Background
The prevalence of infertility is estimated to be 12–15% in couples of childbearing age [
1,
2]. Even though the development of assisted reproductive technology (ART) has helped many couples achieve pregnancies, the rates of pregnancy and live births among all ART-treated couples is still not satisfactory. How to improve assisted reproductive outcome has become a critical topic for both infertile couples and clinicians.
Physical activity is generally considered to be a health promoting behaviour as it is associated with reduced risks of cardiovascular disease, diabetes and cancers [
3,
4]. A number of epidemiological studies have focused on the effects of physical activity on fertility, but no consensus has been reached until now [
5‐
7]. Most of the investigators reported that moderate physical activity benefits female fertility [
5‐
7], whereas high intensity and frequency of physical activity increase subfertility and infertility, mainly ovulatory infertility [
6], and increasing vigorous physical activity is associated with delayed time to spontaneous pregnancy [
7]. Similar conclusions were also reached in athletes, who have been shown to have a higher prevalence of reproductive dysfunction compared with non-athletes [
8,
9].
Physical activity during pregnancy has been shown to improve reproductive outcome [
10,
11]. However, no consensus has been reached regarding the effect of physical activity before ART on pregnancy outcome [
12‐
14]. A population-based cohort study [
12] showed that women engaged in physical activity for ≥4 h/week had a 40% decreased likelihood of live birth in in vitro fertilization (IVF) cycles compared with women not regularly engaged in physical activity. This conclusion was supported by some other investigations focusing on IVF or intracytoplasmic sperm injection (ICSI) cycles [
13,
15,
16]. Whereas a randomized controlled trial (RCT) [
17] showed no differences in pregnancies or live births between the physical activity intervention and control groups. Gaskins et al. [
14] also found that time spent in moderate to vigorous physical activities and total metabolic equivalent task hours before IVF were not associated with probability of implantation, clinical pregnancy or live birth.
Therefore, in this systemic review and meta-analysis, we aimed to evaluate the association between maternal physical activity before IVF/ICSI cycles and reproductive outcomes, to provide a comprehensive analysis of the current data and a context for how to counsel infertile couples and physicians trying to improve the success rate of assisted reproductive treatment.
Discussion
To our knowledge, this is the first meta-analysis to evaluate the relationship between female physical activity before IVF/ICSI cycles and ART outcomes. We found that physical activity before IVF/ICSI cycles is associated with better assisted reproductive outcomes, mainly based on the increase in the rates of clinical pregnancy and live birth, and also a small but not statistically significant increase in the implantation rate, whereas the miscarriage rate was not associated with physical activity in women before ART cycles. These results suggested that exercise before IVF/ICSI cycles may clearly help physically inactive women to improve their chance of a successful pregnancy.
The effect of exercise on fertility and IVF outcomes has been a subject of considerable dispute. A recent population-based cohort study indicated that moderate exercise is associated with better clinical outcomes, regardless of BMI [
7]. In 2008, the (United States Department of Health and Human Services (USDHHS) released in the “Physical Activity Guidelines for Americans” which recommended at least 150 min of moderate-intensity physical activity per week for pregnant women without obstetric/medical complications [
30]. However, until now, no population-based RCT was conducted to show whether moderate physical activity prior to IVF/ICSI was beneficial to ART outcomes. The only RCT enrolled 38 overweight/obese women who received lifestyle intervention (exercise and weight-loss diet) or standard treatment for 5–9 weeks before oocyte pick-up. The results showed no significant improvement of clinical pregnancy and live birth in intervention patients compared with control. However, the small sample size in that study may be insufficient to draw a solid conclusion. In this meta-analysis, we provided a comprehensive analysis of the current data and found a 1.96-fold and 1.94-fold increase of clinical pregnancy rate and live birth rate, respectively, in physical active women compared with physical inactive women. Whereas no significant differences were found in implantation rate and miscarriage rate. According to the GRADE criteria, the combined results were assessed with moderate to low quality evidence, since most included studies were prospective cohort studies, and cofounders like dietary pattern, duration of physical activity as well as male factors in patients would affect clinical outcomes. Additionally, heterogeneity among studies existed in every clinical outcome. Considering that the following two aspects may affect the combined results: 1) age and BMI were critical confounders affecting IVF/ICSI outcomes [
27‐
29], but were not controlled in some of included studies; 2) Intensity of physical activity was not consistent among included studies, we performed sub-group analysis to further evaluate the relationship between physical activity before IVF/ICSI cycles and clinical outcomes. As a result, we found very similar pooled effects in comparison with the overall results, suggesting that the effect of physical activity on pregnancy outcome was independent of age and weight loss. Additionally, in most of the included studies, patients reduced their physical activity when pregnancy was achieved. This may help to exclude the effect of physical activity during pregnancy on assisted reproductive outcomes, and this has been shown in several studies [
31‐
33].
The mechanisms by which physical activity prior to IVF/ICSI cycles improves pregnancy outcome may be very complex, and no molecular pathway has been identified. The most relevant determinants of regular physical activity on reproductive outcome seem to be mainly related to its effect on the clinical pregnancy rate. This may be due to the following reasons: First, physical activity performed to improve health status may promote changes in energy balance, which, in turn, is tightly correlated with the reproductive system [
34]. Second, physical activity may improve the assisted reproductive outcome through insulin sensitization, restore ovarian function [
35] and sensitize the ovary to clomiphene citrate during simple ovulation induction [
36]. Many infertile women are characterized by obesity, which is positively associated with insulin resistance, for instance, in polycystic ovary syndrome [
37,
38]. Regular physical activity is also known to be an effective therapeutic intervention to improve glucose homeostasis and insulin sensitivity [
39]. Endometrial insulin resistance should also be noted because studies have proved that a reduction in insulin resistance at the endometrial level induced by insulin-sensitizing agents leads to changes in the expression of glucose transporter endometrial protein [
40] and is associated with a declining risk for miscarriage and implantation failure in IVF cycles during clinical observation [
41]. Additionally, regular physical activity can help relieve stress and anxiety, which have been shown to be important risk factors affecting the assisted reproductive outcome [
33,
42].
The strength of this review is that we focused on the effect of a very common lifestyle choice on assisted reproductive outcome, and the conclusion is very critical when providing consultation to infertile couples. All included studies established a good contrast between no regular physical activity and regular physical activity, and the intensity of physical activity was within a normal range. This excluded the effect of high-intensity physical activity (e.g., that performed by athletes) on pregnancy outcome. This study also has several limitations. First, the number of included studies was limited. The information on physical activity was based on the memories of the infertile couples. However, all physical activity information of the participants was recorded before the IVF/ICSI cycles, and they did not know whether they would become pregnant, which suggests the truthfulness of the results. Second, the questionnaires in the included studies differed from each other and bias cannot be excluded, even though all studies established a reliable contrast regarding different levels of physical activity. Third, we could not eliminate the effects from other confounders, such as etiology of infertility, dietary habits and psychological factors, which may also affect assisted reproductive outcomes.
To provide better evidence regarding the relationship between physical activity before ART and clinical outcomes, carefully controlled and sufficiently powered intervention studies are needed, frequency, intensity and duration of physical activity and potential confounders like dietary pattern, male factor etc. should be considered.