Supermicrosurgery with perforator‐to‐perforator anastomoses for lower limb reconstructions – A systematic review and meta‐analysis

Supermicrosurgical flaps based on perforator‐to‐perforator microanastomoses have been described for lower limb reconstruction. This approach offers the benefit of raising short pedicles while sparing axial vessels, which effectively enables complex reconstructive techniques in comorbid patients at high risk of reconstructive failure. The aim of our study is to assess the surgical outcomes of perforator‐to‐perforator based flaps in comparison to conventional free flaps for reconstructions of the lower limb district, through a systematic review of literature and meta‐analysis.


| INTRODUCTION
Lower limb soft tissue defects consistently overload plastic surgery practices, due to the heavy burden of trauma-related injuries, posttraumatic infections, diabetic foot-related issues, and oncologic outcomes for cancer resections.Microsurgery is now routinely performed for the treatment of these defects, having moved away from conventional workhorse flaps to sophisticated perforator-based free tissue transfers (May et al., 1981), including the antero-lateral thigh (ALT) flap and the superficial circumflex iliac perforator (SCIP) flap (Berner et al., 2020).The latest frontier in microsurgery is currently represented by the use of supermicrosurgical techniques (Koshima, 2008;Koshima et al., 2010).Microanastomosing vessels and single nerve fascicles as small as 0.3-0.8mm in caliber is just one of the benefits (Badash et al., 2018), though this requires manipulation of 30-80 μm needles and micro-sutures as small as 13-0 (Yamamoto et al., 2017).
The technical advancement of anastomosing <1-mm vessels has made possible what was previously considered unachievable within the field, cementing the success of microsurgery and setting the bar of satisfactory results higher than ever.Perforator-to-perforator supermicrosurgery is the next evolution of the technique, which consists in using perforators as the recipient vessels for microanastomoses, thus overcoming the need for dissecting axial vessels (Hong, 2009;Mureau & Hofer, 2008).The benefits of this approach include an easier flap dissection and the ability to spare main vascular pedicles in lower limbs, which is a considerable advantage since lower limb reconstruction patients commonly present impaired vascularity, either due to comorbidities and/or trauma-related injuries (de Laat et al., 2018).
Nevertheless, the technical difficulties make these types of procedures particularly complex and place them beyond the reach of most routine plastic surgery practices and reconstructive units around the world.
The results of modern microsurgical procedures have improved over the years, with low donor site morbidity and high flap success rate (Park & Chang, 2016).It stands to reason whether supermicrosurgery procedures are associated with a higher rate of flap-related complications and are thus riskier than the more commonly performed, "standard" microsurgical procedures (with axial vessel anastomoses).
We conducted a systematic review summarizing the state of the art of scientific literature on supermicrosurgery using perforatorto-perforator anastomoses for reconstructions of lower limb defects, and performed a meta-analysis comparing the flap-related outcomes and complications of microsurgical flaps based on axial vessels versus perforator-to-perforator based flaps.

| MATERIALS AND METHODS
We performed a systematic review of literature in accordance with PRISMA guidelines.We searched for publications on PubMed, Embase, Web of Science (including Science Citation Index and Conference Proceedings Citation Index), and Cochrane Library databases to identify all publications regarding lower limb reconstructions using perforator-to-perforator supermicrosurgery techniques.For PubMed, the following search term strategy was used: (perforator-to-perforator The following data were extracted from the manuscripts included in the final tally: number of patients, gender, mean age, comorbidities, indication for surgery, location of defect, size of defect, type of flap used for reconstruction, recipient vessels used, flap complications, donor-site complications and follow-up.In the meta-analysis, the following outcomes were evaluated: flap survival rate (total flap losses to total number of flaps ratio) and partial flap loss rate.Studies with a sample size of four or more flaps and that directly compared the outcomes for perforator-to-perforator anastomosis versus axial vessel anastomosis were included in the metaanalysis.The level of evidence for included studies was evaluated using the Oxford Centre for Evidence-Based Medicine (OCEBM) (Howick et al., 2011), and the Oxford quality scoring system (Jadad Score) (Jadad et al., 1996).

| Statistical analysis
When available, hazard ratios (HRs) were recorded from each study.
When not available, hazard ratios and their standard errors were computed according to the methods described by Tierney et al. (2007).All data was provided with a confidence interval (C.I.) of 95%.Statistical significance was considered at p-value <.05.Several separate metaanalyses were performed for each outcome.Given the small number of studies involved, it was not possible to reasonably assess study heterogeneity.Considering that, a hierarchical Bayesian approach with heterogeneity and informative priors was used, according to guidelines (Higgins et al., 2009), and as previously published in articles under similar circumstances (Iocca et al., 2017;Proietti et al., 2018Proietti et al., , 2020)).
T A B L E 1 Quality assessment of included studies and overview of patient population and outcomes.Funnel-plots, Egger's regression test were used to assess publication bias for the flap success rate (Egger et al., 1997).All analyses were conducted with R version 4.0.4,using the metafor and adaptMCMC packages.

| Study selection
From 483 starting citations, we identified 63 articles following the first screening based on the assessment of titles and abstracts.Any citation deemed not relevant to the systematic review endpoints was excluded.During the second screening, duplicates were excluded and manuscripts not meeting the inclusion criteria or meeting the exclusion criteria were discarded, only leaving 34 articles.After full-text assessment, any manuscript that did not provide clinical data of a patient population undergoing supermicrosurgical free flaps for lower limb reconstruction, and namely data on flap-related outcomes, was excluded.Thus, 16 manuscripts remained and are listed in Table 1 ( Goh et al., 2015;Hong, 2009;Hong et al., 2013Hong et al., , 2014;;Hong & Koshima, 2010;Kim et al., 2010Kim et al., , 2015;;Kim & Kim, 2012;Koh et al., 2018;Power et al., 2022;Scaglioni et al., 2021;Seo et al., 2015Seo et al., , 2018;;Suh et al., 2016;Tashiro et al., 2014;Yoon et al., 2014).Flowchart representation of the search strategy with the included and excluded articles is depicted in Figure 1.stumps, vascular malformations and immune disorders, which were reported in 100 cases (9.6%) (Figure 2).Defects were most commonly located in the lower third of leg and/or foot, which was reported in 554 patients (52.9%).Other less common locations include the knee and upper third of leg (14.2%) followed by the middle third of leg (10.1%).The least common location was the thigh, which was reported in just 41 cases (3.9%), but location was unknown in 197 cases (18.9%).Most defect sizes were moderate (between 10 and 50 cm 2 ), and were covered with flaps which were 84 cm 2 in mean size.Reconstructive flaps included the SCIP, the ALT, the Upper Medial Thigh (UMT), the Inferior Gluteal Artery Perforator (IGAP), the Superior Gluteal Artery Perforator (SGAP), the Peroneal Artery  F I G U R E 3 Distribution of all complications in p-2-p supermicrosurgical procedures.

| Patient characteristics
F I G U R E 4 Forest plot depicting partial, total and overall flaprelated complications.

| Meta-analysis
Among the 16 identified articles from the systematic review, 5 had an OCEBM level of 3 (i.e., non-randomized controlled cohort/follow-up studies) and 11 had a level of 4 (i.e., case-series, case-control, or historically controlled studies).No study had a Jadad score above 0 as none of the featured articles met the methodological requirements of a clinical trial by objective criteria.While all 16 articles were initially considered for the meta-analysis, only three were included after review of full-text manuscripts in accordance with inclusion and exclusion criteria, as they were the only manuscripts to successfully compare perforator-to-perforator and regular microsurgical anastomoses in lower limb reconstructions.The meta-analysis compared 482 perforator-2-perforator flaps to 451 conventional free flaps, assessing flap-related complications.Overall flap complications had a HR of 1.41 (0.94-2.11; 95% C.I.).Supermicrosurgical and conventional microsurgical reconstructions were not associated with statistically significant differences ( p = .89)(Figure 4).Egger's test did not demonstrate presence of publication bias (Figure 5).

| DISCUSSION
Supermicrosurgery has been gradually adopted worldwide in cases where adequate major vessels are not available as recipient vessels, only leaving small caliber vessels and nerves for anastomosis.The surgical innovation that is supermicrosurgery has made certain advancements possible, including improved nerve coaptation (Iida et al., 2014), secondary free flap salvage (Cereceda-Monteoliva et al., 2018), and lymphatico-venular bypasses in lymphatic surgery (van Mulken et al., 2018).The potential of supermicrosurgery shines particularly bright with the implementation of perforator-to-perforator free tissue transfers (Zeiderman & Pu, 2021).A perforator-to-perforator approach has been proven to be a reliable and safe technical approach to soft tissue defect coverage (Escand on et al., 2023).The most relevant finding from our study is that no statistically significant differences could be found between flap complications in conventional and perforatorto-perforator anastomoses, despite the notorious technical difficulties of the latter.A relevant advantage of perforator-to-perforator based flaps in lower limbs is related to the infamously complex nature of the lower limb reconstruction population, which if often characterized by either comorbidities which damage peripheral blood vessels and microcirculation, or by impaired limb vascularity caused by injury to one of the main lower limb vessels which is typical in patients with a history of trauma and/or amputation.In fact, our study population was characterized by diabetes mellitus in as many as 43.0% of patients, followed by peripheral vascular disease in 14.5% and high blood pressure in 12.0%.
The study is however limited by the fact that neither the vascularity of the affected limb in trauma cases nor the effect of comorbidities on flap-related complications could be assessed due to a lack of data provided by the full-text manuscripts selected for analysis.Nevertheless, having optimal recipient vessels can be challenging, and a perforator-to-perforator technique can overcome the need for intact axial vessels, especially in circumstances where their blood supply supports the vascularity of the entire limb and thus cannot be expendable.Additionally, its overcomes the need for tedious dissections of long pedicles which is ought to reduce morbidity associated with long and delicate dissections.We speculate that a perforator-to-perforator approach might also help avoiding the risk of avascular complications to the distal segment related to thrombosis of the anastomosis.Furthermore, clinical appearance of the perforator (good pulsation) and blood velocity in its lumen (>15 cm/s with duplex scans) have been recognized as the best predictors of success for a super-microanastomosis, thus more invasive pre-operative diagnostic tests (i.e., angiography) could be avoided and applied only for selected cases.With these two factors, in fact, flap survival rate is reported to be more than 90% in patients with peripheral vascular disease (Hong, 2009;Hong et al., 2014).Escand on et al.'s findings resonate with our own, as the main takeaway from their study suggests that case series of supermicrosurgical procedures used in lower limb reconstructions have a success rate comparable to that which can be found in standard microsurgical procedures.Even so, our findings are limited by the fact that they could F I G U R E 5 Funnel plot for flap-related complications.

Cereceda
not highlight disadvantages directly related to technical aspects of supermicrosurgical perforator-to-perforator flaps.This is particularly relevant since several authors have addressed some concerns with the technique and its possible limitations.In fact, the ability to raise perforator-to-perforator flaps and to successfully revascularize them requires long learning curves, experience and very specific training (Liu, 2013), all of which are neither accounted for nor addressed by any of the included manuscripts, in which surgeries were performed by senior surgeons with recognized expertise in the field.We believe that expertise is the mother of success in microsurgery, as in surgery overall.It is likely that the high flap success rate and low flap complication rates have been achieved by the authors of the manuscripts assessed in this review through their expertise and surgical capabilities honed over the years.
Additionally, Badash et al. (2018) discussed an alleged increased inherent risk of thrombosis in small-caliber perforator thrombosis which ultimately leads to flap necrosis.This could be due to several reasons: a higher risk of caliber mismatch due to uneven lumens (Fensterer et al., 2014;Zhou et al., 2010), turbulent blood flow due to vessel wall irregularity leading to turbulence-induced thrombogenicity (Doh et al., 2021;Krijgh et al., 2021), or vessel wall collapse due to low inner lumen pressures, since blood flow rate proportionally decreases with reduction in vessel diameter (Drzewiecki et al., 1997).
Finally, another concern has been raised by Kim et al. (2015), according to whom the use of perforator-to-perforator should be limited to small-sized flaps, as a single low-caliber vessel may not be adequate to perfuse across large flap areas.While these concerns have not been properly substantiated, as demonstrated by the fact that flaps in our meta-analysis had a mean size of 84 cm 2 , we still believe that the relevance of these concerns should be weighed in future evidence regarding the use of perforator-to-perforator flaps.
Finally, while our findings were characterized by relatively a short follow-up period of 16.4 months on average.We believe that this was still enough to assess the surgical safety of perforator-to-perforator technique since most free flap failures occur within the first 48 h from surgery (Novakovic et al., 2009).Late failures (defined as vascular compromise occurring after the first 48 h) are mainly linked to infection or mechanical stress/pressure on the pedicle and around the anastomosis (Wax & Rosenthal, 2007).Our study was conclusive in proving the safety of perforator-to-perforator supermicrosurgery for lower limb reconstructive surgery.Nevertheless, we believe that the meta-analysis was also limited by the low quality of overall evidence, as suggested by OCEBM and Jadad Score results, summarized in Table 1.This is purportedly due to lack of high-level evidence in this specific field of plastic surgery, which will hopefully be addressed by implementing clinical trials in research to come.

| CONCLUSION
The use of perforator-to-perforator supermicrosurgery has been compared to conventional free flap transfers for soft tissue reconstruction of the lower limbs.Our evidence supports the safety of surgical outcomes, with acceptable flap complication rates.Nevertheless, these findings are limited by poor overall quality which must be addressed and used to encourage higher-level evidence in the field.

CONFLICT OF INTEREST STATEMENT
Nothing to disclose.

[
MeSH] OR supermicrosurgery [MeSH]).For Embase, we used: ((perforator AND to AND perforator/exp) OR (supermicrosurgery*) AND (lower AND limb [MeSH] OR (leg [MeSH]) OR (foot [MeSH]))).For Web of Science, we used: (Topic = (perforator AND to AND perforator*) OR Topic = (supermicrosurgery*)).Finally, for Cochrane Library ((perforator AND to AND perforator*) OR (supermicrosurgery*)) were used.All citations were screened through their titles and abstracts, duplicates were removed, and then full-text manuscripts were assessed according to the following inclusion criteria: only humanbased topics and manuscripts written in English were to be taken into consideration.Case reports and case series with fewer than three patients, non-clinical articles addressing supermicrosurgery training or instrumentation and clinical cases addressing lymphedema treatment or treatment of anatomical districts other than the lower limb were used as exclusion criteria for this review.Review search started on March 2022 and ended in July 2022, and was conducted by G.F. and L.P. The two reviewers independently reviewed the titles and abstracts yielded by this comprehensive search and subsequently selected articles based on the predetermined inclusion and exclusion criteria.Disagreements were resolved by a third reviewer (P.P.) or through consensus-based discussion with a fourth reviewer (F.L.T.).

F
I G U R E 1 Flow diagram representation of the search strategy used for the systematic review and meta-analysis, in accordance with PRISMA guidelines.F I G U R E 2 Distribution of various indications for p-2-p supermicrosurgery for soft tissue coverage of the lower limbs.Perforator, the Soleus Perforator, the Medial Sural Artery (MSA), the Posterior Interosseous Artery Perforator (PIAP), the Anterior Tibial Artery perforator (ATAP), the Thoraco-Dorsal Artery Perforator (TDAP), the Latissimus Dorsi (LD), and the Radial Forearm flaps.The most commonly reported reconstructive option was the SCIP flap, followed by the ALT flap.Median flap thickness was 5.5 mm (range: 2.5-12.0mm), while pedicle length was 4.1 cm (range: 1.0-8.0cm).Patient comorbidities and other demographics and surgical characteristics are reported in -Monteoliva et al. (2018) conducted a meta-analysis counting 47 manuscripts to assess success rate of reconstructive supermicrosurgery in various anatomical districts.Out of 698 flaps, 69.2% were used for soft tissue coverage of the lower limb.They reported 3.84% of cumulative partial flap loss rate and an overall vascular complication rate 5.93% which resulted in complete or partial flap loss.When specifically addressing lower limb complications, they found that 22 flaps experienced partial flap loss (4.6%) while 18 had complete flap loss (3.7%).These findings are in line with our own, based on 1047 perforator-to-perforator flaps, where total flap failure occurred in 71 cases (6.8%) and partial flap failure in 47 cases (4.5%).