From April 1, 2019 to April 30, 2022, we included 22 children who had a stroke out of 714 children with sickle cell disease followed in the pediatrics department of the Gabriel Toure university hospital center, or 3.08%.

The male gender represented 54.5%, or a sex ratio of 1.2. The age group of 1 to 5 years was the most affected with 40.9%, followed by 6 - 10 years with 36.4%. The average age of the patients was 7 years with extremes ranging from 1 year to 15 years and a median of 5 years. The mothers were not educated in 45.5% of cases, while the level of primary education represented 40.9% of cases among the fathers. Patients not living in Bamako were 41%.

Patients had a history of headache in 63.5% before the occurrence of clinical stroke. Stroke was the circumstance of discovery of sickle cell anemia in 50% of cases. Disturbance of consciousness was the reason for consultation in six cases (27%), followed by convulsions in 18% of cases (Table 1). The clinical signs (on admission) are dominated by loss of consciousness and hemiparesis with 45.5% each, followed by sensitivity disorders with 32% (Table 2). Pneumonitis was associated in three patients or 14% of cases.

We noted 95.5% SS form and 4.5% Sβ⁰thalassemia form. The hemoglobin level was less than 7 g/dl in 59% of cases with an average level of 6.7 g/dl ± 1.7. Hyperleukocytosis was present in 21 patients or 95.4% with an average rate of 20,063 white blood cells/mm³ ± 9.8.103/mm³ (Table 3). Transcranial Doppler ultrasound was performed in five patients before the occurrence of the stroke, two of whom were pathological. All our patients suffered from an ischemic stroke and the sylvian artery was affected in ten cases out of 22, or 45.5% (Table 4).

All our patients were transfused urgently, then a monthly transfusion program was put in place after their discharge. The rest of the treatment was based on hyperhydration (100%), hydroxycarbamide (86%), Physiotherapy (73%), oxygen therapy (45.5%). Patients stayed in hospital between 21 and 30 days in 36% of cases. The lethality was 22.7%. The survivors presented serious after-effects in 22.5% of cases and mild after-effects in 45.4% of cases. Hemiparesis was present as a sequela in 45.4% of cases at the time of hospital discharge (Figure 1). There

Others*: ptosis (1), hallucination (1) agitation (1), strabismus (1).

were 9% who experienced no after-effects. The patients were followed monthly to assess their progress and adapt their care.

During our study, we recorded 22 cases of stroke in 714 children with sickle cell disease, including nine retrospective cases and 13 prospective cases. The frequency of stroke in this population is 3.08%. This result is similar to that of Ouattara A [7] who found a frequency of 2.8% in 2010.

During this study, stroke occurred in both sexes, however the male sex was predominant with 54.5%, i.e. a sex ratio of 1.2. This result is close to that of Rakotoharimanana V C [8] , who found a sex ratio of 1.6. According to DeBaun MR et al. [9] male gender is a risk factor for microvasculopathy. The age group of 1 to 5 years was the most affected with 40.9% followed by 6 - 10 years with 36.4%. The average age of the patients was 7 years with extremes ranging from 1 year to 15 years and a median of 5 years. The predominance of this age group could be explained by the increase in blood circulatory speeds in the cerebral arteries in children between 3 and 12 years old with a peak around 7 - 8 years old. This result is similar to that of a study carried out in Senegal by Lemine SOM et al. [10] which found an average age of 7 years with extremes ranging from 10 months to 18 years. However, Coulibaly F Y found more pathological transcranial Dopplers in the age group of 1 - 10 years but with a high frequency between 6 - 10 years [11] . Patients living outside Bamako represented 41%, these are patients far from structures specializing in the management of sickle cell disease, thus delaying the diagnosis, management, and prevention of complications of the disease.

Half of the patients (eleven) were diagnosed with sickle cell disease after the occurrence of the stroke, for the other half, five were regular in follow-up and six were irregular. Which means that any delay in the diagnosis and management of sickle cell disease is a factor leading to severe complications. Ouattara A [7] , found that 61.5% of patients were undiagnosed before the stroke. Headaches were the most common neurological antecedents (63.5%). They can be isolated or associated with other neurological signs such as behavioral disorder, paresis or convulsion. They can be the presumptive sign of transient ischemia which can lead to a complete infarction. Thus, any neurological sign, even a change in behavior in a sickle cell patient, should lead to a search for a stroke [12] . Disturbance of consciousness was the most frequent reason for consultation (27.3%), followed by convulsions (18.2%). Ouattara A [7] found a predominance of convulsions (46.2%), followed by hemiplegia (23.1%). The appearance of the convulsive seizures varied: tonic seizures, atonic seizures, myoclonic seizures or even tonic-clonic seizures. Their appearance can be explained in particular by cerebral anoxia, anemia, hyperthermia and infections. The mode of installation of the deficit was gradual in 54.5% of cases and abrupt in 45.4%. Pallor was the most common clinical sign (73.2%), which is due to the chronic anemia that accompanies this disease. The most common neurological signs were disturbance of consciousness and hemiparesis with 45.4% each, followed by disturbances of sensitivity in 31.8% of cases. This result is different from that of Rakotoharimanana V C [8] who found a predominance of hemiplegia in 79% of cases, followed by dysarthria in 63% of cases. According to Bernaudin F et al. [13] , the clinical signs most frequently presented by sickle cell patients during stroke are hemiparesis, aphasia or dysphasia with or without convulsions [13] . We noted 95.5% of SS form and 4.5% of Sβ⁰thalassemia form, this could be explained by the high number of SS compared to others in the general population. Driscoll MC et al. in 2003, in a study which included 3425 patients, found that the frequency of strokes was 7.1% in patients with SS hemoglobin, 1.1% in patients with Sβ⁰thalassemia hemoglobin, 0.6% for S/β⁺thalassemia and no cases in patients with SC hemoglobin [14] . However, Coulibaly FY [11] found a higher frequency of pathological Doppler in SS than in Sβ⁰thalassemia and normal velocities in SC. Our patients had a hemoglobin level < 7 g/dl in 57.1% of cases and between 7 and 9 g/dl in 42.8% of cases, with an average level of 6.7 g/dl ± 1.7. This drop in hemoglobin level leads to an increase in cerebral blood flow with acceleration and disruption of the flow at the level of the carotid siphon, the end of the internal carotid artery and the flexions of the cervical internal carotid artery, which favors damage. endothelial with hyperplasia of the intima and media therefore leading to stenosis [15] . It has also been shown that a sudden and significant drop in hemoglobin level would cause an infarction linked to tissue hypoxia by a significant drop in local cerebral blood flow [16] . The vast majority of our patients (95.4%) had hyperleukocytosis (leukocytes > 10,000/mm³) with an average rate of 20,063/mm³ ± 9,800/mm³. This hyperleukocytosis has been very often described in sickle cell patients, and the high number of polynuclear neutrophils is a pejorative element in the severity of sickle cell disease. Their presence in the post-capillary venules strongly suggests (because of their volume), their major participation in the circulatory slowdown initiating vaso-occlusion. It has also been shown that sickle cell red blood cells are capable of interacting with leukocytes and particularly neutrophils [17] . All our patients (100%) had an ischemic stroke. This predominance of ischemic lesions was also found by OUATTARA A [7] and RAKOTOHARIMANANA [8] with 90.9% and 87.5% respectively. In our study, the Sylvian artery (right and left) was the most affected (45.5%). In our study, the Sylvian artery (right and left) was the most affected (45.5%). This result is similar to that of BASSE AM et col which found involvement of the sylvian artery in 49 cases out of 64 ischemic strokes (76.5%) [18] . However, according to the literature, stroke in children with sickle cell disease is linked to progressive stenosis of the large vessels, particularly the carotid system and the middle and anterior cerebral arteries in their proximal segment [16] . The majority of our patients (77.2%) did not have transcranial Doppler ultrasound before the occurrence of the stroke due to lack of follow-up, whereas for the prevention of stroke, transcranial Doppler ultrasound must be carried out systematically once a year in all sickle cell patients from the age of two. Among the five patients who performed transcranial duplex ultrasound, two were pathological and three normal. According to Françoise Bernaudin, 20% of children with sickle cell disease will have ischemic lesions without pathological transcranial Doppler ultrasound [15] .

All our patients were able to be urgently transfused with packed red blood cells. Then a transfusion program with phenotyped blood was started. This program consisted of a simple transfusion and/or transfusion exchanges with monitoring of the hemoglobin level through blood count, hemoglobin electrophoresis, and serum ferritin with the aim of lowering and maintaining the hemoglobin level S around 30%. All patients who benefited from the transfusion program showed good progress. This is similar to the result of the STOP I study, in fact, in this multicenter study (STOP I) randomizing a monthly transfusion program versus simple observation in 130 children with sickle cell disease having a pathological transcranial ultrasound with a speed greater than 200 cm/s showed that the transfusion program aimed at maintaining the HbS level below 30% made it possible to reduce the risk of stroke very significantly by 92% (p < 0.001) [13] . After 24 months of observation, ten strokes and one intracerebral hematoma occurred in the 67 non-transfused patients compared to a single stroke in the 63 transfused [13] . The majority of our patients (86%) benefited from treatment with hydroxycarbamide, even if no study has shown its effectiveness during stroke in sickle cell patients. It is mainly used to reduce recurrences of vaso-occlusive crises but in certain studies such as that carried out in Tunisia, it was indicated for the prevention of relapse of a cerebrovascular accident in two patients [19] . According to Bernaudin et al., in the event of normalization of speeds (<170 cm/s) and in the absence of stenosis at the ARM, exit from the transfusion program with bridging by treatment with hydroxycarbamide can be proposed, subject to good compliance and after an overlap period of at least 3 months [20] . In our study, 68% of patients benefited from functional rehabilitation and we noted good neurological recovery in all these patients. Rehabilitation began as soon as the patients were stabilized. According to Bonan L et al., the objective of rehabilitation is to stimulate the process of cerebral plasticity to maintain or restore altered functions and to prevent the occurrence of complications and to lead the patient to optimal autonomy regardless of the severity of the after-effects [21] . In our study, the case fatality rate was 22.7%. Among the living patients, 22.7% developed serious after-effects, 45.4% developed minimal after-effects and 9.1% completely recovered. This fatality rate could be explained by the delay in diagnosis and treatment linked to the time (consultation) between the start of symptoms and admission to hospital and also by the severity of the brain lesion.