The Capanna technique originally described in 1993, consists of inserting a vascularized fibula within a bone allograft shell to reconstruct large bony defects. This was first utilized in the lower extremity following oncologic resection but can also apply to the upper extremity ( Fig. 4 A–C ). ^, The fibula allows for accelerated healing at the osteosynthesis sites and helps with revascularization of the allograft, whereas the allograft allows for stronger fixation and prevents catastrophic fibula fractures when reconstructing metaphyseal defects. ^, A variation with the fibula placed extramedullary can also be performed when there is adequate skin or soft tissue envelope for a bulky construct, with potential benefits being facilitation of intramedullary nail fixation, and placement of the fibula closer to recipient vessels. The Capanna technique has been well-described for use in humerus, femur and tibia reconstructions, with union success rates upwards of 90% and significantly higher success rates when compared with allograft alone, along with excellent long-term function based off the Musculoskeletal Tumor Society (MSTS) score when union is achieved. ^,,,,,,,,,

( A ) Radiograph and computed tomography (CT) depicting a chondrosarcoma of the proximal humerus in an 18 year-old. ( B ) Image of the Capanna technique with the free fibula placed inside an allograft for proximal humerus reconstruction. The peroneal vascular pedicle can be seen passing through a fibula trough longitudinal made in the allograft. ( C ) Radiograph depicting good healing of the allograft-intramedullary fibula construct.

From our 20-year experience, we have found the Capanna technique most efficacious in pediatric reconstructions of the humerus, femur, and tibia. The technique is best used for children above the age of 8 to minimize the risks of long-term limb length discrepancies. New options such as combining a Capanna technique with a proximal epiphyseal fibular transfer may reduce these problems in the future. Outcome studies by Misaghi and colleagues suggested that the Capanna technique for pediatric tibial reconstruction is superior to results obtained with allograft alone.

Reconstruction of the tibia for adult traumatic or oncologic defects may be achieved by transferring the fibula as a free or pedicled VBG from the ipsilateral extremity. This may be performed in conjunction with an allograft as described in the Capanna technique or in a double barrel configuration given the diameter mismatch as well as the need for structural support. ^,,, In the majority of cases of trauma and oncologic resection, we prefer to leave the native fibula if possible, as it helps with overall structural support of the lower leg, preserves length, and avoids loss of blood flow; however, if the patients contralateral fibula is not available or there is a high risk for microvascular failure, a pedicled fibula can be used for tibial reconstruction. For proximal and middle defects, the ipsilateral fibula pedicled using the proximal blood supply may be suitable, whereas for distal defects, the distal blood supply based on retrograde flow through the distal anastomosis between the posterior tibial and peroneal artery is used.

The Capanna technique can be used to reconstruct segmental femoral defects beyond 20 cm. Additonally the fibula can be used as a vascualrized onlay graft to heal cases of chronic femoral nonunion. ^,,,,,,,, In instances of shorter length defects, segmenting the fibula such as in a double barreled configuration to match the diameter of the femur and to provide additional strength has also been well-documented. In osteonecrosis of the femoral head the free vascularized fibula has been shown to be a reliable method to delay or even avoid THA. ^,,

The shape and size of the fibula in relation to the diaphysis of the radius and ulna make it the ideal option for segmental forearm bone loss. It can be divided into 2 segments to reconstruct both the radius and ulna simultaneously in cases of military or combat trauma. The fibula can also be configured with a prosthesis to reconstruct the proximal radius and the elbow joint, as seen in Fig. 5 A–D . Studies have shown excellent union rates from 80% to 100% More distal reconstruciotn are associated with poorer functional outcomes due to distal radial unlnar joint involvement. ^,,,,,

( A ) Radiographs depicting plasma cell tumor of radial head and neck. ( B ) Images of free fibula harvested with a skin paddle and configured with a prosthesis to reconstruct the proximal radius. ( C ) Radiographs at 3-year follow-up depicting well-healed proximal radius, prosthesis positioning. ( D ) Images at 3-year follow-up showing good range of motion.

The use of the fibula for reconstruction of the humerus has also been well-described, whether in its use for intercalary defects, as an onlay for pathologic fractures, or with an allograft such as for glenohumeral arthrodesis. Use of an allograft with the fibula in defects close to the anatomic neck of the humerus has been reported to obtain stable fixation and better tendinous reattachment of the rotator cuff and deltoid. ^,,, The humerus has a significantly greater diameter than that of the fibula, making bony fixation sometimes difficult, and a double-barreled approach may help in matching diameters. Studies have shown failure rate of up to 40% in regards to late fracture when the vascularized fibula is used for large intercalated diaphyseal humeral defects. ^,,,,,,, However, with use of rigid compression plating for fibula fixation rather than use of screws, high union rates of 95%, low fracture rate of 8%, and good functional MSTS scores are acheivable.

Given the fibula’s ability to be osteotomized into multiple segments connected by a common peroneal artery, multiple metacarpal or metatarsal defects can be reconstructed with this VBG. In the event of significant mismatch between the diameter of fibula and metacarpal or metatarsal bony diameters, splitting the fibula may provide a better match without compromising outcome. ^,,

Bony defects in children may involve the physis. In such cases, bone reconstruction should allow for axial growth to avoid significant limb length discrepancies. In these cases, transfer of the vascularized proximal fibular epiphysis enables bony reconstruction and the potential for longitudinal growth. The use of the proximal fibula with the epiphysis is most commonly used in oncologic defects of the distal radius, distal ulna, proximal humerus, and proximal femur. The blood supply for the physis is supplied by branches of the anterior tibial artery and lateral geniculate artery, with the epiphyseal pedicle arising from the main tibial artery just prior to entering the anterior compartment. The fibula epiphysis was first used to reconstruct the distal radius as described by Pho in 1981. Studies by Innocenti, with follow-up of over 20 years, showed expected fibula growth rates after transplant. ^, An international multi-institutional study has also demonstrated that the proximal humerus can also be reconstructed successfully with 88.9% of patients experiencing longitudinal growth albeit with a 40.9% fracture rate. Use of the fibula epiphysis to reconstruct the proximal tibia can be successful via both microvascular and pedicled transfer. Additionally, it has been reported to reconstruct both the distal or proximal femur. ^,, Enthusiasm for this technique has recently been tempered by a long-term outcome study suggesting that lower extremity femoral reconstruction often resulted in early failure with only 1 out of 6 patients maintaining the construct until adulthood.

The MFC VBG has proven to be a versatile tool in reconstructive surgery, with applications spanning various anatomic regions particularly the extremities. In the upper extremity, the graft’s thin, pliable nature allows for easy shaping and contouring, making it ideal for complex hand and wrist reconstructions. The MFC has been used successfully to treat recalcitrant scaphoid nonunion, with studies reporting success rates as high as 94%. Additional uses in the upper extremity include reconstruction of recalcitrant defects of the humerus, clavicle, metacarpals and forearm defects. A large systematic review of 248 cases found union rates have been reported to average 98.7%. Although it is harvested with cortical bone, it provides less structural stability than the fibula or scapula; thus stable fixation across the defect site is a necessity when using an MFC graft to ensure bony union.

Lower limb reconstruction has also benefited significantly from MFC grafts. Cavadas and colleagues reported a 93% success rate in achieving bone fusion in foot and ankle reconstructions using this technique. The graft’s ability to provide both structural support and vascular supply makes it particularly useful in cases of avascular necrosis or in patients with compromised local vascularity. Its application in treating complex hindfoot defects and recalcitrant bony nonunions of the lower extremities has been well-documented, offering a reliable solution for challenging cases. Additionally, a case series by Saad and colleagues further supported this claim, showing promising results in treating avascular necrosis of the talus using a MFC periosteal free flap. The study demonstrated significant improvement in patient outcomes and MRI evidence of resolution of marrow edema and soft tissue inflammation in cases of talus AVN.

A variation of the MFC vascularized bone flap, as mentioned, includes using the transverse branch of the DGA to create an osteochondral flap, otherwise known as the MFT flap. The flap derives cartilage from the proximal trochlea that articulates with the patella, making it particularly suitable for complex carpal reconstructions. While the MFC flap is traditionally used for nonunions with or without segmental bone loss, the MFT flap is specifically indicated for proximal pole scaphoid nonunions that cannot be reconstructed via conventional techniques and in advanced Kienbock’s disease in young patients ( Fig. 9 A–E ). ^, Articles report high success rates for MFT flaps in carpal applications, such as in a case series of 7 patients, albeit with 2 failures due to graft resorption, by Alolabi and colleagues demonstrating that union was achieved in 5 patients at a mean of 12 weeks, with increased grip strength and absence of pain at mean follow-up time of 32 months.

( A ) Radiographs of a 20 year-old mechanic with Kienbock’s disease with a fracture through the midportion of the lunate. ( B ) T1-weighted and T2-weighted MRI images revealing for collapsed and fragmented lunate with enhancing bone marrow edema images. ( C ) CT images revealing for the fragmented lunate compromising the lunocapitate and radiocarpal surfaces. ( D ) Images showing the MFT vascularized osteochondral graft. ( E ) Radiographs revealing for well-healed lunate.

The 1,2 ICSRA flap first described by Zaidemberg and colleagues for the treatment of scaphoid nonunion has vessels identified on the extensor retinaculum and followed distally to their anastomosis with the radial artery at the anatomic snuffbox. After opening the first and second extensor compartments away from the 1,2 ICSRA vessel to establish the radial and ulnar boundaries of the flap, the bone flap is then designed with its center 1.5 cm proximal to the radiocarpal joint ( Fig. 11 ).

Images showing the 1,2 ICSRA vascularized bone grafts in situ.

Zaidemberg and colleagues reported on a cohort of 11 patients with chronic scaphoid nonunion with a 100% rate of radiographic bony union with an average follow-up of 6.2 weeks. A larger cohort of 48 patients with scaphoid nonunions found radiographic union was achieved in only 71%. In this study, analysis of the scaphoid nonunions that did not heal with the 1,2 ICSRA found the failures were primarily seen in those patients with midcarpal instability, humpback deformity, and proximal pole avascular necrosis. The challenge of treating avascular necrosis with the 1,2 ICSRA was further illustrated in a study exploring VBGs in treating Preiser’s disease. Of the 8 pedicled VBGs to the avascular scaphoid, 6 consisted of the 1,2 ICSRA flap, and while symptomatic pain relief was achieved, the proximal scaphoid was still not revascularized. With the proper indications applied and patients with advanced carpal collapse are excluded, recent studies have demonstrate that use of the 1,2 ICSRA flap can reliably result in bony union rates of nearly 90%. ^,

The 2,3 ICSRA flap was first described by Sheetz and colleagues in an anatomic study where the supplying artery was found to originate from the posterior division of the AIA running directly over Lister’s tubercle between the second and third extensor compartments. Distally, it anastomoses with the radial artery and thus can be utilized as a retrograde pedicled flap for carpal nonunions. Through a longitudinal incision centered over Lister’s tubercle, the 2,3 ICSRA vessel is identified on the extensor retinaculum and followed distally to their anastomosis with the radial artery at the anatomic snuffbox or the dorsal intercarpal arch. After opening the second and third extensor compartments away from the 2,3 ICSRA vessel to establish the radial and ulnar boundaries of the flap, the bone flap is then designed with its center 2.1 cm proximal to the radiocarpal joint.

While both the 1,2 ICSRA and 2,3 ICSRA represent viable options for treating scaphoid nonunions, there has been a relative paucity of studies examining the outcomes of the 2,3 ICSRA. A constraint of the 2,3 ICSRA is that as the pedicle runs right over Lister’s tubercle, it can be challenging to assess the recipient scaphoid nonunion prior to elevating the 2,3 ICSRA. However, this can be solved by basing the 2,3 ICSRA along its transverse connection to the radial artery at the snuffbox instead of the dorsal intercarpal arch, taking advantage of the distal perfusion arch along the dorsal radius. One reported benefit of the 2,3 ICSRA is the greater rotation arc given its slightly longer pedicle, and one study of 72 scaphoid nonunion patients reported a 92% bony union rate when treated with the 2,3 ICSRA flap. The 2,3 ICSRA flap was preferred in 52 patients versus the 1,2 ICSRA flap in 20 patients reportedly due to the longer pedicle allowing inset with less tension on the vascular pedicle, and there was no difference in the rates of bony union. More recently, a comparison study of 32 total scaphoid non-union patients compared 12 1,2 ICSRA and 20 2,3 ICSRA patients and compared patient-reported outcomes using a patient questionnaire consisting of the Modified Mayo Wrist Score (MMWS) and Patient-Rated Wrist Evaluation (PRWE) sent 1 year after surgery. PRWE scores were similar between the 2 groups; however, MMWS was significantly between groups with the all 1,2 ICSRA patients reporting a good-excellent score while only 45% of 2,3 ICSRA achieving the same result, with the difference attributable to range of motion.

The fourth extensor compartmental artery (4ECA) and fifth extensor compartmental artery (5ECA) both arise from the posterior division of the AIA and branch off separately at a proximal point. Because the 5ECA was found to contribute minimal nutrient vessels to the bone and the distal 4ECA was often very small, most studies utilizing these VBGs have combined them into the 4 + 5ECA where the proximal branchpoint is preserved such that the bone supplied by the 4ECA is perfused via the retrograde 5ECA. The bone flap is designed with its center 1.1 cm proximal to the radiocarpal joint.

Moran and colleagues reported on 26 patients treated with the 4 + 5ECA VBG for the management of Kienbocks disease. Long-term result showed that 77% of patients had no further evidence of lunate collapse on plain radiographs and 71% displayed evidence of lunate revascularization on MRI. Better than 90% reported that pain improved and grip strength improved to 89% of the contralateral hand after 31 months of follow-up. In a more recent study of 45 patients with Kienbocks disease (21 patients Lichtman Stage 2 and 24 patients Lichtman Stage 3A), half of each cohort underwent capitate shortening with 4 + 5ECA VBG and the other half underwent capitate shortening alone. The authors found that although there were no differences in stage 2 patients, stage 3A patients treated with capitate shortening and 4 + 5ECA demonstrated significantly improved pain reduction, increased ROM, increased grip strength, preserved carpal height, and better MMWSs than those treated only with capitate shortening. Additionally, the patients not treated with the 4 + 5ECA flap were more likely to progress to more advanced disease required scaphocapitate arthrodesis.