Large (≥3-Millimeter) Parenchymal Brain Calcified Cysticerci Are More Often Associated with Hippocampal Atrophy than Smaller Ones in Seizure-Free Individuals with a Single Lesion

ABSTRACT.

Calcified cysticerci are often associated with hippocampal atrophy (HA). While most studies suggest that repetitive seizures cause HA in these patients, others have demonstrated that HA may also occur in persons without epilepsy. Little is known about mechanisms triggering HA in seizure-free individuals with calcified cysticerci. Here, we aimed to assess whether the size of the calcification is associated with HA. Using a population-based design, we selected apparently seizure-free individuals with a single calcified cysticercus in whom interictal paroxysmal activity and other causes of HA have been discarded. A total of 55 individuals (mean age, 58.3 ± 13 years, 62% women) fulfilled inclusion criteria. Unadjusted and multivariate models were fitted to assess the association between the size of the calcification dichotomized into <3 mm and ≥3 mm (exposure) and the presence of HA (outcome). Sixteen participants (29%) had HA, which was asymmetric in eight (50%) cases. Hippocampal atrophy was noted in 11/20 (55%) participants with large calcifications and in 5/35 (14%) with small calcifications (P = 0.001). A multivariate logistic regression model showed a significant association between the presence of large calcifications and HA, after adjustment for relevant confounders (odds ratio: 7.78; 95% CI: 1.72–35.1). Participants with calcifications ≥3 mm in diameter were 7.8 times more likely to have HA than those with smaller ones. Study results open avenues of research for the use of agents to prevent HA progression.

INTRODUCTION

Calcified lesions are the end stage of most parenchymal brain cysticerci and may occur either spontaneously or as the result of treatment with cysticidal drugs.¹ These lesions, traditionally considered inert and not causing clinical manifestations or remote structural damage of the nervous system, have been associated with recurrent seizures, chronic headaches, or even transient focal neurological deficits.^2–4 In addition, there is increasing evidence of a link between calcified neurocysticercosis (NCC) and hippocampal atrophy (HA).^5–7 Despite the lack of well-conducted prospective studies, biological plausibility strongly favors NCC as the cause (exposure) and HA as a late effect (outcome).⁵ However, pathogenic mechanisms leading to HA in patients with calcified NCC are still elusive. Surgical case series of patients with NCC and intractable epilepsy suggest that HA occurs as the result of repetitive seizures inducing subsequent hippocampal damage.^8,9 In contrast, population-based studies have shown that seizures are not necessary for HA to occur in the setting of calcified NCC.¹⁰ In these cases, it has been suggested that some genetic predisposition or, more likely, the occurrence of inflammatory events related to periodic exposure of the immune system to parasitic antigens trapped within the calcification may account for the development of HA.^5,7,11

In this view, the larger the calcification, the greater the quantity of antigens that would be released in each of the remodeling phenomena experienced by the calcification and the higher the risk of late HA. However, this has not been demonstrated so far. In this study, we attempted to assess whether large calcifications are more often associated with HA than smaller ones in seizure-free individuals with a single calcified cysticercus in the brain parenchyma who also had no evidence of subclinical paroxysmal epileptiform activity on a scalp electroencephalogram (EEG) and no other demonstrable cause of HA.

MATERIALS AND METHODS

Epidemiological background.

Atahualpa is a rural Ecuadorian village where human cysticercosis is highly endemic (affecting approximately 10% of the adult population),² but active transmission has been arrested without any specific intervention.¹² Therefore, the only neuroimaging finding in Atahualpa residents with NCC is the presence of parenchymal brain calcifications.² This characteristic provides optimal grounds for the conduction of population studies on this form of the disease.

Study population.

Characteristics of Atahualpa residents have previously been described.¹³ Most of the population is homogeneous regarding ethnicity, levels of education, living conditions, socioeconomic status, and dietary habits. About 20% of the houses have open latrines for feces disposal, and evidence of open-air defecation still exists. Domestic pig farming is common, with more than 600 pigs in the village, as demonstrated in the last census conducted by our group.¹⁴ Corralling of pigs is noted in about 60% of cases, and the other pigs are allowed to roam free. Despite the pig-raising culture of Atahualpa residents, they are not frequent pork consumers, because most pigs are sold in neighboring villages or reserved for special occasions.^12,14

Study design.

By use of a population-based cross-sectional design, Atahualpa residents aged ≥40 years were identified by means of annual door-to-door surveys, and those who signed a comprehensive informed consent were enrolled. Demographics, level of education, and clinical interviews for the diagnosis of epilepsy were collected at baseline, ensuring a complete clinical database in all cases. Likewise, conditions that may be associated with HA (traumatic brain injury, dementia, severe sleep apnea, stroke, hypoxic encephalopathy, and severe depression, among others) were investigated.¹⁵ For this, we used interviews and procedures previously described for participants of the Atahualpa Project Cohort, such as neurological examinations, the Montreal Cognitive Assessment, a single-night diagnostic polysomnography, the depression axis of the Depression Anxiety Stress Scale-21, and a brain magnetic resonance image (MRI) (see below).^16–18

Participants were invited for the practice of an unenhanced computed tomography (CT) of the head, and those who had parenchymal brain calcifications, which are highly suggestive of NCC according to the last revised version of internationally accepted diagnostic criteria,¹⁹ were eligible for this study. Then, individuals received a brain MRI and a scalp EEG, and those who had these three mandatory exams were conditionally included. Individuals with a diagnosis of epilepsy (see below) and those who had more than one parenchymal brain calcification or a scalp EEG showing paroxysmal epileptiform activity were subsequently excluded from analysis. The study was approved by the Ethics Committee of Hospital-Clínica Kennedy, Guayaquil, Ecuador (FWA 00030727). Research was conducted in accordance with the ethical principles of the Declaration of Helsinki. Aggregated data are available from the corresponding author upon reasonable request.

Recognition and diagnosis of persons with epilepsy.

Interviews were conducted by personnel of the Atahualpa Project, including rural doctors who were trained to identify individuals with a seizure disorder using a field instrument described and validated by Placencia et al.^20,21 During a second phase, certified neurologists (blinded to results of the survey) interviewed all individuals who were screened as suspected cases. All participants with a diagnosis of epilepsy (or a single seizure) were excluded from analysis.

Computed tomography.

All CTs were performed with a Philips Brilliance 64 CT scanner (Philips Medical Systems, Eindhoven, The Netherlands). The slice thickness was 3 mm, with no gap between slices. All women of childbearing age underwent a pregnancy test before the exam, and those who were pregnant had the CT delayed until delivery. A neurologist and a neuroradiologist independently read all CTs with attention to the presence of parenchymal brain calcifications. Rounded and homogeneous nonphysiological supratentorial calcifications, not associated with other neuroimaging findings suggestive of alternative etiologies and not explained by any other causes, were considered to be of cysticercotic origin.¹⁹ Images corresponding to all identified lesions were sent to a cysticercosis expert for further confirmation. Interreader agreement was excellent (k = 0.92), and disagreements were resolved by consensus. For the purposes of this study, calcifications were separated into two groups, those that measured <3 mm in diameter (small), which were visualized in only one CT slice (Figure 1), and those with a diameter of ≥3 mm, which were often visualized in more than one slice (Figure 2).

Figure 1.

Unenhanced computed tomography scans of the head in four different study participants showing small (<3-mm) calcifications in the brain parenchyma (arrows).

Figure 2.

Unenhanced computed tomography scans (CTs) of the head in two different study participants (upper and lower rows) showing large (≥3-mm) parenchymal brain calcifications that are seen in more than one CT slice.

Electroencephalogram recordings and interpretation.

One-hour scalp EEGs were performed using a Nihon Kohden EEG-1200 digital machine (Nihon Kohden Corporation, Tokyo, Japan) and collected using the international 10-20 electrode configuration, with the addition of T1 and T2 electrodes, as previously described by our group.²² Studies included eye opening, eye closure, hyperventilation, photic stimulation, wakefulness, and sleep (when possible). All exams were reviewed by an epilepsy board-certified neurologist blinded to clinical data and neuroimaging findings. Participants with evidence of focal slowing or epileptiform discharges, including temporal intermittent rhythmic delta activity, were excluded.

Magnetic resonance imaging.

All MRIs were performed with the use of a Philips Intera 1.5T machine (Philips Medical Systems, Eindhoven, The Netherlands), in accordance with a previously defined protocol that included two-dimensional multislice turbo spin echo T1-weighted, fluid attenuated inversion recovery, T2-weighted, and gradient-echo sequences in the axial plane, as well as a T1-weighted inversion recovery sequence (acquisition time, 3:31 minutes; repetition time, 2.250 ms; echo time, 11 ms; and inversion time, 400 ms) oriented in the coronal plane. The latter was perpendicular to the long axis of the temporal bone with a slice thickness of 4 mm and no gap between slices.¹⁰ This sequence was used for hippocampal evaluation by the use of the Scheltens’ medial temporal atrophy scale.²³ This visual scale grades the width of the choroid fissure, the width of the temporal horn, and the height of the hippocampus on a five-point rating scale ranging from 0 to 4. Scores of up to 1 point for persons <75 years of age and 2 points for persons ≥75 years were considered age-related changes. Each temporal lobe was rated separately, and any asymmetry of ≥1 point was noted. Two certified neuroradiologists independently read all MRIs blinded to other information. Interreader agreement was very good (k = 0.81), and disagreements were resolved by consensus.

STATISTICAL ANALYSES

Data analyses were carried out using STATA version 18 (College Station, TX). In unadjusted analyses, continuous variables were compared by linear models and categorical variables were compared by the χ² or Fisher exact test as appropriate. Hippocampal atrophy was used as the dependent variable (outcome), and the size of calcified cysticerci (exposure) was used as the independent variable dichotomized into small (<3 mm) and large (≥3 mm). Logistic regression models were fitted to assess the association between the exposure and the outcome, after adjustment for age, sex, and level of education.

RESULTS

Of 933 individuals aged ≥40 years enrolled in the Atahualpa Project Cohort, 776 (83%) received an unenhanced CT of the head. The remaining 157 either declined consent for this exam, were severely disabled, emigrated, or died between enrollment and the invitation. In addition, individuals with evidence of one or more of the abovementioned conditions known to be associated with HA were not considered for this study.

Calcified parenchymal brain cysticerci were noted in 90 of 776 participants (11.6%; 95% CI: 9.5–14%). Of them, 64 (71%) had a single calcification and were considered eligible. Eight participants were further excluded because they had epilepsy and one because of lack of MRI, leaving 55 subjects for analysis. Of note, one of the eight excluded persons with epilepsy also had epileptiform discharges on EEG, one had severe obstructive sleep apnea, one had a history of traumatic brain injury, and another did not have a brain MRI.

The mean age of 55 participants was 58.3 ± 13 years, 34 (62%) were women, and 29 (53%) had elementary school education only. Parenchymal brain calcifications were located in the parietal lobes (n = 19), the frontal lobes (n = 17), the occipital lobes (n = 10), the temporal lobes (n = 4), and deep in the brain (n = 5). There were no individuals with infratentorial calcifications in this series. Lesions were more often located in the right (n = 33) than in the left (n = 22) cerebral hemisphere; however, there were no differences in the left or right location of calcifications according to their specific position within the brain parenchyma. Calcifications were small (<3 mm in diameter) in 35 (64%) participants and ≥3 mm in diameter in the remaining 20 (36%).

Sixteen participants (29%) had HA, which was asymmetric in eight (50%); seven (88%) of them had the calcified lesion on the same side as the atrophy. Participants with HA were older (67.2 ± 10.3 years versus 54.7 ± 12.3 years; P <0.001) and more often had elementary school education (75% versus 44%; P = 0.034) than those with normal hippocampi, but there were no differences in the percentages of women (63% versus 62%) across groups.

Hippocampal atrophy was noted in 11 of 20 participants with large calcifications and in 5 of 35 with small calcifications (55% versus 14%; P = 0.001). Univariate analysis did not show significant differences across categories of calcification size in any of the investigated covariates (Table 1).

Table 1.

Characteristics of the study populations across size categories of parenchymal brain calcifications (univariate analysis)

Variable	Small Calcifications (N = 35)	Large Calcifications (N = 20)	P-Value
HA, n (%)	5 (14)	11 (55)	0.001*
Age, years (mean ± SD)	56.8 ± 12	61 ± 14.6	0.254
Being Women, n (%)	19 (54)	15 (75)	0.128
Primary School Education, n (%)	16 (46)	13 (65)	0.168

HA = hippocampal atrophy.

^*Statistically significant result.

An unadjusted logistic regression model showed a significant association between the presence of large calcifications and HA (odds ratio [OR]: 7.33; 95% CI: 2.01–26.7). This significance persisted in a multivariate logistic regression model, after adjustment for age, sex, and level of education (OR: 7.78; 95% CI: 1.72–35.1); increasing age was the single covariate that remained independently significant in this model but did not impact the abovementioned association (Table 2).

Table 2.

Multivariate logistic regression model showing a significant association between the presence of large (≥3-mm) calcifications and HA

HA Covariate	OR	95% CI	P-Value
Large Calcifications	7.78	1.72–35.1	0.008*
Age	1.09	1.01–1.17	0.022*
Being Women	0.85	0.17–4.21	0.847
Primary School Education	1.49	0.27–8.06	0.645

HA = hippocampal atrophy; OR = odds ratio.

^*Statistically significant result.

DISCUSSION

This population-based study, conducted in apparently seizure-free community-dwelling middle-aged and older adults living in an area endemic for cysticercosis, shows a robust association between the presence of calcifications of ≥3 mm and HA, which persists after adjustment for relevant confounders. Participants with large calcifications were 7.8 times more likely to have HA than those with small ones.

Parenchymal brain cysticerci have been associated with HA for years, the first report dating back to 1998.²⁴ That case report described a 47-year-old man with intractable temporal lobe epilepsy and a large calcified cysticercus located within the ipsilateral hippocampus who improved after surgical resection of the hippocampus and the calcification. This preliminary report was followed by several studies that linked NCC, epilepsy, and HA, suggesting that recurrent seizures or status epilepticus may lead to the subsequent development of HA in cysticercosis patients, which, in turn, exacerbates the seizure disorder.^8,9,25–29 In many patients, cysticerci were not necessarily located within or near the hippocampus, suggesting a remote deleterious effect of parasites on hippocampi.^30,31 More recent evidence suggests that HA in patients with parenchymal brain cysticerci and seizures takes some time to develop, since it is most commonly observed in patients with calcified parasites rather than in those with viable cysts.⁷

The abovementioned studies came from clinical or surgical case series of patients presenting with epilepsy (intractable in many cases). However, it has also been postulated that HA may occur in NCC patients without a history of seizures.^30,32 Further support for this hypothesis came from population studies showing a strong association between asymptomatic (seizure-free) subjects with calcified cysticerci in the brain parenchyma and HA.^10,33 The hypothesis favoring a nonepileptic origin of HA in individuals with NCC postulates that periodic remodeling of calcifications with release of trapped parasitic antigens to the brain parenchyma and subsequent exposure to the immune system may account for recurrent bouts of inflammation that trigger remote hippocampal damage.^5,7 This possibility is supported by experimental evidence showing local inflammatory changes in the brain parenchyma surrounding cysticerci associated with axonal swelling and spheroid formation at a considerable distance from the parasites (including the hippocampus). Those axons originate in neurons located near the parasites and end in either the hippocampi themselves or in brain areas that further connect with hippocampi, such as the cerebral cortex or nuclei deep in the brain. This might result in partially deafferented hippocampi, which, in turn, may provide a substrate for HA to develop.³⁴

In this context, our study adds information to the knowledge on the association between NCC and HA by showing that the size of calcified parasites has an important role in the occurrence of HA in apparently seizure-free subjects with NCC. Nevertheless, other pathogenic mechanisms (genetic predisposition, intensity of the inflammatory response) may also account for the association between the size of calcifications and the presence of HA.¹¹

Major strengths of our study include the population-based design, the unbiased selection of participants, and the stringent inclusion criteria that reduced the risk of confounders related to a history of seizures and the presence of more than one calcification, as well as the exclusion of other conditions that may be associated with HA. We must acknowledge, however, that the field instrument used for the recognition of individuals with suspected seizures may not detect some cases of complex partial seizures and that a 1-hour scalp EEG may miss paroxysmal activity in some cases. In addition, underreporting of seizures due to stigma may be a problem in studies aiming to determine history of seizures/epilepsy in individuals living in rural settings.³⁵ This is less likely to have occurred in the present population, since compliance with field personnel is high, as previously described by our group.³⁶ The cross-sectional design is another limitation of the study, since this precluded the assessment of the direction of the association between the main variables investigated. However, biological plausibility strongly suggests that the direction of this relationship goes from the size of calcified parasites to hippocampal damage, since the opposite is unlikely. Although the lack of hippocampal volume measurements may be perceived as another potential weakness, the Scheltens’ medial temporal atrophy scale has proved reliable for visual grading of HA in comparison with volumetric assessment.³⁷ The Scheltens’ scale may not be appropriate for categorizing subfield HA, but it is adequate for estimating the presence of atrophy. In addition, the presence of a single parenchymal brain calcification does not provide a definitive diagnosis of NCC according to current criteria,¹⁹ but given the location and morphological characteristics of the included calcifications and the exclusion of other pathologies that may be potentially associated with nonphysiological supratentorial calcifications, other diagnostic considerations seem unlikely according to a panel of experts of the Infectious Diseases Society of America and the American Society of Tropical Medicine and Hygiene.³⁸ Immune diagnostic tests were not performed because results of either antibody or antigen detection may be unreliable in patients presenting only with calcified cysticerci.^39,40

There is evidence on the effects of bisphosphonates (i.e., etidronate) in reversing the calcification process in the brain.^41,42 In addition, a recent experimental study found that etidronate significantly reduced the likelihood of residual calcifications in pigs with NCC also treated with cysticidal drugs (The Cysticercosis Working Group in Perú; unpublished data). Smaller calcifications may be associated with lesser amounts of antigens released to the brain parenchyma in each of the abovementioned remodeling processes. In this view, etidronate may be advised to reduce the size of parenchymal brain calcified cysticerci, thus preventing further hippocampal damage. Additional studies are needed to corroborate this hypothesis.

In summary, the present study shows a significant association between large parenchymal brain calcified cysticerci and HA in apparently seizure-free community dwellers living in a community of endemicity. Subsequent longitudinal studies are needed to evaluate the progression of HA according to the size of the calcification. In addition, the present study opens avenues of research for the use of agents that may reduce the size of calcifications to reduce the progression of HA.