Retrospective Hospital-based Study of Opportunistic Neuroinfections in HIV/AIDS
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.293440
Source of Support: None, Conflict of Interest: None
Keywords: Cryptococal meningitis, highly active anti-retroviral therapy, opportunistic neuroinfection, progressive multifocal leucoencephalopath
The emergence of acquired immunodeficiency syndrome (AIDS) pandemic constitutes a great public health challenge. According to the recent available data, approximately2 million people are infected with human immunodeficiency virus (HIV) in India. AIDS-related opportunistic infections (OIs) cause significant morbidity and mortality. A host of new OIs have been added to the existing list earlier dominated by tuberculosis. Many studies on HIV have described the spectrum of OI worldwide, including in India., Tuberculosis is the most commonly reported OI among HIV-infected individuals in India., Other OIs include oral candidiasis, cerebral toxoplasmosis, cytomegalovirus retinitis, herpes zoster, and cryptococcal meningitis.,,,,,
Neurological manifestations are seen throughout the course of HIV infection from early till terminal stages. Symptomatic neurologic dysfunction develops in more than50% of the individuals infected with HIV. Neuropathological lesions are documented at autopsy in approximately 90% ofthe cases due to neurotropism of HIV and poor penetration of antiviral drugs through the blood–brain barrier.[11 Neurological manifestations, natural course, and outcome of HIV disease could be different in India from other countries because of prevailing endemic infections, poverty, lack of awareness, and inability to take highly active anti-retroviral therapy (HAART). This study was carried out to assess the spectrum of opportunistic neuroinfections in HIV-infected patients and to assess the relation of neuroinfections to HAART status and factors predicting the outcome in patients with OI.
This retrospectivestudy was conducted at a teaching hospital over a period of 14 years from 2001 to 2014. Patients with HIV infection confirmed by enzyme-linked immunosorbent assay (ELISA) were included in the study. The diagnosis of AIDS and AIDS-related OIs was made using World Health Organization (WHO) guidelines. Detailed demographic, clinical, laboratory, and radiological data were obtained from case records. Information on the laboratory investigations including hemogram, renal, and liver parameters were obtained. Findings of the chest radiogram and cerebral imaging using computerized tomography or magnetic resonance imaging were documented. Cerebrospinal fluid (CSF) analysis was performed based on clinical and radiological findings. CSF evaluation included India ink preparation, fungal culture, estimation of adenosine deaminase (ADA), Venereal Disease Research Laboratory (VDRL), toxoplasma serology, Gram's stain, and bacteriological culture. Anemia was defined when the hemoglobinlevel was less than 12 g/dL in women and less than 13 g/dL in men. Hyponatremia was considered with sodium levels less than 130mEq/L.
Patients with opportunistic neuroinfections were grouped into those having cryptococcal, tubercular, toxoplasma, and progressive multifocal leucoencephalopathy (PML). They were grouped under meningoencephalitis of undetermined etiology when the search for pathogen was negative. When a patient had more than one pathogen, they were grouped as having mixed meningitis. Their clinical presentations were divided into acute, subacute, and chronic. Status and duration of HAART were collected and patients were grouped into those on HAART and HAART-naïve. Outcome of the patients was grouped as improved, partially improved, or dead. Neurological status at discharge was stratified using modified Rankin score (mRS) as good functional status (mRS <2), moderate impairment (mRS2–4), and severe impairment as mRS 5.
Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS) software for Windows version 20. Descriptive data were tabulated, and mean, standard deviation, and median of the data were calculated. Chi-square and Fisher's exact tests were used for categorical variables. Student's t-test and analysis of variancewere used for continuous variables. For all the tests, P value of 0.05 or less was considered significant.
During the study period, 277 patients with HIV and AIDS with neurological manifestations were treated. Among them, 167 patients (60.28%) had meningoencephalitis, of whom 131 were men (78.44%) and 36 were women (21.56%) with a male-to-female ratio of 3.63:1. Their age ranged from 16 to 67 years (39.25 ± 8.85 years). The age of men (39.47 ± 8.24 years) and women (38.44 ± 10.87 years) was similar. The etiological factors detected were isolated cryptococcal meningitis in 67 patients, tuberculosis in 32, toxoplasmosis in 10, and PML in 16. In 18 patients, causative pathogen could not be identified. About 24 patients had concurrent infection with two pathogens [Table 1].
A total of 68 patients had been diagnosed to have HIV infection earlier while HIV status was detected during the admission for OI as AIDS-defining illness in 99 patients (59.3%). Among these 99 patients detected to have HIV infection, 40 had cryptococcal meningitis (11 with coexistent tubercular meningitis), 20 had isolated tubercular meningitis, 16 had PML, 8 had toxoplasmosis, and 15 had meningoencephalitis of undetermined etiology. Among the 68 patients with known HIV infection, 38 were on HAART and a total of 129 patients were HAART-naïve. Clinical presentation was acute in 97 (58.1%) patients, subacute in 53 (31.7%), and chronic in 17 patients [Table 1]. Clinical features of the patients with different etiologies of meningoencephalitis are shown in [Table 1]. Headache was the most common symptom (85.8%), followed by fever (65.7%), altered sensorium (37.1%), and seizures (25.1%). Sixteen patients had hemiparesis and 44 had cranial nerve deficits [Table 1].
Localized abnormalities in cerebral imaging were seen in 49 patients (29.3%) consisting of granulomas in 21 patients (12.5%), PML in 16 (9.58%), infarctions in 9 (5.3%) patients, and features of acute disseminated encephalomyelitis (ADEM) in 3 patients. In all, 28 (16%) patients had hyponatremia and 108 had anemia (64.67%). Renal dysfunction during hospital stay occurred in 19 patients (11.37%) and 11 (6.58%) had hepatic dysfunction.
Fifteen patients died during admission (8.98%). Mortality was the highest in patients with mixed meningitis (16.7%) followed by meningoencephalitis of undetermined cause (11.1%), cryptococcal meningitis (10.4%), and neurotuberculosis (6.2%). Hospital stay was shorter in patients who died (9.53 ± 9.14 days; median 5 days) in comparison to those who recovered (14.34 ± 12.03 days; median 10 days) and those who partially improved (17.76 ± 20.03 days; median 9 days).
Cryptococcal meningitis was the most common cause of meningoencephalitis in 88 patients (52.7%). Among these patients, 67 had isolated cryptococcal meningitis in whom the presentation was subacute to chronic in 76% [Table 1]. About 38 of the 67 patients with cryptococcal meningitis had been diagnosed to have HIV seropositivity earlier, among whom only 11 (28.9%) were on HAART [Table 2]. Two patients had cerebral infarctions and the remaining 65 patients had normal cranial imaging. Ten patients developed transient azotemia during treatment with amphotericin, three had seizures, and one had hepatic dysfunction. In all, 52 (77.6%) improved with treatment, 8 (11.9%) partially improved, and 7 died while on treatment.
Among the 56 patients with neurotuberculosis, 32 patients had only tubercular meningoencephalitis, of whom 28 had acute presentation and 4 had sub-acute presentation [Table 1]. Fifteen of the 32 patients with tuberculosis were on HAART. About 13 patients had tuberculoma, 6 had arteritis, 1 patient had ADEM, and 12 had normal cerebral imaging. Five patients each had seizures and hepatic dysfunctions. A total of 23 (71.8%) recovered with treatment, 7 (21.9%) partially improved, while 2 died during hospital stay.
Among 24 patients with mixed meningitis, 21 patients had tubercular and cryptococcal meningitis and 3 patients had tubercular meningitis with bacterial meningitis. Thirteen had known retroviral infection, while 11 were newly detected. Six patients were on HAART and 18 were drug-naïve [Table 2]. Hyponatremia was present in eight patients. Twenty improved with treatment and four died. All the 10 patients with cerebral toxoplasmosis had acute presentation and HIV status was identified after admission in 8. Seven patients improved completely and three partially improved with treatment.
Among patients with known HIV infection, subacute presentation was common (45.6%), while in the newly diagnosed cases acute presentation was seen in majority (P = 0.0004). Cryptococcal meningitis was the most commonOI in patients with known HIV infections (P < 0.0001; [Table 2]). Among the patients on HAART, tuberculosis was more frequent, whereas cryptococcal meningitis was more frequent in HAART-naïve patients (P = 0.004). All the patients with PML were newly diagnosed to have HIV/AIDS. There was no significant difference in the outcome in relationship to HAART therapy and nature of HIV illness [Table 2].
There is a gradual decline in overall incidence and prevalence of HIV infection and occurrence of opportunistic neuroinfections due to the availability of HAART and increasing awareness of AIDS. Developing countries continue to havepersons with HIV infection and there is paucity of recent studies on opportunistic neuroinfections. The prevalence of HIV infection in India is estimated to be 0.26% amounting to about 2.1 million HIV patients makingIndia the third largest country for HIV infection. Due to the decline in immunity and fall in CD4 cells, patients in AIDS are prone to several opportunistic systemic and nervous system infections. The occurrence of neurological manifestation is reported to be about 16%–70%., In this study, opportunistic neuroinfection occurred in 60%.
Central nervous system (CNS) could be involved primarily due to direct viral invasion or due to opportunistic neuroinfections. Opportunistic neuroinfections can occur at any stage of disease including cryptococcal meningitis, tuberculosis, syphilis, toxoplasmosis, JC virus infections, amoebic meningoencephalitis, and cytomegalovirus infections. PML occurs during the late stages of disease. CNS lymphoma, cerebrovascular accident, peripheral neuropathy, and AIDS dementia complex are other neurological complications. Tuberculosis is the most common systemic OI occurring in AIDS in India. The most common neuroinfections noted in this study are cryptococcal meningitis andneurotuberculosis similar to a study from Indian subcontinent which reported cryptococcal meningitis in 46% and neurotuberculosis in 30%. The most common type of mixed meningitis wasconcurrenttuberculous and cryptococcal meningitis. This study reports 59.3% of neurological OI occurring as AIDS-defining illness in patients with neuro-AIDS, with cryptococcal meningitis being the most common. There is a variable prevalence of systemic and nervous system OI from different countries with pneumocystis carini being the most common in the West. About a tenth of our patients with meningoencephalitis had no organism detected in CSF. This group would have probably been smaller if more sensitive microbiological assays were available.
The majority of the studies report younger age and male predominance in relation to neuroinfections as observed in this study. Opportunistic neuroinfections can present as acute, subacute, or chronic forms of meningoencephalitis or as cerebral granulomas. Tuberculosis and cryptococcal meningitis present as subacute to chronic meningitis. In this study, we found that acute presentation was more common in tuberculosis and subacute to chronic presentation in patients with cryptococcal meningitis. Headache and fever were the most common symptoms in all neuroinfections and tended to occur more frequently in patients with cryptococcal meningitis and toxoplasmosis. Seizures and hemiparesis were common in tuberculosis compared with other groups similar to a previous study.
Cerebral imagingin HIV patients revealsfeaturesof cerebral edema, exudates and granulomata, symmetrical or asymmetrical white matter hyperintensities, infarction, cerebral atrophy, and space occupying lesions. Contrast-enhanced cerebral imaging is essential in the diagnosis of OIs. In this study, the most common imaging abnormality revealed granulomas in 21, PML in 16, and cerebral infractions in 9 patients.
In our study, we observed tuberculosis to bemore frequentin patients on HAART and cryptococcal meningitis more common in HAART-naïve patients. The majority of our patients were HAART-naïve (77.24%). This is attributed to lack of awareness, low socioeconomic status, and social stigma of the disease in our country.
PML is an uncommon neuroinfection and is caused by JC virus which affects oligodendroglial cells and is characterized by cognitive deficits, language abnormality, gait ataxia, sensorimotor deficits, and rarely seizures. Western literatures have described the presence of PML in up to 5% cases. A recent report reported an incidence of PML of 1.2% in HIV-positive patients followed up in HAART clinic. PML was the AIDS-defining illness in 10 of their 18 patients. PML occurred as AIDS-defining illness in 16 patients in our study. Higher incidence of PML in our cohort is probably due to patients with neurological consequence secondary to HIV/AIDS being included in this study.
Studies from the West report higher mortality in patients with tuberculosis. In our study, mortality was higher in cryptococcal meningitis group, similar to previousIndian studies., This is probably due to delay in diagnosis, severity of infection, and drug intolerance. Patients with tuberculosis meningitis had better outcome and no death occurred due to toxoplasmosis or PML during hospital stay. Due to retrospective nature of the study, complete data were unavailable regarding the disease burden including CD4 count and viral load. In addition, in some patients, causative organism could not be identified. Prospective studies evaluating disease severity and use of sensitive microbiological assays with long-term follow-up would address these issues.
We report the clinical spectrum of opportunistic neuroinfections in a large cohort of patients with HIV/AIDS. Cryptococcal meningitis was the most common AIDS-defining illness in these patients. Tuberculosis was more frequent in patients on HAART and these patients had better prognosis. In-hospital mortality was highest among patients with cryptococcal and mixed meningitis.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2]