Epilepsy Outcomes and Hippocampal Volumes in Children with Multiple Parenchymal Neurocysticercoses.

Background: We aimed to study the radiological and epilepsy outcomes of children with multiple neurocysticercosis (NCC) and the relationship with hippocampus volume. Study Design: Observational study. Materials and
Methods: Children between 5 and 18 years of age with NCC lesions and completed more than 2-year follow-up were enrolled. Magnetic resonance imaging (MRI) brain with a predefined protocol was done as per clinical indication, and bilateral hippocampi volume was measured with a region of interest-based manual method. Proportion of children who had radiological resolution of lesions, proportion of children who were seizure-free for at least 3 months after antiepileptic drugs withdrawal, and mean hippocampi volume were calculated.
Results: During the study period, 229 children followed up in pediatric NCC clinic were enrolled, the mean age was 10.5 ± 3 years, and 129 (56%) were boys. A total of 159 (65%) children had single-lesion NCC, and 79 (35%) had multiple lesion (> 2) NCC. Resolution of lesions was seen in 8% children with multiple NCC, significantly less as compared to the single NCC group (8% vs. 27%, P = 0.006). Multiple-lesion NCC group had a higher rate of calcification as compared to children with a single lesion (66% vs. 47%, P = 0.008). Twenty-nine (37%) children with multiple-lesion NCC were seizure free for at least 3 months after antiepileptic drug withdrawal. Mean combined hippocampal volume in children with single-lesion NCC was 2.96 (0.62) mL and multiple-lesion NCC was 2.88(0.43) mL.
Conclusion: Children with multiple-lesion NCC have lower resolution, and higher calcification during follow-up in comparison to single-lesion NCC. Frequency of children who remained seizure-free 3-month post AED withdrawal was similar in single- and multiple-lesion groups. Copyright:

INTRODUCTION

Neurocysticercosis (NCC), the most common parasitic infestation of the central nervous system (CNS), is endemic in developing countries like India, and is a common cause of acquired epilepsy worldwide.[12] NCC remained a neglected disease and was added by World Health Organization (WHO) to the list of major neglected tropical diseases in 2010.[3] An estimated 26% to 54% cases with active epilepsy in the developing world are due to NCC.[4]

The calcified NCC, a result of the host’s inflammatory response to the larval form of Taenia solium, is the most common form of the disease, and a continued risk factor for subsequent seizures. On the contrary, the resolution of NCC lesion is associated with good outcome and resolution of epilepsy in the majority.[5] Recent observations have suggested calcified NCC as a risk factor for enduring epileptogenesis.[67] Perilesional gliosis, reactivation with the presence of perilesional edema surrounding the calcified lesion, and hippocampal sclerosis contribute to enduring epileptogenic state.[78] In addition, a correlation between hippocampal sclerosis and NCC lesions, particularly, calcified lesions, has been proposed.[91011] Whether it is a mere coexistence, or a pathogenic relationship exists is still controversial.

The clinical and radiological outcomes of single-lesion NCC have been reported in various studies; however, similar studies in multiple-lesion NCC are lacking.[512] We hypothesized that children with multiple-lesion NCC have higher seizure recurrences, less resolution, and less seizure freedom post-antiepileptic drug (AED) withdrawal in comparison with single-lesion parenchymal NCC.[512,13] Hence, we studied the proportion of children with multiple-lesion NCC having seizure freedom post-AED withdrawal at 2 years follow-up. Besides, we also studied the hippocampal structural changes in children with calcified NCC by comparing the mean hippocampal volumes of children with calcified and non-calcified NCC, who were completing 2 years since the first symptom.

MATERIALS AND METHODS

This hospital-based study was conducted from January 2016 to May 2017 in an outdoor clinic of a tertiary care center. The inclusion criteria were: children of either sex, aged between 5 and 18 years, with parenchymal NCC lesions and completed at least 2 years follow-up after the first symptom. Children having additional structural brain lesions that are adjudged to be independent risk factors for epilepsy, pre-existing hippocampal atrophy, or malformation detected at the time of diagnosis were excluded from the study. Consecutive children fulfilling the inclusion criteria were enrolled in the study after informed consent of the parents. Data were prospectively collected by interview with parent /caretaker/ and the child and through clinic records. A thorough history, which included details of symptoms, first seizure/ diagnosis of NCC and drug-free period/ seizure freedom, and other details were recorded.

Children who required a magnetic resonance imaging (MRI) (indication-drug withdrawal or continuing seizures) were subjected to MRI brain to look for the resolution of NCC lesions. All the MRI’s were done using a 3T scanner and included 3D T1 (T1 MPRAGE/ SPGR BRAVO), inversion recovery, Cor T2, axial T2, FLAIR, and SWI images. An expert neuroradiologist reviewed the MRI’s, and the number, location, presence of perilesional edema, stage of the lesions were recorded. The hippocampal volume measurement was done by “region of interest” (ROI)-based manual method. The right, left, and combined hippocampal volumes were recorded. The epilepsy outcome, the relationship between NCC number and stage were studied. We also studied the possible association between NCC number (and impact of calcification) and hippocampal volumes. For this, the mean hippocampal volume of children with multiple NCC was compared with age and sex-matched controls with single NCC.

The primary outcome of the study was to assess the frequency of children with multiple NCC lesions, who had radiological resolution of lesions at 2 years after the first symptom. The secondary outcome variables were the frequency of children with multiple NCC lesions who were seizure free for at least 3 months after AED withdrawal compared with children with single NCC lesion. The mean hippocampal volume in children with multiple NCC and those with single NCC, completing 2 years after the first symptom was compared. In addition, hippocampal volumes were compared in children with calcified NCC and those without calcified NCC completing 2 years after the first symptom.

Statistical methods

The data was collected and analyzed using the Statistical Package for the Social Sciences (SPSS) (version 23.0 IBM). For the descriptive analysis, frequencies of categorical variables were computed along with their 95%CI. Means and medians were calculated for numerical variables. Distribution of data was assessed using the Levene’s test for equality, and the “t test.” P Value was taken accordingly. Epilepsy outcomes in NCC were assessed, and it was studied whether there is any relationship between NCC number at onset and outcome. Secondly, we studied the possible association between NCC number (and impact of calcification) and hippocampal volumes. For this, the mean hippocampal volume of children with multiple NCC was compared with age and sex-matched controls with single NCC using the “t test”.

RESULTS

Records of 445 children registered in the NCC clinic between 2012–2015 were reviewed, among these 196 children, who had incomplete follow-up and 20 children with coexisting neurological disorder were excluded. During the study period from January 2017 to May 2018, 229 children who had >2 years of follow-up were enrolled in the study [Figure 1]. The median age of the study group was 11 years (IQR 8–13, range 4–17) and 129(56%) were boys. The median duration of follow-up of the study group was 30 months (IQR 24–36), ranging from 12 to 132 months.

Figure 1

Flowchart of the study participants

The median age at the presentation to the clinic was 7 years (IQR 5–10). All except six children (97.4%) had at least one episode of seizure before presentation. These six children presented with isolated headache. One or more episode of status epilepticus was present in 21(9.2%) children, and 17(7.4%) children had status epilepticus as the first presentation. The median number of seizure episodes were 2 (IQR 1–5) with the range varying from 1–50. Seizures were controlled with a single AED in 211 (94.6%), and with 2 AED’s in 12 children (5.4%). The median duration for which the children had received AEDs was 27 months (IQR 24 – 35) with a range varying between 2 and 120 months. One course of an antihelminthic agent was received by 223 (97%) and steroids by 212(92%) children.

Radiology at diagnosis

The diagnosis of NCC was established by CT scan in 139 (60.6%), by MRI in 59 (25.7%) children and by both modalities in rest of the 31 (13.5%) children. Of the enrolled children, 150 (65.5%) had single lesion and, 79 (34.5%) children had multiple (≥2) lesions (2 lesions in 45, 3 in 12, more than 3 in rest). The most common type of lesion at presentation was ring-lesion (colloidal stage), present in 193(84.3%) children; followed by live lesions (vesicular stage) seen in 65 (28.4%); and 58 (25.3%) had calcified lesions at diagnosis . Of the 312 lesions detected, 120 (52%) were in parietal lobe, 110 (48%) in frontal, 30 (13%) in temporal, 41(18%) in occipital lobe, 6(2.6%) in basal ganglia and 5 (2.2%) in cerebellum.

Epilepsy outcome

At 2 years, among 229 children, 91(40%) children reported seizure recurrence. One hundred (43.7%) children were off AED’s, and 93(40.6%) amongst them were off AEDs for at least 3 months. The median duration off AED’s was 8.5 (IQ 5–14) months in this subgroup. Thirty-five children (15.2%) had failure of AED withdrawal. Among multiple-lesion NCC, 29(37%) were off AED for at least 3 months, and among single-lesion NCC, 64 (43%) were off AED for at least 3 months. Nineteen (24%) children with multiple-lesion NCC had a failure of AED withdrawal as compared to 16(10.6%) children with single-lesion NCC [Table 1]. There was no statistically significant difference in these groups (p-value =0.38). Radiological resolution of the lesion was associated with successful AED withdrawal [Table 2]. A higher number of seizures at presentation, higher seizures recurrences, lack of radiological resolution and the presence of calcified lesions strongly predicted failure of AED withdrawal [Table 2].

Table 1

Comparison of epilepsy and radiological outcome in children with single-lesion and multiple-lesion NCC

Variables	Single lesion (n = 150)	Multiple lesion (n = 79)	P Value
Duration of follow-up in months median (IQR)	30(24–35)	31(24–45)	NS
Epilepsy outcome
Off AED for at least 3 months (n = 93)	64(42.7;35–51)§	29 (36.7;26–48)	0.38
Off AED (any duration)	68(45.3;37–54)§	26 (32.9;23–44)	0.07
Total seizures (mean)	3.8 ± 5.3	7.3 ± 10	0.006
Seizures in last 2 years (mean)	1 ± 2	2 ± 5	0.13
Duration on AED in months (mean)	31 ± 16.6	34.5 ± 18.	0.16
Duration off AED in months (mean)	10.2 ± 8.2	11.5 ± 9.1	0.49
Failure of AED withdrawal	16(10.7;6–17)§	19(24.1;15–35)	0.13
Radiological outcome
Number of patients with follow-up imaging	103(68.7;61–76)§	50(63.3;52–74)
Complete resolution of all lesions	28/103 (27.3;19–37)§	4 /50 (8;2–19)	0.006*
Calcified lesions	70 /150 (46.7;38–55)§	52 (65.8;54–76)	0.008*
Calcification at diagnosis	29/150 (19.3;13–27)§	29 (36.7;26–48)	0.007*
Calcification of lesion/s during follow-up	44(29.3;22–37)§	30(38;27–50)	0.19
Appearance of new NCC lesions on follow-up	23 (22.3;15–32)§	11(22;11.5–36))	1.0
Hippocampus volume (mL), mean (SD)
Right	2.9 ± 0.63	2.93 ± 0.43	0.75
Left	2.94 ± 0.63	2.83 ± 0.45	0.52
Combined	2.96 ± 0.62	2.88 ± 0.43	0.62

IQR = interquartile range, AED = antiepileptic drugs, NCC = neurocysticercosis, SD = standard deviation

*Fischer exact test

§(%; 95%CI)

Table 2

Difference between those who were off-AEDs for at least 3 months and those on AEDs

Variable	Off AEDs (n = 93)	On AEDs (n = 136)	P Value
Male	62(66.67%)	67(49.26%)	0.009
Number of lesion, n (%)
Single	64 (68.8%)	86 (63.2%)	0.383
Multiple	29 (31.2%)	50 (36.8%)
Number of lesions (mean ± SD)	1.3 ± 0.6	1.5 ± 1	0.159
Complete resolution of lesion	23 (26.4%)	09 (6.6%)	<0.001
Calcified lesions	42 (45.16%)	80 (58.82%)	0.042
Calcification on follow-up	21(22.58%)	49(36.03%)	0.082
Perilesional edema at diagnosis	83 (89.25%)	117 (86.03%)	0.472
Appearance of new lesion	11(11.8%)	18(13.23)	0.753
Total seizure (mean ± SD)	2.6 ± 3	6.5 ± 9	<0.001
Seizures in last 2 years (mean ± SD)	0.27 ± 0.6	2.03 ± 4.3	<0.001
Duration on AEDs (mean in months ± SD)	26 ± 15.5	36 ± 17	<0.001
Failure of AED withdrawal	07(7.5%)	28(20.6%)	0.008

AED = antiepileptic drug

Radiology at follow-up

Of the 229 children enrolled, follow-up imaging was available in 153 (67%) children, including the 53 children who underwent MRI for hippocampal volume assessment. Of these 153 children, 74 (48%) had CT, and 79 (52%) had MRI as their follow-up imaging. Radiological resolution of lesion/s was seen in 32 (21%) children, 122(53%) had calcified lesions, 70 (46%) had calcification of lesions during follow-up [Figure 2]. At the follow-up imaging, 34 (22%) children had appearance of new lesions on the MRI/CT scans.

Figure 2

Axial T2-weighted image (A) showing the presence of two nodular T2 hypointense lesions in the right frontal lobe with no apparent perilesional edema. (B) Axial T2-weighted section at a high frontal level showing a calcified nodular lesion in the left frontal lobe with perilesional edema around the lesion. Coronal T2-weighted image (C) depicting the ROIs drawn covering both the hippocampi indicating symmetrical signal and bulk

Follow-up imaging was available in 103 (86%) children with single-lesion NCC and 50 (63%) children with multiple-lesion NCC. Out of the 103 children with single-lesion NCC, 28 children (27%) had radiological resolution. Whereas, among 50 children with multiple-lesion NCC who had follow-up imaging, only four (8%) had radiological resolution (P = 0.006) [Table 1].

Calcified lesions were present in 29 cases each in single and multiple-lesion groups at diagnosis (P = 0.007) [Table 1]. At 2-year follow-up images, a total of 70 children (47%) with single-lesion NCC and 52 children (66%) with multiple-lesion NCC had calcification (p-value = 0.008). The appearance of new lesions on follow-up imaging was seen in 23 children (22%) with single-lesion NCC and 11 children (22%) with multiple-lesion NCC.

Lesion load and hippocampal volume

Fifty-three children underwent MRI brain during the study period to look for resolution of lesions, in these children sequences required to calculate hippocampal volume were acquired simultaneously, and volume was calculated by region of interest-based manual method [Figure 2]. Volume assessment was done in 41 children (In 12 children hippocampal volume could not be calculated due to technical issues leading to loss of recorded data). Of the analyzed 41 MRIs, 19 children had single lesion, and 22 children had multiple-lesion NCC on their initial imaging.

Among the 19 children with single-lesion NCC, the mean hippocampal volume of right side was 2.98 + 0.63 mL, left side was 2.94 + 0.63 mL and the combined average volume was 2.96 + 0.62 mL. In the 22 children with multiple-lesion NCC, mean hippocampal volume of the right side was 2.93 + 0.43 mL, the left side was 2.83 + 0.45 mL and the combined average volume was 2.88 + 0.43 mL. No statistically significant difference in mean hippocampal volume was observed between the two groups (P > 0.05). We found one child with multiple NCC with left-sided medial temporal lobe sclerosis, one with atrophic cerebral hemisphere and hippocampus, whereas one had signal changes in hippocampus without atrophy.

Mean hippocampal volume on the right side was 2.94 + 0.52 mL, left side was 2.83 + 0.52 mL, and the combined volume was 2.9 + 0.51 mL in 32 children with calcified NCC. Out of nine children with ring lesions (non-calcified lesions) mean hippocampal volume of the right side was 2.9 + 0.7 mL, left side was 3.04 + 0.62 mL, and combined volume was 3.01 + 0.58 mL. However, no statistically significant difference in mean hippocampal volume observed between the two groups.

DISCUSSION

A core finding of this study was that the radiological resolution of multiple-lesion NCC was significantly lower compared to those who had single-lesion NCC. Secondly, a significantly higher percentage of children had calcification of lesion in the follow-up imaging in the multiple-lesion NCC group. Although the number of children who were off AEDs for 3 months was similar in both the multiple-lesion NCC and single-lesion NCC group, the trend was suggesting a higher seizure burden in children with multiple-lesion NCC.

The resolution rate, although higher in the single NCC group is much lower than what is reported in the literature. In the study by Sharma et al.[13] as many as 71.3% of individuals with solitary NCC had a resolution at 6 months. The difference in resolution rates could be due to different imaging modalities and settings in which children were seen. Our findings are in keeping with a previous report from our center in which the resolution rate was 29% in 176 children who underwent follow-up CT head.[512] With increasing resolution of MRI imaging and the ability to detect even the calcified lesions, the proportion of complete resolution of NCC lesions is likely to be lower in future studies.

Patients with multiple lesions were noted to have a higher rate of calcified lesions as compared to patients with a single lesion, both at diagnosis and in follow-up imaging (65.8% vs. 46.6%). Multiple-lesion group and single-lesion group had similar rates of new lesion appearance. The appearance of a new lesion on follow-up imaging was also significantly high at 22% as compared to 2–12% in published literature.[514] There is a need for prevention of reinfection among the children with NCC and a need for reemphasizing better hygiene practices.[15]

In our study, 40% of the children were off AED for 3 months, and there was no difference in the single-lesion and multiple-lesion NCC group. Factors associated with favorable epilepsy outcome were-higher rate of lesion resolution; the lesser rate of calcification; male sex and lower number of total seizures in the preceding 2 years. It is evident and intuitive that children who have a resolution of lesions are more likely to have a better outcome.

In a study conducted by Rajshekhar et al.,[16] in India, in 185 patients with solitary NCC, 84.9% of patients were seizure free at a mean follow-up of 66 months. Seizure recurrence was seen in 28 patients (15.1%) at varying duration after AED withdrawal. This difference in seizure recurrence percentage can be due to a longer duration of follow-up in their study and high risk of seizure recurrence in the immediate period after AED withdrawal. The burden of drug-refractory epilepsy was minuscule in our cohort with nearly all children being controlled on single AED, indicating that whereas NCC is a significant risk factor for epilepsy, seizure control is usually easy.[17]

The main burden of NCC is epilepsy and is responsible for nearly 50% of the cases with focal seizures in India.[16] There were trends to show a higher epilepsy burden in children with multiple NCC. They had higher mean total seizure burden compared with those with solitary NCC. However, the proportion of patients who were off AED and seizure free for 3 months were similar in the two groups. Seizure recurrence after AED withdrawal was higher in multiple-lesion group as compared to single-lesion group (21.6% vs. 12.7%). Overall, the seizure recurrence after AED withdrawal was 15.7%, which is similar to the data published in the literature.[518,19]

Adults with medial temporal lobe epilepsy undergoing surgical management for hippocampal sclerosis were reported to have a high incidence of calcified NCC.[20] A similar observation has been reported from Brazil and Indian subcontinent.[2122] In a population-based study from Athapula, adult population >60 years of age screened for NCC. Those who had calcified NCC underwent MRI for measurement of hippocampal volume and controls were selected from the same community.[23] Nineteen among 28 individuals with calcified NCC and 4 controls were reported to have bilateral or unilateral HS and had ipsilateral calcified NCC.[23]

Bianchin et al.[24] studied a cohort of 290 surgically treated patients with MTLE-HS and found that 38.6% of them had calcified NCC lesions. In patients with single calcified NCC lesions, the lesion matched the side of hippocampal sclerosis in 74.1% of patients a finding suggesting an anatomical relationship between NCC and MTLE-HS. Rathore et al.,[21] a epilepsy surgery center reported calcified NCC lesions were associated with drug-refractory epilepsy in 40%, hippocampal sclerosis in 42%. Presence of temporal lobe lesions was associated with hippocampal sclerosis and anterior temporal lobectomy with surgical removal of the lesion was associated with better seizures outcome. In contrast, Gama et al.[25] studied 63 patients with mesial temporal lobe epilepsy and found no association of hippocampal sclerosis and NCC. These all studies suggested the role of NCC in the pathogenesis of hippocampal sclerosis.[26] Inflammation-mediated or persistent epileptogenesis are the two possible mechanisms reported with HS in NCC.[911,26]

In our study, the region of interest manual method was used to measure hippocampal volumes. Salmah et al.[27] reported the mean hippocampal volume of children with epilepsy and normal controls. In our study, the hippocampal volume was similar to the controls, as reported by Salmah et al.[27] There was no statistically significant difference in the mean hippocampal volume of children with single and multiple lesion. No statistically significant difference was noted in the mean hippocampal volume of children with calcified and non-calcified lesions. We did find MTLS in one child with multiple NCC in our study. Our preliminary findings do not support an association between MTLS and NCC. However, the numbers in our study were small, and the follow-up duration was short. A longer follow-up of a larger cohort of NCC patients is needed to study this association.

The main limitation of the study was that outcomes were assessed at short follow-up and 3 months duration was possibly insufficient for success or failure of AED withdrawal. As we know seizure recurrences in children with NCC are unpredictable and may be associated with prolonged quiescence in between, a longer follow-up would help answer the question.[28]

In conclusion, children with multiple-lesion NCC have a poor radiological resolution in comparison with children with a single lesion, with higher rate of lesion calcification. There were trends of adverse epilepsy outcomes in the group with multiple lesions. Lastly, hippocampal volumes and the number of NCC or the presence of calcified lesions in children had no observed association.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.