Abstract
Toxoplasmosis encephalitis is a severe, but often misdiagnosed complication in patients after bone marrow transplantation (BMT). We describe the unique computed tomography (CT) and magnetic resonance (MR) imaging features of cerebral toxoplasmosis in two bone marrow recipients and compare them to the cases in the literature. To our knowledge, this is the first report analyzing the appearance of cerebral toxoplasmosis on diffusion-weighted MR imaging (DWI).
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
Introduction
Toxoplasmosis is an opportunistic protozoal infection primarily affecting patients with immune deficiencies. The incidence varies geographically, with a higher frequency of seropositivity in the general population in Central Europe than in North America [1, 2]. Toxoplasmosis is most commonly seen in human immunodeficiency virus (HIV) -positive patients and is a rare but almost always fatal complication in patients following allogeneic bone marrow transplantation.
Disseminated toxoplasmosis after bone marrow transplantation (BMT) has been reported in 1% to 7.6% of allogeneic transplant recipients [3–5]. Due to the deranged cellular and humoral immune response during the first year after BMT, patients usually do not show typical clinical or imaging signs of cerebral infection [4, 6]. The final diagnosis of intracranial toxoplasmosis is usually made at autopsy.
Since 1992, approximately 100 cases of toxoplasmosis encephalitis in patients after BMT have been reported in the literature [6–15]. The diagnosis was based on imaging findings, polymerase chain reaction (PCR) and/or brain biopsy, and standardized treatment with pyrimethamine and sulfodiazine was given in some of the cases. In the majority of the published cases, the diagnosis was established at autopsy without specific anti-microbial therapy being given prior death.
We describe two proven cases with isolated cerebral toxoplasmosis in patients following BMT with a fatal outcome despite antitoxoplasmosis therapy.
Case reports
Case 1
A 41-year-old woman had allogeneic BMT for treatment of acute myelocytic leukemia (AML). Immediately after BMT, she received intravenous steroids to prevent graft-versus-host disease (GVHD). Two months later, based on increasing CMV-copies in the blood, cytomegaly virus (CMV) infection was diagnosed and successfully treated with gancyclovir. At day +119 after BMT, the patient was hospitalized because of apathy, abnormal fatigue and reduced general condition. Her leukocyte count was 6,400/μl and the thrombocyte count was 26,000/μl. Computed tomography (CT) scan showed three hyperdense lesions with mild perifocal edema located in the grey-white matter junction and in the deep central nuclei. The initial magnetic resonance (MR) examination two days later revealed 34 lesions in both hemispheres and in the cerebellum, with a maximum size of 2 cm. On fluid-attenuated inversion recovery (FLAIR)-weighted MR images most lesions were hyperintense, and some lesions were slightly hypointense centrally (Fig. 1a). On unenhanced T1-weighted images two lesions were hyperintense, two lesions were hypointense and all other lesions were isointense (Fig. 1b). The lesions showed no contrast enhancement on post-contrast images and no mass effect. On diffusion-weighted imaging (DWI), the lesions with hyperintensity on T1-weighted images showed low signal with a hyperintense rim consistent with subacute hemorrhage. The other lesions were isointense on DWI. The findings were suggestive of toxoplasmosis encephalitis and treatment with pyrimethamine and sulfodiazine was initiated. Despite therapy, the patient showed neurologic deterioration and was somnolent 5 days after the onset of symptoms. Follow-up CT and MR scans 8 days and 10 days after the initial CT scan revealed significant progression with more than 40 lesions supra- and infratentorially. All lesions were hyperintense on unenhanced T1-weighted MR images on follow-up scan. On gradient-echo MR images, the lesions were of mixed signal-intensity centrally and surrounded by a hypointense rim (Fig. 1c). On DWI, all lesions showed a low signal with a hyperintense rim (Fig. 1d). To rule out other infectious agents, stereotactic biopsy of a subcortical lesion was performed. Histology showed necrotizing encephalitis compatible with toxoplasmosis and the antitoxoplasmic therapy was continued. Nevertheless, the patient died 18 days after hospitalization. Brain autopsy revealed widespread hemorrhagic-necrotizing toxoplasmosis encephalitis of the cerebral hemispheres, cerebellum, basal ganglia and pons (Fig. 1e). There was no sign of toxoplasmosis in liver, spleen or lung immunohistochemically.
Case 2
A 63-year-old man underwent two autologic BMTs two and three years prior to presentation, and finally received an allogeneic BMT for treatment of multiple myeloma. Nine months after the last BMT, he was admitted to the hospital because of thrombotic-thrombocytopenic purpura and activation of a known chronic hepatitis B. In addition, he had a chronic extensive GVHD of skin, oral mucosa, eyes and liver which was treated with aprednisolon (1 mg/kg/day). His leukocyte count was 6,000/μl. Six weeks after hospitalization (day +280 after BMT) he showed increasing fatigue and listlessness. One week later, he developed a palsy of the VII cranial nerve and a left-sided spasticity. Initial CT scan showed multiple hypodense lesions in both cerebral hemispheres and in the cerebellum, with a maximum size of 2 cm (Fig. 2a). MR examination revealed more than 20 lesions that showed hyperintensity on T2-weighted images (Fig. 2b,c). None of the lesions showed contrast enhancement and only some of the lesions were surrounded by mild edema (Fig. 2d). On DWI, two types of lesions were seen. Two lesions showed a low signal with a hyperintense rim and were hyperintense on T1-weighted images. These imaging findings were consistent with hemorrhagic lesions. All other lesions showed high signal on DWI and low signal on the ADC map, and were iso- or slightly hypointense on T1-weighted images consistent with restricted diffusion (Fig. 2e,f). The MRI findings were suspicious for toxoplasmosis encephalitis and antimicrobial therapy with pyrimethamine, sulfodiazine and dalacin was started immediately. After a short-term clinical improvement, the patient showed rapid deterioration and died 14 days after the onset of neurologic symptoms. Brain autopsy revealed multifocal hemorrhagic-necrotizing toxoplasmosis encephalitis. There was no evidence of toxoplasmosis in other organs.
Discussion
Toxoplasmosis encephalitis is a rare, but severe complication in bone marrow recipients usually appearing between the 2nd and 6th month after BMT [4, 16]. The clinical symptoms are non-specific, with fever, seizures, headaches or altered mental status the most common manifestations [5, 8]. Cerebrospinal fluid (CSF) analysis is often inconclusive and PCR for toxoplasmosis DNA in different organs, including the brain in BMT patients, is reported to be positive only in 75% of infected organs [11, 17]. These facts emphasize the importance of recognizing the patterns of toxoplasmosis encephalitis on initial imaging examinations. In HIV-positive patients, the imaging findings in toxoplasmosis are well known. MRI usually shows multifocal lesions with hypo-, iso- or hyperintensity on T2 and iso- or hypointensity on T1-weighted images [18]. After application of gadolinium, the lesions typically show ring or nodular enhancement and, commonly, edema and hemorrhage are present.
One of the main differences between patients with HIV and patients after BMT is the fact that toxoplasmic lesions are often initially hemorrhagic in BMT population, whereas hemorrhagic transformation will be only seen in HIV population after initiation of the antitoxoplasmic therapy. Enhancement of the lesions will be present in all patients with HIV and improves characterization of a lesion seen on the unenhanced images, but may be absent in BMT group [19].
To the best of our knowledge, we have found seven reports describing the CT and MRI features of toxoplasmosis encephalitis after BMT in the English literature to date, for a total of 26 patients [5, 6, 10–14]. The time interval between BMT and onset of cerebral toxoplasmosis for these patients was between 14 and 689 days (mean, 154.3±178.5 days). Twelve patients survived without relapse, and the survival time of the 13 patients who died due to toxoplasmosis ranged from 10 days to 456 days (mean: 219.5±224.2 days). The death of one patient was not due to cerebral toxoplasmosis. In two cases, diagnosis was made postmortem by brain autopsy; in all other cases diagnosis was based on neuroradiological and clinical criteria (19 cases) or brain biopsy (5 cases). All patients with suspected or confirmed toxoplasmosis encephalitis received a therapy regimen with pyrimethamine and sulfadiazine (22 cases) or pyrimethamine and clindamycin (4 cases). In our two cases, the onset of cerebral symptoms due to toxoplasmosis was on day 119 (Case 1) and on day 280 after BMT (Case 2), and the diagnosis was based on CT and MR imaging findings; in one patient, a brain biopsy was performed. Both patients died 18 and 14 days, respectively, after symptom onset despite immediate antitoxoplasmic therapy with pyrimethamine and sulfodiazine.
In 24 of the 26 patients described in the literature, MR examination showed multiple lesions in the basal ganglia and, subcortically in supra- and infratentorial location. In two patients, MRI showed only one and two lesions in the cerebral hemispheres, respectively. Depending on the enhancement characteristics of the lesions and the presence of edema, MRI revealed two different features. In 16 patients, the lesions showed the typical MR appearance of cerebral toxoplasmosis with nodular or ring-enhancement surrounded by moderate to extensive edema. In contrast, the lesions in 10 patients showed no enhancement and there was no edema. In seven cases, some lesions were hyperintense on non-enhanced T1-weighted images, suggesting hemorrhage. In our two cases, MRI revealed multiple lesions in a supra- and infratentorial location that were also non-enhancing and showed no or mild edema. In our first patient, all lesions were hemorrhagic, whereas in the second patient, only two out of more than 20 lesions were hemorrhagic.
In a previous study involving seven patients with cerebral fungal or viral infection after BMT, Yuh et al. also reported a lack of contrast enhancement and edema in patients with low total white blood cell count (WBC), whereas in one patient with a normal total WBC, all lesions showed marked contrast enhancement and edema [20]. Therefore, they explained the lack of contrast enhancement as a result of reduced inflammatory response due to the impaired humoral and cellular immunity, preventing the passage of gadolinium through the blood-brain barrier. This hypothesis is supported by two other studies dealing with the MRI features of toxoplasmosis encephalitis. In a study by Maschke et al., patients with non-enhancing lesions also had low leukocyte counts compared to patients with a typical appearance of cerebral toxoplasmosis and normal leukocyte counts [12]. In addition, a recent study of cerebral toxoplasmosis in immunocompetent patients showed marked linear or ring enhancement of all brain lesions [21].
Because of the limited data, a correlation between contrast enhancement and leukocyte count in the 26 cases from the literature was not possible. However, it is remarkable that in the two patients from our institute, leukocyte counts were within normal limits and the lesions showed no contrast enhancement and only discrete or no edema. These findings suggest that the WBC is not the only predictor for the inflammatory response against cerebral toxoplasmosis. In general, patient 2 was at higher risk of developing a toxoplasmosis infection because he received immunosuppressive therapy with high-dose corticosteroids for treatment of a chronic GVHD. The literature reveals that prior GVHD is a risk factor for toxoplasmosis after BMT [15, 22]. The corticosteroid therapy in our patient may also be responsible for the lack of enhancement and edema of the toxoplasmic lesions. Only a few studies have commented upon the influence of corticosteroids on the enhancement of brain lesions. They reported a reduced enhancement after application of corticosteroids, but the effect on the blood-brain barrier is still unclear [23–25]. Patient 1 did not receive any immunosuppressive medication and we have no explanation for the lack of enhancement and edema of the brain lesions on her initial CT and MR scan. On follow-up CT and MR scan all lesions were hemorrhagic; therefore, we cannot comment on enhancement. The extensive hemorrhagic transformation of the lesions in this patient was probably due to the distinct depression of thrombocytes.
With regard to the outcome of the 10 patients with atypical appearance of cerebral toxoplasmosis and our two cases, the overall mortality was 76% (9/12 patients). In the 16 patients with typical ring-enhancing lesions, the overall mortality rate was 38% (6/16 patients). These findings suggest that typical MRI features of toxoplasmosis encephalitis (i.e., contrast enhancement of the lesions) indicate an inflammatory response due to a more intact immune system and, therefore, a higher probability of survival.
Another fact influencing the outcome of the patients with toxoplasmosis encephalitis is the time of onset of disease. According to a retrospective study of 110 cases of toxoplasmosis after BMT, the outcome was better in the subjects with a late onset of disease (>100 days) which is obviously due to recovery of the immune system in this period of time. The outcome was worst in disseminated infection (64% of fatal cases had disseminated toxoplasmosis) [8]. In contrast to these findings, both of our patients showed a fulminant and fatal course of disease despite late onset (day 119 and 180 after BMT) and isolated cerebral manifestation of toxoplasmosis.
The literature regarding DWI of cerebral toxoplasmic lesions in BMT patients is sparse. We only found a single case report by Ionita et al. describing few hyperintense lesions on DWI in a patient; however, the apparent diffusion coefficient (ADC) was not measured [6]. On DWI of our patients, two types of lesions were seen: lesions with low signal on DWI and a hyperintense rim consistent with hemorrhage, and lesions hyperintense on DWI with reduced ADC consistent with restricted diffusion. Restricted diffusion was found in brain abscesses [17]. The same imaging features have been described in AIDS patients with toxoplasmosis encephalitis [26].
In conclusion, cerebral toxoplasmosis should be considered in the differential diagnosis of bone marrow recipients presenting with multiple lesions in CT and MRI even if there is an absence of contrast enhancement and edema. Normal white blood cell count and the late onset of disease do not exclude toxoplasmosis. DWI showed two different patterns depending on whether hemorrhagic transformation occurred. The role of DWI in the differentiation of toxoplasmosis from other infections/neoplastic lesions should be more extensively investigated.
References
Jones J, Kruszon-Moran D, Wilson M (2003) Toxoplasma gondii infection in the United States, 1999–2000. Emerg Infect Dis 9:1371–1374
Zuber P, Jacquier P (1995) Epidemiology of toxoplasmosis: worldwide status. Schweiz Med Wochenschr Suppl 65:19–22
Slavin MA, Meyers JD, Remington JS, Hackman RC (1994) Toxoplasma gondii infection in marrow transplant recipients: a 20 year experience. Bone Marrow Transplant 13:549–557
Derouin F, Gluckman E, Beauvais B et al (1986) Toxoplasma infection after human allogenic bone marrow transplantation: clinical and serological study of 80 patients. Bone Marrow Transplant 1:67–73
Small TN, Leung L, Stiles J et al (2000) Disseminated toxoplasmosis following T cell-depleted related and unrelated bone marrow transplantation. Bone Marrow Transplant 25:969–973
Ionita C, Wasay M, Balos L, Bakshi R (2004) MR imaging in toxoplasmosis encephalitis after bone marrow transplantation: paucity of enhancement despite fulminant disease. Am J Neuroradiol 25:270–273
Wolff D, Becker C, Kubel M et al (1998) Second unrelated bone marrow transplantation without additional conditioning therapy after engraftment failure. Bone Marrow Transplant 21:325–327
Mele A, Paterson PJ, Prentice HG, Leoni P, Kibbler CC (2002) Toxoplasmosis in bone narrow transplantation: A Report of two cases and systematic review of the literature. Bone Marrow Transplant 29:691–698
Gonzalez MI, Caballero D, Lopez C et al (2000) Cerebral toxoplasmosis and Guillain-Barre syndrome after allogeneic peripheral stem cell transplantation. Transpl Infect Dis 2:145–149
Dietrich U, Maschke M, Dörfler A, Prumbaum M, Forsting M (2000) MRI of intracranial toxoplasmosis after bone marrow transplantation. Neuroradiology 42:14–18
Bleggi-Torres LF, De Medeiros BC, Werner B, Pasquini R, De Medeiros CR (1999) Unusual presentation of toxoplasmosis after BMT. Bone Marrow Transplant 23:855–856
Maschke M, Dietrich U, Prumbaum M et al (1999) Opportunistic CNS infection after bone marrow transplantation. Bone Marrow Transplant 23:1167–1176
Tefferi A, O'Neill BP, Inwards DJ (1998) Late-onset cerebral toxoplasmosis after allogeneic bone marrow transplantation. Bone Marrow Transplant 21:1285–1286
Brinkman K, Debasi S, Sauerweun R et al (1998) Toxoplasma retinitis/encephalitis 9 months after allogeneic bone marrow transplantation. Bone Marrow Transplant 21:635–636
Chandrasekar PH, Momin F (1997) Disseminated toxoplasmosis in marrow recipients: a report of three cases and a review of the literature. Bone Marrow Transplant 19:685–689
De Medeiros BC, De Medeiros CR, Werner B et al (2001) Disseminated toxoplasmosis after bone marrow transplamtation: report of 9 cases. Transpl Infect Dis 3:24–28
Held TK, Kruger D, Switala AR et al (2000) Diagnosis of toxoplasmosis in bone marrow transplant recipients: comparison of PCR-based results and immunohistochemistry. Bone Marrow Transplant 25:1257–1262
Chang L, Cornford ME, Chiang FL et al (1995) Radiologic-pathologic correlation. Cerebral toxoplasmosis and lymphoma in AIDS. Am J Neuroradiol 16:1653–1663
Malcolm PN, Howlett DC, Saks A et al (1999) MRI of the brain in HIV-positive patients: what is the value of routine intravenous contrast medium? Neuroradiology 41:687–695
Yuh WTC, Nguyen HD, Gao F et al (1994) Brain parenchymal infection in bone marrow transplantation patients: CT and MR findings. Am J Roentgenol 162:425–430
Vastava PB, Pradhan S, Jha S et al (2002) MRI features of toxoplasma encephalitis in the immunocompetent host. Neuroradiology 44:834–838
Martino R, Maertens J, Bretagne S et al (2000) Toxoplasmosis after Hematopoietic stem cell transplantation. Clin Infect Dis 31:1188–1194
Zaki HS, Jenkinson MD, Du Plessis DG et al (2004) Vanishing contrast enhancement in malignant glioma after corticosteroid treatment. Acta Neurochir 146:841–845
Enzmann DR, Britt RH, Placone BC et al (1982) The effect of short-term corticosteroid treatment on the CT appearance of experimental brain abscesses. Radiology 145:79–84
Troiano RA,Hafstein MP, Zito G et al (1985) The effect of oral corticosteroid dosage on CT enhancing multiple sclerosis plaques. J Neurol Sci 70:67–72
Batra A, Tripathi RP, Gorthi SP (2004) Magnetic resonance evaluation of cerebral toxoplasmosis in patients with the acquired immunodeficiency syndrome. Acta Radiol 45:212–221
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Mueller-Mang, C., Mang, T.G., Kalhs, P. et al. Imaging characteristics of toxoplasmosis encephalitis after bone marrow transplantation: report of two cases and review of the literature. Neuroradiology 48, 84–89 (2006). https://doi.org/10.1007/s00234-005-0018-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00234-005-0018-3