Lm. Andersson et al., Higher HIV-1 RNA cutoff level required in cerebrospinal fluid than in blood to predict positive HIV-1 isolation, J MED VIROL, 62(1), 2000, pp. 9-13
HIV-1 can be isolated from the vast majority of blood samples taken from HI
V-1-seropositive patients not treated with antiretroviral drugs. Isolation
rates from cerebrospinal fluid (CSF) samples are considerably lower, rangin
g between 20-70%. The objective of this study was to determine the cutoff l
evels for HIV-I RNA that would yield a positive predictive value greater th
an or equal to 90% for positive virus isolation from CSF and blood. Quantit
ative HIV-1 RNA PCR (Amplicor HIV monitor, version 1.0, Roche Diagnostic Sy
stems) and virus isolation were used to examine 303 CSF samples and 278 pai
red blood samples from 157 HIV-1-seropositive patients. Patients on antiret
roviral treatment provided 140 of the CSF samples and 131 of the blood samp
les. CSF samples that were positive by culture numbered 137 of 303 (45%), a
s compared with 216 of 278 (78%) blood samples. In the case of samples take
n from patients with antiretroviral treatment, 28% were positive by culture
from CSF and 63% from blood. As expected, mean HIV-1 RNA levels were highe
r in CSF and blood samples positive by culture than in samples negative by
culture. A cutoff level of >5,000 HIV-1 RNA copies/ml was required to yield
a positive predictive value for positive virus isolation from CSF samples
of greater than or equal to 90%, whereas the cutoff level for blood samples
was just above the detection limit of the assay (>200 HIV-1 copies/ mi). J
. Med. Virol. 62:9-13, 2000. (C) 2000 Wiley-Liss, Inc.