In 2011 the European Society for Human Reproduction and Embryology published the first consensus on poor ovarian response and introduced ESHRE’s diagnostic criteria for POR, also known as the Bologna criteria. The diagnostic criteria provide guidelines for homogeneity and standardization of research in the field of POR. However, with the clinical application of the Bologna criteria in recent years, some of the shortcomings of the criteria have led to in-depth consideration, and there is controversy as to whether the criteria need to be revised. This article will explain the current use and research of the Bologna criteria in the context of the current situation.
1. Background of POR consensus
POR is a state of poor ovarian response to gonadotropin stimulation, mainly characterized by low follicle development in ovarian stimulation cycles, low peak blood estrogen, high cycle cancellation rate, low number of eggs obtained and low clinical pregnancy rate. The incidence of POR has been reported in the literature to be approximately 9% to 24%. The pathophysiological mechanisms of POR are not fully understood, and the emergence of many unanticipated PORs during the course of treatment has become a difficult clinical problem. Therefore, the prediction and management of POR has become a hot research topic in the field of ovulation promotion in assisted reproductive technologies.
However, studies on POR before the publication of the ESHRE consensus were in a confusing state of poor homogeneity and low credibility, most notably due to the lack of internationally accepted definitions and diagnostic criteria for POR. For example, prior to 2011, there were 47 randomized controlled studies of POR, but 41 different definitions were used, and one study group even used 2 definitions of POR at the same time. What is more, none of the criteria within these 41 definitions was used in more than 50% of the studies, while age and sinus follicle count (AFC) were included as criteria in only 9% of the studies [4].
Following various calls, in 2011 ESHRE proposed the first international diagnostic criteria for POR, also known as the Bologna criteria. The popularity of this research topic is evident from the 98 citations of the criteria once they were proposed to date. The Bologna criteria helped overcome the difficulties of previous POR studies by specifying population inclusion criteria for randomized controlled studies and meta-analyses of POR, laying the foundation for homogeneity and standardization of future POR studies.
2. The Bologna Criteria
2.1 Bologna criteria in the POR consensus proposed by ESHRE in 2011
(1) Advanced age (≥40 years) or presence of other risk factors for ovarian hyporesponsiveness.
(2) POR in the previous in vitro fertilization (IVF) cycle (≤3 eggs obtained with conventional stimulation protocol).
(3) Abnormal ovarian reserve function test [AFC <5 to 7 or Anti-Mullerian Hormone (AMH) <0.5 to 1.1 μg/L].
POR is diagnosed when at least 2 of the above 3 criteria are met, and is directly diagnosed when a patient has 2 cycles of the maximum dose ovarian stimulation regimen and still has POR.
By definition, POR reflects ovarian responsiveness and therefore at least one cycle of stimulation is necessary for the diagnosis. However, patients older than 40 years combined with an abnormal ovarian reserve function test have been substituted for ovarian stimulation as an indication of inadequate ovarian reserve function, so these patients are defined as having expected POR.
The Bologna criteria well combine the 3 main clinical features with simplicity, clarity of definition and reproducibility, but their feasibility and validity need to be validated by a large number of prospective clinical studies.
2.2 Cut-off values of the Bologna criteria
2.2.1 Age
The Bologna criteria emphasize the importance of age on ovarian reserve function in women. A study that included 3825 women with correlation between age and poor ovarian response demonstrated that the incidence of POR increased with age. In women older than 40 years, the incidence of POR reached 50% [5].
However, some women of advanced age between 40 and 42 years can still produce sufficient follicles and eggs, while some younger women are not protected from POR [4]. Therefore, advanced age (≥40 years) can be the most relevant risk factor for POR, but it needs to be combined with other clinical manifestations to confirm the diagnosis.
The selection of 40 years as a cut-off value is still controversial. Age is the best predictor of egg quality and the most important independent influence in ongoing pregnancy. Age is also the only predictor of live birth rate [6-7]. Normally, an abrupt decline in follicular pool occurs around 37 years of age [8]. However, under IVF stimulation cycles, there is no significant decline in ovarian responsiveness until the age of 43 years, and the incidence of poor ovarian response amounts to
Other risk factors other than POR proposed by the ESHRE consensus include: genetic or acquired disorders that affect ovarian reserve function and response to ovarian stimulation, such as abnormalities in the number and structure of chromosomes, genetic mutations (e.g., Tunner syndrome and mutations in the FMR1 gene); previous history of pelvic inflammatory disease, endometriosis, and surgery for ovarian cysts. History of radiotherapy (especially alkylating chemotherapeutic agents) can also lead to varying degrees of ovarian insufficiency. Shortened menstrual cycles are also a risk factor for POR.
The risk factors for ovarian hyporesponsiveness are still controversial and therefore the ESHRE consensus does not have a very detailed definition [19]. As new evidence accumulates and is confirmed, further refinement of the relevant risk factors is needed to further revise the POR criteria. Of course, it is undeniable that no single risk factor can directly and accurately predict POR, and therefore, the ESHRE consensus suggests that POR should be diagnosed in those aged <40 years with a history of at least 1 previous conventional stimulation cycle por combined with a history of abnormal ovarian reserve function test in a high-risk group with low ovarian response.
3. Homogeneity of the population included in the Bologna criteria
In response to the ESHRE consensus that POR can be diagnosed by meeting at least 2 of the 3 criteria, a very pointed question has been raised as to whether the POR diagnostic criteria can really guarantee the selection of a highly homogeneous POR population in future studies. Since the criterion is 2 out of 3, patients selected according to this criterion may have 8 different subtypes. In addition, the Bologna criteria focus on the number of eggs more than the quality of eggs, but clinical prognosis is not only related to the number of eggs retrieved, but also to the quality of eggs. Therefore, there is the possibility of differences in baseline and diverse clinical prognosis in clinical trials of POR, leading to potential bias [20].
In response to the clinical heterogeneity defined by POR, it has been suggested that stratified randomized controlled trials need to be applied inside future RCT studies to facilitate the selection of homogeneous populations with more accurate and simple criteria. However, such restricted RCT studies may lead to non-randomization in stratification and increase the systematic error, and the inclusion of stratified randomization also greatly increases the operational difficulty of RCT studies. Its validity has yet to be determined by further studies.
4. Prospects for POR clinical studies using Bologna criteria
So far, the ESHRE Bologna criteria have provided a good benchmark for clinical studies of POR, but they have also exposed the lack of homogeneity of the population included in the criteria for POR and the low predictability of pregnancy outcomes. Therefore, we need to validate the Bologna criteria using a rigorously designed RCT combined with a meta-analysis of RCT studies with small samples. The POR population meeting the Bologna criteria will be considered for subtype analysis if there are significant prognostic differences in the results of high-quality studies [23].