Prevalence and diagnosis of congenital uterine anomalies in women with reproductive failure: a critical appraisal.

BACKGROUND The prevalence of congenital uterine anomalies in women with reproductive failure remains unclear, largely due to methodological bias. The aim of this review is to assess the diagnostic accuracy of different methodologies and estimate the prevalence of congenital uterine anomalies in women with infertility and recurrent miscarriage (RM). METHODS Studies from 1950 to 2007 were identified through a MEDLINE search; all relevant references were further reviewed. RESULTS The most accurate diagnostic procedures are combined hysteroscopy and laparoscopy, sonohysterography (SHG) and possibly three-dimensional ultrasound (3D US). Two-dimensional ultrasound (2D US) and hysterosalpingography (HSG) are less accurate and are thus inadequate for diagnostic purposes. Preliminary studies (n = 24) suggest magnetic resonance imaging (MRI) is a relatively sensitive tool. A critical analysis of studies suggests that the prevalence of congenital uterine anomalies is approximately 6.7% [95% confidence interval (CI), 6.0-7.4] in the general population, approximately 7.3% (95% CI, 6.7-7.9) in the infertile population and approximately 16.7% (95% CI, 14.8-18.6) in the RM population. The arcuate uterus is the commonest anomaly in the general and RM population. In contrast, the septate uterus is the commonest anomaly in the infertile population, suggesting a possible association. CONCLUSIONS Women with RM have a high prevalence of congenital uterine anomalies and should be thoroughly investigated. HSG and/or 2D US can be used as an initial screening tool. Combined hysteroscopy and laparoscopy, SHG and 3D US can be used for a definitive diagnosis. The accuracy and practicality of MRI remains unclear.