Large-defect Resurfacing: A Comparison of Skin Graft Results Following Sarcoma Resection and Traumatic Injury Repair

Introduction. Soft tissue sarcomas are rare neoplasms, and most plastic surgeons do not commonly resurface large tissue defects after a wide resection of these tumors. Objective. The purpose of this study is to elucidate the clinical results of large skin grafts after wide sarcoma resection by comparison with grafts for traumatic skin defects. Materials and Methods. A retrospective review was performed of patients who received skin grafts > 50 cm² after traumatic injury or wide sarcoma resection from 2014 to 2016. Patient medical records were reviewed; graft take rate, graft loss, and days to complete epithelialization were compared between the 2 groups. Results. In the sarcoma group (n = 8), 5 grafts were partially lost; the sarcoma group mean graft take rate of 67.5% ± 30.0% was significantly lower than that of the trauma group (n = 7) at 99.6% ± 1.1%. The mean time to complete epithelialization from the skin graft placement in the sarcoma group was 113.3 ± 66.0 days, which was significantly longer than that of the trauma group (40.3 ± 38.0 days). Wounds located around the shoulder joint in 2 sarcoma group patients did not heal even after 300 days of conservative treatment; 1 required a secondary flap. Conclusions. The results of skin grafting for resurfacing large defects after sarcoma resection are inferior to those for traumatic injury repair. Skin grafts may fail because the blood supply for the wound bed is impaired during resection. Furthermore, due to the wound bed movement, epithelialization over muscles of the shoulder joint is difficult to achieve, and skin grafts in this region will likely fail.

RELATED CONTENT

Comparison of Efficacy and Complications Between Negative Pressure Wound Therapy and Conventional Mechanical Fixation in Skin Grafts: A Retrospective Analysis

Autologous Skin Grafting in a High-throughput Clinic Evaluation of Two Novel Handheld Devices

Defect Reconstruction of an Infected Diabetic Foot Using Split- and Full-thickness Skin Grafts With Adjuvant Negative Pressure Wound Therapy: A Case Report and Review of the Literature

Coverage procedures for large soft tissue defects include primary closure, skin grafts (SGs), local flaps, regional flaps, and free tissue transfer. Of these options, skin grafting is a simple, effective procedure for wound coverage commonly used in trauma, wound infection, and resection of skin carcinoma.¹ Skin grafting can also be used for wound coverage after wide resection of soft tissue sarcomas,² and adequate soft tissue reconstruction is critical to surgical success.^3,4 

Because sarcoma is a rare carcinoma,⁵ patients with a suspected soft tissue or bone sarcoma are generally referred to the regional soft tissue sarcoma unit to be managed by a multidisciplinary team of sarcoma specialists.⁶ Therefore, the use of SGs in resurfacing large tissue defects is generally unfamiliar to most physicians except those in the regional units.

The wound bed after sarcoma resection has different characteristics from those of other conditions in which skin grafting is indicated. Because sarcomas arise from subcutaneous or deeper tissues and are excised along with at least 1 to 2 cm of the normal tissue surrounding the tumor, the wound is far deeper than after skin carcinoma removal.⁷ There is generally no wound bed preparation in sarcoma resection because, in principle, reconstruction should be achieved as a 1-stage procedure at the time of resection.² Resection wounds are made by deliberate, sharp dissection in the operating room and are sterile, unlike wounds from trauma, burn, or infection, which can be contaminated, crushed, and lacerated.

Based on these facts, the authors think skin grafting after sarcoma resection should be given particular attention; however, there is little information about its use in resurfacing large defects. The purpose of this study is to elucidate the characteristics and clinical results of large SGs for the reconstruction of tissue defects resulting from sarcoma resection by comparing them to SGs for traumatic skin defects.

With approval of the Kyushu University Institutional Review Board for Clinical Research (Fukuoka, Japan) and written informed consent from patients, a retrospective review was performed of patients who received SGs > 50 cm² for wound closure of sarcoma resection defects or traumatic skin defects from January 1, 2014, to December 31, 2016. Patient medical records were reviewed, and the following data were retrieved: age, gender, comorbidities, smoking status, size and site of the SG, secondary procedures, postoperative immobilization method, duration of immobilization, and day of complete epithelialization.

The following specific data in the sarcoma group were reviewed: extent of resection, histology, and patient’s current status. The following specific data in the trauma group were also reviewed: type of injury, number of debridements, and time from injury to skin grafting. 
 

Operative procedure in the sarcoma group
All tumors were resected with a wide margin and with overlying skin. The first author performed all skin grafting procedures and followed the patients. Skin grafting was performed immediately after sarcoma resection when exposed vital structures or tissues that have poor blood supply (such as tendon without paratenon, cartilage, and ligament) could be covered with local skin or muscle. Split-thickness skin graft (STSG) harvesting was done with an electric dermatome (Keisei Medical Industrial Co, Ltd, Tokyo, Japan) set at 0.4 mm and meshed using a mesh dermatome (Keisei Medical Industrial Co, Ltd). In patient 6, the authors converted the annular-shaped wound into a fusiform-shaped defect; 2 triangles were excised from opposite ends of the wound and used for a full-thickness skin graft (FTSG, Burow’s graft).⁸ 
 

Operative procedure in the trauma group
Wound cleansing and debridement were performed several times until necrotic tissue was cleared and covered using a negative pressure wound therapy (NPWT) dressing. The first author performed all but 1 (patient 12) procedure and followed the patients. The STSGs were harvested in the same manner as the sarcoma group. The authors did not debride granulation tissue immediately before skin grafting. Granulation tissue was washed with saline and the SG was placed over it. In patient 9, the defect was relatively small (58.9 cm²); a FTSG was harvested from the ipsilateral groin, with primary closure of the donor site.
 

Clinical photographs
Clinical photographs were taken once or twice a week at the time of the dressing change while patients were in the hospital. After discharge, the patients were examined once or twice each month, and clinical photographs were taken at that time.
 

Graft fixation
Grafts were placed at the recipient site and fixed with skin staples or with simple interrupted sutures using 4-0 nylon and multiple tacking sutures. A nonadherent dressing (ADAPTIC Non-Adhering Dressing; KCI, an Acelity Company, San Antonio, TX) was applied under a standard black polyurethane foam (GRANUFOAM Dressing; KCI, an Acelity Company; or Renasys; Smith & Nephew, Hull, UK), and NPWT was set to continuous negative pressure between -100 mm Hg and -125 mm Hg. Negative pressure wound therapy for securing the grafts was discontinued on postoperative days (PODs) 3 to 8, and saline wet-to-dry dressings were continued until the wound healed.
 

Assessment of graft fixation
A graft was considered successfully fixed (graft fixation) if it was attached to the wound and immobile. Graft fixation was judged at the time of NPWT removal (PODs 3-8). The graft fixation rate was determined by physician evaluation and recorded in the medical chart. In cases in which there was no description of a physician evaluation in the medical chart, clinical photographs were reviewed, and graft fixation rate was calculated by analyzing the photographs of the recipient sites using Photoshop software CS5, Version 12.0 (Adobe Systems Inc, San Jose, CA). The number of pixels within the total SG area was determined. Then, the number of pixels within the fixed graft areas were determined. The percentage of graft fixation was calculated by dividing the number of pixels in the fixed graft by the total SG site pixels and multiplying the quotient by 100.⁹

Assessment of graft take
Clinically, a SG that had a healthy color and remained attached to the wound bed after NPWT removal was considered a take. Graft take rate was determined by physician evaluation and recorded in the medical chart. In cases in which there was no description of a physician evaluation in the medical chart, clinical photographs were reviewed, and graft take rate was calculated by analyzing the photographs of the recipient sites using the photo editing software. In 9 patients (all patients in the trauma group and patients 3 and 8 in the sarcoma group), nearly all SGs remained attached after NPWT removal, and graft take was almost 100%. In these patients, the graft fixation rate approximately equaled the graft take rate. In the sarcoma group, 6 grafts initially were fixed, but a considerable part of the graft peeled off or gradually became necrotic during the clinical course (patients 1, 2, 4, 5, 6, and 7). In these patients, the authors reviewed clinical photographs retrospectively, and graft take rate was determined when maximum graft loss was observed (graft take was smallest). Therefore, the time interval from operation to evaluation of graft take rate differs. Graft take rate was calculated using clinical photography and analysis with the photo editing software. The number of pixels within the total SG area was determined. Then, the number of pixels within the graft take areas was determined. The percentage of graft take was calculated by dividing the number of pixels in the graft take by the total SG site pixels and multiplying the quotient by 100.⁹

Additional procedures
Information on additional procedures, such as the application of a NPWT dressing to promote granulation, debridement of necrotic tissue performed in the operating room, regrafting, and flap coverage performed after skin grafting, were retrieved from the medical records.
 

Time interval to complete epithelialization
Epithelialization was considered complete when the physician concluded there was no exudate present. The day of complete epithelialization was retrieved from the clinical chart. Complete epithelialization was not achieved with conservative treatment in patients 6 and 7, and they were excluded from this analysis.

Two intervals were calculated: time to complete epithelialization from wound creation (day of tumor resection in the sarcoma group and day of injury in the trauma group), and time to complete epithelialization from SG placement. The day of skin grafting was considered the same day as wound creation in the sarcoma group; however, in the trauma group, the day of skin grafting is different from wound creation because skin grafting was performed after several rounds of wound bed preparation.
 

Follow-up
After hospital discharge, patients were examined once or twice monthly until epithelialization was complete. After that time, patients were seen once or twice annually.

Statistical analysis
Differences in clinical characteristics between the 2 groups were compared using the t test, unpaired t test, or chi-squared test where appropriate. The significance level was set at P < .05.

Patient characteristics
There were 8 patients who underwent skin grafting after sarcoma resection (sarcoma group) and 7 for traumatic skin defects (trauma group). Seven STSGs and 1 FTSG (patient 6) were performed in the sarcoma group; 6 STSGs and 1 FTSG (patient 9) were performed in the trauma group. The age, gender, location, wound bed, comorbidities, size of defects, histology (sarcoma group), and type of injury (trauma group) are shown in Table 1 and Table 2. There were no patients with peripheral vascular disease. One patient had uncomplicated, well-controlled diabetes mellitus in the sarcoma group. Two patients in the sarcoma group and 2 in the trauma group had hypertension, and 2 patients in the sarcoma group had hyperlipidemia. There were no smokers in the sarcoma group but 3 in the trauma group. No patients were pre- or post-radiation therapy in the sarcoma group.

Average postoperative follow-up period in the sarcoma group ranged from 12 to 38 months (average, 21 months). All patients were alive without tumor recurrence during follow-up. One patient had lymph node metastasis and underwent excision on POD 150 (patient 3). In the trauma group, the postoperative follow-up period ranged from 10 to 38 months (average, 23 months). There was no statistically significant difference between the 2 groups in terms of age, sex, type of SG, or defect size (Table 3).
 

Wound bed preparation in the trauma group
The number of debridements in the trauma group varied from 1 to 4, and the interval from day of injury to skin grafting was 13 to 34 days. In patient 10, the exposed knee joint was covered with a medial gastrocnemius flap.
 

Characteristics of the wound bed
In the sarcoma group, 7 SGs were applied on the muscle stump exposed after sarcoma resection, including 1 with exposed tendon. Skin was grafted onto deep fascia in patient 2 (Table 1). There was no major nerve and vascular exposure in the operative field after sarcoma wide resection.

In the trauma group, all 7 SGs were placed on well-formed granulation tissue over subcutaneous tissue (4) or muscle (3), including 2 with exposed tendon (patients 10 and 13) (Table 2). There was 1 superficial infection (patient 11). No major nerve and vascular injury except for a sensory cutaneous nerve (patient 10) was noted.

Graft fixation
Graft fixation rate is shown in Table 4. All but 1 graft (patient 6) was fixed promptly, and the fixation rate was 100%. In patient 6, sutures fixing the SG were lost partly because of extensive motion of the scapula and surrounding muscle; the graft fixation rate was 86.2%.

Postoperative immobilization
The method and duration of postoperative immobilization is shown in Table 4. The joint was immobilized with a brace, sling, or external fixator for longer than 2 weeks in all but 2 patients whose wound was in the proximal thigh around the hip joint (patients 12 and 15). In patients 6 and 7, the shoulder was immobilized with a sling for 3 weeks postoperatively. 

Skin graft take and loss 
Graft take rate of each patient is shown in Table 4. In the trauma group, almost perfect graft take was achieved. In patient 11, wound culture was performed because superficial infection was suspected at the time of dressing change 5 days before the SG procedure. The culture was positive for Pseudomonas aeruginosa and methicillin-susceptible Staphylococcus aureus. Ceftazidime was administered on the day of surgery and continued for 7 days; the graft take rate was 100%.

Graft take rate in the sarcoma group was significantly lower than in the trauma group: 67.5% ± 30.0% and 99.6% ± 1.1%, respectively (P = .019) (Table 3). In the sarcoma group, 5 grafts were partly lost (patients 1, 2, 4, 6, and 7; Table 5). Two SGs (patients 2 and 4) were lost solely because of wound bed ischemia. In patient 1, besides wound bed ischemia, the distal part of the wound was depressed and the overlying SG peeled off and was lost. In patient 6, the wound bed appeared to be healthy muscle (Figure 1A); however, the SG was lost over time. Graft loss area was the largest on POD 46, and the graft take rate was 19.7% (Figure 1B).

In patient 7, the biceps, deltoid, and triceps muscles were exposed after wide resection of the tumor (Figure 2A), and skin was grafted onto these muscles. Deltoid muscle, which was located at the center of the wound, became thin and ischemic during the tumor resection and became necrotic on POD 16 (Figure 2B); skin necrosis followed. Despite the absence of wound ischemia or infection, skin grafted onto healthy muscle was lost over time (Figure 2C). Graft loss area was the largest on POD 41 (Figure 2D), and the graft take rate was 39.2%. 

Additional procedures
In the sarcoma group, additional procedures were performed in 3 cases (Table 5). 

In patient 6, epithelialization of the medial surface of the scapula was not achieved because this area slid under the skin and abraded the surface with shoulder motion (Figure 1C, 1D). Flap coverage was performed on POD 301, and the wound healed. In patient 7, epithelialization occurred from the periphery but did not extend beyond the border between the deltoid, biceps, and triceps muscles (Figure 2D). The wound muscle moved and changed shape depending on shoulder position (Figure 2D, 2E); epithelialization was not complete by POD 400. In the trauma group, additional procedures were performed in 1 case (Table 5). As mentioned, 2 patients in the sarcoma group did not achieve epithelialization with conservative treatment. 

Time to complete epithelialization 
Two patients in the sarcoma group whose wounds did not achieve complete epithelialization were excluded from this analysis. The mean time interval to complete epithelialization from wound creation of the remaining 5 patients in the sarcoma group was 113.3 ± 66.0 days and was 66.4 ± 44.4 days in the trauma group (Table 3). There was no statistically significant difference between the 2 groups (P = .155). The mean time interval to complete epithelialization from SG placement of the 5 patients in the sarcoma group was 113.3 ± 66.0 days and 40.3 ± 38.0 days in the trauma group. The time interval to complete epithelialization from SG placement was significantly shorter in the trauma group (P = .016).

Initially, the authors expected equal or superior clinical results for SGs in resurfacing large defects after sarcoma resection when compared with traumatic defects, because those wounds are sterile with edges made by deliberate, sharp dissection. However, this study shows there is a greater chance of SG failure after sarcoma resection. Considering the initial SG fixation was appropriate in both groups, and almost perfect graft take was achieved in the trauma group, the inferior results are considered to come from wound bed properties after sarcoma resection. Improvement of treatment outcome is warranted.

Disadvantages for skin grafting after sarcoma resection with respect to wound characteristics are: (1) wound depth in wide resection, reaching muscle body or deep fascia, and the exposed muscle stump moves significantly (patients 6 and 7); (2) the possibility that remaining tissues might be ischemic and become necrotic, or the overlying SG fails (patients 1, 2, and 4); and (3) the wound is depressed during tumor resection, and overlying skin is peeled off after removal of the bolster dressing (patient 1).

In the present study, bacterial contamination in the trauma group did not cause a significant problem. Traumatic wounds are contaminated and lacerated at the time of injury, and bacterial colonization may increase during NPWT application.¹⁰ However, graft take was almost perfect and mean graft take rate was significantly higher in the trauma group than in the sarcoma group. These results indicate that traumatic wounds, after serial debridement and wound bed preparation using NPWT, are more suitable for skin grafting compared with wounds immediately after sarcoma resection. Despite the sterility of sarcoma resection, the wound might contain depressed areas or ischemic muscle. From another perspective, the assumed advantage of sterility in the sarcoma group was not a significant factor in SG success.

If delayed reconstruction is performed for large-defect resurfacing after sarcoma resection, the wound bed bacterial load will increase compared with immediate reconstruction after resection.¹⁰ Considering the good clinical results in the trauma group, delayed reconstruction does not negatively impact skin grafting. Necrotic muscle will become apparent and can be removed during the waiting period, and graft loss, because of muscle necrosis, will be avoided. In this study, the time interval to complete epithelialization from SG placement was significantly longer in the sarcoma group, even after excluding 2 patients whose complete epithelialization was not achieved with conservative treatment. Delayed reconstructive resurfacing of large-defect areas after sarcoma resection is expected to shorten the time interval to complete SG epithelialization. In the field of skin carcinoma surgery, Thibault and Bennett¹¹ reported a delay of 2 to 8 days between skin carcinoma removal and FTSG placement resulted in a lower incidence of SG necrosis than no delay or a delay of 1 day before grafting. It also is reported that staged sarcoma excision and reconstruction is acceptable for managing major soft tissue defects.¹²

The period required for wound healing (time from day of tumor resection to complete epithelialization in the sarcoma group, and time from day of injury to complete epithelialization in the trauma group) was not significantly different between the 2 groups. It is unclear whether the total treatment period for epithelialization is shortened after performing second-look debridement and wound bed preparation. Furthermore, this method requires 2 or 3 operations. In general, reconstruction is performed as a 1-stage procedure at the time of sarcoma resection.² The present authors recommend considering delayed skin grafting in cases in which the wound is large, deep, and suspected to contain ischemic muscle.

Despite 3-week sling immobilization of the arm in adduction and internal rotation in patients 6 and 7, these grafts were lost without wound ischemia or infection. Furthermore, epithelialization was not achieved after long-term conservative treatment. As part of the wide resection for sarcoma, these wounds were deep and the muscle stump was exposed. Those wounds are far deeper than tissue defects after skin carcinoma removal, because sarcomas arise from subcutaneous or deeper tissues, and they are excised along with normal tissue surrounding the tumor. The wounds in these 2 patients were located around the shoulder joint, which has multidirectional mobility and the greatest range of motion of any joint in the body.¹³ In addition, because the shoulder girdle complex consists not only of the glenohumeral joint but also of the acromioclavicular, sternoclavicular, and scapulothoracic articulations, it is difficult to immobilize with a sling alone.¹³ 

In patient 6, the defect was over the scapula. Because the scapulothoracic articulation is a sliding junction between the deep aspect of the scapula and thoracic rib cage, the positional relationship between the thoracic rib cage and scapula changes because of thoracic spine motion.^14,15 The medial side of the patient’s scapula slid under the posterior aspect of the SG, resulting in abrasion of the SG with scapulothoracic movement. This motion also inhibited reepithelialization, leading to failure of conservative treatment. To prevent this effect, thoracic rib cage and thoracic spine motion also must be prevented by immobilization in a shoulder spica cast.¹⁶ 

In patient 7, the shoulder sling did not provide sufficient immobilization of the wound bed muscle for SG take. The deltoid muscle contracts as the prime mover during shoulder motion, and the long head of the biceps and lateral head of the triceps muscles move separately, depending on direction of shoulder motion.¹³ Shear movement occurred between the biceps, deltoid, and triceps muscles, and epithelialization did not extend beyond the borders of these muscles. 

Based on the aforementioned observations, SGs for resurfacing defects around the shoulder immediately after sarcoma resection with sling immobilization for 3 weeks and NPWT application as a bolster were unsuccessful, even if the wound bed were healthy muscle, which is generally thought to be a suitable wound bed for SGs. The authors considered the mechanical breakdown of the connections between the SG and wound bed occurred because of muscle movement before the formation of firm attachments and a new blood supply, resulting in gradual graft loss.¹⁷ More strict and prolonged immobilization using a shoulder spica cast might have improved graft take rate. However, this cast is essentially a body cast, very uncomfortable for patients, with the accompanying risk of pressure ulcers and respiratory problems.¹⁸ The authors do not use shoulder spica cast immobilization after skin grafting around the shoulder because of the concern about complications. 

Recently developed materials might improve graft take after sarcoma resection, including fibrin sealant for graft fixation.¹⁹ Fibrin glue provides an immediate fibrin network that stabilizes the graft and facilitates graft nutrition by serum imbibition with subsequent vascular ingrowth.²⁰ The present authors fixed grafts with skin staples or simple interrupted sutures using 4-0 nylon and placed multiple tacking sutures, but the result was not satisfactory. Fibrin glue has been shown improve percentage of SG take, especially in difficult grafting sites associated with movement.¹⁹ Fibrin glue may facilitate vascular ingrowth and prevent mechanical breakdown of the connections between the SG and wound bed that occur with severe muscle movement. Currently, there are no reports on the use of fibrin glue for SG fixation to the wound after sarcoma resection. Progress in such investigations are expected in the future.

Various bioengineered and synthetic products are candidates for better SG take.^21,22 Acellular dermal allografts from deepithelialized cadaveric skin seem to be the best option, but these are not approved in Japan. The present authors use acellular bilayer dermal substitutes (Integra Dermal Regeneration Template; Integra LifeSciences, Plainsboro, NJ) for wound bed preparation in relatively small superficial defects for softer, more pliable, and hypopigmented skin.²¹ By using these products during wound bed preparation, there is the possibility to improve graft take via granulation tissue formation over the depressed region or exposed small tendons. However, it is uncertain whether granulation tissue successfully developed over wounds abraded with scapular movement as seen in patient 6 or with extensive muscle shear movement as in patient 7. Further study is needed to examine whether large defects after sarcoma resection are successfully treated with skin grafting using dermal matrices and bioengineered skin substitutes for wound bed preparation. 

Successful use of NPWT for securing SGs, especially in wounds with exudative, irregular, or mobile recipient beds, and in difficult anatomic locations, has been reported.²³ Studies have shown that 3- to 5-days’ application of NPWT is sufficient for good graft take^9,23,24; the present authors used NPWT for 3 to 8 days for skin bolstering in this series. However, graft take rate in the sarcoma group was disappointing. In this study, the SG was attached to the wound bed at the time of NPWT removal in both groups. Negative pressure was successfully used as a SG bolster even in mobile wounds around the shoulder or depressed regions during its application. Dong et al²⁵ reported that, in a murine model, the mechanical strength of the SG at POD 9 is more than 3-times stronger than that at POD 3. It is possible that prolonged use of NPWT keeps SGs attached to the wound beds until strength increases enough to prevent mechanical breakdown of the connections between the SG and the wound bed. It is worthwhile to examine the effect of prolonged use of NPWT on graft take after sarcoma resection.

The authors feel more attention should be paid to wound bed condition to improve graft take after sarcoma resection. Graft take rate may be improved if the wound could be flattened by excising additional muscle around the deepest part. The authors believe muscle with a dark red color, showing no contractions with electrocautery stimulation, or broken into several discrete bundles and having a mop-like appearance should be debrided. In addition, a large amount of muscle necrosis occurred in the gastrocnemius muscle of patient 4, requiring debridement of necrotic muscle and a second SG. The blood supply of the gastrocnemius muscle is characterized by a single vascular pedicle that penetrates the muscle belly at its proximal pole; other muscles with this vascular pattern are the rectus femoris and tensor fasciae latae.²⁶ These muscles might be susceptible to pedicle injury during tumor wide resection and would likely become ischemic. 

There are several risk factors for SG failure, such as radiation therapy,²⁷ older age,²⁸ smoking,²⁹ diabetes mellitus,³⁰ and vascular disease.³¹ Although there was no statistically significant difference between the ages of the 2 groups in the current study, the number of patients aged 40 years of age or younger was larger in the trauma group. Regarding older age, Thourani et al²⁸ reported that age > 55 years has a negative impact on SG take in burn patients; however, age was not a risk factor for lower limb SG failure.³⁰ There is the possibility that the younger age of the trauma group positively influenced the SG take and time for wound healing.

In general, smoking negatively affects wound healing.²⁹ There were 3 smokers in the trauma group but none in the sarcoma group, so the negative effect of smoking in the sarcoma group could not be verified.

Ramanujam et al³⁰ reported that diabetes mellitus alone was not significantly associated with prolonged healing, but the presence of comorbidity significantly increased the risk of delayed healing. In the current study, 3 patients whose SG take rate was < 50% were a 50-year-old man without comorbidity (patient 1), a 62-year-old man without comorbidity (patient 6), and a 72-year-old man with uncomplicated, well-controlled diabetes mellitus (patient 7). The authors believe failure of a SG and epithelialization over muscles cannot be explained by the comorbidity. Further studies are needed to confirm this finding.

At this point, the most reliable way to cover a large skin defect after sarcoma resection around the shoulder joint is a skin flap.³² Early surgical intervention when SG failure and delayed reepithelialization occur after sarcoma resection also should be considered. An adjuvant method to improve graft take in resurfacing defects after sarcoma resection should be established. Hyperbaric oxygen therapy might be beneficial as an adjunctive therapy for compromised SGs in these situations.³³ Unfortunately, this therapy is not available at the authors’ institute. 

The current study has a number of limitations, including its retrospective nature, small sample size, and it was conducted at a single institution. However, there is value in investigating the characteristics of skin grafting for resurfacing large defects after wide resection in patients with sarcoma, because these procedures rarely are performed except in specialized centers. Furthermore, experience with these reconstructions is limited. This paper aimed to inform the specificity and difficulty of skin grafting to deep wounds after sarcoma resection. 

Sarcoma resection wounds have special characteristics. There might be depressed and ischemic areas because of tumor resection, and skin grafted in these areas will fail. Inevitably, deep tissues and muscle body are exposed, and wound bed movement might be significant. Skin grafts around the shoulder, especially, are at high risk for loss and for difficult healing with only conservative treatment. 

Authors: Masuo Hanada, MD, PhD^1,2; Hideki Kadota, MD, PhD¹; Sei Yoshida, MD¹; Naohide Takeuchi, MD, PhD²; Takamitsu Okada, MD, PhD²; Yoshihiro Matsumoto, MD, PhD²; and Yasuharu Nakashima, MD, PhD²

Affiliations: ¹Department of Plastic Surgery, Kyushu University Hospital, Fukuoka, Japan; and ²Department of Orthopaedic Surgery, Kyushu University Hospital

Correspondence: Masuo Hanada, MD, PhD, Department of Plastic Surgery, Kyushu University Hospital, Department of Orthopaedic Surgery, Kyushu University Hospital, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812-8582 Japan; mhanada@ortho.med.kyushu-u.ac.jp

Disclosure: The authors disclose no financial or other conflicts of interest.