11.1 Introduction and Epidemiology

It is a real challenge to write a chapter on pediatric stroke, given that available information, experience, and scientific evidence are quite limited. As opposed to the approach to stroke in the adult population, the approach to stroke in childhood often lacks a clear scientific outline, being based on small case series and reports of isolated cases. There have, however, been reports of ischemic stroke cases in children as far back as the eighteenth century.

Approaching this entity as a single condition is always a problem. It should be envisaged differently for each stage, because causes, symptoms and prognoses are different at each stage of childhood. As opposed to general opinion, pediatric stroke is among the top 10 causes of death in children in the United States. ¹ It occurs much oftener than is generally thought, incidence being between 2, 3 to 13 in 100,000 children. From a historic point of view, this incidence is on the rise, perhaps due to more awareness of the problem (and subsequently, more specialists suspecting this diagnosis) and also to the fact that patients at risk for stroke live longer now. ^{2 ,​ 3} Another important contributing factor is progress in imaging techniques that diagnose stroke cases. Some of these techniques are increasingly common, like MRI.

Unfortunately it looks like pediatric stroke is still underestimated in spite of its frequency. In most of the cases that have come our way we have observed a low level of suspicion, with the subsequent delay in diagnosis that goes beyond 24 hours, as will be discussed later.

Stroke is so frequent that it even competes with tumors of the central nervous system in pediatrics for the first places in incidence lists. Although in most case reports on this subject patients are older than 28 days, it is 17 times more frequent in newborn period; in this population, a stroke is especially likely to occur.

As opposed to the situation in the adult population, AIS incidence is slightly higher than the incidence of acute hemorrhagic stroke. There is a clear male predominance, approximately 60%. There are clear epidemiological ethnic differences that have not been well defined so far. ⁴

On facing a pediatric patient with AIS, marked differences with adult cases appear. That is why we have decided to write a separate chapter about this group. Not only is AIS more frequent than hemorrhagic stroke, there is also a noticeable arterial-venous ratio of 3/1 as a cause for ischemic stroke in children. The ratio in newborns is 2/1. ⁵ Venous pathology, then, plays a great role in this often-forgotten disease of children. Nonspecific symptoms associated with stroke must ALWAYS awaken suspicion of this disease in a child with nonspecific neurological symptomatology.

11.2 Clinical Symptoms

The clinical manifestations of stroke in childhood are clearly linked to the corresponding stage in the child’s life. It becomes difficult, therefore, to organize this complex subject.

Cases of intrauterine stroke have long been known and reported ⁶ but always as isolated cases with no evident symptoms, which were diagnosed on account of clinical manifestations days or weeks after birth or as autopsy findings in children who died soon after birth.

The symptoms may appear suddenly or gradually, they may associate neurological impairment or not, they may include seizures, etc., according not only to the cause but to the age of the patient.

In the newborn period, ischemic stroke tends to present with seizure and rarely, if ever, with any noticeable focal neurological deficit. The onset is usually insidious, which is not the case with older children. Focal neurological deficits become more apparent from a few monthsto a year later.

According to some reports, evident signs of neurological deficit that are related to perinatal or neonatal brain ischemia may be delayed. In 18 out of 22 cases, lack of use of one hand is related to probable perinatal ischemia, while in 12 out of 22 cases, permanent alterations of language, cognition or behavior develop during a long-term follow-up. ⁷

In children older than one year, early and unusual dominance of one hand may indicate a previous stroke. The disease may still appear associated with seizures, fever or coma. This symptomatology becomes less frequent as children get older. Adolescents present more regularly with acute-onset hemiparesia, with or without seizures, a clinical picture that often regresses. Fever, seizures, and headaches are less frequent. ⁸

Stage of childhood is not the only determining cause; other factors and the affected territory also play a role in the presentation of symptoms. In the case of ischemic arterial strokes, the affectation of proximal arteries in the anterior sector is better tolerated than in adults, perhaps because vascular anastomotic circuits, intracranial and extracranial, are notoriously more present. The incidence of such anterior strokes is supposed to reach 71% of cases in children. It should be underlined that in 8% of cases both anterior territories are associated, ⁹ which is a remarkable phenomenon.

Venous strokes are apt to have a more insidious, progressive and unspecific onset, and they are associated to vague symptomatology. Intracranial hypertension may be associated. Focal symptoms are not always present, and they depend on the affected territory, which is nearly impossible to diagnose by clinical means only. Some symptomatologic associations linked to brain venous thromboses have been described, but the latter do not always result in symptoms, because of the particular complexity of brain anatomy and venous hemodynamics in children.

Transitory ischemic attacks do not present with lipohyalinosis as in adults; if they occur at all, they are associated with hemiplegic migraine or other manifestations of brain insufficiency due to moyamoya disease (which increases brain metabolic requirements) or to arteriovenous fistulae originating a local circulatory steal phenomenon.

11.3 Etiology

The causes of ischemic stroke in childhood are invariably manifold and not well-known. In most cases (80%), however, a specific cause for stroke can be found if the case is adequately studied.

The study of risk factors is most important in these cases in order to understand and supervise the great amount of lab tests and imaging that will be suggested.

According to the Canadian Pediatric Ischemic Stroke Registry (2017), nearly half the cases of stroke develop in previously healthy children, but in newborns, strokes are part of an acute systemic disease (infectious or not), arise in prothrombotic conditions, or are a consequence of maternal disorders. Outside the newborn period, other more specific pediatric factors come into play, like arteriopathy (50%), congenital or acquired heart disease (24%), and prothrombotic or hematological disorder (20–50%). There may be a previous trauma, especially in cases of arterial dissection and also in some venous cerebral thromboses. ^{10 ,​ 11}

Venous conditions usually are associated with infections, dehydration, head trauma or congenital prothrombotic states. In sum, 50% of the cases associate with a risk factor, 75% with more than one and in 25% no risk factors are to be found. ⁴

In order to diagnose correctly, the fundamental steps are: a painstaking case history, a careful physical examination (it must be borne in mind, however, that this is a neurological emergency), then specific imaging, adequate cardiological assessment and a complete study of coagulation. In children it is especially important to differentiate conditions of venous etiology from those of arterial etiology, as well as thrombotic disorders from embolic disorders, as shown in table Table 11‑1

Congenital cyanotic cardiopathies are a common cause of stroke in children, on account of their compensating polycytemia’s determining both local thrombosis (arterial or venous) and arterial embolism. Acquired cardiopathies are also a relatively frequent cause of embolism in brain arteries. This applies to both valvular disease like rheumatic disease, mitral prolapse, auricular myxoma, valvular prostheses and myocardial disease, like myocardiopathies and arrhythmias.

Traumatic intracranial arterial conditions, classically considered infrequent, deserve a whole new chapter. They are thought to be one of the leading causes of stroke in childhood, particularly in older children. ¹² Basically, dissections can be divided into intracranial or extracranial, and may in turn be either subintimal or subadventitial. Through a lesion of the intima of a cerebral artery blood enters between the intima and the media layer. As a consequence, the arterial lumen becomes narrower, and blood flow is reduced. Distal embolism may arise from that lesion. Once blood forces its way through the whole media layer, the dissection becomes adventitial, and there is risk of both ischemic and hemorrhagic stroke; those are extremely severe cases with high morbimortality.

Arterial dissections have always been linked to previous trauma, but we often see in practice that this event does not appear clearly in the case history and therefore stroke goes unsuspected and undiagnosed, with the subsequent loss of time while investigating wrong etiologies.

It must be remembered, too, that a third of strokes in childhood occur in the context of infection, viral or bacterial. Arterial vasculitis can originate stroke in children. Numerous agents like nonspecific bacteria, tuberculous bacilli, Mycoplasma, Chlamydia, herpesvirus, HIV and fungus can cause arterial vasculitis when they invade the central nervous system (CNS). Meningitis often associates venous thrombosis of cortical and deep communicant veins, while HNT infections tend to affect the transverse sinuses, the sphenoparietal sinuses of Brechet and the cavernous sinus that run through the region.

Nevertheless vasculitis is not a particularly frequent cause of ischemic stroke in children, as opposed to what is generally assumed.

Hematological diseases may cause brain infarcts, either arterial or venous. Both acquired hematological disorders (like disseminated erythematosus lupus and iatrogenic diseases) and congenital hematological conditions (such as hereditary deficiency of coagulation inhibitors, such as proteins C or S or antithrombin III, increase of clotting proteins or mutations of the prothrombin gene) increase the possibility of ischemic stroke on account of hypercoagulability or a prethrombotic state. They constitute a clear risk for stroke in childhood.

Nearly 4% of the population has a mutation of the prothrombin gene (20210) and 2 to 15% of the white population evidences a resistance to natural clotting of activated protein C, which is associated with a mutation of the factor V Leiden. Both circumstances dramatically increase the occurrence of thrombosis in that population. In these prothrombotic cases, both veins and arteries of the brain are diseased. ¹³

Hyperhomocysteinemia is a well-known risk factor for stroke in children. The mechanism involves a C677T mutation that produces a 5–10 methytetrahydrofolate reductase (MTHFR) variant, which in turn decreases the amount of available 1.5-methyltetrahydrofolate for the conversion of homocystein into methionine.

An increase in lipoprotein (a) can result in prothrombotic states because it competes with plasminogen; it inhibits fibrinolysis and increases the risk of thrombosis.

Deficiency of proteins C and S may be acquired, for example by sepsis and viral infections like varicella.

Other prothrombotic abnormalities less frequently associated with ischemic stroke are thrombocytosis, dysfibrinogenemia and increase of factor VIII. ¹³

Such prothrombotic alterations may coincide with other events favoring thrombosis (arterial or venous manipulation, infections, sideropenic anemia), thus clearly boosting the risk for ischemic stroke.

The Registry of the Canadian Stroke Network evidences a clear increase of risk for patients with a mutation of factor V Leiden, high levels of lipoprotein (a) or a deficiency of protein C and antiphospholipid antibodies. ¹³ An association of these factors obviously heightens the risk of an ischemic stroke occurring in childhood.

The association of other pathological entities can also increase the risk for stroke in children. That is the case with falciform anemia, noninfectious immunologic vasculitis as part of a connective tissue disease (such as systemic lupus erythematosus, polyarteritis nodosa, Henoch-Schönlein purpura, Takayasu arteritis), fibromuscular dysplasia, metabolic diseases (such as ornithine transcarbamylase deficiency, methylmalonic aciduria and MELAS) and moyamoya disease. Detailed analysis of these diseases falls beyond the scope of this chapter.