Solitary cerebral parenchymal cysticercosis: A prospective comparative study with computed tomography and magnetic resonance imaging
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.125272
Source of Support: None, Conflict of Interest: None
Aim: To present a prospective series of 86 patients with solitary cerebral cysticerci who underwent both contrast computed tomography (CT) and contrast magnetic resonance imaging (MRI), and to correlate and compare the imaging findings using these two modalities. Materials and Methods: Lesion characteristics, staging of the cyst and perilesional hypodensity on CT and hyperintensity on MRI were analyzed, and the lesion was described with regard to the appearance of the scolex, cyst fluid, cyst wall and the perilesional area. Results: Patients were largely children and adolescents (mean age 17.4 years, range 6-52) with 52.3% males. MRI was performed, on average, 10.7 days after CT. MRI was more sensitive than plain CT scan in detecting cysticercal lesions (P = 0.003), but there was no statistically significant difference between contrast CT and MRI. None of the patients were detected with an alternative disease on serial MRI. No cyst showed significant mass effect. Non-contrast CT was less sensitive than either contrast CT or MRI in detecting the scolex (P = 0.011), but no difference was seen between the latter two modalities. Qualitative imaging characteristics of the cysticercus on MRI and CT scan are described. All cysts were round in shape, with an average diameter of <10 mm, and maximum diameter of 19 mm. Conclusion: In the present prospective series, contrast CT was nearly as sensitive as MRI in detecting solitary cerebral cysticerci. Thus, in highly resource-limited settings contrast CT may be sufficient for the diagnosis and management of neurocysticercosis.
Keywords: Computed tomography scan, magnetic resonance imaging, neurocysticercosis, solitary cerebral cysticercosis
Neurocysticercosis (NC), caused by the infection of human central nervous system (CNS) by cysts of the pork tapeworm Taenia solium, is the most common and important parasitic disease of the CNS.  Currently it is estimated that 20 million people harbor neurocysticercosis worldwide.  Developing countries shoulder a disproportionately large share of the global burden of epilepsy, and recent advances in diagnostic techniques have brought to light the fact that many patients hitherto thought to be suffering from "idiopathic" epilepsy actually harbor NC, which has been implicated in the causation of seizures. Solitary cysticercal granulomata (SCG) are a common manifestation of NC in India and account for 60-71% of all cases of NC. , Development of imaging techniques, computed tomography (CT) and magnetic resonance imaging (MRI)-enabled ante-mortem diagnosis of neurocysticercosis, resulting in increasing recognition of the high prevalence of the disease worldwide.  Imaging helps to select appropriate treatment and evaluate the response. This study aims to correlate and compare the imaging findings on CT (both non-contrast and contrast) and MRE and also to elucidate diagnostic features and facilitate early and appropriate treatment.
Patients were prospectively enrolled at the tertiary national referral center for neurological diseases in south India. After obtaining written informed consent from patients or relatives, all patients presenting with new-onset focal or generalized seizures were subjected to non-contrast (ncCT) and contrast-enhanced (ceCT) CT brain followed by non-contrast (ncMRI) and contrast (cMRI) MRI. Patients with an SCG on ceCT were included in the study. Diagnosis of cysticercosis was based on the revised diagnostic criteria for NC proposed by del Brutto et al.  Patients were excluded from the study if they had received albendazole or praziquantel in the past, or had evidence of other lesions on CT.
ncCT and ceCT scans were performed using serial axial scans at a slice thickness of 5 mm with the orbito-meatal line as a reference. Non-ionic iodinated contrast (300 mg/mL) was injected intravenously at a dose of 1 mL/kg for children and 40 mL for adults. Gadolinium-enhanced MRI of the brain was performed on a 1.5 Tesla MR system (Magnetom Vision 7252, Siemens AG, Erlangen, Germany) using a circular polarized head coil according to a pre-determined protocol, which included pre- and post-contrast fast spin-echo (SE) T1-weighted sequences (time to pulse repetition [TR] 650, time to echo [TE] 12, number of excitations [n] 1), proton-density imaging (TR 4800, TE 22, n 1), T2-weighted imaging (TR 4800, TE 90, n 1), and FLAIR (TR 9000, TE 119, inversion time [TI] 1200, n 1). Imaging in the axial plane was performed using 5-mm-slice thickness with an interslice gap of 0.5 mm, 24 × 24 cm field of view, and a matrix size of 256 × 256. Post-contrast MT spin-echo imaging was performed with gadolinium diethylene triaminopentaacetic acid (Gd-DTPA, 0.1 mmol/kg IV) as contrast.
Lesion characteristics, staging of the cyst and the presence of perilesional hyperintensity on conventional FLAIR, PD, T2W and T1W images were analysed by a trained observer under the supervision of a neuroradiologist. The lesion description included: Presence of scolex, cyst fluid, cyst wall and perilesional area. Staging of the cysticercal lesion on MRI was done using standard criteria. ,,,
SPSS version 15.0 was used for statistical analysis. Descriptive statistics like mean, standard deviation, standard error, frequencies and percentages were used to express data. The independent sample t test was used to compare continuous variables between two groups. Chi-square test and one-way ANOVA were employed for categorical variables.
During the study period 86 patients with SCG and new-onset seizures had CT and MRI. There were 45 men (52.3%) and 41 women, ages between 6 and 52 years, with a mean age of 17.4 years. Children and adolescents (<18 years) accounted for the majority (56.8%). MRI was performed on an average 10.7 days (±6.5) after CT scan.
ncCT scan failed to reveal any lesion in 30 patients which were subsequently detected only on a ceCT. MRI proved more sensitive than CT scan: ncMRI detected additional lesions in 5 patients and a total of 16 additional lesions were detected on cMRI. In the light of additional information from MRI, we reviewed the CT scans of such patients again, but did not note any findings to suggest presence of SCG. About 85.4% of subjects had a single cyst on cMRI, while 2, 3, 4 and 5 cysts were seen in 9.8%, 1.2%, 2.4%, and 1.2% of subjects, respectively. cMRI was more sensitive than ncCT scan in detecting cysticercal lesions (P = 0.003) [Figure 1]a and b. However, there was no statistically significant difference between ceCT and MRI in the detection of cysticercal lesions. No patient with a diagnosis of SCG on CT scan had an alternative disease diagnosed on MRI.
All the lesions were round in shape and none of the cysts showed significant mass effect or midline shift. In most the cyst diameter was <10 mm and CT scan slightly overestimated the cyst diameter as compared to MRI; however, the difference was not significant between the two imaging modalities [Figure 2]a. The maximum diameter of the cyst was 19 mm. Of the 102 cysticercal lesions identified, 2.5% were in stage 1, 71.6% in stage 2, and 25.9% in stage 3. All lesions were cerebral in location, with the exception of a single intraventricular cyst. The lesion location was: parietal in 51.2%, frontal in 40.2%, occipital in 4.9%, and temporal in 3.7% of patients [Figure 2]b.
A scolex was seen in 52.4% of MRI, 36.5% of ncCT and 51.2% of ceCT. The difference between ncCT and ceCT was significant (P = 0.011; [Figure 2]c), but there was no difference between ceCT and MRI. The scolex was visualized in 33 patients on noncontrast T1 sequence, usually appearing isointense to grey matter, but was hyperintense in 3.1% of images. Similarly, the scolex was visualized as an enhancing nodule (hyperintense to grey matter) in 61.9% of postcontrast T1W images while on T2W and FLAIR sequences, it was hypointense. The ncCT images showed the scolex to be isodense to grey matter, with significant enhancement on ceCT. Nearly all cysts had hypointense contents on pre-and post-contrast T1W images (98.4% precontrast, 93.7% postcontrast), while all appeared hypodense on CT scan. On T2W sequences the cysts had largely hyperintense contents (93.9%). FLAIR demonstrated inversion of cyst contents in 30.8% of lesions. The lesion wall appeared predominantly isointense on precontrast T1W sequences (93.8%), while it appeared hypointense on T2W and PD sequences (83.3%). On ncCT the lesion wall was iso- to mildly hyper-dense in most (95.2%) scans. On postcontrast T1W sequences, most cysts showed ring enhancement with the cyst contents not enhancing with gadolinium contrast (93.7%), 6.3% showed disc enhancement with enhancement of the cyst contents as well as the wall. Around 92.1% of ceCT scans showed ring enhancement, with the remaining showing disc enhancement. The area of cerebral parenchyma around the lesion showed hypointensity in a majority of pre- and postcontrast T1W images (70.8% and 71.4%, respectively). Only 3.2% of postcontrast T1W images showed a hyperdensity in the perilesional area. Perilesional hyperintensity was seen around the cyst on a majority of T2, PD, and FLAIR sequences (98.5%). On CT scan, a similar proportion (76.2%) showed a hypodensity around the lesion, suggestive of perilesional edema. Mass effect or midline shift due to edema was not seen.
Neuroimaging is essential to the diagnosis of neurocysticercosis. Recent advances in the detection of this disease by imaging techniques have resulted in a more accurate appreciation of the prevalence of the infection worldwide. With modern imaging techniques, as many as 70% of all cases of systemic cysticercosis show evidence of nervous system involvement. 
Like in this study, earlier studies have also recorded a similar proportion of young people among their patients with cerebral cysticercosis.  Similarly, earlier studies have also suggested that MRI is more sensitive than ncCT in detecting SCG lesion. ,,, However, there was no significant difference in detecting additional lesions between ceCT and MRI (9.3%). In this study MRI only helped in confirming the diagnosis of NC, and did not identify an alternative disease process in any of the patient even on follow-up imaging. In one study, thin-section ceCT (2.5-mm sections) was found to be as sensitive as MRI.  It is quite possibly that if 2.5-mm sections had been the CT imaging protocol in our study, the sensitivity of ceCT would have been loser to that of cMRI. MRI, with its better image definition, multiplanar imaging capabilities, excellent depiction of tissue contrast, artefact-free visualization of posterior fossa structures and sensitivity to flow effects, is a powerful tool for the evaluation of NC. , MRI is also superior to CT in the demonstration of different stages of cyst activity. , Conventional spin-echo (SE) T2- and T1-weighted images are the most sensitive sequences for detecting intracranial cystic lesions, including cysticerci. The combined use of T2- and T1-weighted imaging is ideal for demonstrating cysts and scolices regardless of the evolutionary stage of NC, probably due to excellent contrast between the scolex and the cyst. In contrast, FLAIR detected only 77.4% of cysticerci in a porcine study. 
CT scan is more sensitive for detecting calcification, the inactive or healed stage of NC.  Punctate calcific foci on CT scan are a very common finding in persons residing in endemic areas, seen in 14-20% of CT scans.  Spin-echo MRI sequences are insensitive to calcification: gradient-echo sequences may be better in this regard.  Intraventricular and subarachnoid cysts, and peri-cystic inflammation and gliosis are better delineated by MRI. , This paradoxical situation precludes the recommendation of a single imaging modality as a universal method for the diagnosis of NC.  In our patients, ceCT scan was nearly as sensitive as MRI, and thus in highly resource-limited settings (i.e. most of the region endemic for NC) ceCT may be recommended as an adequate imaging modality. This finding is in line with previous recommendations, where either ceCT or MRI has been recommended as the initial imaging modality in suspected SCG. , Earlier authors have suggested that ncMRI and cMRI are equally sensitive for the detection of parenchymal cysts  in this study, although more lesions were detected with cMRI, this was not statistically significant. The administration of contrast material to our patients improved the sensitivity of CT scan, as well as of MRI. It has been suggested that without contrast, it is difficult to identify cysticerci <1 cm in size, particularly when they are isointense with brain parenchyma.  In an earlier study from India, contrast was administered to only six of 16 patients with SCG and enabled clear visualization of the lesion in one. 
The differential diagnosis for neurocysticercosis includes abscess, tuberculosis, neoplasm (primary or metastatic), and other parasitic infections.  NC has been reported to be the commonest cause of small single ring-enhancing lesions.  In our cohort none of the cases diagnosed to have SCG on ceCT scan had a diagnosis of tuberculoma or any other lesion on serial MRI. Hence, although MRI is a more sensitive tool in characterizing NC, a ceCT scan may be adequate in arriving at a diagnosis of NC. In the cohort of patients with SCG studied by Rajshekhar et al., none had evidence of raised intracranial pressure, the deficits were stable, the size of the lesion was <20 mm with regular margins and either "disc" (uniform) or "ring" (peripheral) enhancement on contrast. Perilesional edema was mild to moderate in severity.  In our cohort the average size of cysts was <10 mm, with all cysts less than 20-mm diameter and devoid of mass effect or midline shift, similar to the observations in other studies. ,,,
In cerebral parenchymal cysticercosis, the lesions are present cortically or subcortically, predominantly in the frontal and parietal regions. The scolex is seen in approximately half the cysts in both CT and MRI. ncCT scan was significantly less sensitive in identifying the scolex, but no difference was noted between ceCT and MRI. In another study based on the same cohort, we have shown that the scolex disappeared within three months in 92% of the patients. At 24 months, the scolex was not seen in any patient.  This is due to degeneration of the cyst contents including the scolex, which is often invisible in the later stages of cyst degeneration. Although proton-density imaging has been reported to be the most sensitive sequence for the scolex,  it demonstrated the scolex in only 41.9% of our patients. In agreement with Martínez et al., the hyperintense cyst contents on T2 images precluded the visualization of the scolex in many patients, and the T2 sequence demonstrated the scolex as a hypointense mural nodule in only 41.8% of patients. Scolices were best visualized on contrast-enhanced MRI, followed by noncontrast T1 images. In the latter, the scolex appeared largely isointense to grey matter.
The contents of the cyst appeared hypointense on T1W sequences, hyperintense on T2W sequences and hypodense on CT scan, reflecting the fluid nature of the contents.  However, inversion of the FLAIR signal from the cyst contents occurred in only 30.8% of patients, probably due to the fact that cyst degeneration had already started in most patients by the time of presentation leading to a rise in protein concentration within intracystic fluid. A small proportion (6.1%) of patients showed T2-hypointense cyst contents, suggesting that these lesions were in an advanced stage of degeneration at the time of the initial MRI. ,
The region of cerebral parenchyma around the cysticercus (perilesional area) showed T1-hypointense, T2-hyperintense and CT-hypodense oedema resulting from inflammation in a majority of patients. Contrast enhancement was seen in all patients; this was of the ring-enhancement pattern in most. Contrast enhancement indicates inflammation around the cysticercus and the degree of enhancement corresponds to the extent of perilesional edema, which is an indicator of inflammation. , Enhancement of the cyst contents as well as the wall (so-called disc-enhancing pattern, suggestive of later stages of degeneration)  was seen in 6.5% of patients on post-contrast sequences.
The significance of the low-signal capsule on T2W images as well as the slight hyperdensity on CT scans is unknown. The T2-hypointensity may reflect a paramagnetic substance such as free radicals generated during active phagocytosis by the abundant macrophages in the capsule wall. 
In the present study, cysts were largely in the early and late degenerating stages: This is in accordance with current theories of epileptogenesis in NC.  The pathologic stages described by Escobar.  have their imaging correlates: these reflect underlying changes in the disease process and host response. The evolution of cysticercal granulomata through the various stages of degeneration when studied with serial MRI has been detailed by us elsewhere. 
We are grateful to the Indian Council of Medical Research (ICMR) for funding this study.
[Figure 1], [Figure 2]