Muscle-only versus chimeric muscolocutaneous gastrocnemius pedicled ﬂap in complex orthoplastic reconstructions of the knee region: a retrospective study

Background: The knee region represents a challenging area of soft tissue reconstruction. Specifically, in the context of total knee arthroplasty (TKA) or following high energy trauma with fractures and hardware fixation, soft tissue defects can expose critical structures such as joint, bone or tendon, besides the implant/plates itself, with dramatic consequences in terms of postoperative infection and hardware contamination. Methods: A retrospective study was conducted on a prospectively maintained database from January 2016 to February 2021. Inclusion criteria involved all patients who underwent an implant-associated infection of the knee and upper third of the leg coupled with a soft tissue reconstruction (STR) using the traditional gastrocnemius muscle (GM) pedicled flap or the chimeric GM-MSAP (medial sural artery perforator) flap. Results: 38 patients were included (Group A, GM flap, 22 patients; Group B, chimeric GM-MSAP flap, 16 patients). No statistically significant differences were detected in terms of age, comorbidities, defect size, follow-up and flap complications. A statistically significant difference was seen among the groups in terms of successful flap re-raise (required because of a persistent infection of the implant or in a two-stage procedure setting, including the use of a cemented spacer) in favor of the GM-MSAP group. Conclusion: The chimeric GM-MSAP, being safer to reraise if required, can be a significantly more powerful tool in those cases in which a two-stage procedure is planned or when there is a high probability for secondary intervention need, reducing the need to convert in either free flap coverage or amputation.


Introduction
The knee region represents a challenging area of soft tissue reconstruction. The coverage should be stable, while preserving mobility. Specifically, in the context of total knee arthroplasty (TKA) or following high energy trauma with fractures and hardware fixation, soft tissue defects can expose critical structures (such as joint, bone or tendon) besides the implant/plates itself, with dramatic consequences in terms of postoperative infection and hardware contamination (1,2). Considering the increasing numbers of such procedures, the incidence of wound breakdown, infection and major surgery complications is expected to raise accordingly (2).
When possible, a local coverage should be preferred, leaving the free tissue transfer as a back-up option (2,3).
The gastrocnemius muscle is the workhorse flap to reconstruct soft tissue defects in the knee and upper leg area, especially when hardware, bone, tendon or joint capsule are exposed (3). The ease of harvest and the minimal donor-site morbidity, together with its capacity to fill dead space are the major advantages of this technique, which has proven to be effective also in posttraumatic injuries with exposed framework and TKA salvage (4).
However, gastrocnemius muscle is sometimes insufficient to reach supero-lateral and supra patellar defects, being sometimes particularly narrow or tendinous in its distal part, making coverage less stable. Moreover, the need to skin grafting makes postoperative immobilization and healing longer (4,5). Furthermore, another limit of muscle-only gastrocnemius flap resides in the fact that the muscle component may incur into shrinkage and atrophy over time, making re-raise difficult when hardware materials (including plates and screws, TKA or temporary cemented spacer) need to be exchanged or removed.
Despite its description long time ago (3), the musculocutaneous gastrocnemius flaps has had limited success, probably due to its bulkiness and inset difficulties when harvesting the whole skin over the calf area. Lately, chimeric principles and perforator/propeller flaps recently opened new tailoring possibilities in complex knee reconstructions (6). Indeed, chimeric flaps can incorporate multiple tissue types (e.g. muscle and skin), which are potentially connected only by branching or perforating vessels (7). A chimeric gastrocnemius-MSAP flap can therefore serve for the double aim of filling a deep defect, while at the same time addressing skin shortage with the fasciocutaneous component (8).
Such property may be critical when secondary procedures are needed, demanding further flap reraise due to framework exchange or resistant infection. This work compares for the chimeric GM-MSAP flaps and the traditional GM pedicled flaps in specifically complex orthoplastic scenarios, where TKA explant following infection or multiple-stage surgeries due to chronic osteomyelitis were necessary, with limited skin availability and complex infectious conditions. Surgical technique and orthoplastic planning have been described. Outcomes and complications have been critically analyzed. Moreover, all patients without complete follow-up (at least 6 months since last procedures) were excluded from analysis. Patient's demographic data and comorbidities were gathered from medical and anaesthetic charts. Operative notes were screened for technique details.

Material and Methods
Hospital letters and outpatient reports were used to evaluate the operative time, the hospital stay, as well as flap complications and the number of surgeries each patient underwent.

Orthoplastic approach
The duration of infection and the cause dictated the treatment concept of the orthopaedic team (one-stage revision vs. two-stage procedure) (9,10). In one-stage revision procedures the STR was performed directly after the debridement, antibiotics and either implant retention, implant exchange or implant removal procedures. One-stage revision procedures were routinely performed in the acute setting of implant infection (<6 weeks from onset of symptoms/diagnosis). In patients who underwent a two-stage procedure, the flap was used at the first timepoint after implant removal and spacer implantation (as coverage over the spacer) or at the second time point, after spacer removal, implant placement and final coverage. The flap timing was dictated by patient needs and time of presentation to our hospital (same patients have already done one stage of the surgery in other centres).

Surgical Technique
GM and GM-MSAP flaps were harvested according to previous literature (3,8). Once the flap was harvested and rotated into the defect, the skin paddle, could be further propelled up to 90° to extend proximally and distally the flap coverage surface. Donor site was closed directly or skin grafted, depending on skin paddle size.
Flap re-raise for second stage procedures were generally performed 6-8 weeks after debridement and cement spacer placement. When the skin paddle was present, no undermining between the skin and the muscle was performed. Outpatient follow-up was performed at one, three or six weeks after STR, followed by monthly assessments for at least 6 months.

Outcome analysis
A successful STR was considered when no further flap procedures were necessary to ensure bone or hardware coverage.
Complications were listed as major and minor. Major complications were considered full or partial flap necrosis (at least 1/3 of the flap, implying new plastic surgery procedures), while minor complications were considered partial flap necrosis (less than 1/3 of the flap, and maintained vascularization allowing STSG). Early complications were defined as complications occurring within 6 weeks after flap surgery (first or second stage). Late complications were defined as complications arising between 6 weeks and up to 4 years.
The study was conducted accordingly to the guiding principles following the Declaration of Helsinki of 1975. Informed consent was obtained from all patients, including approval for scientific publication and photographic\video documentation.
The manuscript was written in accordance with the STROBE checklist. (11)

Statistical analysis
All investigated parameters were statistically analysed (average, range, standard error of the mean). Patients' groups were compared using independent two-sided t-tests for means, Mann-Whitney U tests for medians and two-sided Chi-square or Fisher's exact test as appropriate to analyze categorical variables. The assumption of normality was verified using the Shapiro-Wilk test. Statistical significance was set at a p-value <0.05. Statistical analysis was performed using GraphPad Prism (version 8.0, GraphPad software, La Jolla, CA).

Results
More than 180 soft tissue reconstructive procedures using pedicled flap on the leg were screened. After application of inclusion criteria, 39 patients were included. Patients were divided in two groups depending on the presence of the MSAP skin paddle (Group A, GM flap; Group B, chimeric GM-MSAP flap). Please refer to Table 1-2 for patient's data and characteristics.

GM Group
The mean age of patients in GM group was 60  13 y.o (average  standard error of the mean, SEM). 9 patients out of 22 (41%) required a GM flap coverage following knee trauma and internal fixation. Among these, 7 benefited of a one-stage procedure including surgical debridement, hardware ablation and soft tissue coverage, while in 2 patient the bone debridement required the use of a cemented spacer in a two-stage procedure (Masquelet procedure).
13 patients (59%) presented an infection following a TKA. Among these, a one-stage procedure was attempted in 9 out of 13 patients (the TKA was either completed replaced or only the mobile parts were changed, according to orthopedic decision), while in 4 patients a two-stage procedure was performed including the temporary use of a cemented spacer. Please refer to table 1 for GM group patients results.

GM-MSAP Group
The mean age of patients in GM-MSAP group was 64  16 y.o (average  SEM). 4 patients out of 16 (25%) required a GM-MSAP flap coverage following knee trauma and internal fixation. Among these, 3 benefited of a one-stage procedure, while in 1 patient a two-stage procedure (Masquelet procedure) was attempted.
12 patients (75%) presented an infection following a TKA. Among these, a one-stage procedure was attempted in 5 out of 12 patients, while in 7 patients a two-stage procedure was performed.
In case of two-stage procedures, flap coverage was performed concomitantly with the first stage in 7 out of 8 patients (88%).
Among flap complications, we recorded 1 case of flap wound dehiscence and 1 case of distal flap necrosis, both managed successfully without need of further major surgical procedures. One case of donor site infection was recorded following skin graft and was addressed with a superficial debridement and a new STSG. The donor site was closed primarily in 7 patients (44%), while in the remaining 9 a skin graft was necessary.
We recorded 5 cases of persistent infection following the flap coverage procedure (in 3 cases following a one-stage implant exchange) and flap re-raise was necessary in 11/16 patients (68%). The average number of surgeries per patient was 3.5.
Nevertheless, the flap re-raise, eventually associated to a further implant exchange or an extensive debridement/washout when implant infection was persistent, was successful in all cases and no further flap procedures were necessary (Fig. 2).
Please refer to table 2 for GM-MSAP group single patient details.

Groups Comparison
No statistically significant differences were detected in terms of age, BMI, ASA score, diabetes and follow-up. Similarly, the number of smokers was comparable in the two groups, despite a positive trend seen in favor of the GM group (p 0.08), which however did not correlate with the incidence of persistent material infection or flap complications (table   3).
The defect size and the operative time were super-imposable in the two groups (p value 0.9). No significant differences were seen in the distribution of one-stage vs two-stage procedures among groups, despite a trend towards a greater number of staged procedures recorded in the GM-MSAP group (50% vs 27% in the GM group).
In terms of flap complications, no significant differences were seen among the two groups, as both flaps were reliable and provided efficient tissue coverage (p 0.4) (Figure 3).
The number of persistent infections of the implant following the STR showed a discrete trend towards a decrease in infection in the GM-MSAP group (p value 0.1), but did not reach statistical significance. The need for flap re-raise was highly comparable in the two groups, despite a higher need in the GM group (p value > 0.99), matching the higher number of persisting infection in such group. The number of total surgeries and unplanned surgeries per patient amongst groups did not show significant differences. Importantly, a statistically significant difference was seen among the groups in terms of successful flap re-raise (p < 0.05) in favor of the GM-MSAP group. Final limb salvage, even including supplementary procedures such as free tissue transfer was 100% in the GM-MSAP group and 90% in the GM group (Figure 4).

Discussion
TKA is a recognized procedure for the management of disabling knee arthritis with successful outcomes resulting in marked pain relief and improved patient functionality.
Studies have cited survivorship of TKA of over 90%, 80%, and 70% at 10, 15, and 20 years, respectively (10). With an increasing elderly population, the number of primary TKAs is projected to increase 673% by 2030, and revision total arthroplasty will likely mirror this trend, especially as patients continue to live longer. In this contest, the incidence of wound healing complications and supplementary major surgery in the long term is increasing as well (12) (8).
As well, the risk of IAOM (Implant-Associated OsteoMyelitis) after fracture fixation depends on the type and site of injury, and particularly on the soft tissue damage (13). It is especially high in patients with open fractures of the tibia, multiple injuries, high-energy trauma, vascular injury, and late admission to a trauma treatment centers. While in the thigh and in the upper limb it is often possible to close the skin directly and muscles provide a healthy cover, in the leg, high-energy fractures can disrupt the limited soft tissues, producing major compromise (14). Over the tibia, even when skin closure can be achieved, surrounding tissues are often insufficient to guarantee sufficient vascularity and coverage, with the risk of bacterial adherence and biofilm formation on the framework (15).
Adequate coverage of the knee and the upper third of the leg can be extremely challenging in case of foreign material infection (following TKA or internal fixation) or bone infection with osteomyelitis (1,14). Extensive aggressive debridement is of paramount importance for infection eradication, however, vascularized soft tissues around the implant critically impact on the reconstruction as they provide blood flow to the area, allowing delivering systemic antimicrobial agents, immune cells, and antibodies (16).
As an evolution of the originally described myocutaneous GM flap (3), Hallock described the potential use of the pedicled chimeric GM MSAP flap in 2008 (6). These concepts were further developed by other groups, exploiting flap chimerism (17) and as a useful tool for knee reconstruction where the defect is too large or too complex for a single medial gastrocnemius muscle flap. Indeed, the perforator nourishing the skin paddle could serve for tailored reconstructions according to the propeller flap principles (8). However, despite case reports and small series, no article specifically focused on knee complex prosthetic surgery and revision surgery in case of implant contamination. Moreover, no study in literature quantified or compared outcomes between simple gastrocnemius muscle flap and its chimeric modification including the skin paddle. This paper analyses two patient cohorts in term of age, patients comorbidities, orthopedic surgical indication for knee prosthetic revision and soft tissue defect size.
Flap surgery was ideally performed in the first surgical step in order to maximize the time for the soft tissue to heal and integrate. Moreover, an early flap placement over a spacer could assure well-vascularized tissue to act synergistically with the antibiotic-loaded cement to deliver antimicrobial agents to the site of infection. However, this was not always possible as patients were often referred after an initial surgical attempt were an exchanged TKA was already placed (and contaminated after a wound dehiscence) or a spacer has already been placed but soft tissue were precluding a safe definitive Despite the promising results, some limitations need to be acknowledged. Particularly, due to small numbers of patient enrolled in the study, the statistical significance our findings is underpowered and therefore larger studies are required in the future to validate our results.

Conclusion
Implant-associated infections of the knee and upper third of the leg often require multiple re-interventions and represent a real limb-threating complication.
This study compares for the outcomes of gastrocnemius muscular flap and chimeric musculo-cutaneous MSAP in soft tissue reconstruction of the knee.
Both techniques provided efficient coverage in this complex scenario. According to our results and clinical experience, the chimeric musculo-cutaneous MSAP flap can be a significantly more powerful tool in those cases in which a two-stage procedure is planned or when there is a high probability for secondary intervention need, as the skin paddle of the flap guarantee a more reliable scarring process (skin-skin vs skin-grafted muscle),  Thanks to the skin paddle, a tension free skin closure was achieved (F).