Outcomes of operative versus non-operative interventions on rectus abdominis diastasis: a combination of a retrospective analysis and a systematic review

Background There is no current consensus on the treatment of Rectus abdominis diastasis (RAD). This study aimed to assess the recurrence rates and adverse events after operative or non-operative intervention.Methods We performed a retrospective study and a systematic review. Patients with a diagnosis of RAD who underwent the mesh placement procedure were eligible. Data were collected from The Research Patient Data Registry. Then, a systematic literature search of Pubmed, Embase, and The Cochrane Library databases was performed. A modified Downs and Black checklist for randomized and non-randomized studies of healthcare interventions (DB) was used to evaluate the methodological quality of the studies.Results Data of 82 patients was collected. The 5-year cumulative incidence of recurrence of abdominal hernia was high (females 37.0% vs males 43.4%, P = 0.557). Of 3908 citations, 27 studies describing 2,372 patients were included into the systematic review. In non-operative group, seven studies (25.93%) with a total of 603 subjects reported no patient relapsed after a follow-up of 12 months; a 40% recurrence rate was reported after a follow-up of 5 years in one study, the adverse events associated with operation mainly included seroma (5.13%), a sensibility disorder of the abdominal skin (2.26%), and minor scarring (1.56%). Muscle strength training, such as training of the pelvic floor muscles, abdominal muscles, rectus abdominis, internal/external oblique, or transverse abdominal muscles, was performed in all included non-surgery studies, neither adverse events nor recurrence was reported.Conclusion Operative intervention may be effective at restoring the functions of the rectus muscles and at improving the stability of abdominal wall in the short-term, while non-operative intervention has a limited effect, especially for those RAD combined with hernia. tendon; OR: ASI: Aponeurosis Shortening

= 0.557). Of 3908 citations, 27 studies describing 2,372 patients were included into the systematic review. In non-operative group, seven studies (25.93%) with a total of 603 subjects reported no patient relapsed after a follow-up of 12 months; a 40% recurrence rate was reported after a follow-up of 5 years in one study, the adverse events associated with operation mainly included seroma (5.13%), a sensibility disorder of the abdominal skin (2.26%), and minor scarring (1.56%). Muscle strength training, such as training of the pelvic floor muscles, abdominal muscles, rectus abdominis, internal/external oblique, or transverse abdominal muscles, was performed in all included non-surgery studies, neither adverse events nor recurrence was reported.Conclusion Operative intervention may be effective at restoring the functions of the rectus muscles and at improving the stability of abdominal wall in the short-term, while non-operative intervention has a limited effect, especially for those RAD combined with hernia.

Background
Rectus abdominis diastasis (RAD) is characterized by widening of the linea alba and usually presenting with a "bulge" in the midline when intra-abdominal pressure is increased. This phenomenon at present was well-recognized as the laxity of the ventral abdominal musculature 1,2 .
There is no current consensus on the treatment of RAD, nor is there unity on the repair techniques among the surgical and physical and rehabilitative communities 19-21 , although a large body of literature has been generated on the restoration of some degree of strength of the midline in the postpartum setting, and many different innovative methods for surgical treatment to repair the defeat are available: vertical incision from the xiphoid process to the umbilicus or a low transverse skin incision, the repair of diastasis recti combined with liposuction, with or without using mesh to repair the rectus abdominis diastasis as well, etc. Based on the current literature, the promising results of new procedures with mesh placement and recurrent rate after long-term follow-up have not been clearly reported yet 22,23 .
Therefore, evaluating the effect of both physical therapy and repair surgery interventions is important, because RAD is wide-spread, significantly downgrading the quality-of-life (QoL) and increasing the costs of health care burden as well.
To better understand what is about the therapy and its outcomes of RAD, we conducted a retrospective study concerning the mesh placement and the systematic review summarizing operative-interventions and non-operative interventions in patients with RAD. Our result possibly may aid our understanding of RAD and help physiotherapists, surgeons and patients with RAD to make intervention strategy decision when encountering RAD patients.

Methods
This study followed a two-step procedure: (1) a retrospective analysis summarizing the outcomes of repairing of abdominal wall and hernia with mesh based on the data extracted from the Research Patient Data Registry (RPDR) in Massachusetts, USA; (2) a systematic literature review of intervention studies evaluating the effects of operative versus non-operative strategies on patients with RAD. The current review was registered on PROSPERO and can be accessed at www.crd.york.ac.uk/PROSPERO/display_record.asp? (CRD42018115608). We applied the PRISMA reporting guidelines for this review.

Materials and Methods
This retrospective study analyzed the outcomes of patients with RAD who underwent mesh placement from six hospitals within the Partners Healthcare System in Massachusetts, United States. The review board of RPDR approved this study protocol (#2018P003056). Patients with a diagnosis of RAD who underwent the mesh placement procedure were eligible. RAD was defined as an inter-rectus distance (IRD) of or more than 22 mm, three centimeters above the umbilicus measured in a relaxed state 24 .
Exclusion criteria were the implantation of mesh was not used for the repairing of abdominal wall and hernia. Identified individual-level data of eligible patients from January 2011 to November 2018 was obtained from RPDR. Data including demographic information (age, body mass index (BMI), race or ethnic group), social history (smoking status, alcohol abuse, depressive disorder), diagnosis (including contaminant conditions possibly associated with RAD: hernia, low back pain, pelvic and perineal pain,

The outcome indexes
The primary outcome was the recurrence of RAD and hernia, and the second outcome was a composite of concomitant conditions (hernia, low back pain, pelvic and perineal pain, incontinence, strain of muscle, fascial and tendon (SMFT), depressive disorder and the peri-operative characteristics (operation type, anesthesia type, mesh type, skin closed method and adhesions in abdominal cavity noted on operate notes).

Statistical analysis
All statistical analyses were performed using IBM SPSS Statistics, version 20. Data were summarized as mean and standard deviation or frequencies and percentages where applicable. Chi-square test or Fisher's exact test were used to analyze categorical variables and Independent-Samples T test was used for continuous variables. binary logistic regression analysis to determine the association between the potential factors and the primary outcome, and the odds ratio and 95% confidence interval were estimated from BLRA. To reduce confounding bias, we adopted a forward stepwise regression analysis (α entry = 0.10, α removal = 0.10). A P-value <0.05 was considered statistically significant.

Characteristics of the patients
Of 82 patients who were reviewed at the 6 clinical settings from January 2011 through November 2018, 45 were female who received 54 mesh placement procedures totally and 37 were male who received 53 procedures. These patients were all diagnosed with RAD combined with one or more hernia. The mean age undergoing first time hernia repair in males was significantly older than that of females (mean 61.2 years old vs. 49.2 years old, t = 4.654, P = 0.000). The mean BMI values were greater than 30 kg/m 2 in both groups (mean 31.8 vs. 31.6, χ 2 = 0.147, P = 0.883). The number of individuals smoking at index repair and using alcohol were high in both groups (37.8% vs. 31.1%, 54.1% vs. 48.9%, respectively) ( Table 1).

The outcomes
The 5-year cumulative incidence of recurrence of and RAD and hernia was 37.0% in female group compared with 43.4% in male group (χ 2 0.450, P = 0.557, Figure 1). While the recurrence after a oneyear follow-up in male group was higher than in female group (24.53% vs. 5.56%, χ 2 0.7.571, P = 0.007, Figure 1). We failed to identify risk factors contributing to recurrence after the repair with mesh when practicing binary logistic regression analysis by using 17 independent variables including age, race, concomitant conditions, mesh type, suture type, etc. As for the composite second outcomes, ventral hernia was most common type (27.8% vs. 40.7%, P = 0.104) and there was no difference in hernia number and type between two groups (P = 0.176). The rate of concomitant low back pain were higher than 30% as well in both groups (35.1% vs. 33.3%, P = 1.000), whereas depressive disorder and major depressive disorder in female group were significantly higher than that in male groups (51.1% vs. 27.0%, P = 0.041; 46.7% vs. 21.6%, P = 0.022, respectively). Over onequarter cases had noted adhesion in abdominal cavity and there was no significant difference between two groups (29.7% vs. 35 Table 1). The most common anesthesia type was general anesthesia and the suture was used to close skin commonly in both groups (79.2% vs. 85.2%, 86.8% vs. 88.9%, respectively). In female group, risk of coexisting with low back pain in patients with pelvic and perineal pain was 27 times greater than that of those without pelvic and perineal pain (OR 28.653; 95% CI 2.924, 280.753; P =0.004); and patients with concomitant SMFT were more likely to have low back pain than those without SMFT as well (OR 9.604; 95% CI 0.763, 120.858; P =0.08).

Materials and Methods
This study was conducted according to the Modified Downs and Black checklist for randomized and non-randomized studies of healthcare interventions (DB) 25 .

Study selection
Titles and abstracts were reviewed independently by three reviewers (Li YX, Fan L and Li Q) to determine the eligibility of the papers for inclusion. The research team discussed and resolved issues together if there were any disagreements. Studies meeting all the following criteria were included:1) the study included at least one intervention that aimed to improve RAD; 2) the study was designed as a prospective, retrospective, clinical trial, or case report; 3) Full-text journal articles in English.
Cadaver studies and animal studies were excluded.

Data extraction and outcome measures
Outcomes of the systematic review were overall presence/absence of RAD, RAD size/width, patient satisfaction, recurrence rate, physical activity levels, back pain, functional impairments, quality of life, follow-up information, and adverse events. Data on the type of study and sample size, the nationality of the enrolled patients, and the patients' age, sex, BMI (Kg/m 2 ), the types of anesthesia used during the surgical operation were also extracted.

Quality assessment
All included papers were evaluated for methodological quality using DB, which has a total maximum modified DB score of 28 and consists of four factors: external validity, confounders, bias, and power 25 . Articles with a DB score were grouped into the following quality levels: strong (≥ 21), moderate (14-20), limited (7-13), or poor (7 or less) 26 . The final set of papers varied in outcome reporting; therefore, a meta-analysis was not possible.

Study identification
The initial database search yielded 3906 articles, and two were obtained by tracking the references.
After removing 739 duplicates, 3169 studies were screened for further analysis. Since 3102 did not match the review criteria, 67 full-text articles were screened for eligibility; of these, 40 were excluded later. Therefore, 37 studies were included in this systematic review. The flow diagram and reasons for study exclusion are shown in Fig. 2.

Characteristics of the included studies
The characteristics of all 2372 patients from the 37 included studies are presented in Tables 1 and   2

Quality assessment
Of the studies examining surgery intervention, 11 were of moderate quality, 13 were of limited quality, and three were of poor quality; of the studies examining non-surgery intervention, four were of strong quality, four were of moderate quality, and two were of limited quality. The average modified DB score of the included studies for the surgery intervention group was 11.7 (range 6-20), and that for the non-surgery group was 18.4 (range 12-26).

Type of anesthesia
Fifteen studies described the type of anesthesia, ten used general anesthesia, two used local anesthesia, and three used general anesthesia combined with epidural anesthesia. There were no complications due to anesthesia reported in any of the included studies.

Operative intervention
Eight studies (29.63%) reported the use of endoscopy during abdominoplasty surgery; the remaining studies (70.37%) were performed using a vertical skin incision from the xiphoid process to the umbilicus or a low transverse skin incision. The mesh augmentation technique was used to reinforce the linea alba with or without associated hernias in nine studies (33.33%), and the type of mesh used included polypropylene, self-adhesive synthetic, composite, and resorbable vinyl mesh.
In eight studies (29.63%), liposuction was combined with traditional abdominoplasty and repair of the rectus sheath defect to reduce complications and provide fast recovery; the volume of abdominal lipoaspirate ranged from 50 to 2800 ml, and the maximum volume reported in these studies was 7800 ml. Five studies (18.52%) reduced recurrence by adopting a special suture method, such as the triangular mattress suture, continuous suture, the Venetian blinds application technique, braided suture, buried suture, and figure-of-eight suture combined with the wearing of self-adherent silicone strips.

Non-operative intervention
Muscle strength training, such as training of the pelvic floor muscles, abdominal muscles, rectus abdominis, internal/external oblique, or transverse abdominal muscles, was performed in all included non-surgery studies. Physical interventions including head lift, pelvic lock or tilt, plank, superman, and double leg raise were induced into the intervention program in two studies. The frequency was from one to 14 times per week, and the period of intervention was from 8 weeks to 15 months unequally.

Outcomes
The conclusions from six studies (60%) in the non-surgery group were positive, and the remainder (40%) were negative. The indexed outcomes included one or more of the following: inter recti distance (IRD), BMI, abdominal strength, pain, quality-of-life (QOL), pelvic floor distress index (PFDI), and activation endurance (hold seconds). In the surgery group, 26 studies (96.30%) obtained positive results over the short-term, and just one study (3.70%) reported negative results over the long-term.
The indexed outcomes in this group included one or more of the following: the width of the diastasis of the recti abdominis muscles, ventral hernia pain, quality-of-life, Oswestry Disability Index, abdominal wall function, cosmetic outcome, satisfaction, Aponeurosis Shortening Index (ASI), back pain after surgery, and recurrence rate. None of the included studies demonstrated that physical training can be effective in improving the abdominal wall function of patients suffering RAD combined with hernia.

Recurrence rates
In the surgery group, there were seven studies (25.93%) including a total of 603 subjects with a reported follow up of 12 months; no patient relapsed; after a follow up of 24 months, the three remaining studies (11.11%) reported a 0% recurrence in a total of 268 subjects. Van reported a 40% recurrence rate of standard plication of the abdominal wall with absorbable material after a follow up of 5 years in a retrospective study of 40 patients attending a follow-up ultrasound investigation and with a total sample size of 70 women. No residual or recurrent diastasis was reported among 228 subjects receiving experimental training in the non-surgery group.

Summary of results
This systematic review found a limited strength of evidence that abdominoplasty and diastasis recti repair with or without mesh and/or liposuction may be effective at restoring the function of the rectus muscles, improving the stability of the abdominal wall and upgrading the quality of life in the shortterm. However, the post-operative recurrence rate in patients who underwent either traditional standard vertical plication or modern endoscopic surgery increased rapidly with the time, a rate of up to 40 percent was reported in one study, and this high recurrence rate also has been confirmed in our retrospective study.
Our study also found a moderate strength of evidence that physical training including strength training of the pelvic floor muscles, transverse abdominal muscle, core stabilization, and rehabilitation therapy have limited effectiveness for the rehabilitation of diastasis rectus muscles, especially for patients suffering from RAD combined with hernia.
Among all twenty-one types of complications reported in the included studies, the rate of seroma was the highest at five percent, which was higher than other reported 59 . A large body of evidence shows that the use of drains in abdominoplasty may increase the seroma rate, while surgeons are unwilling to change their clinical practice of using drains 60 . Meanwhile, thrombosis and pulmonary embolism were reported in these studies, but their final outcomes and the reason leading to the severe adverse events were not reported. Therefore, an understanding of how to prevent these adverse events due to diastasis rectus abdominus repair surgery is also needed and will require further, in-depth research.
Therefore, the problems of how to reduce the recurrence rate and adverse events related to procedure remains although surgeons have tried innovative suture techniques or materials, such as triangular mattress suture, smooth running absorbable polydioxanone suture, nylon, or polydioxanone and/or different meshes, or in combination with liposuction abdominoplasty, etc. In this retrospective study, we found that, obesity was an important characteristic in both male and female patients with RAD. In addition, the rate of noted adhesion in abdominal cavity was high, similar to previous study 61 . These adhesions played an important role in increasing the difficulty of operation performing due to extensive adhesions. Whether the adhesion in abdominal cavity was related to the recurrence of RAD and hernia was unknown. In the field of fascia research, researchers found that a previously unknown feature of human anatomy, termed a newfound "organ", exists in numerous tissues, including the submucosae of the entire gastrointestinal tract and urinary bladder, the dermis, the peribrochial and peri-arterial soft tissues, and the fascia; this may play an important role in the mechanical functioning of many or all tissues and organs, cancer metastasis, edema, and fibrosis 62 .
Other researchers found that telocytes exist in human fascia, which may contribute to some common diseases such as stiffening of limbs, inflammatory diseases, sclerotic and the progression of fibrotic, etc 63, but the exact role of these cells is unknown 64 . It was also reported that fascia lesions were closely related with chronic pain 65 . We confirmed that women with SMFT were also more likely to coexisting with LBP than those without SMFT, which is consistent with myofascial pain theory as well 26 . We demonstrated that RAD is most common among obesity population and more than 33 percent of them were coexisting with low back pain. In addition, women with SMFT had more than 9 times of risk of coexisting with low back pain than those without SMFT; we also found that the volume of intraabdominal and superficial fat being removed during abdominoplasty surgery is closely associated with the outcome. Is there a close relationship between the fat removed by surgeons and the newfound "organ" with telocytes? The authors want to boldly make a hypothesis here, and further study is warranted to examine this most possibly association.
In conclusion, this systematic review found a moderate strength of evidence that physical training has limited effectiveness for the rehabilitation of diastasis rectus muscles, especially for those patients suffering from RAD combined with hernia, and abdominoplasty and diastasis recti repair with or without mesh and/or liposuction may be effective at restoring the functions of the rectus muscles and improving the stability of the abdominal wall and quality of life in the short-term; however, up to now, the long-term durability and the adverse events are unsatisfactorily, these all need further, in-depth research.

Limitations
Several limitations existed in this study. Firstly, the methodological quality, especially in the surgery group, was weak. Secondly, a meta-analysis of the studies was not performed given the lack of consistency in the interventions. Thirdly, the results of the review may be influenced by publication bias because we did not contact the authors about other possibly existing unpublished analyses, nor did we search for unpublished studies.

Recommendations for future research
RAD is common and may be associated with low back pain and impaired health-related quality of life.
Fortunately, mounting evidence indicates that surgery repair and physical training have helped many patients in improving their quality of life. However, why is RAD very common among people with high BMI, why does the diffuse fusiform bulge reappear to a large extent even when mesh is used, and why is the outcome possibly better when abdominoplasty combined with liposuction is used rather than the standard method? Whether or not there is a closely relationship between surface and deep fascia lesions? Many puzzles remain to be explored. Therefore, future research should pay more attention to the pathogenesis of RAD based on a multidisciplinary co-operative team of experts, including at least the departments of plastic surgery, radiology, ultrasound imaging, and pathology as well as the fascia research center.

Conclusion
This review found a limited strength of evidence that abdominoplasty and diastasis recti repair with or without mesh and/or liposuction may be effective at restoring the functions of the rectus muscles and improving the stability of the abdominal wall and quality of life in the short-term; a moderate strength of evidence was found that strength training has a limited effect, especially for those patients suffering RAD combined with hernia. To maximize long-term efficacy, a multidisciplinary cooperative team including experts from at least the departments of plastic surgery, radiology, ultrasound imaging, and pathology as well as the fascia research center is urgently needed because abdominoplasty represents the fourth most-common aesthetic surgery procedure according to the

Ethics approval and consent to participate
The review board of the Research Patient Data Registry (RPDR) approved this retrospective study protocol (#2018P003056). As for the systematic review, the ethics approval is not required.

Consent for publication
Not applicable.

Availability of data and materials
As a review study, the data were extracted from the Research Patient Data Registry (RPDR) in Massachusetts, USA; as for the systematic review, the data were extracted from Pubmed and The Cochrane Library databases. The data supporting our findings can be obtained by emailing the corresponding author.

Competing interests
The authors declare that they have no competing interests.

Funding
Not applicable.  Tables       The comparison of hernia recurrence rate after repair procedure with mesh.