January 7, 2018
Dear Dr. Khan,
We are happy to see Dr. Bø et al.’s interest in the Society of Obstetricians and Gynecologists of Canada (SOGC)/Canadian Society for Exercise Physiology (CSEP) 2019 Canadian Guideline for Physical Activity throughout Pregnancy (1,2), and appreciate the opportunity to respond to their concerns. We understand these concerns to include: variations in the studies included in Davenport et al 2018 (3), our assessment of the overall quality of the evidence supporting the recommendation on pelvic floor muscle training (PFMT), values and preferences of pregnant women and other key stakeholders, and the methodology used to develop the recommendation on urinary incontinence in comparison to other reviews on the topic.
The Guideline Consensus Panel was struck to include key stakeholders in the fields of maternal and fetal health. This panel consisted of 19 representatives of the SOGC and CSEP, methodologists (AGREE II, GRADE, statistics, library science), researchers with expertise in prenatal exercise, frontline clinicians who care for pregnant women including physicians specializing in maternal/fetal medicine, obstetrics, family medicine and sports medicine, a midwife and a public health nurse. The input of pregnant women, as well as other experts (including pelvic health specialists), were solicited both formally and informally throughout the process. Indeed, we were grateful for the positive feedback of Dr. Dumoulin who reviewed our systematic review on urinary incontinence and PFMT prior to submission to BJSM (see Acknowledgement section) (3). In addition to broad expertise, this particular Guideline was developed utilizing the rigorous and transparent methodology outlined by the Grading of Recommendations Assessment, Development and Evaluation(GRADE) framework(4)and the Appraisal of Guidelines for Research & Evaluation II (AGREE II) instrument (5). The Guideline Consensus Panel chose this approach as it is considered the “gold standard” for guideline development.
A detailed explanation of the decisions informing the Guideline was included within the Guideline document, as well as a separate methodology paper (1,2,6). Briefly, the Guideline Consensus Panel ranked outcomes as “critical”, “important” or “not important” during a two day meeting in October 2015. Prior to the meeting, 10 pregnant women (stakeholders) from across Canada filled out a survey to rate the outcomes that were most important to them both on a Likert scale, and by open ended questions. Six women were multiparous (and thus more likely to have previously experienced urinary incontinence either pre- or post-natally as suggested by Dr. Bøand colleagues) and four nulliparous; none of them listed urinary incontinence as an outcome which would inform their decisions regarding prenatal exercise. Each proposed outcome was discussed by the expert panel with input from the pregnant women, followed by a vote to obtain consensus. Urinary incontinence was voted by the panel as an “important” but not “critical” outcome when making decisions regarding prenatal exercise. For context, some outcomes rated as “critical” were preeclampsia and miscarriage, and some rated as “not important” were likelihood of breastfeeding and newborn length. In total, 37 “critical” or “important” outcomes were considered in the development of the recommendations. It is important to highlight that it is unusual among GRADE informed guideline processes to make a specific recommendation for an outcome rated as “important” and not “critical”. However, the addition of a recommendation regarding urinary incontinence was prioritized when developing the Guideline recommendations based on the evidence identified in the systematic review (3).
Although other study designs were included in the systematic reviews, only RCTs informed the specific recommendations of the Guideline (6). Further, as the Guideline was specific to exercise, only RCTs including exercise as the intervention (named “exercise-only RCTs”) were included (note that standard care was not considered a co-intervention). Studies that included co-interventions (e.g., Stafne 2012 (7)) were included in our supporting systematic reviews (see Online Supplement Figure 1 (3)); however, as these studies provided indirect evidence they were not considered for the Guideline development.
The objective grading of evidence was overseen by three methodologists and confirmed by experts on the panel. While exercise-only RCTs reporting on development of urinary incontinence during pregnancy included “moderate” quality evidence, evidence on symptom severity was deemed to be of “low” quality due to downgrading for risk of bias (performance and attrition bias) and inconsistency (Online Supplement Table 2). In both of these cases, there were fewer than 2,000 women included in each forest plot which also influenced our rating. From a methodological perspective, the quality of the evidence is “low”, which means our confidence in the estimate of the effect is limited. Following GRADE, there are only 5 scenarios where a recommendation based on “low” quality evidence may be deemed a strong recommendation and the recommendation for urinary incontinence does not meet any of those scenarios (See GRADE handbook Table 6.3 (4)). Therefore, a weak recommendation is appropriate based on the “low” quality evidence.
The formal and informal feedback (including from pelvic floor specialists) from some stakeholders regarding pelvic floor muscle training (PFMT) was that there was concern regarding a broad recommendation to all pregnant women to engage in PFMT. Their sentiment was that women with tense pelvic floors may not benefit from this recommendation. In GRADE terms, this means that there is “uncertainty” and “variability” in terms of values and preferences of women regarding practicing PFMT which further supports a weak recommendation. Based on the “low” quality of empirical evidence, the ranking by our experts of urinary incontinence as an “important” outcome (not “critical”), and the uncertainty and variability of the feedback on the utility of the recommendation of PFMT for pregnant women, the Consensus Panel agreed on a weak recommendation. Based on the rigor of our methodology, we stand by this recommendation.
Dr. Bø and colleagues indicated that they were concerned that the Guideline may mislead pregnant women. We would like to point out there is a public facing version of the guidelines for pregnant women available from CSEP that does not mention strength of recommendations for this reason (available as a tear sheet and online:https://csepguidelines.ca/guidelines-for-pregnancy/). Regarding providers, this is the standard way of reporting guidelines and we have provided guidance as to how to interpret the information for those who may need it.
We would like to thank the authors for pointing out that we cited Mørkved 2007 (8) rather than 2003 (9). Both papers report on the same cohort of women, and we can confirm that the numbers included in the forest plot are correct. As mentioned above, Stafne 2012 (7) included a co-intervention (i.e. dietary counselling) and therefore provided indirect evidence and was not eligible to inform the recommendation in the Guideline. Regardless, even if this study was eligible for inclusion, our recommendation would not change based on the points described above.
In the current Guideline, Recommendation 4 suggests that “Pregnant women should incorporate a variety of aerobic and resistance training activities to achieve greater benefits. Adding yoga and/or gentle stretching may also be beneficial”. The language in this recommendation was chosen deliberately – the first sentence states “should” while the second says “may”. While the majority of studies included aerobic exercise alone or in combination with resistance training, a number of studies also examined yoga (primarily on mental health outcomes (10)), as well as stretching and PFMT. Rather than simply suggesting that women add PFMT to their aerobic and resistance training as was done with yoga and stretching, the panel chose to develop a recommendation specific to PFMT as outlined above based on the promising (but not without limitations) body of evidence. The panel agreed this specificity was the most rigorous representation of the available evidence.
Diastasis recti was considered a critical outcome by the Guideline Consensus Panel because the panel agreed that this condition could weaken the core musculature and potentially affect the woman’s ability to be physically active over the short and long term. In the Guideline document the paragraph regarding diastasis recti is included under the section of “Considerations for Implementation” and, as stated, is based on expert opinion of the Guideline Consensus Panel because there was no exercise-only RCTs available regarding this outcome. However, the statement that pregnant women with diastasis recti should avoid abdominal exercise (such as abdominal curl-ups) was not referring to postpartum women (this Guideline is specific to physical activity during pregnancy) but was based on a cautionary statement by Mota et al (2015) (11); “Caution: Only the immediate effect of the exercises was measured; therefore, these data cannot be used to suggest effectiveness of treatment as an intervention or as regular exercises”. Because of this statement and because of the lack of evidence, the Guideline Consensus Panel agreed that that caution is warranted until further evidence becomes available.
While the evidence regarding the safety and benefits regarding prenatal exercise (including many different types of exercise) is mounting, there still remain substantial gaps in the literature. This includes additional rigorous RCTs yielding high-quality evidence. We look forward to future studies from the authors’ respective groups (and others) which will inform more refined recommendations in future iterations of the Guideline.
Margie H Davenport, PhD, University of Alberta
Michelle F Mottola, PhD, Western University
Stephanie-May Ruchat, PhD, Université du Québec à Trois-Rivières
Gregory A Davies, MD, Queen’s University
Veronica J Poitras, PhD, Ottawa, ON
Casey E Gray, PhD, Children’s Hospital of Eastern Ontario Research Institute
Alejandra Jaramillo, MSc, Ottawa, ON
Nick Barrowman, PhD, Children’s Hospital of Eastern Ontario Research Institute
Kristi B. Adamo, PhD, University of Ottawa
Mary Duggan, CAE, Canadian Society of Exercise Physiology
Ruben Barakat, PhD, Universidad Politécnica de Madrid
Phil Chilibeck, PhD, University of Saskatchewan
Karen Fleming, MD, Sunnybrook Health Sciences Centre
Milena Forte, MD, Mount Sinai Hospital
Jillian Korolnek, RM, Toronto, Ontario
Taniya S Nagpal, BSc, Western University
Linda G Slater, MLIS, University of Alberta
Deanna Stirling, BScN, Middlesex-London Health Unit
Lori Zehr, PhD, Camosun College
- Mottola MF DM, Ruchat SM, Davies GA, Poitras VJ, Gray CE, Jaramillo A, Barrowman N, Adamo KB, Duggan M, Barakat R, Chilibeck P, Fleming K, Forte M, Korolnek J, Nagpal T, Slater L, Stirling D, Zehr L. 2019 Canadian Guideline for Physical Activity throughout Pregnancy. Br J Sports Med. 2018;52:1339-46.
- Mottola MF DM, Ruchat SM, Davies GA, Poitras VJ, Gray CE, Jaramillo A, Barrowman N, Adamo KB, Duggan M, Barakat R, Chilibeck P, Fleming K, Forte M, Korolnek J, Nagpal T, Slater L, Stirling D, Zehr L. 2019 Canadian Guideline for Physical Activity throughout Pregnancy. JOGC. 2018;40(11):1549-59.
- Davenport MH NT, Mottola MF, Skow R, Riske L, Poitras VJ, Garcia AJ, Gray CE, Barrowman N, Meah VL, Sobierajski F, James M, Nuspl M, Weeks A, Marchand AA, Slater LG, Adamo KB, Davies GA, Barakat R, Ruchat SM. . Prenatal exercise (including but not limited to pelvic floor muscle training) on and urinary incontinence during and following pregnancy: A systematic review and meta-analysis. Br J Sports Med. 2018;52:1397-1404.
- Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. 2011;64(4):383-94.
- Brouwers MC, Kho ME, Browman GP, Burgers JS, Cluzeau F, Feder G, et al. AGREE II: advancing guideline development, reporting, and evaluation in health care. Prev Med. 2010;51(5):421-4.
- Davenport MH RS, Mottola MF, Davies GA, Poitras VJ, Gray CE, Jaramillo A, Barrowman N, Adamo KB, Duggan M, Barakat R, Chilibeck P, Fleming K, Forte M, Korolnek J, Nagpal T, Slater L, Stirling D, Zehr L. 2019 Canadian Guideline for Physical Activity Throughout Pregnancy: Methodology. JOGC. 2018;40(11):1468-83.
- Stafne SN, Salvesen KA, Romundstad PR, Torjusen IH, Morkved S. Does regular exercise including pelvic floor muscle training prevent urinary and anal incontinence during pregnancy? A randomised controlled trial. BJOG. 2012;119(10):1270-80.
- Morkved S, Salvesen KA, Schei B, Lydersen S, Bo K. Does group training during pregnancy prevent lumbopelvic pain? A randomized clinical trial. Acta Obstet Gynecol Scand. 2007;86(3):276-82.
- Morkved S, Bo K, Schei B, Salvesen KA. Pelvic floor muscle training during pregnancy to prevent urinary incontinence: a single-blind randomized controlled trial. Obstet Gynecol. 2003;101(2):313-9.
- Davenport MH MA, Mottola MF, Skow R, Meah VL, Poitras VJ, Garcia AJ, Gray CE, Barrowman N, Riske L, Sobierajski F, James M, Nagpal T, Marchand AA, Nuspl M, Slater LG, Barakat R, Adamo KB, Davies GA, Ruchat SM. The impact of pre-natal exercise on both pre-natal and post-natal anxiety and depressive symptoms: a systematic review and meta-analysis. Br J Sports Med. 2018;52:1376-1385.
- Mota P PA, Carita AI, Bo K. The immediate effects on inter-rectus distance of abdominal crunch and drawing-in exercises during pregnancy and the postpartum. J Orthop Sports Phys Therapy. 2015;45(10):781-8.