Ischemic Perinatal Stroke: Challenge and Opportunities

Terminology, definition and classification

There is no unanimity about the terminology, definition, or classification of perinatal strokes. Many pathological conditions can lead to occlusions of cerebral arterial or venous structures causing focal disruption of blood supply and stroke in the perinatal period. An expert panel convened by the National Institutes of Health in 2006 focused attention on the ischemic variety of strokes in late preterm (previously called ‘near-term’) and term infants, and proposed a definition and a classification based on the following considerations (1).

(1) In the perinatal period, the predominant pathology leading to stroke is one of ischemia, most often from arterial or venous thrombosis.

(2) The timing of the vascular event leading to a stroke is almost never known. Therefore, strokes are best classified based on the age at establishing the diagnosis, without implying the gestational or postnatal age at which the vascular event took place.

Ischemic perinatal stroke (IPS) was defined as: ‘Ischemic perinatal stroke is 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 28th postnatal day, and confirmed by neuroimaging or neuropathological studies.’

Three subgroups were proposed based on the age at diagnosis of stroke (1).

1. Fetal ischemic stroke (FIS): diagnosed before birth using fetal imaging methods, or in stillbirths, based on neuropathologic examination.

2. Neonatal ischemic stroke (NIS): diagnosed after birth, on or before the 28^th postnatal day (including in preterm infants).

3. Presumed perinatal ischemic stroke (PPIS): diagnosed after the 28th day of age, in whom it is presumed (but not certain) that the ischemic event occurred sometime between 20th week of fetal life through the 28th postnatal day.

Ischemic perinatal stroke in preterm infants

Ischemic strokes are rare in preterm infants (4, 5). Common cerebrovascular disorders in preterm infants are, intra and peri-ventricular hemorrhage, parenchymal hemorrhage, and white matter injury from vascular or non-vascular causes. In one case-controlled study that covered a 10-year span, the investigators identified 31 cases of IPS (5). The majority of strokes (25/31, or 81%) involved the middle cerebral artery distribution. Involvement of one or more lenticulostriate branches was most common among infants with a gestational age of 28–2 weeks, but main branch involvement was seen only in those with a gestational age of >32 weeks. Risk factors predisposing to strokes in preterm infants include fetal twin-to-twin transfusion syndrome, abnormal fetal heart patterns, and neonatal hypoglycemia.

Risk factors

Since the causal pathways and the nature of their interactions leading to stroke are poorly understood, the reported risk factors must be considered as associations, and not necessarily causative factors. Thus, in individual IPS cases, one can only speculate about their causal relationship (1, 6–10).

All pregnancies are considered natural pro-thrombotic states, because during pregnancy the physiologic processes lead to a hypercoagulability state of the blood. Up to 67/100 000 pregnant women develop cerebral infractions (11). Increased proclivity for clot formation during pregnancy is secondary to accelerated platelet-to-vessel interactions related to accelerated thrombin generation, and decreased thrombolysis (7, 7, 8).

Many inherited and acquired thrombophilias in the mother and/or the fetus can cause enhanced clot formation in the maternal or fetal vasculature, or in the utero-placental unit; the topic has been extensively reviewed (12–16). However, a majority of pregnant women with a history of thrombophilia remain healthy, with a very low incidence of stroke in their offspring.

The ‘pro-inflammatory’ status of pregnancy also increases the potential for prothrombotic interactions between inflammatory and coagulation pathways (11). Superimposed infections (6, 14) can exacerbate these tendencies and promote thrombotic episodes in either the maternal or fetal circulations when there is with a concomitant maternal or fetal thrombophilia (16). Endothelial disruption associated with preeclampsia can also function as a pro-thrombotic stimulus.

The developing fetus maintains a delicate equilibrium between bleeding and thrombosis. This balance can be shifted toward excessive thrombotic or bleeding tendency, depending upon the segment of the coagulation cascade that is perturbed. Ischemic perinatal stroke has also been reported with fetal polycythemia, twin-to-twin transfusion syndrome in preterm infants, prolonged maintenance of systemic or umbilical vascular catheters (5), persistent pulmonary hypertension, congenital heart disease with right-to-left shunting; arterio-venous malformations; dehydration, and meningitis (2–6, 17–21). Male sex, non-Hispanic black ethnicity, and maternal autoimmune conditions are other non-specific risk factors for IPS (2–4).

Thrombotic episodes on the fetal side of the placenta can potentially lead to embolic phenomenon in the fetal brain. Some investigators propose that patients with symptomatic IPS should be screened for inherited prothrombotic defects. However, the costs versus benefits of such evaluations are unclear.

Diagnostic issues

Fetal strokes are often diagnosed serendipitously during routine prenatal ultrasound examination (22). They may also be discovered during neuropathology examination of stillborn infants.

Newborn with IPS can present with focal or generalized seizures, apnea, hypotonia, or episodes of duskiness, irritability, and poor feeding. Ischemic perinatal stroke is more frequently seen in males, and in black infants compared with white infants (2). Because the MCA is the most common vessel involved, right congenital hemiplegia is seen more often than the left. Infants with IPS may remain asymptomatic and be discharged as ‘healthy.’ Such infants may manifest later in infancy or childhood with unilateral weakness, seizures, or developmental delays (8).

Neuroimaging studies are the most important tools for the confirmatory diagnosis of IPS (23–25). Although cranial ultrasound and computed tomography (CT) scans can help detect a neonatal IPS in the first few days of life, it is possible to miss anterior or posterior lesions. Conventional T2-weighted magnetic resonance imaging (MRI), magnetic resonance angiography (MRA), and diffusion-weighted imaging (DWI) remain the principal methods for establishing the diagnosis of IPS. Ischemic tissue in the newborn brain can be detected using DWI with high sensitivity during the first 2 days after the appearance of the symptoms. Diffusion-weighted imaging sensitivity declines after 5 days from appearance of symptoms. However, as the lesions become less visible on DWI, they become more visible on the conventional T1- and T2-weighted MRI. Magnetic resonance imaging features enable prognostication, especially of motor outcomes.

Vascular ischemic lesions in the middle cerebral artery distribution, affecting tissue in the hemisphere, the posterior limb of the internal capsule, and the basal ganglia (lenticulostriate vessels) tend to cause hemiplegia, irrespective of the size of the infarct (23). Involvement of the cerebral peduncles seen on early DWI is also associated with the development of hemiplegia.

Ancillary tests

There is no consensus about how many (or which) tests are to be carried out to find various thrombotic conditions in an infant with IPS, or in the mother or the family of an affected infant. In all cases of IPS, the clinician needs to obtain a detailed history of past pregnancy outcomes, and family history of thromboembolic conditions. A routine screening for thrombotic conditions is not recommended for all pregnant women; however, some experts recommend screening of women at risk, such as those with a past pregnancy loss, another affected child with IPS, preeclampsia, or prevalence of thrombotic disorders in the pedigree (12, 13, 15, 21).

An evaluation for the underlying thrombotic conditions in the IPS infant may be valuable, especially because this can help to ‘rule-out’ inherited or acquired thrombotic conditions. One may consider screening for battery of conditions, depending upon ones access to resourceful laboratories. Some suggested tests (13–16, 21) include: activated protein C activity; free and total protein S activity; antithrombin III activity; lipoprotein (a); total plasma homocysteine; factor V Leiden, prothrombin 20210; lupus anticoagulant; antiphospholipid antibodies, fibrinogen; plasminogen; and factor VIIIC; β glycoprotein antibodies; and two mutations of methylene tetrahydrofolate reductase (MTHFR C677T and A1298C).

One study employing these screening procedures discovered that prothrombotic risks were present in 28 of 51 (55%) mothers, and in 30 of 60 (50%) children (21). It is important to appreciate that while these tests are helpful in family counseling, there are no clear implications for recurrence, which is considered to be < 5% (12, 21).

Besides the above tests, clinical examination and appropriate scanning tests need to be considered to detect congenital heart malformations and to assess vascular anatomy, especially in the neck, and an evaluation of thrombotic lesions in systemic vessels. Examination of the placenta when available would be of great value to uncover chorioamnionitis and placental vacular pathologies. Similarly, autopsy examination can be extremely useful in defining the timing and the evolution of the vascular event leading to stroke.

Therapy and outcome

At present only symptomatic treatment of infants with IPS is recommended, and the roles of thrombolytic therapies are yet to be tested fully in infants. While mortality from IPS is rare, neurological deficits or epilepsy occurs in 50–75% of survivors (26–29). Hemiplegia or hemiparesis, with sensorimotor impairments are the commonest deficits. Deficits in language, vision, cognition, and behavior occur in 20–60%. In most infants focal neurological deficits emerge after early infancy and new deficits can evolve over several years. More than 80% of PPIS infants have hemiparesis. Thus, long-term follow-up with standardized measures are required in all IPS and PPIS cases.

Conclusions

Over the past decade, IPS has emerged as an important cause of brain injury in the perinatal period. It has become a major cause of cerebral palsy (CP). The timing of the vascular episode leading to IPS is almost never known; thus scientists and clinicians ought to consider designating the subcategory of IPS based only on the age at establishing the diagnosis, and not attempt to imply that a particular vascular event occurred at a certain gestational or postnatal age. Similarly, although numerous risk factors have been associated in cases of IPS, much remains to be learned about the interactions among the risk factors and their causal pathways leading to vascular occlusions. Thus, in individual cases of IPS, it is nearly impossible to establish the cause and effect relationships between associated risk factors and the vascular pathology leading to stroke.

Radiographic investigations discussed above are the gold standard to make a definitive diagnosis of neonatal or presumed perinatal stroke. In addition, tests to uncover rare thrombotic conditions, also as discussed above, can be helpful. The yield from such studies is small; but the findings can aid in counseling the family about risks for recurrence, and for strokes in other siblings. Evidence-based information is lacking at present to make stronger recommendations for extensive screening of parents and other family members of affected patient to detect inherited thrombotic disorders.

Since IPS is a complex disorder spanning many specialties, healthcare workers in diverse disciplines need to be trained to recognize this disorder and coordinate their efforts in evaluating and treating such infants, and planning for a comprehensive follow-up. Much research is needed to develop uniform standards and criteria for IPS diagnosis using neuroimaging methods, and to further improve the definition and classification.