Interleukin (IL)-8 concentrations were analyzed in 70 cerebrospinal fluid (CSF) samples from patients with meningitis of different etiologies and in 34 normal CSF samples. Patient groups included those with pyogenic meningitis, viral meningitis, self-resolving aseptic meningitis without a specific diagnosis, and meningitis of other etiologies and normal CSF from patients with and without neurologic disease. All samples from patients with pyogenic meningitis (18) but only 3 from patients with meningitis of other etiologies and with CSF polymorphonuclear leukocyte (PMNL) counts > or = 80% had IL-8 levels > or = 2.5 ng/mL. IL-8 was above the normal level (< or = 0.5 ng/mL) in samples from 5 of 13 viral and 8 of 23 self-resolving aseptic meningitis patients and in 7 of 13 samples from patients with meningitis caused by other microorganisms. There was a significant relationship between IL-8 levels and CSF PMNL counts in patients with nonpyogenic meningitis. The data suggest a possible role of IL-8 as PMNL chemotactic factor in different infections of the subarachnoid space, not only in pyogenic meningitis.

We assayed tumor necrosis factor-alpha (TNF-alpha) and interleukin-1 beta (IL-1 beta) concentrations in CSF from patients with meningitis of different etiologies and tested the usefulness of these assays for differentiating between pyogenic meningitis and aseptic meningitis of different etiologies. We used a monoclonal-antibody ELISA to test 125 CSF specimens from patients whose cases were classified as follows: pyogenic meningitis (n = 20), viral meningitis (n = 22), self-resolving aseptic meningitis for which no specific diagnosis was made (n = 25), meningitis due to other infectious agents (n = 11), and neoplastic meningitis (n = 5); we also tested normal CSF from healthy patients (n = 20) and those with neurological diseases (n = 22). Levels of TNF-alpha were above 200 pg/mL in 16 of 20 patients with pyogenic meningitis, but not in patients in the other groups. Levels of IL-1 beta were above 100 pg/mL in 15 of 20 patients with pyogenic meningitis and in one patient with a brain abscess. A positive correlation between levels of these cytokines and different inflammatory parameters was noted, whereas an inverse relationship with the duration of symptoms was observed. With regard to diagnosis, measurement of TNF-alpha and IL-1 beta levels showed sensitivities of 84.2% and 78.9%, respectively; specificities of 100% and 99%, respectively; a positive predictive value of 1 and 0.93, respectively; and a negative predictive value of 0.97 and 0.96, respectively.

We assayed 229 CSF samples from 180 adults with meningitis of different etiologies for adenosine deaminase activity (ADA) and evaluated the usefulness of this assay in the differential diagnosis of aseptic meningitis. Cases of meningitis were classified as tuberculous meningitis (TBM), pyogenic meningitis, viral meningitis, self-resolving aseptic meningitis without a specific diagnosis, meningitis associated with other infections, and neoplastic meningitis. We also tested 117 CSF specimens for which parameters were normal. We chose a cutoff point of 10 IU/L on the basis of our results and found elevated ADA levels in 50% of the patients with TBM (no differences between patients with AIDS and those who did not have AIDS were observed). Among samples from patients with aseptic meningitis, we observed high ADA levels in only two of five of the patients with neurobrucellosis. Therefore, we concluded that in cases of aseptic meningitis, a CSF ADA level of > or = 10 IU/L has a sensitivity of 48%, a specificity of 100%, a positive predictive value of 1, and a negative predictive value of 0.91 as a diagnostic criterion for TBM or neurobrucellosis. ADA levels were also > 10 IU/L in 30% of the patients with pyogenic meningitis, but this diagnosis was easily excluded on other grounds.

The prevalence of osteopenia in HIV-infected patients is high. However, the mechanisms implicated in bone mass loss in HIV infection are unclear. Because of this, we analyzed serum free testosterone and vitamin D3 hydroxylated metabolites in HIV-infected patients, with and without antiretroviral treatment, and the relation between them and osteopenia. Seventy-four HIV-infected patients were selected because they had frozen sera available at a date close to a DEXA evaluation. Free testosterone, 25(OH)D3, and 1,25(OH)2D3 were determined in frozen serum. There were no differences in free testosterone, 25(OH)D3, and 1,25(OH)2D3 levels between patients with and without osteopenia. 25(OH)D3 levels in naive and HAART-treated patients were 26.2 (10.3-32.8) and 33.1 (20.6-46.8) ng/ml, respectively (p = 0.04). 1,25(OH)2D3 levels in naive and HAART treated patients were 60.3 (49.2-80.8) and 85.5 (68-111.6) pmol/liter (p = 0.01). Free testosterone levels in 9 naive men and in 50 HAART-treated men were 42.6 (24.1-67.3) and 69.2 (47.5-112.1) pmol/liter, respectively (p = 0.04). In conclusion, HIV-infected patients with and without osteopenia showed similar levels of vitamin D metabolites and free testosterone. However, antiretroviral drug-naive patients showed lower serum levels of vitamin D metabolites and free testosterone than HAART-treated patients.