Impact of Rectus Diastasis Repair on Abdominal Strength and Function: A Systematic Review

Rectus diastasis plication performed during abdominoplasty aims to narrow the widened linea alba and return the rectus muscle bellies to their anatomic position. It is unclear whether plication improves abdominal strength and function. This systematic review summarizes the effect of rectus plication on abdominal strength, function, and postoperative complications. A comprehensive search of CINAHL, Embase, Medline and Web of Science was performed. Screening and data extraction were performed in duplicate. Data were extracted from the included articles, and outcomes were analyzed categorically. A total of 497 patients from seven articles were included. Mean age was 44.5 years (range 20.5-72) and 94.4% were female. Three articles reported abdominal strength measurements, with two showing significant improvement. Four articles used the SF-36 survey, all demonstrating improvement in physical function subscale postoperatively. An additional six instruments were used to assess functional outcomes, of which four demonstrated significant improvement. The overall complication rate was 17.0%. Rectus plication is commonly performed during abdominoplasty to improve abdominal form and function. While the literature to date is encouraging with respect to functional outcomes, improvements in abdominal strength are less consistent. Heterogeneity in patient population, outcome measures, and comparison groups limit the strength of our conclusions. Future research should include a large comparative study as well as a protocol for standardizing outcomes in this population.

by Emanuelsson et al. compared two surgical techniques -double rowed Quill suture plication and retrorectus polypropylene mesh -to patients undergoing a three-month physiotherapy training program. All three groups improved their baseline abdominal strength per the Biodex system-4, which entails measuring patient-applied force against the system at various positional angles [10]. However, the operative groups surpassed the physiotherapy group in visual analog scale (VAS) and patient-perceived strength and saw greater improvements in the Biodex-4 system ratings. There were no differences between the two surgical techniques in terms of subjective strength or functional outcomes [10].
The current literature regarding abdominal strength following ARD repair is inconclusive and to the best of our knowledge, there is no systematic review on the subject. The objective of this study is to determine whether ARD repair improves abdominal wall strength and function in patients undergoing abdominoplasty compared to the patient's baseline or other interventions.

Protocol and Registration
This systematic review adheres to PRISMA guidelines [16] and was registered a priori to Open Science Framework (OSF, 10.17605/OSF.IO/H9JB3) and can be found in Appendix A.

Search Strategy
A comprehensive search of CINAHL, Embase, Medline and Web of Science was completed from database inception to April 22, 2020, and was checked regularly for new relevant articles, with the assistance of a health science librarian. Our sample search strategy can be found in Appendix B. The search was limited to English language and human studies. A manual search of the included study's references was completed to ensure relevant articles were not missed.

Article Selection
The following inclusion criteria were applied: 1) primary research; 2) patients undergoing abdominoplasty with open ARD repair; 3) abdominal strength or functional outcomes were reported; and 4) mean follow-up was greater than six months. Articles were excluded if: 1) they were case reports, opinion pieces, editorials and non-primary research (e.g., systematic reviews, scoping reviews, commentaries); 2) no strength or functional outcomes were reported; 3) a laparoscopic technique was used for ARD repair; 4) full text was not available; and 5) they were not written in English. When outcomes of interest were incompletely reported, an attempt was made to contact the authors for this data.

Study Screening
Title and abstract and full-text screening were performed in duplicate by two independent reviewers. Discrepancies at the title/abstract stage resulted in automatic inclusion in the next stage, and discrepancies at full text were resolved by consensus between the reviewers. Further discordance at this stage was then settled by the senior author. Reasons for exclusion at both the title/abstract and full-text stage were recorded. At each stage, reviewer agreement was assessed by calculating Cohen's Kappa (κ) statistic [17].

Data Extraction
Data were extracted independently by two reviewers into an online collaborative spreadsheet (Google, California, USA). Data extracted included study characteristics (design, date, location, sample size, demographics, time horizon, and level of evidence), description of the population, intervention, comparator (if applicable), relevant outcomes, and time horizon.

Statistical Analysis
Due to the nature of the outcomes reported in the included articles, the analysis and results are presented in a descriptive fashion. Outcomes were grouped into the following categories: strength (primary outcome), functional, and complications. Categorical data were reported using frequencies and percentages. Continuous data were reported with weighted means, median and range.

Risk of Bias
Risk of bias for randomized control trials was assessed using the CLARITY group Cochrane standardized risk of bias assessment for randomized controlled trials [18]. For non-randomized trials, the Methodological Index for Non-Randomized Studies (MINORS) was used, with a minimum score of zero, and a maximum score of 16 for non-comparative studies or 24 for comparative studies [19].

Included Articles
The search yielded 420 articles, of which seven were included for analysis ( Figure 1). Moderate agreement was achieved at the title/abstract stage (κ=0.761, 95% CI 0.590 to 0.932), and perfect agreement at the fulltext stage (κ=1.00). Of the included articles, four were randomized control trials (Level I) [10], one was a retrospective cohort study (Level III) and two were prospective case series (Level IV). Two of the included randomized trials were overlapping reports of the same cohort of patients, with varying follow-up periods and outcome measurements [10,20,21]. The mean MINORS score for the two non-comparative articles was 9.5/16 (SD 0.71, range 9-10), which represents a moderate risk of bias. For the one non-randomized comparative study, the MINORS score was 15/24, which represents a moderate risk of bias. Study demographics can be found in Table 1, and PICOT summaries of included studies can be found in Table 2

Patient Characteristics
There were 497 unique patients across all studies. The mean age was 44.5 years (SD 4.1, range 20.5-72) and 94.4% were female (469/497 patients). The median follow-up was 12 months (range 3-78 months), with one article having a mean follow-up less than 12 months. The pre-operative ARD width ranged from 3.0 to 10.0 cm. The mean preoperative body mass index (BMI) was 25.7 kg/m 2 (SD 1.69, range 17.2-36).

Abdominal Strength Outcomes
Details of strength and functional outcome scales can be found in Table 3. Objective and subjective measurements of abdominal strength were reported in three studies, and the details of each study can be found in  [2] found no significant difference in postoperative abdominal endurance between those who received plication and those who did not.

Functional Outcomes
Functional outcome measures were used in all seven studies; details of each study can be found in Table 4. The most consistent result was an improvement in the physical function subscale of SF-36 in all studies that used this instrument [6,10,21,23]. The next most commonly used scale, the DRI, showed improvement in one study [23] but no improvement in another [2]. The remaining functional scales (VAS, NHP, Modified Sahlgrenska Excess Skin Questionnaire, EuraHS-QoL) showed significant improvement after plication, except for the EuroQoL-5.

Complications
Post-operative complications were reported in five articles in patients who had received rectus plication (276 patients), with an overall complication rate of 17.0% (47/276). Of the 47 complications recorded, the three most common complications were seroma (12 patients, 27.7%), wound dehiscence (11 patients, 23.4%), and minor bleeding episodes (10 patients, 21.3%). An outline of the remainder of complications is found in Table 5.

Comparative Articles: Plication vs. No Plication Abdominoplasty
Two articles [2,6] compared patients who received rectus plication during their abdominoplasty to those who did not. Staalesen et al. found no significant differences between groups for the SF-36, the Modified Sahlgrenska Excess Skin Questionnaire, or the EuroQoL-5D (p > 0.05) [6]. Similarly, Wilhelmsson et al. [2] demonstrated no significant differences between the plicated and non-plicated groups in postoperative abdominal muscle endurance (p = 0.53), lung function (p = 0.25) or physical function (p = 0.35). The only difference observed was a significant improvement in postoperative running in the plicated group, which was not observed in the non-plicated group (pre-op DRI running score: 36, post-op: 8, p = 0.04) [2]. Unfortunately, neither article reported on complications and therefore we were unable to assess whether adding rectus plication affects the complication rate.

Discussion
The purpose of this study was to determine whether ARD improves abdominal wall strength and function in patients undergoing abdominoplasty. Overall, rectus plication did not consistently improve abdominal strength. In two of the three reporting studies [10,23], there was a significant improvement in both objective and subjective abdominal strength compared to the patient's baseline. However, a third study by Wilhelmsson et al. demonstrated no significant difference [2]. Regarding functional outcomes, the most consistent result was an improvement in the physical function subscale of SF-36 in all studies that used this instrument [6,10,21,23]. The DRI scale showed improvement in one study [23] but no improvement in another [2]. Four of the remaining five functional scales demonstrated a significant improvement after plication, with the only exception being no significant improvement in EuroQoL-5 scores. With respect to postoperative complications, the overall complication rate in this review was 17.0%. This is consistent with current literature on postoperative complications following abdominoplasty ranging from 10 to 20% [24,25].
For abdominal wall defects, such as ventral hernias, it has been shown that restoring muscular continuity improves truncal strength and abdominal wall function [26,27]. Although a ventral hernia and a severe ARD, in theory, could have a similar impact on abdominal wall function, the biomechanics of ARD repair in abdominoplasty have not been studied. We postulated that ARD repair would enhance abdominal strength and function following abdominoplasty. However, the existing literature is of too poor quality and studies are too heterogeneous to make any strong conclusions. Additionally, the existing literature does not differentiate based on the severity of rectus diastasis, and therefore we are unable to determine the degree of defect that results in weakness, pain, or disability. It is possible that larger improvements could be expected in those with more severe diastasis, but this cannot be determined from the current literature.
Furthermore, rectus plication could theoretically increase operative risk due to longer operative time and an increase in intraabdominal pressure; however, there were no data comparing complication rates between abdominoplasty alone versus abdominoplasty with plication, so this could not be assessed in this review.
The major limitation of this review is the heterogeneity and quality of outcome measures. Abdominal strength is measured in newton-metres (Biodex System-4), seconds (endurance), and subjectively with a visual analogue scale. This heterogeneity was not anticipated for our primary outcome, which was chosen a priori. The Biodex System-4 has proven to be a favourable outcome measure with high reliability and external validity [28]; however, only one article used this tool for assessment of abdominal strength [10].
Limitations of this method include cost and access to the equipment required [28]. The endurance testing uses a validated stamina scale but requires a trained physiotherapist and can be influenced by patient motivation [2,23]. Functional outcomes were measured using seven different scales assessing quality of life and physical functioning with a series of subjective patient-reported questionnaires. These tools are influenced by patient perception and are therefore limited in both their reliability and validity compared to objective measurements [29,30]. Only one of the seven scales (the Modified Sahlgrenska Excess Skin Questionnaire) was modified to be specific for the targeted patient population [6]. Another scale was specific for a similar patient population (the EuraHS-QoL for hernia patients) [22]. The most consistently reported functional outcome scale was the SF-36, which has been widely used in the scientific literature across a variety of patient populations [31]. However, the studies included in this review report a limited amount of SF-36 data, and therefore we were limited in the data available for pooled analysis [6,10,21,23]. To address this heterogeneity in outcome reporting, outcome standardization, which is being undertaken in several areas of plastic surgery, would be beneficial [32][33][34][35].
In addition to heterogeneity in outcome measures, there was substantial variation in the control groups used. Rectus plication was compared with no plication [2,6], mesh repair [10], endoscopic repair [22], and/or physiotherapy [10], and in three studies [7,21,23], there was no comparison group. This made it difficult to compare outcomes across studies even when the same outcome measure was used. Finally, there was variability in patient population (for example, post-partum women versus post bariatric surgery patients) [6,23]. Overall, heterogeneity precluded a meta-analysis.
One strength of our review is that there were three level I RCTs and two of the remaining three studies collected data prospectively. However, all three RCTs had high risk of bias in at least one domain, and two had high risk of bias in more than one domain. Of the non-randomized non-comparative trials [7,23], the mean MINORS score was 9.5/16, indicating a moderate risk of bias. For the one study with a comparison group, the MINORS score was 15/24, which again represents a moderate risk of bias. Furthermore, our literature search only yielded seven articles, which demonstrates the lack of research to date. This small number and their moderate risk of bias impacts the strength of conclusions that can be drawn from this review.

Conclusions
Rectus plication is commonly performed during an abdominoplasty to improve form and function. While the literature to date is encouraging with respect to functional outcomes, improvements in strength outcomes are less consistent. Substantial between-study heterogeneity in patient population, outcome measures and control group limit the strength of our conclusions. Future research should involve a large, three-armed trial comparing abdominoplasty with rectus plication, abdominoplasty without rectus plication, and nonoperative management (physiotherapy). Outcomes should include both objective and subjective strength outcomes, as well as patient-reported functional outcomes.

Appendices Appendix A. Protocol
Functional Improvement in Abdominal Strength Following Rectus Diastasis Repair for Abdominoplasty: A Systematic Review Protocol (OSF)

Study Information
Hypotheses: Patients who have undergone abdominoplasty or panniculectomy with repair of the rectus diastasis will have a significant improvement in their abdominal strength.

Design Plan
Study type: Meta-Analysis -A systematic review of published studies. Blinding: No blinding is involved in this study. Is there any additional blinding in this study? Systematic screening of the literature will be completed independently by two reviewers, who will be blinded to the other person's decisions.

Sampling Plan
Existing Data: Registration prior to analysis of the data Explanation of existing data: Not applicable.

Data Collection Procedures
Inclusion: Patients who are undergoing abdominoplasty or panniculectomy; Patients who had surgical repair of rectus abdominus diastasis; Studies that report outcome parameters relating to abdominal strength or functional measure of core strength; All time horizons. Exclusion: Case reports, opinion pieces, editorials and non-primary research (e.g., systematic reviews, scoping reviews, commentaries); Animal or pre-clinical studies; Studies in languages other than English; Overlapping reports of the same study. Data Collection: A study-specific data extraction form will be developed using Google Sheets (Google, California, USA). The form will be created a priori and any required modifications will be made by the reviewers in the extraction phase. The form will include study characteristics (design, date, location, sample size, demographics and level of evidence), specifics of abdominoplasty procedure, and reported study outcomes (function, abdominal strength, pain, disability, complications, etc.). The reported outcomes will be specific to each of the studies included in our review. Reviewers will independently collect data from each of the studies included using the form.

Analysis Plan
Statistical models: We will summarize study characteristics using frequencies and percentages for categorical data and means and standard deviation or median and interquartile range for continuous data. We will summarize populations, interventions, comparison groups (if applicable) and reported outcomes in a table and we will present authors' conclusions about the intervention graphically if relevant. Missing data: We will contact the authors of the study for any missing data.

Exploratory analysis: No response
Level of evidence: Level of evidence will be classified as follows: Level I (randomized controlled trials/systematic reviews/meta-analysis of RCTs), Level II (cohort study), Level III (case control study), Level IV (case series/case report), or Level V (opinion/survey/qualitative study). Risk of bias in individual studies: Risk of bias for randomized control trials will be assessed using the Cochrane Risk of Bias tool. For nonrandomized trials, the Methodological Index for Non-Randomized Studies (MINORS) will be used to assess risk of bias. Risk of bias across studies: The GRADE criteria will be used to consider risk of bias, inconsistency, indirectness, imprecision and likelihood of publication bias. This will be used to assess the quality of evidence, and taken into consideration for clinical recommendations.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.