Advanced magnetic resonance imaging and spectroscopy in a case of neurocysticercosis from North America

Abstract

Neurocysticercosis (NCC) is a parasitic infection caused by Cysticercus cellulosae, the metacestode of pork tapeworm (Taenia solium). NCC is one of the most common public health problems worldwide. We present a patient harboring a bilobed ring-enhancing lesion with a presumed diagnosis of brain metastasis, who returned to the USA after traveling to an endemic region. The diagnosis of NCC was established based on a characteristic resonance of succinate on proton magnetic resonance spectroscopy. Also, higher mean diffusivity and lower fractional anisotropy along with relative cerebral blood volume were observed from the lesion compared to contralateral normal brain regions. Multiparametric analysis may improve the differential diagnosis of ring-enhancing intracranial lesions such as NCC.

Introduction

Neurocysticercosis (NCC) is the most common parasitic infection of the central nervous system (CNS). About 50 million people get infected worldwide, and about 50,000 die of NCC every year. It is endemic in Asia, South America, and Sub-Saharan Africa.^1,2 Due to high immigration and travel rates from endemic zones, the prevalence of NCC has been increasing in developed countries, including the USA where it has become a diagnostic challenge.^3–5 NCC results when the Cysticercus cellulosae, a metacestode of pork tapeworm (Taenia solium) encysts in the brain of humans and swine. ⁶ The parasite most frequently lands in the brain parenchyma but can also be found inside the ventricles and subarachnoid spaces. Once implanted, the parasite endures in its dormant condition for many years but eventually undergoes degeneration and induces granuloma formation, focal scarring, and calcification. The clinical manifestations of NCC are often nonspecific and highly variable, depending on the number, topology, evolutionary stage of the cyst, and the host’s immune response to the parasite. ⁷ Cerebral parenchymal NCC presents with increased intracranial pressure, hydrocephalus, and focal neurological deficits and seizures in 50–80% of cases. ⁸

Magnetic resonance (MR) imaging along with serological tests and the patient's clinical symptoms help establish the definite diagnosis of NCC.^9,10 Four stages of development and regression of NCC have been described on imaging and histopathology, and these are grouped as vesicular or cystic, colloidal, nodular-granular, and nodular-calcified stage. Different evolutionary stages of NCC are usually recognized by characteristic appearances on conventional neuroimaging.^11–15 However, a degenerating colloidal stage in the absence of scolex (a pathognomonic feature of NCC), with surrounding edema and contrast enhancement, may mimic brain abscesses, toxoplasmosis, tuberculomas, gliomas, or even brain metastases. Because of the confounding appearance on conventional neuroimaging, it is not always possible to distinguish NCC accurately from other intracranial mimics.^16,17 Lately, the differential diagnosis of NCC has greatly improved by combined use of proton MR spectroscopy and diffusion and perfusion MR imaging techniques.^{11,13,18–24}

Here, we report a patient with a bilobed ring-enhancing lesion with surrounding edema clinically suspected to be a brain metastasis. The diagnosis of NCC was not initially considered in the list of differential diagnosis. However, combined use of metabolic and physiologic MR imaging assisted us in making an accurate diagnosis of NCC and prevented invasive surgical intervention in this patient. This case argues that NCC should be considered as a differential diagnosis, even in those regions of the world that are considered non-endemic for NCC, and also emphasizes the importance of metabolic and physiologic MR imaging in making an accurate diagnosis.

Case presentation

A 60-year-old female was brought to the emergency department of our tertiary care teaching hospital after experiencing a 10-minute episode of generalized tonic-clonic seizure. When she arrived at the hospital, her symptoms had resolved, and she was amnestic to the event. The patient had no prior history of neurological disorders, trauma, hypertension, diabetes, unintentional weight loss, or decreased appetite. Five weeks prior to the onset of her seizure, the patient had traveled to a NCC endemic region. On physical examination, the patient was alert, conscious, and cooperative. Vital signs were normal. Neurological, fundoscopic, and ophthalmologic examinations were also normal, with no signs of meningeal irritation or papilledema. The laboratory and biochemical data were within the reference range.

Subsequently, the patient underwent 3T MR imaging of the brain. The neuroimaging protocol included anatomical imaging with and without gadolinium-based contrast agent. Additional sequences included: single voxel (10 mm × 10 mm × 10 mm) proton MR spectroscopy using a spin echo PRESS (point-resolved spectroscopy) sequence (repletion time (TR)/echo time (TE) = 2000/30 ms) with water suppression by means of chemical shift selective saturation pulses, diffusion tensor imaging (DTI) using 30 noncollinear/noncoplanar directions with a single-shot spin-echo, echo-planar read-out sequence (TR/TE = 5000/86 ms, b = 0, 1000 s/mm²) with parallel imaging, and dynamic susceptibility contrast perfusion-weighted imaging (DSC-PWI) using T2*-weighted gradient-echo EPI sequence (TR/TE = 2000/45 ms, number of measurements = 45). The detailed parameters of these sequences have been described previously.^25,26 On post-contrast T1-weighted images, a bilobed ring-enhancing lesion without any evidence of a scolex was observed in the anterior aspect of right temporal lobe. As visible on the corresponding T2-FLAIR image, there was a local mass effect with surrounding vasogenic edema extending into the subcortical white matter (Figure 1).

Figure 1.

Axial post-contrast T1-weighted image demonstrating a bilobed ring-enhancing lesion in the right temporal lobe (white arrow). On corresponding axial T2 FLAIR image, the lesion is surrounding by vasogenic edema extending into the subcortical white matter regions. Compared to contralateral normal brain parenchyma, higher mean diffusivity along with lower fractional anisotropy are noted from the lesion. On the corresponding cerebral blood volume map, the lesion demonstrates low blood volume.

As shown in Figure 2, single voxel MR spectroscopy demonstrated decreased N-acetyl aspartate (NAA; 2.02 ppm) without significant elevation of choline (Cho; 3.2 ppm)-to-creatinine (Cr; 3.02 ppm) ratio. Additionally, we observed a prominent singlet at 2.4 ppm consistent with succinate (Suc) along with a broad resonance centered at 1.3 ppm consistent with lipids (Lip) and lactate (Lac). The position of the Suc peak at 2.4 ppm was assigned by using the chemical shift position of Cr resonance centered at 3.03 ppm as an internal reference.

Figure 2.

Axial post-contrast T1-weighted image demonstrates the position of ¹H MRS voxel centered on a bilobed ring-enhancing lesion. The corresponding spectrum exhibiting various resonances. Note the presence of the succinate (Suc) peak centered at 2.4 ppm.

Motion and eddy current corrected DTI-derived maps (mean diffusivity (MD) and fractional anisotropy (FA)) and DSC-PWI-derived cerebral blood volume (CBV) maps and T2-FLAIR images were co-registered to post-contrast T1-weighted images using our previously reported method.^26,27 A semi-automatic method was used to segment the contrast-enhancing region (CER) of the lesion. Representative anatomical images, MD, FA, and CBV maps are shown in Figure 1. The median values of DTI and DSC-PWI indexes from CER and contralateral normal brain parenchyma are presented in Table 1. In comparison to contralateral normal brain parenchyma regions, higher MD and lower FA and relative CBV (rCBV) values were observed from the lesion. The combined analysis of anatomical, metabolic, and physiologic MR imaging characteristics from the lesion in our case suggested the diagnosis of NCC in the colloidal stage of its development. The final diagnosis of NCC was based upon the qualitative enzyme-linked immunosorbent assay test of the patient’s serum that was found to be positive for immunoglobulin G antibodies to cysticercosis and complete resolution of the lesion on follow-up imaging.

Table 1.

Summary of DTI and DSC-PWI derived parameters.

Parameters	MD (mm2/s)	FA	rCBV
Contrast-enhancing region of lesion	1.11 × 10−3	0.16	0.91
Contralateral normal brain	0.70 × 10−3	0.21	1.35

All values are reported as medians.

DTI: diffusion tensor imaging; DSC-PWI: dynamic susceptibility contrast perfusion-weighted imaging; MD: mean diffusivity; FA: fractional anisotropy; rCBV: relative cerebral blood volume.

The patient was administered dexamethasone (4 mg orally every 12 hours) for seven days, levetiracetam (500 mg orally every 12 hours), and albendazole (an anti-parasitic drug, 400 mg orally every 12 hours) for two weeks. At a three-month follow-up visit, the patient had no recurrence of seizures. Conventional MR imaging showed that the bilobed ring-enhancing lesion had reduced in size. At a seven-month follow-up visit, the patient continued to be seizure free, and the lesion continued to resolve on conventional neuroimaging. At the 16-month follow-up, there were no signs of the lesion on anatomical images.

Discussion

We demonstrated the combined potential of ¹H MRS, DTI, and DSC-PWI in accurately diagnosing NCC in a patient who had returned to the USA after traveling to a NCC endemic region. A characteristic resonance of Suc on ¹H MRS was observed. Also, higher MD, lower FA, and lower rCBV values were observed from the NCC lesion compared to contralateral normal brain parenchyma. Collectively, these spectroscopic and imaging findings were useful in establishing the diagnosis of NCC in our case, and hence helpful in avoiding an unnecessary craniotomy. The clinical manifestations of NCC are often nonspecific, and many cases remain seronegative either due to the presence of a dead parasite or not so strong elicitation of antibody response. ²⁸ Moreover, not all cases of NCC present with a typical neuroimaging finding of a cyst with eccentrically placed scolex.^19,29 In view of nonspecific clinical symptoms and equivocal findings on conventional neuroimaging, it is often a challenging task to distinguish NCC accurately from other intracranial mass lesions.^16,17 This becomes even more difficult in those parts of the world where the incidence of parasitic CNS infections, including NCC, are not very common, as happened in our case. A combination of metabolic and physiologic imaging assisted in establishing the diagnosis of NCC in the present case.

Some excellent review articles^11,15 and book chapters ³⁰ have extensively described the potential role of different MR imaging pulse sequences in providing structural characterization and details of different stages of NCC. Neuroimaging findings in each stage reflect underlying pathophysiological processes of the disease and host immune response. In brief, the vesicular stage is viable and composed of a well-defined fluid-filled cystic cavity that appears hyperintense on T2-weighted images and hypointense on T1-weighted images relative to normal brain parenchyma, whilst the scolex is seen as an eccentrically placed nodule, hypointense on T2-weighted images and hyperintense on T1-weighted images. There is usually no perifocal edema and abnormal contrast enhancement on post-contrast T1-weighted images. The colloidal stage is characterized by degenerating parasite and inflammatory process. The cystic fluid turns turbid, and the cyst wall becomes thickened. On T2-weighted images, the cystic fluid and surrounding edema appear hyperintense, whereas the cyst wall and scolex appear isointense or hypointense with ring-shaped enhancement. In the nodular-granular stage, the parasite retracts, and its fluid content is absorbed. Fibrosis develops with time, progressively occupying the entire lesion. The lesion appears isointense to hypointense with or without a central hyperintense signal on T2-weighted images and appears as isointense on T1-weighted images. On post-contrast T1-weighted images, the lesion appears as a homogeneously or ring-shaped enhancing nodule with or without surrounding edema. Finally, the nodule undergoes mineralization and subsequently calcification. This may result from partial dystrophic calcification of the parasite or from the presence of a cysticercal calcareous capsule. The calcified lesion appears as isointense to hypointense on both T2- and T1-weighted images. Some calcified lesions may demonstrate persistent contrast enhancement for at least one year.

Using ex vivo one-dimensional ¹H MRS and two-dimensional correlation spectroscopy (2D-COSY) experiments on cysticercus fluids aspirated from cysts located in swine brains, some previous studies have documented a characteristic metabolite pattern for NCC.^19,31 The investigators reported the presence of several metabolites in the cyst fluid such as amino acids (leucine, isoleucine, valine), Suc, Lac, Lip, Cr, and Cho, thus providing a benchmark for assignment of the metabolites that are likely to be observed in a focal NCC lesion in vivo. Using ¹H MRS, some prior studies from NCC patients have documented the presence of all those metabolites that were seen on previously reported ex vivo ¹H MRS experiments.^30,32,33 In the present case, we observed the resonances of NAA, Cr, Cho, Lip+Lac, and Suc. NAA is a neuronal marker and is not expected to be present in a NCC lesion. ³⁴ We believe that NAA was a contamination from the adjoining brain parenchyma due to partial volume effect from the selected voxel. The resonances of Cr, Cho, and Lip+Lac are nonspecific and can be seen both from NCC lesions as well as from normal brain parenchyma regions. However, a prominent singlet at 2.4 ppm consistent with Suc (which is considered as a putative marker of parasitic cysts) assisted in turning the differential diagnosis of the lesion in favor of NCC in the present case. Using biochemical assays, some studies have documented the presence of Suc in cysticercus cysts. ³⁵ It has been reported that phosphoenolpyruvate formed during glycolysis partially transforms to oxaloacetate and undergoes a cascade of reactions to form Suc. ³⁶ Interestingly, the peak of Suc can also be seen in pyogenic brain abscesses that are teemed with obligate aerobic bacterial populations. ³⁷ However, it is also well documented that the Suc peak, if present, is always accompanied by a peak corresponding to acetate in those brain abscesses. ³⁷ Furthermore, a study has reported that the presence of Suc alone or higher Suc than acetate levels from an intracranial lesion indicates the diagnosis of NCC. ³² Thus, the absence of an acetate peak on ¹H MRS ruled out the prospect of a probable diagnosis of pyogenic brain abscess in our case.

Multiple studies have also reported the promising role of diffusion and perfusion MR imaging in the differential diagnosis of intracranial mass lesions.^{11,18,32,38–42} Compared to brain abscesses and tuberculomas, higher MD or apparent diffusion coefficient (ADC) values have been reported in NCC in previous studies and hence may facilitate the distinction of NCC from other intracranial mass lesions.^20–22,43 It is believed that higher ADC/MD in the NCC lesions than abscesses and tuberculomas might be related to lower cellularity, protein contents, and viscosity associated with cysticercus cysts. Consistent with the findings from earlier studies, higher MD values were observed from the lesion compared to contralateral normal brain parenchyma in our case. Furthermore, lower FA values were also observed from the NCC lesion compared to contralateral normal brain regions. Using DTI, Gupta et al. ¹⁸ showed wide variations in MD and anisotropic indexes from NCC lesions, depending upon the evolutionary stage of the cysts, with nodular-calcified and nodular-granular stages showing decreasing trends in MD and increasing trends in anisotropic indexes than in vesicular and colloidal stages. While analyzing diffusion MR imaging data from different stages of NCC lesions, Santos et al. ⁴¹ also reported a wide range of ADC/MD values (0.54–1.41 × 10⁻³ mm²/s) from the lesions in the colloidal stage of their development, suggesting the presence of a variable amount of proteinaceous contents within these lesions. In close agreement with this study, we also observed an MD value of 1.11 × 10⁻³ mm²/s in our case. Although, brain abscesses usually exhibit low FA values, some investigators have also reported high FA from brain abscesses. In a study, higher FA values were also observed in brain abscesses in comparison to NCC. ²²

In the present case, a lower rCBV value was observed from the contrast-enhancing lesion compared to the contralateral normal brain region. This finding is in agreement with a previous study that reported lower rCBV from NCC lesions compared to tuberculomas. ²³ In general, elevated rCBV values are observed from gliomas and tuberculomas secondary to neovascularization (increased microvascular density).^44–46 Taken together, these findings suggest that rCBV may be a useful parameter to differentiate NCC from tuberculomas and gliomas. While some atypical forms of fungal abscesses are known to exhibit high rCBV, ²⁷ most brain abscesses are characterized by low rCBV.^27,47,48 A similar pattern of low rCBV in brain abscesses and NCC lesions suggests that a differential diagnosis may not always be possible based on perfusion imaging alone. Recently,^49,50 newer neuroimaging techniques such as constructive interference in steady state, susceptibility-weighted imaging, and arterial spin labeling have been proposed to improve the diagnosis of NCC. We believe that combining the unique strengths of different imaging techniques in a multiparametric approach may allow for a more precise diagnosis of NCC.

Until recently, NCC was considered as a rare entity in developed countries. However, together with the growing number of immigrants from endemic areas, there has been an increase in the number of patients with cysticercosis in some of these countries.^3–5 Also, increased tourism and international trading affairs have rendered people from nonendemic areas more vulnerable to acquire this parasitic disease. The most frequent pattern of NCC expression in immigrants or travelers is the presence of a single cysticercus lesion, suggesting that the common form of disease acquisition is through sporadic contact with Taenia carriers who are food handlers. Otherwise, travelers would have more often presented with severe infections, which are typically observed in Taenia carriers who ingest huge quantities of Taenia solium eggs from eating raw pork.^51,52

In conclusion, the rising prevalence of brain infections, including NCC, in the developed world is mainly driven by enhanced migration rate from disease-endemic areas and the ease of international travel. Under these circumstances, comprehensive evaluation of the brain lesions with multimodality imaging as done in our case is valuable in providing a “virtual biopsy” guiding the clinical decision making. The multiparametric analysis may be helpful in making an accurate diagnosis of NCC, thus preventing unnecessary surgical intervention.

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

ORCID iD

Sanjeev Chawla https://orcid.org/0000-0001-6978-7284