BK virus associated meningoencephalitis in an AIDS patient treated with HAART

Neurological complications associated with HIV-1/AIDS are being recognized with a high frequency that parallels the increased number of AIDS cases [1]. Since the introduction of HAART, morbidity and mortality secondary to primary and secondary neurological opportunistic diseases in HIV-1/AIDS patients have significantly decreased [2, 3]. However, neurocognitive impairments continue to occur in high frequencies, even in countries, with a free and universal access program to HAART, such as Brazil [4].

Recently, atypical presentations of brain diseases, such as JC virus granule cell neuronopathy, and those related to the reconstitution of the immune system after initiation of HAART have been reported in a growing basis [5–7]. These unusual neurological pictures in AIDS patients represent new diagnostic and therapeutic challenges [7, 8].

BKV meningoencephalitis is a rare polyomaviral infection with fatal outcome when associated with AIDS [9–11]. The clinical picture is devastating, resulting in death from multi-organ failure [12]. Here, we describe the clinical course of an AIDS patient with presumed BKV meningoencephalitis who showed substantial improvement after modification of his HAART regime.

In April 2004, a 43-year-old HIV-1 positive heterosexual male was admitted to the hospital complaining of bilateral headache of moderate intensity accompanied by speech, gait and memory disturbances. HIV-1 infection was diagnosed in July 2003 after an episode of cryptococcal meningitis; with a documented CD4⁺ lymphocyte count of 6 cells/mm³. His past medical history includes an episode of pancreatitis secondary to lopinavir/ritonavir (July, 2003), and an episode of tuberculous meningitis (October, 2003). On admission, antiretroviral medications included zidovudine, lamivudine, and efavirenz (antiretroviral scheme modified after pancreatitis in August, 2003). Other medications included: fluconazole, TMP-SMX, isoniazid, ofloxacin, ethambutol, and pyrazinamide (rifampin was discontinued and replaced by ethambutol and ofloxacin after a marked increased of hepatic enzymes in November, 2003)

Initial examination revealed mental confusion, dysarthria and ataxia. Baseline cranial CT scan showed two hypodense lesions with mass effect and no contrast enhancement in the left temporo-parietal and right occipito-parietal areas. His CD4⁺ lymphocyte count was 37 cells/mm^3, with an undetectable viral load. Presumptive cerebral toxoplasmosis was diagnosed, and a treatment with sulphadiazine-pyrimetamine, folinic acid, and dexamethasone was started. After 14 days of therapy, the patient's neurological status and CT scan findings remained unchanged. Analysis of the cerebrospinal fluid (CSF) obtained on day 14 after admission showed 12 leukocytes/mm³ (79% lymphocytes, 10% monocytes), glucose of 46 mg/dl and a protein level of 146 mg/dl. Herpes simplex encephalitis was suspected, and acyclovir replaced the medications for toxoplasmosis. After a week on this therapeutic regime, the patient's neurological status remained unaltered, and PCR analysis of CSF was performed, resulting negative for all human herpes viruses, JCV, and for Toxoplasma gondii. Nevertheless, PCR for polyomavirus BKV resulted positive.

From this data, acyclovir and dexamethasone were discontinued, and the patient underwent MRI-guided stereotactic brain biopsy of the lesion found in the right occipital lobe. Intraoperative MRI findings were similar to those previously seen by CT scan (Figure 1A). Histopathological examination showed thickened leptomeninges with a lymphocytic infiltrate that extended perivascularly (Figure 2A). The underlying cortex showed mild astrocytosis with prominent hypertrophic nuclei and bi-nucleated forms (Figure 2A). PCR examination of the brain tissue sample also showed the presence of BKV DNA in the absence of any other polyomaviruses (Figure 2C and 2D). The patient was reclassified as with presumptive BKV subacute meningoencephalitis. His therapeutic regime included the replacement of efavirenz for atazanavir-ritonavir along with the administration of zidovudine/lamivudine, leaving the patient exclusively with HAART. Urine analysis and renal function tests were performed, showing <10 leucocytes/field, undetectable red blood cells and proteins, a serum creatinine of 0.8 mg/dl, and a BUN of 16 mg/dl. These values, discarded the presence of renal and urinary abnormalities that are normally present in the context of BKV clinical infection. Four weeks after admission, neurological manifestations improved considerably, and the patient was discharged.

Five months after discharge, the patient complained of mild headaches. A new MRI reported improvement of the two lesions previously seen; however, new areas of increased signal intensity of the white matter appeared (Figure 1B). Based on a rising CD4⁺ lymphocyte count of 144 cells/mm³, and in the absence of any other co-infectious agent, "encephalopathy of unknown origin" secondary to immune reconstitution was suspected. We kept close follow-up without any other intervention, and the patient gradually recovered.

Seven months after discharge, MRI lesions and CD4⁺ lymphocyte count further improved (Figure 1C.). Since then, and until May 2007, the patient has been in the same HAART regime, with no neurological complaints; his renal and liver function have always been unaltered, and his last CD4 ⁺ lymphocyte count was of 336 cells/mm³ with a viral load bellow 400 copies/ml.

We report the first in-vivo case of BKV-associated meningoencephalitis in an AIDS patient who showed clinical improvement and regression of brain lesions while on HAART.

BKV infection is generally a benign condition in immunocompetent patients, and the causative agent of "polyomavirus associated nephropathy" and hemorrhagic cystitis in immunocompromised individuals [12, 14]. However, there is a growing body of evidence demonstrating its neurotropism [14–16]. Among AIDS patients, only 3 cases of BKV- meningoencephalitis have been described, and all of them reported post-mortem [9–12].

Initially, the morphology of the lesions and the cytochemical profile of the CSF samples led us to empirically diagnose the patient with cerebral toxoplasmosis and herpes simplex encephalitis afterwards. However, the lack of response to treatment and the presence of BKV DNA in CSF samples and brain tissue led us to reclassify the patient as with BKV meningoencephalitis.

Compared to the other 3 cases reported, the presence of BKV DNA in the CSF is a common denominator of true neurological disease. As previously described, the detection of BKV genome in brain samples is unspecific, as it is normally present in 3%–6% of HIV-1 infected patient, even without neurological symptoms [17]. Our patient's histopathological description is identical to the ones previously reported, showing diffuse areas of increased signal [9, 11]. Progressive multifocal leukoencephalopathy was ruled out because of the normal morphology of oligodendrocytes. In terms of the radiological changes, our case is similar to the previous ones, showing areas of increased signal intensity of the periventricular white matter (MRI) [11]. Other descriptions include increased meningeal contrast enhancement along with increased meningeal thickness (MRI) [9], and marked internal hydrocephalus and periventricular lucencies (CT scan) [10]. None of the previous reports showed features suggesting immune reconstitution inflammatory syndrome (IRIS).

Our main limitation to conclude that BKV is the causative agent of the neurological disorders is the lack of demonstration of the virus in the brain tissue sample, either by immunohistochemistry or by in-situ hybridization. Tissue samples extracted by stereotactic surgery are limited, and most of the times insufficient to run all diagnostic tests. These limitations were not faced in other reported cases, as all of these were done post-mortem.

Another important difference is the lack of renal/systemic involvement in our patient, when compared to the other cases previously reported. Our patient' renal work up included serial urianalysis and renal function tests that remained normal throughout the course of the disease (creatinine 0.8 mg/dl, BUN of 16 mg/dl, in April, 2004, creatinine of 0.9 mg/dl, BUN of 28 mg/dl in October, 2004,). As an outpatient his renal function has continued in the normal ranges (creatinine of 0.9 mg/dl, BUN of 40 mg/dl, May 2007). The containment of the disease was possibly related to a partial effect of HAART, similar to what has been reported for other AIDS-associated neurological complications in the post-HAART era, opening a broader spectrum of AIDS-associated neurocognitive disorders [18].

In this case, the change of atazanavir-ritonavir for efavirenz might have helped to reactivate the immune system, improving the symptoms and morphology of the lesions associated with BKV meningoencephalitis. The rationale of this change was based upon our previous experience and the results of an unpublished work describing the benefits of protease inhibitors in the restoration of the CD4 count when compared to non-nucleoside analogs (Riddler SA, Haubrich R, DiRienzo G, et al. A prospective, randomized, phase III trial of NRTI-, PI-, and NNRTI-sparing regimens for initial treatment of HIV-1 infection: ACTG 5142. Program and abstracts of the XVI International AIDS Conference; August 13–18, 2006; Toronto, Canada. Abstract THLB0204). Unfortunately, the restoration of the immune system is not innocuous, and as previously described, deleterious effects are found in the context of polyomavirus infection, introduction of HAART, and a rising CD4⁺ count [19]. Immune reconstitution inflammatory syndrome (IRIS), as the latter phenomenon is known, affects 15% to 45% of patients receiving HAART. CNS involvement hás been reported to occur in the presence of tuberculosis (33%), cryptococcosis (4.2%–15.9%), JC vírus (unknown frequency); being manifested radiologically as the extension or worsening of a previous condition or by the apparison of lesions that are enhanced by contrast, with or without involvement of the meninges and sometimes accompanied by hydrocephalus [6]. In our patient, the newly developed cerebral lesions and the headaches spontaneously resolved.

In conclusion, our report demonstrates that BKV might be the causative agent of meningoencephalitis in AIDS patients, and that this virus must be investigated in higher frequencies in HIV-1 positive patients with neurological manifestations now that its neurotropism has been better documented. In our patient, a previously reported life-threatening and disseminated disease progressed to a mild and focalized neurological condition in the presence of HAART.