
 aCL +679582.06 (0.56–7.5)0.36
 aCL −43340  
 LAC +406341.8 (0.56–5.7)0.36
 LAC −67661  
 Anti-β2 GPI +346282.4 (0.72–8.2)0.19
 Anti-β2 GPI −62557  
SLE subjects     
 aCL +303272.9 (0.38–+∞)0.61
 aCL −27126  
 LAC +132113.7 (0.59–23.8)0.23
 LAC −43241  
 Anti-β2 GPI +252231.0 (0.17–6.4)1.00
 Anti-β2 GPI −26224  
Table 5. Association of aPL with APS according to the assay performed*
 n+ aPS− aPSOR (95% CI)P
  • *

    Relative risk for APS based on the presence of aCL, LAC, and anti-β2GPI are shown for all subjects and for the subset with SLE. aPL = antiphospholipid antibody; APS = antiphospholipid syndrome. See Table 4 for additional definitions.

All subjects     
 aCL +6744233.2 (1.5–7.1)0.006
 aCL −431627  
 LAC +403195.1 (2.1–12.2)0.0003
 LAC −672740  
 Anti-β2 GPI +3417170.68 (0.29–1.6)0.40
 Anti-β2 GPI −623725  
SLE subjects     
 aCL +3011194.6 (1.2–17.6)0.03
 aCL −27324  
 LAC +13859.9 (2.5–39.2)0.002
 LAC −43637  
 Anti-β2 GPI +258172.0 (0.57–6.9)0.35
 Anti-β2 GPI −26521  

Comparison of SLE and non-SLE subjects with APS revealed a strong association between the presence of anti-β2 GPI and SLE. Among all APS subjects (Table 3), anti-β2GPI positivity increased significantly (trend test, P = 0.005) as subject classification progressed from primary (4/24,17%) to SLE-like (5/13, 38%) to SLE (8/13, 62%), suggesting a strong association of anti-β2GPI with SLE. This is in contrast to aCL positivity, which remained relatively constant across these 3 groups (trend test, P = 0.93).

SLE subjects.

Overall, 25 (48%) SLE subjects had detectable anti-β2GPI, 13 (23%) had a positive LAC assay, and 30 (53%) had detectable aCL. There was no significant difference in anti-β2GPI positivity between subjects classified as SLE-no APL/no APS (6/19, 32%) and SLE-aPL/no APS (11/19, 58%), suggesting that presence of anti-β2GPI does not increase detection of aPL in SLE subjects. Moreover, among SLE subjects, anti-β2GPI had a weaker association with APS (P = 0.35) than aCL (P = 0.03) or LAC (P = 0.002). Therefore, as also described for the entire cohort, anti-β2GPI did not demonstrate a significant association with APS in pediatric SLE subjects.

Among all subjects with APS, the strong association of anti-β2GPI with SLE suggested a potential role for anti-β2GPI as a marker of disease activity in SLE. To test this hypothesis, we evaluated the association of anti-β2GPI level with the Systemic Lupus Erythematosus Disease Activity Index and with anti-dsDNA titer, but found no statistically significant association.

aPL-associated thrombocytopenia.

APS manifested as chronic thrombocytopenia in 17 of 58 (29%) subjects with APS. These 17 individuals included 5 with primary APS, 4 with SLE-like APS, and 8 with SLE. Of these 17 subjects, further classification into the categories of primary APS, SLE-like APS, and SLE with APS did not reveal a significant group difference for aCL positivity, but did reveal a difference for anti-β2GPI positivity. Specifically, there were no subjects with primary APS manifesting as isolated chronic thrombocytopenia with detectable anti-β2GPI. The majority of subjects (75%) with SLE and APS manifesting as chronic thrombocytopenia had anti-β2GPI, and 25% of SLE-like APS subjects with chronic thrombocytopenia had anti-β2GPI. Trend testing demonstrated statistical significance for the association of anti-β2GPI with SLE and SLE-like disease (P = 0.04), in contrast to aCL positivity, among the subset with a diagnosis of APS due to chronic thrombocytopenia only (Figure 3).

thumbnail image

Figure 3. Anticardiolipin antibody (aCL) and anti–β2-glycoprotein I (anti-β2GPI) antibody results for subjects with antiphospholipid syndrome (aPS) manifested as chronic thrombocytopenia. Prevalence of antiphospholipid antibodies by detection of either aCL or β2GPI among subjects manifesting aPS as chronic thrombocytopenia according to the three major aPS subgroups. * Indicates significant association of antiβ2GPI with SLE and SLE-like disease (P = 0.04, trend test) as compared with aCL.

Download figure to PowerPoint

Five subjects initially diagnosed with isolated chronic idiopathic thrombocytopenia purpura (ITP) subsequently developed SLE, including 4 females and 1 male. Average age at development of thrombocytopenia was 9.7 years (SD 0.13), and average time between onset of thrombocytopenia and diagnosis of SLE was 1.5 years (SD 0.27). Among subjects with APS manifesting as chronic thrombocytopenia only, there was a strong association of anti-β2GPI with SLE (75%) and SLE-like disease (25%), but not with primary APS.


  1. Top of page
  2. Abstract
  7. Acknowledgements

The association between aPL and thrombosis has been well established in both adults and children with SLE and primary APS. However, many questions remain regarding the antigen specificity of aPL, the role of aPL in APS pathogenesis, and appropriate aPL assays for the clinical setting. Screening for anti-β2GPI has generated enthusiasm for better characterization of this condition. The aims of this study were to characterize the clinical and antibody features of APS in a pediatric population and to determine if anti-β2GPI detection improves identification of subjects at risk for thrombosis or other manifestations of APS.

Although all APS subjects in this cohort do not meet the recently described stringent criteria developed for diagnosis of APS in adults (3), they do demonstrate accepted clinical manifestations of the syndrome (15–17). The most frequent APS manifestation seen in this cohort was chronic thrombocytopenia. Although not specific for APS, chronic thrombocytopenia is a well-recognized complication in the spectrum of APS. The prevalence of chronic thrombocytopenia in pediatric APS has not previously been reported, however the prevalence of 29% in our cohort is similar to that previously reported in adult series (18). The positive association of APS with Caucasian and Hispanic ethnicity and negative association with Asian ethnicity, may reflect referral bias or may suggest the presence of underlying genetic factors. Many of the subjects with primary APS were Caucasian; therefore, the large number of subjects with primary APS in this cohort may have biased the association of APS with Caucasian ethnicity.

Although the literature contains numerous case reports of children with primary APS, there are no published series, and the prevalence of primary APS in childhood is not known. This report adds 25 subjects to the literature on primary APS in children. We observed anti-β2GPI antibodies in only 17% (4/24) of these subjects with childhood-onset primary APS, in contrast to the 80% previously reported for adult-onset primary APS (5). This may reflect the influence of age or differences in clinical manifestations.

Most published series on the prevalence of aPL in pediatric SLE are small, ranging from 18–59 subjects (16, 19–22). Our finding of 53% positivity for aCL, 23% positivity for LAC, and 67% positivity for aPL in general in childhood SLE is similar to previous reports (16, 19, 20, 22, 23). No published series to date has reported the prevalence of anti-β2GPI in pediatric SLE, however our finding of anti-β2GPI in 48% of pediatric SLE subjects is consistent with the observed prevalence of anti-β2GPI in adult SLE cohorts, which ranges from 4%–89% (6, 7, 24) depending on whether or not the patients met criteria for APS. As expected, many patient sera were positive by only 1 or a limited number of aPL assays. Our findings confirm earlier observations that if any 1 of the standard assays are not performed, a subset of aPL may be missed. This is most notable for aCL detection, which we conclude should be included in all evaluations for aPL; and least notable for anti-β2GPI detection.

Earlier reports have suggested a greater specificity of anti-β2GPI for predicting APS than other aPL assays (24), however in this cohort, anti-β2GPI were not seen as frequently as aCL or LAC and demonstrated weaker correlation with APS than other aPL assays. Published reports suggest that LAC is a better predictor of thrombosis than other aPL assays (25, 26); however, our findings did not support this observation. Although screening for anti-β2GPI was determined to be insensitive for prediction of APS risk in general, there was a strong association between the presence of anti-β2GPI and occurrence of stroke. Larger studies are needed to confirm this association and to evaluate a potential role for anti-β2GPI in the pathogenesis of CNS thrombosis.

Separate analysis of SLE subjects did demonstrate a significant association of aCL and LAC with APS. However, in contrast with Cabiedes et al (24), who demonstrated an association between anti-β2 GPI and APS in adult SLE, we found no significant difference in prevalence of anti-β2GPI among these groups in children (45% in SLE with no APS versus 62% in SLE with APS). One of the most interesting associations identified in this study was the strong association of anti-β2GPI with SLE, and its ability to discriminate between primary and secondary APS in pediatric subjects with chronic thrombocytopenia. Chronic ITP, a common pediatric condition, is a reason for concern because of its potential evolution to SLE. Our observations suggest that there may be a role for anti-β2GPI detection in identification of the subset of children with ITP who are at risk for future development of SLE. This needs to be confirmed in a prospective study. If confirmed, the presence of anti-β2GPI may serve as a useful prognostic marker and indicator for more aggressive immunosuppressive therapy.

One limitation of this study is potential referral bias. Although subjects were not recruited systematically, referral patterns in northern California, combined with efforts to recruit from the only other 2 pediatric rheumatology centers in the area, support our belief that a broad spectrum of subjects with rheumatic diseases has been included in this cohort. This study was neither designed to address the correlation of aPL levels with disease activity nor response to therapy, and prospective longitudinal studies are needed to answer those questions. Finally, background positivity for anti-β2GPI in healthy children was not determined, but is assumed to be quite low, as previously reported for healthy adults (7).

In summary, this study extends the results of previous reports in several ways. First, in addition to SLE, this cohort includes children with either SLE-like or primary APS and therefore adds significantly to the limited literature available on primary APS in children and adolescents. Second, this is the first report on the prevalence of anti-β2GPI in a pediatric cohort and suggests that β2GPI screening is less sensitive than aCL or LAC assays for detection of APS. Furthermore, among subjects with anti-β2GPI, neither isotype nor antibody level was clinically significant. Third, our data suggests a role for β2GPI detection in predicting evolution to SLE in children with chronic ITP, a common pediatric condition.


  1. Top of page
  2. Abstract
  7. Acknowledgements

The authors thank Elena Bleumers for technical assistance and Dr. Diane Wara for critical review of the manuscript.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • 1
    Nimpf J, Bevers EM, Bomans PH, Till U, Wurm H, Kostner GM, et al. Prothrombinase activity of human platelets is inhibited by beta 2-glycoprotein-I. Biochim Biophys Acta 1986; 884: 1429.
  • 2
    Schousboe I. beta 2-Glycoprotein I: a plasma inhibitor of the contact activation of the intrinsic blood coagulation pathway. Blood 1985; 66: 108691.
  • 3
    Wilson WA, Gharavi AE, Koike T, Lockshin MD, Branch DW, Piette JC, et al. International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum 1999; 42: 130911.
  • 4
    Roubey RA. Immunology of the antiphospholipid antibody syndrome. Arthritis Rheum 1996; 39: 144454.
  • 5
    Cabral AR, Cabiedes J, Alarcon-Segovia D. Antibodies to phospholipid-free beta 2-glycoprotein-I in patients with primary antiphospholipid syndrome. J Rheumatol 1995; 22: 18948.
  • 6
    Sanfilippo SS, Khamashta MA, Atsumi T, Amengual O, Bertoaccini ML, D'Cruz D, et al. Antibodies to beta 2-glycoprotein I: a potential marker for clinical features of antiphospholipid antibody syndrome in patients with systemic lupus erythematosus. J Rheumatol 1998; 25: 21314.
  • 7
    Viard JP, Amoura Z, Bach JF. Association of anti-beta 2 glycoprotein I antibodies with lupus-type circulating anticoagulant and thrombosis in systemic lupus erythematosus [see comments]. Am J Med 1992; 93: 1816.
  • 8
    Brandt JT, Triplett DA, Alving B, Scharrer I. Criteria for the diagnosis of lupus anticoagulants: an update. Thromb Haemost 1995; 74: 11851190.
  • 9
    Alarcon-Segovia D, Perez-Vazquez ME, Villa AR, Drenkard C, Cabiedes J. Preliminary classification criteria for the antiphospholipid syndrome within systemic lupus erythematosus. Semin Arthritis Rheum 1992; 21: 27586.
  • 10
    Alarcon-Segovia D, Cabral AR. The concept and classification of antiphospholipid/cofactor syndromes. Lupus 1996; 5: 3647.
  • 11
    Piette JC, Wechsler B, Frances C, Papo T, Godeau P. Exclusion criteria for primary antiphospholipid syndrome [letter]. J Rheumatol 1993; 20: 18024.
  • 12
    Piette JC. Towards improved criteria for the antiphospholipid syndrome. Lupus 1998; 7: S14957.
  • 13
    Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982; 25: 12717.
  • 14
    Agresi A. An introduction to categorical data analysis. New York: Wiley & Sons; 1996.
  • 15
    Harris EN. A reassessment of the antiphospholipid syndrome. J Rheumatol 1990; 17: 7335.
  • 16
    Seaman DE, Londino AV Jr, Kwoh CK, Medsger TA Jr, Manzi S. Antiphospholipid antibodies in pediatric systemic lupus erythematosus. Pediatrics 1995; 96: 10405.
  • 17
    Von Scheven E, Athreya BH, Rose CD, Goldsmith DP, Morton L. Clinical characteristics of antiphospholipid antibody syndrome in children. J Pediatr 1996; 129: 33945.
  • 18
    Cuadrado MJ, Mujic F, Munoz E, Khamashta MA, Hughes GR. Thrombocytopenia in the antiphospholipid syndrome. Ann Rheum Dis 1997; 56: 1946.
  • 19
    Gattorno M, Buoncompagni A, Molinari AC, Barbano GC, Moreeale G, Stalla F, et al. Antiphospholipid antibodies in paediatric systemic lupus erythematosus, juvenile chronic arthritis and overlap syndromes: SLE patients with both lupus anticoagulant and high-titre anticardiolipin antibodies are at risk for clinical manifestations related to the antiphospholipid syndrome. Br J Rheumatol 1995; 34: 87381.
  • 20
    Ravelli A, Caporali R, Di Fuccia G, Zonta L, Montecucco C, Martini A. Anticardiolipin antibodies in pediatric systemic lupus erythematosus. Arch Pediatr Adolesc Med 1994; 148: 398402.
  • 21
    Shergy WJ, Kredich DW, Pisetsky DS. The relationship of anticardiolipin antibodies to disease manifestations in pediatric systemic lupus erythematosus. J Rheumatol 1988; 15: 138994.
  • 22
    Berube C, Mitchell L, Silverman E, David M, Saint Cyr C, Laxer R, et al. The relationship of antiphospholipid antibodies to thromboembolic events in pediatric patients with systemic lupus erythematosus: a cross-sectional study. Pediatr Res 1998; 44: 3516.
  • 23
    Love PE, Santoro SA. Antiphospholipid antibodies: anticardiolipin and the lupus anticoagulant in systemic lupus erythematosus (SLE) and in non-SLE disorders: prevalence and clinical significance. Ann Intern Med 1990; 112: 68298.
  • 24
    Cabiedes J, Cabral AR, Alarcon-Segovia D. Clinical manifestations of the antiphospholipid syndrome in patients with systemic lupus erythematosus associate more strongly with anti-beta 2-glycoprotein-I than with antiphospholipid antibodies. J Rheumatol 1995; 22: 1899906.
  • 25
    Mackworth-Young C. Antiphospholipid antibodies: more than just a disease marker? Immunol Today 1990; 11: 605.
  • 26
    Horbach DA, van Oort E, Donders RC, Derksen RH, de Groot PG. Lupus anticoagulant is the strongest risk factor for both venous and arterial thrombosis in patients with systemic lupus erythematosus: comparison between different assays for the detection of antiphospholipid antibodies. Thromb Haemost 1996; 76: 91624.