Background In both in vitro fertilisation (IVF) and intracytoplasmic sperm injection (ICSI), selection of the most competent embryo(s) for transfer is generally based on morphological criteria. However, many women fail to achieve a pregnancy after transfer of good quality embryos. One of the presumed causes is that such morphologically normal embryos show an abnormal number of chromosomes (aneuploidies). In preimplantation genetic screening (PGS), embryos are analysed for aneuploidies and only embryos that are euploid for the chromosomes tested are transferred. This technique has been suggested and used to improve pregnancy rates for the following indications: (i) advanced maternal age, (ii) repeated IVF failure, (iii) repeated miscarriage and (iv) testicular sperm extraction (TESE)-ICSI. Although PGS is used more and more often, its effectiveness is still unclear. Objectives To assess the effectiveness of PGS in terms of live births in women undergoing IVF or ICSI treatment. Search strategy We searched the Cochrane Menstrual Disorders and Subfertility Group Trials Register, the Cochrane Central Register of Controlled Trials (CENTRAL) (Cochrane Library Issue 1, 2005), MEDLINE (1966 to present) and EMBASE (1980 to present) (searched March 2005) and reference lists of articles. We also contacted authors for providing additional data when necessary. Selection criteria Trials for all four suggested indications as mentioned above were sought. All relevant published randomised controlled trials were selected. They were eligible for inclusion if the comparison dealt with IVF/ICSI with PGS versus IVF/ICSI without PGS. Data collection and analysis Relevant data were extracted independently by two authors. All trials were screened and analysed according to predetermined quality criteria. Validity was assessed in terms of method of randomisation, completeness of follow-up, intention-to-treat analysis and presence or absence of blinding. The primary outcome measure was live birth rate per woman. Secondary outcome measures were the proportion of women reaching embryo transfer, mean number of embryos transferred per transfer, clinical pregnancy rate, multiple pregnancy rate, miscarriage rate, ongoing pregnancy rate, proportion of women reaching embryo transfer after cryopreservation and proportion of women whose child has a congenital malformation. Main results Two randomised controlled trials met our predetermined eligibility criteria. These trials used PGS for advanced maternal age. The primary outcome of live birth rate per woman was not significantly different in the PGS and control groups, though data were only available from one study. The live birth rate was 11% (21 out of 199) in the PGS group, versus 15% (29 out of 190) in the control group (OR 0.65; 95% CI 0.36 to 1.19). For a control group rate of 15%, these data suggest a live birth rate using PGS of between 4% and 17%. Ongoing pregnancy rate was provided in both studies. This was not significantly different with a combined odds ratio of 0.64 (95% CI 0.37 to 1.09). For a control group rate of 20%, this suggests an ongoing pregnancy rate using PGS of between 8% and 21%. Authors' conclusions To date there is insufficient data to determine whether PGS is an effective intervention in IVF/ICSI for improving live birth rates. Available data on PGS for advanced maternal age showed no difference in live birth rate and ongoing pregnancy rate. However, only two randomised trials were found, of which one included only 39 patients. For both studies comments on their methodological quality can be made. Therefore more properly conducted randomised controlled trials are needed. Until such trials have been performed PGS should not be used in routine patient care.