Texte intégral / Full text

Résumé

Introduction : L'accident vasculaire cérébral (AVC) est l'une des complications les plus graves de la drépanocytose. Au Niger, les données disponibles sont descriptives. Cette étude avait pour objectif d'identifier les facteurs associés aux AVC chez les patients drépanocytaires suivis au Centre national de référence de la drépanocytose (CNRD) de Niamey.

Méthodes : Il s'agissait d'une étude cas-témoins menée au CNRD du 1er mai au 15 juillet 2023, à partir de la revue des dossiers de suivi. Les cas étaient des drépanocytaires (SS, Sβ, SC) victimes d'AVC ; chaque cas était associé aux deux patients sans AVC qui le suivaient dans la file active. Les facteurs associés ont été recherchés par régression logistique binaire multiple (procédure pas à pas descendante).

Résultats : Au total, 348 patients ont été inclus (116 cas et 232 témoins). Tous les cas présentaient la forme homozygote SS et 74,1 % avaient 15 ans ou moins. Dans le modèle final, l'AVC était associé au suivi irrégulier (ORa = 4,00 ; IC 95 % : 2,15–7,43) et aux antécédents de transfusion (ORa = 61,67 ; IC 95 % : 23,38–162,61). Les mères sans emploi (ORa = 0,45 ; IC 95 % : 0,23–0,89) et l'anémie consignée dans le dossier (ORa = 0,11 ; IC 95 % : 0,03–0,35) étaient inversement associées à l'AVC. Le test de Hosmer-Lemeshow indiquait une bonne adéquation du modèle (p = 0,449).

Conclusion : Le suivi irrégulier, les antécédents de transfusion et la profession de la mère étaient associés à l'AVC chez les drépanocytaires. Le caractère apparemment protecteur de l'anémie doit être interprété avec prudence. Le renforcement d'un suivi régulier, avec un rappel actif des patients perdus de vue, est prioritaire.

Mots-clés : Accident vasculaire cérébral ; drépanocytose ; facteurs associés ; étude cas-témoins ; Niger.

Introduction

Stroke is defined as "rapidly developing clinical signs of focal or global disturbance of cerebral function, with symptoms lasting more than 24 hours or leading to death, with no apparent cause other than of vascular origin" [1]. Stroke occurs most often in people with hypertension, diabetes, kidney disease or advanced age, and also in patients with sickle cell disease. Sickle cell disease, the most common genetic disease worldwide, affecting around 50 million people [2], is a leading cause of stroke in both children and adults [3]. According to the World Health Organization (WHO), stroke affects nearly 8% of children with sickle cell disease worldwide [4].

Ohene-Frempong et al. reported that the spontaneous risk of stroke increases with the age of the patient with sickle cell disease. The probability of a first stroke was estimated at 11% by the age of 20 years, 15% by 30 years and 24% by 45 years in patients with SS and Sβ⁰ genotypes [5]. In sub-Saharan Africa, a systematic review of 10 articles from seven countries found a prevalence ranging from 2.9% in Nigeria to 16.9% in Tanzania [6]. The severity of stroke lies in the risk of disabling complications. Stroke prevention in sickle cell disease relies on early detection of stroke risk with transcranial Doppler ultrasound and the initiation of lifelong hydroxyurea treatment.

In Niger, an increase in the number of strokes among patients with sickle cell disease has been observed, hence the need to identify the associated factors that might explain this rise. The main associated factors reported in the literature are age, sex, the type of haemoglobinopathy, severe anaemia and transfusions [5, 7–13]. Most of these studies were prospective longitudinal or retro-prospective studies.

In Niger, existing data come from descriptive studies. To shed more light on the associated factors, we conducted a case-control study of clinical factors, such as the frequency of vaso-occlusive crises (which we hypothesised to be the main risk factor), and of sociodemographic factors.

The aim of our study was to identify the factors associated with stroke in patients with sickle cell disease in a hospital setting at the National Reference Centre for Sickle Cell Disease in Niamey in 2023.

Methods

The study was conducted at the National Reference Centre for Sickle Cell Disease (CNRD). The CNRD is a public scientific, technical and cultural institution under the technical supervision of the Ministry of Public Health and Hygiene (MSP/HP). Located in Niamey, it receives patients from all regions of Niger.

This was a case-control study carried out at the CNRD from 1 May to 15 July 2023.

Cases were patients with sickle cell disease (SS, Sβ, SC) who had had a stroke and who had a usable medical record with an identification number in the active follow-up list. Controls were patients with sickle cell disease (SS, Sβ, SC) from the same list who had not had a stroke, with two controls per case.

Cases were selected consecutively from the active follow-up list, starting from 2014, until the planned number was reached. Each time a case was identified, the two patients immediately following it in the list were selected as its controls.

Sample size was calculated with the STATCALC function of Epi-Info® version 7.2.5.0 using the Kelsey method, with the following parameters: 95% confidence level, 80% power, two controls per case, a frequency of vaso-occlusive crisis (VOC) of 54.9% [14] and an expected odds ratio of 2. This gave a minimum sample of 348 patients, comprising 116 cases and 232 controls.

The dependent variable was the occurrence of stroke. The main exposure of interest was the frequency of vaso-occlusive crises.

Data were collected with a questionnaire through a review of the records of the active follow-up list. Data were entered with Epi-Info® 7.2.5.0 and analysed with Stata® version 14.

A descriptive analysis with calculation of proportions was performed. Odds ratios with their 95% confidence intervals and p-values were estimated.

Simple binary logistic regression was then performed. All variables associated with stroke with a p-value ≤ 0.20 were entered into the multiple binary logistic regression model, adjusting for sociodemographic, clinical, biological and genetic characteristics. The main exposure of interest was also forced into the initial model.

A backward stepwise procedure was used for the multivariable analysis. The final model retained only independent variables with a p-value < 0.05. The goodness of fit of the final model was assessed with the Hosmer–Lemeshow test.

Administrative authorisations were obtained from the Ministry of Public Health and Hygiene, the coordination of the BFELTP programme and the CNRD. Ethically, the database was anonymised, and patient confidentiality and anonymity were respected.

Results

A total of 348 patients with sickle cell disease were selected. Overall, 74.1% of cases and 60.8% of controls were aged 15 years or younger. Among stroke cases, 52.6% were male, compared with 44.0% of controls. All stroke cases (100%) had the homozygous SS form. In addition, 31.0% of cases lived in rural areas, compared with 44.0% of controls (Table I).

VOC: vaso-occlusive crisis; ACS: acute chest syndrome; HbF: foetal haemoglobin.

In simple binary logistic regression, only father's education and the presence of HbF had p-values above 0.20 (Table II).

OR: odds ratio; CI: confidence interval; VOC: vaso-occlusive crisis; ACS: acute chest syndrome; HbF: foetal haemoglobin.

After adjustment, the factors associated with stroke in patients with sickle cell disease at the CNRD in Niamey in 2023 were anaemia, maternal employment, irregular follow-up and transfusion history (Table III).

aOR: adjusted odds ratio; CI: confidence interval. Hosmer–Lemeshow test: p = 0.449 (> 0.05), indicating an adequate fit of the final model.

Discussion

This study, conducted at the CNRD in Niamey in 2023, included 348 patients. The factors significantly associated with stroke were irregular follow-up, anaemia, transfusion and maternal employment.

The study showed that irregular follow-up was a predictor of stroke. This is an original finding, as this factor has not been reported in the available literature. Regular follow-up is essential in the management of sickle cell disease because it allows complications to be detected early. Stroke screening relies on transcranial Doppler ultrasound, performed annually when the result is normal and every six months when it is abnormal. Irregular follow-up therefore prevents monitoring of the vasculopathy as well as of certain biological factors that favour stroke, such as leucocytosis and thrombocytosis.

Although transcranial Doppler is the reference investigation for stroke-risk screening in children with sickle cell disease, it could not be included as a variable in our analysis. This examination was only introduced into the routine follow-up assessment in 2017, whereas our study period began in 2014, so the information was not available for all included patients. An important avenue for future research would be to assess, in a study covering the period after 2017, the direct association between the actual performance of transcranial Doppler and the occurrence of stroke.

The study also found that anaemia was inversely associated with stroke, a result that contrasts with that reported by Munube et al. in Uganda [7]. This discrepancy may be explained by limitations inherent in our data source, which was restricted to follow-up records. Some patients may have received an emergency transfusion on admission or in the three months preceding the stroke, and the corresponding blood counts, recorded in inpatient files, were not accessible for this study. This apparently protective effect of anaemia (aOR = 0.11) must therefore be interpreted with caution, all the more so because anaemia was positively associated with stroke in the crude analysis (OR = 2.62) and its direction reversed only after adjustment for transfusion history, which was very strongly associated with stroke; this pattern suggests confounding between anaemia and transfusion rather than a true protective effect. This counter-intuitive association may be explained by recent transfusions not documented in the follow-up files, which could have normalised haemoglobin levels while reducing the actual risk.

Our results also showed that patients whose mothers were employed were more frequently affected by stroke. This may be because working mothers spend less time with their children, so that therapeutic education instructions may be poorly applied. These mothers may also more often miss sickle cell follow-up appointments, which are essential for the detection, prevention and monitoring of major complications of the disease, including stroke.

A history of transfusion was also found to be a predictor of stroke in patients with sickle cell disease: transfused patients were more exposed to stroke. This is consistent with the results of two studies conducted in Uganda and the United States [5, 7]. It may be explained by the fact that repeated transfusions, especially with whole blood, provide additional white blood cells and platelets, which contribute to thrombus formation and thus to stroke.

However, after adjustment, no independent association was found between stroke and the main exposure of interest, the frequency of vaso-occlusive crises. This may be because most patients seek care only for major crises or when a first warning sign appears, such as motor deficit or aphasia. Minor crises do not require a consultation and are managed at home.

Strengths and limitations

Like any study, this one has strengths and limitations.

Unlike the cross-sectional studies usually conducted, the case-control design allowed different associated factors to be identified. Two new factors not previously reported in the literature were thus identified: irregular follow-up and maternal employment.

The study covered a well-defined period (2014–2021), during which cases were identified and matched with two controls each. It was, however, a hospital-based study limited to patients followed at the CNRD, and not all patients who have a stroke are admitted to this facility.

The retrospective design, based on existing records, exposes the study to information bias, as some of the information collected may no longer be up to date. This bias particularly concerns minor vaso-occlusive episodes managed at home without consultation and emergency transfusions, which are inconsistently recorded in medical files. Such under-reporting may have affected the accuracy of some variables in our analysis, especially undocumented transfusions, as discussed above.

In addition, data on stroke recurrence and on patients' vital status during follow-up were not systematically available in the records reviewed, which prevented us from exploring the factors associated with recurrence or mortality.

These results call for stronger follow-up strategies for patients with sickle cell disease, both at the CNRD and nationwide. Since irregular follow-up emerged as a major predictor, particular attention should be paid to active recall through a follow-up register at the CNRD to identify patients who have missed their appointments, although its implementation depends on the availability of dedicated human and logistical resources. Strengthening therapeutic patient education, involving all those who care for the child and not only the mother, appears to be an essential lever.

Conclusion

In this study, irregular follow-up, transfusion history and maternal occupation were associated with stroke, while anaemia recorded in the follow-up file was inversely associated with it, a finding that requires cautious interpretation. All affected patients had the homozygous SS form, and most were aged 15 years or younger. These findings highlight the need for regular, strengthened follow-up to prevent this severe complication.

*Correspondence: MOUSSA HASSANE Fati

moussafati75@yahoo.com

Available online: November 9, 2026

¹ Department of Public Health, Faculty of Medicine, Joseph KI-ZERBO University, Ouagadougou, Burkina Faso

² National Reference Centre for Sickle Cell Disease, Niamey, Niger

³ Comlan Alfred Quenum Regional Institute of Public Health, Ouidah, Benin

© Journal of African Clinical Cases and Reviews 2026

References

1. Béjot Y, Touzé E, Jacquin A, Giroud M, Mas JL et al. Épidémiologie des accidents vasculaires cérébraux. Med Sci (Paris). 1 août 2009 ;25(8‑9):727‑32. 10.1051/medsci/2009258-9727.

2. Diallo M et al. Aspects épidémio-cliniques et thérapeutiques des accidents vasculaires cérébraux chez les enfants drépanocytaires de 6 mois à 15 ans au département de pédiatrie du CHU Gabriel Toure. [Internet] [Thésis]. USTTB ; Disponible sur : https://www.bibliosante.ml/handle/123456789/5909 ; 2023 [cité 13 févr. 2023].

3. Kirkham FJ, Lagunju IA et al. Epidemiology of Stroke in Sickle Cell Disease. J Clin Med. 18 sept 2021;10(18):4232. [PubMed]. 10.3390/jcm10184232.

4. WHO EMRO | Accident vasculaire cérébral | Thèmes de santé [Internet]. Disponible sur https://www.emro.who.int/fr/health-topics/stroke-cerebrovascular-accident/index.html [cité 17 juill 2023].

5. Ohene-Frempong K, Weiner SJ, Sleeper LA, Miller ST, Embury S, Moohr JW, et al. Cerebrovascular Accidents in Sickle Cell Disease: Rates and Risk Factors. Blood. 1 janv 1998;91(1):288‑94. https://www.sciencedirect.com/science/article/pii/S0006497120549082 [2023-02-13].

6. Marks LJ, Munube D, Kasirye P, Mupere E, Jin Z, LaRussa P, et al. Stroke Prevalence in Children With Sickle Cell Disease in Sub-Saharan Africa: A Systematic Review and Meta-Analysis. Glob Pediatr Health. 14 mai 2018;5:2333794X18774970. Disponible sur: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5954575/ [cité 2023-05-01].

7. Munube D, Katabira E, Ndeezi G, Joloba M, Lhatoo S, Sajatovic M, et al. Prevalence of stroke in children admitted with sickle cell anaemia to Mulago Hospital. BMC Neurol. 17 sept 2016;16(1):175. https://doi.org/10.1186/s12883-016-0704-2 [consulté 2023-03-29].

8. Guitton C. La vasculopathie cérébrale de l'enfant drépanocytaire. MISE AU POINT.

9. Gargot J, Parriault MC, Adenis A, Clouzeau J, Dinh Van KA, Ntab B, et al. Low Stroke Risk in Children With Sickle Cell Disease in French Guiana: A Retrospective Cohort Study. Front Med (Lausanne). 2022;9:851918. Disponible sur : https://www.frontiersin.org/articles/10.3389/fmed.2022.851918 [consulté 2023-02-12].

10. Rodrigues DOW, Ribeiro LC, Sudário LC, Teixeira MTB, Martins ML, Pittella AMOL, et al. Genetic determinants and stroke in children with sickle cell disease. J Pediatr (Rio J). 2016;92(6):602‑8. 10.1016/j.jped.2016.01.010. [PubMed].

11. Webb J, Kwiatkowski JL et al. Stroke in patients with sickle cell disease. Expert Review of Hematology. 1 juin 2013;6(3):301‑16. Disponible sur: https://doi.org/10.1586/ehm.13.25. Consulté le 2023-04-29.

12. Rallo MS, Akel O, Kalakoti P, Sun H et al. Characteristics and outcomes of stroke hospitalizations in patients with sickle cell disease and moyamoya syndrome. Journal of Stroke and Cerebrovascular Diseases [Internet]. 1 oct 2022;31(10). Disponible sur : https://www.strokejournal.org/article/S1052-3057(22)00399-8/fulltext [cité 13 févr 2023].

13. Srisuwananukorn A, Raslan R, Zhang X, Shah BN, Han J, Gowhari M, et al. Clinical, laboratory, and genetic risk factors for thrombosis in sickle cell disease. Blood Adv. 12 mai 2020 ;4(9):1978‑86. 10.1182/bloodadvances.2019001384. [PubMed].

14. Mbassi ADH, Dongo FD, Ngo SU, Mafo FV, Alima YA, Njom NAE, et al. Aspects épidémiologiques, cliniques et thérapeutiques des crises vaso-occlusives chez les enfants drépanocytaires en milieu hospitalier à Yaoundé. Health Sci Dis. 2017;18(4):89-97.

How to cite this article: Moussa Hassane F, Sopoh G, Ousseini M, Hamidoun Tahirou H, Boureima Soumana M, Yoda H, Yanogo P. Factors associated with stroke in patients with sickle cell disease at the National Reference Centre for Sickle Cell Disease, Niamey, Niger, 2023. Jaccr Public Health 2026.

Télécharger le PDF