Resistance to antiretroviral drugs in human immunodeficiency virus (HIV) disease may be discordant between blood and cerebrospinal fluid (CSF). However, there is no method by which patients who are likely to harbor resistant HIV in the CSF can be noninvasively identified. Activated monocytes are known to traffic through the brain and perivascular microglia are considered to "turnover" regularly from bone marrow-derived monocytes. Monocytes lack certain kinases necessary to metabolize some antiretroviral drugs, making it possible that monocytes, could deliver antiretroviral drugs to the brain. Low monocyte counts in the peripheral blood, however, would be expected to lead to decreased trafficking and turnover of monocytes, with less drug delivery to the brain. The authors hypothesized that this would increase the likelihood of drug resistant HIV in the central nervous system. To test this, 24 matching CSF and plasma samples that had been prospectively collected and stored from patients treated with nucleoside analogue reverse transcriptase inhibitor drugs were assessed for genotypic resistance. Those CSFs, with evidence of resistance mutations, were compared to those without for peripheral blood monocyte count, hemoglobin, CD4 cell count, and zidovudine (ZDV) use. The same analyses were repeated on the plasma samples. There were 11 CSFs with evidence of resistance mutations. The peripheral blood monocyte count was significantly lower in the CSF resistant group (0.29 +/- 0.16) versus (0.52 +/- 0.21) x 10(9)/L (P <.001). There was no difference between the groups according to hemoglobin, CD4 cell count, total white cell count, or use of ZDV. There was no difference between resistant and sensitive plasma samples according to peripheral blood monocyte count. To further test the hypothesis, the authors determined the concentrations of ZDV, stavudine, and abacavir in monocytes after each drug had been added to monocyte cultures. There was a significant decline in the concentration of each drug in the supernatant, implying that it had been "taken up" by the monocytes. These preliminary data suggest that peripheral blood monocytes may be important in delivery of antiretroviral drugs to the brain and the development of resistance.