In brief, it involves centrifuging the ejaculated SB431542 cell line semen in a 45–90% colloidal silica density gradient to separate progressively motile, HIV-free sperm from the infected NSCs and seminal plasma that remain in the supernatant. Specific precautions, as described elsewhere, are taken to minimize the risk to staff and the risk of cross-contamination of uninfected gametes and embryos (e.g. samples are handled within a separate high-security laboratory) [23], and semen parameters are assessed according to World

Health Organization criteria [24]. Following initial centrifugation, the ‘wash’ process of aspirating the supernatant and re-suspending the sperm pellet in fresh medium before further centrifugation was classically performed three times to maximize the clearance of NSCs, before the preparation of a final swim-up. From September 2008, a modified protocol of two washes, with a swim-up performed only if the patient was not on HAART or if the sample was prepared with significant debris (white blood cells etc.), was adopted. This was introduced to

maximize the post-wash sperm yield Fluorouracil with no increase in residual virus demonstrated [25]. As a quality control for the procedure, an aliquot of washed sperm (approximately 100 mL) was subsequently tested for detectable HIV RNA prior to the sample being used for treatment. A nucleic acid sequence-based amplification (NASBA; Biomerieux, Basingstoke, UK) was initially used and from June 2007 a Roche (Burgess Hill, UK) polymerase chain reaction (PCR) assay was performed in view of the improved time efficiency of this assay, with a detection limit of >25 HIV-1 RNA

copies per 106 sperm. From 2004, it became mandatory for couples Cyclooxygenase (COX) to freeze a washed, negative sample as back-up in case residual HIV was found in a post-wash sample which would otherwise have necessitated cycle cancellation. Correlation analyses on nonparametric continuous variables (between sperm parameters and CD4 cell count, VL, years since diagnosis and years of HAART use) were performed using Spearman rank correlation where two related variables (i.e. variables measured on the same subject without necessarily utilizing the same unit) were being assessed. The degree of association is expressed as the correlation coefficient, with a value of between +1 and −1. A value near 0 suggests no correlation and the variables are independent with no effect on each other. A positive value suggests a positive correlation existing between two variables to a perfect positive correlation at +1, and a negative value suggests a negative correlation existing between two variables to a perfect negative correlation at −1. The significance of any correlation is expressed as a P-value. The Mann–Whitney U-test was also used to detect an effect of the categorical variables VL (detectable vs. undetectable), CD4 cell count and use of antiretrovirals on sperm parameters.