Who needs ICSI? A nationwide UK survey on ICSI use.

Background: Intracytoplasmic sperm injection (ICSI) has been a milestone in the treatment of male factor infertility. However ICSI is more expensive, demands more expertise, and involves more risk than conventional in vitro fertilisation (IVF). Currently there are large nationwide differences in ICSI usage, with some centres using ICSI for 21% of their IVF cycles and others for more than 80%. This is, most likely, due to differences in ICSI selection criteria but there are limited data on the criteria used. We have therefore carried out a national survey in the UK, the first, as far as we are aware, to examine different criteria used and their effect on ICSI usage and treatment outcomes. Methods: Centres which offer ICSI were identified using the Human Fertilisation and Embryology Authority (HFEA) website. Questionnaires were then posted to all centres which offer the procedure. Each centre received the questionnaire twice; the first was sent to the HFEA person responsible and a month later, a follow-up questionnaire was sent to the centre's lead embryologist. Data were also extracted from the HFEA website. Results: 71 centres were identified and questionnaires returned from 43 (61%). When deciding to use ICSI, 43 (100%) of centres used sperm count, 93% used sperm motility, 76% used sperm morphology and 72% used anti-sperm antibodies. All centres stated that they would offer ICSI after failed fertilisation with conventional IVF and 38% of centres offered ICSI on patient request. No centres reported using other criteria for selection. The absolute values chosen for each criterion varied hugely between centres. Compared with the 2010 World Health Organization (WHO) guidelines of normal semen analyses, 32% of centres used a higher count, 50% a higher motility and 59% a higher morphology. Based on the WHO criteria, 27% of centres would use ICSI for sperm that were normal by all WHO criteria. Between centres, no significant difference in ICSI fertilisation rates was found. However, there was a significant negative correlation between increased ICSI usage and fertilisation rates by conventional IVF (p = 0.0058). Data obtained from the HFEA website failed to demonstrate an increase in live birth rate in centres using ICSI more frequently. Conclusion: ICSI usage varied widely, due to large differences in the ICSI selection criteria used, with many centres using ICSI for patients with normal semen parameters. Centres which used more ICSI did not report higher live birth rates. No evidence was found to suggest that higher ICSI usage increased overall fertilisation rates. These findings highlight the need for guidelines on when to use ICSI.