The Bologna criteria for poor ovarian response: the good, the bad and the way forward.

The management of poor ovarian response (POR) remains one of the most significant challenges of in vitro fertilization. Numerous interventions have been proposed, yet few have been shown to be beneficial. The most important problem in evaluating the available evidence has been the lack of a standardized definition of POR. The Bologna criteria for POR have been recently introduced to provide a framework allowing future research in this field to be performed on a relatively homogenous population. However, it has been suggested that the population described by the Bologna criteria might not be sufficiently homogenous and for this reason stratified randomization should be used in relevant randomized controlled trials (RCTs). Stratified randomization, besides its advantages, also has important shortcomings and for this reason it should be used only when there is clear evidence mandating such a design. Currently, there is insufficient data to support such practice in RCTs performed on the population described by the Bologna criteria for POR. Until such evidence becomes available, the scientific community should aim at evaluating interventions for POR according to the Bologna criteria in RCTs of sufficient sample size, with proper allocation concealment and masking.