LETTERS TO THE EDITOR Ovarian reserve screening: a scientific and ethical analysis

Sir, In a recent article, Tremellen and Savulescu (2014) ‘argue in support of screening the general population for diminished ovarian reserve (OR)’. They state that OR ‘is determined by the net result of the initial size of the primordial follicle endowment, the natural rate of atretic loss and the impact of any insult that may hasten follicle depletion’. However, OR is not defined, nor do they mention that OR is influenced by the rate at which primordial follicles are activated and leave the primordial follicle pool. They remark that anti-Müllerian hormone [AMH] and antral follicle count [AFC] are excellent quantitative markers of OR, but possibly do not reflect oocyte quality and therefore the chances of immediate natural conception. However, a poor AMH or AFC result does reflect a marked reduction in the chances of IVF-related conception. We support the call for methods to determine the OR as a means of informingwomenabout their likely future fertility and reproductive lifespan,particularly thosewithamarkedly reducedOR. It is notourpurpose here to comment on the practical or ethical issues of screening for OR, but there is a need for critical analysis of what is meant by the term OR and the limitations of the use of AMH or AFC as a measure of it. In a recent paper (Findlay et al., 2014), we argue that the term ‘ovarian reserve’ should be confined to the primordial pool of follicles. AMH and AFC only measure the pool of growing antral follicles and not the actual size of the primordial pool, and as such reflect the potential for production of oocytes within a relatively short (6 month) period of time. We propose that the follicles measured by AMH or AFC should be classed as ‘ovulatory potential’ as distinct from ‘ovarian reserve’. It is important to remember that primordial follicles do not produce AMH and cannot be detected by ultrasound. The use of serum AMH as a measure of the size of the primordial pool is basedonly on a correlation between a histological assessment of primordial follicle number in the ovary and the level of AMH in healthy adult women (Hansen et al., 2011). At present, we do not have a method other than histological analysis, of measuring directly the size of the primordial pool of follicles. The basis of the confusion about the meaning of OR and its influence on fertility most probably reflects the different interpretations by clinicians, scientists and the general public of what is meant by the ‘ovarian reserve’. We believe that agreeing on what we mean by terms like OR should be clear and consistent so as not to confuse researchers, clinicians, the media and women about their future fertility and likelihood of pregnancy. While we strongly believe that the term ‘ovarian reserve’ should be confined to the pool of primordial follicles, we are open to debate about how to define the cohort of growing antral follicles detected by AMH or AFC in terms of future fertility.We suggest ‘ovulatory potential’ (Findlay et al., 2014), highlighting both the potential of those follicles to develop further to ovulatory status (as, for example, in assisted conception) and their transient nature. Tremellen and Savulescu (2014) propose ‘total fertility potential’, which would appear to include the primordial pool of follicles. In addition to confusion about themeaningofOR, thereare limitations to using serum AMH as a measure of the ‘total fertility potential’, as defined by Tremellen and Savulescu (2014). Problems with AMH assay methodology have been reviewed recently by Dewailly et al. (2014). There are at least as many great issues with AFC measurement (Iliodromiti et al., 2014). There is enormous variability in AMH levels between individuals of the same age (Kelsey et al., 2011) and little prospective data assessing changes over time in individual women, although overall there does seem to be a relationship to time to menopause (Tehrani et al., 2013). Tremellen and Savulescu (2014) point out the influence of oral contraceptives on serum AMH levels; there is growing evidence that a wide range of health and endocrine factors can also impact AMH, complicating interpretation (Lawrenz et al., 2012; Su et al., 2013; Titus et al., 2013; van Dorp et al., 2014). Interpretation further requires recognition of the changing relationship between AMH and ovarian function through puberty, adolescence and early adulthood. In conclusion, whilst we recognize the potential value of a screening tool for OR for reproductive aged women, we emphasize that a test to measure OR does not currently exist. We look forward to further debate on the definition of the OR and the methods to assess it.