The physical examination at admission was dominated by hemiplegia (93.3%), aphasia (73.3%) and seizures (53.3%). The neurological signs observed are presented in Table 1. Other signs related to the underlying pathologies were found. These include heart rhythm disturbances (40%), cyanosis (13.3%), pallor (33.3%) and splenomegaly (33.3%).

For the positive diagnosis of stroke, the brain scan could only be carried out in 5 cases, i.e. a completion rate of 33.3% and it was an ischemic stroke in all 5 cases. It showed carotid involvement in three cases, of which two were left and one right; one occurred in the right vertebro-basilar territory; and one case was of mixed involvement (vertebro-basilar and carotid). The etiological research made it possible to carry out cardiovascular assessments including cardiac ultrasound and electrocardiogram with respective completion rates of 60% and 53.3%. Cardiac ultrasound revealed 2 cases of tetralogy of Fallot. The ECG revealed the existence of complications such as heart rhythm disturbances in a case of mitral disease.

The etiologies highlighted were mainly thromboembolic complications in the context of pathologies such as sickle cell disease (40%), heart disease (30%), corticosteroid-resistant nephrotic syndrome with significant hypoalbuminemia (6.7%) and post-varicella angiitis (6.7%). The etiologies highlighted are listed in Table 2. The stroke made it possible to discover the unknown underlying

conditions in 5 patients including 3 cases of sickle cell anemia (20%) and 2 cases of congenital heart disease such as tetralogy of Fallot (13.3%).

Regarding the treatment, functional rehabilitation was practiced in 80% of patients by the physiotherapy service. In 2 patients (13.3%), the regression of the motor deficit was spontaneous. Other treatments were administered for either symptomatic or etiological management in certain cases. These included parenteral rehydration by ringer lactate infusions associated or not with abundant drinks (40%) and transfusion with packed red blood cells (20%) in sickle cell patients, heparin therapy (13.3%) and corticosteroid therapy (6.7%) for cases of nephrotic syndrome and anticonvulsants based on phenobarbital for cases of convulsions (26.7%) (See Table 3).

The evolution was marked by the occurrence of death in the acute phase in a sickle cell patient due to an infectious complication, and by an improvement on the neurological level with complete regression of the motor deficit for the other 14 cases (93 .3%) over a 3-month follow-up.

Stroke is a rare pathological entity in children [7] , it is 10 to 20 times less common than in adults [5] [6] . Also in our series, the hospital frequency is only 1.5 cases per year. The incidence of childhood stroke varies greatly from one study to another. In a study conducted in Congo by Tohodjede Y et al. in 2018 [8] , 16 cases were found over 10 years, while in 2015 Ndiaye M et al. in Senegal reported 73 cases over 10 years [9] . Sounga Bandzouzi P.E. et al. in 2021 in Pointe-Noire (Congo) listed 17 cases over a period of 35 months [10] . These dissimilarities between the incidences are due to methodological differences such as the multicenter nature or not, the retrospective or prospective nature and the means used for confirmation of the stroke.

The average age of stroke onset in our series is 4 years, the age group of 3 to 5 years being the most represented with 60% of cases. Bartolome AM, Pontigon

AM. and Moharir M. found two frequency peaks, one between 2 and 3 years and the other between 10 and 11 years [11] . Sounga Bandzouzi P et al. found in their study an average age of 11 years with the extremes of 7 to 15 years [10] . For the study by Tohodjede Y. et al. the average age of the patients was 8 years with extremes of 1 and 15 years [8] . These different results show that all ages can be affected by stroke. A male predominance is noted in most studies. In our series, this predominance was highlighted with a sex ratio of 1.5. Bournoussi W [12] in Morocco found 75% boys, i.e. a sex ratio of 3. A sex ratio of 2.2 was reported in the studies by Ndiaye in Senegal [9] and Ouchen in Morocco [13] .

The majority of our patients (73.3%) had a chronic pathology with little or no follow-up. In our context of a country with limited resources, patient monitoring poses problems due to the financial cost it involves (consultation costs, additional tests and medications while medical care coverage is exceptional). The underlying pathologies found in our patients were represented by sickle cell disease (40%), acquired or congenital heart disease (20%), HIV infection (13.3%) and chronic kidney disease (6.7%). Our results are consistent with the risk factors for ischemic stroke noted by other authors in the literature, namely cardiac causes, hematological causes (anemia and sickle cell anemia) and infectious causes such as retroviral infection [10] [14] [15] .

Several authors report a predominance of hematological causes, notably sickle cell disease and anemia. Indeed, sickle cell disease is the most common cause of ischemic stroke in children with a risk approximately 300 times greater than in children of the same age without sickle cell disease. It causes cerebral vasculopathy [16] [17] . Heart disease represents a significant cause of stroke in children, it is found in 25% of cases [18] . Indeed, some of our patients (20%) had heart disease as other authors have reported [8] [9] [12] [13] . During our study, 13.3% of our children were HIV positive. As the literature says, HIV is among the viral infections that can cause cerebral angiitis responsible for cerebral infarction [19] . A specific type of cerebral vasculopathy characterized by arterial ectasia and aneurysmal formation was first recognized in HIV-infected children and young adults as a cause of cerebral infarction and cerebral hemorrhage [20] [21] . Other bacterial or viral infections in children can be complicated by cerebral angiitis responsible for cerebral infarction, the most common of which at this age is post-varicella angiitis which is characterized by necrotizing arterial damage [22] . In our patients, the stroke occurred after chickenpox in one case.

Nephrotic syndrome was one of the etiologies of childhood stroke implicated in our series (6.7%). Indeed, a risk of thromboembolic complications exists in the case of nephrotic syndrome but its incidence is far lower in the pediatric population [23] . It would be around 3% [24] .

At admission, hemiplegia was the dominant sign (i.e. in 93.3% of cases) in our study. It was associated or not with other signs such as aphasia (73.3%), convulsions (40%) and headaches (13.3%). The frequency of hemiplegia is noted by other authors [6] [10] [24] at 62%, 53% and 82.6% respectively. The same signs of stroke were also noted by Souga et al. [10] but in proportions different from ours (hemiplegia, language disorder, convulsions and headaches), respectively in 53%, 53%; 11.7% and 29.4% of cases. Barnoussi in Morocco in 2014 noted in his study of 8 cases, aphasia in 3 children and an altered state of consciousness with convulsive seizures in 2 other patients [12] . Concerning the positive diagnosis of stroke, the brain scan could only be performed in 33.3% of cases in our series. The diagnosis of stroke was based on anamnestic, clinical and evolutionary arguments for other cases. This low rate is linked to the high cost of this examination in a country where the average household income is low. But these were cases of ischemic stroke in all cases. Sounga’s study also noted a predominance of ischemic stroke [10] . In Africa, several authors report a predominance of ischemic stroke and sickle cell disease remains the dominant etiology in the majority of cases [8] [14] [15] . But the literature and studies carried out on other continents highlight the predominance of hemorrhagic strokes in children where arteriovenous malformations are the main cause in 14% to 57% [22] .

The management of chronic conditions is not easy in countries with limited resources. Still, socio-economic aspects can modulate the prognosis of chronic conditions by impacting their management [25] . In our study, it was not possible to determine by brain scan whether there was no case of hemorrhagic stroke. In the literature, children with ischemic stroke have a better prognosis than those with hemorrhagic stroke [6] . In Sounga’s study, the majority of patients had fully recovered from their motor deficit and had no cognitive after-effects, but no neuropsychological test was carried out [10] . Under normal conditions, a child with a chronic illness requires regular and specialized monitoring involving numerous medical or paramedical specialists and the multidisciplinary consultation for these children with chronic illnesses has among other objectives the prevention and detection of complications or disorders associated with the chronic pathology [25] [26] . In our series, all patients had recovered from their neurological deficit over a three-month follow-up except for the case of death which occurred in the acute phase. But the cognitive side had not been evaluated nor the search for recurrences beyond 3 months.