Additionally, we do not have information about the reasons for late presentation or delayed initiation of HAART – for some individuals, their HIV diagnosis may have been delayed despite presenting for care within other medical services – and have no indication whether treatment delays were a consequence of patient- or clinician-led decisions, or as a consequence of unexpected CD4 decline. Other factors that may influence outcomes, such

as adherence, presence of symptoms (HIV- and non-HIV-related) and immune activation, are potentially important but not collected by UK CHIC so we cannot exclude them as confounders. In particular, while the CD4 cell count slope is of potential importance when analysing immune recovery on HAART, the short duration of pre-HAART follow-up in late presenters (median 0.1 years) would mean that most patients have insufficient numbers of CD4 cell count results find more available to permit a calculation of the slope. Finally, although selleck products it can be argued that CD4 percentage may be a more robust marker of immune status than the absolute CD4 cell count, most treatment guidelines

are based on the absolute CD4 cell count, and so we have used this parameter in our analysis. A previous analysis of our cohort showed absolute CD4 cell count to be a more robust predictor of clinical events than CD4 percentage [19]. Our results demonstrate that, once successful HAART is established, treatment of late presenters and starters is similarly successful as in ideal starters. Although reassuring, these findings do not challenge current treatment recommendations because of the possibility of confounding, relatively short follow-up, exclusion of subjects with poor early outcome (within the first 3 months of HAART) and lack of consideration of other endpoints pertinent to guidelines (such as toxicity and resistance development). In summary, the difference in

48-week treatment outcomes between late presenters and late starters suggests that factors other than CD4 cell count alone may be driving Pregnenolone adverse outcomes on HAART. Importantly, although late presenters demonstrated some inferior treatment outcomes to late starters and ideal starters during the first year of HAART, by year 2, differences in virological, immunological and clinical outcomes among the three groups were smaller. These findings suggest that, after 1 year of successful HAART, late presenters and late starters may eventually expect similar treatment outcomes to those who start HAART in accordance with current guidelines. Longer term follow-up is required to explore whether these small differences are further reduced with time on HAART. The UK CHIC Study is funded by the Medical Research Council, UK (grants G0000199 and G0600337).