Ethics statement

The present study was approved by the Mozambique National Bioethical Committee of Health (51/CNBS/2017) and by the administrative authorities from Zambézia province and community leaders. The study also received approval from the ethics committee of the Klinikum rechts der Isar, Technical University of Munich, Germany (537/18 S-KK).

Prior to the interviews and blood sampling, written consent were obtained from the enrolled participants. Written consent from children aged below 12 years and illiterate participants were obtained in the presence of a guardian or witness, who signed on their behalf. For participants aged between 12 and 17 years a written assent was obtained in addition to their parents’ consent. Each participant was assigned a unique identification study number and all data and samples were handled confidentially using this study number from this point onward.

Participants who suffered from epileptic seizures/epilepsy or other neuropsychiatric disorders were referred to the local health centre for diagnosis confirmation and follow-up according to the Mozambique Ministry of Health guidelines [16].

Study area and population

The study was conducted in seven villages of Mocuba district (Fig 1) located in Zambézia province, central Mozambique (16°50’ 13” S, 36°59’ 14” E). The district has an area of 8,803 km² and an altitude of 200–400 meters above sea level.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Location of the study area.

Created by Patricia Noormahomed based on EarthExplorer (https://earthexplorer.usgs.gov).

https://doi.org/10.1371/journal.pntd.0010606.g001

The total population is estimated at 422,681. There is an illiteracy rate of 62.5% with most of the population (82.6%) living in rural areas. Moreover, the disease profile of Mocuba is dominated by infectious diseases including many NTDs with higher prevalence estimates. As an example, the district has an HIV prevalence of 9% and 39% of children are infected by at least one intestinal parasite [17–21].

Study design and sampling

A community-based cross-sectional study was conducted between March and May 2018. The sampling frame was constructed using information of the National Institute of Statistics [22]. For the sample size calculation, it was estimated that 15% of households would have at least one member with cysticercosis based on a study conducted in a northwestern district of Mozambique [14]. Assuming a 95% confidence level and 20% precision, and correcting for a finite population of households in the study area [22], a sample size of at least 543 households was obtained by using the formula n = Z²p(1—p)/d² [23], where n is the sample size required, Z is the z–score for the desired confidence level, p is the expected proportion of households, and d is the desired precision relative to the expected proportion of households. The sampling procedure followed a two-stage household-based design, whereby Primary Sampling Units (PSUs) consisted of villages, and households were chosen during the second stage. Prior to selection of households, the village authorities were approached and the purpose of the study was explained. Once the permission was granted, a systematic random sampling with probability proportional to size (PPS) technique was used to select households. Two eligible participants from each household were recruited randomly into the study (Fig 2). When only one member was available in a selected household, either by absence or refusals of other members, participants were recruited from the nearest household. Inclusion criteria were living in the household and being at least 2 years old. People who had taken alcoholic beverages in the previous 24 hours were excluded. In total, 1,086 participants from 565 households were included in the study.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Data collection workflow of the study sample.

https://doi.org/10.1371/journal.pntd.0010606.g002

Questionnaire survey and vignettes

A pre-designed questionnaire developed by the Cysticercosis Working Group for Eastern and Southern Africa (CWGESA) was used, with some modifications, that included four vignettes on neuropsychiatric disorders [24,25]. The four vignettes included chronic headache, epileptic seizures/epilepsy (generalized and focal) and psychotic disorders, and were confirmed with a neurologist (ASW). This was the major tool used to establish a diagnosis of neuropsychiatric (= neurological and psychiatric) disorders in the study. The questionnaire also covered information on socio-demographic and risk factors associated with cysticercosis such as age, gender, education, sources of drinking water, use of latrines, rearing pigs and pig management practices, consumption of pork, awareness about TSC transmission, risk-related habits and practices.

Furthermore, direct observation to visualize the existence of latrines and usage and pig husbandry system in the household was made by the enumerators. The questionnaire designed in Portuguese language was used for the study, translated to the local language (Lomué) and back translated to Portuguese.

The survey was carried out by locally selected enumerators with knowledge of the local language, experience with community surveys, and basic training on TSC transmission, prevention and clinical signs/symptoms of selected neuropsychiatric disorders. The training of the enumerators was done via workshop sessions and pre-tested prior to commencement of the study.

Vignette 1. Assessment of chronic headache

To assess chronic headache, participants were asked if they had a history of daily headache that occurred for at least 15 consecutive days and for longer than three months. We did not ask for severity or any specific characteristics of headaches in order to keep it more general and allow for secondary causes of chronic headache such as NCC.

Vignette 2. Assessment of generalized tonic-clonic seizures

To assess generalized tonic-clonic seizures (GTCS), we asked the participant if he/she had ever experienced a sudden loss of consciousness that caused him/her to fall to the ground or whether he/she was told about such an event by someone else. We also asked if the eyes were turned upwards, the body became rigid and/or involuntary violent movements of the arms and legs were observed. In addition, we asked if they were told or noticed that they were drooling and/or urinating and/or biting their tongue. We also asked if the person was asleep or slowed down or even confused after the actual epileptic seizure and could not remember what happened.

Vignette 3. Assessment of focal seizures

To assess focal seizures (FS), participants were asked whether they had ever noticed or been told about uncontrolled movements in body parts such as an arm or leg without loss of consciousness or before loss of consciousness. For the latter, we also asked about warning signs such as a strange smell (aura) before the epileptic seizure.

Vignette 4. Assessment of psychotic disorders

To assess psychotic behaviour and/or hallucinations, we asked the participant if he/she was informed of any changes in behaviour, such as becoming very unstable, appearing anxious and/or doing inappropriate things, such as undressing or harming others for no reason. They were also asked if they heard strange voices that no one else could hear, or if they saw things that were not real to others. They were also asked if they had the impression of being followed.

Blood sampling

A 5 ml sample of venous blood was obtained by venepuncture from each of the recruited individuals after consenting [26]. The samples were immediately stored in cooling boxes at 4°C and transported to the Mocuba district Hospital Laboratory, where they were centrifuged for 5 min at 1500 r.p.m. to obtain serum. Serum was stored at -20°C until shipment to the Parasitology Laboratory at Faculty of Medicine, Eduardo Mondlane University (UEM) in Maputo where samples were stored at -80°C until further processing.

Serological testing

T. solium circulating antigens (Ag) were detected using the monoclonal antibody based B158C11A10 Enzyme Linked Immunosorbent Assay (Cysticercosis Ag-ELISA, ApDia, Belgium). Ag index was calculated as the mean of optical density (OD) of each serum sample divided by the cut-off. In addition, IgG antibodies (Ab) against T. solium larva were determined via Western Blot (LDBIO Diagnostic, Lyon, France). A sample was considered positive if at least 2 bands (6–8 kDa; 12 kDa; 23–26 kDa; 39 kDa; 50–55 kDa) were visible, according to the manufacturer’s instructions.

Statistical analysis

Data were captured in EpiData Entry Client version 4.3 and analysed using STATA version 13.0 (Stata Corp., College Station, TX). Descriptive analyses were first performed by calculating frequencies and percentages for categorical variables. Age was categorized into age groups (2–14, 15–24, 25–54 and >55 years) for the purpose of analysis. Using Chi-square test, bivariate analysis was performed to assess associations between a positive Ag-ELISA and/or Western Blot assay results and several factors. These factors included: socio-demographic and neuropsychiatric diseases; knowledge about T. solium; transmission risk-related variables and pig management variables. Chi-square tests were also used to assess difference in seroprevalence by district. Logistic regression models were performed to assess the association of seroprevalence of T. solium with the four vignettes. Models were run unadjusted and adjusted for age, gender, education and occupation. The significance level was set at p<0.05.

Serology

Our findings revealed that human cysticercosis was prevalent in the setting of our study (10.3%), although lower than previously reported in Angónia district and Beira city, Central Mozambique [14,15]. The prevalence is still worrying as it demonstrates the potential for a greater spread of the disease, which can be caused by several factors, including hygiene and sanitation. The higher prevalence in some municipalities such as Mocuba (17.7%) in comparison with those with lower prevalences (e.g. Alto Benfica, 3.6%), may be explained by sociodemographic factors, which should be analysed in future studies.

Previous studies from Mozambique found in general higher prevalence of either antigens (15%) or antibodies (10%) to T. solium larva than the ones reported in this study [14,15], and were consistent with studies done in other settings such as Cameroon [27], Nigeria [28], Vietnam [29,30], Zambia [31] and Democratic Republic of Congo [5], which reported similar seroprevalence rates. In our study, only 12.5% of the screened sera tested simultaneously positive to both serological assays. The differences in the prevalence of antigens (4.9%), that indicates a current infection, and antibodies (6.2%), which indicates a previous or current exposure to infection, may be influenced by the number, location of cysts and their larval stage. Furthermore, antibodies can persist longer, even after the clearance of the infection, while the presence of antigens is only detected when viable parasites are present [1–3,10,32,33]. Moreover, the western blot assay we used detected only three out of the seven glycoproteins described to be specific for cysticercosis [1,32,34]. So, it is possible that the prevalence of antibodies found in our study was underestimated.

In addition to that, the differences in study design, HIV serostatus characteristics of the study participants (immunocompetent or immunocompromised), serological assays used for each study, and parasite genetic diversity (which ultimately influenced the sensitivity and specificity of the serological assays) can explain the differences found within studies in the same region and in other settings [5,9,10,35]. Of note, that to date, we do not have systematic studies done in Mozambique accessing the existing genetic variants of T. solium, except for the one in which phylogenetic studies of cysts from Central Mozambique revealed the existence of the Latin variant [32]. As Mozambique has been exposed to migration from Europe and Asia, we cannot exclude the possibility that Mozambique has both variants circulating in different regions and further studies should be done in order to provide relevant information about this subject.

Cysticercosis serology and neuropsychiatric disorders

Our study also demonstrated that chronic headache (49.5%), generalized epileptic seizures (27%), focal epileptic seizures (18%), and psychosis (18.3%) affected a substantial proportion of study participants irrespective of their cysticercosis serostatus. Our findings are consistent with data from 2018 hospital records, indicating that Mocuba district had the highest number of patients (7,416; 75.3%) in the entire province (9,844) with mental disorders. Out of these 7,416 mental health out-patients, 1,863 (25.1%) had epilepsy, 250 (3.4%) had mental retardation and 1,401 (18.9%) had psychosis [36].

In our study, epileptic seizures were significantly more prevalent in younger participants (2–14 years) and decreased with age (p<0.001). This finding is consistent with recent mental health data in Mozambique showing that among people living with epilepsy, 54% were children aged up to 15 years, although the underlying causes of these conditions are not well defined [21,37]. Similar studies in African countries also found that epileptic seizures and epilepsy were the main leading causes of mental health diagnoses in out-patients’ consultations and that the majority of people affected are children and adolescents [14,16,37,38]. Although we did not find significant associations between serology and epileptic seizures, as with psychosis, our findings therefore present a very high prevalence of epileptic seizures (both generalized and focal) that occur independently of seropositivity to cysticercosis and thus could be caused by other underlying diseases such as cerebral malaria, head trauma or perinatal hypoxia in children, to name a few.

When compared with other similar studies, our findings here corroborate with results from Bangoua, Cameroon [27], which did not find any statistically significant difference in the seropositivity to T. solium larva between people with epileptic seizures (1/61, 1.6%) and those without (5/323, 1.5%). Similarly, another study in the Democratic Republic of Congo [5] revealed that there was no difference in seropositivity between people with neurological disorders and those without. In addition, there was no association between cysticercosis and epilepsy in a study conducted in the Gambia [39]. To the contrary, it is well documented in the literature from Latin America, Asia and Africa, including the study done in Angónia district, that seropositivity to circulating antigens or antibodies of cysticercosis were associated with epilepsy in community-based studies [14,40,41,42]. Furthermore, the odds ratio from a study in Nigeria revealed that people with epilepsy in the study area with a relatively high cysticercosis prevalence, were two times more likely to be seropositive than those without [28].

In our study, chronic headache was significantly associated with circulating antigens (p = 0.013). In fact, other studies performed in Mozambique, Burkina Faso and Tanzania also found that seropositivity to cysticercal antigens was associated with severe headache [14,15,42,43]. Nevertheless, in other settings such as in Anzoátegui in Venezuela, Vietnam and Nabo Village in Tiandong, China, no statistical association was found between these two variables [44–46]. Our findings may reflect the fact that headache may be correlated to the pathogenesis of the disease in this study setting.

In addition to the high burden of epileptic seizures and chronic headache, we also found that psychotic manifestations were highly prevalent in the setting of our study (18.3%), confirming the anecdotal reports from the Mocuba district community leaders of having a high proportion of community residents with those conditions. As with the other two disorders, psychotic disorders do not seem to be related to cysticercosis seropositivity either. Psychotic disorders are understudied in Africa, as revealed by a recent scoping review [47], that found only nine studies in which the prevalence of psychotic disorders was analysed in Africa from 1984 to 2020. The available reports describe a prevalence ranging from 1% to 4.4% in rural regions. The only published population-based study done in 2003 in a rural (Cuamba) and urban (Maputo) area of northern and southern Mozambique, respectively, found that seizure disorders (4.4% vs 1.6%), psychosis (4.4% vs 1.9%) and mental retardation (1.9% vs 1.3%) were higher in rural compared to urban areas [24] and well below the results of our study and recent hospital data from Zambézia Provincial Directorate [36]. It is difficult to explain this high burden. Apart from the possibility of other CNS pathogens causing epileptic seizures/epilepsy and/or psychosis, such as Toxoplasma gondii, Toxocara spp. and Onchocerca volvulus (the latter was recently identified in Zambézia province) [20,21,48,49], we think that the prolonged armed conflict that affected this region for decades in some of the study villages [50,51], combined with the stress it provokes in the affected population, might be playing a role as causative agent for this high prevalence of psychosis. Indeed, the effects of armed conflict on mental health within the affected communities is well documented in the literature, as it causes disruption of socio-economic and political infrastructure including the appropriate delivery of health care services, which was the case in our study setting and other regions of Mozambique [50–53].

Risk factors and multivariate logistic regression

In our study, patients of older age were more likely to be seropositive for the parasite. Regarding age, the results are important because they will allow us to understand the transmission dynamics of TSC in the study region, and to identity the age groups at higher risk. It is likely that infection events accumulate as people age [54], and that T. solium cysticercus antibodies might persist for several years [55]. Similarly to our results, studies from Tanzania [43], Ecuador [56], Zambia [31], Peru [57], Burkina Faso [58] and Vietnam [29], including a systematic review [59], revealed an association between increasing age and seropositivity to cysticercosis, except for a study conducted by Noormahomed et al. [15], in which age was not significantly associated with the seroprevalence of the helminths studied, including cysticercosis.

Furthermore, in concordance with the present study, there was no statistically significant difference between gender and seropositivity in Ecuador [56] and Zambia [31]. To the contrary, in Tanzania [43], females were more likely to be seropositive, while in Burkina Faso and Vietnam this was true for males [29,58]. On the other hand, neither gender nor age presented a statistically significant association with seropositivity to cysticercosis in other studies [30, 44,46, 60].

Analysing the knowledge of T. solium, although our results did not present a statistically significant association with TSC seropositivity, in other countries such as Nigeria [28] poor knowledge of cysticercosis and improper pork preparation were identified as main risk and behavioural factors contributing to the high prevalence.

Other potential risk factors such as consumption of pork [61], pig keeping, use of unsafe water, faecal disposal and types of latrine were not statistically significant in our study, similarly to the study done in Angónia district [14]. This was expected since the study participants share the same environmental conditions, and are equally exposed to the same risk factors and consequently equally exposed to the infection. Furthermore, although there is a fraction of Muslim people amongst our study participants, we did not expect our findings to be affected by the religion, since cysticercosis is acquired mainly through water or food contaminated with T. solium eggs, and consequently anyone can acquire the infection regardless of the consumption of pork meat or professed religion. In some studies, food, water and hygiene related factors were significantly associated with cysticercosis seropositivity, including consumption of undercooked pork [44], and using the same water source [43]; although in other studies, factors such as consumption of raw meat and/or vegetables [30] and hygiene (presence of latrine) [57] were not associated with cysticercosis seropositivity.

Strengths and limitations

The strength of this study is reflected in the fact that, to the best of our knowledge, this is the first assessment done in Mozambique aiming to define the seroprevalence of cysticercosis and its possible association with some selected neuropsychiatric disorders. It is therefore the first pilot community-based study representative of the Mocuba district population including rural and urban settings and all age groups except those under 2 years old. This baseline information will be useful to define strategies for control and mitigation of TSC in this district and to effectively diagnose and treat neuropsychiatric disorders that might be associated with cysticercosis. Additionally, further studies can be developed in other districts and provinces, taking into account the population at risk.

One of the limitations was the lack of clinical examination and diagnostic tools required to confirm the diagnosis of neuropsychiatric disorders. Also, patients with neuropsychiatric disorders with positive or negative serology for cysticercosis did not undergo CT or MRI examination in order to confirm or not the diagnosis. Regarding the serological assays, these present limitations in terms of sensitivity and specificity as well as on the discrimination of past or current infection. Additionally, serological screening for malaria, toxoplasmosis, toxocariasis, and onchocerciasis, that may cause neuropsychiatric disorders, was not done. Thus, it is not possible to draw firm conclusions about the fraction of participants whose neuropsychiatric signs/symptoms could have been due to cysticercosis/NCC or its sequelae.