The head nodding syndrome—Clinical classification and possible causes


  • William Matuja and Erich Schmutzhard contributed equally to the conceptual work of the manuscript.

Address correspondence to Dr. Andrea S. Winkler, Department of Neurology, University of Ulm, Oberer Eselsberg 45, 89081 Ulm, Germany. E-mail:


Purpose: In the 1960s in Tanzania, L. Jilek-Aall observed a seizure disorder characterized by head nodding (HN). Decades later, “nodding disease,” reminiscent of what was seen in Tanzania, was reported from Sudan. To date this seizure disorder has not been classified and possible causes still remain obscure.

Methods: In a prospective study in southern Tanzania, we evaluated 62 patients with HN. Selected patients underwent blood (n = 51) and cerebrospinal fluid (CSF) (n = 48) analyses. Others were chosen for MRI (n = 12) and EEG (n = 10).

Results: Seizure type was classified as “head nodding only” and “head nodding plus,” the latter being combined with other types of seizure (n =34). During HN, consciousness was impaired in 11 patients (17.7%) and supportive signs of epileptic seizures were described by 15 (24.2%) patients. Precipitating factors were confirmed by 11 (17.7%) patients. Fifty-six (90.3%) patients had at least one relative with epilepsy. EEG confirmed interictal epileptic activity in two patients and unspecific changes in four patients. MRI showed hippocampus pathologies (n = 5) and gliotic changes (n = 5). Skin polymerase chain reaction (PCR) positivity for Onchocerca volvulus was significantly associated with lesions on MRI. However, PCR of the CSF was negative in all cases.

Conclusions: We present a comprehensive clinical description of the “HN syndrome,” possibly a new epilepsy disorder in sub-Saharan Africa. MRI lesions and their association with positive skin PCR for O. volvulus despite negative PCR of the CSF is intriguing and deserves attention. Furthermore, the high prevalence of hippocampus sclerosis and familial clustering of epilepsy may point toward other potential pathogenetic mechanisms.

In southern Sudan, a progressive disease characterized by head nodding (HN), mental retardation and stunted growth has been noticed since the mid 1990s and was reported in 2003 as “nodding disease” (Lacey, 2003). Possible causes have remained obscure. An interesting observation, however, is that 93% of the children affected are infested with O. volvulus, a filarial worm, which is transmitted by the black fly, Simulium spp, breeding near fast-flowing rivers (Harding, 2003).

Interestingly, some decades earlier, in 1962, L. Jilek-Aall described several children with attacks of “nodding head,” some of them preceding the development of “grand-mal” seizures (Aall, 1962). In further publications, she elaborated on this observation (Aall-Jilek, 1964, 1965). The area in which L. Jilek-Aall observed the HN cases, the Mahenge area in the Ulanga district in southern Tanzania, is counted amongst the regions most heavily infested with O. volvulus (Mwaiko et al., 1990). Independently of the reports of Jilek-Aall a seizure type reminiscent of HN—“dropping the head in the pan”—was described among the Bassa and Kpelle in Liberia, where epilepsy is highly prevalent (Gerrits, 1983).

In various regions with a high prevalence of onchocerciasis, the number of people with epilepsy has been reported to be higher compared to hypo- or nonendemic areas—Mexico (Casis Sacre, 1938; Balanzario Rosas, 1942), Sudan (Haseeb et al., 1962), Uganda (Kipp et al., 1992; Ovuga et al., 1992; Kipp et al., 1994; Kaiser et al., 1996, 1998), Burundi (Newell et al., 1997), Tanzania (Rwiza et al., 1992; Jilek-Aall, 1995; Maegga, 1998), Cameroon (Boussinesq et al., 2002) and Nigeria (Dozie et al., 2006). In some of these areas an entity termed the Nakalanga syndrome consisting of epilepsy, stunted growth, and mental retardation, reminiscent of the “nodding disease” in Sudan, was described (Ovuga et al., 1992; Newell et al., 1997) and HN attacks were mentioned explicitly in an area highly endemic for O. volvulus in West Uganda (Kaiser et al., 2000). However, not all studies on the relationship between O. volvulus and epilepsy have shown positive results (Twum-Danso, 2004). A meta-analysis of all available studies was only able to demonstrate a nearly significant association between O. volvulus and epilepsy (Druet-Cabanac et al., 2004).

To date, the cause of the HN syndrome as observed in Tanzania and Sudan remains unclear and a relationship with onchocerciasis may only be speculated on. In our study, we collected details on semiology of HN seizures and other associated clinical features. In selected patients, we performed blood and CSF analyses, O. volvulus polymerase chain reaction (PCR) in skin and CSF, EEG, and MRI in order to characterize this new disease entity.


Study site

The study was conducted in The Mahenge Epilepsy Clinic, which was founded in 1960 by Jilek-Aall and which today sees over 900 people with epilepsy (Jilek-Aall, 2003). It is situated in the Government Hospital of Mahenge, which cares for approximately 32,000 inhabitants of the Vigoi division, Ulanga district, southern Tanzania. People with epilepsy (PWE) reach the clinic mainly on foot. Patients who are too ill on the day they are due to attend the clinic are visited by the nurse on the motorbike provided for him. He will see the patient for required treatment and bring him/her the medication. Each patient is supposed to come to the clinic once a month for a check-up and for a month's supply of medication. For many years phenobarbitone and phenytoin have been the only antiepileptic medication available at the Government Hospital in Mahenge. The hospital of Mahenge is the only hospital for patients of the Vigoi region, which extends over mountainous terrain. There is a referral hospital in Ifakara, in the lowlands, which is seldom used as access is difficult. Most patients are brought to the clinic by relatives who have heard about the treatment of epilepsy by word of mouth. The clinic has been there for many decades. It has been visited by several overseas doctors throughout the years, has always been headed by Jilek-Aall, and has always had the same nurses. Thus the service has gained the trust of the PWE. We believe that there are few barriers nowadays for accessing the Epilepsy Clinic, so that most of the PWE will eventually reach the clinic.

Participants and data collection

Patients known to suffer from HN seizures were recruited from The Mahenge Epilepsy Clinic in a consecutive way. A “HN questionnaire” was drafted and used in a face-to-face interview performed with a final-year medical student, a translator, at least one witness, and the patient himself. Physical examination with neurological evaluation was performed by three experienced neurologists (ES, WM, ASW).


HN represents a repetitive short loss of neck muscle tone resulting in a nodding of the head, sometimes associated with a short loss of muscle tone of the upper extremities. Loss or impairment of consciousness may be present, but not always. To date HN is not mentioned in any classification and it remains unclear whether it represents a seizure disorder and if so, whether it belongs to the group of generalized or partial seizures.

We further specified whether brain damage or signs of focal neurological lesions were present. A subgroup of HN seizures was termed “HN plus” syndrome consisting of additional partial or generalized non-HN epileptic seizures.

Analysis of skin

In 51/62 (81.6%) patients, skin snips were taken from the right and left spina iliaca posterior superior. Eleven patients did not consent to further investigations. If there were skin changes supportive of O. volvulus, a skin snip was taken from that site too. Microscopic evaluation of microfilaria (mf) was performed at 40× magnification 1 and 4 h later. If no mf was seen, the well was checked again after 12 h. The mf density per person was calculated. The O. volvulus PCR was performed after a modified version of that used by Zhang et al. (Zhang et al., 2000). Methods were optimized; quality controls and specificity tests with different parasites were performed to exclude cross reactivity. Gel electrophoresis was applied to visualize the amplification products.

Analysis of blood

In 51/62 (81.6%) patients, total white and differential blood counts were performed according to conventional, standardized methods. The IgG4-enzyme-linked immunosorbent assay (ELISA) for O. volvulus was performed according to Lal and Ottesen (1988). Factors greater than 10 were considered positive.

Analysis of CSF

Forty-eight/62 (77.4%) patients with HN underwent CSF analysis (14 patients did not give consent). Cells were counted in the Fuchs-Rosenthal chamber. Their number was given as cells/μl CSF. Total protein was measured. Glucose, albumin, and IgG (unspecific and specific for O. volvulus; n = 47) concentrations of serum and CSF were determined. The antibody index (AI) for O. volvulus was calculated according to the method of Reiber and Lange (n  = 19; Reiber & Lange, 1991). AI > 4 was considered positive. The ELISA and PCR for O. volvulus were performed according to the methods mentioned for blood and skin.

Performance of EEG and MRI

Selected patients underwent MRI (n = 12) and interictal EEG (n = 10). MRI and EEG facilities were located over 300 km away in Dar es Salaam. Traveling on dangerous nontarmac roads was difficult, thus only some of those originally interviewed could be examined with EEG and MRI. Patients were chosen randomly taking into consideration informed consent and the aim to recruit equal numbers in the groups “HN only” and “HN plus” syndrome. EEGs were undertaken at The Regency Medical Center, Dar es Salaam, using a digital Biologic Ceegraph computer system. Silver-silver chloride electrodes were applied using colloidal solution according to the 10-20 placement system. Activation techniques, including hyperventilation for 3 min and photic stimulation were performed. Analysis, reading, and interpretation was done by two experienced neurologists (IU and WM). The MRI scans were performed in the Department of Radiology at The Aga Khan Hospital, Dar es Salaam, using MRI 1.5 tesla GE. All scans were evaluated by two experienced radiologists (JD and TG).

Ethical clearance

The study was cleared by the Ethics Committee of the Muhimbili University College of Health Sciences, University of Dar es Salaam. Oral informed consent was given by all patients allowing us to transcribe clinical data into the HN questionnaire. Written and witnessed informed consent was obtained from the patients or, in case of children or noncompetent patients, their respective parents or next of kin for all invasive procedures.



The average age at the time of interview was 14.9 ± 4.3 years (median age: 14 years, range: 8–32 years) with a 3:2 female preponderance. Most of the patients (58/62) belonged to the Pogoro tribe. At the time of interview and at the time of diagnosis of epilepsy, most of the patients were between 11 and 15 years (43.5% and 48.4%). In 50% of all patients, the first HN attack happened between the ages of 6 and 10 years, followed by the age group 11–15 years (37.1%). Only one patient had his first HN seizure when aged over 15 years.

Seizure classification

Twenty-eight (45.2%) patients had HN only, most of them without brain damage or without focal neurological signs. Another 28 patients suffered from “HN plus” with one other type of seizure apart from HN. The majority—19 patients—had generalized seizures without brain damage or focal neurological signs. “HN plus” with two other types of seizure was found in six (9.7%) patients (see Table 1).

Table 1.   Distribution of diagnostic groups
  1. aTwo patients had additional focal neurological signs.

  2. bPatient had additional focal neurological signs.

“Head nodding only” 2845.2
 HN only without brain damage and without focal neurological signs2337.1
 HN only with brain damage but without focal neurological signs34.8
 HN only with focal neurological signs but without brain damage23.2
“Head nodding plus”2845.2
 with one other type  
 of seizure  
 HN without brain damage + generalized seizures without brain damage1930.7
 HN with brain damagea + generalized seizures with brain damage711.3
 HN without brain damage + complex partial seizures11.6
 HN with brain damage + complex partial seizures11.6
“Head nodding plus”69.7
 with two other types  
 of seizure  
 HN without brain damage + generalized seizures without brain damage + complex partial seizures46.5
 HN with brain damageb + generalized seizures with brain damage + complex partial seizures11.6
 HN with brain damage + generalized seizures with brain damage + pseudoseizures11.6
Total 62100

Descriptive features of HN seizures

In all cases, a loss of muscle tone of the neck and consequently a nodding of the head was reported. Additional loss of muscle tone of the upper extremities was noted in 37 (59.7%) patients while HN. Consciousness was impaired in 11 (17.7%) patients.

A regularly provoking factor for HN seizures was given by 11 (17.7%) patients. Food was the commonest precipitant in all diagnostic groups, quoted by 9 of the 62 patients. In the diagnostic group “HN plus” with two other types of seizure, food made up a significant proportion (2/28 “HN only,” 4/28 “HN plus” with one other type of seizure, 3/6 “HN plus” with two other types of seizure; Fisher's exact test, p = 0.035). Two patients with “HN only” got their seizures during cold weather or bathing in cold water. No other precipitants were reported. It was noticeable that in January during the rainy season the frequency of HN seizures increased in many patients.

Supportive signs of epileptic seizures were described by 15 (24.2%) patients while HN (see Table 2). There was no statistical difference between the diagnostic groups (5/28 “HN only,” 10/28 “HN plus” with one other type of seizure, 0/6 “HN plus” with two other types of seizure; Kruskal–Wallis test, p = 0.218). Of the 62 HN patients, 27 (43.5%) reported postictal reorientation after HN seizures, the duration ranging from a few minutes up to 2 h (average duration 51.6 ± 47.4 min). Tiredness and sleeping were the commonest symptoms (25/62), followed by headache and confusion (4/62). Three patients stated less common symptoms such as crying and dizziness. There was no statistical difference regarding symptoms during postictal reorientation between diagnostic groups (12/28 “HN only,” 12/28 “HN plus” with one other type of seizure, 3/6 “HN plus” with two other types of seizure; Kruskal–Wallis test, p = 0.983).

Table 2.   Supportive signs of epileptic seizures in people with head nodding
  “Head nodding plus”
 “Head noddingwith generalized
Supportive signsonly” (%)seizures (%)
Drooling of saliva3 (10.7)3 (8.8)
Urine incontinence2 (7.1)5 (14.7)
Drooling and urine0 (0) 2 (5.9)
 incontinence0 (0) 2 (5.9)
Total5 (17.9)10 (29.4)
Did not comment13 (46.4)8 (23.5)

In most patients, HN attacks appeared prior to the other types of seizure and continued together. In 26 (92.9%) patients of the group “HN plus” with one other type of seizure, the HN seizure started 2.5 ± 4.1 years before the other seizure appeared. In two patients of this group, seizure history started with generalized seizures. In seven (25.0%) patients, HN seizures stopped a few months to a few days before the first generalized seizure occurred. A decreasing frequency of HN seizures after generalized seizures had started was reported by three (10.7%) patients. In five (17.9%) patients, HN was noted before each generalized seizure. The other patients did not report a specific order of appearance.


With the exception of three (4.8%) patients with “HN only” all were on antiepileptic therapy during the time of the study. Twenty-seven (45.8%) patients were taking a combination therapy of phenobarbitone and phenytoin with an average dose of 76.9 ± 50.4 mg and 170.4 ± 65.4 mg, respectively. Twenty-four (40.7%) patients were taking phenytoin and six (10.2%) patients phenobarbitone alone with an average dose of 191.3 ± 51.5 mg and 76.7 ± 35.0 mg, respectively. Carbamazepine was started in two (3.4%) patients. Before treatment, the average frequency of HN seizures was 22.1 ± 21.7 seizures per month in all patients. After treatment it was 10.9 ± 11.5 HN seizures per month. The reduction was highly significant (t-test, p < 0.001). In the 34 patients with other types of seizure apart from HN, reduction in the frequency of seizures other than HN was reported in 15 (44.1%) and no change in 5 (14.7%) patients. Three patients (8.8%) experienced worsening of the frequency when on treatment. They were all on a combination therapy of phenytoin and phenobarbitone. In 11 patients, no further details were available.

Family history

Fifty-two (83.9%) patients confirmed having at least one family member suffering from any kind of epilepsy, including HN seizures. Only one relative in the family with epilepsy was reported by 21 (40.4%) patients; the other 31 (59.6%) had two or more family members with epilepsy. Within the “HN only” group, there were 3 relatives with HN only, 4 with “HN plus” syndrome, and 37 with other types of seizures, mainly generalized in nature. In patients with the “HN plus” syndrome, 6 relatives had HN only, 8 suffered from the “HN plus” syndrome, and 41 relatives reported other types of seizures, again mainly generalized in nature. In terms of family history, there was no  significant difference between the three diagnostic groups (see Table 1; Kruskal–Wallis test, p = 0.533).

Laboratory results

Thirty-one (60.8%) of 51 patients had mf visible on microscopic investigation of the skin. The mean mf density in the skin of these people was 3.6 ± 3.9 mf/mg skin. The PCR showed that 12 of the 20 people without microscopically visible mf had traces of O. volvulus DNA in the skin raising the number of people carrying the parasite to 43 (84.3%). All but two patients (3.9%) had normal total white blood cell counts (reference range by Lugada et al., 2004). Fourteen (27.5%) patients had an elevated neutrophil count and 28 (54.9%) showed elevated percentages of eosinophils. Forty-four (86.3%) patients had a positive result in the O. volvulus serum ELISA. In 48 patients, spinal tap was performed. Three (5.9%) showed an elevated cell count (>5 cells/μl) with 6, 8, and 28 cells/μl, mainly lymphocytes and monocytes. The CSF of the patient with 28 cells/μl was contaminated with blood. Glucose ratio, protein content, and albumin ratio were all normal. Three patients (5.9%) had an IgG-Index above the laboratory reference range of 0.67 with 0.69, 0.78, and 0.96. In 19 patients, the AI was calculated from the serum and CSF results of the O. volvulus ELISA, only one patient showed a borderline increase of 2.6. CSF PCR was negative in all patients.

MRI results

Of the seven patients with “HN only,” four had a normal scan. Of these, one had shown frontal release signs and upper motor neuron signs bilaterally on neurological examination. In the three other patients, hippocampus pathologies were seen (two bilateral). In two patients, there were additional gliotic lesions (see Table 3). One patient showed upper motor neuron signs unilaterally, whilst the other two had normal neurological examination. None of the patients in the “HN plus” group had a normal MRI. For the exact distribution of seizure types, mental state, neurological signs, and MRI pathologies see Table  3.

Table 3.   Characteristics of patients undergoing MRI
  Neurological/mentalAsymmetrical MRISymmetrical MRI  
Age (yrs)GenderexaminationpathologiespathologiesMf skin/mgPCR skin
  1. aDot like hyperintensities on T2-weighted images and FLAIR.

  2. UMN, upper motor neuron; rt, right; lft, left; HC, hippocampus; gwm-j, gray/white matter junction; MF, microfilariae; neg°/pos°, no further counting available; PCR, polymerase chain reaction; NAD, nothing abnormal detected.

Patients with “head nodding only”      
16FNADGliosisa rt frontalHC sclerosis rt>lft5.7Pos
18MFrontal release signs,NoNo1Neg
   UMN signs lft>rt    
10FNADNoNo0Not tested
14FNADNoNoNeg°Not tested
15MNADHC sclerosis rtNo0Pos
9FUMN signs rtGliosisa rt frontalHC atrophy rt>lftPos°Not tested
Patients with “head nodding plus”
 syndrome (with other types
 of seizure)
4MNADNoGliosisa bilateral lft>rt1.5Pos
     frontal subcortical  
16FNADGliosisa lft temporo-No6.2Pos
    occipitoparietal gwm-j   
13MNADNoGliosisa bilateral frontal0Neg
16FNADSignal enhancement HC lftNo0Pos
21MDiffuse brain damage,NoHC sclerosis bilateral0Pos
   no focal neurology    

In total, three HN patients irrespective of their type had gliotic lesions and another three had hippocampus pathologies either uni- or bilaterally. Two patients showed both types of lesions. There was a significant association between lesions on MRI and positive skin PCR (chi-square test, p = 0.023), but not between MRI pathologies and mf density (Mann–Whitney U test, p = 1.0). All four patients with normal MRI had either a negative skin PCR or, if PCR was not available, a negative skin snip microscopy for O. volvulus. Six of the eight patients with lesions on MRI had a positive skin PCR and one patient was not tested, but was positive on microscopy. Only one patient with gliotic lesions seemed to be free from infection with O. volvulus (see Table 3). Interestingly, the “HN plus” group showed significantly more lesions on MRI compared to the “HN only” group (Mann–Whitney U test, p = 0.047; see Table  3).

EEG results

Four (40.0%) of 10 patients with HN with available EEG results had a normal interictal EEG. Of these one patient was suffering from generalized seizures and HN seizures, the others from HN only. In two of the six patients with HN and abnormal EEG, interictal epileptic activity could be recorded, showing intermittent generalized slowing and sharp wave activity. One patient with “HN only” had neurological signs, and the other had neither focal neurological deficits nor significant mental impairment. EEG findings in the other patients were rather unspecific; they are listed in Table 4. Only one patient with slow basic rhythm in the group “HN plus” was mentally retarded.

Table 4.   EEG results (interictal)
  “Head nodding plus”
 “Head noddingwith generalized
EEG findingsonly” (%)seizures (%)
  1. aMultiple answers possible.

Slow basic rhythm0 (0.0)2 (33.3)
Diffuse slowing0 (0.0)1 (16.7)
Unilateral slowing1 (20.0)0 (0.0)
Intermittent gen. slowing0 (0.0)1 (16.7)
Intermittent gen. slowing1 (20.0)1 (16.7)
 and sharp waves  
Normal3 (60.0)1 (16.7)
Totala5 (100) 6 (100) 
Number of patients55


Our clinical as well as electrophysiological findings support the hypothesis of a seizure disorder, although other possibilities such as negative myoclonus cannot be excluded with certainty without the appropriate diagnostic tool at hand. Almost 25% of our patients reported supportive signs of epileptic seizures, 18% had impaired consciousness during the HN seizure, 43.5% mentioned symptoms suggesting postictal reorientation and 2 of the 10 patients who underwent EEG showed interictal epileptic activity. Complex partial as well as generalized seizures were present in association with HN seizures in over half of the patients. HN attacks mainly preceded the later evolving complex partial and generalized seizures. This further supports the hypothesis of a true seizure disorder. Food and cold weather as precipitating factors for HN seizures, which were present in 18% of our patients, have already been described (Jilek et al., 1970 ; Lacey, 2003). The significance of these signs however remains unclear. A nutritional factor seems to be unlikely as HN happened before and while eating and over 80% of our patients reported no precipitating factor at all.

As to the classification of HN, it has been associated with various epilepsy syndromes such as myoclonic astatic epilepsy (Kubota et al., 2004), myoclonic epilepsy in infancy (Aso et al., 1994), juvenile myoclonic epilepsy (Aso et al., 1994), unclassified photosensitive epilepsy in children (Aso et al., 1994), West's syndrome (Pranzatelli, 2002) and complex partial seizures of temporal lobe origin (Brockhaus & Elger, 1995), but also less severe syndromes such as benign myoclonus of early infancy (Maydell et al., 2001), and spasmus nutans (Aung et al., 1996). Most of those have a specific pattern on EEG, however EEG may be unspecific or even normal, as seen in our patients (Brockhaus & Elger, 1995; Maydell et al., 2001; Pranzatelli, 2002; Kubota et al., 2004). Interestingly, a report from Japan on atonic epileptic drop attacks showed HN without falling as an aberrant pattern of the drop attacks. The associated EEG pattern showed generalized spike-and-slow wave complexes (Oguni et al., 1997). The latter was also seen in two of our patients.

In terms of MRI findings, the high prevalence of hippocampus sclerosis in our patients (42%) is intriguing. In children, complex partial seizures of temporal lobe origin were found to include HN as a semiologic feature, which normally is not seen in adults (Brockhaus & Elger, 1995). Furthermore some of our patients, over time, showed an evolvement of the HN seizures either into complex partial or generalized seizures. In West Uganda, which is also highly endemic for O. volvulus, HN seizures with and without clinical features of complex partial seizures have been observed (Kaiser et al., 2000).

Taking our results together, we were unable to ascertain whether HN seizures are focal or generalized in nature. There clearly is evidence for both. Hippocampus pathology in five patients and unilateral slowing in one patient would support a focal origin, whereas there was interictal generalized epileptic activity in two patients.

In our study, we found familial clustering of HN and generalized seizures, so a genetic background of HN may be possible and indeed some of the above quoted seizure disorders are known to have a genetic component. Other genetic syndromes such as the Aicardi and the Pitt-Rogers-Danks syndrome not only show myoclonic seizures as a core feature, but also a varying combination of psychomotor and mental retardation, stunted growth, and dysmorphic facial features, reminiscent of the Nakalanga syndrome, which has been observed in areas highly endemic for O. volvulus, and “nodding disease” as described in Sudan (Abe et al., 1990; Ovuga et al., 1992; Newell et al., 1997; Fagan et al., 1998; Lacey, 2003).

Whether O. volvulus, as an infectious agent, plays a role in the genesis of HN is debatable. Similar to the Nakalanga syndrome and “nodding disease” in Sudan, 40% of our  patients were mentally retarded, half of them severely. Patients also were small in stature and some faces showed seemingly dysmorphic features. Interesting, however, is the fact that despite their very young age, more than half of the patients with HN had gliotic lesions on MRI and that lesions in general were associated with positive results on skin PCR. PCR of the CSF, however, was negative in all patients, and there was no major abnormality in other CSF parameters, including the AI for O. volvulus. The high prevalence of O. volvulus in the skin and antibodies in the serum of our patients indicates infection with the parasite. The elevated eosinophil count in the blood further points toward helminthic infestation, which however may be unspecific. Considering the fact that PCR results of the CSF and AI for O. volvulus were negative, invasion of mf, past or present, into the central nervous system could not be demonstrated. Further studies should however be encouraged in order to clarify this point.

Our study represents a comprehensive report on clinical details as well as possible pathogenetic aspects of the HN syndrome. Causes of HN according to our results may be three-fold; it may represent complex partial seizures with corresponding hippocampus sclerosis; a genetic syndrome as supported by high clustering of family members with epilepsy seems possible; as to the role of O. volvulus, results are controversial. Evidence of invasion of the parasite into the central nervous system, however, could not be ascertained. Further studies are definitely needed in order to better understand the etiology of this potentially new epilepsy disorder.


We are very grateful to the staff of The Mahenge Epilepsy Clinic as well as to the staff of The Mahenge District Hospital who were ever so helpful. Our special thanks goes to all the patients and their relatives, who participated in our study. In the context of our study, one should not conclude without mentioning the tireless efforts of Professor Louise Jilek-Aall, founder of the Mahenge Epilepsy Clinic, in caring for the people with epilepsy of the Mahenge area and initiating multinational research projects to raise awareness of their suffering. The study was supported by the Savoy Epilepsy Foundation, Quebec, Canada. ASW was supported by the Center for International Migration (CIM), Frankfurt, Germany.

Conflict of interest: “We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.” The authors have no conflicts of interest to declare in relation to this study.