The closing years of the 20th Century saw an unprecedented explosion in the knowledge base of reproductive medicine. In turn, this knowledge has led to new approaches to the diagnosis and treatment of disease. The scientific foundation of reproductive medicine continues to expand rapidly in the “post-genomic” era, but at the same time there is an urgent need to translate scientific advances into clinical practice.

Estrogens are essential regulators of female fertility. Estrogen biosynthesis is dependent upon expression of the cytochrome P450 aromatase, the highest levels of expression of which are detected in the human placenta and the granulosa cells of the mature ovarian follicle. Consistent with this pattern of expression, in non-pregnant premenopausal women, the ovaries have been shown to be the primary source of estradiol. In recent years there has also been an increased interest in the role(s) played by estrogens in regulation of adipose tissue, the vasculature and bone and as paracrine regulators of tissue function in postmenopausal women. In the premenopausal years the uterus is a target for estrogen and the cyclical event of menstruation is a consequence of the sequential exposure of the endometrium to estrogen and progesterone (reviewed in references 5 and 6). Endometrial breakdown occurs as a consequence of progesterone withdrawal with the demise of the corpus luteum (CL) in the absence of pregnancy.

Only 50% of all conceptions result in a live birth (Figure 1). Human reproduction can therefore be viewed as being remarkably inefficient. However, the diametrically opposite opinion that it is in fact a very efficient and selective process, designed to optimize the outcome of pregnancy, deserves to be stated.

Between 65% and 90% of clinically recognized miscarriages are due to chromosome abnormalities, the occurrence of which is more closely related to basal follicle-stimulating hormone (FSH) levels rather than to maternal age alone. In contrast to women suffering a sporadic miscarriage, women who recurrently miscarry often, but not exclusively, lose pregnancies with a normal chromosome content. Using a combination of conventional Geimsa banding and the recently introduced technique of comparative genomic hybridization it has been reported that 54% of pregnancy losses amongst women with recurrent miscarriage are euploid. The challenge we face is to identify the causes of pregnancy loss amongst those couples who recurrently lose such euploid conceptions.

Whilst many conventionally held beliefs as to the aetiology and treatment of women with recurrent miscarriage have not withstood critical scrutiny, significant progress has been made. This progress is the subject of this review, which will also highlight potential areas of future research.

Percutaneous trans-catheter embolization has been practised by radiologists for well over 20 years. In many different clinical situations a great variety of embolization materials or agents has been used in all parts of the body, but generally these procedures have been performed rarely. An important indication is severe bleeding not responding to conservative measures, where the alternative treatment would involve major surgery. Embolization has also been used in tumours, particularly where they are hypervascular, when the role has often been to debulk and devascularize immediately prior to surgery. The third main indication is in arteriovenous malformations and fistulae. Thus, it is somewhat surprising that it was not until 1995 that uterine artery embolization (UAE) was first advocated as a treatment for uterine fibroid disease.

Embryonic stem (ES) cells are a primitive cell type derived from the inner cell mass (ICM) of the developing embryo. When cultured for extended periods, ES cells maintain a high telomerase activity, normal karyotype and the pluripotential developmental capacity of their ICM derivatives. Such capacity is best demonstrated by mouse ES cells which can contribute to all tissues of the developing embryo following either their injection into host blastocysts or tetraploid embryo complimentation (for a review see Robertson). For both practical and ethical reasons it is not possible to inject human ES cells into blastocysts for the development of a term fetus. However, when injected beneath the testicular capsule of severe combined immunodeficient (SCID) mice, human ES cells form teratomas comprising tissue representatives of all three embryonic germ layers (ectoderm, mesoderm and endoderm) thus attesting to their pluripotency. Based upon morphological criteria, neuronal, cardiac, bone, squamous epithelium, skeletal muscle, gut and respiratory epithelia are readily identifiable within the human ES-cell-derived teratomas. With the demonstrated capability to isolate and maintain pluripotent human ES cells in vitro, their ability to give rise to tissue representatives of all three embryonic germ layers and the technical advances made possible by research on mouse ES cells, a rapid increase in human ES cell research aimed at drug discovery and human cell and gene therapies has occurred. Indeed in the mouse, dissociated embryoid bodies (EBs) have already been demonstrated capable of repopulating the haematopoietic system of recipient animals (for a review see Keller) and mouse ES cells are currently being used in attempts to repair mouse neural degenerative lesions.