Background Our ARV Network's 2009 audit highlighted the large proportion of patients with a CD4 count <350. A re-audit was designed to provide more information on patients with CD4 counts in this range.

Methods We conducted a retrospective review of case notes for all patients in the network starting ARVs in 2010. For analysis of CD4 counts the cohort was divided into two groups; those diagnosed within 1 year of starting treatment and others.

Results 114 patients started ARVs in 2010 from four centres in our network. 62 (54.4%) were male. Mean age was 38.4 years (range 17–62). Ethnicity data showed only 37.7% were white with the majority being Black (54.4%). 6 (7.0%) of 85 patients had a major NNRTI resistance mutation. Mean nadir CD4 count was 222 (range 5–610). 101 (88.6%) patients had a CD4 count under 350. 106 reasons for low CD4 count were recorded. 65 patients (64.3%) had low CD4 counts because of late diagnosis, 15 (14.8%) had declined ARV when initially offered while 10 (9.9%) had been lost to follow-up. The patients starting Rx within 1 year of diagnosis (no=67) had a lower mean nadir CD4 count compared to those diagnosed earlier (no=47) (162 cells vs 271 cells, p<0.5). There was no difference between the two groups in the number of patients having a pre-treatment resistance test, the mean CD4 rise 6 months after treatment initiation and the proportion of patients having an undetectable viral load 12 months after treatment initiation. At 6 months the mean CD4 count had risen from 222 at treatment initiation to 360, but 54 (47.4%) still had a CD4 count under 350. The main reasons for this were poor immune recovery in 80.7%, poor adherence 7%, poor attendance 5.3%.

Discussion A proportion of our cohort started ARVs with a low CD4 count mainly due to late diagnosis. This is an important barrier to ARV initiation and needs to be addressed and our audit data would support the need for extra support and resources directed to earlier HIV diagnosis.