Imaging correlates of serum enzyme-linked immunoelectrotransfer blot (EITB) positivity in patients with parenchymal neurocysticercosis: results from 521 patients

Abstract

Background

The presence of perilesional edema among patients with parenchymal neurocysticercosis (pNCC) of various lesion subtypes has not been correlated with results of serum enzyme-linked immunotransfer blot (EITB) for cysticercal antibodies.

Methods

In total, 521 patients with pNCC were classified into solitary cysticercus granuloma (SCG), multiple lesions, at least one of which was an enhancing granuloma (GMNCC), solitary calcified cysticercal lesion (SCC) and multiple calcified cysticercal lesions (CMNCC). The proportion of EITB positivity among each lesion subtype and its association with perilesional edema were determined.

Results

There were significantly higher positive EITB results in patients with GMNCC (90/111, 81.1%) compared with other lesion types. Perilesional edema was associated with positive EITB in patients with CMNCC. On univariate analysis, perilesional edema and GMNCC were associated with EITB positivity. On multivariate analysis, only GMNCC (OR 7.5; 95% CI 3.5 to 16.2) was significantly associated with EITB positivity.

Conclusions

In patients with pNCC, the presence of perilesional edema is associated with a higher probability of a positive EITB result in patients with CMNCC, suggesting a synchronicity in the mechanisms associated with formation of perilesional edema and the antibody response in this subtype. In patients with enhancing granulomas, edema is not an independent predictor of a positive EITB, suggesting that the enhancement itself is associated with a strong antibody response.

Introduction

Neurocysticercosis (NCC) is a common parasitic infection of the central nervous system, caused by the larval form of the cestode Taenia solium. The larva lodges most commonly in the brain parenchyma but can also be seen in the extraparenchymal space (in the subarachoid space, ventricles or scalp). Solitary cysticercus granuloma (SCG) is the commonest form of NCC in Indian patients.^1–6 Seizures are the most frequent manifestation of degenerating parenchymal NCC (pNCC) or its calcified residue.^1,2 They occur intermittently and might be related to the release of antigens from a degenerating cyst leading to focal edema and parenchymal irritation.

Serological testing is an important criterion for diagnosing NCC in patients with brain imaging findings suggestive of cysticercal lesions.^3,7 Presently, lentil lectin glycoprotein enzyme-linked immunoelectrotransfer blot (LLGP-EITB) assay is considered to be the best serological test with which to diagnose NCC.^8–13 In LLGP-EITB, antibodies are tested against seven cyst glycoprotein antigens—GP50, GP42-39, GP24, GP21, GP18, GP14 and GP13—the latter four being grouped into the 8kDA family.^7,8,9,11 In patients with extra-parenchymal NCC and multilesional pNCC, the LLGP-EITB has higher sensitivity, vis-à-vis those with other subtypes of NCC.^8,9,11 EITB fares poorly in patients with solitary calcifications and in patients with SCG.^12–15 While the impact of multiple enhancing parenchymal lesions and live cysts on the sensitivity of the EITB has been established, the influence of other factors such as the presence of perilesional edema on neuroimaging have not been investigated.^11,16,17

Perilesional edema is common and is associated with episodic seizure activity in patients with calcified pNCC and SCG.^17–22 The genesis of perilesional edema has been attributed to the innate immune response and to the intermittent release of antigens from the degenerating larva or calcified lesions.^18,19,22 As edema is secondary to inflammation, we hypothesized that its presence might be associated with a positive EITB result.

The aim of this study was to measure the association between perilesional edema, lesion subtype and seropositivity to the EITB in a large group of patients with definitive pNCC in whom the interval between the imaging and the test was ≤30 days.

Materials and Methods

The study protocol was approved by the Institutional Review Board of Christian Medical College, Vellore.

Patients

Patients who underwent EITB and were diagnosed with pNCC by CT or MRI or both, from 2001 to 2018, were eligible for inclusion in the study. As perilesional edema wanes rapidly and can resolve completely within weeks, a time interval of ≤30 days between brain imaging and the EITB test was chosen as an inclusion criterion to correlate the imaging findings with the EITB result. Among 634 pNCC patients for whom data were available, 521 fulfilled this inclusion criterion.

Imaging correlates

Investigators who interpreted imaging findings were blinded to the EITB results and those who interpreted EITB tests were blinded to the imaging findings. The number of pNCC lesions identified by CT or MRI and the presence of perilesional edema was recorded.

Perilesional edema was diagnosed based on the hypodensity around the lesion seen on CT or hyperintensity around the lesion on T2W/T2W FLAIR sequences of MRI.

Lesions seen at imaging were categorized into degenerating cysts or granuloma (enhancing), live cysts (non-enhancing) and calcifications. Based on these categories of lesions, four lesion subtypes of NCC patients were created: solitary cysticercus granuloma (SCG); solitary cysticercal calcified lesion (SCC); more than one enhancing lesion or a combination of enhancing lesion/s and live cyst/s with or without calcified lesions (GMNCC); >1 calcified lesion (CMNCC) (Figure 1). Diagnosis of SCG was made by previously validated criteria while that for multilesional NCC on imaging was made according to the criteria proposed by Del Brutto et al. and Carpio et al.^3,6,7

Figure 1.

(A) Axial contrast enhanced TIW MR images of a 42-y-old man with seizures showing multiple ring-enhancing lesions with scolices within (white arrow) suggestive of GMNCC lesion subtype. (B) Axial T2W MR image of the same patient showing three calcified lesions (black arrows) in the right temporal, left frontal and left temporal lobes. His EITB was positive. (C) Axial contrast enhanced T1W MR image of a 15- y-old male with a ring-enhancing SCG and edema in the left frontal lobe. His EITB was positive. (D) Axial contrast CT image of an 8-y-old girl with seizures and CMNCC showing two calcifications without any enhancement, with hypodensity around the right frontal lobe calcification indicative of edema. Her EITB was positive. (E) Axial contrast CT image of a 28-y-old man with SCC showing a solitary calcification in the left frontal lobe without edema. His EITB was negative.

EITB assay

Patients’ sera were analyzed with EITB to detect antibodies against T. solium metacestode glycoprotein antigens (GP50, GP38, GP24, GP21, GP18, GP14 and GP13). The antigens were prepared from naturally infected pigs obtained locally, as described previously.¹² Although 29 kDa antigen was reported in a previous publication from Christian Medical College and Hospital, Vellore, it was not detected in any of the patients in the current study.¹² The EITB was considered to be positive if any one band among the seven described by Tsang et al.⁹ was seen. The number of positive bands and their molecular weights were noted for each patient.

Statistical analysis

Data were analyzed using SPSS (version 21.0; IBM, Bangalore), R (version 4.0.3) and Stata (version 16.1; StataCorp LP, College Station, TX, USA). To study the association between the EITB result and brain imaging, the data were further grouped into those in whom the imaging test interval was ≤5 d vs those with an imaging test interval of 6–30 d. The crude association between perilesional edema and EITB positivity was measured using prevalence proportion ratios with their 95% CIs. The median number of bands in each lesion subtype was calculated and compared using Kruskal–Wallis test and Student's t test. χ² test or Fisher exact test was used to test associations between the various categorical variables and the EITB test result. Student's t test was used to study associations between the continuous variables and the EITB test result. Variables that were found to be significant on univariate analysis were subjected to logistic regression analysis. A p value of <0.05 was considered significant.

Results

Study population

There were 521 eligible pNCC patients including 330 (63.3%) males and the median age was 22 y (IQR 12–32 y). A total of 353 (67.8%) patients had perilesional edema. Most patients had SCG (n=304, 58.3%), followed by GMNCC (n=111, 21.3%), SCC (n=66, 12.7%) and CMNCC (n=40, 7.7%).

NCC subtype and EITB results

No EITB band (negative EITB) was detected in 209 (40.1%) and one or more EITB bands (positive EITB) were detected in 312 (59.9%). The pNCC lesion subtype was significantly associated with the EITB result (p<0.001). More participants with SCC were negative on EITB (65.2%) than those with CMNCC (42.5%), SCG (41%) and GMNCC (18.9%) (p<0.001). Ninety of the 111 patients in GMNCC subtype had a positive EITB test, which was significantly higher than that in the other groups (p<0.001).

The median number of positive bands among 521 patients was 1 (IQR 0–2), but varied across groups (Table 1). Lesion subtype was significantly associated with band count (p<0.001), with GMNCC (3; IQR 1–4) having a higher number of bands than other subtypes. The number of positive bands was also significantly higher in patients with multiple lesions compared with solitary lesions (p<0.0001).

Table 1.

Lesion type and EITB results

Type of NCC	Number (%)	EITB positive (%)	Median number of bands (IQR)*
All NCC	521 (100)	312 (59.9)	1 (0–2.5)
SCG	304 (58.3)	176 (57.9)	1 (0–2)
SCC	66 (12.7)	23 (34.9)	0 (0–1)
GMNCC	111 (21.3)	90 (81.1)*	3 (1–4)*
CMNCC	40 (7.7)	23 (57.5)	1 (0–2)

CMNCC, multiple calcified cysticercal lesions; GMNCC, multiple parenchymal cysticercal lesions, at least one of which is enhancing; NCC, neurocysticercosis; SCC, solitary calcified cysticercal lesion; SCG, solitary cysticercus granuloma.

*GMNCC vs other subtypes p<0.001 (for EITB positivity and median number of bands).

On multivariate analysis, GMNCC lesion subtype was an independent predictor of a positive EITB result. The presence of enhancing lesions on brain imaging, as well as multiple calcifications, was also associated with a significantly higher likelihood of positive EITB compared with SCC, as shown in Table 2.

Table 2.

Univariate and multivariate analysis of factors predictive of positive EITB

Variable	EITB positive (n=312) (%)	EITB negative (n=209) (%)	p value on univariate analysis	OR (95% CI)	p value on multivariate analysis
Mean age (y)	22.8±13.8	23.7±13.3	0.75	0.99 (0.98 to 1.01)
Male:female	198:114	132:77	0.71	1.01 (0.7 to 1.46)
Type of lesion
GMNCC (n=111)	90 (81.1)	21 (18.9)	<0.001	7.52 (3.5 to 16.16)	<0.001
SCG (n=304)	176 (57.9)	128 (42.1)		2.62 (1.15 to 5.99)	0.02
CMNCC (n=40)	23 (57.5)	17 (42.5)		2.24 (1.19 to 4.23)	0.01
SCC (n=66)	23 (34.8)	43 (65.2)		Reference
Edema present
Yes (n=353)	227 (64.3)	126 (35.7)	0.003	0.81 (0.51 to 1.29)	0.38
No (n=168)	85 (50.6)	83 (49.4)
Interval from scan to EITB
≤5 d	223 (59.8)	150 (40.2)	0.94	1.04 (0.69 to 1.58)	0.88
>5 d	89 (60.1)	59 (39.9)

Bold, Significant p value <0.05 in both univariate and Multivariate analysis.

CMNCC, multiple calcified cysticercal lesions; GMNCC, multiple parenchymal cysticercal lesions, at least one of which is enhancing; SCC, solitary calcified cysticercal lesion; SCG, solitary cysticercus granuloma.

Perilesional edema and EITB results

The percentage of patients with perilesional edema in the GMNCC, SCG, CMNCC and SCC lesion subtypes was 79.3%, 80.3%, 35.0% and 12.1%, respectively. Patients with enhancing lesions (GMNCC or SCG) had a significantly higher proportion of lesions with perilesional edema compared with those with no enhancing lesions (CMNCC or SCC) (p<0.0001). Among the patients with only calcifications, those with CMNCC had a significantly higher association of perilesional edema compared with SCC (p=0.004). There was no association between gender or the median number of bands and perilesional edema.

Overall, 227 (64.3%) of 353 patients with perilesional edema had a positive EITB result while 85 (50.6%) of 168 patients without perilesional edema had a positive EITB result (p=0.003). Perilesional edema was associated with a positive EITB result in patients with CMNCC and in those without enhancing lesions (Table 3). The prevalence proportion ratios for perilesional edema and EITB positivity with each lesion subtype are shown in Table 4.

Table 3.

EITB result and perilesional edema according to lesion subtype

Lesion subtype	EITB positive	EITB negative	p value
SCG	176	128
Edema present (n=243) (%)	141 (58)	102 (41.9)	0.92
Edema absent (n=61) (%)	35 (57.4)	26 (42.6)
SCC	23	43
Edema present (n=8) (%)	3 (37.5)	5 (62.5)	0.86
Edema absent (n=58) (%)	20 (34.5)	38 (65.5)
GMNCC (n=111)	90	21
Edema present (n=88) (%)	72 (81.8)	16 (18.2)	0.69
Edema absent (n=23) (%)	18 (78.2)	5 (21.7)
CMNCC (n=40)
Edema present (n=14) (%)	11 (78.6)	3 (21.4)	0.047
Edema absent (n=26) (%)	12 (46.1)	14 (53.8)
Enhancing lesions (SCG and GMNCC)
Edema present (n=331) (%)	213 (64.4)	118 (35.6)	0.83
Edema absent (n=84) (%)	53 (63.1)	31 (36.9)
Non-enhancing lesions (SCC and CMNCC)
Edema present (n=22) (%)	14 (63.6)	8 (36.4)	0.03
Edema absent (n=84) (%)	32 (38.1)	52 (61.9)

Bold font indicates p value <0.05.

CMNCC, multiple calcified cysticercal lesions; GMNCC, multiple parenchymal cysticercal lesions, at least one of which is enhancing; SCC, solitary calcified cysticercal lesion; SCG, solitary cysticercus granuloma.

Table 4.

Prevalence proportion ratio (PPR) for perilesional edema and EITB positivity according to lesion subtype

Lesion subtype (number)	Number of patients with perilesional edema (%)	PPR (95% CI)	p value
SCG (n=304)	243 (80.3)	1.22 (0.48 to 3.08)	0.67
SCC (n=66)	8 (12.1)	0.98 (0.77 to 1.2)	0.84
GMNCC (n=111)	88 (79.3)	1.04 (0.82 to 1.3)	0.79
CMNCC (n=40)	14 (35)	1.7 (1.04 to 2.8)	0.04

CMNCC, multiple calcified cysticercal lesions; GMNCC, multiple parenchymal cysticercal lesions, at least one of which is enhancing; SCC, solitary calcified cysticercal lesion; SCG, solitary cysticercus granuloma.

Among patients with enhancing lesions, perilesional edema was not associated with a positive EITB result. On multivariate analysis, presence of perilesional edema was not an independent predictor of a positive EITB result (Table 2).

Discussion

EITB for NCC

EITB has been established as a sensitive and specific assay for NCC over the past 3 decades.³,^8–10,^12–14 The EITB bands diagnostic of NCC are all ≤50 kDa.^9,12 However, a drawback of the EITB test is that it is an indicator of exposure but not of active infection, unlike the antigen ELISA.^8,10 EITB is less sensitive for detecting SCG than Multiple cysticercal lesions or extraparenchymal NCC, as has been reported earlier in Indian patients.^12,13 In the current study, a positive EITB test result was found in 60% of NCC patients compared with EITB being positive in 90% of Peruvian patients with NCC.¹¹ This discrepancy is likely due to a larger proportion of patients with solitary and calcified lesions in our patient population in comparison with the higher proportion of patients with live cysts and extraparenchymal lesions in the study conducted by Arroyo et al.¹¹ In addition, multivariate analysis reported here confirmed that patients with the GMNCC subtype have a higher rate of positive EITB. Because the EITB band patterns and positivity might differ in different populations, it is important to establish seropositivity rates on EITB and band pattern analysis in different geographical areas of the world where NCC is endemic.

EITB results and lesion subtypes

It has been well documented that patients with only calcific NCC lesions (SCC or CMNCC) and those with SCG have a low positivity on EITB.^12,14,23 In the current study, 40% of NCC patients had a negative EITB test result, even although it was performed within 30 d of brain imaging. In some of these patients, the negative results could be a reflection of the low level of antibody response to a solitary lesion (SCG/SCC) or low antigenic challenge from calcified lesions (CMNCC). The other possibility is that the parenchymal lesions are confined to an ‘immune-privileged’ zone with restricted access to the humoral immune response due to an intact blood–brain barrier. Additionally, T. solium oncospheres, postmaturation, have been shown to evade immune detection by binding with host plasma proteins or by expressing surface proteins similar to those of the host.²⁴ Other studies have shown that the cyst can express mitochondrial enzymes and acetyl cholinesterase that are similar to those in the host.^25,26 Collectively, these features can result in the cyst being recognized as ‘self’ by the immune system and therefore is not attacked.²⁵ Furthermore, slow degeneration of parenchymal cysts results in gradual depletion of the antigenic reservoir that might prevent a host response, particularly when the number of lesions is low or when there are calcifications.²⁵

Perilesional edema in NCC and EITB results

The current study is the first to report the correlation of perilesional edema and a positive EITB result among all the subtypes of NCC. Univariate analysis indicates a correlation between positive EITB and the presence of perilesional edema, which is particularly significant when the EITB is performed ≤5 d after brain imaging. However, perilesional edema was not an independent predictor of positive EITB on multivariate analysis in the entire cohort. The presence of perilesional edema was significantly associated with a positive EITB test only in the CMNCC subtype.

Perilesional edema is present in a very high proportion of patients with enhancing pNCC lesions presenting with seizures, provided the imaging is performed within a few days or weeks of the occurrence of seizures. We studied the incidence of edema in patients with SCG and noted that 187/195 (95.9%) had edema in imaging performed within 30 d of seizures at initial presentation. There was complete resolution of edema at a median interval of 3 mo in 72 (61.5%) of the 117 patients with persistent SCG (Rajshekhar et al., unpublished data). We also noted that perilesional edema wanes rapidly, often before the resolution of the enhancement of the lesion. It was also postulated that episodic seizure activity may be provoked by intermittent release of antigens from the pNCC lesion and that this is associated with an increase in enhancement and worsening of the edema, both being markers of inflammation.²⁷ It has been reported that perilesional edema resolves over 6 wk.¹⁸

Edema is also noted around calcified lesions in up to 30–50% of patients presenting with seizures.^17–20,^22,28 Seizures and edema have both been recognized as episodic phenomena associated with a breach in the blood–brain barrier and edema often heralds symptoms such as seizures or headache.¹⁸ Activation of focal inflammation by intermittent release of antigens from a calcified granuloma may cause perilesional edema and seizures.^17,19 Fujita et al.²¹ performed positron emission tomography scans on nine patients with degenerating cysts or perilesional edema or both and identified expression of a mitochondrial membrane translocator protein that is upregulated in astrocytes and microglia in inflammatory conditions of the brain. Their findings suggested a low level of continuous inflammation that showed intermittent exacerbations, which probably provoke seizures and are also associated with perilesional edema.

Edema was not an independent predictor of a positive EITB result in patients with enhancing lesions (GMNCC and SCG). These enhancing lesions are more likely to have a positive EITB result compared with calcified lesions. This suggests that in patients with enhancing lesions, the enhancement, which is the imaging correlate of inflammation, is in itself associated with a positive EITB, irrespective of the presence or absence of edema. In patients with multiple calcified lesions (CMNCC), on the other hand, where inflammation is less frequent and the imaging correlate of the same is perilesional edema (and mild enhancement of the calcified lesion on MRI), it is understandable that lesions with edema are more likely to be associated with a positive EITB.

Herrick et al.¹⁷ demonstrated that perilesional edema around calcified lesions and the number of bands on Western blot were independent predictors of seizure recurrence in a retrospective case control analysis of 38 patients (11 patients with seizure recurrence and 27 controls without seizure recurrence) who had at least 1–2 viable cysts in addition to other NCC subtypes on baseline MRI. Subjects with edema who had ≥4 bands on baseline Western blot had a higher volume of edema at follow-up compared with those who had 1–3 bands. In contrast to these findings, there was no correlation between the median number of bands and the presence of perilesional edema in our patient population. This could be attributable to the lower number of median bands in our study (one) compared with four bands in their study population. The low prevalence of viable cysts among our patients could explain the lower median number of bands.

Limitations

Although we had data available on 521 patients with regard to perilesional edema and EITB, data on symptoms could not be retrieved for most cases; 211 patients had documented seizures within 7 mo of their EITB. Our study was not designed to correlate the change in perilesional edema at follow-up and a synchronous EITB test.

Conclusions

Patients with multiple and enhancing granulomas have a higher EITB positivity than those with solitary and calcified lesions. In those with enhancing granulomas, edema is not an independent predictor of a positive EITB, suggesting that the enhancement itself is associated with a strong antibody response. In those with multiple calcified lesions, perilesional edema is associated with a higher probability of a positive EITB. Our results suggest a synchronicity in the mechanisms associated with enhancement and perilesional edema and the antibody response in patients with pNCC.

Authors’ contributions

VR and RKM designed the study protocol and carried out the assessment of radiological images; AT and JM carried out data collection; PV, BEP and ST carried out the EITB assays; GR and EJ carried out the statistical analysis; PV and RKM drafted the manuscript; DD, HC and VR critically revised the manuscript. All authors read and approved the final manuscript. VR and RKM are the guarantors of the paper.

Funding

This work was supported by the National Institute of Neurological Diseases and Stroke (https://www.ninds.nih.gov/) in the USA [R21NS077466] with HC as the Principal Investigator (PI) and by the Department of Biotechnology (www.dbtindia.nic.in) in India [BT/MB/BRCP/06/2011] with VR as the PI. The patients included in this study during 2016–2018 were part of a study funded by the National Institute of Neurological Diseases and Stroke (https://www.ninds.nih.gov/) [1R01NS098891, dated 15 July 2016]. HC is funded by the Canada Research Chair in Epidemiology and One Health [CRC 950–231857].

Competing interests

None.

Ethical approval

The procedures followed in this study are in accordance with the ethical standards of the Helsinki Declaration (1964, amended most recently in 2008) of the World Medical Association. The design of the work has been approved by the Institutional Review Board of Christian Medical College, Vellore.

Data availability

Data available on request.