The antiphospholipid syndrome (APS) is an autoimmune multisystem disease characterized by thromboembolic events, pregnancy morbidity, and hematological, dermatological, neurological, and other manifestations in the presence of elevated titers of antiphospholipid antibodies (aPL) [1]. Antiphospholipid syndrome may occur as an isolated clinical entity (primary APS) or in association with other diseases, mainly systemic lupus erythematosus (SLE) . It occasionally occurs with other autoimmune conditions, infections, and malignancies. Pediatric APS due to de novo production of aPL occurs anytime from the neonatal period through childhood and adolescence. Neonatal APS, manifesting with thrombotic events, is a rare complication; it may be associated with maternal aPL, in which case it is attributed to transplacental passage of aPL, or it may be found in newborns whose mothers do not have aPL. The rapid development of multiple organ thromboses and microthromboses, known as catastrophic APS (CAPS) , has been reported in children [2].

In recognition of the unique issues surrounding the diagnosis and treatment of pediatric APS patients, a pediatric APS task force was developed in preparation for the 15th International Congress on aPL (www.​apsistanbul2016.​org) (North Cyprus, September 2016). This multidisciplinary group consists of pediatric rheumatologists, pediatric neurologists and neurodevelopment specialists, pediatric hematology-coagulation specialists, and neonatal-perinatal medicine specialists. The objectives of this task force were to review the current knowledge on the pathogenesis, clinical and laboratory features, diagnosis, classification, and treatment of the pediatric APS to generate recommendations for future research and to suggest modifications to the APS classification criteria for enhanced applicability to children and neonates. Members of this task force have published two extensive reviews of the literature [3, 4].

Hemostasis, the normal reparative process for damaged vasculature, is a three-stage process (vessel damage and platelet response, thrombin generation, and fibrinolysis) with the following participants: cells or cellular remnants (platelets, monocytes, endothelial cells, and microparticles), proteins (hemostatic and fibrinolytic), and vascular endothelium, together known as the cell-based model of hemostasis [5–7].

Inhibitory mechanisms that modulate hemostasis include blood proteins (proteins C and S, antithrombin [AT], α₂-macroglobulin, and tissue factor pathway inhibitor [TFPI]); vessel wall proteins (thrombomodulin, endothelial protein C receptor, and TFPI) or chemicals (nitrous oxide [NO], CD39 [ectonucleotidase]); prostacyclin; and proteoglycan-related glycosaminoglycan molecules (chondroitin, dermatan sulfate, or heparin sulfate) [8, 9].

Children differ from adults in the participants and the inhibitory mechanisms of hemostasis, thereby influencing the epidemiology, clinical manifestations, and management of thrombosis.

Assessment of Virchow’s triad of stasis, vascular injury, and hypercoagulability is inherent in the assignment of thrombotic risk. The antithrombotic milieu in a child slowly acquires characteristics of an adult with the onset of puberty. Compared to adults, children are relatively protected from thrombosis because of the decreased potential for thrombin generation, increased α₂-macroglobulin, and the antithrombotic potential of the vascular wall. Children may acquire additional risk factors in adolescence, including smoking, birth control pills, and pregnancy. Preexisting risk factors such as obesity and dyslipoproteinemia, which may have been present earlier, have an amplified effect. As the vascular endothelium decreases in antithrombotic potential, vascular injury from play activity and sports may impose additional risk.

Antiphospholipid antibodies coincident with inherited thrombophilia (for isolated or combinations of genetic mutations) are associated with increased relative risk of thrombosis. The increased risk of thrombosis associated with aPL and factor V Leiden mutations in adults and children has been discussed elsewhere [55–58].

Antiphospholipid syndrome plays an important role in pediatric venous and arterial thromboembolic events, including stroke [59, 60], but is rare among all patients diagnosed as having APS: onset prior to the age of 15 occurred in only 2.8% of all patients with APS [61]. Pediatric and neonatal APS have been addressed by few studies [62–64].

Although the criteria of obstetric morbidity are not applicable to prepubertal children, and despite lack of validation in children and neonates, the same diagnostic criteria are used for classification of pediatric APS [65]. Nonetheless, pediatric APS presents with symptoms similar to adults [66]. Katzav et al. [67] showed that patients with childhood-onset APS presented with more episodes of chorea and jugular vein thrombosis than did adults [67]. Arterial thrombosis is more common in children [63, 68]. In a cohort of 28 children with APS [69], the most common initial manifestations were venous thrombosis, stroke, and thrombocytopenia. Lupus anticoagulant was detected in 96% of those tested. Seven of the 24 patients with vascular thrombotic events had recurrences. Hereditary thrombophilia was more common in children who experienced a single episode of APS (8 [53.3%] of 15 patients) than in those who experienced recurrences (2 [28.6%] of 7 patients). However, only two patients in the latter group (28.6%) received anticoagulants after the first manifestation, compared with 12 (70.6%) of the 17 patients without recurrences. Infants with perinatal stroke usually display a monophasic disease, and other manifestations of APS do not develop later.

Based on very small studies, levels of IgA anticardiolipin (aCL) antibody are lower in children with APS than in adults, whereas IgG anti-β₂-glycoprotein-I (aβ₂GPI) levels are highest in preschool children [69, 70].

An international registry of pediatric APS , established in 2004 [63], currently documents standardized clinical, laboratory, and therapeutic data of 140 children with definite APS seen at 24 pediatric university centers. Thrombotic events include venous thrombosis in 60%, arterial thrombosis in 32%, small vessel thrombosis in 6%, and mixed arterial and venous thrombosis in 2% of the first 121 patients [63]. During a follow-up period of 6.1 years, 19% developed recurrent thrombosis [63] and 5% developed catastrophic APS.

Perinatal stroke is a special entity in children with APS [71, 72]. Perinatal arterial stroke (PAS) may result from focal arterial or venous thrombosis or emboli occurring between 20 weeks of fetal life and the 28^th postnatal day. Diagnosis may be delayed; reported prevalence is 1:4000–5000 live births [72]. Risk factors may relate to both fetal and neonatal disorders as well as maternal and placental conditions. The relative roles of genetic and acquired thrombophilia in the pathogenesis of PAS are controversial. Positive aPL tests can be found in both neonates with PAS or their mothers, or both [60]. Berkun et al. [72] presented a cohort of 12 neonates with cerebral thromboembolism and persistent aPL whom they followed prospectively with repeated antibody testing. In 10/12 cases aPL levels decreased to normal within a median of 2.5 years. Notably, following the diagnosis of perinatal cerebral event, 11 of the patients received no anticoagulant therapy; a single infant who suffered from perinatal cerebral sinus venous thrombosis was treated with low-molecular-weight heparin (LMWH) until the age of 6 months. In this cohort, maternal aPL was detected in only two cases, not necessarily correlating with the antibodies found among the affected infants. Similarly, within an Israeli PAS cohort of eight mother-infant pairs [60], aPL discordance was noted between mothers and their offspring: in three cases, only the mother had aPL; in four, only the infant; and in one, both mother and infant. These findings indicate that maternal antibodies are not the only pathogenic factor for neonatal APS. An important finding was the absence of recurrent thrombosis or other APS manifestations, despite lack of prolonged anticoagulation; also, there was gradual disappearance of aPL. Timing of perinatal brain injury introduces multiple, often competing, factors in brain maturation and development that will ultimately determine outcome.

Epidemiologic pediatric studies show that, from 1 year to puberty, pediatric patients are less prone to thrombosis than are adults. Low titer and transient aPL [72] are frequent within the pediatric population, but thrombosis is rare. Overall, perinatal stroke in children with aPL may not require anticoagulant therapy unless other risk factors prevail.

A few registry-based studies suggest that neurological manifestations occur in 16–22% of pediatric APS cases [63, 73]. They include thrombosis (arterial and venous); immune-mediated inflammatory change; and secondary complications or symptoms, for instance, seizures, headaches, focal neurological deficits/signs, and cognitive changes. Psychiatric symptoms, including psychosis and mood disturbance, have also been described [63, 74].

Regarding lupus-associated APS, the most common neurological manifestation is thrombosis, including both arterial stroke and venous sinus thrombosis [74]. The 121-patient pediatric APS registry found a 16% prevalence of headache, 9% of cognitive dysfunction, and 10% of psychosis [75]. Comparison of lupus patients with and without aPL suggests an association between aβ₂GPI and neuropsychiatric manifestations (p = 0.02). Among 32 patients with both primary and other autoimmune disease-associated aPL, and that defined neurological manifestations only as epilepsy, chorea, or aseptic meningitis, seizures were the most common neurological manifestation, occurring in five (16%) [73].

In studies documenting the rate of aPL positivity in neurologic cohorts, there is no increase of aPL in pediatric headache patients compared to controls [76]. Approximately 30% of 142 consecutive pediatric epilepsy patients in one cross-sectional cohort study were positive for aPL, but selection bias and lack of controls in this study limit generalizability of the findings [77]. In another study of 80 pediatric epilepsy patients, only three (3.8%) were positive for aPL, with aCL found in one (1.3%) [78].

In pediatric APS, neurological complications are common and may be significant. Future studies should provide detailed documentation of functional and structural neurological changes , including detailed cognitive and neurological assessments and MRI studies using advanced metrics.

Assessment of the risk posed by acquired or extrinsic factors is essential to treatment decisions [79, 80]. The treatment of “at-risk” children may not always include anticoagulants but may include dietary changes, behavior modification, and/or treatment of associated conditions (none of which, however, are evidence-based). Endothelial activation and endothelial damage associated with inflammation may shift the balance toward thrombosis. Thus, treatment of an underlying disease may be of benefit. In an otherwise-healthy, low-risk child with a postinfectious event, this prothrombotic situation may be transient; however, in children with underlying rheumatic diseases, anticoagulation therapy may need to be continued until disease activity is quiescent. Deitcher and Carman [81], studying adults, suggested that “lifelong” and “long-term” treatment are different, the latter implying that the issue will be revisited. Risk stratification and individualized therapy determine how long a patient is treated.

In children, identification of risk factors and provision of preventive health, and short-term treatment of transient events, may be more important than in adults because of developmental hemostasis [82].

A recent European project, SHARE (Single Hub and Access Point for Pediatric Rheumatology in Europe), systematically evaluated published data on APS in pediatric populations to provide 14 evidence-based recommendations for the diagnosis and treatment. When available, these recommendations will help practicing physicians and facilitate improvement and uniformity of care of children with APS [83].