Morbidity of the Donor Site and Complication Rates of Breast Reconstruction with Autologous Abdominal Flaps: A Systematic Review and Meta-Analysis

Background Numerous studies have evaluated the use of autologous abdominal tissue for breast reconstruction; nevertheless, complications and donor site morbidity rates vary significantly. The study aims to compare the literature regarding morbidity of the donor site and complication rates of breast reconstruction with autologous abdominal flaps. Methods The databases of MEDLINE, EBSCO, Scopus, Wiley Library, and Web of Sciences were searched for studies that compared different flaps in terms of complications and donor site morbidity. The procedures studied included pedicled transverse rectus abdominis myocutaneous ﬂap (pTRAM), free TRAM (fTRAM), deep inferior epigastric perforator (DIEP), and superﬁcial inferior epigastric artery perforator (SIEA) flaps. A total of 34 studies were included. Of these, 28 were retrospective studies and 9 were prospective cohort studies. Results When compared to DIEP, fTRAM flaps were found to have a decreased incidence of flap fat necrosis, hematoma, and total thrombotic events, yet a higher risk of donor site hernia/bulging. pTRAM flaps were also associated with an increased risk of hernia/bulging at the donor site, as well as wound infection, yet flap hematoma was less common. On the other hand, SIEA flaps showed the lowest risk of donor site hernia/bulging while still having a high risk of wound infection. Conclusion fTRAM procedures comparatively had the least complications. However, regarding flap choice, patients would benefit most from a case-by-case analysis, taking into consideration individual risk factors and preferences.


Introduction
Breast cancer is the most common cancer among women. According to the World Health Organization, in 2020, there were 2.3 million women diagnosed with breast cancer [1]. After mastectomy, breast reconstruction surgery is considered a crucial step toward more comprehensive breast cancer treatment and management [2]. For women undergoing breast reconstruction surgery, two main points are considered: timing and type of reconstruction. Breast reconstruction can be done either at the time of mastectomy (immediate reconstruction) or later (delayed reconstruction). Based on the type of procedure, breast reconstruction can be classified into implant-based reconstruction (IBR) or autologous reconstruction (AR) [3]. roughout the years, multiple donor sites and options have been described for AR. However, abdominal-based flaps continue to be the mainstay for AR [4]. In 1982, Hartrampf et al. described the transverse rectus abdominis myocutaneous (TRAM) flap, which provides a soft, ptotic, and aesthetically pleasing reconstruction similar to the natural breast [5]. However, the abdominal wall integrity is compromised due to the use of rectus muscle and fascia, resulting in abdominal bulge and hernia, which is recognized as a limitation of the TRAM flap [6]. To reduce abdominal donor site morbidity by harvesting less muscle, the TRAM flap has undergone numerous modifications, resulting in different techniques such as muscle-sparing TRAM (MS-TRAM), deep inferior epigastric perforator (DIEP), and superficial inferior epigastric artery (SIEA) flaps [7][8][9]. Sailon et al. conducted a systematic review in 2009 to determine the difference in patient selection, flaprelated complications, and donor site morbidity. ey found a statistically significant difference between DIEP and MS-TRAM flaps in fat necrosis and total necrosis, but not partial necrosis rates, whereas for donor site morbidity, there was no statistically significant difference. Also, the study interprets the lack of a statistically significant difference due to the insufficient sample size, resulting in low statistical power and type II error [10]. Egeberg et al. conducted a meta-analysis in 2012 to compare the donor site morbidities of SIEA, DIEP, and MS-TRAM flaps and found a 20% reduced risk of bulging when DIEP flaps were used compared to MS-TRAM flaps. However, there was no statistically significant difference. Also, a limitation in the included studies was the small sample size [11]. Another study found no statistically significant difference in flaprelated complications (vessel thrombosis, partial or total flap loss, and fat necrosis) or donor site morbidity (abdominal bulge) between the MS-TRAM and DIEP flaps. e study concluded that determining flap superiority is challenging due to different harvesting techniques between surgeons and different measures used to evaluate the donor site morbidity. us, higher-quality studies are needed to compare different types of flaps and determine the flap choice for each patient [12]. erefore, this systematic review and meta-analysis aim to review and compare the literature regarding the morbidity of the donor site and the complication rates of breast reconstruction with autologous abdominal flaps.

Study Selection.
English studies that assessed the complications following breast reconstruction with autologous abdominal flaps were included in this systematic review. Studies met the inclusion criteria if they had more than ten patients and compared flap complication outcomes, or donor site morbidities and complications for pedicled transverse rectus abdominis myocutaneous flaps (pTRAM), free TRAM (fTRAM), deep inferior epigastric perforator (DIEP), and superficial inferior epigastric artery perforator (SIEA) flap in adults 18 years or older. Review articles, metaanalysis/systematic reviews, case reports, economic analyses, animal studies, cadaver studies, narrative reviews, and editorials were excluded. Studies that exclusively evaluated only one type of autologous abdominal flap or reported no outcome of interest were also excluded. Articles were retrieved and screened by four independent investigators, and a fifth investigator then resolved any disagreements.

Data Extraction.
Full-text articles were retrieved and screened. A single investigator retrieved data from each report. General information was collected from the articles, including author, country, design, sample size, age of patients, and year of publication. e target outcomes of the incidence of donor site complications, rates of each type of complication, and risk factors for complications were extracted. In addition, implant complications were also extracted.

Statistical Analysis.
Statistical analysis was conducted using R software (RStudio 2021.09.0 Build 351 for Windows). Generally, the analysis consists of two approaches. First, we explored the single proportions of all the flap and donor site complications in each intervention arm using the metaprop package in R. e overall proportions of complications were transformed using logit transformation in a random-effects environment. Statistical differences in the proportions were assessed using a Q test to test for groupbased differences. Second, a network meta-analysis of the primary outcomes was carried out using the netmeta package [15]. A random-effects network meta-analysis was employed using the frequentist approach [16]. e network approach was utilized to combine data from studies with two or more arms and to provide reliable direct and indirect comparisons of the treatment effects. is would optimize the estimated precision and provide an insight into the relative ranking of the risk of complications while accounting for the association between effect sizes [17]. Effect sizes were expressed as the relative risk (RR) and the respective 95% confidence interval (95% CI) to determine the ratio between the incidences of a complication in a given surgical arm versus the control arm. In the present network model, the DIEP arm was considered the common 2 e Breast Journal comparator (control arm). e meta-analysis was not performed for the complications which are reported in ≤3 studies. Heterogeneity testing consisted of a within-design heterogeneity assessment using an I 2 test and a betweenstudy inconsistency analysis. Inconsistency was defined as the variation between direct and indirect evidence. It was assessed locally using net splitting and globally by constructing a full design-by-treatment interaction randomeffects model [18]. League tables were produced to express all the potential pairwise comparisons of surgical complications in off-diagonal cells [19]. Furthermore, the effect estimates of the risk of complications were visually assessed by producing network forest plots. e risk of complications was also ranked using P scores (range 0-1), where higher ranks (close to 1) indicated that the surgical technique was associated with a lower risk of complications. Assessment of publication bias was performed via producing comparison-adjusted funnel plots and was corroborated by Egger's test. A p value of <0.05 indicated statistical significance [20].

Literature Findings.
A total of 2106 publications were found after a thorough search, including 454 from MED-LINE databases, 209 from EBSCO, 2 from Scopus, 1035 from Web of Science, and 406 articles from Wiley Library. For this systematic review, we initially retrieved 69 full-text publications. However, after implementing the exclusion criteria, only 34 articles between January 2000 and November 2021 were included ( Figure 1). Out of the included studies, 5 were prospective studies [21][22][23][24][25] and 29 were retrospective chart reviews [4,9,12,. None of the selected studies were randomized controlled trials. Studies included data of 60551 patients, of whom 20052 (33.12%) underwent fTRAM, 10298 (17.01%) underwent pTRAM, 29259 (48.32%) underwent DIEP, and 942 (1.56%) underwent SIEA surgeries. More details about the characteristics of studies and patients are demonstrated in Table 1.

Network Structure.
e establishment of the network was not possible for distinct complications which comprised 2 intervention arms only (flap intraoperative arterial thrombosis and intraoperative venous thrombosis). Additionally, the following complications were excluded because they were reported in ≤3 studies: fat necrosis, hematoma, seroma, abdominal fat necrosis, and umbilical necrosis at the donor site, as well as seroma, abdominal flap necrosis, and post-operative arterial/venous thrombosis at the flap site. e network structures of the remaining outcomes are illustrated in Figure 2, and the numbers of patients in each comparison are listed in Table 2. All the networks comprised direct comparisons among different surgical techniques except donor hernia/bulge (no direct pairwise comparisons between pTRAM and SIEA) and flap total thrombotic events (the outcome was not investigated in pTRAM procedures, Figure 2).

Testing for Heterogeneity and Inconsistency.
e analysis of within-design heterogeneity, which reflects the variation in true effect sizes between studies with the same design, indicated a statistically significant heterogeneity in the effects sizes of flap fat necrosis (Q � 88.07, p < 0.0001), flap hematoma (Q � 47.49, p < 0.0001), and partial flap loss (Q � 43.94, p < 0.0001). On the other hand, the variation between designs (inconsistency) was significant in four outcomes, including flap hematoma (Q � 35.21, p < 0.0001), flap total thrombotic events (Q � 15.32, p), and partial (Q � 14.85, p � 0.005) and total flap loss (Q � 12.70, p � 0.048).
ese findings supported the application of random-effects models in the network analysis (Table 2).

e Risks of Morbidity of the Donor Site and Flap Complications.
e outcomes of the risk of morbidity at the donor site and flap complications are depicted in Figure 3. Considering DEIP procedures as a reference intervention, fTRAM flaps reduced the risks of flap fat necrosis (RR � 0.60, 95% CI, 0.38 to 0.95), hematoma (RR � 0.54, 95% CI, 0.32 to 0.89), and total thrombotic events (RR � 0.18, 95% CI, 0.05 to 0.63); nevertheless, they were associated with higher risks of hernia/bulge (RR � 1.43, 95% CI, 0.92 to 2.23).
pTRAM flaps were associated with significantly higher risks of hernia/bulge (RR � 3.34, 95% CI, 1.79 to 6.23) and wound infection (RR � 1.45, 95% CI, 1.08 to 1.95) at the donor site. However, the risk of flap hematoma was lower after pTRM flap (RR � 0.52, 95% CI, 0.29 to 0.93). Finally, SIEA flap increased the risk of wound infection compared to DIEP (RR � 4.47, 95% CI, 2.02 to 9.88). More details about the risks of morbidity/complications after different surgeries relative to each other are provided in the net league tables (Table 3).

Risk Ranking.
e P score analysis was implemented in the present study to quantify the certainty that a given intervention is better than another intervention, averaged over all other interventions. is type of analysis has been used as an alternative to SUCRA scoring which has been used in the Bayesian network models [52]. Based on P scores, results indicated that the highest ranks of reducing the risk of hernia/bulge and wound infection at the donor sites were apparent after SIEA and DIEP surgeries, respectively (Table 4). Regarding flap complications, fTRAM surgeries ranked the best in terms of reducing the risk of five complications, including fat necrosis, mastectomy flap necrosis, total thrombotic events, and partial and total flap loss. In addition, SIEA procedures ranked first in reducing the risk of flap wound infection and hernia/bulge (Table 5).
3.6. Publication Bias. Visual assessment of funnel plots indicated that the eligible pairwise comparisons were asymmetrically distributed around the effect estimate, where small-sized studies (with higher standard errors) were more likely to be published when they reported high risks of complications at the donor site (hernia/bulge and wound infection, Figure 4). However, the results of Egger's test indicated a lack of statistically significant publication bias (p � 0.088 for hernia/bulge and p � 0.343 for wound infection).
ere was no risk of publication bias for the e Breast Journal

Discussion
Microsurgical breast reconstruction with abdominal-based flaps is a challenging and complex procedure and an excellent choice for patients who want to avoid implants. However, it has a high risk of wound and surgical complications at the donor and recipient sites [53]. Hence, the literature has reported no statistically significant differences between abdominal-based breast reconstruction flap-related complications or donor site morbidity [12]. As a result, higher-quality research is required to evaluate the different choices of flaps for each patient. us, we have conducted this systematic review and meta-analysis to examine and evaluate the literature on donor site morbidity and complication rates in breast reconstruction using different autologous abdominal flap techniques, including TRAM, pTRAM, fTRAM, DIEP, and SIEA. Our study includes data from 60551 patients, 20052 (33.12%) of whom have undergone fTRAM procedure, 10298 (17.01%) pTRAM, 295259 (48.32%) DIEP, and 942 (1.56%) SIEA flaps. is is in agreement with the literature that the most popular and reliable free flaps are fTRAM and DIEP [54]. We found a tendency toward a higher likelihood of abdominal bulge/ hernia after pTPAM compared to DIEP flaps, which supports the literature that losing muscle makes the fascial abdomen vulnerable to weakness [11,55]. It is well evident in the literature that following pTRAM, vascular-related complications (e.g., flap loss and fat necrosis) are more likely to occur due to the tunneling of a pedicled flap up the chest, causing blood flow restriction by kinking or compression of the superior epigastric arteries, or that the pedicle blood supply offers decreased perfusion [34,36,56]. Nevertheless, in our study, we found that pTRAM increased the risk of hernia/bulge (RR � 3.34, 95% CI, 1.79 to 6.23) and wound infection (RR � 1.45, 95% CI, 1.08 to 1.95) at the donor site. On the other hand, DIEP flaps benefit from the improved blood supply provided by the inferior epigastric veins and the flow dynamics provided by the internal mammary or thoracodorsal systems. Although free abdominally based flaps have a lower risk of complications, pTRAM flaps may still be appropriate for certain patients, especially when microsurgery is not an option due to surgeon skill, patient comfort, or patient desires based on riskweighted judgments.
In this study, fTRAM has a much higher incidence of abdominal bulge/hernia than other types of autologous abdominal breast reconstruction. is is consistent with the literature since obese patients, in comparison with nonobese, had a higher incidence of abdominal bulge/hernia with fTRAM flaps than with DIEP flaps [57]. As a result, if feasible, other techniques (e.g., DIEP flaps) need to be considered in this group of patients. Due to greater tension being placed on the weaker abdominal fascia, obesity may worsen and increase the likelihood of abdominal bulge/ hernia after fTRAM flaps [32]. Nevertheless, fat necrosis, mastectomy flap necrosis, total thrombotic events, and partial and total flap loss were the least common complications associated with fTRAM surgeries.
ere are no established criteria to aid in determining which vessels will provide adequate tissue perfusion; thus, it is recommended to conduct a high-quality study to determine which imaging techniques are best to assess the vessels' flow and anatomy to reduce the incidence of these complications.  Regarding flap fat necrosis, the risk of publication bias was considerable (both visually and statistically). For hernia/ bulge and wound infection at the donor site, the risk of publication bias was visually obvious (although not statistically significant). An essential part of our statistical analysis is assuming that patients undergoing abdominal autologous reconstruction are equivalent candidates.
ere was no selection bias in choosing which flap to use. erefore, to reduce the magnitude of this assumption, studies that exclusively evaluated only one type of autologous abdominal flap were excluded, and only studies that evaluated the different types of flaps concurrently were included. Another 8 e Breast Journal potential source of bias is the lack of standardization of the surgical techniques due to the surgeon's particular variation of each flap procedure. e main strength of this study is that it compares the four most common types of abdominal autologous flaps with donor site morbidity and complication rates in breast reconstruction. Most studies mentioned in the literature compared two flaps only. us, a future systematic review and meta-analysis are needed to evaluate the quality of the published data and compare them to each other to determine the appropriate flap choice for each patient.
Furthermore, although this study did not pay attention to the nonclinical related variables to the patient, such as preoperative physical activity, it is crucial to note that the patient's expectations and objectives are essential in selecting the appropriate flap. Due to a lack of extractable data, we could not adjust for any preoperative risk variables. More high-quality prospective cohort trials are required to estimate how other patient-related variables can impact the complication rates and overall patient satisfaction. Another drawback in our systematic review and meta-analysis is the lack of randomized control trials, owing to ethical considerations. Despite that, we recommend conducting a highquality randomized control trial that will be able to address the biases that were present in the published studies. Another limitation of our study is that the aesthetic results for the breast and the patient's satisfaction were not compared, which is an essential factor in the consideration of flap choice due to the insufficient data and standardized tools used to assess those outcomes in the literature.

Conclusion
e use of the abdominal wall for autologous breast reconstruction is an optimized procedure for both aesthetic and functional outcomes. Our findings demonstrate that the morbidity of the donor site and complications following abdominally based breast reconstruction techniques depend on the flap type and some patient-related factors. When possible, fTRAM reconstructions should be performed as they have a lower risk of complications than any other flap type compared in this study. However, using SIEA in obese patients may reduce hernia/bulge donor site risk. erefore, selecting an autologous abdominal flap is a multifactorial decision that should be patient-oriented rather than procedure-oriented to minimize donor site morbidity and complications.

Data Availability
Data are available on request to the corresponding author.

Conflicts of Interest
e authors declare no conflicts of interest.