Molecular diagnosis of central nervous system opportunistic infections and mortality in HIV-infected adults in Central China

This hospital-based study was conducted at Zhongnan Hospital of Wuhan University, China, between 1 December 2012 and 30 September 2014. Patients were eligible for the study if they had: (1) confirmed with HIV infection; (2) three or more symptoms of meningitis and/or encephalitis (headache, seizure, nausea/vomiting, alteration in consciousness, photophobia, or a focal neurological deficit); and (3) agreed to do a lumbar puncture(LP), as determined by the attending physician. Exclusion criteria for enrollment included: (1) patients with delirium, such as sepsis due to a non-neurological infection or metabolic abnormality; (2) patients with peripheral neuropathy and psychosis rather than medical explanation for their symptoms; or (3) patients with a documented immunosuppressive except for HIV infection or neurosurgical illness. Meningitis was diagnosed if there was a combination of meningeal irritation symptoms and signs and the presence of more than five white blood cells in CSF obtained by lumbar puncture. Exclusion criteria included malignancies, various inflammatory conditions, sarcoidosis, and infectious secondary to procedures. Meningoencephalitis was defined as any brain parenchymal involvement (e.g. seizures), behavioral change, confusion, dysphasia and other cortical deficits, with signs of meningeal irritation. Encephalitis was defined as fever with altered sensorium with CSF pleocytosis and raised protein with or without focal neurological signs in whom malaria, bacterial or fungal meningitis were excluded. Trained radiologists and neurologists were also invited to help identifying patients with a suspected diagnosis of meningitis, encephalitis or meningoencephalitis. The study was approved by Zhongnan Hospital of Wuhan University Ethics Committee.

Demographic information, such as age, gender, marital status, risk factors, duration between HIV infection and hospital admission, clinical features and past medical history were collected on a standardized report form by the trained physician. In China, lumbar puncture (LP) is routine practice for patients with a suspected CNS infection prior to antibiotic treatment. Other important medical assessment were computed tomography (CT) or magnetic resonance imaging (MRI) for patients with a focal brain lesion or suspected brain hemorrhage. But whether the imaging examination ultimately used depends on the patients’ or their family members’ willingness and attitude.

All CSF samples were collected via lumbar puncture and stored at −20 °C for molecular diagnostics. DNA was isolated by using the easyMAG Instrument (bioMérieux). Primers amplifying a conserved sequence of viral DNA polymerase (BALF5) gene and a fluorogenic probe were used at a primer concentration of 0.9 μM and a probe concentration of 0.25 μM. Amplification was carried out in an ABI PRISM 7900 HT Sequence Detector (PE Applied Biosystems) and the standard curve was created with Sequence Detection System software by plotting the CT values against known CMV, EBV, JCV, VZV, HHV-6, HSV or TB-DNA concentrations. The assay has a detection limit of 500 genome equivalents/ml.

In this study, 57 patients with CNS diseases out of 658 hospitalized AIDS (≥14 years old) were admitted to Zhongnan Hospital of Wuhan University with a complaint of new or recurrent neurological or psychiatric symptoms were included (Fig. 1). Of these 57 patients, complete clinical information was available in 54 (94.7%). Forty-one cases of meningitis, six cases of encephalitis and seven cases of meningoencephalitis were found in this series. Characteristics of the 54 patients are shown in Table 1. The mean age was 38 years and 64.8% of participants were male. The main occupational population was farmer (70.4%). And the main HIV acquisition route was sexual contact (79.6%). At the time of enrollment, a median duration of 45 days was observed. The median CD4⁺ T cell count was 31 cells/μl.

Pathogens were detected in 47 (87.0%) CSF samples. The number of patients diagnosed with one or more pathogens is shown in Table 2. Cryptococcus neoformans was the most common pathogen found (63.0%) and the most common copathogen (22.2%). There were high rates of detection of CMV (22.2%) and tuberculosis (13.0%). We also detected VZV (3.7%), JCV (1.9%), EBV (1.9%) and HHV-6 (1.9%) in this patient population. No case of HSV were detected by PCR.

Of the 34 cases with cryptococcus, three were negative on India Ink (8.8%), and four were negative on CrAg (11.8%).

Of five PCR diagnosed toxoplasmosis in this series, all responded clinically and radiographically within four weeks of initiation of pyrimethamine and clindamycin combination therapy.

Among the 12 CMV-DNA positive patients, 8 (66.7%) serum samples were negative for CMV-DNA and the average level of CMV-DNA in CSF was 2.80 ± 1.15 Log₁₀copies/ml.

Among five patients diagnosed toxoplasmosis, brain CT or MRI was used for four patients. The representative ring enhancing lesions were found in all the four patients.

For the EBV-DNA positive patient, mass lesions were not found. For the JCV-DNA positive patient, typical white matter changes were found, which was consistent with the PCR for JCV.

Among 12 CMV-DNA positive patients, brain CT or MRI was used for 6 patients and none was found with brain parenchymal lesion.

In our study population, the total mortality rate was 22.2% (12/54). Table 3 summarizes the potential risk factors of death in our study population. The odds of death together with 95% confidence intervals were reported for each covariate. In the multivariate analysis, pathogen-based treatment and high glasgow coma scale (GCS) score were associated with lower odds of death. Conversely, patients with multiple pathogens infection, seizures or intracranial hypertension showed higher odds of death.