KEY POINTS
- 1.
Perinatal arterial ischemic stroke most often results from the convergence of multiple stroke risk factors specific to the perinatal period and has a low risk of recurrence.
- 2.
Neonatal arterial ischemic stroke refers to the most common presentation of perinatal stroke when focal seizures or diffuse neurologic signs lead to diagnosis soon after birth. A subset of patients remain undiagnosed in the neonatal period and present as having presumed perinatal arterial ischemic stroke, with emerging neurologic deficits later in infancy.
- 3.
Cardiac disease, although present in a minority of cases, is a risk factor for stroke recurrence and should be investigated with echocardiography after perinatal stroke diagnosis.
- 4.
In the absence of multiple thromboembolic events and/or a family history of thrombosis, extensive thrombophilia evaluations are generally not recommended because they do not predict recurrence or alter management.
- 5.
Management of perinatal stroke should focus on supportive care with the primary goal of promoting adequate perfusion to the brain and minimizing extension of the injury. Although the presentations of perinatal stroke and hypoxic-ischemic encephalopathy may overlap, treatment with hypothermia should not be delayed if hypoxic-ischemic encephalopathy is suspected.
Introduction
The perinatal period represents one of the highest-risk times of life for stroke, and the consequences are significant in terms of neurologic morbidity across the lifespan. In 2007 an international workshop convened by the National Institutes of Neurological Disorders and Stroke defined ischemic perinatal stroke as a “group of heterogeneous conditions in which there is focal disruption of cerebral blood flow secondary to arterial or cerebral venous thrombosis or embolization, between 20 weeks of fetal life through the 28th postnatal day, confirmed by neuroimaging or neuropathologic studies.” Perinatal stroke can be divided into distinct syndromes based on the pathophysiology, presentation, and developmental stage at which it occurs. Neonatal arterial ischemic stroke (NAIS), cerebral sinovenous thrombosis (CSVT), and neonatal hemorrhagic stroke typically present acutely in the neonatal period, whereas presumed perinatal arterial ischemic stroke (PPAIS), periventricular venous infarction, and presumed perinatal hemorrhagic stroke occur pre- or perinatally but frequently do not manifest until later in infancy ( Fig. 50.1 ).
Epidemiology
Ischemic perinatal stroke occurs in 1 in 2300 to 5000 live births, with most presenting within the first week of life, significantly higher than the weekly risk of stroke in an adult with known risk factors. Arterial strokes represent the majority of these, with most presenting acutely in the neonatal period (NAIS). The incidence of CSVT is less common, occurring in 2.6 per 100,000 births. , A male predominance has been demonstrated in multiple studies of perinatal stroke. , Hemorrhagic stroke is significantly less common in the perinatal population and is often excluded from studies of perinatal stroke, but one study identified a population prevalence (excluding isolated germinal matrix hemorrhage) of 6.2 in 100,000 live births over a 10-year period, also with a male predominance.
Pathogenesis
The cause of perinatal stroke is thought to be multifactorial, with convergence of multiple risk factors specific to the peripartum period creating conditions ripe for thrombosis and clot embolization to the cerebral arteries. First, pregnancy is a hypercoagulable state. Hemostatic changes occur throughout pregnancy, with increases in procoagulant factors, decreases in anticoagulant factors, and decreases in intrinsic fibrinolytic activity. These changes are essential for maintaining placental function during pregnancy and defending against hemorrhage during delivery, while predisposing to thrombosis. Thrombosis of the placental vessels that normally occurs as the placenta separates at the time of birth may lead to direct embolization into the fetal circulation. Second, the fetal circulation allows for direct embolization of a venous clot across right-to-left cardiac shunts (such as patent foramen ovale) to the aorta and directly to the cerebral circulation. The rapid and complex transition from fetal to postnatal circulation results in significant alterations in blood flow that may contribute to the pathogenesis of perinatal stroke. Third, in addition to thrombosis at the level of the placenta, the placenta may play a larger role in neonatal stroke through cytokine release, thereby creating an inflammatory, procoagulant environment that promotes systemic thrombus formation.
Multiple additional risk factors specific to the pregnancy and the peripartum period have been associated with perinatal stroke, without a clear causative role being established. , , The most consistently demonstrated risk factors include chorioamnionitis, maternal fever, and neonatal sepsis or meningitis, highlighting an important role of inflammation in perinatal stroke. Male sex has also been a consistently demonstrated risk factor for perinatal stroke, although the reason for this association is unknown. , , Numerous other risk factors have been proposed but without being consistently demonstrated. Proposed prepartum risk factors include primiparity, prior fetal loss, intrauterine growth restriction/small for gestational age, oligohydramnios, gestational hypertension or preeclampsia, and maternal diabetes, among others. , , , Proposed intrapartum risk factors include prolonged rupture of membranes, prolonged second stage of labor, complicated delivery, abnormal fetal heart tracing, birth asphyxia, fetal hypoglycemia, and a low 5-minute Apgar score (less than 7), though whether these are the cause or the result of the perinatal brain injury is unclear. , , , The risk of perinatal stroke has been shown to increase significantly when multiple risk factors are present.
The idea that multiple risk factors confined to the perinatal period converge to create a high-risk time for perinatal stroke is consistent with the low recurrence rate of perinatal stroke. However, there are some risk factors that confer ongoing stroke risk to the child, and their identification is essential to proper management. Hypercoagulable disorders have been thoroughly investigated in studies of perinatal stroke, and the results are inconsistent. Early studies examining the role of prothrombotic risk factors in perinatal stroke demonstrated high rates of prothrombotic abnormalities in infants with perinatal stroke. A large case-control study of infants with perinatal stroke showed that 68% of neonates with stroke had at least one abnormality on thrombophilia testing compared with 24% of age- and sex-matched healthy controls. In two studies of mother-child pairs, more than 50% of both the infants with perinatal stroke and their mothers had prothrombotic abnormalities on testing in the neonatal period. , These abnormalities included low protein C or S, elevated lipoprotein(a), elevated homocysteine, methylene tetrahydrofolate reductase mutations, factor V Leiden, prothrombin G20210A mutations, or antiphospholipid antibodies. A prospective study of children with perinatal stroke showed that although the risk of a recurrent symptomatic thromboembolic event was low, about 3%, the risk of a second event was higher if prothrombotic risk factors were identified. A meta-analysis including six studies that exclusively evaluated perinatal stroke likewise concluded that thrombophilia was a risk factor for neonatal arterial ischemic stroke, but the authors noted the paucity of eligible studies with control subjects, which limited the conclusions that could be drawn. Despite these findings, more recent, larger studies have suggested that there is only a minimal association between perinatal stroke and thrombophilia. , Importantly, abnormal testing in these studies did not predict stroke recurrence, leading the authors to conclude that hypercoagulability evaluations should not be routinely completed in infants with perinatal stroke. Even if disordered coagulation is identified in the perinatal period, it is often transient, confined to the perinatal period or related to the thrombus itself. Regardless, multiple thromboembolic events and/or a family history of thrombosis should prompt consideration of a thorough thrombophilia evaluation.
In contrast, cardiac disease, although present in a minority of cases, is a risk factor with a clearer causal link to stroke and is associated with a substantially increased recurrence risk. Imaging studies of children with congenital heart disease have found that 10% to 30% of infants can have stroke, with at least half occurring preoperatively. , In another study of children with congenital heart disease, the stroke recurrence rate was 14% after a neonatal sentinel stroke. In addition, compared with older children, neonates are at higher risk for stroke associated with extracorporeal membrane oxygenation.
Although arteriopathy is the most common risk factor for childhood arterial ischemic stroke, its role in perinatal stroke is minimal. Rare cases of arteriopathic stroke in the perinatal period have been published, with etiologies felt to be congenital or traumatic. Incidence of arteriopathy in association with perinatal stroke may be underestimated due to inconsistency in the use of vascular imaging.
Presentation
Perinatal stroke can present at different times from fetal to postnatal life, with distinct clinical presentations depending on when the stroke is discovered. Therefore perinatal stroke can be subclassified into different types, including fetal, neonatal, and presumed perinatal arterial ischemic stroke.
Fetal strokes are typically hemorrhagic, although arterial ischemic stroke is increasingly being diagnosed with advancement of imaging including high-quality fetal ultrasound and fetal magnetic resonance imaging. In these cases, pregnancy often proceeds normally and the child’s parents may present before the child is born for counseling on prognosis.
Neonatal stroke is the subclassification of perinatal stroke that applies to children who present in the in the first 28 days of life. Neonatal stroke typically presents with acute symptomatic seizures or diffuse neurologic signs on the first days of life ( Table 50.1 ). Seizures are focal and often refractory to treatment in the first few days of life, but they typically spontaneously remit outside of the acute period. Diffuse neurologic signs most commonly include abnormal tone or decreased level of arousal. Respiratory distress and feeding difficulties can also be part of the presentation of perinatal stroke. Focal findings are uncommon in the neonatal period. Typically, these children will be discharged home appearing well, although seizures may recur and neurologic deficits may emerge over time.
| Presenting Signs | Percentage |
|---|---|
| Seizure | 72 |
| Level of consciousness | 39 |
| Tone | 38 |
| Focal neurologic signs | 30 |
| Respiratory difficulty | 26 |
| Feeding difficulty | 24 |
The diagnosis of PPAIS is given to children who present after the first 28 days of life with clinical or radiologic evidence of stroke that is presumed to have occurred in the perinatal period (between the 20th week of fetal life and the 28th day of postnatal life). Typically, pregnancy, delivery, and postnatal periods are medically unremarkable, and these children are often discharged from the nursery without suspected neurologic concerns. Often, PPAIS presents around 3 to 12 months of age with emerging focal deficits which may include early handedness, unilateral fisting, increased tone on one side, or visual field cut. Alternatively, children may present later in childhood with focal seizures, and they may or may not have subtle focal findings on a careful neurologic examination. These clinical findings often prompt neuroimaging that reveals evidence of a chronic infarction, which in the absence of a history of acute-onset neurologic deficits after the first month of age allows for a retrospective diagnosis of PPAIS. PPAIS is thought to occur by the same mechanisms as neonatal stroke, and why different children present in different ways is a question that deserves further investigation.
Evaluation and Management
Stroke should be on the differential diagnosis for any term infant presenting with focal seizures or encephalopathy soon after birth. An algorithm for the diagnosis and management of perinatal stroke is presented in Fig. 50.2 . There can be some overlap in the presentation of infants with hypoxic-ischemic encephalopathy (HIE) and stroke. One study comparing infants with these diagnoses found a lower frequency of sentinel events and acidosis in the perinatal stroke group compared with the HIE group. Nevertheless, HIE has a specific time-dependent treatment in hypothermia, and therefore, if HIE is suspected this should take precedence. It remains unclear whether hypothermia is beneficial or harmful for infants with stroke. Treatment with hypothermia should not be delayed when there is suspected HIE.
The diagnosis of perinatal stroke is made by neuroimaging. Magnetic resonance imaging (MRI) is the gold standard for evaluating infants with suspected stroke. Serial MRI can also provide important information about the evolution of the lesions ( Fig. 50.3 ). Although computed tomography is fast and widely available, if provides far less information than MRI and has the additional concern of radiation exposure. Diffusion weighted imaging will demonstrate infarcted brain tissue very early in the course. We recommend including sequences that are sensitive for blood products such as gradient echo or susceptibility weighted imaging. Vessel imaging with magnetic resonance (MR) angiography and venography can be included to identify arteriopathies or CSVT, which can be difficult to distinguish based on clinical presentation alone. All of these studies can be performed without the use of gadolinium, although MR venography will benefit from contrast administration.
