Clinical and etiological profile of epilepsy at the regional hospital center of Tahoua (Niger): A 4‐year retrospective study

Abstract Objective We aimed to evaluate epilepsy management at the regional hospital center (RHC) of Tahoua (Niger) to determine the intrahospital deficiencies to optimize to improve the management of people with epilepsy. Methods A descriptive retrospective study was carried out at the Psychiatric Unit of the RHC of Tahoua between January 1, 2016 and December 31, 2019. We collected from the registers of consultation all patients followed by nurse technicians in mental health for epilepsy whose diagnosis was made by nonspecialist physicians or internists. The study included patients with epilepsy who had a medical follow‐up at least 6 months. Results Of the 2022 patients seen during the period of the study, 246 patients were consulted for epilepsy with a hospital frequency of 12.2%. The mean age was 22.38 years, with a slight predominance of men (57.7%). Only generalized tonic‐clonic seizures (95.1%) and focal‐aware seizures (4.9%) were reported. The main etiologies were cerebral malaria (18.7%), bacterial meningitis (8.1%), and head trauma (4.9%). In 60.2% of cases, the etiology was undetermined, but the etiological diagnosis investigation was incomplete (CT scan not done). Only the first‐generation antiepileptic drugs were used. Only 2.8% of the patients had drug‐resistant epilepsy, and 97.2% of the patients had controlled seizures. Conclusion The study shows a predominance of infectious causes, particularly cerebral malaria. We found a high proportion of epilepsies with unknown etiologies with incomplete workup. The RHC of Tahoua should facilitate access to the CT scan for people with epilepsy to improve etiological diagnosis investigation.


of 9
TOUDOU-DAOUDA AND IBRAHIM-MAMADOU Until May 2018, the public health infrastructures of Niger consisted of 1057 integrated health centers, 2466 health huts, seven regional hospital centers, seven mother and child health centers, and four tertiary care hospitals with national referral maternity (OMS, 2018).
Niger's health system also included 387 private establishments (such as 346 medical offices, 36 clinics, two nonprofit hospitals, and two private centers specializing in ophthalmology and trauma), 120 private pharmacies, 11 laboratories, and 19 private health schools. In Niger, people with epilepsy are mainly cared for by nonphysician health care workers, nonspecialist physicians, and non-neurologist physicians because, until May 2018, the country had only seven neurologists who worked in the hospitals of Niamey (capital of Niger). Nonphysician health care workers are mainly represented by nurse technicians in mental health who are trained in the management of epilepsy and psychiatric diseases. In the Tahoua region (Niger), the majority of people with epilepsy are followed at the Psychiatric Unit of the Regional Hospital Center (RHC) of Tahoua by nurse technicians in mental health assisted by nonspecialist physicians and internists. In the case of diagnostic difficulty, the patient is referred to Niamey for neurological consultation, and for eventual additional examinations useful to make an etiological diagnosis. This demonstrates not only the unavailability of diagnostic tools such as electroencephalogram (EEG) and neuroimaging exams but also the lack of neurologists at the RHC of Tahoua.
In Niger, only one study was carried out on the clinical characteristics of epilepsy in a tertiary care referral center of Niamey (Assadeck et al., 2019). Thus, the present study was designed to evaluate epilepsy management at the RHC of Tahoua to determine the intrahospital deficiencies to optimize to improve the management of people with epilepsy. health districts, and regional hospital centers where they play the role of psychiatrists when these latter are not available.

Patients
The study included people of all ages who were followed at the Psy-

Ethical approval
The local Ethics Committee of RHC of Tahoua (Niger) considered the study to be a service evaluation and ruled that no formal ethics approval was required for this study because patient management was TOUDOU-DAOUDA AND IBRAHIM-MAMADOU 3 of 9 not affected. The written informed consent was waived by the local Ethics Committee because of the retrospective and anonymous nature of the data.

Statistical analysis
In the descriptive analysis of the data, patient characteristics were expressed as percentages for the qualitative variables and mean ± SD for the quantitative variables. The chi-square test of Pearson and Fisher's exact test were used to compare the proportions of the qualitative variables. p-Values <.05 were considered statistically significant.
All statistical analyses were performed using the IBM SPSS statistical software package version 22.0 (SPSS Inc.).

Sociodemographic characteristics
Of the 2022 patients seen at the Psychiatric Unit of the RHC of Tahoua

Clinical characteristics
The mean age of first seizures was 18.96 ± 13.36 years (range: 0 and 65 years) ( Table 2). The age of first seizures was less than 10 years in 29.2% of the patients and less than 20 years in 59.7% of the patients. Generalized tonic-clonic seizures (95.1%) and focal aware seizures (4.9%) were the two seizure types identified in this study. Loss of acquisitions or psychomotor decline was the most frequent associated clinical sign (6.5%), followed by the hemicorporeal motor deficit (1.6%).
Generalized tonic-clonic seizures were significantly more frequent in patients aged 20-39 years (p = .001) ( Table 3). We did not find a relationship between gender and seizures type.

Etiological characteristics and risk factors
Cerebral malaria was the most common etiology (18.7%), followed by bacterial meningitis (8.1%) and head trauma (4.9%) ( Table 2). In 60.2% of the patients, the etiology was undetermined, but the etiological diagnosis investigation was incomplete (cerebral CT scan not done). Head injuries were significantly more frequent in men than in women (7.7% vs. 1%; p = .015) as well as toxic causes (6.3% vs. 0%; p = .009). Significantly, cerebral malaria (p < .001) and bacterial meningitis (p = .001) were more common in patients younger than 20 years. Poststroke epilepsies were significantly more frequent in patients older than 40 years (p < .001). Cerebral malaria was more frequent in TA B L E 2 Clinical, etiological, and therapeutic characteristics as well as the outcomes during follow-up visits of the patients (n = 246)
VPA was the most used AED (42.3%). The CBZ-PB combination was the most used (14.6% of all cases).

Outcomes during follow-up visits
The patients were followed at a frequency of 3 months. Significantly, we found an association between drug resistance and bacterial meningitis (p < .001) and hypoxic-ischemic encephalopathy due to birth asphyxia (p < .001) ( Table 5). Patients with drug-resistant epilepsy were significantly on dual therapy (p < .001).

DISCUSSION
The present study shows a hospital frequency of epilepsy of 12.

p-Value
Seizures type, n (%) GTCS Note: Peasants: The farmers, breeders, and butchers were considered as peasants.
Niamey (Assadeck et al., 2019). On the other hand, a predominance of the female sex has been reported in other studies (Osuntokun et al., 1987(Osuntokun et al., , 1982Rwiza et al., 1992;Simms et al., 2008). The mean age of first seizures was 18.96 years in the present study.
The age of first seizures was less than 10 years in 29.2% of the patients and less than 20 years in 59.7% of the patients. Higher proportions for an age of first seizure less than 20 years have been reported in Burkina Faso (Nitiéma et al., 2012;[65.8%]) and Kenya (Munyoki et al., 2010;[77.8%]). On the other hand, a lower proportion than our findings for an age of first seizures less than 10 years has been reported in Cameroon (Kamgno et al., 2003;[24.6%]).
As in the present study, many studies from sub-Saharan Africa showed a predominance of generalized tonic-clonic seizures (Assadeck et al., 2019;Igwe et al., 2014;Kamgno et al., 2003;Mmbando et al., 2018). However, a predominance of focal-onset seizures has been reported in other studies in sub-Saharan Africa (Newton & Gero, 1984;Osuntokun et al., 1987;Van Der Waals et al., 1983). The report of only two seizure types in the present study could be explained by the fact that epilepsy management had been provided by health workers not well experienced in the management of this disease. Certain types of focal to bilateral tonic-clonic seizures tend to be considered as generalized seizures.
In the present study, an etiology has been identified in 39.8% of the patients. Infectious causes were the most frequent (26.8%), followed by the structural causes (9.3%). Cerebral malaria was the most common infectious cause (18.7%), followed by bacterial meningitis (8.1%).
The most common structural causes were head trauma (4.9%) and hypoxic-ischemic encephalopathy due to birth asphyxia (2.8%). In the study carried out in Niamey at the National Hospital of Niamey, the structural causes were the most frequent dominated by cerebrovascular disease and head injuries (Assadeck et al., 2019). Two Nigerian studies reported a predominance of head injuries and febrile convulsions in childhood as the main etiological factors (Ogunrin et al., 2014(Ogunrin et al., , 1987. As in the present study, the predominance of infectious causes has been reported in a study from Togo (Balogou et al., 2007). In the present study, the predominance of cerebral malaria in patients living in urban areas than those living in rural areas could be explained by the fact that in Niger, there are more stagnant waters in cities than in the countryside. These stagnant waters are a source of reproduction for mosquitoes, which are the vector agents of malaria.
In the present study, only the first-generation AEDs (VPA, CBZ, and PB) were prescribed. A Nigerian study also reported the prescription of only the first-generation AEDs (Igwe et al., 2014). As in the study carried out at the National Hospital of Niamey (Assadeck et al., 2019), VPA was the most AED used in the present study. In Mali, a study reported phenytoin as the most AED used (Nimaga et al., 2002). In the present study, newer generation AEDs were not used because they are expensive for patients who are farmers in the majority of cases and have limited access to these drugs. Although first-generation AEDs are not expensive, the rupture of these treatments in pharmacies is a recurrent problem in Niger, exposing patients to the risk of epileptic seizures due to drug withdrawal. However, this notion of the rupture of AEDs was not reported in the present study. Possibly, this informa-tion was neglected and not noted in the registers of consultation by nurse technicians in mental health during follow-up visits. We found only 2.8% of cases of drug-resistance in the present study. In the literature, it is estimated that 30%--40% the of patients with epilepsy will develop drug-resistant epilepsy (Kalilani et al., 2018). Our results are very far from the data in the literature. This low rate of drug-resistance in the study presence could be explained by (1)

Limitations of the study
The first limitation of the study is sample size. Second, the unavailability of complementary examinations for an etiological investigation of epilepsy at the RHC of Tahoua such as neuroimaging exams. This explains the high proportion of patients with epilepsy without a defined etiology. Third, the retrospective nature of this study explains why some details were not provided especially seizure frequency in patients with drug-resistant epilepsy. Fourth, the low prevalence of drug resistance suggests that certain types of focal-onset seizures are unrecognized because the medical follow-up was provided essentially by nurse technicians in mental health.

CONCLUSIONS
Our study shows a predominance of infectious causes (26.8%), certainly due to a high prevalence of infectious diseases in Niger, such as malaria and meningitis. Also, the study found only 2.8% of drug resistance, which is far and away very low compared to the literature data.
On the other hand, the present study shows limited access to diagnostic tests of epilepsy at the RHC of Tahoua, such as the CT scan explaining why 60.2% of the patients did not have an identified etiology. To improve the quality of epilepsy management at the RHC of Tahoua, it is necessary to facilitate for people with epilepsy the accessibility to the cerebral imaging such as CT scan. It is also necessary to organize continuous training sessions for the health workers who are already in their workstation (nurse technicians in mental health, nonspecialist physicians, and internists) to improve their knowledge of epilepsy management. Finally, young people must be encouraged to specialize in neurology, and the fight against infectious diseases must be well undertaken.