Brain-Eating Amoebae: Predilection Sites in the Brain and Disease Outcome

Acanthamoeba spp., Balamuthia mandrillaris, and Naegleria fowleri are pathogenic free-living amoebae (1). They are well known to produce fatal central nervous system (CNS) infections, and pathogenic Acanthamoeba spp. can also produce blinding keratitis, which is often associated with the inappropriate use of contact lenses. All three genera are known as amphizoic amoebae, due to their ability to exist as parasitic organisms and to inhabit natural environments as free-living organisms. In nature, Acanthamoeba seems to be most ubiquitous; it can inhabit a variety of environments and has been isolated from soil, water, and air. B. mandrillaris is rather selective, living in the soil, and it has been rarely isolated from water (1–3). Naegleria fowleri, being a thermophilic protist, prefers warm water such as hot springs in temperate zones and lakes in the tropics (4, 5). Acanthamoeba spp. and B. mandrillaris are known to have two stages in their life cycles, i.e., a vegetative trophozoite stage and a dormant cyst form. N. fowleri exhibits a transient flagellate form in addition to the trophozoite and cyst forms (1–6). These forms are interchangeable, depending on the environmental conditions. Among the various forms, the trophozoite form is often the infectious one. These amoebae cause two distinct clinical entities, namely, granulomatous amoebic encephalitis (GAE), caused by pathogenic Acanthamoeba spp. and B. mandrillaris, and primary amoebic meningoencephalitis (PAM), caused by N. fowleri. GAE and PAM are distinguished by their etiologies, risk factors, duration of illness, clinical features, and laboratory and imaging findings (6). N. fowleri is the only known pathogenic species in the genus Naegleria, which consists of over 40 species, that causes human disease, while B. mandrillaris is the only isolated species in the genus Balamuthia. The genus Acanthamoeba is classified into 20 genotypes (T1 to T20) (1–3, 7, 8). These amoebae and their associated infections have garnered increasing scientific and medical attention in recent years due to their poor prognoses, i.e., less than 5% of patients survive if early intervention is not initiated (1, 6). In addition to poor prognoses, cases of amoebic meningoencephalitis are often underreported, misreported, and underrecognized globally, due to lack of awareness, lack of available diagnostic tools, lack of wide distribution of knowledge regarding public health issues and risk factors, especially in developing countries, and the similarity of symptoms to those of other common CNS infections, such as viral and bacterial meningitis. In addition, the pathogenesis and pathophysiology of CNS infections due to the aforementioned free-living amoebae are incompletely understood. For example, PAM is an acute infection that lasts only a few days, while GAE is a chronic to subacute infection that lasts up to several months. Given the nasal route of entry, N. fowleri is likely to have an intimate correlation with the frontal lobe, due to the anatomical proximity of the olfactory bulb to the frontal lobe; the olfactory bulb is terminal to the olfactory neuroepithelium of the nasal passage, traversing the cribriform plate to the brain (1, 6). Although the intranasal route is the route of infection, current administration of drugs (such as amphotericin B) against PAM is via the intravenous route, which causes significant toxicity to other tissues and requires high doses to reach the site of infection at sufficient concentrations to kill the parasite. In contrast, pathogenic Acanthamoeba and B. mandrillaris spread hematogenously and possibly distribute in the frontal lobe, the temporal lobe, and the parietal lobe, likely through the middle cerebral artery, as these cortices are among the main regions for middle cerebral artery supply (9). By comparing the available reported cases of CNS infections due to free-living amoebae, the aim of the present study was to determine the principle sites of infection within the brain, the diagnostic methods employed (premortem and postmortem), and the available treatment regimens, with examples of successful outcomes, with the goal of increasing awareness for the improved management of amoebic meningoencephalitis.

Free-living pathogenic amoebae are now well-recognized agents of brain infections leading to GAE and PAM. GAE is a chronic infection that can last up to several months, while PAM is an acute fulminant infection that lasts a few days (1, 6). It is intriguing to see the distinctive difference of the chronicity in pathogenicity of these amoebae. For example, Acanthamoeba and B. mandrillaris likely enter the host via the lower respiratory tract and/or skin breaks (1, 6). In contrast, N. fowleri enters the host via the nasal route. Recently, another route of entry has been included, i.e., via organ transplantations, with recipients of organ donations acquiring amoebic meningoencephalitis from the donor diagnosed postmortem with amoebic meningoencephalitis of the same genotype (15–18). This is important because amoebae are ubiquitous and nonresponsive to antibiotics, and organ recipients are already rendered immunosuppressed; therefore, any entry of these pathogenic free-living amoebae may have devastating consequences. Although data on risk factors were not available for all cases reviewed in this study, some factors were observed to dictate the susceptibility of patients to amoebic meningoencephalitis. For GAE due to Acanthamoeba, immunosuppression appeared to be a factor (1, 6, 19, 20), while B. mandrillaris was shown to infect immunocompetent individuals in addition to immunocompromised patients (1, 3). Preceding cutaneous lesions are often liable to GAE caused by both types of amoebae. PAM usually occurred in immunocompetent children and young adults (1, 6, 7). However, all patients had history of activities in or near fresh water sources such as swimming pools or hot springs, including recreational activities and religious practices such as ablution, or health care practices such as the use of neti pots. Eliciting a thorough patient history is absolutely paramount for the accurate diagnosis of PAM, and public health preventive measures such as water treatment should be undertaken for high-risk populations.

Neuroimaging studies revealed the locations of lesions in the frontal, parietal, and temporal lobes in most cases of GAE, but lesions in the frontal lobe were much more frequent for N. fowleri. Neuroimaging modalities can have false-negative results, however, and the specificity of neuroimaging for the diagnosis of amoebic meningoencephalitis has yet to be evaluated. In the absence of accurate diagnoses and effective treatments, both diseases often result in death. N. fowleri was found in the CSF more often than the other two types of amoebae, most likely due to its motile flagellated form. However, diagnosis with biopsy samples may be hindered by the inoculum size and the magnitudes of inflammation and necrosis in the tissue sections. In addition to the aforementioned factors, to untrained eyes the morphology of trophozoites in tissue sections bears a close resemblance to that of macrophages, which are also common in acute inflammatory responses. The other challenges in diagnosis include a wide spectrum of differential diagnoses, e.g., brain tumors, multiple sclerosis, lupus encephalitis, progressive multifocal leukoencephalopathy, stroke, meningitis from other causes (viral, tuberculous, or pyogenic), and cerebral toxoplasmosis (1, 6). A recent case of cerebral toxoplasmosis complicated by GAE caused by both Acanthamoeba and B. mandrillaris highlighted the complex nature of the disease, especially because both types of amoebae are known to act as reservoir hosts for many microorganisms (1, 6, 21–23). It is more intriguing that Acanthamoeba and B. mandrillaris meningoencephalitis cases present as vascular diseases (masquerading as cerebral vascular occlusions or aneurysms). This is most likely due to the ability of amoebae to produce endothelial damage, resulting in cytokine release, crossing of the blood-brain barrier, granulomatous inflammation, thromboembolic events, increased vascular permeability, and ultimately necrosis.

For chemotherapeutic strategies, current available delivery routes include intravenous, oral, and intrathecal administration. Systemic antimicrobial treatment has its limitations, however, due to its adverse effects and reduced delivery, together with delayed diagnosis. Other concerns include poor pharmacodynamic and pharmacokinetic profiles of available drugs, solubility, CNS penetration, drug-drug interactions, patient medical conditions, patient tolerance, and Acanthamoeba susceptibility to amoebicidal agents (24). In the case of PAM, the amphotericin B deoxycholate preparation is preferable to the liposomal formulation against N. fowleri infections, although it has no effect on Acanthamoeba or B. mandrillaris (25, 26). More recently, miltefosine has shown promising results with respect to bioavailability and drug-drug interactions (25). Of note, the major groups of azoles and macrolides are amoebistatic rather than amoebicidal. Additionally, nephrotoxic and hepatotoxic effects due to the use of drugs among patients with compromised renal and liver functions (such as transplant patients) may further complicate treatment. Potential drug delivery systems that directly target the inoculation sites of amoebae by circumventing the need for optimal blood-brain barrier penetration should be the focus of future studies, to increase the odds of survival for patients with PAM while minimizing adverse effects and disease complications. Overall, a complete understanding of the pathogenic mechanisms and the role of the immune system and development of novel chemotherapeutic approaches to drug delivery (27, 28) are important for the rational development of antiamoebic therapy.

Despite advances in clinical recognition, diagnostic methods, and treatment approaches, the mortality rates associated with CNS infections due to amoebae have remained high. Although neuroimaging findings reveal common areas for lesions, the locations may not be consistent and may vary depending on the causative agent. High levels of clinical suspicion are important, especially for refractory cases of meningoencephalitis, for rapid diagnosis of the infection, which is a prerequisite for successful treatment. The fact that only a few individuals, among all of the hosts exposed to these amoebae, develop infections suggests the possible presence of underlying predisposing factors. Future research is needed to define genetic, immunological, pathogenic, and environmental factors that contribute to deadly amoebic meningoencephalitis. Moreover, the abilities of pathogenic amoebae to host other microbial pathogens as reservoirs and to act as hyperparasites have enhanced their capacity as pathogens of increasing importance to human and animal health.

This work was supported by Sunway University (Malaysia).

We declare no conflicts of interests regarding the publication of this paper.