Multiple systematic reviews and meta-analyses have
considered the question of whether antenatal physical activity, compared with
no such intervention, can improve pregnancy outcomes. Reported benefits of a
physical activity intervention include a reduction in gestational weight gain
and the risk of adverse pregnancy outcomes such as preeclampsia and gestational
diabetes mellitus (GDM). Such reviews have informed national and international
pregnancy care guidelines, with recommendations that pregnant women engage in
regular physical activity in pregnancy.

The principle of systematic reviews and meta-analyses is to
synthesise the available evidence, generating an aggregate effect estimate from
multiple studies. As part of this process, assessment of individual trial
quality is generally undertaken, commonly utilising the Cochrane Risk of Bias
tool 2.0, to assess risk of bias across multiple domains, including the impact
of missing outcome data. A central tenet of a rigorous randomised trial is the
inclusion of outcome data from all participants recruited and randomised in an
intention-to-treat (ITT) analysis. Failure to undertake an ITT analysis
introduces bias reflecting not only the extent of missing outcomes but also the
degree of “missingness”, which can be related to the nature of the
intervention. This is of particular relevance when considering physical
activity interventions, where the underlying characteristics and clinical
outcomes of those who comply are likely different from those who do not. Furthermore,
exclusion of participants who are non-compliant with an intervention limits the
generalisability of the findings into clinical practice. The aim was to conduct
a methodologically rigorous systematic review to evaluate physical activity
interventions in pregnancy on clinical pregnancy outcomes and to assess the
impact of including studies at increasing risk of bias on the validity of
recommendations for clinical practice guidelines.

This systematic review considered all randomised trials that
evaluated the effect of an antenatal physical activity intervention alone
(i.e., the intervention did not include dietary or behavioural components). Systematic
search of PubMed, PubMed Central, Ovid Medline, Embase, Cochrane Central
Register of Controlled Trials, and CINAHL from inception to 14 December 2023. Trial
quality was assessed using the Cochrane Risk of Bias tool. Independent of this,
studies were grouped based on degree of deviation from the intention to treat
principle. Sequential meta-analysis was performed in which greater degrees of
potential bias were allowed. Between intervention group comparisons used,
relative risks or mean differences with 95% confidence intervals for
dichotomous outcomes and continuous outcomes, respectively.

Overall, the quality of trial reporting was low. Only 5
trials (12.5%) were performed and analysed in keeping with the intention to
treat principle. When considering only those trials performed rigorously, there
was no evidence that antenatal physical activity improves pregnancy outcomes or
limits gestational weight gain (WMD −0.60kg; 95% CI −2.17, 0.98 WMD −0.60kg;
95% CI −2.17, 0.98).

Authors found no evidence that antenatal physical activity
is associated with improved pregnancy outcomes or reduced gestational weight
gain when considering only trials considered to have no or negligible risk of
bias. The effect of including trials with increasing potential for bias shifted
the estimates of effect further from the null while simultaneously introducing
greater heterogeneity.

Study findings are in contrast with those of published
systematic reviews and meta-analyses that have been used to inform national and
international guidelines for pregnancy care. In particular, recently published
systematic reviews and meta-analyses have reported that antenatal physical
activity interventions are associated with a reduction in gestational diabetes,
hypertensive disorders of pregnancy, and gestational weight gain. These
aggregate data meta-analyses included all trials, irrespective of individual
trial risk of bias or rigour.

Authors have demonstrated in review that a small minority of
randomised trials have been conducted in a methodologically rigorous way, the
majority of trials having substantial risk of bias, particularly in relation to
performing an ITT analysis. When considering only trials at no or negligible
risk of bias, there is no effect of antenatal physical activity interventions
on gestational weight gain or pregnancy and birth outcomes for women and their
infants. In contrast, the inclusion of trials with significant methodological
flaws results in misleading findings of benefit. Furthermore, almost half of
the trials they identified recruited fewer than 100 women. It is well known
that smaller studies often show different, commonly larger, treatment effects
than those reported by larger randomised trials. Smaller studies, particularly
those with fewer than 100 participants, have been shown to inappropriately skew
aggregate data meta-analysis results towards beneficial effects. This has
implications for evidence-based clinical practice guidelines that rely on
systematic reviews and meta-analyses suffering from artificially inflated
effect estimate.

Study findings cast doubt on the validity of national and
international clinical practice guidelines recommending antenatal physical
activity for improvement of pregnancy outcomes. A rigorous assessment of trial
quality is required prior to incorporation of evidence synthesis into such clinical
practice guidelines, as study found a significant proportion of included trials
were of poor quality and biased aggregate estimates of effect.

Source: Amanda J. Poprzeczny,
Andrea R. Deussen, Megan Mitchell; BJOG: An International
Journal of Obstetrics & Gynaecology, 2025; 0:1–15 https://doi.org/10.1111/1471-0528.18084