Case N° 1.

Female, aged 11 years, who has never been at school. Hypertonic convulsive seizures reportedly started at 30 months of age, and HN began at the age of six. Attention deficits were noticed by her mother 32 months prior to HN onset. About 24 months after the onset of NS, other symptoms emerged including difficulties to concentrate, hyperactivity, memory deficit-s, wandering behavior and four-limb deformities.

At the clinic, the patient presented with a behavioral disorder, agitation, aphasia, left spastic hemiparesis, left inwardly turned foot, tendon retractions of the two lower limbs more severe on the left and stumbling gait. Her memory and language were also impaired. We equally noted moderate developmental delay and cognitive decline, retarded growth, and stiffness of the joints and limbs. Finally, this patient was bulimic but with a normal weight-for-age index (Z-score = -1.19, percentile = 11.7%). This is the only case in which a HN was witnessed by the research team; the seizure was triggered by noise, lasted for about 30 seconds, and consisted of about 8 HN/min.

Case N° 2.

Female, aged eight years, with pronounced learning difficulties at school, which got worse after HN started. She reportedly experienced attention deficits shortly before HN onset, but the exact timing could not be specified. HN started at the age of five and persisted to date. The main HN trigger was cold. The patient was mostly lucid but often had aggressive episodes, reported to have started three months after HN onset. Memory impairment, visual and auditory hallucinations, began 24 months after HN onset. Prolonged insomnia and saddened mood were noted one and four months after HN started, respectively. The rest of the physical examination was unremarkable.

Case N° 3.

Female, aged 16 years, whose HN started at the age of eight. Although this patient was an average school student, she had dropped out of school after HN onset. About 12 months following HN onset, she started experiencing learning difficulties and memory deficits. Auditory hallucinations started 24 months after HN onset. Prolonged mood changes, including episodes of sadness, also developed but the timing was unclear. A few months after the onset of HN, generalized myoclonic seizures started and were still present at the time of this study.

The frequency of HN could not be determined, and no specific HN trigger was identified. The neurological examination found a monoparesis with amyotrophy and a right claw hand, a right hand drop, and spastic, stumbling gait.

Case N° 4.

Female, aged eight years, who reported onset of HN 11 months before this study, which led to her dropping out of school. Already at age five, a number of signs had appeared and were still present during our neurological examination: reduced concentration, blank starring episodes and excessive sleep. HN started in 2019 with an estimated frequency of 11–15 per minute, and was often triggered by meals taken during cold weather. Learning difficulties, little eating, lower than expected weight for age (Z-score = -3.60, percentile = 0.1%), and other symptoms appeared almost simultaneously with HN. Overall, this patient was ill-looking with obvious growth retardation (Z-score = -2.12) and loss of interest in activities. The rest of the clinical examination was unremarkable.

Case N° 5.

Female, aged 15 years, unschooled, and with a family history of epilepsy (brother). At the age of 12, nodding seizures started with a frequency of 5 HN/min, which was the same seizure frequency reported at the time of the survey. There were no specific triggers for the HN. Excessive somnolence was noted by her mother prior to HN onset, with worsening thereafter. Twelve months after HN onset, she experienced progressive weakness of both lower limbs and the right upper limb which evolved into a spastic syndrome, resulting in an unstable gait confirmed on clinical examination.

Socio-demographic characteristics and background.

The age of onset of HN ranged from five to 12 years, falling within the WHO probable case definition [2] and similar to what had been reported by several other authors [6,11–14,26]. This would correspond to an appearance of cases in CAR between 2011 and 2019 suggesting a recent and probably still incidental event. Families had a low socio-economic status like the majority of NS cases described both in East Africa since its discovery [3,13,18] and in Cameroon [6].

Deliveries went well in all our cases and all babies cried immediately at birth, which is in agreement with other studies [6,26]. Parents did not associate HN seizures with epilepsy, which explains the lack of prior use of AED.

Winkler’s classification, HN frequency and triggers.

Two patients were classified “HN plus” according to the classification by Winkler et al. (2008). The associated epilepsy in each case was generalized epilepsy (one was myoclonic) as in most of the studies describing “HN plus” in Uganda [11,13,26] and in Tanzania [14] with variable proportions. In addition, these HN-associated seizures occurred earlier in our study compared to what was observed elsewhere [14,27].

The frequency of HN appeared to be lower in our population than in most studies conducted in East Africa. Higher frequencies were reported in some studies [18] and this is probably why the WHO refers to 5 to 20 HN/min in its definition of probable cases. However, lower frequencies were also observed elsewhere, notably in Cameroon [6]. The frequency of these seizures could be related to the severity of the disease. This is a difficult criterion to assess [6,26] and therefore often omitted [10].

In our study, as in several others, either a meal [11,13,18], or cold weather or a cold breeze [6,11,13] was the trigger for HN. In one of our cases, the mother insisted that HN were triggered when the child “ate his meal at cold hours”. Seizures were triggered mainly during morning and evening meals.

Growth retardation and sexual development.

One patient (20%) was small for their age compared to 40.9% (9/22) and 26.7% (4/15) in two series in Uganda [11,26]. It is a fairly common symptom and found in almost every series [6,18]. Physical deformities including limbs that can be significant were described in some cases in our series and were responsible for gait disorders and functional impairment. These deformities were reported in somewhat older NS cases in Uganda [27].

No cases of delayed sexual development were identified among our NS cases. This would also be a sign encountered in variable proportions in the literature [2,11,18,26].

Developmental delay and cognitive decline, psychiatric disorders.

Developmental delay and cognitive decline described in the literature [6,11–13,26,28] were also observed in our series. Only two cases (40%) showed no cognitive decline. As in Ugandan [27] and Tanzanian [14] studies, most of these signs occurred after the onset of HN (between 12 and 24 months) suggesting that this is a natural history of the disease. Cognitive impairments were relatively frequent in the different NS populations and the proportions varied according to location: 100% in Sudan [12], 45.4% in Uganda [11], and 40.3% in Tanzania [28]. The severity of these cognitive impairments also varied within each study. For example, although 100% of children had cognitive impairment in Sudan, it was severe in only 9.5% of these cases, compared to 45.4% in Uganda and 48% in Tanzania among children with cognitive impairment. Deficiencies were perhaps less severe in Central African countries, such as Cameroon [6] and CAR (this study), where cognitive deficiencies were mostly mild to moderate.

The presence of psychiatric signs (60% in our study) is quite variable from one study to another, ranging from 0% [26] to 50% [11] in the same country (Uganda). These signs as well as prolonged periods of depressive mood appeared about 24 months after HN, which is similar to what was described in Uganda [27].

Last but not least, the fact that some parents reported prodromal signs (somnolence, diminished attention, blank starring, etc.) during the months or years preceding HN onset aligns with the natural history of NS as proposed by Idro et al [27].

Onchocerciasis endemicity.

Several studies reported a high number of NS cases in areas highly endemic for onchocerciasis [17,18,29]. No data on Onchocerca volvulus endemicity were available in this area at the time of the present study (December 2019). In January 2020, we conducted a follow-up study in the same area to assess the prevalence of onchocerciasis in children aged six to nine years, based on Ov16 antibody prevalence. Results were recently published and the overall Ov16 seroprevalence was 8.9% [21]. This means that there is an ongoing Onchocerca volvulus transmission in the Landja area, but the endemicity of onchocerciasis in the area is low. However, the overall prevalence of onchocerciasis antibodies was 20% in Kodjo, a village close to rapids on the Ubangui river [21]. This suggests that Onchocerca volvulus transmission is high and ongoing in remote villages along the Ubangui river and the burden of NS could be higher there, but the prevalence of the disease is unknown in this area where there are security concerns and problems of access.

Diagnostic certainty (WHO Classification).

According to the WHO classification [2], only one of our cases was confirmed and the other four were probable. As in other studies, it was difficult to confirm a case of NS. For example, only three out of 21 cases were confirmed in Sudan [12] and none in 15 cases in Uganda [26]. Probable cases could therefore be sufficient to diagnose the disease and to manage the patients. According to the WHO classification and literature data, the diagnosis of NS may be uncertain in some cases, particularly those with focal neurological signs or symptoms (case N° 5). However, such focal abnormalities in persons with NS were already described in some series, particularly in Uganda [11,13] and Tanzania [14] but they were very uncommon. Lastly, the grouping of cases in space (associated with other clinical criteria) seems to be a very important criterion to support the diagnosis of NS. It is important to recall that an expert consensus meeting in Gulu in 2015 had already raised several limitations and shortcomings of the WHO classification and suggested a modification [15].

Electroencephalogram.

Among our patients, four had an abnormal EEG. In patients with NS, EEG results were abnormal in different proportions according to the studies, ranging from 60% [14] to 100% [11]. Most of these anomalous patterns in the literature were obtained in the interictal phase as was the case in our study [12–14]. In some cases the EEG trace changed during the ictal phase (i.e. during HN) [12], but this was not observed in our cases.

Apart from a diffuse slowing observed in two of our patients, the abnormalities noted on the EEG were variable. Only two patients had slow spikes. The various other graphoelements we found could be consistent with diffuse brain damage due to different pathologies. Also, we caught a HN seizure during the EEG exam but there was no change in the EEG tracing.

These results suggest that NS may have several electroencephalographic manifestations in this population that may correspond either to a wide variety of clinical forms or to various stages of disease progression. In addition, it is important to note that there is still no consensus of what a “typical” EEG result might be in a patient with NS.

Limitations of the study.

The aim of this study was to describe the clinical features of NS cases in CAR but a few limitations exist. Because of the difficulty of access to certain areas, we only surveyed one of the six districts of the Landja area and the number of NS cases identified is too low to draw definite conclusions. Therefore, it is likely that the figures presented in this study are far from the real prevalence of NS. There may also be a recall bias, especially in the reconstruction of the history of the disease by parents or guardians, which may have impacted the timing of onset, duration of symptoms and potentially the diagnosis itself. It is also possible that we missed additional NS cases because 39% of cases detected with the screening questionnaire were not seen by the neurologist. Two main reasons were mentioned by the families: some parents said they were discouraged by their children’s state of health and did not want to make any more efforts; other parents were simply absent or unavailable to take their children to the health center. To our knowledge, validated questionnaires for screening and confirmation of NS cases are not available. We therefore used non-validated questionnaires that were inspired by all the studies carried out by experts in the field. There was a variability between interviewers during the screening phase because the investigators were organized in three teams but all interviewers were trained before the study, had the same degree of education, and asked only multiple choice questions. The two confirmation questionnaires and the EEG questionnaire were administered by the same team, thus eliminating inter-interviewer variability.