Reply to: Comment on: “The Effects of Menstrual Cycle Phase on Exercise Performance in Eumenorrheic Women: A Systematic Review and Meta-Analysis” and “The Effects of Oral Contraceptives on Exercise Performance in Women: A Systematic Review and Meta-analysis”

We thank Lei et al. [1] for their interest in our articles [2, 3]. In their letter, Lei et al. [1] questioned why certain studies [4–8] had not been included in our reviews. The reason for this is because our research question pertained to the effects of ovarian steroids on exercise performance in eumenorrheic women [2] and oral contraceptive users [3] and was not intended to look at these relationships in response to ambient heat stress or other conditions. In this response, we will clarify the search and selection strategy of our reviews to address the misunderstanding between ourselves and Lei et al. [1]. (i) Our search strategies reflected our research question and although the studies [4–8] listed by Lei et al. [1] in their letter were not included, these omissions do not reflect an incomplete search in the context of our research aims. We admit that no search is entirely exhaustive; however, we respectfully disagree with Lei et al. [1] that we have overlooked their studies [4–8]. Rather, they were outside the scope of our reviews [2, 3]. As such, we stand by our search strategies which were aligned with our research questions and intended outcomes and were constructed according to Cochrane Library guidelines. (ii) As identified by Lei et al. [1], we included two studies [9, 10] in our menstrual cycle review [2] that investigated the influence of menstrual cycle phase on exercise under thermally stressed conditions, but not others [4–8]. It is important to clarify, however, that only the thermoneutral data were extracted from these studies [9, 10], in agreement with our research question. These studies [9, 10] were included as they were identified through our research specific, pre-defined search strategy, and were not specifically sought by using search terms related to thermoregulation (for a full list of search terms see Sect. 2.2 [2]), and were deemed eligible based on our inclusion criteria, namely: (i) no specific interventions (e.g., dry/humid heat) were investigated; and (ii) only studies that had the primary or secondary objective of assessing changes in exercise performance across the menstrual cycle were included (e.g., not exercise performance during the menstrual cycle under different conditions of temperature). In contrast, the studies [4–8] listed by Lei et al. [1] were either (i) not identified using This reply refers to the letter available at https ://doi.org/10.1007/ s4027 9-020-01384 -8 commenting on the parent articles available at https ://doi.org/10.1007/s4027 9-020-01317 -5 and https ://doi. org/10.1007/s4027 9-020-01319 -3.


Dear Editor,
We thank Lei et al. [1] for their interest in our articles [2,3]. In their letter, Lei et al. [1] questioned why certain studies [4][5][6][7][8] had not been included in our reviews. The reason for this is because our research question pertained to the effects of ovarian steroids on exercise performance in eumenorrheic women [2] and oral contraceptive users [3] and was not intended to look at these relationships in response to ambient heat stress or other conditions. In this response, we will clarify the search and selection strategy of our reviews to address the misunderstanding between ourselves and Lei et al. [1].
(i) Our search strategies reflected our research question and although the studies [4][5][6][7][8] listed by Lei et al. [1] in their letter were not included, these omissions do not reflect an incomplete search in the context of our research aims. We admit that no search is entirely exhaustive; however, we respectfully disagree with Lei et al. [1] that we have overlooked their studies [4][5][6][7][8]. Rather, they were outside the scope of our reviews [2,3]. As such, we stand by our search strategies which were aligned with our research questions and intended outcomes and were constructed according to Cochrane Library guidelines. (ii) As identified by Lei et al. [1], we included two studies [9,10] in our menstrual cycle review [2] that investigated the influence of menstrual cycle phase on exercise under thermally stressed conditions, but not others [4][5][6][7][8]. It is important to clarify, however, that only the thermoneutral data were extracted from these studies [9,10], in agreement with our research question. These studies [9,10] were included as they were identified through our research specific, pre-defined search strategy, and were not specifically sought by using search terms related to thermoregulation (for a full list of search terms see Sect. 2.2 [2]), and were deemed eligible based on our inclusion criteria, namely: (i) no specific interventions (e.g., dry/humid heat) were investigated; and (ii) only studies that had the primary or secondary objective of assessing changes in exercise performance across the menstrual cycle were included (e.g., not exercise performance during the menstrual cycle under different conditions of temperature). In contrast, the studies [4][5][6][7][8] listed by Lei et al. [1] were either (i) not identified using our research specific, a priori search terms [8]; (ii) not indexed in the databases we searched [5]; or (iii) not deemed eligible based on our inclusion criteria [4,6,7]. As such, we believe that the scope of our review precluded the studies listed by Lei et al. [1] In terms of future work, we agree that a more explicit list of eligibility criteria and/or some exemplars of certain criteria would be beneficial and help to avoid misinterpretation between authors and their readers. We note that Lei et al. [1] are active researchers in the field of thermoregulation, and believe their work is more suited to a specific meta-analysis investigating the effects of menstrual cycle phase on exercise performance in eumenorrheic women and oral contraceptive users in response to ambient heat stress or as part of a wider review of the effects of menstrual cycle phase on performance in response to different environmental conditions.
In regard to the quality assessment used in the menstrual cycle paper, we cited two studies [11,12] to underpin the inclusion of urinary ovulation detection kits in our approach. The purpose of urinary ovulation detection kits is to confirm ovulation, which is only part of the criteria to establish eumenorrhea. Eumenorrhea is a multifaceted term that reflects several concurrent requirements, namely menstrual cycle lengths between 21 and 35 days resulting in nine or more consecutive periods per year; evidence of the mid-cycle luteinising hormone surge; the correct hormonal profile (see Fig. 1, [2]); and no hormonal contraceptive use in the 3 months prior to recruitment. As such, we believe that it is appropriate to downgrade an assessment of the research quality of studies with serum hormone analysis, but without urinary ovulation detection kits. If we consider studies comparing the follicular (i.e., low oestrogen and progesterone) with the luteal (i.e., higher oestrogen and progesterone) phase, blood samples will have been drawn at these two timepoints and might show that ovarian steroid concentrations are within the correct ranges for these phases. However, this design would not confirm that ovulation has taken place, which is a key indicator of eumenorrhea, as the mid-luteal rise in oestrogen and progesterone can occur in the absence of ovulation [13]. Moreover, the phase "ranges" for serum oestrogen and progesterone are broad and overlapping; for example, the follicular, ovulatory and luteal phases are characterised by oestradiol concentrations of 72-529, 205-786 and 235-1309 pmol•L −1 , respectively As such, urinary ovulation detection kits provide additional context for interpreting hormonal data. Lastly, it is important to note that urinary ovulation detection kits are prospective (i.e., are used to predict phases), while blood samples are retrospective (i.e., used to confirm phases). Indeed, Janse de Jonge et al. [12] recommended that a combination of methods be used to verify menstrual cycle phase, namely calendar-based counting, urinary ovulation detection kits and serum oestrogen and progesterone concentrations. Collectively, it is evident that urinary ovulation detection kits are an integral part of the research design of menstrual cycle studies.
We would like to clarify several points from the following statement made by Lei et al. [1] based on the paper by Janse de Jonge et al. [12]: "The authors refer to a previous paper as their reference for this, yet up to 50% of (highly) physically active women may have luteal phase-deficient and/or anovulatory cycles that would provide a false-positive result with this methodological recommendation." First, Janse de Jonge et al. [12] cite one paper [13] that showed that it is not uncommon for false positive results to occur when participants interpret urinary ovulation detection kits at home. Secondly, Janse de Jonge et al. [12] state that there is a high prevalence (30%) of anovulation and luteal phase deficiency in physically active women. Thirdly, Janse de Jonge et al. [12] state that the prevalence of anovulatory and luteal phase deficiency can be as high as 50% in heavily exercising women. As such, contrary to the statement by Lei et al. [1] above, we do not believe that Janse de Jonge et al. [12] were stating that anovulatory and/or luteal phase deficient cycles result in false-positive urinary ovulation detection kit results. Instead, we agree with Janse de Jonge et al. [12] when they conclude that urinary ovulation detection kits are useful for determining ovulation and the mid-luteal phase, but do not preclude luteal phase deficient cycles. We assert that this conclusion further highlights the need for blood sample verification alongside urinary ovulation detection kits.
Our reviews [2,3] were intended to address a specific research question and we fully acknowledge that all systematic reviews and meta-analyses are limited by their scope. There is a need for further studies, and reviews, in sport and exercise science with women as participants. Furthermore, a universal set of standards is required to improve the quality of future studies in this area and to consistently and appropriately appraise the quality of existing studies.

Declarations
Funding No sources of funding were used to assist in the preparation of this letter. Eimear Dolan is supported by a research grant (2019/05616-6) from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP).