Immune Reconstitution Inflammatory Syndrome of the Brain: Case Illustrations of a Challenging Entity

By Rushing, Elisabeth J Liappis, Angelike; Smirniotopoulos, James D; Smith, Alice B; Henry, James M; Man, Yan-Gao; Nelson, Ann M

Abstract Immune reconstitution inflammatory syndrome represents a spectrum of clinicopathologic entities encountered in human immunodeficiency virus-infected patients who have received highly active anti-retroviral therapy. The diagnosis is often challenging, treatment options are limited, and the prognosis is variable. To increase awareness and define the clinicopathologic features, we present our experience with 6 probable cases involving the brain, including 1 autopsy. Clinicopathologic review was supplemented by immunohistochemical analysis. There were 5 men and 1 woman, ranging in age from 34 to 47 (mean, 41; SD, 5.39) years. All patients experienced neurologic deterioration (focal deficits in 5/6) after highly active anti-retroviral therapy. All specimens showed a predominance of CD8^sup +^ lymphocytic inflammation. Concurrent CNS infections included human immunodeficiency virus encephalitis, progressive multifocal leukoencephalopathy, cryptococcal meningitis, and syphilis. One patient died, 1 was lost to follow-up, 2 improved, and 2 showed no substantial clinical changes. Subtle and overlapping features may preclude a definitive diagnosis. To capture all suspected cases, it is important to consider the possibility of this entity. In this study, the degree of CD8^sup +^ inflammation was more pronounced in the single fatal example, and mast cells were not identified in the infiltrates. Although nonspecific, imaging findings may offer clues to early diagnosis.

Key Words: Brain, CD8, Highly active anti-retroviral therapy, Human immunodeficiency virus, Immune reconstitution inflammatory syndrome, Lymphocytes, Mast cells.

INTRODUCTION

The widespread use of highly active anti-retroviral therapy (HAART) has dramatically improved the quality of life and has increased survival of patients with human immunodeficiency virus (HIV) infection. Notwithstanding these advances, HAART can occasionally represent a “double-edged sword” because previously subclinical infections can be unmasked or preexisting untreated opportunistic infections exacerbated as the host inflammatory response is restored. This phenomenon, known both as immune reconstitution inflammatory syndrome (IRIS) and immune reconstitution disease, is defined as a paradoxical clinical decline soon after starting HAART despite appropriate reduction in HIV-RNA levels and improved CD4 T-lymphocyte cell counts (1-7).

Highly active anti-retroviral therapy-induced restoration of a pathogen-specific immune response contributes to the pathologic recognition of antigens associated with a known replicating infection or persistent nonreplicating antigens from a prior infection in IRIS (4). Although histopathologic criteria have not been codified, IRIS is often dominated by CD8^sup +^ T-lymphocyte inflammatory infiltrates. In some instances, inflammation may masquerade as a malignancy or even conceal underlying infectious agent(s), resulting in delayed diagnosis and inappropriate treatment (4, 6).

Estimates indicate that one fourth to one third of HIV-infected patients develop IRIS, underscoring the frequency of this condition. At greatest risk are patients with advanced immunodeficiency (CD4^sup +^ T-cell nadirs <100 cells/ml) at HAART initiation who are placed on potent anti-retroviral regimens such as boosted protease inhibitors (BPIs) or nonnucleoside reverse-transcriptase inhibitor and have a rapid decrease in the viral load (7-10).

The most common organisms associated with this condition are cytomegalovirus (CMV), Mycobacterium tuberculosis or Mycobacterium avium complex, and Cryptococcus neoformans (11-16). In the recent meta-analysis by D’Amico et al (17) and a study of 49 cases by Manabe et al (9), mycobacterial infections were the most common IRIS- associated infectious agents. Although less common than infectious diseases, autoimmune diseases and neoplasia have also been reported with IRIS (9, 10).

Restoration of the immune response may be systemic or localized to specific tissue sites such as the brain (18, 19). In the central nervous system (CNS), atypical presentations of progressive multifocal leukoencephalopathy (PML) in particular have been recognized after HAART initiation (17, 20-23). Somewhat less common are immune reconstitution-associated cryptococcal infections (9, 14, 24).

In most patients, the clinical outcome of immune reconstitution has been beneficial to the host, although several reports have described worsening or fatal outcomes (7, 24). Therapeutic intervention, which is variably effective, is initially directed against the specific infectious agent and may be supplemented by corticosteroids to suppress the exuberant and potentially destructive inflammatory response (9, 10, 25).

Anecdotal experience suggests that CNS IRIS may be underdiagnosed or even misdiagnosed, leading to complications that may have been preventable or reversible (5, 10, 11). Because of advanced immunosuppression, HIV-infected patients often have overlapping or concurrent diagnoses, which reinforce the importance of considering the clinical context in the interpretation of pathologic specimens. The aim of this report is to increase awareness of the protean manifestations of this entity and to highlight the diagnostic challenges as illustrated by our experience with 6 cases.

Although little is known regarding the role of mast cells in CNS infections and inflammatory conditions, recent publications indicate that they may play a role in the pathogenesis of various CNS inflammatory diseases. Mast cells have been found close to multiple sclerosis plaques, and in vitro experiments indicate that mast cells contribute to T-cell activation (26, 27). Accordingly, double labeling with mast cell markers was performed to determine whether they might participate in the T-cell-mediated inflammatory processes of IRIS.

MATERIALS AND METHODS

Clinical Data/Diagnostic Criteria

Cases were received in consultation at the Armed Forces Institute of Pathology between 2002 and 2007. Information regarding clinical presentation, neuroradiologic appearance, therapy, and outcome were extracted from patient records and supplemented by correspondence with contributors. In several instances, clinical information was limited; however, all 6 patients were infected with HIV/AIDS and had a history of HAART therapy. All radiologic data were reviewed independently by 2 neuroradiologists (A.B.S. and J.D.S.). The minimal histopathologic feature interpreted as consistent with IRIS was inflammation, with a predominance of CD8^sup +^ lymphocytes, often accompanied by macrophages and necrosis.

TABLE 1. Clinical and Pathologic Data*

Histology

Five cases were surgical specimens, and a single case was a “brain-only” autopsy specimen. The contributor’s hematoxylin and eosin slides submitted in reference to each of the cases were reviewed and assessed for general architectural features and the presence of infectious agents. In all cases, if they were not provided by the contributor, additional sections prepared from the accompanying paraffin block or unstained sections were stained with histochemical methods to detect microorganisms, including Brown- Brenn, Brown-Hopps, Grocotts-Methenamine-Silver, Warthin-Starry, and Ziehl-Neelsen stains. Luxol fast blue stains and the Bielschowsky method for nerve fibers were used to evaluate demyelination.

Immunohistochemistry

For immunohistochemical staining, sections were deparaffinized and subjected to antigen retrieval according to standard methodology. All antibodies used were from Dako (Carpinteria, CA) and were diluted in PBS as indicated. The panel of antibody reagents included HIVp24 (1:50), glial fibrillary acidic protein (polyclonal; 1:6000), lymphocyte markers (CD3 at 1:8000; CD4 at 1:8000; CDS at 1:200; CD45RB at 1:400), CD68 (1:2000) for macrophages, and the proliferation marker Ki-67 (clone Mib-1; 1:500). For mast cells, a monoclonal mouse anti-human mast cell tryptase antibody (1:100) was used. Lymph node sections were used as a positive control for mast cells.

FIGURE 1. The radiologic spectrum of immune reconstitution inflammatory syndrome. Representative images from Cases 4 to 6 are presented. Case 6 axial fluid attenuation inversion recovery (FLAIR) (A) and postcontrast T1 (B) demonstrate relatively symmetric T2 hyperintensity in the periventricular white matter. There is also subcortical white matter involvement in the left temporal lobe. On the postcontrast imaging, there is no enhancement. Case 4 axial FLAIR (C) and axial postcontrast (D) demonstrate scattered subcortical T2 hyperintensity and punctuate hyperintensity within the corona radiata. There is enhancement on the postcontrast image corresponding to the regions of T2 abnormality, including those of the corona radiata (arrow). Case 5 axial FLAIR (E) and coronal postcontrast (F) images reveal T2 hyperintensity in the left frontal periventricular and subcortical white matter. There is also abnormal signal in the left frontal and right parietal (arrow) subarachnoid spaces. On the postcontrast image, there is leptomeningeal enhancement in the right parietal region corresponding to the abnormal T2 signal. The leptomeninges in the left frontal region also demonstrated enhancement, but the white matter abnormality did not demonstrate enhancement (not shown). Double immunostaining for simultaneous elucidation of HIV and mast cells were performed using our previously published protocol (28). Briefly, the sections were first immunostained for p24 (1:10, Novocastra Laboratories, Ltd, Newcastle, UK) with a black substrate. The immunostained sections were then incubated with the mouse anti-human mast cell tryptase antibody as previously described.

In Situ Hybridization

In selected cases, a JC virus (JCV) probe was used from a 4-kb fragment of JCV cloned into the EcoR1 site of a pBR322 plasma vector (Enzo Diagnostics, New York, NY: NO. BP-847). Tissue sections with a positive JCV probe were detected by streptavidin biotinylated horseradish peroxidase complex using 3-Amino-9-ethylcarbazole, which rendered the positive cells red. Hematoxylin was used as the counterstain.

RESULTS

Clinicopathologic/Radiographic Features

Clinical information is summarized in Table 1. There were 5 men and 1 woman who ranged in age from 34 to 47 (mean, 41; SD, 5.4) years. Neuroimaging findings are illustrated in Figure 1, and details are tabulated in Table 2. None of the inflammatory infiltrates contained mast cells and therefore will not be commented on individually in the case studies that follow (Fig. 2). Representative histopathologic features from selected cases are illustrated in Figures 3 and 4.

TABLE 2. Neuroimaging Findings (MRI)

Case 1

A 47-year-old man with a 10-year history of HIV infection presented with new-onset neurologic symptoms consisting of right- sided weakness and paresthesias after receiving HAART. At therapy initiation, he was prescribed abacavir, lamivudine, zidovudine, and the nonnucleoside reverse-transcriptase inhibitor nevirapine; his CD4 count was 848 (49%) cells/mm^sup 3^, and his HIV-RNA count was 97 copies/ml. The differential diagnosis included hemorrhage into a preexisting lesion, possible astrocytoma or oligodendroglioma versus an atypical lymphoma.

Biopsy of the lesion revealed hypercellular gray and white matter with gliosis, atypical, hyperchromatic astrocytic nuclei, macrophages, and moderate inflammation. The inflammatory infiltrates were largely perivascular with a smaller infiltrating component. Microorganisms were not identified. Scattered multinucleated giant cells suggestive of HIV encephalitis were observed (Fig. 3A). Mural calcification and numerous vessels possessed thickened, hyalinized walls, suggesting a vasculopathy or possible ruptured vascular malformation (Fig. 3B). There were parenchymal and perivascular collections of lymphocytes that were predominantly CD8^sup +^ (Figs. 3C, D), with fewer CD3^sup +^ and CD4^sup +^ cells. Immunostaining with CD-68 highlighted scattered macrophages. The Ki-67 stain did not reveal significantly increased proliferative activity. The HIVp24 antigen was interpreted as nonreactive; however, questions were raised as to whether the stain was technically adequate. Tissue was not available for additional sections. The inflammatory features of the biopsy without obvious etiology were interpreted as probable IRIS.

Case 2

This 35-year-old woman presented with an altered state of consciousness accompanied by severe lower extremity weakness. She was found to be seropositive for HIV. The imaging findings were interpreted as consistent with PML, and the patient was started on didanosine, nevirapine, and indinavir. Although the time interval was not specified, the patient deteriorated after HAART initiation, and new imaging studies showed focal enhancement. Symptoms did not improve with high-dose steroids, prompting surgical intervention to obtain a tissue diagnosis.

Stereotactic biopsy revealed subtle parenchymal pallor and scattered cyst-like structures reminiscent of toxoplasmosis. Because these structures were not present in deeper sections, immunostaining for Toxoplasma was not performed. Elsewhere, a number of cells were found with enlarged nuclei and inclusion-like structures suggestive of PML (Fig. 3E) accompanied by moderate lymphocytic infiltrates. As in Case 1, the inflammation was composed of mostly CD8^sup +^ cells interspersed with occasional CD4^sup +^ and CD3^sup +^ lymphocytes. Luxol fast blue-stained sections revealed patchy parenchymal pallor (Fig. 3F). In situ hybridization was positive with the JCV probe, confirming the diagnosis of PML. Although CD4 counts or other laboratory indices were not available, these unusual biopsy features were thought to represent inflammation typical of IRIS. The patient is alive, but without follow-up details.

FIGURE 2. (A) Tryptase staining (red chromogen) of mast cells in a lymph node-positive control (400 x ). (B) Double immunostained section from Case 5 showing CD8^sup +^ cells and absence of staining for mast cells (400 x ).

Case 3

A 40-year-old Hispanic man was seropositive for HIV and had a CD4 T-cell lymphocyte count of 478 (20%) cells/[mu]l; HIV-RNA was not reported. A supraclavicular lymph node biopsy showed extensive Cryptococcus. Highly active anti-retroviral therapy was initiated with zidovudine, lamivudine, and a BPI regimen of lopinavir/ ritonavir. The clinical presentation and imaging studies suggested the possibility of bilateral intracranial neoplasia. A brain biopsy was obtained.

Hematoxylin and eosin-stained sections from the biopsy showed abundant yeast forms within the leptomeninges accompanied by moderate, predominately perivascular inflammation. Special stains (Grocotts-Methenamine-Silver) supported the diagnosis of meningitis due to C. neoformans (Fig. 3G). Immunohistochemical staining showed a largely CD8^sup +^ T-cell infiltrate (Fig. 3H) with scattered CD4^sup +^ lymphocytes. Gliosis and scattered atypical-appearing astrocytes were observed within the white matter, but inclusions were not seen. Immunolabeling for HIV p24 was nonreactive. The development of clinical meningitis accompanied by moderate, mostly CD8^sup +^ inflammation after HAART initiation was felt to represent probable IRIS.

Case 4

A 46-year-old man originally from Africa was diagnosed with AIDS when he presented with rash, headache, and a rapid plasmin reagin of 1:256. Diagnosed with secondary syphilis, he received intravenous penicillin G for 14 days; the lumbar puncture revealed an elevated protein and negative Venereal Disease Research laboratory. He initiated HAART with abacavir, lamivudine, and a BPI lopinavir/ ritonavir 2 months after discharge. The initial CD4 count was 8 (1.5%), with an HIV-RNA level of 71,000 copies/ml. The viral load was undetectable (<50 copies/ml) 6 weeks after initiation of therapy, and the CD4 count increased to 43 (3.0%) when he experienced an acute onset of word-finding difficulty accompanied by left-arm numbness. Cultures and serological and polymerase chain reaction-based evaluations of the cerebrospinal fluid for possible infectious etiologies, including syphilis, tuberculosis, toxoplasmosis, viral (CMV, herpes simplex virus, JCV), and Epstein- Barr virus-associated lymphoma, were all negative. Representative neuroimaging findings from this patient are shown in Figures 1A and B.

Persistent symptoms unresponsive to a second course of penicillin G prompted referral for a biopsy of an accessible occipital lesion. Pathologic examination revealed microglial nodule encephalitis with CD8^sup +^ lymphocytosis accompanied by a heavy, often perivascular plasma cell infiltrate. Spirochetes were not identified by special stains, but postsyphilitic IRIS was not excluded. Immunostains for p24 were reactive in rare cells. Highly active anti-retroviral therapy was continued, and the patient had a gradual resolution of symptoms over the course of 6 months and repeat magnetic resonance imaging showing improvement in the white matter lesions.

Case 5

This 34-year-old Hispanic man with a history of Kaposi sarcoma, Castleman disease, and prior reactive rapid plasmin reagin presented with headaches and left-sided neurologic deficits after HAART. Pre- HAART CD4 and viral load were not available. Highly active anti- retroviral therapy consisted of a BPI regimen, including lopinavir and ritonavir. The patient was hospitalized and empirically treated for neurosyphilis, herpes simplex encephalitis, and toxoplasmosis. Cerebrospinal fluid studies showed elevated protein (111 mg/dl) with normal glucose and lymphocytic pleocytosis. Investigations for CMV, Toxoplasma, Cryptococcus, syphilis, Epstein-Barr, and herpes simplex virus were negative as well as cerebrospinal fluid cultures for microorganisms, including M. tuberculosis. Cerebrospinal fluid cytology for malignant cells was negative. Computed tomography showed diffuse cerebral edema with progressive upward transtentorial herniation. The patient became apneic and died. Permission was obtained for postmortem examination of the brain. Representative neuroimaging findings from this patient are shown in Figures 1C, D.

FIGURE 3. Selected pathologic features of brain involvement in immune reconstitution inflammatory syndrome (IRIS). The biopsy from Case 1 showed (A) a multinucleated giant cell (arrow; 400 x ) and mural calcification. There were also foci of (B) thickened, hyalinized vessels, suggesting a vasculopathy (200 x ) and perivascular and parenchymal chronic inflammation (C; 400 x ), which was composed of CD8^sup +^ lymphocytes (D; 400 x ), a common feature in IRIS. In Case 2 (E), there was an enlarged nucleus with an inclusion-like structure suggestive of progressive multifocal leukoencephalopathy (arrow; 400 x ). A Luxol fast blue-stained section reveals patchy parenchymal pallor (F; 20Ox). Grocotts- Methenamine-Silver stain demonstrates meningitis due to Cryptococcus neoformans (C; 400 x ). Moderate perivascular CD8^sup +^ T-cell infiltrate was seen in Case 3 (H, 400^sup x ^).

Grossly, the brain was edematous and showed evidence of uncal and tonsillar herniation. The leptomeninges were opacified. sectioning revealed no focal lesions. Microscopic examination revealed a necrotizing meningoencephalitis with abundant collections of macrophages accompanied by exuberant, multifocal parenchymal and dense perivascular infiltrates of chronic inflammatory cells (Fig. 4A), prompting an initial consideration of lymphoma. The infiltrates were present throughout both cerebral hemispheres, slightly greater on the right. The leptomeningeal infiltrate was particularly prominent over the parietal lobes. There was a decreasing gradient of inflammation from the superficial cortex toward the deep white matter. With the exception of leptomeningeal mononuclear inflammation, brainstem and cerebellum were spared. No nuclear or cytoplasmic inclusion bodies were identified. Special stains for microorganisms were negative. The lymphocytic infiltrate showed scattered CD20^sup +^ cells (Fig. 4B) but was predominately CD8^sup +^ (Fig. 4C) with fewer CD3^sup +^ and CD4^sup +^ cells. Immunolabeling for HIV p24 was reactive in rare cells (Fig. 4D). These findings were interpreted as probable IRIS without detectable infection other than HIV. Case 6

This previously healthy 41-year-old Eritrean man presented with a 5-month history of fatigue, tremulousness, intermittent headache, tinnitus, vertigo, and unsteady gait. Cerebrospinal fluid obtained by lumbar puncture revealed increased protein (133 mg/dl), lymphocytic pleocytosis, negative Epstein-Barr viral titer, and HIV detected by polymerase chain reaction. At initiation of HAART, his CD4 count was 3 (7.1%), and HIV RNA was 8,600 copies/ml, and he was treated with a BPI regimen. At presentation, 5 months after the initiation of HAART, the CD4 was 230 (17.6%), and viral load was 963 copies/ml. A stereotactic biopsy was obtained, which consisted of a tiny fragment of brain tissue with mild to moderate perivascular cuffing (Fig. 4E) with mostly CD8^sup +^ (Fig. 4F) and scattered CD4^sup +^ lymphocytes. No nuclear or cytoplasmic inclusions, malignant cells, or microorganisms were identified using the routine panel of special stains. The clinicopathologic features were interpreted as HIV encephalopathy associated with IRIS. After changing the HAART regimen to agents to achieve better CNS penetration, the patient showed remarkable clinical and radiographic improvement.

FIGURE 4. Selected neuropathologic features of immune reconstitution inflammatory syndrome (IRIS). The autopsy findings in Case 5 included a dense perivascular infiltrate of chronic inflammatory cells in the parietal lobe cortex (A; 400 x ). CD20 immunostain highlights perivascular B cells (B; 400 x ). (C) The perivascular and parenchymal lymphocytic infiltrate is predominately CD8^sup +^ T cells (C; 200 x ). An immunostain for human immunodeficiency virus p24 is reactive (D; 400 x ). In Case 6, the stereotactic biopsy showed moderately increased cellularity (E; 200 x ) and a perivascular accumulation of mostly CD8^sup +^ lymphocytes (F; 400 x ).

DISCUSSION

The concept of IRIS began with the observation that certain HIV- infected patients experienced clinical deterioration after HAART therapy that most often occurred within a few months of treatment (9, 10). In comprehensive reviews, Gray et al (1) , Langford et al (2), Scaravilli et al (3), and others observed that despite changing patterns of HIV/AIDS after the introduction of HAART, CNS involvement remains a major cause of death (9). In patients with severe immunodeficiency, the incidence of toxoplasmosis, cytomegalovirus, and HIV encephalitis decreased, whereas the incidence of PML and primary lymphoma remains unchanged. When administered to patients with early stages of immunodeficiency, PML, CMV, and lymphomas decreased, whereas herpes zoster and varicella zoster virus encephalitis increased (2, 3). In addition, approximately 25% of patients were unable to tolerate the drugs, and others developed “burnt-out” lesions from treated forms of HIV encephalitis, PML, and varicella zoster virus encephalitis (3, 9).

Pathologists and radiologists commonly encounter deficiencies in clinical information, thereby making the diagnosis of IRIS especially challenging. First and foremost, entertaining the diagnosis is essential. Radiographic and pathologic features of this condition reflect the wide variety of clinical neurologic abnormalities that result from the complex interaction of inflammatory mediators and pathogens.

As might be expected, radiologic findings are diverse and frequently atypical for the presumed underlying disease process. An unusual imaging appearance in a patient with HIV/AIDS should alert the provider to inquire whether the patient received HAART. As observed by others and noted in some of our cases, the presence of mass effect or enhancement in an unexpected setting may be suggestive of IRIS (1, 4, 10, 13). Similarly, the pathologic diagnosis is subject to potential pitfalls because inflammation can be patchy and may not be apparent in small specimens. In addition to HAART, previous therapy with corticosteroids and/or anti-microbial agents can markedly alter the nature of the inflammatory reaction (8, 10). Accordingly, brain tissue from patients with HIV/AIDS should be submitted for culture, routine histopathologic analysis and subjected to special histologic and immunohistologic stains. In particular, it is important to obtain stains for pathogens such as the JCV, HIV, and C. neoformans and to determine the T-cell subset phenotype and distributions (1, 3-6, 13). Previous studies have shown that T-cell infiltrates, especially cytotoxic (CD8^sup +^) lymphocytes, predominate in IRIS (5, 26).

In the present series, clinical criteria were incompletely documented in several cases. Nevertheless, the history of HAART in concert with biopsy results provided compelling support for the diagnosis of IRIS. For example, although the amount of tissue in biopsy from the 41 -year-old Eritrean man was extremely small, the diagnosis of IRIS was supported by the presence of CD8^sup +^ lymphocytosis and the appropriate clinical context. Other examples that illustrate the critical role of early biopsy are the patient who presented with intracranial hemorrhage and the patient with a history of cryptococcosis. In both instances, the radiographic interpretation was that of intracranial neoplasia. Subsequent biopsies demonstrated CD8^sup +^ infiltrates and reactive changes without evidence of malignancy, thus altering the prognosis and clinical management.

The biopsy from the patient with secondary syphilis contained microglial nodule encephalitis, CD8^sup +^ lymphocytosis accompanied by heavy plasma cell infiltrates. Although syphilitic infection could not be excluded, spirochetes were not detected on special stains. To our knowledge, HAART-associated neurosyphilis has not been reported, but syphilitic infections of the eye have been described in patients with immune reconstitution (16, 19). Focal neurologic deficits and not dementia dominate the clinical presentation in most cases of IRIS. Given the frequency of multiple concurrent infections in HIV-infected patients, it was not surprising to observe p24 staining in 2 of our cases.

Although it is difficult to assess the degree of inflammation in biopsy material in even a semiquantitative manner, our cases and previous studies suggest that the degree of inflammation seems to be predictive of outcome (2, 3, 7). The single known fatality was a patient whose brain contained dense perivascular and diffuse parenchymal infiltrates of T8^sup +^ lymphocytes. Although initial diagnostic considerations included lymphoma and, to a lesser extent, vasculitis, lymphocytic markers facilitated the distinction. Our findings were similar to the massive perivascular and diffuse parenchymal CD8^sup +^/CD4 infiltrates found in 7 fatal cases of HAART-treated patients reported by Langford et al (2). In contrast to their cases, our case lacked diffuse myelin pallor.

Human tissue mast cells may serve as a reservoir of persistent HIV in infected individuals during HAART (26, 27, 29). In our cases, we were not able to demonstrate the participation of mast cells. Double labeling experiments did not detect mast cells as a component of the inflammation observed in any of our cases. Additional studies with larger numbers of cases are needed for confirmation of these results.

In summary, experience with these 6 cases illustrates the diverse and sometimes diagnostically challenging presentation of IRIS. Distinction from other entities should begin with vigilance in all patients with HIV/AIDS. In many cases, brain biopsy provides an important resource for identifying IRIS and diseases that mimic it.

REFERENCES

1. Gray F, Chretien F, Vallat-Decouvalaere AV, et al. The changing pattern of HIV neuropathology in the HAART era. J Neuropathol Exp Neurol 2003;62:429-40

2. Langford TD, Letendre SL, Larea GJ, et al. Changing patterns in the neuropathogenesis of HIV during the HAART era. Brain Pathol 2003; 13:195-210

3. Scaravilli F, Bazille C, Gray F. Neuropathologic contributions to understanding AIDS and the central nervous system. Brain Pathol 2007;17:197-208

4. Shelburne SA 3rd, Hamill RJ. The immune reconstitution inflammatory syndrome. AIDS Rev 2003;5:67-79

5. Shelburne SA 3rd, Visnegarwala F, Darcourt J, et al. Incidence and risk factors for immune reconstitution inflammatory syndrome during highly active antiretroviral therapy. AIDS 2005;19:399-406

6. Shelburne SA, Montes M, Hamill RJ. Immune reconstitution inflammatory syndrome: More answers, more questions. J Antimicrob Chemother 2006;57:167-70

7. Robertson J, Meier M, Wall J, et al. Immune reconstitution syndrome in HIV: Validating a case definition and identifying clinical predictors in persons initiating antiretroviral therapy. Clin Infect Dis 2006;42: 1639-46

8. Venkataramana A, Pardo CA. Immune reconstitution inflammatory syndrome in the CNS of HIV-infected patients. Neurology 2006;67: 383- 88

9. Manabe Y, Campbell JD, Sydnor E, et al. Immune reconstitution inflammatory syndrome: Risk factors and treatment implications. J Acquir Immune Defic Syndr 2007;46:456-62 10. Lipman M, Breen R. Immune reconstitution inflammatory syndrome in HIV. Curr Opin Infect Dis 2006;19:20-25

11. Manzardo C, Del Mar Ortega M, Sued O, et al. Central nervous system opportunistic infections in developed countries in the highly active antiretroviral therapy era. Neurovirology 2005;11:72-82

12. Lortholary O, Fontanet A, Memain N, et al. Cryptococcus Study Group. Incidence and risk factors of immune reconstitution inflammatory syndrome complicating HIV-associated cryptococcosis in France. AIDS 2005;19:1043-49

13. Lawn SD, Bekker LG, Miller RF. Immune reconstitution disease associated with mycobacterial infections in HIV-infected individuals receiving antiretrovirals. Lancet Infect Dis 2005;5:361-73

14. Broom J, Woods M 2nd, Allworth A. Immune reconstitution inflammatory syndrome producing atypical presentations of cryptococcal meningitis: Case report and a review of immune reconstitution-associated cryptococcal infections. Scand J Infect Dis 2006;38:219-21

15. Buckingham SJ, Haddow LJ, Shaw PJ, et al. Immune reconstitution inflammatory syndrome in HIV-infected patients with mycobacterial infections starting highly active anti-retroviral therapy. Clin Radiol 2004; 59:505-13

16. Skiest DJ, Hester LJ, Hardy RD. Cryptococcal immune reconstitution inflammatory syndrome: Report of four cases in three patients and review of the literature. J Infect 2005;51:e289-97

17. D’Amico R, Sarkar S, Yusuff J, et al. Immune reconstitution after potent antiretroviral therapy in AIDS patients with progressive multifocal leukoencephalopathy. Scand J Infect Dis 2007;39:347-50

18. Subsai K, Kanoksri S, Siwaporn C, et al. Neurological complications in AIDS patients receiving HAART: A 2-year retrospective study. J Neurol 2006;13:233-39

19. Gray F, Bazille C, Adle-Biassette H, et al. Central nervous system immune reconstitution disease in acquired immunodeficiency syndrome patients receiving highly active antiretroviral treatment. J Neurovirol 2005;11:16-22

20. Cinque P, Bossolasco F, Brambilla AM, et al. The effect of highly active antiretroviral therapy-induced immune reconstitution on development and outcome of progressive multifocal leukoencephalopathy: Study of 43 cases with review of the literature. J Neurovirol 2003;9:73-80

21. Hoffman C, Horst HA, Albrecht H, et al. Progressive multifocal leucoencephalopathy with unusual inflammatory response during antiretroviral treatment. J Neurol Neurosurg Psychiatry 2003;74: 1142-44

22. Nuttall JJ, Wilmshurst JM, Ndondo AP, et al. Progressive multifocal leukoencephalopathy after initiation of highly active antiretroviral therapy in a child with advanced human immunodeficiency virus infection: A case of immune reconstitution inflammatory syndrome. Pediatr Infect Dis J 2004;23:683-85

23. Martinez JV, Mazziotti JV, Efron ED, et al. Immune reconstitution inflammatory syndrome associated with PML in AIDS: A treatable disorder. Neurology 2006;67:1692-94

24. Ratnam I, Chiu C, Kandala NB, et al. Incidence and risk factors for immune reconstitution inflammatory syndrome in an ethnically diverse HIV type 1-infected cohort. Clin Infect Dis 2006;42:418-27

25. Riedel DJ, Pardo CA, McArthur J, et al. Therapy insight: CNS manifestations of HIV-associated immune reconstitution inflammatory syndrome. Clin Pract Neurol 2006;2:557-65

26. Sundstrom JB, Ellis JE, Hair GA, et al. Human tissue mast cells are an inducible reservoir of persistent HIV infection. Blood 2007;109: 5293-300

27. Theoharides TC, Kempuraj D, Iliopoulou BP. Mast cells, T cells, and inhibition by luteolin: implications for the pathogenesis and treatment of multiple sclerosis. Adv Exp Med Biol 2007;601:423- 30

28. Man YG, Burgar A. An antigen unmasking protocol that satisfies both immunohistochemistry and subsequent PCR amplification. Pathol Res Pract 2003;199:815-25

29. Moloney G, Branley M, Kotsiou G, et al. Syphilis presenting as scleritis in an HIV-positive man undergoing immune reconstitution. Clin Experiment Ophthalmol 2004;32:526-28

Elisabeth J. Rushing, MD, Angelike Liappis, MD, James D. Smirniotopoulos, MD, Alice B. Smith, MD, James M. Henry, MD, Yan- Gao Man, MD, PhD, and Ann M. Nelson, MD

From the Departments of Neuropathology and Ophthalmic Pathology (EJR), Radiologie Pathology (JDS), Gynecologic and Breast Pathology (Y-GM), and Environmental and Infectious Disease Sciences (AMN), Armed Forces Institute of Pathology; Department of Medicine (AL), Division of Infectious Diseases, The George Washington University Medical Center, Washington, District of Columbia; Department of Radiology and Radiological Sciences (JDS, ABS), Uniformed University of the Health Sciences, Bethesda, Maryland; and Department of Pathology (JMH), University of Texas Health Science Center, San Antonio, Texas.

Send correspondence and reprint requests to: Elisabeth J. Rushing, MD, Department of Neuropathology and Ophthalmic Pathology, Armed Forces Institute of Pathology, Washington, DC 20306-6000; E- mail: elisabeth.rushing@gmail.com

Copyright Lippincott Williams & Wilkins Aug 2008

(c) 2008 Journal of Neuropathology and Experimental Neurology. Provided by ProQuest LLC. All rights Reserved.