Acta Pædiatrica ISSN 0803-5253
REVIEW ARTICLE
Neonatal stroke: a review of the current evidence on epidemiology, pathogenesis, diagnostics and therapeutic options NE van der Aa, MJNL Benders, F Groenendaal, LS de Vries ([email protected]) Department of Neonatology, Wilhelmina Children’s Hospital, University Medical Center Utrecht, Utrecht, The Netherlands
Keywords Cerebral sinovenous thrombosis, MRI, Neonatal stroke Correspondence Prof. Dr. Linda S. de Vries, MD, PhD, Department of Neonatology, Wilhelmina Children’s Hospital, UMC Utrecht, KE 04.123.1, PO Box 85090, 3508 AB Utrecht, The Netherlands. Tel: +31-88-7554545 | Fax: +31-88-7555320 | Email: [email protected]
ABSTRACT Neonatal stroke, including perinatal arterial ischaemic stroke and cerebral sinovenous thrombosis, remains a serious problem in the neonate. This article reviews the current evidence on epidemiology, pathogenesis, diagnostics and therapeutic options. Conclusion: Although our understanding of the underlying mechanisms and possible risk factors has improved, little progress has been made towards therapeutic options. Considering the high incidence of neurological sequelae, the need for therapeutic options is high and should be the focus of future research.
Received 20 November 2013; revised 2 January 2014; accepted 10 January 2014. DOI:10.1111/apa.12555
Neonatal stroke can be defined as a cerebrovascular injury that occurs around birth. It can be focal or multifocal and may include both ischaemic and haemorrhagic injury. Neonatal stroke is most frequently referred to as perinatal cerebral injury of ischaemic origin. In an international workshop on this subject, held in 2006, ischaemic perinatal stroke was narrowed down to ‘a group of heterogeneous conditions in which there is focal disruption of cerebral blood flow secondary to arterial or cerebral venous thrombosis or embolisation, between 20 weeks of foetal life through to the 28th postnatal day, confirmed by neuroimaging or neuropathological studies (1). Two common subtypes are perinatal arterial ischaemic stroke (PAIS) and cerebral sinovenous thrombosis (CSVT). However, there is still an ongoing discussion about which other patterns of injury should be regarded as perinatal stroke. Periventricular haemorrhagic infarction (PVHI), which is commonly seen in preterm infants, is frequently regarded as being part of the stroke spectrum (2,3). Another one is haemorrhagic stroke. Finally, presumed perinatal stroke, encompassing both arterial ischaemic stroke and PVHI, is also considered part of the neonatal stroke spectrum, though by definition they are diagnosed beyond the 28th postnatal day. In this article, we review both PAIS and CSVT in terms of their aetiology, clinical presentation and the diagnostic and therapeutic options.
PERINATAL ARTERIAL ISCHAEMIC STROKE Perinatal arterial ischaemic stroke has an incidence of between 1/1600 and 1/5000 (4,5). The wide range of
356
reported incidences may be explained by differences in the definition and diagnosis of PAIS. Some only include PAIS, while others also include presumed perinatal stroke and haemorrhagic stroke. All studies are of retrospective design and therefore depend on the accuracy of the databases used. In view of the fact that magnetic resonance imaging (MRI) scans are not routinely obtained for all infants presenting to neonatal intensive care units with neonatal seizures, the true incidence is likely to be even higher. Perinatal arterial ischaemic stroke may affect both term and preterm born infants. It has been suggested that the incidence of PAIS is higher in preterm born infants, but this might also reflect the routine use of cranial ultrasound in preterm infants (6). Several studies have reported a male predominance of approximately 60%, and PAIS is known to involve the left hemisphere more often (7).
Key notes
Neonatal stroke, including perinatal arterial ischaemic stroke and cerebral sinovenous thrombosis, has a multifactorial aetiology and often results in neurological sequelae. Neonatal neuroimaging plays an important role in both diagnosis and prognosis and may predict motor outcome. Although a lot of progress has been made in identifying possible underlying mechanisms, further research is required with regard to therapeutic options.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
Neonatal stroke
van der Aa et al.
Pathogenesis and risk factors Although several risk factors have been associated with PAIS, little is known about the exact pathophysiological mechanisms responsible for most cases (2). It is generally thought that the patent foramen ovale allows passage of thrombi derived from the placenta or venous circulation, resulting in occlusion of an artery. Most potential risk factors have been identified in small retrospective case series and do not necessarily reflect a causal relation. Maternal risk factors include infertility, primiparity, maternal fever, meconium-stained amniotic fluid, chorioamnionitis, pre-eclampsia and intrauterine growth retardation (8,9). Complicated deliveries, both instrumental and emergency caesarean section, low Apgar scores and hypoglycaemia are more frequently observed in infants with PAIS (6,8,9). Studies have reported prothrombotic factors in more than half of the population studied. In a large case–control study, prothrombotic factors were more often observed in the cases (68%) being studied than in age and sex-matched controls (24%) (10). The presence of prothrombotic factors has also been associated with poor neurological outcome, including development of unilateral spastic cerebral palsy (USCP) (11). It is of interest to note that, despite the presence of these prothrombotic factors, recurrence of arterial ischaemic stroke is rare, suggesting a multifactorial cause of PAIS. Congenital heart disease is an important risk factor for PAIS. Due to the disturbed anatomy, cardiac thrombi are easily formed, which may result in a spectrum of cerebral injury, including PAIS (12). Other factors associated with the development of PAIS are infectious disorders such as meningitis and sepsis, dehydration, arterial dissection, catheterisation and extracorporeal membrane oxygenation, which all increase the risk of thrombus formation. The role of the placenta remains uncertain, as data on placental abnormalities in infants suffering PAIS are scarce, probably because the placenta is often discarded before the onset of clinical symptoms. Clinical presentation The first manifestation of PAIS occurs most frequently during the first week. Seizures are the most common first clinical sign of PAIS, occurring in 70–90% of all infants with PAIS (13). When compared to seizures observed following hypoxic ischaemic encephalopathy (HIE), seizures following PAIS tend to arise later and are more often lateralised (14). Seizures may be subtle and may remain unnoticed, causing a delay in diagnosis. Additional signs may include apnoea, sometimes as clinical sign of a seizure, lethargy, feeding difficulties and hypotonia. Some infants may show asymmetry of tone during the first days to weeks, with hypotonia rather than hypertonia on the affected side. Infants may, however, remain asymptomatic. Imaging studies performed in children presenting with USCP and born at term suggest that approximately half of the infants with PAIS are diagnosed during the neonatal
period (8). Infants who are not diagnosed during the neonatal period, but develop neurological deficits attributable to focal infarction later in infancy, will be diagnosed as presumed perinatal stroke, based on later neuroimaging findings. Clinical presentation of these children will depend on the age at which they are recognised, and the extent of injury, and include early handedness, decreased hand use and rigidity of the upper limb or fisting. Other presentations include seizures, developmental delay and cognitive deficits (3). Diagnosis An amplitude-integrated EEG (aEEG) or full EEG should be obtained in any infant presenting with neonatal seizures and may help to localise the origin of the seizures. Although both single- and two-channel aEEG can be used, the latter may provide additional information of more seizures patterns on the affected side. In the case of PAIS, (a)EEG may further assist in predicting development of motor outcome, which is often worse following the persistence of abnormal background activity on the affected side (15). PAIS is likely in infants without a history of perinatal asphyxia, presenting with focal seizures after 12 h, and neuroimaging is mandatory. As most infants will present with seizures, laboratory studies should be performed to rule out other pathology, including hypoglycaemia, electrolyte disturbances, infection and metabolic syndromes. After confirmation of PAIS, the coagulation status should be evaluated. While prothrombotic mutations may be tested at any time, evaluation of protein-based assays should take place shortly after diagnosis (16). Cranial ultrasound Cranial ultrasound is often the first imaging modality used in infants who are suspected of cerebral injury (Fig. 1A). If performed shortly after the onset of symptoms, scans may still appear normal. In a study of serial cranial ultrasound exams of 47 infants with PAIS, sensitivity increased from 68% in the first 3 days to 87% between days 4 and 10 (17). While the use of the posterior fontanelle improves the detection of PCA stroke, detection of smaller cortical strokes remains challenging (18). Cranial ultrasound is operator dependent, and detection of PAIS may therefore differ between centres. Magnetic resonance imaging Magnetic resonance imaging is the most sensitive imaging modality for detection of PAIS using nonionising radiation and is therefore the gold standard. The most frequently used sequences include T1-weighted (T1WI), T2-weighted (T2WI) and diffusion-weighted imaging (DWI) (Fig. 1B, D). Within the first week after PAIS, T2WI will show high signal intensity in the affected cortex and white matter, which can first be observed from 24 to 48 h onwards (19). The reduced contrast between the cortex and white matter is referred to as the ‘missing-cortex sign’. In contrast, T1WI
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
357
Neonatal stroke
van der Aa et al.
A
B
C
D
E
F
Figure 1 Cranial ultrasound and magnetic resonance imaging (MRI) findings of a full-term infant with a perinatal arterial ischaemic stroke. Cranial ultrasound on day 5 shows an area of increased echogenicity in the territory of the left middle cerebral artery, including involvement of the thalamus (A). An MRI, including T2-weighted imaging (C) and diffusion-weighted imaging (ADC map) (B), was obtained on the same day and showed a main branch middle cerebral artery stroke. A second MRI, acquired at the age of 10 days, showed pseudonormalisation of the diffusion abnormalities (D), but cystic evolution of the white matter (E). At three months of age, multicystic encephalomalacia can be seen in the former stroke area (F). Due to the extent of the injury, this infant was closely followed, and development of a USCP, hemianopia and postnatal epilepsy was observed in the first six months after birth.
will display lower signal intensity in the cortex and white matter. By the end of the first week, the pattern starts to change. T2WI will show a lower signal intensity in the cortex, while a high signal intensity can be observed on T1WI. This is called cortical highlighting. A concurrent change in signal intensity towards the lower signal intensity can be observed in the white matter on T1WI. In large strokes, swelling may be observed, compressing the ventricles and, in some cases, resulting in a midline shift. Diffusion-weighted imaging plays the most important role in the diagnosis of PAIS, due to its high sensitivity for detecting ischaemic lesions in the acute phase, that is, during the first few days after the insult, when signs on T2WI and T1WI can be missed (19). Experimental studies have shown that the rapid shift of water from the extracellular space to the intracellular space, resulting in cytotoxic oedema, can be assessed within minutes as increased signal intensity on DWI. Changes on DWI can, therefore, precede those on conventional T1WI and T2WI and are often seen more clearly on DWI during the first 24–72 h. The changes are characterised by high signal intensity on DWI or low signal intensity on the derived apparent diffusion coefficient (ADC) map. The lowest ADC can be observed around day 3, after which ADC values will slowly increase (19,20). At
358
approximately 6–10 days after the injury, ADC values in the ischaemic tissue appear to have normalised, so-called pseudonormalisation, at which point DWI is of limited value in diagnosing PAIS. Following this period, there is a further increase in the ADC during subsequent days, reflecting the underlying cell lysis and vasogenic oedema, which will eventually result in cystic evolution on T2WI and T1WI. Magnetic resonance imaging plays an important role in the prediction of motor outcome, especially in the development of USCP. On conventional imaging, concomitant involvement of the hemisphere, the basal ganglia and the posterior limb of the internal capsule is highly predictive for development of contralateral USCP (21). Restricted diffusion on DWI at the level of the cerebral peduncle has been interpreted as ‘pre-Wallerian’ degeneration of the descending corticospinal tracts and is a predictor of poor motor outcome (22,23). Finally, diffusion tensor imaging (DTI), which can be used as a marker for white matter integrity, may reveal an asymmetry of the corticospinal tracts, which is highly predictive of the development of a USCP (24,25). Several other MRI sequences may provide additional information following PAIS (Fig. 2). Depending on the technique used, magnetic resonance angiography (MRA)
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
Neonatal stroke
van der Aa et al.
A
B
C
D
Figure 2 Examples of additional magnetic resonance imaging (MRI) sequences that can be used following perinatal arterial ischaemic stroke, shown in an infant with a left middle cerebral artery stroke (A). On the 3D time of flight magnetic resonance angiography, no flow is observed in the left middle cerebral artery (B, arrow). Perfusion-weighted imaging, using arterial spin labelling, shows restricted diffusion in the stroke area with focal regions of hyperperfusion (C, courtesy of Jill de Vis, UMC Utrecht). A fractional anisotropy map, derived from diffusion tensor imaging at three months of age, shows decreased anisotropy in the posterior limb of the internal capsule and an absent external capsule (D).
may provide anatomical information on the cerebral arteries. 3D reconstructions can be used to study the anatomy and may demonstrate anatomical variations or occlusion of a vessel. Phase contrast MRA can be used to quantify the blood flow in large vessels and may indicate asymmetry in carotid blood flow due to hypoperfusion or hyperperfusion (26). Other sequences include MR spectroscopy, perfusionweighted imaging and functional MRI. MR spectroscopy allows quantification of several metabolites, such as Nacetyl-aspartate, choline and creatine. Increased lactate levels and decreased ratios of N-acetyl-aspartate to choline suggest recent ischaemia, but may persist for weeks (27). Studies in adults suggest that perfusion-weighted imaging, such as arterial spin labelling, may be of value in determining the amount of salvageable tissue by looking at the perfusion– diffusion mismatch. Few studies have reported results of perfusion weighted in neonates, and its role following PAIS requires further research (28). Therapy During the acute phase, therapeutic options are limited and mainly involve supportive measures, such as maintenance of normal hydration, electrolytes, glucose, haemoglobin, oxygen and pH levels. Hyperthermia should be prevented, but the
role of therapeutic hypothermia has yet to be determined. This is unlikely to prove useful, due to the delay in onset of clinical symptoms and therefore the missed therapeutic time window. Clinical or subclinical seizures should be treated. aEEG or continuous EEG is required to recognise subclinical seizures and the effect of antiepileptic medication. Guidelines on anticonvulsive therapy vary between centres, but often involve phenobarbital as a first-line treatment. There is no evidence for the use of anticoagulants, and their administration is only recommended in infants with an ongoing cardioembolic source (29). Several therapeutic agents have proved useful in animal stroke models, and these include statins, anti-inflammatory agents and neurotrophic factors (30). Many of these agents still need to be studied in neonatal stroke, although the first neonatal clinical studies are being performed (31). Beyond the neonatal period, therapy aims to treat sequelae and often requires a multidisciplinary approach. Most research has focused on the treatment of motor disabilities. Therapy should be started as soon as asymmetry is recognised and is traditionally aimed at minimising spasticity and maximising range of motion. Severe spasticity can be reduced by oral medication or botulinum toxin injections and is often combined with intensive physical
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
359
Neonatal stroke
van der Aa et al.
therapy. Constraint-induced movement therapy, which addresses the inability to use the affected hand, has shown positive results in children with USCP and can be applied at a young age (32). A pilot study on repetitive transcranial magnetic stimulation (TMS) has shown positive results in children with USCP, but requires further investigation (33). Epilepsy is another sequela that may arise, and this requires appropriate pharmacological treatment. Children with cognitive impairment due to PAIS may require additional attention at school or may need special education. Visual field defects often arise following PCA or large MCA strokes, and this should be taken into account when it comes to school activities such as reading (18,34). Awareness of any social issues, behavioural problems and lowered self-esteem is also necessary, as these are more often observed in children with PAIS and may affect their quality of life. Neurodevelopmental outcome Both short-term and long-term outcomes mainly depend on the extent and location of the stroke. Mortality rates in infants suffering PAIS are low and are often the result of the causal disease, rather than the stroke itself. Recurrence of stroke in the child and its siblings is rare. Data on outcome vary with regard to the incidence of sequelae, due mainly to differences in inclusion criteria and duration of follow-up. Unilateral spastic cerebral palsy is the most frequently observed sequela and has been reported in up to 50% of all cases, depending on stroke location (21). Clinical signs may evolve over time, but may already be present at three months of age (35). MRI plays an important role in predicting motor outcome, as described above. Recurrence of seizures after the neonatal period has been reported in up to 50% of all infants with PAIS and is more common in children who develop USCP (36). Seizures tend to respond well to antiepileptic drugs, and drug-resistant epilepsy is only observed in 10–15% of all cases. Some infants will develop hypsarrhythmia. This is often followed by development of epilepsy at a later age, which sometimes requires surgical removal of epileptogenic tissue (36). There is little consensus on cognitive outcome following PAIS, which may also depend on the duration of follow-up. Studies with serial measurements have shown that cognitive function may decline over time, possibly as a result of the development of postneonatal epilepsy (37–39). Cognitive delay and behavioural problems are more often observed in children with USCP (40). Finally, abnormalities in visual function are not uncommon following PAIS and are, once again, associated with larger lesions and development of USCP (34). MRI shows that these abnormalities are not always related to involvement of the optic radiation or visual cortex, suggesting involvement of other cortical and subcortical structures.
CEREBRAL SINOVENOUS THROMBOSIS Cerebral sinovenous thrombosis (CSVT) has a lower reported incidence than PAIS, with incidences ranging
360
from 0.6 to 12 per 100 000 live births (41,42). It is most likely that this wide range reflects differences in the threshold for performing neuroimaging and the MRI sequences that are used. Furthermore, it is generally assumed that CSVT often remains undiagnosed due to lack of awareness among clinicians, the nonspecific clinical presentation and the difficulty of radiological diagnosis. A higher incidence, of up to 4%, has been reported in preterm infants, but this might also reflect the routine use of cranial ultrasound with Doppler in these infants (43). Pathogenesis and risk factors Cerebral sinovenous thrombosis can be diagnosed when venous blood flow is impaired or absent in one of the cerebral sinuses. Impaired venous drainage often causes increased venous pressure, resulting in increased capillary hydrostatic pressure, vasogenic oedema and, finally, secondary haemorrhagic infarction. Parenchymal lesions are present in 60% of the infants on first imaging, and their location depends on the sinuses involved (44). In contrast to PAIS, no case–control studies have been performed in CSVT, but a number of potential risk factors that may lead to the development of CSVT have been identified. A recent large cohort study reported the presence of a single risk factor in 48% of all cases and multiple risk factors in 39% (45). Pre-eclampsia, which involves a hypercoagulable state, is a frequently reported maternal factor (41,44,46). Other factors include chorioamnionitis and gestational diabetes. The latter is, however, also associated with other risk factors, such as vascular injury, pre-eclampsia and larger infants. Intrapartum factors include a complicated delivery, reported in up to 60% of cases (41), meconium aspiration and intubation at birth (41,44,46,47). Cerebral sinovenous thrombosis is often observed in the context of additional pathologies, such as meningitis, sepsis and dehydration, and male infants are once again more susceptible (41,42,44,46). Some studies have reported a higher incidence of congenital heart defects in infants with CSVT (7,44,46,47). Infants who require extracorporeal membrane oxygenation are at risk of developing CSVT, most likely due to disturbed jugular flow (47). Finally, mechanical compressions by the occipital bone in the supine position have been independently associated with CSVT (48). Prothrombotic factors are frequently found in infants with CSVT, but their role remains unclear (41,42,44). Fitzgerald et al. (44) reported higher incidences of factor V Leiden and MTHFR mutations. A high incidence of MTHFR mutations was also reported by Berfelo et al. (41) in a multicentre study, but all infants were shown to have normal homocysteine levels. Clinical presentation The majority of infants with CSVT present within 48 h after birth, with another 25% presenting later in the first week (41,44,46). Symptoms are often subtle and nonspecific and may include seizures, respiratory distress, lethargy,
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
Neonatal stroke
van der Aa et al.
irritability, apnoea, poor feeding and jitteriness. Seizures of various subtypes occur in approximately two-thirds of cases, with a similar incidence in infants with an early or late presentation (44,46). Comorbidities, which may include risk factors, have been reported in 62% of all cases and include dehydration, meningitis, sepsis and cardiac defects or malformations (44). Diagnosis When infants present with seizures, an aEEG or EEG should be obtained to confirm their presence and their location (41). Additional investigations may be required to detect the presence of the comorbidities described above. These include laboratory studies and cardiac ultrasound. The high incidence of coagulation disorders warrants further investigation on the coagulation status. Neuroimaging serves two purposes, which are imaging of the thrombus and detecting associated parenchymal lesions. Parenchymal lesions are present in 60–80% of all cases and are more common in preterm born infants (49). Cranial ultrasound can be used for both, but has been shown to have a low sensitivity (46,50). In full-term infants, the presence of an intraventricular haemorrhage, especially when observed in combination with a thalamic
haemorrhage, suggests the presence of a CSVT (Fig. 3A). Wu et al. (51) demonstrated that 34% of a cohort of term born infants with an intraventricular haemorrhage had a CSVT. Intraventricular haemorrhages are more common in preterm infants, but a CSVT should be considered if the haemorrhage has an unexpected late onset in and is accompanied by bilateral white matter injury (49). Additional colour flow Doppler may be used to assist in the diagnostic process by demonstrating an absent or decreased flow, but it has a rather low sensitivity of 37–48% (41,52). CT is still used by some centres to diagnose CSVT, but it requires ionising radiation and provides a false negative in half of the cases (46). MRI allows more accurate detection of CSVT and associated parenchymal injury and is considered the gold standard when combined with MR venography. In some cases, the thrombus can already be detected using conventional T1- and T2-weighted imaging (Fig. 3B). MRI may show additional white matter injury due to periventricular congestion, especially in the frontal regions (Fig. 3C). DWI will show restricted diffusion in these areas. In preterm born infants, the white matter injury is known to be more extensive and may result in cystic periventricular leukomalacia (49).
A
B
C
D
Figure 3 Cranial ultrasound and magnetic resonance imaging (MRI) findings of an infant born at term who presented with seizures on day 10 after birth. Cranial ultrasound (A) shows a right-sided IVH and an echodensity in the right thalamus, suggestive of a haemorrhage. Magnetic resonance imaging (MRI) confirmed the presence of a CSVT, showing a large thrombus in the straight sinus (B). This had led to venous congestion resulting in additional white matter injury. This additional injury can be clearly visualised using DWI, showing restricted diffusion in the frontal (so-called iris sign) and parietal white matter predominantly seen in the right hemisphere (C,D).
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
361
Neonatal stroke
van der Aa et al.
Using MR venography, the venous system can be visualised without the use of contrast. Some caution is needed, however, as lower venous flow, a smaller calibre venous sinus and skin moulding can also cause a flow gap in the venous sinuses. These findings should, therefore, be interpreted in the context of additional MR findings and the clinical condition of the infant. Several studies have reported the highest incidence of thrombosis in the superior sagittal sinus, followed by the transverse and straight sinuses (42,44,46), although a more recent study showed involvement of the straight sinus in the majority of the described cases (49). Involvement of multiple sinuses is common and is found in up to 70% of cases (41,44). Like PAIS, information on the coagulation status should be obtained, including the evaluation of protein-based assays shortly after diagnosis (16). Therapy Cerebral sinovenous thrombosis therapy should primarily be aimed at treating any underlying cause that may have led to the development of the thrombus. This includes treatment of dehydration, sepsis, meningitis and significant congenital heart disease. Studies in infants without CSVT show that in the supine position, compression of the occipital bone occurs, which may result in a reduced cerebral venous flow. Adjusting the infant’s pillow decompresses the occipital bone and has been shown to increase flow in the sigmoid and superior sagittal sinus and may offer a noninvasive therapy for infants with CSVT (48). However, this has not yet been studied in this population, and its efficacy still needs to be proven. There is ongoing discussion about whether anticoagulants should be used to prevent propagation of the initial thrombus and to enhance recanalisation. In the absence of randomised control trials, evidence for the use of anticoagulants is limited to case series. Withholding treatment does, however, seem to be associated with an increased risk of thrombus propagation of about 25% (53). Even in the presence of a thalamic haemorrhage, anticoagulation therapy appears to be safe (54). That is why a recent guideline issued by the American College of Chest Physicians recommends administering unfractionated heparin or low molecular weight heparin to infants without an intracranial haemorrhage, followed by low molecular weight heparin for 6–12 weeks (29). In the presence of a significant haemorrhage, anticoagulation therapy may be initiated, but a conservative approach with supportive care and radiological monitoring of the thrombosis at 5–7 days may also be chosen. If propagation of the thrombus is noted at that time, anticoagulation therapy may still be initiated. Neurodevelopmental outcome There are fewer studies on outcome following CSVT than on PAIS, and reported outcomes vary. Death rates tend to be low, in the order of 2–5% (44,45). Higher mortality rates of 19% and 25% have also been reported in two studies, which might be explained by better identification of CSVT cases and differences in the redirection of care (41,55).
362
Long-term outcome studies suggest that 60–80% of all infants develop some kind of deficit (44,45). Motor impairments, including cerebral palsy, are observed in 20–80%. Most children will be able to walk, however, especially in the absence of additional ischaemic injury (56). Describing the cognitive outcome following CSVT is hindered by a lack of the use of standardised assessments and the use of relatively coarse outcome measures, such as ‘moderately abnormal’ and ‘abnormal’ outcome, often performed at a young age. However, available data do show that cognitive outcome is abnormal in 25–73% of all cases (44,49,52). Little is known on visual deficits, although Berfelo et al. (41) reported strabismus and uncoordinated eye movements in 8% of survivors. Postneonatal epilepsy occurs in 15–40% of all infants (42,44,54), and infants who have associated thalamic injury face a particular risk of developing epilepsy. In these cases, an electrical status epilepticus in slow wave sleep is often observed, which may be challenging to treat and is associated with poor cognitive outcome (57).
FUTURE DIRECTIONS During the last decades, considerable progress has been made in the field of perinatal stroke, with regard to epidemiology, potential risk factors, neuroimaging and subsequent outcome. Despite this progress, the pathogenesis remains a black box and therapies are still lacking. One difficulty in studying perinatal stroke is the low incidence of both PAIS and CSVT. Larger number of patients will be required to better understand the pathophysiological mechanisms and to study potential therapies. This may be reached by both better registration and by multicentre collaborations, such as the International Paediatric Stroke Study (13). Such research networks allow adequately powered case–control studies and enable a setting for studying future therapies. Additional insights may come from the application of novel imaging techniques, such as resting state MRI, DTI and serial imaging. While numerous studies have reported the structural changes on MRI following perinatal brain injury, few have studied functional changes in terms of functional connectivity. One study in term born infants with hypoxic ischaemic encephalopathy did indeed show a trend towards declining brain network integration in infants with increasing neuromotor deficits (58). Further studies should point out whether connectivity parameters can indeed be related to functional outcome following perinatal brain injury. This might be particularly interesting in relation to cognitive outcome (59). While several therapies have shown to be effective in experimental stroke studies, their effectiveness and applicability in PAIS still need to be confirmed. The difficulty in treating PAIS is that the most optimal therapeutic time window for starting treatment during the acute phase of hypoxic ischaemic injury is during the first 6 h, while infants with PAIS often present with seizures beyond the first 24 h (14). It is, therefore, most likely that therapies
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
Neonatal stroke
van der Aa et al.
aiming at regeneration and remodelling, which can be applied in the postacute phase, will be most efficient. Such therapies include stem cells, erythropoietin, repetitive TMS, constraint-induced movement therapy (CIMT) and handarm bimanual intensive therapy (30,32,33). Repetitive TMS and CIMT have been applied in older children following PAIS, but it has been suggested that these therapies will be more efficient when applied during a phase when the balance between ipsilateral and contralateral corticospinal tracts is still being established, which is until the age of 12– 18 months. In CSVT, randomised controlled trials are needed to determine the role of anticoagulation therapy.
References 1. Raju TN, Nelson KB, Ferriero D, Lynch JK. Ischemic perinatal stroke: summary of a workshop sponsored by the National Institute of Child Health and Human Development and the National Institute of Neurological Disorders and Stroke. Pediatrics 2007; 120: 609–16. Epub 2007/09/04. 2. Mineyko A, Kirton A. The black box of perinatal ischemic stroke pathogenesis. J Child Neurol 2011; 26: 1154–62. Epub 2011/06/15. 3. Kirton A, Deveber G, Pontigon AM, Macgregor D, Shroff M. Presumed perinatal ischemic stroke: vascular classification predicts outcomes. Ann Neurol 2008; 63: 436–43. Epub 2008/ 02/29 4. Lee J, Croen LA, Backstrand KH, Yoshida CK, Henning LH, Lindan C, et al. Maternal and infant characteristics associated with perinatal arterial stroke in the infant. JAMA 2005; 293: 723–9. Epub 2005/02/11 5. Laugesaar R, Kolk A, Tomberg T, Metsvaht T, Lintrop M, Varendi H, et al. Acutely and retrospectively diagnosed perinatal stroke: a population-based study. Stroke 2007; 38: 2234–40. Epub 2007/06/23. 6. Benders MJ, Groenendaal F, Uiterwaal CS, Nikkels PG, Bruinse HW, Nievelstein RA, et al. Maternal and infant characteristics associated with perinatal arterial stroke in the preterm infant. Stroke 2007; 38: 1759–65. Epub 2007/05/15. 7. Golomb MR, Fullerton HJ, Nowak-Gottl U, Deveber G, International Pediatric Stroke Study G. Male predominance in childhood ischemic stroke: findings from the international pediatric stroke study. Stroke 2009; 40: 52–7. Epub 2008/09/ 13. 8. Wu YW, March WM, Croen LA, Grether JK, Escobar GJ, Newman TB. Perinatal stroke in children with motor impairment: a population-based study. Pediatrics 2004; 114: 612–9. Epub 2004/09/03. 9. Harteman JC, Groenendaal F, Kwee A, Welsing PM, Benders MJ, de Vries LS. Risk factors for perinatal arterial ischaemic stroke: a case–control study. Arch Dis Child Fetal Neonatal Ed 2012; 97: F411–6. 10. Gunther G, Junker R, Strater R, Schobess R, Kurnik K, Heller C, et al. Symptomatic ischemic stroke in full-term neonates: role of acquired and genetic prothrombotic risk factors. Stroke 2000; 31: 2437–41. Epub 2000/10/07. 11. Mercuri E, Cowan F, Gupte G, Manning R, Laffan M, Rutherford M, et al. Prothrombotic disorders and abnormal neurodevelopmental outcome in infants with neonatal cerebral infarction. Pediatrics 2001; 107: 1400–4. Epub 2001/06/05. 12. McQuillen PS, Miller SP. Congenital heart disease and brain development. Ann N Y Acad Sci 2010; 1184: 68–86. Epub 2010/02/12.
13. Kirton A, Armstrong-Wells J, Chang T, Deveber G, Rivkin MJ, Hernandez M, et al. Symptomatic neonatal arterial ischemic stroke: the international pediatric stroke study. Pediatrics 2011; 128: e1402–10. Epub 2011/11/30. 14. Rafay MF, Cortez MA, de Veber GA, Tan-Dy C, Al-Futaisi A, Yoon W, et al. Predictive value of clinical and EEG features in the diagnosis of stroke and hypoxic ischemic encephalopathy in neonates with seizures. Stroke 2009; 40: 2402–7. Epub 2009/ 05/30. 15. Mercuri E, Rutherford M, Cowan F, Pennock J, Counsell S, Papadimitriou M, et al. Early prognostic indicators of outcome in infants with neonatal cerebral infarction: a clinical, electroencephalogram, and magnetic resonance imaging study. Pediatrics 1999; 103: 39–46. Epub 1999/01/26. 16. Nowak-Gottl U, Duering C, Kempf-Bielack B, Strater R. Thromboembolic diseases in neonates and children. Pathophysiol Haemost Thromb 2003; 33: 269–74. Epub 2005/ 02/05. 17. Cowan F, Mercuri E, Groenendaal F, Bassi L, Ricci D, Rutherford M, et al. Does cranial ultrasound imaging identify arterial cerebral infarction in term neonates? Arch Dis Child Fetal Neonatal Ed 2005; 90: F252–6. Epub 2005/04/23. 18. van der Aa NE, Dudink J, Benders MJNL, Govaert P, van Straaten HL, Porro GL, et al. Neonatal posterior cerebral artery stroke: clinical presentation, MRI findings and outcome. Dev Med Child Neurol 2013; 55: 283–90. 19. Dudink J, Mercuri E, Al Nakib L, Govaert P, Counsell SJ, Rutherford MA, et al. Evolution of unilateral perinatal arterial ischemic stroke on conventional and diffusion-weighted MR imaging. AJNR Am J Neuroradiol 2009; 30: 998–1004. 20. van der Aa NE, Benders MJ, Vincken KL, Groenendaal F, de Vries LS. The course of apparent diffusion coefficient values following perinatal arterial ischemic stroke. PLoS ONE 2013; 8: e56784. 21. Mercuri E, Barnett A, Rutherford M, Guzzetta A, Haataja L, Cioni G, et al. Neonatal cerebral infarction and neuromotor outcome at school age. Pediatrics 2004; 113: 95–100. 22. de Vries LS, van der Grond J, van Haastert IC, Groenendaal F. Prediction of outcome in new-born infants with arterial ischaemic stroke using diffusion-weighted magnetic resonance imaging. Neuropediatrics 2005; 36: 12–20. 23. Kirton A, Shroff M, Visvanathan T, deVeber G. Quantified corticospinal tract diffusion restriction predicts neonatal stroke outcome. Stroke 2007; 38: 974–80. 24. Roze E, Harris PA, Ball G, Elorza LZ, Braga RM, Allsop JM, et al. Tractography of the corticospinal tracts in infants with focal perinatal injury: comparison with normal controls and to motor development. Neuroradiology 2012; 54: 507–16. Epub 2011/10/19. 25. van der Aa NE, Leemans A, Northington FJ, van Straaten HL, van Haastert IC, Groenendaal F, et al. Does diffusion tensor imaging-based tractography at 3 months of age contribute to the prediction of motor outcome after perinatal arterial ischemic stroke? Stroke 2011; 42: 3410–4. Epub 2011/10/25. 26. van der Aa NE, Porsius ED, Hendrikse J, van Kooij BJ, Benders MJ, de Vries LS, et al. Changes in carotid blood flow after unilateral perinatal arterial ischemic stroke. Pediatr Res 2012; 72: 50–6. Epub 2012/03/27. 27. Groenendaal F, van der Grond J, Witkamp TD, de Vries LS. Proton magnetic resonance spectroscopic imaging in neonatal stroke. Neuropediatrics 1995; 26: 243–8. Epub 1995/10/01. 28. De Vis JB, Petersen ET, Kersbergen KJ, Alderliesten T, de Vries LS, van Bel F, et al. Evaluation of perinatal arterial ischemic stroke using noninvasive arterial spin labeling perfusion MRI. Pediatr Res 2013; 74: 307–13. Epub 2013/06/26.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
363
Neonatal stroke
van der Aa et al.
29. Monagle P, Chan AK, Goldenberg NA, Ichord RN, Journeycake JM, Nowak-Gottl U, et al. Antithrombotic therapy in neonates and children: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141: e737S–801S. Epub 2012/02/15. 30. Barnette AR, Inder TE. Evaluation and management of stroke in the neonate. Clin Perinatol 2009; 36: 125–36. Epub 2009/ 01/24. 31. Benders MJ, van der Aa NE, Roks M, van Straaten HL, Isgum I, Viergever MA, et al. Feasibility and safety of erythropoietin for neuroprotection after perinatal arterial ischemic stroke. J Pediatr 2013; pii: S0022-3476(13)01380-2. doi: 10.1016/j. jpeds.2013.10.084. [Epub ahead of print]. 32. Aarts PB, Jongerius PH, Geerdink YA, van Limbeek J, Geurts AC. Effectiveness of modified constraint-induced movement therapy in children with unilateral spastic cerebral palsy: a randomized controlled trial. Neurorehabil Neural Repair 2010; 24: 509–18. Epub 2010/04/29. 33. Kirton A, Chen R, Friefeld S, Gunraj C, Pontigon AM, Deveber G. Contralesional repetitive transcranial magnetic stimulation for chronic hemiparesis in subcortical paediatric stroke: a randomised trial. Lancet Neurol 2008; 7: 507–13. Epub 2008/ 05/06. 34. Mercuri E, Anker S, Guzzetta A, Barnett A, Haataja L, Rutherford M, et al. Neonatal cerebral infarction and visual function at school age. Arch Dis Child Fetal Neonatal Ed 2003; 88: F487–91. Epub 2003/11/07. 35. Guzzetta A, Pizzardi A, Belmonti V, Boldrini A, Carotenuto M, D’Acunto G, et al. Hand movements at 3 months predict later hemiplegia in term infants with neonatal cerebral infarction. Dev Med Child Neurol 2010; 52: 767–72. Epub 2009/10/30. 36. Wanigasinghe J, Reid SM, Mackay MT, Reddihough DS, Harvey AS, Freeman JL. Epilepsy in hemiplegic cerebral palsy due to perinatal arterial ischaemic stroke. Dev Med Child Neurol 2010; 52: 1021–7. Epub 2010/05/26. 37. Ballantyne AO, Spilkin AM, Hesselink J, Trauner DA. Plasticity in the developing brain: intellectual, language and academic functions in children with ischaemic perinatal stroke. Brain 2008; 131: 2975–85. 38. Westmacott R, Macgregor D, Askalan R, deVeber G. Late emergence of cognitive deficits after unilateral neonatal stroke. Stroke 2009; 40: 2012–9. 39. van Buuren LM, van der Aa NE, Dekker HC, Vermeulen RJ, van Nieuwenhuizen O, van Schooneveld MM, et al. Cognitive outcome in childhood after unilateral perinatal brain injury. Dev Med Child Neurol 2013; 55: 934–40. 40. Ricci D, Mercuri E, Barnett A, Rathbone R, Cota F, Haataja L, et al. Cognitive outcome at early school age in term-born children with perinatally acquired middle cerebral artery territory infarction. Stroke 2008; 39: 403–10. Epub 2007/12/29. 41. Berfelo FJ, Kersbergen KJ, van Ommen CH, Govaert P, van Straaten HL, Poll-The BT, et al. Neonatal cerebral sinovenous thrombosis from symptom to outcome. Stroke 2010; 41: 1382– 8. Epub 2010/06/05. 42. deVeber G, Andrew M, Adams C, Bjornson B, Booth F, Buckley DJ, et al. Cerebral sinovenous thrombosis in children. N Engl J Med 2001; 345: 417–23. Epub 2001/08/11. 43. Raets MM, Sol JJ, Govaert P, Lequin MH, Reiss IK, Kroon AA, et al. Serial Cranial US for Detection of Cerebral Sinovenous Thrombosis in Preterm Infants. Radiology 2013; 269: 879–86. Epub 2013/08/30.
364
44. Fitzgerald KC, Williams LS, Garg BP, Carvalho KS, Golomb MR. Cerebral sinovenous thrombosis in the neonate. Arch Neurol 2006; 63: 405–9. Epub 2006/03/15. 45. Moharir MD, Shroff M, Pontigon AM, Askalan R, Yau I, Macgregor D, et al. A prospective outcome study of neonatal cerebral sinovenous thrombosis. J Child Neurol 2011; 26: 1137–44. Epub 2011/06/02. 46. Nwosu ME, Williams LS, Edwards-Brown M, Eckert GJ, Golomb MR. Neonatal sinovenous thrombosis: presentation and association with imaging. Pediatr Neurol 2008; 39: 155–61. Epub 2008/08/30. 47. Wu YW, Miller SP, Chin K, Collins AE, Lomeli SC, Chuang NA, et al. Multiple risk factors in neonatal sinovenous thrombosis. Neurology 2002; 59: 438–40. Epub 2002/08/15. 48. Tan MA, Miller E, Shroff MM, Deveber GA, Kirton A. Alleviation of neonatal sinovenous compression to enhance cerebral venous blood flow. J Child Neurol 2013; 28: 583–8. Epub 2012/07/19. 49. Kersbergen KJ, Groenendaal F, Benders MJ, van Straaten HL, Niwa T, Nievelstein RA, et al. The spectrum of associated brain lesions in cerebral sinovenous thrombosis: relation to gestational age and outcome. Arch Dis Child Fetal Neonatal Ed 2011; 96: F404–9. Epub 2011/02/15. 50. Miller E, Daneman A, Doria AS, Blaser S, Traubici J, Jarrin J, et al. Color Doppler US of normal cerebral venous sinuses in neonates: a comparison with MR venography. Pediatr Radiol 2012; 42: 1070–9. Epub 2012/04/26. 51. Wu YW, Hamrick SE, Miller SP, Haward MF, Lai MC, Callen PW, et al. Intraventricular hemorrhage in term neonates caused by sinovenous thrombosis. Ann Neurol 2003; 54: 123–6. Epub 2003/07/03. 52. Grunt S, Wingeier K, Wehrli E, Boltshauser E, Capone A, Fluss J, et al. Cerebral sinus venous thrombosis in Swiss children. Dev Med Child Neurol 2010; 52: 1145–50. Epub 2010/06/25. 53. Moharir MD, Shroff M, Stephens D, Pontigon AM, Chan A, MacGregor D, et al. Anticoagulants in pediatric cerebral sinovenous thrombosis: a safety and outcome study. Ann Neurol 2010; 67: 590–9. Epub 2010/05/04. 54. Kersbergen KJ, de Vries LS, van Straaten HL, Benders MJ, Nievelstein RA, Groenendaal F. Anticoagulation therapy and imaging in neonates with a unilateral thalamic hemorrhage due to cerebral sinovenous thrombosis. Stroke 2009; 40: 2754–60. Epub 2009/06/23. 55. Wasay M, Dai AI, Ansari M, Shaikh Z, Roach ES. Cerebral venous sinus thrombosis in children: a multicenter cohort from the United States. J Child Neurol 2008; 23: 26–31. Epub 2008/ 01/11. 56. Golomb MR, deVeber GA, MacGregor DL, Domi T, Whyte H, Stephens D, et al. Independent walking after neonatal arterial ischemic stroke and sinovenous thrombosis. J Child Neurol 2003; 18: 530–6. Epub 2003/09/19. 57. Kersbergen KJ, de Vries LS, Leijten FS, Braun KP, Nievelstein RA, Groenendaal F, et al. Neonatal thalamic hemorrhage is strongly associated with electrical status epilepticus in slow wave sleep. Epilepsia 2013; 54: 733–40. Epub 2013/03/20. 58. Tymofiyeva O, Hess CP, Ziv E, Tian N, Bonifacio SL, McQuillen PS, et al. Towards the “baby connectome”: mapping the structural connectivity of the newborn brain. PLoS ONE 2012; 7: e31029. Epub 2012/02/22. 59. van den Heuvel MP, Stam CJ, Kahn RS, Hulshoff Pol HE. Efficiency of functional brain networks and intellectual performance. J Neurosci 2009; 29: 7619–24. Epub 2009/06/12.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
REVIEW ARTICLE
Neonatal stroke: a review of the current evidence on epidemiology, pathogenesis, diagnostics and therapeutic options NE van der Aa, MJNL Benders, F Groenendaal, LS de Vries ([email protected]) Department of Neonatology, Wilhelmina Children’s Hospital, University Medical Center Utrecht, Utrecht, The Netherlands
Keywords Cerebral sinovenous thrombosis, MRI, Neonatal stroke Correspondence Prof. Dr. Linda S. de Vries, MD, PhD, Department of Neonatology, Wilhelmina Children’s Hospital, UMC Utrecht, KE 04.123.1, PO Box 85090, 3508 AB Utrecht, The Netherlands. Tel: +31-88-7554545 | Fax: +31-88-7555320 | Email: [email protected]
ABSTRACT Neonatal stroke, including perinatal arterial ischaemic stroke and cerebral sinovenous thrombosis, remains a serious problem in the neonate. This article reviews the current evidence on epidemiology, pathogenesis, diagnostics and therapeutic options. Conclusion: Although our understanding of the underlying mechanisms and possible risk factors has improved, little progress has been made towards therapeutic options. Considering the high incidence of neurological sequelae, the need for therapeutic options is high and should be the focus of future research.
Received 20 November 2013; revised 2 January 2014; accepted 10 January 2014. DOI:10.1111/apa.12555
Neonatal stroke can be defined as a cerebrovascular injury that occurs around birth. It can be focal or multifocal and may include both ischaemic and haemorrhagic injury. Neonatal stroke is most frequently referred to as perinatal cerebral injury of ischaemic origin. In an international workshop on this subject, held in 2006, ischaemic perinatal stroke was narrowed down to ‘a group of heterogeneous conditions in which there is focal disruption of cerebral blood flow secondary to arterial or cerebral venous thrombosis or embolisation, between 20 weeks of foetal life through to the 28th postnatal day, confirmed by neuroimaging or neuropathological studies (1). Two common subtypes are perinatal arterial ischaemic stroke (PAIS) and cerebral sinovenous thrombosis (CSVT). However, there is still an ongoing discussion about which other patterns of injury should be regarded as perinatal stroke. Periventricular haemorrhagic infarction (PVHI), which is commonly seen in preterm infants, is frequently regarded as being part of the stroke spectrum (2,3). Another one is haemorrhagic stroke. Finally, presumed perinatal stroke, encompassing both arterial ischaemic stroke and PVHI, is also considered part of the neonatal stroke spectrum, though by definition they are diagnosed beyond the 28th postnatal day. In this article, we review both PAIS and CSVT in terms of their aetiology, clinical presentation and the diagnostic and therapeutic options.
PERINATAL ARTERIAL ISCHAEMIC STROKE Perinatal arterial ischaemic stroke has an incidence of between 1/1600 and 1/5000 (4,5). The wide range of
356
reported incidences may be explained by differences in the definition and diagnosis of PAIS. Some only include PAIS, while others also include presumed perinatal stroke and haemorrhagic stroke. All studies are of retrospective design and therefore depend on the accuracy of the databases used. In view of the fact that magnetic resonance imaging (MRI) scans are not routinely obtained for all infants presenting to neonatal intensive care units with neonatal seizures, the true incidence is likely to be even higher. Perinatal arterial ischaemic stroke may affect both term and preterm born infants. It has been suggested that the incidence of PAIS is higher in preterm born infants, but this might also reflect the routine use of cranial ultrasound in preterm infants (6). Several studies have reported a male predominance of approximately 60%, and PAIS is known to involve the left hemisphere more often (7).
Key notes
Neonatal stroke, including perinatal arterial ischaemic stroke and cerebral sinovenous thrombosis, has a multifactorial aetiology and often results in neurological sequelae. Neonatal neuroimaging plays an important role in both diagnosis and prognosis and may predict motor outcome. Although a lot of progress has been made in identifying possible underlying mechanisms, further research is required with regard to therapeutic options.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
Neonatal stroke
van der Aa et al.
Pathogenesis and risk factors Although several risk factors have been associated with PAIS, little is known about the exact pathophysiological mechanisms responsible for most cases (2). It is generally thought that the patent foramen ovale allows passage of thrombi derived from the placenta or venous circulation, resulting in occlusion of an artery. Most potential risk factors have been identified in small retrospective case series and do not necessarily reflect a causal relation. Maternal risk factors include infertility, primiparity, maternal fever, meconium-stained amniotic fluid, chorioamnionitis, pre-eclampsia and intrauterine growth retardation (8,9). Complicated deliveries, both instrumental and emergency caesarean section, low Apgar scores and hypoglycaemia are more frequently observed in infants with PAIS (6,8,9). Studies have reported prothrombotic factors in more than half of the population studied. In a large case–control study, prothrombotic factors were more often observed in the cases (68%) being studied than in age and sex-matched controls (24%) (10). The presence of prothrombotic factors has also been associated with poor neurological outcome, including development of unilateral spastic cerebral palsy (USCP) (11). It is of interest to note that, despite the presence of these prothrombotic factors, recurrence of arterial ischaemic stroke is rare, suggesting a multifactorial cause of PAIS. Congenital heart disease is an important risk factor for PAIS. Due to the disturbed anatomy, cardiac thrombi are easily formed, which may result in a spectrum of cerebral injury, including PAIS (12). Other factors associated with the development of PAIS are infectious disorders such as meningitis and sepsis, dehydration, arterial dissection, catheterisation and extracorporeal membrane oxygenation, which all increase the risk of thrombus formation. The role of the placenta remains uncertain, as data on placental abnormalities in infants suffering PAIS are scarce, probably because the placenta is often discarded before the onset of clinical symptoms. Clinical presentation The first manifestation of PAIS occurs most frequently during the first week. Seizures are the most common first clinical sign of PAIS, occurring in 70–90% of all infants with PAIS (13). When compared to seizures observed following hypoxic ischaemic encephalopathy (HIE), seizures following PAIS tend to arise later and are more often lateralised (14). Seizures may be subtle and may remain unnoticed, causing a delay in diagnosis. Additional signs may include apnoea, sometimes as clinical sign of a seizure, lethargy, feeding difficulties and hypotonia. Some infants may show asymmetry of tone during the first days to weeks, with hypotonia rather than hypertonia on the affected side. Infants may, however, remain asymptomatic. Imaging studies performed in children presenting with USCP and born at term suggest that approximately half of the infants with PAIS are diagnosed during the neonatal
period (8). Infants who are not diagnosed during the neonatal period, but develop neurological deficits attributable to focal infarction later in infancy, will be diagnosed as presumed perinatal stroke, based on later neuroimaging findings. Clinical presentation of these children will depend on the age at which they are recognised, and the extent of injury, and include early handedness, decreased hand use and rigidity of the upper limb or fisting. Other presentations include seizures, developmental delay and cognitive deficits (3). Diagnosis An amplitude-integrated EEG (aEEG) or full EEG should be obtained in any infant presenting with neonatal seizures and may help to localise the origin of the seizures. Although both single- and two-channel aEEG can be used, the latter may provide additional information of more seizures patterns on the affected side. In the case of PAIS, (a)EEG may further assist in predicting development of motor outcome, which is often worse following the persistence of abnormal background activity on the affected side (15). PAIS is likely in infants without a history of perinatal asphyxia, presenting with focal seizures after 12 h, and neuroimaging is mandatory. As most infants will present with seizures, laboratory studies should be performed to rule out other pathology, including hypoglycaemia, electrolyte disturbances, infection and metabolic syndromes. After confirmation of PAIS, the coagulation status should be evaluated. While prothrombotic mutations may be tested at any time, evaluation of protein-based assays should take place shortly after diagnosis (16). Cranial ultrasound Cranial ultrasound is often the first imaging modality used in infants who are suspected of cerebral injury (Fig. 1A). If performed shortly after the onset of symptoms, scans may still appear normal. In a study of serial cranial ultrasound exams of 47 infants with PAIS, sensitivity increased from 68% in the first 3 days to 87% between days 4 and 10 (17). While the use of the posterior fontanelle improves the detection of PCA stroke, detection of smaller cortical strokes remains challenging (18). Cranial ultrasound is operator dependent, and detection of PAIS may therefore differ between centres. Magnetic resonance imaging Magnetic resonance imaging is the most sensitive imaging modality for detection of PAIS using nonionising radiation and is therefore the gold standard. The most frequently used sequences include T1-weighted (T1WI), T2-weighted (T2WI) and diffusion-weighted imaging (DWI) (Fig. 1B, D). Within the first week after PAIS, T2WI will show high signal intensity in the affected cortex and white matter, which can first be observed from 24 to 48 h onwards (19). The reduced contrast between the cortex and white matter is referred to as the ‘missing-cortex sign’. In contrast, T1WI
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
357
Neonatal stroke
van der Aa et al.
A
B
C
D
E
F
Figure 1 Cranial ultrasound and magnetic resonance imaging (MRI) findings of a full-term infant with a perinatal arterial ischaemic stroke. Cranial ultrasound on day 5 shows an area of increased echogenicity in the territory of the left middle cerebral artery, including involvement of the thalamus (A). An MRI, including T2-weighted imaging (C) and diffusion-weighted imaging (ADC map) (B), was obtained on the same day and showed a main branch middle cerebral artery stroke. A second MRI, acquired at the age of 10 days, showed pseudonormalisation of the diffusion abnormalities (D), but cystic evolution of the white matter (E). At three months of age, multicystic encephalomalacia can be seen in the former stroke area (F). Due to the extent of the injury, this infant was closely followed, and development of a USCP, hemianopia and postnatal epilepsy was observed in the first six months after birth.
will display lower signal intensity in the cortex and white matter. By the end of the first week, the pattern starts to change. T2WI will show a lower signal intensity in the cortex, while a high signal intensity can be observed on T1WI. This is called cortical highlighting. A concurrent change in signal intensity towards the lower signal intensity can be observed in the white matter on T1WI. In large strokes, swelling may be observed, compressing the ventricles and, in some cases, resulting in a midline shift. Diffusion-weighted imaging plays the most important role in the diagnosis of PAIS, due to its high sensitivity for detecting ischaemic lesions in the acute phase, that is, during the first few days after the insult, when signs on T2WI and T1WI can be missed (19). Experimental studies have shown that the rapid shift of water from the extracellular space to the intracellular space, resulting in cytotoxic oedema, can be assessed within minutes as increased signal intensity on DWI. Changes on DWI can, therefore, precede those on conventional T1WI and T2WI and are often seen more clearly on DWI during the first 24–72 h. The changes are characterised by high signal intensity on DWI or low signal intensity on the derived apparent diffusion coefficient (ADC) map. The lowest ADC can be observed around day 3, after which ADC values will slowly increase (19,20). At
358
approximately 6–10 days after the injury, ADC values in the ischaemic tissue appear to have normalised, so-called pseudonormalisation, at which point DWI is of limited value in diagnosing PAIS. Following this period, there is a further increase in the ADC during subsequent days, reflecting the underlying cell lysis and vasogenic oedema, which will eventually result in cystic evolution on T2WI and T1WI. Magnetic resonance imaging plays an important role in the prediction of motor outcome, especially in the development of USCP. On conventional imaging, concomitant involvement of the hemisphere, the basal ganglia and the posterior limb of the internal capsule is highly predictive for development of contralateral USCP (21). Restricted diffusion on DWI at the level of the cerebral peduncle has been interpreted as ‘pre-Wallerian’ degeneration of the descending corticospinal tracts and is a predictor of poor motor outcome (22,23). Finally, diffusion tensor imaging (DTI), which can be used as a marker for white matter integrity, may reveal an asymmetry of the corticospinal tracts, which is highly predictive of the development of a USCP (24,25). Several other MRI sequences may provide additional information following PAIS (Fig. 2). Depending on the technique used, magnetic resonance angiography (MRA)
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
Neonatal stroke
van der Aa et al.
A
B
C
D
Figure 2 Examples of additional magnetic resonance imaging (MRI) sequences that can be used following perinatal arterial ischaemic stroke, shown in an infant with a left middle cerebral artery stroke (A). On the 3D time of flight magnetic resonance angiography, no flow is observed in the left middle cerebral artery (B, arrow). Perfusion-weighted imaging, using arterial spin labelling, shows restricted diffusion in the stroke area with focal regions of hyperperfusion (C, courtesy of Jill de Vis, UMC Utrecht). A fractional anisotropy map, derived from diffusion tensor imaging at three months of age, shows decreased anisotropy in the posterior limb of the internal capsule and an absent external capsule (D).
may provide anatomical information on the cerebral arteries. 3D reconstructions can be used to study the anatomy and may demonstrate anatomical variations or occlusion of a vessel. Phase contrast MRA can be used to quantify the blood flow in large vessels and may indicate asymmetry in carotid blood flow due to hypoperfusion or hyperperfusion (26). Other sequences include MR spectroscopy, perfusionweighted imaging and functional MRI. MR spectroscopy allows quantification of several metabolites, such as Nacetyl-aspartate, choline and creatine. Increased lactate levels and decreased ratios of N-acetyl-aspartate to choline suggest recent ischaemia, but may persist for weeks (27). Studies in adults suggest that perfusion-weighted imaging, such as arterial spin labelling, may be of value in determining the amount of salvageable tissue by looking at the perfusion– diffusion mismatch. Few studies have reported results of perfusion weighted in neonates, and its role following PAIS requires further research (28). Therapy During the acute phase, therapeutic options are limited and mainly involve supportive measures, such as maintenance of normal hydration, electrolytes, glucose, haemoglobin, oxygen and pH levels. Hyperthermia should be prevented, but the
role of therapeutic hypothermia has yet to be determined. This is unlikely to prove useful, due to the delay in onset of clinical symptoms and therefore the missed therapeutic time window. Clinical or subclinical seizures should be treated. aEEG or continuous EEG is required to recognise subclinical seizures and the effect of antiepileptic medication. Guidelines on anticonvulsive therapy vary between centres, but often involve phenobarbital as a first-line treatment. There is no evidence for the use of anticoagulants, and their administration is only recommended in infants with an ongoing cardioembolic source (29). Several therapeutic agents have proved useful in animal stroke models, and these include statins, anti-inflammatory agents and neurotrophic factors (30). Many of these agents still need to be studied in neonatal stroke, although the first neonatal clinical studies are being performed (31). Beyond the neonatal period, therapy aims to treat sequelae and often requires a multidisciplinary approach. Most research has focused on the treatment of motor disabilities. Therapy should be started as soon as asymmetry is recognised and is traditionally aimed at minimising spasticity and maximising range of motion. Severe spasticity can be reduced by oral medication or botulinum toxin injections and is often combined with intensive physical
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
359
Neonatal stroke
van der Aa et al.
therapy. Constraint-induced movement therapy, which addresses the inability to use the affected hand, has shown positive results in children with USCP and can be applied at a young age (32). A pilot study on repetitive transcranial magnetic stimulation (TMS) has shown positive results in children with USCP, but requires further investigation (33). Epilepsy is another sequela that may arise, and this requires appropriate pharmacological treatment. Children with cognitive impairment due to PAIS may require additional attention at school or may need special education. Visual field defects often arise following PCA or large MCA strokes, and this should be taken into account when it comes to school activities such as reading (18,34). Awareness of any social issues, behavioural problems and lowered self-esteem is also necessary, as these are more often observed in children with PAIS and may affect their quality of life. Neurodevelopmental outcome Both short-term and long-term outcomes mainly depend on the extent and location of the stroke. Mortality rates in infants suffering PAIS are low and are often the result of the causal disease, rather than the stroke itself. Recurrence of stroke in the child and its siblings is rare. Data on outcome vary with regard to the incidence of sequelae, due mainly to differences in inclusion criteria and duration of follow-up. Unilateral spastic cerebral palsy is the most frequently observed sequela and has been reported in up to 50% of all cases, depending on stroke location (21). Clinical signs may evolve over time, but may already be present at three months of age (35). MRI plays an important role in predicting motor outcome, as described above. Recurrence of seizures after the neonatal period has been reported in up to 50% of all infants with PAIS and is more common in children who develop USCP (36). Seizures tend to respond well to antiepileptic drugs, and drug-resistant epilepsy is only observed in 10–15% of all cases. Some infants will develop hypsarrhythmia. This is often followed by development of epilepsy at a later age, which sometimes requires surgical removal of epileptogenic tissue (36). There is little consensus on cognitive outcome following PAIS, which may also depend on the duration of follow-up. Studies with serial measurements have shown that cognitive function may decline over time, possibly as a result of the development of postneonatal epilepsy (37–39). Cognitive delay and behavioural problems are more often observed in children with USCP (40). Finally, abnormalities in visual function are not uncommon following PAIS and are, once again, associated with larger lesions and development of USCP (34). MRI shows that these abnormalities are not always related to involvement of the optic radiation or visual cortex, suggesting involvement of other cortical and subcortical structures.
CEREBRAL SINOVENOUS THROMBOSIS Cerebral sinovenous thrombosis (CSVT) has a lower reported incidence than PAIS, with incidences ranging
360
from 0.6 to 12 per 100 000 live births (41,42). It is most likely that this wide range reflects differences in the threshold for performing neuroimaging and the MRI sequences that are used. Furthermore, it is generally assumed that CSVT often remains undiagnosed due to lack of awareness among clinicians, the nonspecific clinical presentation and the difficulty of radiological diagnosis. A higher incidence, of up to 4%, has been reported in preterm infants, but this might also reflect the routine use of cranial ultrasound with Doppler in these infants (43). Pathogenesis and risk factors Cerebral sinovenous thrombosis can be diagnosed when venous blood flow is impaired or absent in one of the cerebral sinuses. Impaired venous drainage often causes increased venous pressure, resulting in increased capillary hydrostatic pressure, vasogenic oedema and, finally, secondary haemorrhagic infarction. Parenchymal lesions are present in 60% of the infants on first imaging, and their location depends on the sinuses involved (44). In contrast to PAIS, no case–control studies have been performed in CSVT, but a number of potential risk factors that may lead to the development of CSVT have been identified. A recent large cohort study reported the presence of a single risk factor in 48% of all cases and multiple risk factors in 39% (45). Pre-eclampsia, which involves a hypercoagulable state, is a frequently reported maternal factor (41,44,46). Other factors include chorioamnionitis and gestational diabetes. The latter is, however, also associated with other risk factors, such as vascular injury, pre-eclampsia and larger infants. Intrapartum factors include a complicated delivery, reported in up to 60% of cases (41), meconium aspiration and intubation at birth (41,44,46,47). Cerebral sinovenous thrombosis is often observed in the context of additional pathologies, such as meningitis, sepsis and dehydration, and male infants are once again more susceptible (41,42,44,46). Some studies have reported a higher incidence of congenital heart defects in infants with CSVT (7,44,46,47). Infants who require extracorporeal membrane oxygenation are at risk of developing CSVT, most likely due to disturbed jugular flow (47). Finally, mechanical compressions by the occipital bone in the supine position have been independently associated with CSVT (48). Prothrombotic factors are frequently found in infants with CSVT, but their role remains unclear (41,42,44). Fitzgerald et al. (44) reported higher incidences of factor V Leiden and MTHFR mutations. A high incidence of MTHFR mutations was also reported by Berfelo et al. (41) in a multicentre study, but all infants were shown to have normal homocysteine levels. Clinical presentation The majority of infants with CSVT present within 48 h after birth, with another 25% presenting later in the first week (41,44,46). Symptoms are often subtle and nonspecific and may include seizures, respiratory distress, lethargy,
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
Neonatal stroke
van der Aa et al.
irritability, apnoea, poor feeding and jitteriness. Seizures of various subtypes occur in approximately two-thirds of cases, with a similar incidence in infants with an early or late presentation (44,46). Comorbidities, which may include risk factors, have been reported in 62% of all cases and include dehydration, meningitis, sepsis and cardiac defects or malformations (44). Diagnosis When infants present with seizures, an aEEG or EEG should be obtained to confirm their presence and their location (41). Additional investigations may be required to detect the presence of the comorbidities described above. These include laboratory studies and cardiac ultrasound. The high incidence of coagulation disorders warrants further investigation on the coagulation status. Neuroimaging serves two purposes, which are imaging of the thrombus and detecting associated parenchymal lesions. Parenchymal lesions are present in 60–80% of all cases and are more common in preterm born infants (49). Cranial ultrasound can be used for both, but has been shown to have a low sensitivity (46,50). In full-term infants, the presence of an intraventricular haemorrhage, especially when observed in combination with a thalamic
haemorrhage, suggests the presence of a CSVT (Fig. 3A). Wu et al. (51) demonstrated that 34% of a cohort of term born infants with an intraventricular haemorrhage had a CSVT. Intraventricular haemorrhages are more common in preterm infants, but a CSVT should be considered if the haemorrhage has an unexpected late onset in and is accompanied by bilateral white matter injury (49). Additional colour flow Doppler may be used to assist in the diagnostic process by demonstrating an absent or decreased flow, but it has a rather low sensitivity of 37–48% (41,52). CT is still used by some centres to diagnose CSVT, but it requires ionising radiation and provides a false negative in half of the cases (46). MRI allows more accurate detection of CSVT and associated parenchymal injury and is considered the gold standard when combined with MR venography. In some cases, the thrombus can already be detected using conventional T1- and T2-weighted imaging (Fig. 3B). MRI may show additional white matter injury due to periventricular congestion, especially in the frontal regions (Fig. 3C). DWI will show restricted diffusion in these areas. In preterm born infants, the white matter injury is known to be more extensive and may result in cystic periventricular leukomalacia (49).
A
B
C
D
Figure 3 Cranial ultrasound and magnetic resonance imaging (MRI) findings of an infant born at term who presented with seizures on day 10 after birth. Cranial ultrasound (A) shows a right-sided IVH and an echodensity in the right thalamus, suggestive of a haemorrhage. Magnetic resonance imaging (MRI) confirmed the presence of a CSVT, showing a large thrombus in the straight sinus (B). This had led to venous congestion resulting in additional white matter injury. This additional injury can be clearly visualised using DWI, showing restricted diffusion in the frontal (so-called iris sign) and parietal white matter predominantly seen in the right hemisphere (C,D).
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
361
Neonatal stroke
van der Aa et al.
Using MR venography, the venous system can be visualised without the use of contrast. Some caution is needed, however, as lower venous flow, a smaller calibre venous sinus and skin moulding can also cause a flow gap in the venous sinuses. These findings should, therefore, be interpreted in the context of additional MR findings and the clinical condition of the infant. Several studies have reported the highest incidence of thrombosis in the superior sagittal sinus, followed by the transverse and straight sinuses (42,44,46), although a more recent study showed involvement of the straight sinus in the majority of the described cases (49). Involvement of multiple sinuses is common and is found in up to 70% of cases (41,44). Like PAIS, information on the coagulation status should be obtained, including the evaluation of protein-based assays shortly after diagnosis (16). Therapy Cerebral sinovenous thrombosis therapy should primarily be aimed at treating any underlying cause that may have led to the development of the thrombus. This includes treatment of dehydration, sepsis, meningitis and significant congenital heart disease. Studies in infants without CSVT show that in the supine position, compression of the occipital bone occurs, which may result in a reduced cerebral venous flow. Adjusting the infant’s pillow decompresses the occipital bone and has been shown to increase flow in the sigmoid and superior sagittal sinus and may offer a noninvasive therapy for infants with CSVT (48). However, this has not yet been studied in this population, and its efficacy still needs to be proven. There is ongoing discussion about whether anticoagulants should be used to prevent propagation of the initial thrombus and to enhance recanalisation. In the absence of randomised control trials, evidence for the use of anticoagulants is limited to case series. Withholding treatment does, however, seem to be associated with an increased risk of thrombus propagation of about 25% (53). Even in the presence of a thalamic haemorrhage, anticoagulation therapy appears to be safe (54). That is why a recent guideline issued by the American College of Chest Physicians recommends administering unfractionated heparin or low molecular weight heparin to infants without an intracranial haemorrhage, followed by low molecular weight heparin for 6–12 weeks (29). In the presence of a significant haemorrhage, anticoagulation therapy may be initiated, but a conservative approach with supportive care and radiological monitoring of the thrombosis at 5–7 days may also be chosen. If propagation of the thrombus is noted at that time, anticoagulation therapy may still be initiated. Neurodevelopmental outcome There are fewer studies on outcome following CSVT than on PAIS, and reported outcomes vary. Death rates tend to be low, in the order of 2–5% (44,45). Higher mortality rates of 19% and 25% have also been reported in two studies, which might be explained by better identification of CSVT cases and differences in the redirection of care (41,55).
362
Long-term outcome studies suggest that 60–80% of all infants develop some kind of deficit (44,45). Motor impairments, including cerebral palsy, are observed in 20–80%. Most children will be able to walk, however, especially in the absence of additional ischaemic injury (56). Describing the cognitive outcome following CSVT is hindered by a lack of the use of standardised assessments and the use of relatively coarse outcome measures, such as ‘moderately abnormal’ and ‘abnormal’ outcome, often performed at a young age. However, available data do show that cognitive outcome is abnormal in 25–73% of all cases (44,49,52). Little is known on visual deficits, although Berfelo et al. (41) reported strabismus and uncoordinated eye movements in 8% of survivors. Postneonatal epilepsy occurs in 15–40% of all infants (42,44,54), and infants who have associated thalamic injury face a particular risk of developing epilepsy. In these cases, an electrical status epilepticus in slow wave sleep is often observed, which may be challenging to treat and is associated with poor cognitive outcome (57).
FUTURE DIRECTIONS During the last decades, considerable progress has been made in the field of perinatal stroke, with regard to epidemiology, potential risk factors, neuroimaging and subsequent outcome. Despite this progress, the pathogenesis remains a black box and therapies are still lacking. One difficulty in studying perinatal stroke is the low incidence of both PAIS and CSVT. Larger number of patients will be required to better understand the pathophysiological mechanisms and to study potential therapies. This may be reached by both better registration and by multicentre collaborations, such as the International Paediatric Stroke Study (13). Such research networks allow adequately powered case–control studies and enable a setting for studying future therapies. Additional insights may come from the application of novel imaging techniques, such as resting state MRI, DTI and serial imaging. While numerous studies have reported the structural changes on MRI following perinatal brain injury, few have studied functional changes in terms of functional connectivity. One study in term born infants with hypoxic ischaemic encephalopathy did indeed show a trend towards declining brain network integration in infants with increasing neuromotor deficits (58). Further studies should point out whether connectivity parameters can indeed be related to functional outcome following perinatal brain injury. This might be particularly interesting in relation to cognitive outcome (59). While several therapies have shown to be effective in experimental stroke studies, their effectiveness and applicability in PAIS still need to be confirmed. The difficulty in treating PAIS is that the most optimal therapeutic time window for starting treatment during the acute phase of hypoxic ischaemic injury is during the first 6 h, while infants with PAIS often present with seizures beyond the first 24 h (14). It is, therefore, most likely that therapies
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
Neonatal stroke
van der Aa et al.
aiming at regeneration and remodelling, which can be applied in the postacute phase, will be most efficient. Such therapies include stem cells, erythropoietin, repetitive TMS, constraint-induced movement therapy (CIMT) and handarm bimanual intensive therapy (30,32,33). Repetitive TMS and CIMT have been applied in older children following PAIS, but it has been suggested that these therapies will be more efficient when applied during a phase when the balance between ipsilateral and contralateral corticospinal tracts is still being established, which is until the age of 12– 18 months. In CSVT, randomised controlled trials are needed to determine the role of anticoagulation therapy.
References 1. Raju TN, Nelson KB, Ferriero D, Lynch JK. Ischemic perinatal stroke: summary of a workshop sponsored by the National Institute of Child Health and Human Development and the National Institute of Neurological Disorders and Stroke. Pediatrics 2007; 120: 609–16. Epub 2007/09/04. 2. Mineyko A, Kirton A. The black box of perinatal ischemic stroke pathogenesis. J Child Neurol 2011; 26: 1154–62. Epub 2011/06/15. 3. Kirton A, Deveber G, Pontigon AM, Macgregor D, Shroff M. Presumed perinatal ischemic stroke: vascular classification predicts outcomes. Ann Neurol 2008; 63: 436–43. Epub 2008/ 02/29 4. Lee J, Croen LA, Backstrand KH, Yoshida CK, Henning LH, Lindan C, et al. Maternal and infant characteristics associated with perinatal arterial stroke in the infant. JAMA 2005; 293: 723–9. Epub 2005/02/11 5. Laugesaar R, Kolk A, Tomberg T, Metsvaht T, Lintrop M, Varendi H, et al. Acutely and retrospectively diagnosed perinatal stroke: a population-based study. Stroke 2007; 38: 2234–40. Epub 2007/06/23. 6. Benders MJ, Groenendaal F, Uiterwaal CS, Nikkels PG, Bruinse HW, Nievelstein RA, et al. Maternal and infant characteristics associated with perinatal arterial stroke in the preterm infant. Stroke 2007; 38: 1759–65. Epub 2007/05/15. 7. Golomb MR, Fullerton HJ, Nowak-Gottl U, Deveber G, International Pediatric Stroke Study G. Male predominance in childhood ischemic stroke: findings from the international pediatric stroke study. Stroke 2009; 40: 52–7. Epub 2008/09/ 13. 8. Wu YW, March WM, Croen LA, Grether JK, Escobar GJ, Newman TB. Perinatal stroke in children with motor impairment: a population-based study. Pediatrics 2004; 114: 612–9. Epub 2004/09/03. 9. Harteman JC, Groenendaal F, Kwee A, Welsing PM, Benders MJ, de Vries LS. Risk factors for perinatal arterial ischaemic stroke: a case–control study. Arch Dis Child Fetal Neonatal Ed 2012; 97: F411–6. 10. Gunther G, Junker R, Strater R, Schobess R, Kurnik K, Heller C, et al. Symptomatic ischemic stroke in full-term neonates: role of acquired and genetic prothrombotic risk factors. Stroke 2000; 31: 2437–41. Epub 2000/10/07. 11. Mercuri E, Cowan F, Gupte G, Manning R, Laffan M, Rutherford M, et al. Prothrombotic disorders and abnormal neurodevelopmental outcome in infants with neonatal cerebral infarction. Pediatrics 2001; 107: 1400–4. Epub 2001/06/05. 12. McQuillen PS, Miller SP. Congenital heart disease and brain development. Ann N Y Acad Sci 2010; 1184: 68–86. Epub 2010/02/12.
13. Kirton A, Armstrong-Wells J, Chang T, Deveber G, Rivkin MJ, Hernandez M, et al. Symptomatic neonatal arterial ischemic stroke: the international pediatric stroke study. Pediatrics 2011; 128: e1402–10. Epub 2011/11/30. 14. Rafay MF, Cortez MA, de Veber GA, Tan-Dy C, Al-Futaisi A, Yoon W, et al. Predictive value of clinical and EEG features in the diagnosis of stroke and hypoxic ischemic encephalopathy in neonates with seizures. Stroke 2009; 40: 2402–7. Epub 2009/ 05/30. 15. Mercuri E, Rutherford M, Cowan F, Pennock J, Counsell S, Papadimitriou M, et al. Early prognostic indicators of outcome in infants with neonatal cerebral infarction: a clinical, electroencephalogram, and magnetic resonance imaging study. Pediatrics 1999; 103: 39–46. Epub 1999/01/26. 16. Nowak-Gottl U, Duering C, Kempf-Bielack B, Strater R. Thromboembolic diseases in neonates and children. Pathophysiol Haemost Thromb 2003; 33: 269–74. Epub 2005/ 02/05. 17. Cowan F, Mercuri E, Groenendaal F, Bassi L, Ricci D, Rutherford M, et al. Does cranial ultrasound imaging identify arterial cerebral infarction in term neonates? Arch Dis Child Fetal Neonatal Ed 2005; 90: F252–6. Epub 2005/04/23. 18. van der Aa NE, Dudink J, Benders MJNL, Govaert P, van Straaten HL, Porro GL, et al. Neonatal posterior cerebral artery stroke: clinical presentation, MRI findings and outcome. Dev Med Child Neurol 2013; 55: 283–90. 19. Dudink J, Mercuri E, Al Nakib L, Govaert P, Counsell SJ, Rutherford MA, et al. Evolution of unilateral perinatal arterial ischemic stroke on conventional and diffusion-weighted MR imaging. AJNR Am J Neuroradiol 2009; 30: 998–1004. 20. van der Aa NE, Benders MJ, Vincken KL, Groenendaal F, de Vries LS. The course of apparent diffusion coefficient values following perinatal arterial ischemic stroke. PLoS ONE 2013; 8: e56784. 21. Mercuri E, Barnett A, Rutherford M, Guzzetta A, Haataja L, Cioni G, et al. Neonatal cerebral infarction and neuromotor outcome at school age. Pediatrics 2004; 113: 95–100. 22. de Vries LS, van der Grond J, van Haastert IC, Groenendaal F. Prediction of outcome in new-born infants with arterial ischaemic stroke using diffusion-weighted magnetic resonance imaging. Neuropediatrics 2005; 36: 12–20. 23. Kirton A, Shroff M, Visvanathan T, deVeber G. Quantified corticospinal tract diffusion restriction predicts neonatal stroke outcome. Stroke 2007; 38: 974–80. 24. Roze E, Harris PA, Ball G, Elorza LZ, Braga RM, Allsop JM, et al. Tractography of the corticospinal tracts in infants with focal perinatal injury: comparison with normal controls and to motor development. Neuroradiology 2012; 54: 507–16. Epub 2011/10/19. 25. van der Aa NE, Leemans A, Northington FJ, van Straaten HL, van Haastert IC, Groenendaal F, et al. Does diffusion tensor imaging-based tractography at 3 months of age contribute to the prediction of motor outcome after perinatal arterial ischemic stroke? Stroke 2011; 42: 3410–4. Epub 2011/10/25. 26. van der Aa NE, Porsius ED, Hendrikse J, van Kooij BJ, Benders MJ, de Vries LS, et al. Changes in carotid blood flow after unilateral perinatal arterial ischemic stroke. Pediatr Res 2012; 72: 50–6. Epub 2012/03/27. 27. Groenendaal F, van der Grond J, Witkamp TD, de Vries LS. Proton magnetic resonance spectroscopic imaging in neonatal stroke. Neuropediatrics 1995; 26: 243–8. Epub 1995/10/01. 28. De Vis JB, Petersen ET, Kersbergen KJ, Alderliesten T, de Vries LS, van Bel F, et al. Evaluation of perinatal arterial ischemic stroke using noninvasive arterial spin labeling perfusion MRI. Pediatr Res 2013; 74: 307–13. Epub 2013/06/26.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
363
Neonatal stroke
van der Aa et al.
29. Monagle P, Chan AK, Goldenberg NA, Ichord RN, Journeycake JM, Nowak-Gottl U, et al. Antithrombotic therapy in neonates and children: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141: e737S–801S. Epub 2012/02/15. 30. Barnette AR, Inder TE. Evaluation and management of stroke in the neonate. Clin Perinatol 2009; 36: 125–36. Epub 2009/ 01/24. 31. Benders MJ, van der Aa NE, Roks M, van Straaten HL, Isgum I, Viergever MA, et al. Feasibility and safety of erythropoietin for neuroprotection after perinatal arterial ischemic stroke. J Pediatr 2013; pii: S0022-3476(13)01380-2. doi: 10.1016/j. jpeds.2013.10.084. [Epub ahead of print]. 32. Aarts PB, Jongerius PH, Geerdink YA, van Limbeek J, Geurts AC. Effectiveness of modified constraint-induced movement therapy in children with unilateral spastic cerebral palsy: a randomized controlled trial. Neurorehabil Neural Repair 2010; 24: 509–18. Epub 2010/04/29. 33. Kirton A, Chen R, Friefeld S, Gunraj C, Pontigon AM, Deveber G. Contralesional repetitive transcranial magnetic stimulation for chronic hemiparesis in subcortical paediatric stroke: a randomised trial. Lancet Neurol 2008; 7: 507–13. Epub 2008/ 05/06. 34. Mercuri E, Anker S, Guzzetta A, Barnett A, Haataja L, Rutherford M, et al. Neonatal cerebral infarction and visual function at school age. Arch Dis Child Fetal Neonatal Ed 2003; 88: F487–91. Epub 2003/11/07. 35. Guzzetta A, Pizzardi A, Belmonti V, Boldrini A, Carotenuto M, D’Acunto G, et al. Hand movements at 3 months predict later hemiplegia in term infants with neonatal cerebral infarction. Dev Med Child Neurol 2010; 52: 767–72. Epub 2009/10/30. 36. Wanigasinghe J, Reid SM, Mackay MT, Reddihough DS, Harvey AS, Freeman JL. Epilepsy in hemiplegic cerebral palsy due to perinatal arterial ischaemic stroke. Dev Med Child Neurol 2010; 52: 1021–7. Epub 2010/05/26. 37. Ballantyne AO, Spilkin AM, Hesselink J, Trauner DA. Plasticity in the developing brain: intellectual, language and academic functions in children with ischaemic perinatal stroke. Brain 2008; 131: 2975–85. 38. Westmacott R, Macgregor D, Askalan R, deVeber G. Late emergence of cognitive deficits after unilateral neonatal stroke. Stroke 2009; 40: 2012–9. 39. van Buuren LM, van der Aa NE, Dekker HC, Vermeulen RJ, van Nieuwenhuizen O, van Schooneveld MM, et al. Cognitive outcome in childhood after unilateral perinatal brain injury. Dev Med Child Neurol 2013; 55: 934–40. 40. Ricci D, Mercuri E, Barnett A, Rathbone R, Cota F, Haataja L, et al. Cognitive outcome at early school age in term-born children with perinatally acquired middle cerebral artery territory infarction. Stroke 2008; 39: 403–10. Epub 2007/12/29. 41. Berfelo FJ, Kersbergen KJ, van Ommen CH, Govaert P, van Straaten HL, Poll-The BT, et al. Neonatal cerebral sinovenous thrombosis from symptom to outcome. Stroke 2010; 41: 1382– 8. Epub 2010/06/05. 42. deVeber G, Andrew M, Adams C, Bjornson B, Booth F, Buckley DJ, et al. Cerebral sinovenous thrombosis in children. N Engl J Med 2001; 345: 417–23. Epub 2001/08/11. 43. Raets MM, Sol JJ, Govaert P, Lequin MH, Reiss IK, Kroon AA, et al. Serial Cranial US for Detection of Cerebral Sinovenous Thrombosis in Preterm Infants. Radiology 2013; 269: 879–86. Epub 2013/08/30.
364
44. Fitzgerald KC, Williams LS, Garg BP, Carvalho KS, Golomb MR. Cerebral sinovenous thrombosis in the neonate. Arch Neurol 2006; 63: 405–9. Epub 2006/03/15. 45. Moharir MD, Shroff M, Pontigon AM, Askalan R, Yau I, Macgregor D, et al. A prospective outcome study of neonatal cerebral sinovenous thrombosis. J Child Neurol 2011; 26: 1137–44. Epub 2011/06/02. 46. Nwosu ME, Williams LS, Edwards-Brown M, Eckert GJ, Golomb MR. Neonatal sinovenous thrombosis: presentation and association with imaging. Pediatr Neurol 2008; 39: 155–61. Epub 2008/08/30. 47. Wu YW, Miller SP, Chin K, Collins AE, Lomeli SC, Chuang NA, et al. Multiple risk factors in neonatal sinovenous thrombosis. Neurology 2002; 59: 438–40. Epub 2002/08/15. 48. Tan MA, Miller E, Shroff MM, Deveber GA, Kirton A. Alleviation of neonatal sinovenous compression to enhance cerebral venous blood flow. J Child Neurol 2013; 28: 583–8. Epub 2012/07/19. 49. Kersbergen KJ, Groenendaal F, Benders MJ, van Straaten HL, Niwa T, Nievelstein RA, et al. The spectrum of associated brain lesions in cerebral sinovenous thrombosis: relation to gestational age and outcome. Arch Dis Child Fetal Neonatal Ed 2011; 96: F404–9. Epub 2011/02/15. 50. Miller E, Daneman A, Doria AS, Blaser S, Traubici J, Jarrin J, et al. Color Doppler US of normal cerebral venous sinuses in neonates: a comparison with MR venography. Pediatr Radiol 2012; 42: 1070–9. Epub 2012/04/26. 51. Wu YW, Hamrick SE, Miller SP, Haward MF, Lai MC, Callen PW, et al. Intraventricular hemorrhage in term neonates caused by sinovenous thrombosis. Ann Neurol 2003; 54: 123–6. Epub 2003/07/03. 52. Grunt S, Wingeier K, Wehrli E, Boltshauser E, Capone A, Fluss J, et al. Cerebral sinus venous thrombosis in Swiss children. Dev Med Child Neurol 2010; 52: 1145–50. Epub 2010/06/25. 53. Moharir MD, Shroff M, Stephens D, Pontigon AM, Chan A, MacGregor D, et al. Anticoagulants in pediatric cerebral sinovenous thrombosis: a safety and outcome study. Ann Neurol 2010; 67: 590–9. Epub 2010/05/04. 54. Kersbergen KJ, de Vries LS, van Straaten HL, Benders MJ, Nievelstein RA, Groenendaal F. Anticoagulation therapy and imaging in neonates with a unilateral thalamic hemorrhage due to cerebral sinovenous thrombosis. Stroke 2009; 40: 2754–60. Epub 2009/06/23. 55. Wasay M, Dai AI, Ansari M, Shaikh Z, Roach ES. Cerebral venous sinus thrombosis in children: a multicenter cohort from the United States. J Child Neurol 2008; 23: 26–31. Epub 2008/ 01/11. 56. Golomb MR, deVeber GA, MacGregor DL, Domi T, Whyte H, Stephens D, et al. Independent walking after neonatal arterial ischemic stroke and sinovenous thrombosis. J Child Neurol 2003; 18: 530–6. Epub 2003/09/19. 57. Kersbergen KJ, de Vries LS, Leijten FS, Braun KP, Nievelstein RA, Groenendaal F, et al. Neonatal thalamic hemorrhage is strongly associated with electrical status epilepticus in slow wave sleep. Epilepsia 2013; 54: 733–40. Epub 2013/03/20. 58. Tymofiyeva O, Hess CP, Ziv E, Tian N, Bonifacio SL, McQuillen PS, et al. Towards the “baby connectome”: mapping the structural connectivity of the newborn brain. PLoS ONE 2012; 7: e31029. Epub 2012/02/22. 59. van den Heuvel MP, Stam CJ, Kahn RS, Hulshoff Pol HE. Efficiency of functional brain networks and intellectual performance. J Neurosci 2009; 29: 7619–24. Epub 2009/06/12.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 356–364
