Comparison of Musculocutaneous and Fasciocutaneous Free Flaps for the Reconstruction of the Extensive Composite Scalp and Cranium Defects

Abstract Composite scalp and cranium defects, which require microsurgical reconstruction, result from tumor resection, radiation, trauma, severe burn injuries, and rarely vasculitic disorders. In the current study, the authors aim to compare the outcomes of the fasciocutaneous flaps and musculocutaneous free flaps used for the reconstruction of extensive composite scalp and cranium defects. From 2010 to 2017, 21 patients who underwent composite scalp and cranium defect reconstruction with a free flap were retrospectively identified. Eighteen patients had squamous cell carcinoma, 2 patients had meningioma, and 1 patient had Ewing sarcoma. Thirteen musculocutaneous free flaps including latissimus dorsi and vertical rectus abdominis flaps and 9 free fasciocutaneous flaps including radial forearm and anterolateral thigh flaps were used. Only 1 flap loss was encountered. No neurologic impairment in postoperative period was reported. The mean length of stay in the hospital, the duration of surgery, and total volume of blood transfusion for the fasciocutaneous flap group were significantly shorter than those for musculocutaneous flap group. No flap atrophy was reported in fasciocutaneous flap group. Reconstruction of the composite scalp and cranium defects with fasciocutaneous free flaps allows shorter hospitalization, less blood transfusion and less flap atrophy than those of musculocutaneous flaps. To this respect, their usage should be prioritized in such challenging patients.

Abstract: Composite scalp and cranium defects, which require microsurgical reconstruction, result from tumor resection, radiation, trauma, severe burn injuries, and rarely vasculitic disorders. In the current study, the authors aim to compare the outcomes of the fasciocutaneous flaps and musculocutaneous free flaps used for the reconstruction of extensive composite scalp and cranium defects. From 2010 to 2017, 21 patients who underwent composite scalp and cranium defect reconstruction with a free flap were retrospectively identified. Eighteen patients had squamous cell carcinoma, 2 patients had meningioma, and 1 patient had Ewing sarcoma. Thirteen musculocutaneous free flaps including latissimus dorsi and vertical rectus abdominis flaps and 9 free fasciocutaneous flaps including radial forearm and anterolateral thigh flaps were used. Only 1 flap loss was encountered. No neurologic impairment in postoperative period was reported. The mean length of stay in the hospital, the duration of surgery, and total volume of blood transfusion for the fasciocutaneous flap group were significantly shorter than those for musculocutaneous flap group. No flap atrophy was reported in fasciocutaneous flap group. Reconstruction of the composite scalp and cranium defects with fasciocutaneous free flaps allows shorter hospitalization, less blood transfusion and less flap atrophy than those of musculocutaneous flaps. To this respect, their usage should be prioritized in such challenging patients.
Key Words: Craniectomy, free flap, scalp reconstruction (J Craniofac Surg 2018;29: 1947-1951 C omposite scalp and cranium defects, which require microsurgical reconstruction, result from tumor resection, radiation, trauma, severe burn injuries, and rarely vasculitic disorders. [1][2][3] These soft tissue and bone defects are distinct with regard to anatomy, complication rates, and reconstructive options. Scalp and cranium protect the underlying dura and brain. Once this protection is lost, the patients face the risks of meningitis, brain abscess, cerebrospinal fluid (CSF) leak, and seizures. Prior or planned radiotherapy makes the scenario even worse. Local or regional flaps generally fall short given the fact that the defects created are generally beyond these options. Staged approach using tissue expansion is not feasible due to malignant nature of the process. Free flaps strengthen the plastic surgeons' hand and bring diversity to the reconstruction alternatives. Many techniques have been defined in the literature so far and most of them comprise the same surgical principals. [4][5][6][7][8] The advantages of free flap reconstruction depend on various factors including but not limited to etiology and size of the defect, donor site morbidity, and involvement of surrounding structure. 9 Controversies exist to decide a cranioplasty should be added.
In the current study, the authors aim to share their experiences and to compare the outcomes of the fasciocutaneous flaps and musculocutaneous flaps used for the reconstruction of extensive composite scalp and cranium defects.

MATERIALS AND METHODS
The current study was conducted according to the Declaration of Helsinki. Informed consent was obtained from each patient. From 2010 to 2017, 21 patients who underwent composite scalp and cranium defect reconstruction with a free flap were retrospectively identified. The medical records of patients were reviewed for demographics, tumor pathology, defect location, flap choice, length of stay in hospital, and complications. Patients with a minimum postoperative follow-up of 12 months were included for the study. Patients were excluded from the study if they had a reconstruction other than a free flap, if they had only scalp defect reconstructed with a free flap, if they had only split thickness cranium defect or if they had combined defects of the periorbita and/or midface regions. The patients meeting these criteria were grouped into 2 categories according to the composition of the flap: fasciocutaneous or musculocutaneous.

Surgical Procedure
Initially, the resection of tumor or the debridement of nonviable tissue was undertaken. The cutaneous malignancy was resected with 1 to 2 cm surgical margins. The degree and extent of cranium involvement was determined by both clinically and radiologically. Trying to have 1 cm oncologically safe bone margin, a circular craniectomy was performed. Dura was removed by the neurosurgery team. Duraplasty with fascia lata graft was added if brain parenchyma was exposed. Brain parenchyma resection was also decided by neurosurgery team. The neck dissection, when necessary, was performed. The harvest of the flap was initiated once we were sure of that the margins were tumor-free or all the necrotic scalp and bone were removed in patient with osteoradionecrosis. Recipient vessels for anastomosis were then prepared. The main vessels of choice were superficial temporal vessels. In patient with venous insufficiency, we explored the neck for any suitable vein.

Statistical Analysis
Statistical analysis of the data was performed by using SPSS 17 (SPSS Inc, Chicago, IL). The length of stay in the hospital, duration of the surgery, amount of blood transfusion, flap atrophy leading to wound dehiscence, and complications were compared between fasciocutaneous and musculocutaneous reconstruction groups by using a t test. P 0.05 was considered significant.

Demographic Findings
Twelve of the patients were males and 9 were females. Average patient age was 54.9 years (range from 15 to 82 years). No significant difference existed between 2 groups in terms of age ( Table 1). The mean follow-up time was 37 months (13-64 months). Three of the patients had previous unsuccessful closure attempts to scalp.

Tumor Characteristics
Eighteen patients had squamous cell carcinoma (SCC), 2 patients had meningioma, and 1 patient had Ewing sarcoma. One of the patients with meningioma was operated because of exposure of alloplastic implant, which was implanted 2 years ago during tumor removal ( Fig. 1). One of the patients with SCC was operated because of osteoradionecrosis of the cranium, which was caused by previous high-dose radiotherapy (Fig. 2). Remaining 19 patients had tumor invasion of both skin and cranium. Twelve patients with SCC had parenchymal involvement. Seven patients had prior radiotherapy and 8 patients took postoperative radiotherapy ( Table 2).

Flap and Reconstruction Characteristics
Twenty-two free flaps were performed for scalp and cranium reconstruction (Tables 1-2 Thirteen musculocutaneous flaps and 9 fasciocutaneous flaps were used. In 4 patients with LD flap reconstruction, skin island was designed in such a way that it covered thoracodorsal vessels (Figs. 2 and 5). Transverse cervical vessels were chosen for the patient with osteoradionecrosis of the scalp because of prior radiotherapy and sacrifice of both superficial temporal arteries (Fig. 2). Only 1 flap loss was encountered. All of the rough surfaces of the musculocutaneous flaps were skin grafted 3 weeks later. No cranioplasty was intended except for 1 patient with recurrent meningioma within the posterior cranial fossa (Fig. 6).

Hospitalization
The mean length of stay in the hospital following surgery for the fasciocutaneous flap group was significantly shorter than that of the    Table 2). Cumulative hospitalization involving the primary reconstructive procedure and secondary procedures like skin grafting and flap reshaping were also significantly higher in musculocutaneous flap group (14.9 AE 2.7 days vs 21.3 AE 3.6 days, respectively; P < 0.05).

Complications
Neither dural fistula nor CSF leakage was seen. We did not encounter any neurologic impairment in postoperative period. Two patients had venous outflow problem in the immediate postoperative period and were taken back to the operating theater for venous reanastomosis with a vein graft to external jugular vein. We observed 2 skin graft lyses and 2 seroma formations at the donor site in both groups. The amount of blood transfusion in the form of erythrocyte suspension was significantly higher in musculocutaneous flap group (P < 0.05) ( Table 2). During 7-year follow-up, 3 patients died because of distant metastasis and 1 patient died because of a myocardial infarction (Table 3).
According to the experiences from this series, we devised a simple algorithm prioritizing free fasciocutaneous flaps for the reconstruction of extensive scalp and cranium defects (Fig. 7).

DISCUSSION
In the current study, patients with fasciocutaneous flaps had significantly shorter duration of hospitalization than the patients with musculocutaneous flaps. When compared to each other, RFF and ALT flap showed no significant difference in terms of duration of hospitalization. In their large series, Sweeny et al 9 showed that radial forearm free flap was associated with few complications and shorter duration of hospitalization when compared to LD and VRAM flaps. They used radial forearm free flap more often in the frontal region whereas we used it in the parietal region predominantly. In contrary to our results, they concluded that choice of microsurgical free flap did not affect the length of hospitalization. Regarding the flap choice, a clear advantage of fasciocutaneous flaps was the shorter duration of surgery. The major reasons to use a fasciocutaneous flap in our series were relative ease of flap harvest, allowing 2-surgical team approach and no need for repositioning the patient unlike an LD flap would need. In a study comparing the outcomes of younger and older patients requiring scalp reconstruction with free LD and ALT flaps, Simunovic et al 10 found no difference between groups. They implied that duration of surgery was reduced when ALT flap was used. Although duration of surgery does not seem to be related with perioperative complications in our series, it clearly decreases the frustration of the microsurgeon.
Another advantage of fasciocutaneous flaps was that they are prone to less atrophy. Flap atrophy generally results from denervation of muscle. In addition, postoperative radiotherapy itself may cause flap atrophy or even may worsen it. In our series, we encountered flap atrophy only in the musculocutaneous flap group although it did not cause any late complications. Sosin et al 1 reported 3 patients with free flap atrophy following scalp and calvarium reconstruction. Two of them were fasciocutaneous ALT and ulnar forearm flaps and the remaining one was LD flap. However, they mainly observed flap atrophy in patients with previous unsuccessful reconstruction attempts with LD flap. Others reported that ALT flap underwent less atrophy than the LD flap and for that reason they shifted their main flap choice for scalp reconstruction over the years. 11 Radial forearm flap was the most rewarding flap in terms of blood loss, hospitalization, and duration of the surgery. Its main constraint comes from the enormousness of the defect size. In our series, if the dimension of the scalp defect was more than 112 cm 2 , another free flap option was taken into consideration. In their wide series of RFF experience, Sweeny et al 9 did not use RFF for the scalp defects, which were more than 101 cm 2 . Such a limitation can be overcome by using another fasciocutaneous flap, ALT flap. Scalp reconstruction even with 420 cm 2 ALT flap size has been recently reported. 12 Its length of vascular pedicle can be safely   extended with a meticulous dissection toward the origin of lateral circumflex femoral artery. As previously described, 1,13 LD flap was the most commonly used free flap option in our microsurgical scalp reconstruction series. The main difficulty that we experienced during the inset of the flap was the large muscle bulk and greater surface area of LD flap. We could perform skin-grafting of the rough surface during the primary surgery but we wanted to be sure of free flap survival. After the perioperative risk for microvascular complications was decreased, we performed the skin-grafting as a secondary procedure. We could also not incorporate a skin island to LD flap thereby harvesting a muscle-only flap. However, the skin island behaves like a monitor for flap survival and decreases the need for skin graft which may lead to postradiotherapy wound healing problems in this subset of high grade-malignancy patients. In addition, extended skin island design offers protection over the vascular pedicle itself. This design allows native skin island of flap to cover the microanastomoses and decreases the tension over the vascular pedicle. Otherwise it necessitates skin grafting, which leads to unsightly appearance and scar contracture.
Preoperative and postoperative radiation therapies were not found to be significantly associated with the development of complications although Chao et al 14 found that a combination of the 2 was associated with increased late recipient site complications. Moreover, intense radiotherapy for the treatment of invasive skin cancer may cause fatal osteoradionecrosis of the skull, which is very rarely reported especially in the parietooccipital region. Once the diagnosis of osteoradionecrosis is established, aggressive radical surgery involving 3 important stages is mandatory to prevent further complications; eradication of infected and necrotic tissues, restoration of functioning viable tissues, and achieving an acceptable cosmetic result. 15 Free flaps may not be readily used because the recipient vessels in the head and neck may be affected by the extensive radiation and previous surgeries as in the patient in Figure 2. Therefore, we had to explore the posterior neck to find suitable vessels. Transverse cervical vessels emerge as an excellent source of recipient in vessel-depleted head and neck. In addition, this maneuver removed us from traumatic zone of intense radiation.
Cranial bone reconstruction is important for the protection of the intracranial contents and for contouring the cranial vault. Cranial bone can survive even when it is not vascularized. Therefore, vascularized soft tissue is the key. Knowing the fact that reconstruction of the bony defect is not feasible without stable soft tissue coverage, 16 our reconstruction algorithm avoids cranioplasty together with scalp and cranium reconstruction. Simultaneous cranioplasty during the immediate reconstruction has some controversies in the literature. Although the patients with cranial bone defect may suffer from headache, dizziness, loss of concentration, depression, anxiety, intolerance to noise, and vibration and neuromotor weakness, there are concerns for high rates of tumor recurrence, low survival rates, the need for postoperative radiotherapy, and the risk for surgical complications. 17 Size and location of the bone defect, radiotherapy, and expected intracranial pressure are important determinants for considering bone reconstruction. Afifi et al 18 performed a retrospective review of patients with simultaneous alloplastic cranioplasty and free flap reconstruction and reported a major complication rate of 38%. Chao et al 14 demonstrated that there was no significant difference between simultaneous or scalp-only reconstructions in terms of per operative complications although they observed a trend toward greater complications when cranioplasty was performed at the time of soft tissue reconstruction.
In conclusion, reconstruction of the extensive, composite scalp, and cranium defects with fasciocutaneous free flaps allows shorter hospitalization, less blood transfusion, and less flap atrophy than those of musculocutaneous flaps. That's why we recommend to use RFF in defects <112 cm 2 and to use ALT flap in defects >112 cm 2 . Considering the extensive and destructive nature of the primary lesion, the necessity of multidisciplinary approach, which is an example of neuroplastic surgery, 19 should be emphasized. The importance of transverse cervical vessels should also be underlined as recipient vessels in vessel-depleted head and neck.