Granulomatous Amoebic Meningoencephalitis in an Immunocompetent Patient

Home , Acanthamoeba, Balamuthia mandrillaris, Sappinia, Sappinia diploidea

OBSERVATION Granulomatous Amoebic Meningoencephalitis in an Immunocompetent Patient

Rodrigo Alencar e Silva, MD; Stanley de Almeida Arau´jo, MD; Izabela Faria de Freitas e Avellar, MD; Jose´ Eymard Homem Pittella, MD, PhD; Jose´ Teotoˆnio de Oliveira, MD; Paulo Pereira Christo, MD, PhD

Objective: To report a case of granulomatous amoebic Results: This study shows the diagnosis of B mandril- encephalitis caused by Balamuthia mandrillaris. laris encephalitis suspected from a cerebral biopsy specimen and confirmed by immunohistochemical and poly- Design: Case report. merase chain reaction studies.

Conclusions: This study demonstrates that the diagno- Setting: University hospital. sis of amoebic encephalitis represents a clinical chal- lenge and confirming diagnoses are made, in most cases, Patients: An adult female patient without any appar- after death. High suspicion, histopathologic examina- ent suppressor immune system factor had central ner- tion, and indirect immunofluorescence, polymerase chain vous system infection caused by B mandrillaris. reaction, and cytokine studies from tissue and cerebrospinal fluid are the main devices to reach the diagnosis. Main Outcome Measures: Clinical, neuroimaging, and pathology findings. Arch Neurol. 2010;67(12):1516-1520

RANULOMATOUS AMOEBIC ache in the frontal region that had started encephalitis (GAE) is a 8 days earlier and evolved into a holocra- rare and sporadic central nial headache, whose intensity progres- nervous system infection sively worsened. The patient had had vom- caused by free-living iting over the previous 2 days. There was amoeba.G However, the disease has gained no report of fever at the onset of symp- importance because of the growing num- toms, alcohol abuse, diabetes mellitus, or ber of immunodepressed patients, its diffi- other comorbidities, except for systemic ar- cult diagnosis, lack of adequate treatment, terial hypertension. Additional examina- and high level of mortality. tion revealed mild neck stiffness but no cra- Most cases have been reported in the nial nerve abnormalities or focal signs. United States, Australia, and Europe, Funduscopy was normal. The patient had which might be because of better identi- no lymphadenomegaly, fever, or skin al- fication of patients at these centers and/or terations. A cranial computed tomo- publication bias. To our knowledge, only graphic scan showed discrete sulcus oblit- 2 cases have been described in Brazil, one eration (Figure 1A). The patient had a associated with AIDS and the other, with 1 lumbar puncture for analysis of the cere- alcoholism as a risk factor. brospinal fluid, which showed the follow- Most cases of GAE are related to host im- ing results: elevated opening pressure (44 munosuppression and are caused by amoe- 2,3 cm of water), protein level of 131 mg/dL, bae of the genus Acanthamoeba. A case of glucose level of 57 mg/dL, and 39 nucle- Author Affiliations: GAE caused by Balamuthia mandrillaris in ated cells (69% lymphocytes, 12% neutro- Departments of Clinical an adult immunocompetent patient is re- Neurology (Drs Silva, Oliveira, phils, 15% monocytes, and 4% plasma ported herein. The radiologic and histo- and Christo) and Pathology and cells). Results of gram staining, acid-fast pathologic findings are also described. Forensic Medicine (Drs Arau´ jo, staining, and oncotic cytologic analysis were Avellar, and Pittella), Clinics negative. The VDRL test results were non- Hospital, and Program of REPORT OF A CASE reactive and the cultures for fungi and My- Infectious Diseases and Tropical Medicine, Medical School cobacterium tuberculosis were negative. (Dr Arau´ jo), Federal University A 47-year-old woman from Belo Hori- The initial diagnostic hypothesis was vi- of Minas Gerais, Belo zonte, Minas Gerais, Brazil, was admitted ral meningitis. The patient was kept un- Horizonte, Minas Gerais, Brazil. to the emergency service reporting a head- der observation and showed stabilization

(REPRINTED) ARCH NEUROL / VOL 67 (NO. 12), DEC 2010 WWW.ARCHNEUROL.COM 1516

5 cm

A D A E

L L

5 cm

Figure 1. Cranial computed tomography and magnetic resonance imaging. A, Cranial computed tomography showing discrete obliteration of the sulci. B, Cranial computed tomography showing hypodensity in the left cerebellar hemisphere. C and D, Fluid-attenuated inversion recovery magnetic resonance imaging showing hyperintensity in the left cerebellar hemisphere and frontal, temporal, and occipital lobes. E, Heterogeneous contrast enhancement.

of clinical symptoms over the subsequent 2 days. On the treatment was maintained. The patient had a decreased third day after admission, the patient developed focal signs level of consciousness and underwent orotracheal intu- characterized by horizontal nystagmus, ataxic gait, and bation on the same day. dysmetria on the left side. A cranial computed tomo- A left lateral suboccipital craniectomy was per- graphic scan was repeated and showed hypodensity in formed to obtain a biopsy specimen of a lesion in the left the left cerebellar hemisphere (Figure 1B). The diagnos- cerebellar hemisphere. Blood cell count, coagulogram, tic hypothesis of herpes encephalitis was raised and treat- erythrocyte sedimentation rate, renal and hepatic func- ment with acyclovir, ampicillin, and dexamethasone was tion test results, antinuclear factor results, chest radiog- introduced. The headache improved after 24 hours, but raphy, transthoracic echocardiogram, and serologic test the cerebellar signs persisted. The patient developed results for human immunodeficiency virus were nega- drowsiness after 3 days and underwent nuclear mag- tive and/or showed no alterations. After the procedure, netic resonance imaging of the brain, which showed mul- the patient had an increase in intracranial pressure, loss tiple brain lesions on both cerebral hemispheres, rang- of brainstem reflexes, and hemodynamic instability and ing in diameter from 0.5 to 3 cm. Heterogeneous contrast died 15 days after hospital admission. enhancement was observed especially in the left cerebel- The histopathologic analysis revealed extensive areas lar hemisphere (Figure 1C, D, and E). An empirical treat- of necrosis and hemorrhage in the cerebellum, fibrinoid ment for neurotuberculosis and neurotoxoplasmosis was necrotizing panarteritis, some thrombosis, granuloma- tried. Treatment with rifampicin, isoniazid, pyrazin- tous lymphoplasmacytic inflammatory infiltrate, foamy amide, ethambutol, sulfadiazine, and pyrimethamine was macrophages, isolated multinucleated giant cells, and in- introduced. Ampicillin, acyclovir, and dexamethasone cipient formation of perivascular granulomas. The in-

(REPRINTED) ARCH NEUROL / VOL 67 (NO. 12), DEC 2010 WWW.ARCHNEUROL.COM 1517

C D

Figure 2. Histopathologic examination of the cerebellum. A, Fibrinoid necrotizing panarteritis, with thrombosis, granulomatous lymphoplasmacytic inflammatory infiltrate, and foamy macrophages (hematoxylin-eosin, original magnification ϫ100). B, Isolated multinucleated giant cells and incipient perivascular granulomas (hematoxylin-eosin, original magnification ϫ200). C, Presence of diverse structures (isolated or forming small clusters, marked by asterisks) with morphologic characteristics of amoeba trophozoites seen in the vessel wall and areas with and without an inflammatory reaction (hematoxylin-eosin, original magnification ϫ200). D, Detail showing a vessel completely filled with trophozoites (hematoxylin-eosin, original magnification ϫ400).

flammatory infiltrate extended focally to the adjacent lep- drill baboon that died of encephalitis at the San Diego tomeninges. Different structures (isolated or forming small Zoo.7 This species causes encephalitis in both immuno- clusters) with the morphological characteristics of amoeba depressed and immunocompetent individuals as well as trophozoites were identified in the vascular wall and in in animals.4,8 B mandrillaris and various species of Acan- areas with and without an inflammatory reaction thamoeba are opportunistic agents that cause the clini- (Figure 2). Part of the material was sent to the Armed cal presentation of GAE in debilitated and malnour- Forces Institute of Pathology, Washington, DC. Immu- ished patients; different types of immunocompromised nohistochemical and polymerase chain reaction studies patients, including those with AIDS; and children.5 In con- were then performed and revealed the presence of tro- trast with Acanthamoeba, which preferentially occurs in phozoites of the free-living amoeba B mandrillaris. immunocompromised patients, Balamuthia is also seen in immunocompetent patients, particularly children.9 By COMMENT 2007, approximately 150 cases of infection with Bala- muthia had been reported worldwide since the recogni- The involvement of free-living amoebae in human dis- tion of the disease in 1990.8 eases was only recognized after 1965, when the first fa- Balamuthia and Acanthamoeba species show a ubiq- tal cases of meningoencephalitis were described in Aus- uitous distribution, with organisms being found in soil, tralia and, almost at the same time, the United States.2 water, heaters, and air-conditioning units.10 The former Free-living amoeba species causing central nervous sys- species is probably transmitted by inhalation of air- tem injury include Naegleria fowleri, Acanthamoeba spe- borne cysts or by direct contamination of skin lesions.8 cies, and B mandrillaris.4,5 More recently, Sappinia diploidea The few patients with GAE caused by B mandrillaris had has been implicated as a rare cause of meningoencepha- not been exposed to water but had a history of skin le- litis in humans.6 B mandrillaris is a free-living amoeba sions before the emergence of neurological symp- that was first identified in 1989 in the brain of a man- toms.11-14 In a review including 24 patients with B man-

(REPRINTED) ARCH NEUROL / VOL 67 (NO. 12), DEC 2010 WWW.ARCHNEUROL.COM 1518

©2010 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 drillaris infection, 92% had skin lesions.4 The present Magnetic resonance imaging reveals hyperintense mul- patient had no skin lesions and the probable route of in- tifocal lesions on T2-weighted images that show a vasion of the pathogen might have been the respiratory heterogeneous or ringlike enhancement. The lesions are tract, followed by hematogenic dissemination to the cen- preferentially located in the diencephalon, brainstem, tral nervous system. and structures of the posterior fossa.20 A brain biopsy is The risk factors of Balamuthia encephalitis are not well important for the diagnosis of GAE and can be part of established. In a series of 10 patients, 5 had comorbidi- the treatment in cases in which no dissemination of iso- ties such as diabetes, heart disease, a previous splenec- lated lesions to the brain is observed.10 tomy, nephrotic syndrome associated with prolonged ste- Various alternatives have been proposed for the treatment roid treatment, and possible lymphoma. Five of the 10 ofGAE.Combinationtreatmentswithpentamidine,flucona- patients had been exposed to soil.8 Alcoholism and pro- zole, sulfadiazine, flucytosine, azithromycin, and clarithro- longed antibiotic therapy have also been considered to mycin have been described.4,21 This combination was ad- be risk factors.15 Balamuthia encephalitis was a fatal com- ministered to 2 patients who successfully recovered and plication in 2 patients who underwent renal transplant there was no evidence of recrudescence of the disease 2 and at the University of Mississippi and it appears that the 6 years after the onset of symptoms in these cases.21 The donor had asymptomatic Balamuthia encephatitis.16 Be- prognosis is generally obscure and mortality is approxi- cause of this, organ transplant should also be consid- mately 100%.4 ered as a risk factor for this infection. There is a pre- Granulomatous amoebic encephalitis caused by B man- dominance of cases in young (Ͻ15 years) and elderly drillaris is a rare disease and barely known among phy- (Ͼ60 years) individuals, which may be attributed to some- sicians. Because of the lack of pathognomonic symp- what weaker immune systems. However, no risk factors toms and diagnostic difficulties, numerous complementary could be identified in most cases reported in the litera- tests are generally performed and different empirical treat- ture7 or in the present case. ments are indicated, as in the present case. The diagnos- The diagnosis of GAE in the living requires caution tic hypothesis of GAE caused by B mandrillaris should and suspicion, since its symptoms mimic various types therefore be considered in the case of symptoms of acute of encephalitis. Few appropriate laboratory tests for and subacute meningoencephalitis, even in countries diagnosis are available and many physicians are where this infection is rarely described, such as Brazil, unaware of the disease.8,10 The methods available for the and in immunocompetent patients who do not have any diagnosis of GAE caused by B mandrillaris include his- risk factors. A biopsy should always be performed be- tologic analysis of hematoxylin-eosin–stained speci- cause of the possibility of treatment if the disease is di- mens and detection of amoebae in tissue samples and agnosed at an early stage. serum antibodies using indirect immunofluorescence. Cell culture, polymerase chain reaction, and cerebrospi- 4 Accepted for Publication: March 15, 2010. nal fluid analysis may also be done. Immunohisto- Correspondence: Stanley de Almeida Arau´ jo, MD, Rua chemical analysis permits the detection of trophozoites, Carangola 82 apto 604, Santo Antoˆnio, Belo Horizonte, which are difficult to identify by hematoxylin-eosin Minas Gerais 30130-100, Brazil (stanleyaa@gmail staining in areas of necrosis or when macrophages are .com). abundant. Immunohistochemical analysis also can Author Contributions: Study concept and design: Silva, show granular antigens inside macrophages and the Arau´ jo, Avellar, Oliveira, and Christo. Acquisition of data: blood vessel wall.17 Most cases are identified in retro- 4,10 Silva, Arau´ jo, Avellar, Oliveira, and Christo. Analysis and spective postmortem studies. Recently, the California interpretation of data: Silva, Arau´ jo, Avellar, Pittella, Ol- Department of Public Health study revealed that in iveira, and Christo. Critical revision of the manuscript for Balamuthia encephalitis, cerebrospinal fluid may show important intellectual content: Silva, Arau´ jo, Avellar, Pit- elevations in the levels of cytokines interleukin 6 and tella, Oliveira, and Christo. Study supervision: Silva, Arau´ jo, interleukin 8, which could differentiate it from other 18 Avellar, Pittella, Oliveira, and Christo. types of encephatitis. There is also a triplex real-time Financial Disclosure: None reported. polymerase chain reaction assay for Naegleria, Acanth- amoeba, and Balamuthia developed by the Centers for Disease Control and Prevention.19 REFERENCES An anatomopathologic examination of central ner- 1. Silva-Vergara ML, Da Cunha Colombo ER, De Figueiredo Vissotto E, et al. Dis- vous system lesions requires diligence and experience. seminated Balamuthia mandrillaris amoeba infection in an AIDS patient from Brazil. In the present case, this analysis involved consideration Am J Trop Med Hyg. 2007;77(6):1096-1098. of important differential diagnoses, such as primary 2. Foronda AS. Infecço˜es por Amebas de Vida Livre. Tratado de Infectologia.3rd angiitis of the central nervous system, vasculitis, giant ed. Vol 2. Saõ Paulo, Brazil: Editora Atheneu; 2005:1461-1470. cell arteritis, and infection by viruses, fungi, and other 3. Martıńez AJ, Guerra AE, Garcıá-Tamayo J, Ce´spedes G, Gonza´lez-Alfonzo JE, Vis- vesvara GS. Granulomatous amebic encephalitis: a review and report of a spon- protozoa. There are no specific characteristics in clini- taneous case from Venezuela. Acta Neuropathol. 1994;87(4):430-434. cal, laboratory, or radiologic findings for the diagnosis 4. Cuevas P M, Smoje P G, Jofre´ M L, et al. Granulomatous amoebic meningoen- of GAE. Computed tomography and magnetic reso- cephalitis by Balamuthia mandrillaris: case report and literature review [in Spanish]. nance imaging generally show 1 or more contrast- Rev Chilena Infectol. 2006;23(3):237-242. 5. Oddo´ B D, Ciani A S, Vial C P. Granulomatous amebic encephalitis caused by enhanced lesions that are nonspecific and can be seen Balamuthia mandrillaris: first case diagnosed in Chile [in Spanish]. Rev Chilena in other conditions such as fungal or bacterial infec- Infectol. 2006;23(3):232-236. tions, tuberculoma, toxoplasmosis, or neoplasms.10 6. Gelman BB, Popov V, Chaljub G, et al. Neuropathological and ultrastructural fea-

(REPRINTED) ARCH NEUROL / VOL 67 (NO. 12), DEC 2010 WWW.ARCHNEUROL.COM 1519

©2010 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 tures of amebic encephalitis caused by Sappinia diploidea. J Neuropathol Exp 14. Valverde J, Arrese JE, Pie´rard GE. Granulomatous cutaneous centrofacial and Neurol. 2003;62(10):990-998. meningocerebral amebiasis. Am J Clin Dermatol. 2006;7(4):267-269. 7. Visvesvara GS, Martinez AJ, Schuster FL, et al. Leptomyxid ameba, a new agent 15. Katz JD, Ropper AH, Adelman L, Worthington M, Wade P. A case of Balamuthia of amebic meningoencephalitis in humans and animals. J Clin Microbiol. 1990; mandrillaris meningoencephalitis. Arch Neurol. 2000;57(8):1210-1212. 28(12):2750-2756. 16. CDC: rare infection passed on by Miss. organ donor. www.physorg.com 8. Balamuthia amebic encephalitis California, 1999-2007. JAMA. 2008;300(21):2477- /pdf180469832.pdf. Accessed March 1, 2010. 2479. doi:10.1001/jama.300.21.2477. 17. Guarner J, Bartlett J, Shieh WJ, Paddock CD, Visvesvara GS, Zaki SR. Histo- 9. Visvesvara GS, Moura H, Schuster FL. Pathogenic and opportunistic free-living pathologic spectrum and immunohistochemical diagnosis of amebic amoebae: Acanthamoeba spp., Balamuthia mandrillaris, Naegleria fowleri, and meningoencephalitis. Mod Pathol. 2007;20(12):1230-1237. Sappinia diploidea. FEMS Immunol Med Microbiol. 2007;50(1):1-26. 18. Schuster FL, Yagi S, Gavali S, et al. Under the radar: Balamuthia amebic encephalitis. 10. Deol I, Robledo L, Meza A, Visvesvara GS, Andrews RJ. Encephalitis due to a Clin Infect Dis. 2009;48(7):879-887. free-living amoeba (Balamuthia mandrillaris): case report with literature review. 19. Qvarnstrom Y, Visvesvara GS, Sriram R, da Silva AJ. Multiplex real-time PCR Surg Neurol. 2000;53(6):611-616. 11. Galarza M, Cuccia V, Sosa FP, Monges JA. Pediatric granulomatous cerebral am- assay for simultaneous detection of Acanthamoeba spp., Balamuthia mandril- ebiasis: a delayed diagnosis. Pediatr Neurol. 2002;26(2):153-156. laris, and Naegleria fowleri. J Clin Microbiol. 2006;44(10):3589-3595. 12. Intalapaporn P, Suankratay C, Shuangshoti S, Phantumchinda K, Keelawat S, Wilde 20. Singh P, Kochhar R, Vashishta RK, et al. Amebic meningoencephalitis: spec- H. Balamuthia mandrillaris meningoencephalitis: the first case in southeast Asia. trum of imaging findings. AJNR Am J Neuroradiol. 2006;27(6):1217- Am J Trop Med Hyg. 2004;70(6):666-669. 1221. 13. Pritzker AS, Kim BK, Agrawal D, Southern PM Jr, Pandya AG. Fatal granuloma- 21. Deetz TR, Sawyer MH, Billman G, Schuster FL, Visvesvara GS. Successful treat- tous amebic encephalitis caused by Balamuthia mandrillaris presenting as a skin ment of Balamuthia amoebic encephalitis: presentation of 2 cases. Clin Infect lesion. J Am Acad Dermatol. 2004;50(2)(suppl):S38-S41. Dis. 2003;37(10):1304-1312.

Announcement

New Initiatives: Clinical Trials and Videos. We have embarked on 2 new initiatives: Clinical Trials and video presentations. We welcome manuscripts that describe double-blind, randomized, placebo-controlled clinical trials as our primary area of interest. We plan on expe- diting the review process and time to publication and to include them online ahead of print as these studies are time sensitive and of direct benefit to our patients. We hope you will take advantage of this new initiative. Please refer to the Instructions for Authors when submitting a Clinical Trials paper, including the requirement to reg- ister the trial with an accepted clinical trials site. We plan to utilize videos as part of published papers that highlight and provide convincing information about the observational and visual features of a patient’s neu- rologic findings. Please refer to Instructions for Au- thors for instructions on submitting video presentations.

(REPRINTED) ARCH NEUROL / VOL 67 (NO. 12), DEC 2010 WWW.ARCHNEUROL.COM 1520