Mandibular reconstruction using FFF has been widely utilized since its first report in 1989 . FFF is useful as it allows close to 25 cm of long bone harvest and multiple osteotomies without compromising blood supply . Moreover, since the fibula is distant from the head and neck, the resection and reconstruction teams can work simultaneously. Although Hidalgo mentioned that the blood supply for the cutaneous flap is not good in his first report , the peroneal artery perforators for cutaneous flap perfusion are sufficiently well-sized to achieve anastomosis . However, when it comes to secondary mandibular reconstruction, applying the FFF is still considered to be challenging. Although the potential number of patients who need secondary mandibular reconstructions may be huge, such patients tend to have a history of unfavorable results in previous surgeries and are reluctant to undergo secondary reconstructions due to fear and anxiety . The four patients who underwent secondary mandibular reconstructions at our institution had an average of four surgeries before secondary reconstructions.
Several reports have described secondary mandibular reconstruction using FFF. In India, Kadam et al. performed secondary mandibular reconstructions using FFF in 21 patients and reported flap survival and an improvement in symptoms that had necessitated secondary reconstruction in all patients . In Taiwan, Lin et al. reported 20 secondary mandibular reconstructions using FFF in patients with head and neck cancer and demonstrated its advantages of lesser postoperative complications, such as recipient site infections and plate exposures, by comparing it with 41 secondary mandibular reconstructions using free soft tissue flaps combined with a bridging plate . In both reports, the drawback of secondary mandibular reconstruction was the occurrence of scarring in recipient vessels and surrounding skin-soft tissues. To avoid complications related to vascular and skin-soft tissue scarring, contralateral neck vessels were selected for anastomosis in five out of 21 patients in India. Among the 20 patients from Taiwan, contralateral neck vessels in 13 patients and vessels outside the neck in two patients were selected for anastomosis for the same reasons.
In most cases, patients who require secondary mandibular reconstructions tend to have undesirable histories such as neck lymph node dissections and radiation therapies. In addition, inflammation might have persisted due to poor results from previous surgery. Radiation therapy also leads to catastrophic tissue damage. Postoperative radiation produces irreversible damage to the tissues, causing tissue fibrosis that results in severe scarring [12,13]. Recently, Eriksson et al. reported that radiation therapy causes sustained upregulation of plasminogen activator inhibitor-1, which is the main cause of thrombus formation, resulting in chronic inflammation, mainly in the adventitia . Therefore, for all secondary mandibular reconstructions performed at our institution, the contralateral neck vessels were taken as the recipient vessels to reduce the risk of thrombus formation. A long vascular pedicle is essential to achieve adequate anastomosis when using the contralateral neck vessels. Additionally, secondary mandibular reconstructions usually require a long and rigid bi-cortical bone that enables implant installation after mandibular reconstruction. FFF is an ideal option that satisfies these conditions because it has a good-sized (2 to 3 mm) and lengthy (15 cm) vascular pedicle arising from the peroneal artery and its venae comitantes. FFF can be harvested with an expendable long and rigid bone . Byun et al. reported nine cases of successful secondary palatomaxillary reconstructions using FFF, which indicates that the length of the vascular pedicle in FFF is sufficient to reach the vessels on the contralateral side . Thus, none of the four cases of secondary mandibular reconstructions performed at our institution required vein grafting, and all of them resulted in flap survival. However, two patients developed partial osteonecrosis, and one developed partial skin flap necrosis, out of the eight primary mandibular reconstructions performed at our institution. In both reconstructions that resulted in partial osteonecrosis, the double-barrel technique  was selected, and one of them required vein grafting. Owing to the technique of cutting and making a 180-degree bend in the osteo flap that enables reconstruction of the mandibular ridge, the length of the vascular pedicle becomes too short for safe anastomosis. Although there could have been other technical problems, the double-barrel technique possibly led to a reduction in the length of the vascular pedicle. In addition, this could have been a subjective impression of the surgeon. Arteriosclerosis at the time of vascular anastomosis was conspicuous in patients over 70 years of age, making it difficult to perform both primary and secondary reconstructions.
Although our database allowed us to analyze the demographic differences between primary and secondary mandibular reconstructions using FFF, the present study had several limitations. First, it was retrospective in design with its inherent defects. Second, the sample size was small because it was a relatively infrequent surgery. Finally, it was a single-center study. To reduce bias observed in long-term studies, reconstructive surgery that was performed by the same surgeon in a short period of time was targeted. A multi-center prospective study would be required to overcome these limitations even though there could be potential inconsistencies in surgical techniques.