NEUROCYSTICERCOSIS IN CHILDREN PRESENTING WITH AFEBRILE SEIZURE: CLINICALPROFILE, IMAGING AND SERODIAGNOSIS

INTRODUCTION

Worldwide, one third of the total epilepsy cases arise in childhood andneurocysticercosis (NCC) is the major cause particularly in developing countriesincluding India and Latin America²⁰.³⁰.³⁴. NCC is caused by the metacestode larval stage of Taenia solium,and it affects patients of all ages⁴. In India, NCC is the most common cause of convulsions and hydrocephalus inadults and is the single most common cause of community acquired active epilepsy¹⁶^,²². NCC patients may remain asymptomatic for months to years, and commonly adiagnosis is made incidentally when neuro-imaging is performed. However, itsmanifestations are variable and somewhat different when it occurs in children²⁴^,²⁸.

NCC in children is believed to be infrequent, but, as methods of diagnosis(imaging techniques and immunological tests) have improved, it has become possible toconfirm more cases²¹. The serological tests are considered to be helpful as adjunct tests in order toconfirm clinical diagnosis of NCC where Enzyme linked immunosorbent assay (ELISA) andenzyme linked immunotransfer blotting (EITB) are most frequently used for detection ofantibodies in serum and/or CSF⁹^,¹⁴^,³⁴. Though the latest diagnostic method approved by the Centers for Disease Control(CDC) for a confirmatory diagnosis of NCC is based on EITB, its cost and nonavailability of local suppliers limits its utility in India. In our study a sensitiveELISA kit is employed as a relatively lower cost screening test. There is insufficientinformation about NCC in children from Andhra Pradesh State in India. However, there area couple of reports of T. solium cysticercosis in this state since alot of adult cases of NCC as well as eye cases were reported earlier¹³^,¹⁸^,¹⁹ that indicate a possible prevalence of cysticercosis in children in the sameregion of this country. With this background, the objective of the present study was setto screen for anti-Cysticercus antibodies in serum to estimate the NCC prevalence as anunderlying etiology of afebrile seizure in children by ELISA.

MATERIALS AND METHODS

The present study was conducted in a teaching hospital in Andhra PradeshProvince in India after due approval of the protocol and permission where a total of 61pediatric subjects presenting with afebrile seizure were recruited in a two yearsperiod. Informed consent was obtained from the parents or legal guardians of each of thestudy subjects. Detailed history was then collected from each patient and/orguardian/parents. Each child was clinically examined and 3mL blood samples werecollected. The age range of patients was between one to 15 years with male predominance(male: female = 47:14). CT imaging was performed in all cases. The study subjects wereinitially hypothesized as possible cases of NCC based on two minor plus oneepidemiological criteria as per the revised diagnostic criteria suggested elsewhere⁵.

A commercially procured ELISA kit (UB-Magiwell Cysticercosis Kit^™) was employed for detection of anti-Cysticercus antibodies insera. All the collected sera were tested using the same ELISA kit following themanufacturer's instructions. Antigen coated wells supplied by the kit manufacturer wereincubated with 1:10 diluted patient/control serum (diluted with the serum diluent fluidprovided in the kit). A negative control serum, a low positive control serum and a highpositive control serum (all supplied by the manufacturer) were alsoused for validity of the test. Absorbance was measured at 492nm. A sample was consideredpositive for anti T. solium antibody when the OD value was estimated tobe more than that of the low positive control serum. The sensitivity and specificity ofthe above ELISA for anti-Cysticercus antibody detection in serum was previouslyestimated to be 85% and 94% respectively based on results of the test using sera from agroup of known cases (cases with a definitive diagnosis of NCC) and another groupcomprising healthy normal individuals as described in an earlier study²⁵. Positive control sera were confirmed as positive by radio-imaging features aswell as CDC approved EITB test using a commercially procured kit (ImmuneticsInc, USA).

RESULTS

A positive anti-Cysticercus IgG-ELISA result was observed in 23 of the total61 children indicating an overall positivity of 37.7% (Table 1). The majority of cases with IgG-ELISA positive for Cysticercosispresented with generalized seizure (66.6%), followed by complex partial seizure (31.8%),and simple partial seizure (41.1%) (Table 1).The absorbance (OD₄₅₀) for Low and High positive control sera was measured tobe 0.807 and 1.139 respectively. As per the manufacturer's specification, a case wasconfirmed as positive for anti-Cysticercus IgG antibody when its absorbance was observedto be greater than that of the Low positive control. A case was considered as borderlinewhen its absorbance was observed to be same as that of the Low positive control. Themean OD₄₅₀ values of the ELISA on sera identified to be positive or negativefor anti-Cysticercus-IgG antibodies (n₁=23 and n₂=38respectively) were further analyzed using GraphPad QuickCalcs freestatistical calculators available on web (http://www.graphpad.com/quickcalcs). The two-tailed p value was estimated to be less than0.0001, hence this difference was considered to be extremely statistically significant.Intermediate values used in these calculations: t = 6.0864; df = 16; Standard error ofdifference = 0.080. Confidence interval: 95% confidence interval of this difference:From -0.65452761 to -0.31636439.

The pattern of seizure and other clinical manifestations in the recruitedcases vs. the ELISA-positive cases (n = 61 and 23 respectively) arestated in Figure 1. Among the 23 cases with apositive ELISA result, the relative distribution of different seizure types was found tobe similar to the overall cases. The majority of cases with IgG-ELISA positive forCysticercosis presented with a generalized seizure (52.17%), followed by complex partialseizure (26.08%), and simple partial seizure (21.73%). Headaches were the majorcomplaint (73.91%). Other presentations were vomiting (47.82%), pallor (34.78%), alteredsensorium (26.08%), and muscle weakness (13.04%). There was one hemiparesis casediagnosed to be NCC. Results of the IgG-ELISA in patients with different patterns ofseizure showed predominance of generalized type but no statistical difference wasobserved (Table 1).

Fig. 1

Distribution of clinical signs/symptoms, and seizure pattern in afebrileseizure cases diagnosed to be positive for anti-Cysticercus IgG-ELISA vs. allthe recruited cases.

The CT scan findings i.e., number, distribution of lesions, and respectiveELISA results are summarized in Table 2 and3. The majority (n = 38) of the casespresented with a single space occupying lesion and 15 cases with more than two lesionsbased on CT scan report. In eight cases the CT scan was reported normal; one child foundto be positive by serology.

The distribution of lesions in various parts of the brain showed themajority were in the parietal lobe followed by frontal, occipital, frontoparietal lobes(Table 3).

Of 23 positive cases, only 22 had a CT scan report showing presence oflesion(s) on the brain. The majority of cases had a calcified cyst (nine cases singleand nine cases multiple). Among the cases with single or multiple lesions, the type oflesion and presence or absence of inflammation is shown in Table 4. There was active inflammation around the lesion in 11 cases(one vesicular, nine nodular cases) and four cases with calcified mass. None of thecases with vesicular lesion tested positive for antibody.

Among the positive cases there were 16 male and 7 female children. Theoverall age range of the patients was found to be one to 15 years (median age = 11years). The age range of male cases was one to 14 years (median age = 11 years) and theage range of female cases was from six to 15 years (median age = 12 years).

DISCUSSION

NCC is the main cause of seizures in adults in many developing countriesincluding South and Southeast Asia²³. However, pediatric cases are seldom explored. Our study showed generalizedseizures appeared in the majority of cases compared to partial seizures among thechildren presented to our pediatric epilepsy clinic. Results out of the present studysupported the assumption that NCC can be a potential cause of childhood epilepsyparticularly in tropical countries. In Mexico seizures were reported to be more frequentin children compared to adults, and intracranial hypertension and headaches were morefrequent in adults²⁴. In a study in Mexico, they identified different causes underlying the differentdistribution of seizures and intracranial hypertension in the two patient groups. Butthe situation has not been clarified for the pediatric population universally⁸. In a recent report from another Indian province, 89.66% cases havingneurological manifestations with inflammatory granulomas were found consistent withdiagnosis of NCC¹. However, the serodiagnostic confirmation was not performedin these cases. One strange case in the present series was a one year old patient thatwas diagnosed to be positive by serology. However a recent report of a 2-year-old boywith diagnosis of NCC in Ecuador could support our finding that such a young child canalso acquire this infection. The Ecuador case presented with two parenchymal brainring-enhancing lesions and was confirmed by positive serology as well response toalbendazole therapy where the patient's mother was a Taenia solium carrier⁶. However, in the one year old child the source of infection was not identifiedsince the mother refused to reveal her health history and, in addition that patient wasreferred elsewhere.

It has been accepted that the abrupt onset of clinical signs in human NCCcoincides with intense pathologic reactions to degenerating metacestodes in the brain¹⁵. Following the death of all cysts, provoked seizures disappear, leaving a smallerpercent of patients with residual epilepsy²⁰. In the present study calcified lesions on the brain were observed in themajority of the cases which included either a single lesion or multiple ones, and alsoeither with or without peri-lesional edema. Inflammation surrounding the lesionsindicates the disease to be resulted primarily from the host inflammatory response todying parasites as suggested elsewhere³⁵. Generalized epilepsy was the most common clinical presentation in the presentstudy. Earlier study also showed a predominance of generalized seizures compared toother clinical symptoms and also significantly higher titre in confirmed patients of NCCas compared to patients with other neurological disorders and normal controls²⁶. Though NCC patients present with focal seizures the trend of generalizedseizures to be the major presentation in children having NCC is contradictory to anearlier Indian study³¹. Hence, this study indicates that NCC should be suspected in every child withseizures especially in endemic areas. In the present study, only altered sensorium wasfound in 26.8% children with positive serology testing. Similar incidence was alsoreported in children from the earlier Indian study¹².

In pediatric NCC patients, single colloidal parenchymal cysts were the mostcommon radiologic findings compared with adults in whom multiple viable parasites in thebasal subarachnoidal cisterns or in the ventricles were seen in Mexico²⁴. An increased number of vesicular cysticerci and decreased number of degeneratingcysticerci correlating with aging were noticed in earlier study, whereimmuno-endocrinological factors may be playing a role in susceptibility and pathogenesis⁷. However, in the present study, among the positively diagnosed children, we didnot find any such trend. It may be clearer if we compare the prevalence among bothchildren and adults in our future studies. We found the majority of cases had cyst(s) inthe parietal region of the brain. In an earlier report from India it was hypothesizedthat a child with partial seizures with no obvious causation has a high probability ofharboring one of these three lesions viz., single CT enhancing lesion,single small cerebral calcific CT lesion, and multiple small cerebral calcific CT lesions¹⁷. NCC, single CT enhancing lesion and small single cerebral calcific CT lesiontogether accounted for 40% of etiological factors of seizure as reported in aretrospective study by the same group from India¹⁸. Extraparenchymal NCC is rare in children and carries a poor prognosis²⁹. Also, no extraparenchymal cysticercosis was reported in the present series.However, frequent reporting of cysticercosis in eyes, from the same region of thiscountry indicates the possible prevalence of cysticercosis in children¹⁸.

This is the first report of cysticercal etiology of childhood epilepsy fromthis particular territory. There were a few earlier reports from other provinces in thecountry showing a similar occurrence of cysticercal etiology of childhood seizure¹¹^,³³. NCC has been diagnosed on imaging studies in 34.6% of patients with seizuredisorder of any type, 59.2% of those with a single seizure, 23.7% of those withrecurrent seizure disorder as reported in a recent review²⁷. In the present study, NCC is found to be the etiologic factor in 37.7% among thepatients presenting with seizures and we did not include any adult case in this survey.The majority of our patients were drawn from the rural population with agriculture andanimal farming background. Analysis based on earlier studies indicates that NCC shouldbe suspected in every child with seizures and seroepidemiological study may be concludedfor the ultimate control of the disease²⁶. Thus, NCC must be considered in the differential diagnosis of seizures and awide variety of neurologic disorders, particularly in endemic areas²⁹. We analyzed the social customs and related factors which are peculiar to thisregion and were indirectly responsible for the high incidence of cysticercosis(data not presented). Usually, the highest prevalence rates exist incommunities where there is close contact between man and pigs, hygienic standards arelow, and pork is eaten as raw or undercooked. In Southern India, NCC was found in 2.2%of epilepsy cases. A study from Northern India indicated that cerebral cysticercosis wasfound in 11.1% amongst those attending the neurology clinic and 17.4%-19.2% ofclinically suspected patients. An earlier study from endemic villages report up to 10%or more of the general population having antibodies to T. solium, notnecessarily reflecting the true prevalence of the cysticercosis and leading tomisdiagnosis in a proportion of neurological cases². So, a community approach may help with estimating the true prevalence ofcysticercosis.

Sensitivity and specificity of the presently used commercial ELISA in serumreported to be 85% and 94% respectively in our previous study²⁵. In another Indian study, ELISA kit from the same manufacturer was reported to beof similar sensitivity and specificity³. Previously we demonstrated 80%, and 90% sensitivity of in-house dot-ELISAs whenT. solium metacestode somatic, and ES respectively were used(data unpublished). Specificities of this in house Dot-ELISAs usingT. solium metacestode somatic and ES antigens were observed to be90% in each case. Sensitivity of the employed ELISA kit for anti-Cysticercus antibodydetection as previously estimated was based on results of the test using sera from caseswith a definitive diagnosis of NCC²⁵. However, in the present study the test was employed to screen foranti-Cysticercus antibody in serum from patients presenting with afebrile seizure and inall suspected cases without any definitive diagnosis of NCC. Hence in the presentscreening study, we could diagnose only 37.7% cases to be positive for the said ELISAtest. It is also true that the rest of the cases were negative for the anti-Cysticercusantibody which means the underlying cause of NCC in those cases could not be establishedotherwise possibly there may be other etiology in those cases. However, the importanceof NCC as one of the underlying causes of acute afebrile seizure in pediatric patientscould be highlighted in our study.

Usually, determination of the prevalence of an infection is hampered byvariable sensitivity and specificity of each of the diagnostic tests used, whichidentify a limited or overlapping proportion of its true prevalence. There may bediscrepancy between radio-imaging based on suspicion of NCC vs. a serological testresult. The majority of NCC cases do not always show typical neuro-imaging features andsolitary brain lesion which is a diagnostic challenge¹⁰. Hence it is important for the clinician to choose an appropriate imagingmodality for a proper diagnosis of the etiology in seizure patients. Sensitivity of atest detecting antibodies for the parasite is also related to the number of lesionspresent on the brain as found in our study. There was a statistically significantassociation found between the cases with multiple lesions in brain and theELISA-positivity (p = 0.017). It may be due to a release of moreantigens from multiple larvae that results in mounting a stronger antibody response inthe infected host as suggested in a recent article³⁶.

The presence of antibodies was assessed among all the cases either havingintracranial lesion or no lesion after brain imaging; it was done to employ the use ofserological investigation in the diagnosis of NCC before using imaging. The one childwho had no abnormality on their brain CT but was found to be ELISA positive was ofinterest. This child responded to the albendazole therapy and after follow up, there wasa significant decrease in antibody titre for the above case (data notshown). However, it could only support the fact that the studied populationhad an exposure to the parasite. However, detection of antibodies with a differentrationale may be considered indicative of the active infection. An early and appropriatediagnosis, revealing the biological stage of the parasite would be of value ininitiating an appropriate treatment. Moreover, it would be useful in populationscreening, diagnosis and post-treatment monitoring of NCC especially in the situationswhere imaging techniques are either not available or inconclusive.

LIMITATION

The prevalence of anti-Cysticercus antibodies detected in sera from therandomly selected children with afebrile seizure may not be the true picture if thefigures are extrapolated to a larger population in the territory. The kit detectsantibodies against antigens of the parasite which are not defined. Further study shouldbe conducted to screen antibodies specific to more defined antigens of the larvalparasite.

CONCLUSION

Detecting anti-Cysticercus IgG antibodies in serum could reveal thepotentiality of possible NCC as an underlying cause of afebrile seizures in the studiedpediatric cases. Although, the CT scan is a potent diagnostic tool compared to antibodydetection, the diagnostic decision can be better made when both the modalities areconsidered together. Nevertheless NCC may be suspected as one of the major etiology andhence should be ruled out in every afebrile seizures case, with or without aradioimaging supportive diagnosis, especially in areas endemic fortaeniasis/cysticercosis.

AUTHORS' CONTRIBUTION

Experiments conceived and designed by: PS Sahu, AK Sahu. Experimentsperformed by: PS Sahu, J Seepana. Data Analyzed by: PS Sahu, A Barua. Contributed casesand clinical specimens: S Padela, AK Sahu, S Subbarayudu. Authors of the paper: PS Sahu,J Seepana, AK Sahu.