Diastasis recti abdominis (DRA) is a common problem in child-bearing women caused by increased physical forces on the abdominal wall due to fetal growth and organ migration, and secondarily by changes in the elasticity of connective tissue due to hormonal changes during pregnancy1-3. DRA results in abdo-minal muscle weakness due to an overall increase in the length of the core muscles and soft tissues that make up the abdominal wall and not just an increase in the distance between the rectus abdominis muscles4,5. This hinders the abdominal wall’s ability to maintain posture, stabilize the lumbar spine and pelvis, and has been linked to postpartum low back pain, stress urinary incontinence, pelvic pain, and cosmetic problems6-8.
Conservative management, including exercise programs, abdominal splinting, and manual therapy, as well as various exercise strategies such as transverse abdominal muscle training, pelvic floor muscle activation, Pilates, and core muscle training, have been attempted to prevent and treat DRA9. Studies of the effect of exercise therapy on the management of DRA have reported an average of 0.42 to 1.02 cm of additional inter-recti distance (IRD) reduction in the exercise group compared to the non-exercise group10,11. In most cases, exercise therapy is provided in the form of 1:1 offline lessons. However, due to the lack of time postpartum women have because of childcare demands, compliance is poor, and the high cost makes it unaffordable.
Meanwhile, the use of mobile apps in physical therapy and exercise therapy has been increasing in recent years. In the case of geriatric cardiac rehabilitation, there are reports of the effectiveness of home-based mobile-guided rehabilitation12, and telerehabilitation is also being tried for stroke patients13. If proven effective, telerehabilitation could be a great alternative to traditional exercise therapy for certain conditions. We hypothesize that home-based, mobile-guided exercise therapy will improve IRD, which is a diagnostic marker of DRA, as well as ultrasound elastic modulus and patient-reported outcomes (PROs) related to low back pain and urinary incontinence.
This was a prospective, observational study. We decided to recruit a minimum of 65 participants, which is 1.5 times the reference number, to include a sufficiently large number of participants to exceed the number of participants in studies with comparable designs10. From March to June 2023, postpartum women were recruited after distributing brief educational materials about DRA along with informational content through online communities and social media. The inclusion criteria were women between the ages of 25 and 35 years who underwent only a single vaginal delivery and were 12 weeks to 12 months postpartum. In addition, we excluded patients with a history of cesarean section or abdominal surgery, those who were currently pregnant, those with a history of ventral or umbilical hernia, and those who had difficulty performing exercises because of arthritis of the extremities or intervertebral disc herniation.
A total of 270 women wanted to participate in the study, but 157 did not meet the inclusion criteria (Fig. 1). Thus, 67 participants completed the protocol. This study was approved by the Institutional Review Board of Korea University Guro Hospital (No. 2022GR0291). Written informed consent was obtained from all participants.
Fig. 1. Participant flow diagram.
The following information was collected from the 67 women in the study: age, height, weight, body mass index, chest circumference, abdominal circumference, and delivery information (birth of a child weighing >4 kg, second stage of labor lasting more than an hour, and perineal laceration >2nd grade). In addition, previous incontinence-related symptoms were assessed using the International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form (ICIQ-UI-SF), and back pain-related symptoms were assessed using the Quebec Back Pain Disability Scale (QBPDS). The demographic data of the participants is shown in Table 1.
Table 1 . Demographic characteristics
| Characteristic | Value |
|---|---|
| No. of patients | 67 |
| Age (yr) | 31.6±2.3 |
| Height (cm) | 164.0±5.2 |
| Weight (kg) | 63.0±9.5 |
| Body mass index (kg/m2) | 23.3±2.9 |
| Chest circumference (cm) | 80.0±5.9 |
| Waist circumference (cm) | 84.6±9.4 |
| Information related to delivery | |
| Fetal macrosomia >4 kg | 2 (2.98) |
| 2nd stage of labor >1 hr | 21 (31.34) |
| Perineal laceration >2nd grade | 9 (13.43) |
| ICIQ-UI-SF | 5.0±5.3 |
| QBPDS | 18.0±12.9 |
Values are presented as number only, mean±standard deviation, or number (%).
ICIQ-UI-SF: International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form, QBPDS: Quebec Back Pain Disability Scale.
All the participants underwent a one-time offline, in-person education session before starting the online exercise program. The educational session covered the definition of DRA, its effects, and the importance of recovery. The participants then began the exercise program using the mobile application, which was conducted daily for 8 weeks. One to two exercises and coaching videos were provided daily, resulting in 56 sessions of 20 to 30 minutes over the course of the program. The first video helped participants warm up by including deep breathing, stretching the pelvic girdle in a supine position, stretching the anterior abdominal muscles in a prone position, and stretching the lower extremity muscles, while the second video began with simple strength training exercises and ended with a massage for a cool-down. Given that the participants were postpartum, the intensity of the exercises was guided to a rate of perceived exertion (RPE) of 3 to 4 using the Modified Borg Scale. To accommodate the participants’ daily routines, the exercises were simpler on weekends than on weekdays, and the intensity of the exercises increased towards the end of the protocol. In order to measure compliance as much as possible in a home-based exercise program, participants’ video playing time and whether they played the videos were stored in a database and used to assess compliance. To encourage participation, after the daily exercise videos were played, participants were asked to submit daily feedback notes through the application, which were then answered by an exercise therapist specializing in postpartum recovery exercises. The user interface of the mobile application and a video capture are shown in Fig. 2.
Fig. 2. (A) The user interface of the mobile application. (B) The capture of the provided workout video.
Our exercise program is divided into three main categories: core muscle activation, core muscle stretching, and myofascial release. The exercises are designed to utilize tools that are easily accessible, such as wrappers, massage balls, gym balls, and chairs, and can be performed in different positions, such as supine, prone, or side plank, depending on the muscle being targeted. Supplementary Table 1 shows the structure of the exercise program.
First, ultrasound was used to measure the IRD and shear wave elastography (SWE) at the first visit for the initial educational session and the second visit after 8 weeks of exercise. Ultrasound measurements were performed by a single radiologist with more than 10 years of experience in musculoskeletal imaging. Before the measurement, the patient was placed on a bed, and the level was marked 3 cm above and below the umbilicus at rest. The patient was examined in both the supine rest and head elevation positions, and measurements were taken after a deep breath to facilitate contraction of the rectus abdominis muscle14. For each position and condition, three measurements were taken, and the average value was used for analysis. Similarly, SWE was measured three times at rest and in the head-up position, 3 cm above and below the umbilicus, and averaged. For head elevation, a stadiometer was used to elevate the head by the same amount for each measurement to create a scapular clearance and ensure a clear field of view to measure the IRD. The ultrasonic measurement is shown in Fig. 3. Intraobserver reliability was evaluated by intraclass correlation coefficient model 2 (ICC [2, 1]). PROs reflecting symptoms of back pain and urinary incontinence, which are commonly thought to be associated with DRA, were also measured at baseline and post-intervention. The ICIU-UI-SF15 and QBPDS16 were used for incontinence and back pain, respectively.
Fig. 3. Inter-recti distance measured by ultrasound, 3 cm above the umbilicus (A) at baseline and (B) after 8 weeks of intervention.
The collected data were stored on an encrypted computer accessible only to the statisticians. There were no missing values, except for four participants who dropped out of the study. In our study, we aimed to evaluate changes in DRA using repeated measures data. To analyze this, a linear mixed model (LMM) was chosen because it effectively handles repeated measurements taken at multiple positions and postures for each participant, allowing for the modeling of complex variance structures. This model can account for individual variability and handle the correlation inherent in repeated measures data to provide more accurate estimates. The fixed effects included the participant’s age, height, abdominal circumference, history of fetal macrosomia, perineal laceration, and delayed labor. The degree and direction of changes in the parameters before and after the exercise program were calculated using the coefficient of the visit variable in the LMM. Data exploration showed that the results of ultrasound measurements varied depending on the measurement location, measurement posture, and differences between participants; thus, a random intercept was applied to account for these variables. For SWE and PROs, we used a LMM to compare the pre- and post-exercise values, and only random intercepts for individual differences were applied. In addition, the participants were divided into two groups based on whether their initial IRD met the clinical criteria of DRA, and the slope of the LMM was compared in each group to determine the degree of improvement in the indicators. All statistical analyses were performed using R and RStudio version 4.2.2 (R Foundation for Statistical Computing). LMMs were created using the lme4 package version 1.1-3117 and lmerTest package version 3.1-3 in R18.
As only a subset of participants met the clinical criteria for DRA, a subgroup analysis was performed to determine the treatment effects in the DRA group. Participants were grouped according to whether their initial IRD met the clinical criteria for DRA (≥2 cm). Since IRD was measured using four different methods with different postures and measurement positions, participants were categorized into the DRA group if any of them resulted in an IRD of 2 cm or more. We analyzed whether there was a difference in the degree of improvement in IRD, SWE, ICIQ-UI-SF, and QBPDS between the DRA group and the normal group by comparing the slopes of LMMs.
All participants completed the 8-week exercise program, with no dropouts including the final measurement. Compliance, as measured by whether the exercise videos were played, how long they were played, and the number of daily feedback notes, was 100% for 24 of 67 participants (35.8%) and higher than 80% for 58 participants (86.6%).
For the ultrasound-measured parameters, IRD and SWE, the intraobserver reliability between the three measurements was checked, and an ICC (2, 1) of 0.959 (95% confidence interval [CI], 0.953–0.965) was calculated for IRD and 0.94 (95% CI, 0.931–0.948) for SWE, confirming excellent intraobserver reliability.
The IRD decreased significantly after the intervention (slope, –0.94; 95% CI, –1.52 to –0.36; p=0.002). SWE also showed a significant increase (slope, 7.99; 95% CI, 4.53–11.45; p<0.001). Both PROs also showed a significant improvement after the intervention (ICIQ-UI-SF [slope, –2.58; 95% CI, –2.99 to –2.27; p<0.001], QBPDS [slope, –7.15; 95% CI, –8.22 to –6.09; p<0.001]) (Table 2). Residual versus fitted plots for the linear regression models are shown in Fig. 4.
Table 2 . Effect of online exercise programs on changes in sonographic parameters and patient-reported outcomes
| Variable | Slope | 95% CI | p-value |
|---|---|---|---|
| Inter-recti distance | −0.94 | −1.52 to −0.36 | 0.002 |
| Shear wave elastography | 7.99 | 4.53 to 11.45 | <0.001 |
| ICIQ-UI-SF | −2.58 | −2.99 to −2.17 | <0.001 |
| QBPDS | −7.15 | −8.22 to −6.09 | <0.001 |
CI: confidence interval, ICIQ-UI-SF: International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form, QBPDS: Quebec Back Pain Disability Scale.
Fig. 4. Residual versus fitted plots for the linear regression models. (A) Inter-recti distance. (B) Shear wave elastography. (C) International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form. (D) Quebec Back Pain Disability Scale.
The DRA group was defined as participants with at least one sonographic IRD measurement of 20 mm or more in any location or posture during their first visit, and it included a total of 20 participants. The normal group consisted of the remaining 47 participants, whose IRD measurements were all below 20 mm in every measurement in all locations and postures. The mean initial IRD for each group was 18.0 and 11.4 mm, respectively. When comparing the changes in parameters before and after exercise between the two groups, the DRA group showed an average reduction of 1.58 cm in IRD and an increase of 9.88 in SWE, whereas the normal group showed a reduction of 0.78 cm in IRD and an increase of 7.05 in SWE, indicating a greater improvement in the DRA group. Similarly, the DRA group also demonstrated a greater degree of improvement in PROs (Table 3).
Table 3 . Subgroup analysis for sonographic measurements and patient-reported outcomes
| Variable | IRD | SWE | ICIQ-UI-SF | QBPDS | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Slope | p-value | Slope | p-value | Slope | p-value | Slope | p-value | ||||
| DRA | −1.58 | <0.05 | 9.88 | <0.05 | −3.12 | <0.05 | −7.51 | <0.05 | |||
| Normal | −0.78 | <0.05 | 7.05 | <0.05 | −2.39 | <0.05 | −6.79 | <0.05 | |||
IRD: inter-recti distance, SWE: shear wave elastography, ICIQ-UI-SF: International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form, QBPDS: Quebec Back Pain Disability Scale, DRA: diastasis recti abdominis.
DRA is a common problem in women during pregnancy and childbirth. It usually begins in the second trimester and is most commonly observed in the third trimester2, with an estimated frequency of 66%–100% during pregnancy and approximately 53% after delivery2,19. Risk factors include maternal age, parenting responsibilities, and parity, with conflicting evidence for weight gain and high body mass index20,21. Among the several methods tried to diagnose DRA, ultrasound and caliper measurements are now considered the gold standard. DRA can be diagnosed if the distance between the rectus abdominis muscles measured at the level of the umbilicus is ˃2.7 cm or if it is ˃2 cm when measured 3 to 5 cm above or below the umbilicus22-24. DRA is known to decrease naturally but does not fully recover, with recovery occurring between 1 day and 8 weeks after delivery, after which it plateaus with no further progression25.
There are a few existing studies that have analyzed the effect of exercise therapy on IRD in postpartum women. Awad et al.10 randomized 50 postpartum women to a binder-only group and a binder plus exercise group for 8 weeks and found that the binder-only group achieved an average reduction of 0.37 cm and the binder plus exercise group achieved an average reduction of 0.79 cm, for an additional reduction of 0.42 cm. In addition, Bobowik and Dąbek11 compared 20 women in a 6-week exercise group with 20 in a control group. They found that 95% of the exercise group had a reduction in DRA, while only 15% of the women in the control group had a reduction in DRA, with a significant difference in both groups. An average IRD reduction after 6 weeks of exercise treatment was 2.88 cm. In our study, 8 weeks of exercise therapy resulted in an IRD reduction of 0.94 cm, which is consistent with these studies. Tuttle et al.26 similarly divided 30 patients into four groups: transverse abdominis exercise, kinesio taping, combination, and control, and followed them for 12 weeks. In their study, the exercise and combination groups showed significant improvement compared to the other two groups regardless of posture or IRD measurement location. The exercise-alone group showed a reduction in IRD of 0.94 to 1.09 cm, similar to our study. On the other hand, Keshwani et al.27 divided 32 patients into four groups of eight each, exercise alone, binder alone, a combination of exercise and binder, and control group, and showed a decrease in IRD of 0.93 cm in the exercise alone group and 1.31 cm in the control group, showing less effectiveness of exercise therapy. In their study, patients were recruited from prenatal fitness classes or prenatal educational sessions, so there may be more selection bias than in our study.
Two studies used online exercise programs, with Laframboise et al.28 comparing a 12-week online exercise intervention to a control group of postpartum women, and the results varied depending on where the IRD was measured, with the difference being significant only at the 2 cm level above the umbilicus. Consistent with our results, another study (n=43) analyzing the effectiveness of an online exercise program in postpartum women found a reduction in IRD of 0.24–0.56 cm after the exercise program29, depending on location and posture. A recent meta-analysis analyzing the effect of conservative treatment of DRA reported that overall, postpartum abdominal exercise reduced DRA by 0.4 cm30; however, our study showed a larger reduction of approximately 0.94 cm (95% CI, –1.52 to –0.36).
SWE also showed improvement after the intervention, which is noteworthy in that no previous study has applied it to IRD-related research. We believe that high-quality evidence can be obtained by implementing a higher level of study design in the future. In the case of clinical outcomes, the results were also statistically significant, with ICIQ-UI-SF decreasing by approximately 2.58 points and QBPDS decreasing by approximately 7.15 points, which is better than the results of existing online exercise programs.
The strength of this study is that it is the first study to analyze the impact of an online exercise program on DRA in an Asian population from multiple perspectives. The number of participants is also larger than previous observational studies analyzing the effects of exercise programs. An additional strength is the use of an objective measure of SWE, which has not been used in other studies of DRA. The component of the exercise program is also important. Traditionally, exercise therapy for DRA includes transverse abdominal muscle training, pelvic floor muscle activation, Pilates, and core muscle training to strengthen the abdominal muscles. For institution-based guided training, a professional exercise therapist will guide participants through proper warm-ups and cool-downs, and help set the intensity of the exercise during the workout. To address these issues in an unsupervised setting, we have included warm-up and cool-down exercises before and after the workout videos. However, for participants, the importance of warm-ups and cool-downs may be overlooked, and it is possible that they may not perform these protocols in the workout videos, which could be problematic for postpartum women. Therefore, it is important to make sure that the importance of warm-ups and cool-downs are fully communicated in training sessions, emphasizing the concept of RPE and self-intensity setting.
However, our study had some limitations. First, although we found a significant effect of home-based exercise compared to previous meta-analyses of other exercise therapies, we were unable to make a direct comparison with institution-based guided exercise therapy, as we do not provide exercise therapy with DRA at our institution. The results of the study should be understood and accepted as an observational study, and further studies that include an institution-based treatment group and a control group would be required to increase the level of evidence. Second, compliance issue is important in unsupervised exercise, and it is difficult to evaluate the degree of participation due to the nature of home-based programs. Therefore, we tried to assess and address compliance by documenting whether participants played the videos and for how long, and by writing daily feedback notes and communicating with the exercise therapist. In addition, we did not directly investigate patients’ satisfaction and sense of accomplishment with exercise intervention, but indirectly confirmed it by checking the improvement of clinical scales through clinical scores. However, due to the nature of the method through a mobile application, it was difficult to accurately assess compliance, and it would be helpful to develop technological advancements such as motion recognition simultaneously with video playback in a mobile application or to measure heart rate in conjunction with a wearable device. Third, our analysis lacked consideration of the postpartum period. It has been reported that recovery of DRA begins immediately after delivery and undergoes a recovery process until 8 weeks postpartum, after which natural recovery does not typically occur25. To minimize bias from the natural recovery process, we included postpartum women between 12 weeks and 12 months. However, we were unable to analyze the correlation between the degree of IRD recovery due to exercise and the postpartum period. Further studies that include a detailed analysis of the postpartum period, along with comparisons to a normal group, are expected to provide clinical insights into the optimal timing for initiating IRD treatment.
In conclusion, an 8-week online abdominal muscle exercise program in Korean postpartum women significantly reduced IRD measured by ultrasound and significantly increased SWE. Clinical scores suggestive of symptoms related to urinary incontinence and low back pain also improved significantly. Our results suggest that an accessible, online-based abdominal muscle exercise program may be an effective treatment strategy in postpartum women.
Supplementary Materials can be found at https://doi.org/10.5763/kjsm.2024.42.4.280.
kjsm-42-4-280-supple.pdfNo potential conflict of interest relevant to this article was reported.
Conceptualization, Project administration, Resources, Supervision: JGM. Formal analysis, Visualization: ASC. Sonographic examination: SJH. Investigation, Methodology, Validation: all authors. Writing–original draft: ASC. Writing–review & editing: all authors.
