Abstract
Purpose:
To determine feasibility of a randomized controlled trial (RCT) comparing postpartum rectus abdominis training with transversus abdominis training in reducing the inter-recti distance in patients with diastasis of the rectus abdominis muscles (DRAM).
Method:
A pilot, randomized controlled trial with longitudinal assessment following vaginal delivery at 6 and 12 weeks postpartum was completed. Forty-four women with DRAM (inter-recti distance of ≥ 30mm as measured by digital calipers) were recruited from a regional public hospital in Australia. The standard treatment group (n = 21) was provided with a 5-second transversus abdominis activation exercise in crook lying. The experimental group (n = 23) was provided with a 1-second rectus abdominis crunch exercise. Dosage was between 1–10 repetitions at baseline and between 1–20 at 6 weeks, twice a day, at least 5 days per week, prescribed depending on individual participant ability. The primary outcome measure was inter-recti distance. Feasibility measures were recruitment rate, loss to follow-up, adverse events, and exercise adherence.
Results:
The rectus abdominis group achieved greater reduction of the inter-recti distance at 6 weeks (at and below the umbilicus) and 12 weeks (above, below, and at the umbilicus). Recruitment rate was acceptable (45%) and there were no adverse events, but loss to follow-up was high due to COVID-19 impacts and participants did not return exercise adherence diaries.
Conclusion:
Prescription of rectus abdominis exercise during the early postpartum period following vaginal delivery resulted in a greater reduction of the inter-recti distance at 6 and 12 weeks when compared with transversus abdominis exercise. This pilot trial was impacted by high loss to follow up due to COVID-19 restrictions, but feasibility was otherwise acceptable. The findings of the study will inform future fully powered trials comparing these two exercise types in postpartum women with DRAM.
Key Words: diastasis, muscle, physical therapy, postpartum period, rectus abdominis, transversus abdominis
Résumé
Objectif :
déterminer la faisabilité d’une étude randomisée et contrôlée pour comparer l’entraînement des grands droits à l’entraînement des muscles transverses afin de réduire la distance intermusculaire chez les patientes ayant une diastase des grands droits (DGD).
Méthodologie :
étude pilote randomisée et contrôlée comportant une évaluation longitudinale six et 12 semaines après un accouchement vaginal. Au total, 44 femmes ayant une DGD (distance intermusculaire d’au moins 30 mm, mesurée à l’aide d’une règle à coulisse numérique) ont été recrutées dans un hôpital régional public de l’Australie. Le groupe sous traitement standard (n = 21) a reçu un exercice d’activation des muscles transverses de cinq secondes en position allongée sur le dos, les genoux pliés. Le groupe expérimental (n = 23) a reçu un exercice de demi-redressement assis des grands droits d’une seconde. La posologie de une à dix répétitions pour commencer et de une à 20 répétitions deux fois par jour la sixième semaine, au moins cinq jours par semaine, dépendait de la capacité de chaque participante. La mesure d’issue primaire était la distance entre les muscles. Les mesures de faisabilité étaient le taux de recrutement, la perte au suivi, les événements indésirables et l’adhésion aux exercices.
Résultats :
le groupe d’exercices des grands droits a réduit davantage la distance entre les muscles au bout de six semaines (à l’ombilic et sous l’ombilic) et de 12 semaines (au-dessus et au-dessous de l’ombilic et à l’ombilic). Le taux de recrutement était acceptable (45 %) et il n’y avait pas d’événements indésirables, mais la perte au suivi était élevée à cause des conséquences de la COVID-19, et les participantes n’ont pas remis leur journal d’adhésion aux exercices.
Conclusion :
la prescription d’exercices des grands droits au début de la période postnatale après un accouchement vaginal favorisait une plus grande réduction de la distance entre les grands droits au bout de six et 12 semaines que les exercices des muscles transverses. L’étude pilote a été touchée par une forte perte au suivi à cause des restrictions liées à la COVID-19, mais la faisabilité était autrement acceptable. Les résultats de l’étude éclaireront de futures études comportant un plein échantillonnage pour comparer deux types d’exercices postnatals chez des femmes présentant une diastase des grands droits.
Mots-clés : diastase, muscle, grand droit, période postnatale, physiothérapie, transverse de l’abdomen
Diastasis of the rectus abdominis muscles (DRAM) is a condition where the abdominal muscles separate along the linea alba.1 During pregnancy, hormones cause softening of connective tissues to allow for expansion of the abdominal wall and increased mobility in the pelvis in preparation for birth.2 Between 66%–100% of people have been reported as having experienced DRAM in their third trimester of pregnancy;2 however, the incidence of ongoing DRAM after birth and the best management approach are still not well understood.3
Hormone levels change rapidly postpartum; however, there are latent effects on the tissues for a period of approximately 12 weeks.4 Training programs aimed at reducing the space between the two rectus abdominis (RA) muscle bellies known as the inter-recti distance (IRD) have found that reduction in IRD plateaus at 8 weeks postpartum,4 with minimal changes being detected between measurements from 12–14 weeks postpartum to 6 months postpartum.5 By this stage, the IRD is likely to remain constant unless the person has future pregnancies and deliveries, which would then result in a further window of linea alba softening when the IRD could once again be addressed. Alternatively, surgical management is available.6 However, no further closure is expected to be made with conservative management alone.5 Treatment of the early postpartum patient with DRAM will therefore be different in the first 6–12 weeks postpartum when the focus of therapy will be on reducing the IRD while the linea alba remains malleable. After this time, functional strengthening typically becomes the aim of therapy.4
Almost 90% of physiotherapists who treat early postpartum patients with DRAM report using transversus abdominis (TrA) training as treatment.7 The TrA muscle attaches into the linea alba, while the RA muscle bellies run parallel to it. In the non-pregnant/late postpartum population when the linea alba is firm, contracting the TrA causes shortening of the corset-type structure providing stabilizing forces for the trunk.8 In the early postpartum population when the linea alba is more pliable, the contraction of the left and right TrA will cause shortening of the left and right insertion of TrA separately toward their respective origins, pulling the linea alba further apart and increasing the IRD.9 Alternatively, when RA contracts, the muscle bellies shorten, reducing the gap between them.1,10,11,12,13
Although the complications of DRAM are not well understood,14 there is conflicting literature suggesting a possible correlation with comorbidities, such as low back pain and incontinence. These have a significant effect on quality of life and represent a substantial burden to the health system.15,16 It is also a socially important issue given women’s reported dissatisfaction with the cosmetic effects of persistent DRAM.17
Preliminary data exist to demonstrate biomechanically that RA activation reduces the IRD from resting position and TrA activation increases the IRD when measured with ultrasound.1,10,11 No studies, though, investigating differences between the two, when used as treatments, could be located in existing literature.3,6 There is also evidence suggesting current management of TrA activation is no more effective than receiving no treatment in women from 6 weeks postpartum.18
In the absence of conclusive evidence for treatment with either TrA or RA, a co-contraction between the TrA muscles and the pelvic floor muscles has been cited as a putative reason to continue with TrA training.19 Yet, there is conflicting evidence that this co-contraction exists in the postpartum population.20,21 After initially detected in a sample of 7 participants,20 the study was repeated on 81 subjects where co-contraction was only found in nulliparous participants.21
This pilot study aimed to assess the feasibility of conducting a larger scale randomized controlled trial comparing early postpartum RA and TrA training programs for reducing the IRD in patients with DRAM, following vaginal delivery.
Methods
Design
A single-centre pilot randomized controlled trial was conducted with participants recruited from a regional public hospital in Australia. Patients were screened on the ward for DRAM of two finger widths or greater following vaginal delivery. If DRAM was detected at the level of the umbilicus or 2 cm above/below the umbilicus, details were forwarded to Physiotherapist A (a role shared between two senior physiotherapists) who gained consent, completed baseline measurements of IRD, and administered the Australian Pelvic Floor Questionnaire (APFQ). Eligible and consenting participants were allocated via concealed block randomization, in blocks of four, to one of the two intervention groups by Physiotherapist B (primary investigator), who had no contact with participants. Allocation was generated using a web-based random sequence generator and was stored in a password protected computer spreadsheet by Physiotherapist B. Patient name and group allocation were then given to Physiotherapist C who provided the allocated treatment on the ward. Physiotherapist A remained blinded to patient group allocation, while Physiotherapist C was blinded to IRD measurements and APFQ scores. Participants were blinded to the study hypothesis in that they were not educated on standard treatment versus experimental treatment differences to avoid influencing compliance or motivation. Trial recruitment included all eligible and consenting participants who gave birth at the hospital over a three-month period from November 2019 to January 2020 inclusive.
The standard treatment group received a TrA home exercise programme and the experimental group received a RA home exercise programme. In both groups, repetition dosage was individualized between 1–10 repetitions at baseline by Physiotherapist C based on each participant’s ability to maintain correct technique. Follow-up measurements of outcome variables were then completed at 6 and 12 weeks in an outpatient hospital setting. Physiotherapist A reassessed IRD measures and re-administered the APFQ at both follow-up appointments, while remaining blinded to treatment allocation. Participants were reminded at the beginning of both the follow-up appointments to not disclose their group allocation to the blind assessor. Physiotherapist B checked each participant’s exercise technique and progressed as appropriate to between 1–20 repetitions at the 6-week follow-up appointment. Minor exercise technique corrections were implemented for participants from both groups, where necessary, at the 6-week visit. Pilot trial outcomes are reported in accordance with CONSORT guidelines.22
Participants and centres
Participants were recruited following vaginal delivery as in-patients. As there is poor consensus across the literature on what constitutes a DRAM, for the purpose of this study an IRD of ≥ 30 mm at any measurement point was selected. Inclusion criteria were presence of a DRAM of ≥ 30 mm at one or more of the measurement points of 2 cm above/below/at the level of the umbilicus, vaginal delivery with or without instrument assistance after 37-weeks gestation within the past 72 hours, and age ≥ 18 years. Exclusion criteria consisted of delivery via caesarean section, delivery > 72 hours prior to assessment, presence of serious disease/pathology, or impaired cognition contraindicating involvement in an exercise program. Participants < 18 years of age at the time of birth were excluded due to higher vulnerability.
Interventions
Both treatment groups received a modified version of the hospital’s information booklet provided to women following vaginal birth. The modified version did not include the DRAM education pages which encourage TrA exercise. All women were provided with information on pelvic floor muscle care and exercise, toileting, and graded return to physical activity. As per hospital protocol, any woman who sustained a Grade 3 or 4 perineal tear was referred for outpatient pelvic floor physiotherapy.
Standard treatment group (TrA)
Participants were provided with detailed written information about DRAM and the prescribed TrA exercise. Participants were taught a TrA-activation exercise individually in their hospital room on a flat bed without a pillow; this was completed in a crook lie position. Participants were asked to contract their TrA muscle by a senior physiotherapist using appropriate cues until contraction was achieved and could be reproduced and palpated by the patient without cueing from the therapist. Examples of cueing for both groups are outlined in table 1. Contractions were held for five seconds per repetition, with the number of repetitions prescribed based on ability to sustain this contraction with correct technique.
Table 1.
Examples of Cueing for Each Group
| Transversus abdominis (standard treatment) group | Rectus abdominis (experimental) group |
|---|---|
| Pull your belly button up and toward your spine | Initiate lifting your head from the bed |
| Think about trying to zip up tight jeans | Think about lifting your head ½ way up from a pillow |
| Squeeze the front of your pelvis together | Don’t let the area under your fingers start to dome out (with patient palpation over linea alba) |
| Tighten the muscle under my/your fingers (with therapist and patient palpation) |
Experimental group (RA)
The experimental group was also provided with written information about DRAM and their RA exercise protocol. In the same conditions and position as described above, the experimental group was taught a crunch movement involving a small head lift off the bed with a one second hold to activate RA. Participants were educated on correct technique as well as indicators of fatigue or poor technique, such as doming. Once again, dosage was individualized based on participant exercise quality.
For both groups, dosage commenced between 1–10 repetitions, twice a day at least 5 days a week. A median number of 7 repetitions for each group were prescribed at baseline. A paper exercise diary was provided with these details and instructions to monitor compliance and return the diary at the 6-week follow-up appointment. At this appointment, dosage was re-assessed and prescribed between 1–20 repetitions as appropriate. A median number of 10 repetitions was prescribed in the TrA group and 15 in the RA group at this point. Each exercise was continued for the 12-week duration of the follow-up period. The duration of contractions remained constant throughout the trial. If a participant from either group had a DRAM ≥ 70 mm wide at any measurement point, they were offered Tubigrip compression that had to always be worn, except during hygiene routines. Only one woman (a member of the RA group) had a DRAM > 70 mm and she declined this offer.
The difference in time for repetitions between groups was due to the difference in activation speed of TrA and RA. TrA takes > 2 seconds to activate in the prescribed crook lie position,23 so asking participants to hold for 5 seconds may have resulted in the final contraction being held for 3 seconds. Whereas previous RA curl-up exercise protocols have used a 1-second concentric phase.11,24 In practice, cueing a 1-second RA activation resulted in approximately a 1-second isometric hold with both a 1-second concentric and eccentric phase totaling an approximately 3-second repetition. The slower recruitment and relaxation of the TrA muscle could reasonably mean that despite the difference in instruction, both groups were presumed to be completing similar isometric holds. It is acknowledged that the difference in time taken to complete the prescribed dosages may be a factor influencing motivation and compliance.
Outcome measures
At baseline, demographic data, including age, parity, IRD, and APFQ scores, were collected. The IRD was measured in millimeters at 3 points along the linea alba — at the level of the umbilicus and 2 cm above/below the level of the umbilicus. Measurements were taken using the mean score of 3 measurements at each point along the linea alba using digital calipers with accuracy to 2 decimal places of a millimeter.25 Participants were asked to lie supine without a pillow and with hands by their sides. Participants were then asked to lift their heads just enough to activate RA to a point of it being palpable and distinguishable from the linea alba. Participants completed the APFQ independently at baseline, and at 6 and 12-week follow up, and could ask for clarification at any time from the therapist.
Primary outcomes
High inter- and intra-rater reliability of calipers using a repeated measures technique has previously been established.26 In line with these results, the same therapist who measured each participant at baseline then completed each of the follow-up IRD measures at 6 and 12 weeks for that participant. The primary outcome measure was the millimeter change in IRD from baseline to 6 weeks and 12 weeks postpartum.
Secondary outcomes
The APFQ is a validated outcome measure sensitive to detecting changes in pelvic floor function across four domains — bladder, bowel, prolapse, and sexual function, with higher scores indicating increasing severity of symptoms.27 The APFQ was used to screen for differential effects of each intervention on pelvic floor function.
Feasibility measures
Descriptive statistics were used to assess feasibility, with the following thresholds: recruitment rate 50% of all births, participant loss to follow-up < 15%, no serious adverse events (requiring medical intervention), exercise adherence > 75%.
Sample size
A recruitment period of 3 months was deemed an achievable time frame to collect preliminary data for this pilot study, assuming a DRAM incidence of 60%.14 Based on historical hospital data we estimated that 100 women would deliver at the hospital in this timeframe, with an eligibility rate of 50% (n = 50 participants). There was no a priori sample size calculation, but recruitment was to cease at 3 months or n = 50, whichever occurred first. We acknowledged that the study would likely be underpowered to detect anything other than large between-group effects (standardized mean difference SMD > 0.8), noting moderate effects (SMD > 0.5) would likely still be clinically important.
Due to SARS-CoV-2-related restrictions introduced during the study period, it was decided to cease follow up in line with hospital policy at the time. As a result, loss to follow up was substantial at 6-week follow up and particularly at 12-week follow up (Figure 1).
Figure 1.
Flow of participants through the trial.
Data analysis
Pilot data were analyzed using SPSS, with a view to assessing effect size and variability. Linear mixed models were used to evaluate between-group effects on primary and secondary outcomes, accounting for the repeated measures design and adjusting for baseline scores. Missing data were managed via maximum likelihood estimation within the linear mixed models,28 allowing data from all participants to contribute to outcome modelling in their randomized groups even when lost to follow-up, thus satisfying intention to treat analysis principles.29
Results
Out of 97 women who delivered vaginally at the hospital during the recruitment period, 44 were eligible and provided consent to participate. Twenty-one participants were allocated to the standard treatment (TrA) group and 23 participants to the experimental (RA) group. Participant flow is outlined in Figure 1. Group characteristics were balanced at the commencement of the study (Table 2) and remained balanced at 6 weeks following some early participant attrition. There was no further voluntary participant attrition by 12 weeks; however, all follow-up visits scheduled on or after March 23, 2020 were cancelled due to hospital-determined SARS-CoV-2 restrictions. Each participant was contacted via phone and provided with education and the opportunity to ask questions, as would have occurred in the face-to-face session. The cancelled follow up appointments were all 12-week appointments so exercise progressions were not impacted. Due to the high rate of missing data that needed to be accounted for in analysis, amendments were made to the data analysis section of the registered protocol based on statistician advice.
Table 2.
Baseline Group Characteristics for Participants at Enrolment
| TrA (n = 21)Mean (SD) | RA (n = 23)Mean (SD) | |
|---|---|---|
| Age (Y) | 31 (4.5) | 30 (4.5) |
| Parity | 1.8 (0.6) | 1.7 (0.9) |
| IRD 2 cm above umbilicus (mm) | 35.0 (7.5) | 36.1 (14.7) |
| IRD level of umbilicus (mm) | 34.3 (7.7) | 39.6 (14.8) |
| IRD 2 cm below umbilicus (mm) | 24.7 (9.0) | 28.7 (14.9) |
| APFQ bladder section (0–45) | 7.6 (4.5) | 87.7 (6.1) |
| APFQ bowel section (0–34) | 64.8 (3.9) | 5.1 (4.4) |
| APFQ prolapse section (0–15) | 0.7 (1.6) | 0.7 (0.2) |
| APFQ sexual function section (0–21) | 1.8 (2.2) | 2.1 (3.6) |
| APFQ total score (0–115) | 14.6 (7.9) | 15.6 (14.1) |
Y, years; mm, millimeters; RA, rectus abdominis; TrA, transversus abdominis; SD = standard deviation, IRD = inter-recti distance, APFQ = Australian Pelvic Floor Questionnaire.
Effect of intervention
Primary outcomes
There was a difference in reduction of the IRD (mm) between groups at 6 weeks at the level of the umbilicus and 2 cm below the umbilicus, with the RA group achieving 8.3 mm (95% CI 1.7 to 14.9 mm) and 6.3 mm (95% CI 0.1 to 12.4 mm,) greater reduction respectively. The measurement point 2 cm above the umbilicus showed a trend which again favoured the RA group with 5.0 mm (95% CI −2.2 to 12.2 mm) greater reduction than the TrA group.
At 12 weeks, the RA group achieved greater reduction of the IRD at all 3 measurement points with between-group differences above the umbilicus of 19.2 mm (95% CI 9.1 to 29.3 mm), at the level of the umbilicus 22.9 mm (95% CI 13.6 to 32.1 mm) and 2 cm below the level of the umbilicus 20.2 mm (95% CI 11.5 to 28.8 mm). Each group’s result and between-group differences are outlined in Table 3.
Table 3.
Primary Outcome: Inter-Recti Distance
| Outcome | n (TrA/RA) | Unadjusted means (SE) | Adjusted between-group difference (95% CI)* | |
|---|---|---|---|---|
| TrA | RA | |||
| Above umbilicus (mm) | ||||
| Baseline | 21/23 | 35.0 (1.7) | 36.1 (3.1) | |
| 6 weeks | 11/16 | 26.8 (2.6) | 24.0 (2.1) | −5.0 (−12.2 to 2.2) |
| 12 weeks | 8/4 | 23.4 (2.2) | 14.0 (3.9) | −19.2 (−29.3 to −9.1) |
| At umbilicus (mm) | ||||
| Baseline | 21/23 | 34.3 (1.7) | 39.6 (3.1) | |
| 6 weeks | 11/16 | 27.8 (2.4) | 27.6 (2.3) | −8.3 (−14.9 to −1.7) |
| 12 weeks | 8/4 | 26.0 (1.5) | 18.6 (5.4) | −22.9 (−32.1 to −13.6) |
| Below umbilicus (mm) | ||||
| Baseline | 21/23 | 24.7 (2.0) | 28.7 (3.1) | |
| 6 weeks | 11/16 | 21.3 (2.8) | 21.3 (2.1) | −6.3 (−12.4 to −0.1) |
| 12 weeks | 8/4 | 19.5 (1.8) | 14.3 (6.7) | −20.2 (−28.8 to −11.5) |
Between group differences are adjusted for baseline scores on the outcome. SE = standard error.
Secondary outcomes
At 6 weeks, a between-group difference in the APFQ bladder sub-scale scores was evident, with the TrA group scoring 2.1 points (95% CI 0.00 to 4.2) lower out of the possible total of 45 points, where a lower score represents better bladder function. Differences detected between the bowel, prolapse, and sexual function sub-scales of the APFQ at 6 and 12 weeks did not show a convincing trend in either direction. These outcomes are detailed in Table 4.
Table 4.
Secondary Outcome: Australian Pelvic Floor Questionnaire Scores
| Outcome | n (TrA/RA) | Unadjusted means (SE) | Adjusted between-group difference (95%CI)* | |
|---|---|---|---|---|
| TrA | RA | |||
| Total APFQ score | ||||
| Baseline | 21/23 | 14.6 (1.7) | 15.6 (2.9) | |
| 6 weeks | 11/17 | 8.7 (2.1) | 11.4 (2.7) | 1.9 (−2.4 to 6.3) |
| 12 weeks | 8/4 | 6.1 (1.7) | 21.5 (10.5) | −0.4 (−6.6 to 5.9) |
| Bladder (/45) | ||||
| Baseline | 21/23 | 7.6 (1.0) | 7.7 (1.3) | |
| 6 weeks | 11/17 | 2.9 (0.8) | 5.7 (1.4) | 2.1 (0.0 to 4.2) |
| 12 weeks | 8/4 | 1.1 (0.4) | 10.3 (6.5) | 2.1 (−0.9 to 5.1) |
| Bowel (/34) | ||||
| Baseline | 21/23 | 4.8 (0.8) | 5.1 (0.9) | |
| 6 weeks | 11/17 | 4.9 (1.3) | 3.5 (0.7) | −1.2 (−3.2 to 0.8) |
| 12 weeks | 8/4 | 3.0 (0.6) | 4.8 (1.7) | −1.7 (−4.6 to 1.1) |
| Prolapse (/15) | ||||
| Baseline | 21/23 | 0.7 (0.3) | 0.7 (0.5) | |
| 6 weeks | 11/17 | 2.3 (1.7) | 0.9 (0.6) | −1.3 (−3.7 to 1.0) |
| 12 weeks | 8/4 | 0.5 (0.4) | 2.8 (1.0) | 0.4 (−3.0 to 3.7) |
| Sexual (/21) | ||||
| Baseline | 17/22 | 1.8 (0.5) | 2.1 (0.8) | |
| 6 weeks | 2/7 | 1.5 (1.5) | 1.8 (1.4) | −0.7 (−2.3 to 0.9) |
| 12 weeks | 6/4 | 2.0 (1.6) | 3.8 (2.2) | −0.5 (−2.0 to 1.0) |
Between-group differences are adjusted for baseline scores on the outcome. SE = standard error.
Feasibility measures
The recruitment rate was 45%, with women only excluded due to insufficient DRAM (n = 23) or declining to consent (n = 30). The loss to follow up was 36% at 6 weeks and 73% at 12 weeks attributable predominately to COVID-19 impacts. No adverse events were reported throughout the trial; however, as a precaution, any participant expressing concern over pelvic floor function was referred on for assessment with a pelvic health physiotherapist. These referrals were evenly distributed between RA (n = 3) and TrA (n = 3) groups. Although exercise diaries were supplied, few were returned in either group. Over 80% of participants in each group anecdotally reported meeting the prescribed exercises dosages.
Discussion
The results of this pilot study provide justification for the establishment of a larger scale investigation comparing the efficacies of RA and TrA exercises for reducing IRD in early postpartum patients. Medium-to-large effect sizes (SMDs), along with confidence intervals that excluded zero on several outcomes tend to support the prospect that a home exercise program consisting of a RA crunch exercise may be more effective than TrA exercise for IRD reduction at both 6 and 12 weeks. However, a larger scale study is required to confirm or refute these interim findings given the high rate of loss to follow-up due to COVID-19 interruptions to the trial. These pilot study outcomes are, nonetheless, consistent with previous ultrasound-based studies (using imaging to assess immediate and reversible effects of TrA versus RA activation on IRD) which found that TrA increased IRD, while RA reduced it.1,10,11
Of these previous studies, one included a further outcome measure of linea alba distortion and found that activation of TrA caused less distortion of the linea alba when compared with activation of RA and proposed that while RA activation did decrease IRD, TrA activation may still have a functionally more important role in force transfer across the abdominal wall.11 While these are important findings, it should be noted that the previous study was conducted on patients outside of the early postpartum period after which the linea alba is hypothesized to have returned to a firmer consistency where further IRD decrease is not expected to occur.4 This represents a different patient population with different goals of treatment.
In the first 6 weeks postpartum, there was a difference between groups in bladder symptoms favouring the TrA group. Although statistically significant, the confidence interval was wide and almost included zero, and the effect was not sustained at 12 weeks. However, to be sure that performing RA rather than TrA exercises in this population is not detrimental to pelvic floor function, this outcome should be measured again in any future replication studies.
Confirmation of this pilot study’s trends with a larger sample would question the appropriateness of the current practice of prescribing TrA rather than RA exercises for DRAM in the early postpartum patient. The between-group effects and associated standard errors will be useful for estimating the required sample size to ensure the next trial can be fully powered to detect significant effects for hypothesis testing.
The head-to-head study design could not determine whether RA exercises are truly responsible for IRD reduction, or alternatively if TrA exercises may instead be detrimental to natural recovery. It is also not known whether outcomes would be equivalent or better if both RA and TrA exercises were prescribed together.
Pilot study limitations
Although the pilot study originally expected to retain a larger percentage of participants by 12 weeks, SARS-CoV-2 impacts on staffing availability and patient availability resulted in the loss of several participants from both groups. The study follow up eventually needed to end early due to the introduction of restrictions related to SARS-CoV-2. This does not represent an innate methodological flaw that could have been predicted, avoided, or accounted for in planning. The intention-to-treat analysis using linear mixed models with maximum likelihood estimation to account for missing data is considered suitable for managing large amounts of missing data.30,31
Feasibility measures identified avenues for improving the recruitment rate and verification of exercise adherence. A high rate (40%) of eligible women declined to participate in the study, which could be improved through refining participant information sheets and researcher scripts when describing the study to potential participants. Using smart-phone apps for promoting and measuring exercise adherence may also be more successful than paper-based diaries.
Conclusion
This pilot study provides evidence that prescribing RA activation exercises may be more effective than the current standard practice of prescribing TrA exercises for reducing IRD in early postpartum patients with DRAM. A fully powered trial is feasible when COVID-19 public health restrictions ease, but strategies for improving recruitment rate and documenting exercise adherence should be considered.
Key Messages
1) What is already known on this topic?
When measured on ultrasound, RA activation has an immediate and reversable effect of reduction in the IRD in women with DRAM, whereas TrA activation widens it. However, the effect when used as an exercise program in the 12 weeks following vaginal delivery is not yet clear.
2) What this pilot study adds
This pilot study challenges the existing accepted treatment of TrA exercise as the gold standard for reducing the IRD within the first 12 weeks postpartum. In addition, the pilot study has established a viable protocol for a larger-scale RCT to address this question more comprehensively.
References
- 1.Mota P, Pascoal AG, Carita AI, et al. The immediate effects on inter-rectus distance of abdominal crunch and drawing-in exercises during pregnancy and the postpartum period. J Orthop Sports. 2015;45(10):781–8. 10.2519/jospt.2015.5459. Medline: [DOI] [PubMed] [Google Scholar]
- 2.Thabet AA, Alshehri MA. Efficacy of deep core stability exercise program in postpartum women with diastasis recti abdominis: a randomised controlled trial. J Musculoskelet Neuronal Interact. 2019;19(1):62. [PMC free article] [PubMed] [Google Scholar]
- 3.Michalska A, Rokita W, Wolder D, et al. Diastasis recti abdominis: a review of treatment methods. Ginekol Pol. 2018;89(2):97–101. 10.5603/gp.a2018.0016. Medline: [DOI] [PubMed] [Google Scholar]
- 4.Coldron Y, Stokes MJ, Newham DJ, et al. Postpartum characteristics of rectus abdominis on ultrasound imaging. Man Ther. 2008;13(2):112–21. 10.1016/j.math.2006.10.001. Medline: [DOI] [PubMed] [Google Scholar]
- 5.Mota P, Pascoal AG, Carita AI, et al. Normal width of the inter-recti distance in pregnant and postpartum primiparous women. Musculoskelet Sci Pract. 2018;35:34–7. 10.1016/j.msksp.2018.02.004. Medline: [DOI] [PubMed] [Google Scholar]
- 6.Benjamin DR, Van de Water ATM, Peiris CL. Effects of exercise on diastasis of the rectus abdominis muscle in the antenatal and postnatal periods: a systematic review. Physiotherapy. 2014;100(1):1–8. 10.1016/j.physio.2013.08.005. Medline: [DOI] [PubMed] [Google Scholar]
- 7.Keeler J, Albrecht M, Eberhardt L, et al. Diastasis recti abdominis: a survey of women’s health specialists for current physical therapy clinical practice for postpartum women. J Women’s Health Phys Therapy. 2012;36(3):131–42. 10.1097/jwh.0b013e318276f35f. [DOI] [Google Scholar]
- 8.Pulkovski N, Mannion AF, Caporaso F, et al. Ultrasound assessment of transversus abdominis muscle contraction ratio during abdominal hollowing: a useful tool to distinguish between patients with chronic low back pain and healthy controls? Eur Spine J. 2012;21(6):750–9. 10.1007/s00586-011-1707-8. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Theodorsen NM, Strand LI, Bø K. Effect of pelvic floor and transversus abdominis muscle contraction on inter-rectus distance in postpartum women: a cross-sectional experimental study. Physiotherapy. 2019;105(3):315–20. 10.1016/j.physio.2018.08.009. Medline: [DOI] [PubMed] [Google Scholar]
- 10.Sancho MF, Pascoal AG, Mota P, et al. Abdominal exercises affect inter-rectus distance in postpartum women: a two-dimensional ultrasound study. Physiotherapy. 2015;101(3):286–91. 10.1016/j.physio.2015.04.004. Medline: [DOI] [PubMed] [Google Scholar]
- 11.Lee D, Hodges PW. Behavior of the linea alba during a curl-up task in diastasis rectus abdominis: an observational study. J Orthop Sports. 2016;46(7):580–9. 10.2519/jospt.2016.6536. Medline: [DOI] [PubMed] [Google Scholar]
- 12.Pascoal AG, Dionisio S, Cordeiro F, et al. Inter-rectus distance in postpartum women can be reduced by isometric contraction of the abdominal muscles: a preliminary case–control study. Physiotherapy. 2014;100(4):344–8. 10.1016/j.physio.2013.11.006. Medline: [DOI] [PubMed] [Google Scholar]
- 13.Liaw LJ, Hsu MJ, Liao CF, et al. The relationships between inter-recti distance measured by ultrasound imaging and abdominal muscle function in postpartum women: a 6-month follow-up study. J Orthop Sports. 2011;41(6):435–43. 10.2519/jospt.2011.3507. Medline: [DOI] [PubMed] [Google Scholar]
- 14.Sperstad JB, Tennfjord MK, Hilde G. Diastasis recti abdominis during pregnancy and 12 months after childbirth: prevalence, risk factors and report of lumbopelvic pain. Br J Sports Med. 2016;50(17):1092–6. 10.1136/bjsports-2016-096065. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Benjamin DR, Frawley HC, Shields N, et al. Relationship between diastasis of the rectus abdominis muscle (DRAM) and musculoskeletal dysfunctions, pain and quality of life: a systematic review. Physiotherapy. 2019;105(1):24–34. 10.1016/j.physio.2018.07.002. Medline: [DOI] [PubMed] [Google Scholar]
- 16.Gitta S, Magyar Z, Palancsa M, et al. Prevalence, potential risk factors, sequelae of diastasis recti abdominis and the treatment. Value Health. 2016;19(7):A605. 10.1016/j.jval.2016.09.1488. [DOI] [Google Scholar]
- 17.Saariniemi KM, Salmi AM, Peltoniemi HH, et al. Abdominoplasty improves quality of life, psychological distress, and eating disorder symptoms: a prospective study. Plast Surg Int. 2014;2014. 10.1155/2014/197232. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gluppe SL, Hilde G, Tennfjord MK, et al. Effect of a postpartum training program on the prevalence of diastasis recti abdominis in postpartum primiparous women: a randomized controlled trial. Phys Ther. 2018;98(4):260–8. 10.1093/ptj/pzy008. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gitta S, Magyar Z, Tardi P, et al. How to treat diastasis recti abdominis with physical therapy: a case report. J Diseases. 2016;3(2):16–20. 10.18488/journal.99/2016.3.2/99.2.16.20. [DOI] [Google Scholar]
- 20.Sapsford RR, Hodges PW, Richardson CA, et al. Co-activation of the abdominal and pelvic floor muscles during voluntary exercises. Neurourol Urodyn. 2001;20(1):31–42. . [DOI] [PubMed] [Google Scholar]
- 21.Pereira LC, Botelho S, Marques J, et al. Are transversus abdominis/oblique internal and pelvic floor muscles coactivated during pregnancy and postpartum? Neurourol Urodyn. 2013;32(5):416–19. 10.1002/nau.22315. Medline: [DOI] [PubMed] [Google Scholar]
- 22.Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Ann Intern Med. 2010;152(11):726–32. 10.7326/0003-4819-152-11-201006010-00232. Medline: [DOI] [PubMed] [Google Scholar]
- 23.Selkow NM, Eck MR, Rivas S. Transversus abdominis activation and timing improves following core stability training: a randomized trial. Int J Sports Phys Ther. 2017;12(7):1048–56. 10.26603/ijspt20171048. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Escamilla RF, Lewis C, Bell D, et al. Core muscle activation during Swiss ball and traditional abdominal exercises. J Orthop Sports. 2010;40(5):265–76. 10.2519/jospt.2010.3073. Medline: [DOI] [PubMed] [Google Scholar]
- 25.Benjamin DR, Frawley HC, Shields N, et al. Establishing measurement properties in the assessment of inter-recti distance of the abdominal muscles in a postnatal women. Musculoskelet Sci Pract. 2020;49:102202. 10.1016/j.msksp.2020.102202. Medline: [DOI] [PubMed] [Google Scholar]
- 26.Van de Water ATM, Benjamin DR. Measurement methods to assess diastasis of the rectus abdominis muscle (DRAM): a systematic review of their measurement properties and meta-analytic reliability generalisation. Man Ther. 2016;21:41–53. 10.1016/j.math.2015.09.013. Medline: [DOI] [PubMed] [Google Scholar]
- 27.Baessler K, O’Neill SM, Maher CF, et al. A validated self-administered female pelvic floor questionnaire. Int Urogynecol J. 2010;21(2):163–72. 10.1007/s00192-009-0997-4. Medline: [DOI] [PubMed] [Google Scholar]
- 28.Liu M, Wei L, Zhang J. Review of guidelines and literature for handling missing data in longitudinal clinical trials with a case study. Pharm Stat. 2006;5(1):7–18. 10.1002/pst.189. Medline: [DOI] [PubMed] [Google Scholar]
- 29.McCoy CE. Understanding the intention-to-treat principle in randomized controlled trials. West J Emerg Med. 2017;18(6):1075–8. 10.5811/westjem.2017.8.35985. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hamer RM, Simpson PM. Last observation carried forward versus mixed models in the analysis of psychiatric clinical trials (editorial). Am J Psychiatry. 2009;166(6):639–41. 10.1176/appi.ajp.2009.09040458. Medline: [DOI] [PubMed] [Google Scholar]
- 31.Krueger C, Tian L. A comparison of the general linear mixed model and repeated measures ANOVA using a dataset with multiple missing data points. Biol Res Nurs. 2004;6(2):151–7. 10.1177/1099800404267682. Medline: [DOI] [PubMed] [Google Scholar]