Periosteal medial femoral condyle free flap for metacarpal nonunion

Abstract Background Metacarpal nonunion is a rare condition. The osteogenic capacity of periosteal free flap was investigated in five patients with metacarpal nonunion and impaired bone vascularization. Patients and methods Surgery was performed between 64 and 499 days after the initial bone osteosynthesis. The average age was 27.6 (range 16–32) years. Nonunion was caused by excessive periosteal removal in two patients, extensive open trauma in three. Four nonunions were diaphyseal, one metaphyseal. A periosteal medial femoral condyle free flap was raised on the descending genicular artery for four patients, the superomedial genicular artery for one. After osteosynthesis with a plate, the flap was wrapped around the metacarpal, overlapping the bone proximally and distally. The recipient vessel were the dorsal branch of the radial artery and a vena comitans in the anatomical snuffbox in four patients, at the base of the first webspace in one. Results The flap size ranged from 5 × 3.5 cm to 8 × 4 cm. No postoperative complication occurred. Radiological bone union was achieved 3 to 8 months after surgery. One patient had a full range of motion, one a slight extension lag of the proximal interphalangeal joint, two moderate joint stiffness of the proximal interphalangeal or metacarpophalangeal joint (one requiring plate removal and extensor tenolysis), one severe stiffness that allowed using a hook grip which was the aim of the surgery. Conclusion In case of metacarpal nonunion with impaired bone vascularization, the periosteal medial femoral condyle free flap provides an effective and biomimetic approach to bone healing.


| INTRODUCTION
Metacarpal fractures account for~42% of all hand fractures (Feehan & Sheps, 2006) however, metacarpal nonunion is a rare condition (Ring, 2006). Through the impairment of bone vascularization, extensive open trauma, or excessive periosteum removal during surgery are known sources of nonunion (Choudry et al., 2008). The absence of periosteum decreases the ability of callus formation (Gajendran et al., 2015). The rate of successful nonunion management after traditional means of internal fixation and bone grafting ranges from 70% to 92% in the absence of major skeletal loss (Ring, 2006). In cases of poorly vascularized bone beds and open comminuted fractures, achieving bone union may be compromised and further surgery necessary in 30% (Choudry et al., 2008).
The medial femoral condyle flap was initially described as a pedicled periosteal or corticoperiosteal flap for skeletal reconstruction of the lower limb (Hertel & Masquelet, 1989). It was used as a free corticoperiosteal flap to treat recalcitrant upper limb nonunions, such as clavicle, humerus, and scaphoid (Fuchs et al., 2005;Jones Jr. et al., 2008;Muramatsu et al., 2003). Effectiveness of the technique has been shown by numerous experimental and clinical studies over the past decades confirming the osteogenic properties of periosteum (Camilli & Penteado, 1994;Ortak et al., 2005). Previous reports appear to demonstrate a similar effect on ossification abilities between corticoperiosteal and periosteal-only flaps (Vegas et al., 2012). The aim of this study was to investigate the osteogenic capacity, the advantages, and the outcome of free periosteal-only medial femoral condyle (PMFC) flaps in patients with metacarpal nonunion with impaired bone vascularization.

| Surgical technique
The PMFC flap was harvested through an incision on the distal third of the medial thigh. The vastus medialis fascia was incised posteriorly, allowing exposition of the medial femoral condyle and its periosteal blood supply (Kazmers et al., 2017). The dominant vessel, the descending genicular artery or the superomedial genicular artery, was identified and dissected up to its origin. The periosteal flap was harvested with its pedicle and transferred to the hand. The nonunion site was prepared after removal of hardware, only minimal bone resection was done. The recipient vessels were identified. Metacarpal osteosynthesis was performed with a plate. Cancellous bone graft, either autogenic or allogenic, was inserted between the bone stumps in the presence of bone loss. The periosteum was wrapped around the metacarpal shaft and the plate at the level of the defect. End-toside anastomosis was performed to the radial artery and end-to-end to a vena comitans. Skin was sutured after ensuring perfusion of the flap.
Patients were discharged the day after surgery. Aspirin (100 mg) was given once a day during 1 month. Color Doppler ultrasonography assessed flap perfusion before discharge (day 1) and at suture removal (day 10). Bone union was confirmed by x-rays and/or CT-scan.
T In all patients, the metacarpal was macroscopically devascularized, ivory white, with no periosteum remaining in vicinity of the nonunion.
No bone bleeding was apparent after debridement of the nonunion site. Patient 3 had bone loss to the ballistic nature of the trauma.
The flap was harvested on the descending genicular artery for four patients, on the superomedial genicular artery for one patient.
The size of the flap ranged from 5 Â 3.5 cm to 8 Â 4 cm. Postoperative course was uneventful, no complications occurred. In all cases, bone union was obtained with a mean time of 6 months (range 3-8

| Case report
A bone grafting might be justified (Doi & Sakai, 1994;Georgiadis et al., 1993;Gonzalez del Pino et al., 2004). . We chose to use the PMFC for various reasons: its size is large enough to easily wrap around a metacarpal and bridge a potential bone gap; the donor site morbidity is low (Mehio et al., 2018); harvesting the flap poses no technical difficulty and the vessel size is rather large; no major artery needs to be sacrificed.
Our results demonstrate that adult periosteal flaps are able to promote bone growth on devascularized adult metacarpals. Ongoing smoking does not seem to prevent achieving bone union, nor does the absence of cancellous bone graft. The color Doppler imaging confirmed that the flaps were perfused at 10 days. We cannot exclude that the sturdy plating provided enough stability to allow by itself bone union however, this seems unlikely since bone devascularization was evident during surgery.
The periosteum is involved in all phases of the fracture repair process, providing cells for both osteogenesis and chondrogenesis, participating in intramembranous and endochondral ossification (Thompson et al., 2015). Angiogenesis is critical to the direct intramembranous healing.
The periosteum brings circulating growth factors, migrating bone cell precursors, and vascular endothelial cells in proximity to the nonunion. In this way, it provides the angiogenetic niche necessary for osteogenesis and bone consolidation (Gerstenfeld et al., 2003;Percival & Richtsmeier, 2013). This pathway allows for immediate regeneration of anatomical lamellar bone without remodeling steps as can be raised without vascular compromise is 10 by 4 cm (Hertel & Masquelet, 1989;Sakai et al., 1991), large enough to manage any metacarpal nonunion.
Conventional orthopedic algorithm for the treatment of nonunion is composed of resection of nonunion, osteosynthesis, and bone grafting. This approach provides the shape and the structure but the physiological environment is missing. Therefore, it might seem insufficient particularly when vascularization of the bone is compromised. Enhancing local vascularization can be obtained with the induced membrane technique (Masquelet et al., 2000;Moris et al., 2016). However, this procedure requires two surgeries and does not provide a tissue specifically intended to allow bone formation. We propose a more biomimicry-oriented approach, which achieved union in all five patients despite the lack of initial bone vascularization. In an era where microsurgery is widely available, PMFC flap could be a firstline alternative in cases of poorly vascularized metacarpal nonunion.

| CONCLUSION
The free PMFC flap appears to be a reliable technique to achieve bone union in cases of metacarpal nonunion with poorly vascularized bone.

ACKNOWLEDGMENT
Open access funding provided by Universite de Lausanne.

DATA AVAILABILITY STATEMENT
Data are available on request from the authors.