Published ahead of print on May 27, 2010
doi: 10.3174/ajnr.A2158

American Journal of Neuroradiology 31:E65-E69, August 2010
© 2010 American Society of Neuroradiology

R.F. Cabral^a, P.R. Valle Bahia^a and E.L. Gasparetto^a
aDepartment of Radiology

L. Chimelli^b
bDepartment of Pathology
University Federal of Rio de Janeiro
Rio de Janeiro, Brazil

Immune reconstitution inflammatory syndrome (IRIS) is a clinical deterioration that occurs despite increasing CD4+ T-lymphocyte counts and decreasing plasma human immunodeficiency virus-1 (HIV-1) viral loads. It results from an inflammatory response or “dysregulation” of the immune system to both intact subclinical pathogens and residual antigens. Resulting clinical manifestationsof this syndrome are diverse and depend on the infectious or noninfectious agent involved.¹ Neurologic symptoms and signs of IRIS have been reported in association with clinical and subclinical infections, such as cryptococcosis, tuberculosis, mycobacteriosis, and progressive multifocal leukoencephalopathy,² as well as noninfectious phenomena. There are very rare reports of neurologic IRIS associated with cerebral toxoplasmosis.^3,4

A 26-year-old woman with HIV-1 infection for 8 years presented with a 1-month history of ataxia, left-sided weakness, and hyperreflexia. She was not taking highly active antiretroviral therapy (HAART) regularly and was referred to treatment for cerebral toxoplasmosis 4 years before with clinical and radiologic improvement. Currently, a CT scan showed scattered calcified lesions in the brain parenchyma with no perilesional edema or contrast enhancement. On clinical examination, her mental status was normal, and no papilledema was present. Relevant laboratory tests were a positive HIV-1 test result, a CD4 cell count of 276 x 10⁹ cells/L, and a plasma HIV-1 ribonucleic acid (RNA) load of 105.452 copies/mL. CSF evaluation revealed the following values: 3 cells/mL; protein, 32 mg/dL; glucose, 44 mg/dL; sterile cultures; and negative polymerase chain reaction for JC virus (JCV), cytomegalovirus (CMV), herpes simplex virus (HSV), Toxoplasma gondii, and Epstein-Barr virus. Brain MR imaging showed multiple areas of high signal intensity on fluid-attenuated inversion recovery (FLAIR) images, some presenting nodular or ring enhancement, located mainly in the subcortical white matter and deep gray matter (Fig 1A, –B). The larger lesion, seen in the left frontal lobe, had marked low signal intensity on the T2*-weighted images, suggesting calcification, corroborating the CT findings. Treatment was initiated with sulfadiazine, pyrimethamine, and folinic acid. HAART was restored with tenofovir, lamivudine, lopinavir, and ritonavir. She was treated for 3 weeks without clinical or radiologic improvement.

View larger version (123K): [in this window] [in a new window] Fig 1. A, FLAIR image at admission shows multiple areas of high signal intensity, some with a central area of low signal intensity, located in the subcortical and periventricular white matter. B, T1-weighted image after gadolinium administration demonstrates mild ring enhancement in the left frontal lesion, as well as in the lesion in the head of the right caudate nucleus. C, FLAIR image 1 month later shows significant increase in the size of the areas of high signal intensity in the white matter. D, T1-weighted image after gadolinium administration demonstrates stronger enhancement in the lesions at the left frontal lobe and right caudate nucleus, as well as new nodular areas of enhancement bilaterally.

One month after the beginning of treatment for cerebral toxoplasmosis, she presented with worsening of clinical signs and symptoms. A new brain MR imaging revealed enlargement of most of the lesions, mainly with perilesional high signal intensity on FLAIR images, as well as stronger contrast enhancement (Fig 1 C, –D). At this time, the CD4 cell count increased to 495 cells/mL, and the HIV-1 RNA load decreased to 768 copies/mL. A stereotactic brain biopsy of the left frontal lobe lesion revealed reactive gliosis, collections of histiocytic giant multinucleated cells, and marked perivascular lymphocytic infiltrates with a predominance of CD8 lymphocytes. No fungal elements were identified on periodic-acid-Schiffor Grocott stains; no toxoplasma pseudocysts, tachyzoites, or cytopathic changes were seen in neuronal or glial cells suggestive of infection by CMV, HSV, or JCV. A culture for mycobacterial organisms or fungi was also negative. About a month after the biopsy, even without using steroids, the patient showed significant improvement in clinical status, fulfilling the diagnostic criteria for cerebral toxoplasmosis associated with IRIS.

IRIS is a paradoxic worsening or onset of systemic inflammatory clinical signs and symptoms among patients with HIV/AIDS. It occurs after the initiation of HAART, particularly in those patients with profound immune suppression, and can occur with or without a concurrent opportunistic infection. Whether elicited by an infectious or noninfectious agent, the presence of an antigenic stimulus for development of the syndrome appears necessary. This antigenic stimulus can be intact “clinically silent” organisms or dead organisms and their residual antigens.^1–4 Conventional and advanced MR imaging techniques are useful in evaluating these phenomena. Typical MR imaging patterns of some infections can be modified during IRIS. The most frequent findings are a paradoxic increase in the size and number of lesions, perilesionaledema, and greater enhancement in T1-weighted images post–gadolinium administration. A combination of clinical, laboratory, and MR imaging findings with follow-up studies is recommended in these patients. The use of MR imaging in association with clinical and laboratory data can reduce the number of unnecessary cerebral biopsies in these clinical scenarios.

References

1. Murdoch DM, Venter WD, Van Rie A, et al. Immune reconstitution inflammatory syndrome (IRIS): review of common infectious manifestations and treatment options. AIDS Res Ther 2007;4:9[CrossRef][Medline]
2. Thurnher MM, Post MJ, Rieger A, et al. Initial and follow-up MR imaging findings in AIDS-related progressive multifocal leukoencephalopathy treated with highly active antiretroviral therapy. AJNR Am J Neuroradiol2001;22:977–84[Abstract/Free Full Text]
3. Pfeffer G, Prout A, Hooge J, et al. Biopsy-proven immune reconstitution syndrome in a patient with AIDS and cerebral toxoplasmosis. Neurology 2009;73:321–22[Free Full Text]
4. Tremont-Lukats IW, Garciarena P, Juarbe R, et al. The immune inflammatory reconstitution syndrome and central nervous system toxoplasmosis. Ann Intern Med 2009;150:656–57[Free Full Text]

Reply

Published ahead of print on June 25, 2010
doi: 10.3174/ajnr.A2163

American Journal of Neuroradiology 31:E64, August 2010
© 2010 American Society of Neuroradiology

C. Uggetti^a
aNeuroradiology Unit
Department of Radiology
San Carlo Borromeo Hospital
Milan, Italy

R. La Piana^b
bChild Neurology and Psychiatry
Niguarda Hospital
Milan, Italy

We thank Dr Ramantani and colleagues for their comments on our recent article discussing neuroradiologic findings in Aicardi-Goutières syndrome (AGS).¹

The fact that they, too, report a relatively large series² adds weight to the idea that AGS may not be as rare as once thought. We entirely agree with their view that AGS shares more features with systemic lupus erythematosus than was previously believed to be the case² and also that clinical, genetic, and neuroradiologic findings should be considered together to achieve a better understanding of the complex pathogenesis of this disease. However, as recalled by Ramantani and colleagues, our work, prompted by the lack of a detailed description of neuroimaging in AGS, focused specifically on the neuroradiologic picture of the disease.

In our sample (see On-Line Table 1 in our article¹), we found extra-neurologic involvement in 16 patients (44.4% of the sample) who presented with, among other manifestations, hypothyroidism, chilblain lesions, and celiac disease. As pointed out by Ramantani and colleagues, the prevalence of these features already highlighted in our sample, underlines the need to screen for autoimmune conditions because these can go undiagnosed. Indeed, we currently have a work in progress focusing on this very aspect. As we mentioned in the “Discussion” section of our article,¹ a precise correlation of neuroimaging with genetic and clinical data is needed to improve understanding of AGS and to interpret it as an autoimmune-mediated disorder.

We reiterate that cerebral calcifications are a key finding in the neuroradiologic picture of AGS, typically localized in the basal ganglia, in the deep white matter, and also in the posterior fossa^3–5: as stated in our article, we found calcifications in the dentate nuclei in 11 patients (30.5% of our sample).¹

Regarding the pathophysiologic mechanism responsible for the neuroradiologic picture of AGS, the microangiopathic hypothesis is generally accepted, especially in light of the studies of Barth et al.⁶ As reported in our article,¹ the distributionof cerebral calcifications in AGS, mainly in the basal ganglia and the lobar white matter, may recall the finding of calcifications located along the walls of the arterioles,⁷ thus supporting the hypothesis of a microangiopathic origin of the calcium deposition. However, the pattern of the leukodystrophy is not particularly typical of a leukoencephalopathy of inflammatory microangiopathic origin: in most of our sample, we observed a symmetric homogeneous distribution of the signal-intensity alteration rather than the patchy pattern usually found in microangiopathic autoimmune diseases.⁸ In other words, the white matter signal-intensity alterations may not be interpreted simply as microangiopathic changes, and the microangiopathic mechanism could be limited to the first stages of the disease, when the autoimmune system mediated by interferon- is still active. A more clearly defined and detailed follow-up study of a patient series, comparing the early and subsequent stages of the disease, is needed to obtain more information about the leukodystrophic process in AGS.

The lack of contrast enhancement found in our sample¹ simply confirms the integrity of the blood-brain barrier and does notnecessarily conflict with the hypothesis that AGS has a microangiopathic pathophysiology. Furthermore, in our sample, no pathologic enhancement was noted in patients who underwent neuroradiologic examinations during the acute phase of the disease,⁹ thus reinforcing the idea that the pathophysiology of AGS is probably more complex than that of a typical autoimmune-mediated disease.

References

1. Uggetti C, La Piana R, Orcesi S, et al. Aicardi- Goutières syndrome: neuroradiologic findings and follow-up. AJNR Am J Neuroradiol 2009;30:1971–76[Abstract/Free Full Text]
2. Ramantani G, Kohlhase J, Hertzberg C, et al. Expanding the phenotypic spectrum of lupus erythematosus in Aicardi-Goutières syndrome. Arthritis Rheum 2010;62:1208–12[Medline]
3. Aicardi J, Goutières F. A progressive familial encephalopathy in infancy with calcifications of the basal ganglia and chronic cerebrospinal fluid lymphocytosis. Ann Neurol 1984;15:49–54[CrossRef][Medline]
4. Goutières F, Aicardi J, Barth PG, et al. Aicardi-Goutières syndrome: an update and results of interferon-alpha studies. Ann Neurol 1998;44:900–07[CrossRef][Medline]
5. Lanzi G, Fazzi E, D’Arrigo S, et al. The natural history of Aicardi-Goutières syndrome: follow-up of 11 Italian patients. Neurology 2005;64:1621–24[Abstract/Free Full Text]
6. Barth PG, Walter A, van Gelderen I. Aicardi-Goutières syndrome: a genetic microangiopathy? Acta Neuropathol1999;98:212–16[CrossRef][Medline]
7. Barth PG.. The neuropathology of Aicardi-Goutières syndrome. Eur J Paediatr Neurol 2002;6(suppl A):A27–31[CrossRef][Medline]
8. Aviv RI, Benseler SM, Silverman ED, et al. MR imaging and angiography of primary CNS vasculitis of childhood.AJNR Am J Neuroradiol 2006;27:192–99[Abstract/Free Full Text]
9. Orcesi S, Pessagno A, Biancheri R, et al. Aicardi-Goutières syndrome presenting atypically as a sub-acute leukoencephalopathy. Eur J Paediatr Neurol 2008;12:408–11[CrossRef][Medline]