We cannot draw conclusions in this regard based on our results because of the elevated percentage of samples in which IL-6 plasma levels were under the limit of detection, as has been seen in other studies [10, 15]. Lipid disturbances have also been investigated in relation to the increased cardiovascular risk in patients undergoing cART interruption, although the results are somewhat contradictory. KU57788 Chronic infection, including that produced by HIV, is associated with changes in lipoprotein metabolism. This can lead to proatherogenic dyslipidaemia, especially hypertriglyceridaemia, and decreased HDL-c and LDL-c,

associated with changes in the properties of lipids, rendering them more proatherogenic [23]. In accordance with previous cART interruption studies, we found a decrease in total-c and LDL-c, but also in HDL-c [4-6]. As a result, see more in our study no change in the total-c/HDL-c ratio in patients discontinuing cART was found, in contrast to the SMART study in which an unfavourable change was observed [4]. As far as we know, this is the first study in which patients were treated mainly with NNRTIs, and our data are, at least in part, consistent with

those of the SMART study, in which the strongest HDL-c reduction was found in patients receiving NNRTIs [4]. As has been described, we observed a strong negative correlation between viral load and lipid measurements, supporting a role for HIV in these variables [6]. An interesting finding of our study, described previously in a non-HIV-infected [24] and HIV-infected population [25], is the association between lipid parameters,

especially HDL-c, and MCP-1 and sVCAM, confirmed in the multivariate Tyrosine-protein kinase BLK analysis and maintained over the lengthy follow-up period. Experiments with inflammatory lipopolysaccharide-induced animal models have shown that treatment with ApoA-I, the major component of HDL-c, induces a decrease in MCP-1 and sICAM-1. ApoA-I has modulating effects on MCP-1 expression [26]. Furthermore, it is known that the antioxidant effect occurring through paraoxonase-1, an enzyme contained in HDL-c, inhibits MCP-1 synthesis by endothelial cells [27]. It is likely that the anti-inflammatory effects of HDL-c are attenuated in untreated HIV infection. The negative correlation found between HDL-c and endothelial biomarkers is consistent with the results of studies pointing to a close association between lipids and inflammation pathways, probably mediated by HIV itself [25]. Our study has some limitations, the most important being the small sample, although significant differences were found between arms in some of the parameters. Baseline CD4 cell count differed between arms; however, the role of CD4 count in determining biomarker concentrations has not been clearly documented in previous interruption studies [5-10].