RESUMEN
OBJECTIVE: To describe the prevalence of Y-chromosome microdeletions in a multi-ethnic urban population in London, UK. To also determine predictive factors and a clinical threshold for genetic testing in men with Y chromosome microdeletions. PATIENTS AND METHODS: A retrospective cohort study of 1473 men that were referred to a tertiary Andrology centre with male factor infertility between July 2004 and December 2016. All had a genetic evaluation, hormonal profile and 2 abnormal semen analyses. Those with abnormal examination findings also had targeted imaging performed. RESULTS: The prevalence of microdeletions was 4% (n = 58) in this study. These microdeletions were partitioned into the following regions: Azoospermia factors (AZF); AZFc (75%), AZFb+c (13.8%), AZFb (6.9%), AZFa (1.7%), and partial AZFa (1.7%). A high follicle-stimulating hormone level (P < 0.001) and a low sperm concentration (P < 0.05) were both found to be significant predictors for the identification of a microdeletion. Testosterone level, luteinising hormone level and testicular volume did not predict the presence of a microdeletion. None of the men with an AZF microdeletion had a sperm concentration of >0.5 million/mL. Lowering the sperm concentration threshold to this level retained the high sensitivity (100%) and increased the specificity (31%). This would produce significant cost savings when compared to the European Academy of Andrology/European Molecular Genetics Quality Network and European Association of Urology guidelines. The surgical sperm retrieval (SSR) rate after microdissection testicular sperm extraction was 33.2% in men with AZFc microdeletion. CONCLUSIONS: The prevalence of Y-chromosome microdeletions in infertile men appears to vary between populations and countries. A low sperm concentration was a predictive factor (P < 0.05) for identifying microdeletions in infertile males. A threshold for genetic testing of 0.5 million/mL would increase the specificity and lower the relative cost without adversely affecting the sensitivity. The rate of SSR was lower than that previously described in the literature.