Upper Extremity Diaphyseal Osseous Gap Reconstruction with Free Vascularized Bone Flaps: A Scoping Review

Abstract

Background Reconstruction of upper extremity osseous diaphyseal defects often requires complex reconstructions. In this study, we characterized and summarized the available literature on free vascularized bone flap (VBF) reconstruction for upper extremity diaphyseal defects.

Methods A scoping review using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis extension for Scoping Reviews was conducted. A literature search of major electronic databases was conducted to identify journal articles relating to the management of VBF reconstruction of upper limb long bone defects. Articles with patient-level data were included. Descriptive statistics were performed using Python.

Results Overall, 364 patients were included in this study. The most common indications for VBFs included atrophic nonunion (125, 34.3%), postoncologic resection (125, 34.3%), septic nonunion (56, 15.4%), and trauma (36, 9.9%). Mean defect size was 8.53 ± 5.14 cm. A total of 67 (18.4%) cases had defects < 6 cm, and 166 cases (45.6%) had defects > 6 cm. The fibula was the most utilized VBF (272, 74.73%), followed by the medial femoral condyle flap (69, 18.96%). Overall, primary union rate was 87.1%. Subsequent flap fracture rate was 3.3%. There were only two (0.6%) VBF losses reported in the included cases, and donor-site complications were similarly rare (17, 4.7%).

Conclusion VBF reconstruction is often utilized for postoncologic defects and recalcitrant nonunions. The fibula is the most utilized VBF, but the medial femoral condyle flap is used frequently for smaller defects. VBF reconstruction demonstrates high union rates and low flap fracture rate across indications.

Keywords

upper extremity - osseous defects - fibula - vascularized bone - microsurgery

Upper extremity segmental defects (UESD) are osseous gaps that often arise from postoncologic resections, high-energy trauma, or recalcitrant nonunions. Their treatment requires complex, multidisciplinary interventions. Historically, segmental defects were treated with amputation, but advances in limb salvage have presented surgeons with multiple options for management.[1] Limb shortening, bone transport, nonvascularized bone grafting, and vascularized bone flaps (VBF) are commonly utilized techniques for the treatment of segmental defects, but patient-related, site-dependent, and surgeon-specific limitations influence clinical decision-making.[1] [2] [3]

Currently, treatment of segmental defects is highly variable.[1] Importantly, meticulous planning is required as each approach comes with inherent limitations. Upper extremity shortening is more tolerated than lower extremity shortening, but it is rarely performed due to potential alteration of musculotendinous dynamics.[4] Bone transport requires prolonged external fixation and can be complicated by pin-site infections, joint contractures, and neuropraxia.[2] Nonvascularized bone grafting is useful for smaller segmental defects but requires a vascularized wound bed to optimize osteoconduction and osteoinduction.[5] In addition, nonvascularized bone grafts are prone to resorption and infection, and patients may require serial operations for adequate bony reconstruction.[5] While VBFs require microsurgical expertise and are associated with greater donor-site morbidity, they provide vascularized bone with structural integrity that can achieve bony union in larger defects or fibrotic wound beds.[6] [7] Since they have been shown to have high union rates in the most complex of defects, they warrant closer examination for UESDs to derive a better understanding for their indications of use.[8] [9] [10] [11] [12]

Reviews on the use of VBFs for segmental defect reconstruction describe broad applications, but none have systematically assessed and comprehensively described their use in UESD reconstruction.[11] [12] Prior literature on UESDs have compared reconstructive modalities but often lack the granularity to guide decision-making.[13] [14] While many UESDs encountered in clinical practice may involve adjacent joints, the added biomechanical considerations of joint reconstruction are outside the scope of this review. With this study, we sought to systematically appraise the literature for upper extremity diaphyseal segmental defect reconstruction, with a particular focus on patient-level data.

Methods

Study Design

This work is a scoping review. It was performed using the JBI Manual for Evidence Synthesis and reporting following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) Extension for Scoping Reviews statement.[15] [16] The databases searched include MEDLINE (PubMed), Embase (Elsevier), and Web of Science Core Collection (Clarivate). The search was developed and conducted by a medical librarian, with input from other authors, and included a mix of keywords and subject headings for bone defects of the upper extremity and grafts or flaps. The original searches were conducted on January 11, 2023 and found 3,122 citations. The full, reproducible search strategies for all included databases are located in the [Supplementary Appendix] (available in online version only). Inclusion criteria included upper extremity diaphyseal segmental defects of the radius, ulna, or humerus reconstructed with free VBFs requiring microsurgical anastomosis. Exclusion criteria included nondiaphyseal reconstructions, nonupper extremity defects, nonvascularized reconstructions, review articles, and outcomes studies without patient-level data. Studies with heterogeneous patient populations were included if patient-level data could be derived per the aforementioned inclusion and exclusion criteria, and only applicable patient data were subsequently extracted.

Literature Review

After the search, all identified studies were uploaded into Covidence (Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia, www.covidence.org), a software system for managing systematic reviews, and duplicates were removed by the software (n = 1,287). A final set of 1,835 citations were left to be screened in the title/abstract phase. Study selection was performed independently by two authors (E.E., K.D.). All disagreements were resolved by a separate author (H.I.N.)

For the full-text screening stage, papers were also reviewed by two independent reviewers (H.I.N., J.K.) and were excluded if they did not meet the inclusion criteria. Any conflicts between the two independent reviewers were resolved through discussion between both authors. For papers not published in English that met the inclusion criteria during the title/abstract screening, the abstracts were reviewed for usable data. Due to restrictions in funding, we chose not to have these papers translated.

The study selection process is represented by flowchart as per PRISMA guidelines, which is displayed in the [Supplementary Appendices S1] and [S2] (available in online version only). For data collection, one author (J.K.) collected data from reports independently.

Fig. 1 Central illustration highlighting key findings of the review, including a summary of demographics, indications, and outcomes for vascularized bone flap reconstruction of upper extremity segmental defects. FFF, free fibula flap; IC, iliac crest; MFC, medial femoral condyle; S, scapula; UESD, upper extremity segmental defect; VBF, vascularized bone flap.

Data Items

Data extracted included patient demographics, operative details, and outcomes. Patient age, gender, defect etiology, affected upper extremity bones, number of prior surgeries, and defect size were recorded. If the patient sustained multiple bony defects requiring reconstruction, the larger defect size was used. Prior surgery counts included oncologic resection of malignancies and any surgical management of the segmental defect. Operative details, such as donor site, VBF components, size, recipient vessels, and fixation type were extracted. Ambiguous chronic or recalcitrant nonunions were typically considered atrophic, unless an infectious or hypertrophic etiology was mentioned. Outcomes of interest included time to union, number of surgeries, and primary union. Where studies reported proximal and distal union separately, the longer of the time periods was used. The number of surgeries counted total procedures required, including multistage repairs. Complications reported were VBF loss, delayed union, malunion, nonunion, reoperation, flap fracture, infection, and donor-site complications. Delayed union was coded separately from primary union. Reoperation counts included procedures that were not flap revisions, but results from complications of the flap, such as hematoma removal or revisions of a skin graft. Missing data were noted as “unspecified” and included in the analyses.

Data Analysis

Statistics including the frequency of events and characteristics, sample means, and standard deviations were reported in [Tables 1] [2] [3] [4] [5]. Unadjusted odds ratios and 95% confidence intervals were calculated for primary union and for each complication using surgical indications as exposure events. Only patients with complete data recorded for indication and the outcome in question were included in a given odds ratio calculation. These results were reported in [Table 6]. All statistics were generated using the SciPy and Numpy packages of Python 3.8.12.[17] [18]

Table 1 Patient demographics by defect etiology

	Oncologic	Atrophic	Infectious	Trauma	Other	Total
Subjects	125	125	56	36	22	364
Age (y)
Mean (SD)	26.9 (15.51)	41.65 (14.96)	34.88 (13.75)	32.64 (12.86)	35.53 (15.18)	34.11 (15.88)
Unspecified	13	20	4	0	3	40
Gender
Male	63 (50.4%)	65 (52.0%)	33 (58.93%)	23 (63.89%)	16 (72.73%)	200 (54.95%)
Female	48 (38.4%)	40 (32.0%)	9 (16.07%)	11 (30.56%)	3 (13.64%)	111 (30.49%)
Unspecified	14 (11.2%)	20 (16.0%)	8 (14.29%)	2 (5.56%)	3 (13.64%)	47 (12.91%)
Limb segment
Humerus	61 (48.8%)	69 (55.2%)	11 (19.64%)	13 (36.11%)	1 (4.55%)	155 (42.58%)
Radius	39 (31.2%)	19 (15.2%)	15 (26.79%)	11 (30.56%)	9 (40.91%)	93 (25.55%)
Ulna	9 (7.2%)	26 (20.8%)	9 (16.07%)	4 (11.11%)	8 (36.36%)	56 (15.38%)
Radius and ulna	1 (0.8%)	11 (8.8%)	21 (37.5%)	7 (19.44%)	1 (4.55%)	41 (11.26%)
Unspecified	15 (12.0%)	0 (0.00%)	0 (0.00%)	1 (2.78%)	3 (13.64%)	19 (5.22%)
No. of prior surgeries
Mean (SD)	1.07 (0.43)	1.97 (1.21)	2.62 (1.81)	0.9 (0.91)	2.2 (1.42)	1.79 (1.33)
Range	1.0–4.0	0.0–6.0	0.0–8.0	0.0–3.0	1.0–6.0	0.0–8.0
Unspecified	55	14	6	16	7	98
Defect size (cm)
Mean (SD)	12.34 (3.54)	3.91 (3.4)	9.25 (3.55)	10.69 (3.77)	2.9 (0.14)	8.53 (5.14)
Range	4.0–25.0	0.0–14.0	2.0–16.0	6.0–20.0	2.8–3.0	0.0–25.0
Unspecified	37	38	26	10	20	131

Abbreviation: SD, standard deviation.

Table 2 Vascularized bone flap details by defect etiology

	Oncologic	Atrophic	Infectious	Trauma	Other	Total
Donor site
Fibula	124 (99.20%)	56 (44.80%)	51 (91.07%)	28 (77.78%)	13 (59.09%)	272 (74.73%)
MFC	0 (0.00%)	64 (51.20%)	2 (3.57%)	1 (2.78%)	2 (9.09%)	69 (18.96%)
Iliac crest	0 (0.00%)	1 (0.80%)	0 (0.0%)	4 (11.11%)	0 (0.0%)	5 (1.37%)
Scapula	1 (0.80%)	3 (2.40%)	1 (1.79%)	3 (8.33%)	0 (0.0%)	8 (2.2%)
Rib	0 (0.00%)	1 (0.80%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	1 (0.27%)
Fibula and MFC	0 (0.00%)	0 (0.00%)	1 (1.79%)	0 (0.0%)	0 (0.0%)	1 (0.27%)
Unspecified	0 (0.00%)	0 (0.00%)	1 (1.79%)	0 (0.0%)	7 (31.82%)	8 (2.2%)
Components
Osseous	58 (46.40%)	69 (55.20%)	35 (62.5%)	13 (36.11%)	4 (18.18%)	179 (49.18%)
Osteocutaneous	38 (30.40%)	41 (32.80%)	16 (28.57%)	18 (50.0%)	11 (50.0%)	124 (34.07%)
Osteomyocutaneous	4 (3.20%)	2 (1.60%)	1 (1.79%)	3 (8.33%)	0 (0.0%)	10 (2.75%)
Unspecified	25 (20.00%)	13 (10.40%)	4 (7.14%)	2 (5.56%)	7 (31.82%)	51 (14.01%)
VBF length (cm)
Mean (SD)	15.77 (4.10)	10.82 (4.28)	10.79 (3.14)	13.69 (5.39)	10.18 (1.78)	13.3 (4.69)
Range	7.00–32.00	4.50–22.00	6.0–18.0	3.7–22.0	7.0–12.0	3.7–32.0
Unspecified	31	83	14	5	11	144
Recipient vessels
Radial	27 (21.60%)	15 (12.00%)	21 (37.5%)	9 (25.0%)	3 (13.64%)	75 (20.6%)
Ulnar	6 (4.80%)	8 (6.40%)	11 (19.64%)	1 (2.78%)	1 (4.55%)	27 (7.42%)
Brachial	41 (32.80%)	30 (24.00%)	7 (12.5%)	13 (36.11%)	1 (4.55%)	92 (25.27%)
Radial and ulnar	0 (0.00%)	0 (0.00%)	1 (1.79%)	0 (0.0%)	0 (0.0%)	1 (0.27%)
Other	9 (7.20%)	0 (0.00%)	0 (0.0%)	1 (2.78%)	0 (0.0%)	10 (2.75%)
Unspecified	42 (33.60%)	72 (57.60%)	16 (28.57%)	12 (33.33%)	17 (77.27%)	159 (43.68%)
Fixation
External fixation	3 (2.40%)	1 (0.80%)	1 (1.79%)	5 (13.89%)	1 (4.55%)	11 (3.02%)
Plates and screws	73 (58.40%)	68 (54.40%)	40 (71.43%)	19 (52.78%)	4 (18.18%)	204 (56.04%)
K-wire	9 (7.20%)	2 (1.60%)	0 (0.0%)	0 (0.0%)	1 (4.55%)	12 (3.3%)
Intramedullary nail	0 (0.00%)	9 (7.20%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	9 (2.47%)
Other	0 (0.00%)	1 (0.80%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	1 (0.27%)
Multiple	4 (3.20%)	2 (1.60%)	6 (10.71%)	5 (13.89%)	0 (0.0%)	17 (4.67%)

Abbreviations: K-wire, Kirschner wire; MFC, medial femoral condyle; SD, standard deviation; VBF, vascularized bone flap.

Table 3 Outcomes by defect etiology

	Oncologic	Atrophic	Infectious	Trauma	Other	Total
Time to union (mo)
Any donor site
Mean (SD)	7.33 (5.12)	5.30 (3.60)	5.81 (2.76)	6.04 (4.89)	3.75 (2.42)	6.03 (4.19)
Range	2.00–36.00	1.50–27.00	1.50–13.00	2.00–20.00	1.50–9.00	1.5–36.0
Unspecified	48	30	13	15	12	118
Free fibula
Mean (SD)	7.38 (5.14)	5.73 (2.93)	5.91 (2.74)	6.73 (5.43)	5.67 (3.51)	6.58 (4.28)
Range	2.00–36.00	2.0–12.8	2.5–13.0	2.0–20.0	2.0–9.0	2.0–36.0
Unspecified	48	20	10	12	0	99
MFC
Mean (SD)	–	4.55 (2.73)	–	3.0 (nan)	4.25 (1.06)	4.51 (2.67)
Range	–	1.5–18.0	–	3.0–3.0	3.5–5.0	1.5–18.0
Unspecified	–	10	2	0	0	12
Iliac crest
Mean (SD)	–	5.0 (nan)	–	4.0 (0.82)	–	4.2 (0.84)
Range	–	5.0–5.0	–	3.0–5.0	–	3.0–5.0
Unspecified	–	0	–	0	–	0
Scapula
Mean (SD)	4.0 (nan)	6.67 (2.31)	–	–	–	6.0 (2.31)
Range	4.0–4.0	4.0–8.0	–	–	–	4.0–8.0
Unspecified	0	0	1	3	–	4
Primary union	111 (88.80%)	109 (87.20%)	49 (87.50%)	32 (88.89%)	16 (72.73%)	317 (87.09%)
Complications
VBF loss	1 (0.80%)	0 (0.00%)	1 (1.79%)	0 (0.00%)	0 (0.00%)	2 (0.55%)
Malunion	1 (0.80%)	1 (0.80%)	1 (1.79%)	1 (2.78%)	0 (0.00%)	4 (1.1%)
Nonunion	6 (4.80%)	3 (2.40%)	4 (7.14%)	4 (11.11%)	1 (4.55%)	18 (4.95%)
Reoperation	22 (17.60%)	16 (12.80%)	10 (17.86%)	12 (33.33%)	1 (4.55%)	61 (16.76%)
Flap fracture	9 (7.20%)	3 (2.40%)	0 (0.00%)	0 (0.00%)	0 (0.00%)	12 (3.3%)
Infection	4 (3.20%)	1 (0.80%)	3 (5.36%)	2 (5.56%)	0 (0.00%)	10 (2.75%)
Donor-site complication	6 (4.80%)	11 (8.80%)	0 (0.00%)	0 (0.00%)	0 (0.00%)	17 (4.67%)

Abbreviations: MFC, medial femoral condyle; SD, standard deviation; VBF, vascularized bone flap; nan, not a number (signifying inability to compute a standard deviation).

Table 4 Defect size by indication

	<6 cm	≥6 cm	Unspecified
VBF indication
Oncologic	1 (1.49%)	87 (52.41%)	37 (28.24%)
Atrophic	61 (91.04%)	26 (15.66%)	38 (29.01%)
Infectious	3 (4.48%)	27 (16.27%)	26 (19.85%)
Trauma	0 (0.00%)	26 (15.66%)	10 (7.63%)
Other	2 (2.99%)	0 (0.00%)	20 (15.27%)
Total	67	166	131

Abbreviation: VBF, vascularized bone flap.

Table 5 Defect size by vascularized bone flap

	<6 cm	≥6 cm	Unspecified
VBF selection
Fibula	14 (20.90%)	158 (95.18%)	100 (76.34%)
MFC	50 (74.63%)	1 (0.60%)	18 (13.74%)
Iliac crest	0 (0.0%)	5 (3.01%)	0 (0.0%)
Scapula	3 (4.48%)	0 (0.0%)	5 (3.82%)
Rib	0 (0.0%)	1 (0.60%)	0 (0.0%)
Multiple	0 (0.0%)	1 (0.60%)	0 (0.0%)
Unspecified	0 (0.0%)	0 (0.0%)	8 (6.11%)
Total	67	166	131

Abbreviations: MFC, medial femoral condyle; VBF, vascularized bone flap.

Table 6 Odds ratios for complications by indication

	Oncologic	Atrophic	Trauma	Infectious
Complication
Nonunion	0.94 (0.34–2.62)	0.36 (0.1–1.27)	1.65 (0.52–5.29)	2.73 (0.84–8.89)
Delayed union	0.67 (0.28–1.57)	0.97 (0.05–5.04)	1.59 (0.51–4.96)	1.36 (0.52–3.56)
Malunion	0.62 (0.06–5.99)	0.52 (3.01%)	3.23 (0.32–32.06)	1.7 (00.17–16.69)
Reoperation	0.9 (0.5–1.62)	0.64 (0.34–1.2)	2.99 (1.36–6.56)	0.97 (0.46–2.07)
Graft fracture	5.5 (1.46–20.76)	0.56 (0.15–2.12)	–	–

Note: Bolded metrics represent significant findings, with p < 0.05 on statistical analysis.

Results

Patient Characteristics

A total of 364 individual patients from 54 studies were included for study ([Supplementary Appendix S3], available in online version only). Indications for VBF reconstruction of UESDs were atrophic nonunion in 125 (34.3%) patients, postoncologic resection in 125 (34.3%), septic nonunion in 56 (15.4%), and acute trauma in 36 (9.9%). Patients had an average of 1.8 ± 1.3 prior procedures and presented with an average segmental defect of 8.5 ± 5.1 cm. Humeral defects were observed in 155 (42.6%) patients, radial defects in 93 (25.6%) patients, ulnar defects in 56 (15.4%) patients, and defects of both forearm bones in 41 (11.3%) patients. [Table 1] displays more granular demographic data categorized by defect etiology.

Operative Details

Overall, the free vascularized fibula flap (FFF) was the most utilized VBF, constituting 272 (74.7%) reconstructions. The medial femoral condyle (MFC) was the second most frequently harvested in 69 (19.0%) cases, followed by the scapular (8, 2.2%) and iliac crest (5, 1.4%) flaps. A rib flap was used in 1 (0.3%) case. VBF components were osseous in 179 (49.2%), osteocutaneous in 124 (34.1%), and osteomyocutaneous in 10 (2.8%). The average VBF length harvest was 13.3 ± 4.69 cm. For recipient vessel selection, the radial artery was chosen in 75 (20.6%) patients, the ulnar artery in 27 (7.4%), the brachial artery in 92 (25.3%), and both radial and ulnar arteries in 1 (0.3%). The remaining reports did not specify recipient vessel selection. Plates and screws were the most common fixation method, used to stabilize 204 (56.04%) cases. Reconstructions with Kirschner wires (12, 3.3%), external fixation (11, 3.02%), or multimodal (17, 4.67%) fixation methods were less frequent.

[Table 2] provides details on VBF selection and components, stratified by defect etiology. Postoncologic reconstruction was almost exclusively performed with fibular VBFs (124, 99.2%), with a single use of a scapular flap (1, 0.8%). Component type was varied, with 58 (46.4%) osseous VBF and 38 (30.4%) osteocutaneous VBFs. A mean length of 15.77 ± 4.10 cm was harvested for each VBF, and recipient vessels were the brachial (41, 32.8%) or radial (27, 21.6%) arteries most frequently. Plates and screws were the most utilized (73, 58.4%) form of fixation.

Atrophic nonunions were predominantly reconstructed with MFC (64, 51.20%) or fibular (56, 44.8%) flaps. Pure osseous flaps were used in 69 (55.2%) cases, although cutaneous (41, 32.8%) and myocutaneous (2, 1.6%) components were occasionally incorporated. VBFs provided 10.82 ± 4.28 cm of coverage on average and were commonly anastomosed to recipient brachial (30, 24.0%) and radial (15, 12.0%) arteries. Cases primarily relied on plate and screws for fixation (68, 54.4%) or intramedullary nails (9, 7.2%).

Infected nonunions overwhelmingly relied on fibular flaps (51, 91.07%) consisting of osseous (35, 62.5%) or osteocutaneous (16, 28.6%) components. VBFs had a mean of 10.79 ± 3.14 cm and vascularized with recipient radial (21, 37.5%) or ulnar (11, 19.6%) arteries most frequently. A total of 40 (71.4%) cases were fixed with plates and screws, with multiple fixation methods used in 6 (10.7%) cases.

Patients with traumatic injuries were often reconstructed with fibular (28, 77.8%) or iliac crest (4, 11.1%) flaps. Osteocutaneous VBFs were used in 18 (50.0%) cases, osseous in 13 (36.1%), and osteomyocutaneous in 3 (8.3%). The average VBF length was 13.69 ± 5.39 cm, with anastomosis to brachial (13, 36.1%) or radial (9, 25.0%) arteries most typical. Fixation was mainly plates and screws (19, 52.8%), external fixation (5, 13.9%), or multiple modalities (5, 13.9%).

Outcomes

Across the cohort, primary union was achieved in 317 (87.1%) cases, delayed union was reported for 29 (8.0%) patients, malunion in 4 (1.1%), reoperation in 61 (16.8%), and nonunion in 18 (5.0%). Detailed metrics on time to union by flap choice can be found in [Table 3], among other outcomes. Union occurred within 6.03 ± 4.19 months on average. The VBF was fractured in 12 (3.3%) patients and surgical site infection occurred in 10 (2.8%) patients. VBF loss occurred in 2 (0.6%) cases. Seventeen (4.7%) patients suffered donor-site complications.

Postoncologic reconstructions achieved primary union in 111 (88.8%) cases and nonunion in 6 (4.8%). Union was observed at 7.3 ± 5.1 postoperative month on average. VBF loss was reported in 1 (0.8%) patient, malunion in 1 (0.8%), VBF fracture in 9 (7.2%), infection in 4 (3.2%), and donor-site complications in 6 (4.8%). Donor-site complications included foot weakness in 1 patient, pain in 1, nerve injury in 5, ankle instability in 1, and wound dehiscence in 1 patient. Finally, 22 (17.6%) patients had a second reconstructive procedure. Subgroup analyses of VBF choice by defect etiology and defect size are shown in [Tables 4] and [5].

Patients with atrophic nonunions achieved primary union in 109 (87.2%) cases and nonunion in 3 (2.4%). Union was noted at month 5.3 ± 3.6, on average. No patients had VBF loss, although malunion was reported in 1 (0.8%) case, VBF fracture in 3 (2.4%), infection in 1 (0.8%), and donor-site complications in 11 (8.8%) patients. Reoperation was recommended and performed for 16 (12.8%) patients. Donor-site complications included poor skin graft taken in 3 patients, hematoma in 2, pain in 2, dysesthesia in 2, and fracture in 2 patients. One instance each of seroma, infection, hypergranulation tissue, nerve injury, readmission, ankle instability, ligament injury, and wound dehiscence was reported among patients with donor-site morbidity.

Infectious nonunions had primary union in 49 (87.5%) cases and nonunion in 4 (7.1%). An average of 5.8 ± 2.8 months were required for union. No patients experienced VBF fracture or donor-site complications, with VBF loss noted in 1 (1.8%) patient, malunion in 1 (1.8%), and infection in 3 patients (5.4%). Ten (17.9%) patients required a secondary operation.

Traumatic injuries had primary union in 32 (88.9%) cases, delayed union in 4 (11.1%), and nonunion in 4 (11.1%) cases. Union was typically reported at postoperative month 6.0 ± 4.9. No patients experienced VBF loss, VBF fracture, or donor-site complications. One (2.8%) patient exhibited a malunion and two (5.6%) developed infection. Twelve (33.3%) patients had a secondary operation.

Subgroup analysis for complications by indication revealed an unadjusted odds ratio of 5.5 (1.46–20.76) for flap fracture in oncologic reconstructions. In addition, acute traumatic reconstructions were more likely to require secondary operations with an unadjusted odds ratio of 2.99 (1.36–6.56). Otherwise, nonunion rates were similar between groups ([Table 6]).

Discussion

In this study, we sought to systematically appraise the case literature on VBF reconstruction of diaphyseal UESDs. A graphic of study findings is shown in [Fig. 1]. Overall, the FFF is the most utilized VBFs in the case literature, although the MFC is more commonly used for smaller defects. The most common indications for VBF reconstruction are postoncologic resections, which generate large segmental defects, and chronic nonunions, which are associated with smaller defects but are known to have hostile wound beds. VBFs were much less commonly used for acute traumatic defects, as other segmental defect reconstruction techniques may be preferentially utilized before VBFs. While a majority of VBFs are harvested as osseous-only VBFs, a large minority are harvested as osteocutaneous VBFs, likely for concomitant soft tissue reconstruction or for VBF monitoring. Primary union was achieved in most cases, but delayed union and nonunion were not uncommon. Flap fractures were mostly observed in postoncologic reconstructions, likely attributable to the larger reconstructions required. Donor-site morbidity was minimal, although present, with VBF harvests.

The management of segmental bone defects is varied due to patient-related and surgeon-specific factors. Surgeons tailor their treatment approaches to the defects at hand, and certain treatment paradigms have proven successful. Small defects with vascularized wound beds achieve adequate bony union when reconstructed with many approaches, and efficient reconstructions with minimal donor-site morbidity are used effectively.[19] [20] Conversely, complication rates have been shown to be unacceptably high in cases with large defects or significant soft tissue trauma.[21] In these cases, VBFs may prove more effective than nonvascularized alternatives.

Evidence suggests that vascularized bone generates architecturally durable bone expeditiously. In a canine ulnar autograft model, Shaffer et al demonstrated that vascularized repairs had greater strength and stiffness than nonvascularized repairs.[22] In a subsequent study of a canine spinal surgery model, they demonstrated that vascularized rib flaps maintained their cortical integrity, whereas nonvascularized grafts showed porosity and fragmentation over time.[23] This finding has been challenged in other experiments, which argue that over time, both nonvascularized and VBF have similar strength.[24] They demonstrate that initially, nonvascularized bone is weaker as it undergoes necrosis and revascularization, but that eventually the resultant bone has adequate integrity at 1 year. Nonetheless, this once more stresses the importance of wound bed vascularity to the success of nonvascularized bone grafts, whereas VBFs can heal independently due to their native blood supply.

Clinical studies have similarly demonstrated the efficacy of VBF reconstructions. A landmark study that assessed mandibular reconstruction with VBFs versus nonvascularized bone demonstrated that patients who underwent VBF reconstruction had higher union rates, despite having larger defects and higher rates of radiation.[25] Complex proximal pole scaphoid nonunions associated with avascular necrosis have been shown to have higher union rates when reconstructed with VBFs as opposed to nonvascular alternatives.[26] [27] VBFs have also been shown to be effective when applied to recalcitrant forearm, clavicular, and humeral nonunions.[28] [29] [30] As such, certain VBFs have become workhorse flaps in the reconstructive surgeon's armamentarium.

The FFF is a source of vascularized, structural bone that has proven safe and effective for extremity reconstruction.[31] While it has been broadly applied to head and neck and lower extremity reconstruction, it also serves as an ideal VBF for upper extremity reconstruction.[12] [32] [33] The length and architecture of the bone make it a “like-for-like” match with the radius and ulna, and as such, it does not require flap hypertrophy to adequately bear weight.[31] Up to 26 cm of fibula can be harvested, which makes it the only option for massive segmental defects that often result from oncologic resections.[31] In addition, multiple studies have demonstrated that donor-site morbidity is minimal.[6] [11] [13] The lateral cortex of the fibula provides a robust platform of structural bone for internal fixation. In addition, the VBF is based on the peroneal artery, which has been shown to have reliable skin perforators that afford the option of an osteocutaneous VBF if needed.[13] [34] With its independent blood supply, it can be used effectively in hostile wound beds associated with atrophic and infectious nonunions, although emerging evidence suggests that the MFC may be more poised for reconstruction of these defects.[35]

The MFC corticoperiosteal VBF is ideal for small defects, as shown in this study. Based on the descending genicular artery, the MFC provides corticoperiosteal bone that can be used for extra-articular bony reconstruction.[36] The thin, pliable geometry of the MFC VBF is particularly useful for defects that do not require long structural bone. While the pedicle provides a large zone of periosteal perfusion, a large flap harvest may have unacceptable donor-site morbidity on the femur—as such, smaller harvests are recommended.[36] Evidence is mounting that the MFC is well poised for use in recalcitrant nonunions and hostile wound beds.[37] [38] [39] Recently, a multi-institutional retrospective case series by Guzzini et al reported a 94.9% union rate for MFC reconstructions of long bone recalcitrant nonunions.[40] These findings are corroborated by the high union rates reported in this study.

Ultimately, VBF reconstructions are arduous and time-consuming procedures, although consistency in technique and application can improve efficiency. Careful patient selection is important, and patients who present with complex defects or prior treatment failures may benefit from consideration of VBF reconstructions. Multidisciplinary collaboration is critical as these are complex operations that benefit from two team approaches to improve care. In addition, many nuances with respect to these technically challenging operations remain inadequately characterized due to the rarity of these reconstructions. Technical modifications such as step-cut osteotomies for flap fixation and double-barreled or Capanna technique FFF configurations come with added considerations that require further clinical investigation.

Limitations to this study include the study selection criteria, which were limited to studies with patient-level data. This has an unpredictable effect on the outcomes reported, as these studies were not controlled and are susceptible to many forms of bias. For example, flap failure is traditionally difficult to ascertain in bone-only flaps without implantable Doppler monitoring, and as such, flap failure rates may be underreported. Conversely, this allowed for more granular assessment of the literature. As this is a scoping review, an in-depth analysis of sources of bias was not conducted, detailed inferential statistics were not performed between different types of VBF reconstructions. An analysis of predictors of nonunion was performed, although the results should not be extrapolated broadly as the study design was not intended to explore this question. This study did not systematically assess patient-reported outcomes or functional scores, as these are highly heterogeneous metrics that are very inconsistently reported. Instead, a focus on primary union, as reported in included articles, was the focus of the outcomes analysis. Post-operative protocols for VBF reconstructions were not comprehensively analyzed, although this is a critical component to care. Reporting of protocols was also highly heterogeneous and inconsistent and warrants further investigation. Finally, a robust comparison between VBFs and other reconstructive modalities is needed, and further high-quality research is required in this area.

Conclusion

VBF reconstruction of upper extremity diaphyseal defects is primarily utilized for postoncologic defects and recalcitrant nonunions. The FFF is the most utilized VBF for upper extremity reconstruction, but the MFC VBF is used frequently for smaller defects. VBF reconstruction demonstrates a high flap union rate and low flap fracture rate across indications, making it a valuable tool in the treatment of these challenging defects.

Conflict of Interest

None declared.

Note

S.K.M. serves as a consultant for Restor3D.

• References

• 1 DeCoster TA, Gehlert RJ, Mikola EA, Pirela-Cruz MA. Management of posttraumatic segmental bone defects. J Am Acad Orthop Surg 2004; 12 (01) 28-38
• 2 Dahl MT, Morrison S. Segmental bone defects and the history of bone transport. J Orthop Trauma 2021; 35 (Suppl. 04) S1-S7
• 3 Weiland AJ. Current concepts review: vascularized free bone transplants. J Bone Joint Surg Am 1981; 63 (01) 166-169
• 4 Kusnezov N, Dunn JC, Stewart J, Mitchell JS, Pirela-Cruz M. Acute limb shortening for major near and complete upper extremity amputations with associated neurovascular injury: a review of the literature. Orthop Surg 2015; 7 (04) 306-316
• 5 Klifto CS, Gandi SD, Sapienza A. Bone graft options in upper-extremity surgery. J Hand Surg Am 2018; 43 (08) 755-761.e2
• 6 Vail TP, Urbaniak JR. Donor-site morbidity with use of vascularized autogenous fibular grafts. J Bone Joint Surg Am 1996; 78 (02) 204-211
• 7 Pirela-Cruz MA, DeCoster TA. Vascularized bone grafts. Orthopedics 2013; 17 (05) 407-412
• 8 Avraham T, Franco P, Brecht LE. et al. Functional outcomes of virtually planned free fibula flap reconstruction of the mandible. Plast Reconstr Surg 2014; 134 (04) 628e-634e
• 9 Ackerman DB, Rose PS, Moran SL, Dekutoski MB, Bishop AT, Shin AY. The results of vascularized-free fibular grafts in complex spinal reconstruction. J Spinal Disord Tech 2011; 24 (03) 170-176
• 10 Jones Jr DB, Rhee PC, Bishop AT, Shin AY. Free vascularized medial femoral condyle autograft for challenging upper extremity nonunions. Hand Clin 2012; 28 (04) 493-501
• 11 Chen CM, Disa JJ, Lee HY. et al. Reconstruction of extremity long bone defects after sarcoma resection with vascularized fibula flaps: a 10-year review. Plast Reconstr Surg 2007; 119 (03) 915-924 , discussion 925–926
• 12 Awad ME, Altman A, Elrefai R, Shipman P, Looney S, Elsalanty M. The use of vascularized fibula flap in mandibular reconstruction; a comprehensive systematic review and meta-analysis of the observational studies. J Craniomaxillofac Surg 2019; 47 (04) 629-641
• 13 Hollenbeck ST, Komatsu I, Woo S. et al. The current role of the vascularized-fibular osteocutaneous graft in the treatment of segmental defects of the upper extremity. Microsurgery 2011; 31 (03) 183-189
• 14 Ferreira N, Saini AK, Birkholtz FF, Laubscher M. Management of segmental bone defects of the upper limb: a scoping review with data synthesis to inform decision making. Eur J Orthop Surg Traumatol 2021; 31 (05) 911-922
• 15 Tricco AC, Lillie E, Zarin W. et al. PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med 2018; 169 (07) 467-473
• 16 Peters MDJ, Godfrey C, McInerney P, Munn Z, Tricco AC, Khalil H. Chapter 11: Scoping reviews. In: Aromataris E, Munn Z. (eds). JBI Manual for Evidence Synthesis. JBI; 2020.
• 17 Harris CR, Millman KJ, van der Walt SJ. et al. Array programming with NumPy. Nature 2020; 585 (7825) 357-362
• 18 Virtanen P, Gommers R, Oliphant TE. et al; SciPy 1.0 Contributors. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods 2020; 17 (03) 261-272
• 19 Ring D, Allende C, Jafarnia K, Allende BT, Jupiter JB. Ununited diaphyseal forearm fractures with segmental defects: plate fixation and autogenous cancellous bone-grafting. J Bone Joint Surg Am 2004; 86 (11) 2440-2445
• 20 Luo TD, Nunez Jr FA, Lomer AA, Nunez Sr FA. Management of recalcitrant osteomyelitis and segmental bone loss of the forearm with the Masquelet technique. J Hand Surg Eur Vol 2017; 42 (06) 640-642
• 21 Borzunov DY, Kolchin SN, Malkova TA. Role of the Ilizarov non-free bone plasty in the management of long bone defects and nonunion: problems solved and unsolved. World J Orthop 2020; 11 (06) 304-318
• 22 Shaffer JW, Field GA, Goldberg VM, Davy DT. Fate of vascularized and nonvascularized autografts. Clin Orthop Relat Res 1985; (197) 32-43
• 23 Shaffer JW, Davy DT, Field GA, Bensusan JS. Temporal analysis of vascularized and nonvascularized rib grafts in canine spine surgery. Spine 1989; 14 (07) 727-732
• 24 Dell PC, Burchardt H, Glowczewskie Jr FP. A roentgenographic, biomechanical, and histological evaluation of vascularized and non-vascularized segmental fibular canine autografts. J Bone Joint Surg Am 1985; 67 (01) 105-112
• 25 Foster RD, Anthony JP, Sharma A, Pogrel MA. Vascularized bone flaps versus nonvascularized bone grafts for mandibular reconstruction: an outcome analysis of primary bony union and endosseous implant success. Head Neck 1999; 21 (01) 66-71
• 26 Aibinder WR, Wagner ER, Bishop AT, Shin AY. Bone grafting for scaphoid nonunions: is free vascularized bone grafting superior for scaphoid nonunion?. Hand (N Y) 2019; 14 (02) 217-222
• 27 Ribak S, Medina CEG, Mattar Jr R, Ulson HJR, Ulson HJ, Etchebehere M. Treatment of scaphoid nonunion with vascularised and nonvascularised dorsal bone grafting from the distal radius. Int Orthop 2010; 34 (05) 683-688
• 28 Fuchs B, Steinmann SP, Bishop AT. Free vascularized corticoperiosteal bone graft for the treatment of persistent nonunion of the clavicle. J Shoulder Elbow Surg 2005; 14 (03) 264-268
• 29 Dell PC, Sheppard JE. Vascularized bone grafts in the treatment of infected forearm nonunions. J Hand Surg Am 1984; 9 (05) 653-658
• 30 Beredjiklian PK, Hotchkiss RN, Athanasian EA, Ramsey ML, Katz MA. Recalcitrant nonunion of the distal humerus: treatment with free vascularized bone grafting. Clin Orthop Relat Res 2005; (435) 134-139
• 31 Malizos KN, Zalavras CG, Soucacos PN, Beris AE, Urbaniak JR. Free vascularized fibular grafts for reconstruction of skeletal defects. J Am Acad Orthop Surg 2004; 12 (05) 360-369
• 32 Richard MJ, DiPrinzio EV, Lorenzana DJ, Whitlock KG, Hein RE, Urbaniak JR. Outcomes of free vascularized fibular graft for post-traumatic osteonecrosis of the femoral head. Injury 2021; 52 (12) 3653-3659
• 33 Berend KR, Gunneson EE, Urbaniak JR. Free vascularized fibular grafting for the treatment of postcollapse osteonecrosis of the femoral head. J Bone Joint Surg Am 2003; 85 (06) 987-993
• 34 Garvey PB, Chang EI, Selber JC. et al. A prospective study of preoperative computed tomographic angiographic mapping of free fibula osteocutaneous flaps for head and neck reconstruction. Plast Reconstr Surg 2012; 130 (04) 541e-549e
• 35 Guidi M, Guzzini M, Civitenga C. et al. Multifactorial analysis of treatment of long-bone nonunion with vascularized and nonvascularized bone grafts. J Hand Microsurg 2022; 15 (02) 106-115
• 36 Higgins JP, Bürger HK. Osteochondral flaps from the distal femur: expanding applications, harvest sites, and indications. J Reconstr Microsurg 2014; 30 (07) 483-490
• 37 Kaiser D, Levin LS. Medial femoral condyle free flap for persistent osseous nonunion of the first metatarsophalangeal joint: a preliminary report of a new surgical indication for the medial femoral condyle free flap. Foot Ankle Orthop 2023; 8 (03) 24 730114231191135
• 38 Karunaratne YG, Romeo PB. Vascularized reconstruction of recalcitrant clavicular nonunion with the medial femoral condyle free flap: a systematic review of the literature. J Hand Microsurg 2024; 16 (01) 100016
• 39 DeKeyser GJ, Bailey TL, Higgins TF, Tyser AR. Treatment of recalcitrant femoral shaft nonunion with medial femoral condyle pedicled autograft: technical trick. J Orthop Trauma 2022; 36 (02) e80-e80
• 40 Guzzini M, Ciclamini D, Arioli L. et al. Correlation between risk factors and healing times in long bone nonunions treated with corticoperiosteal flap from the medial femoral condyle. J Reconstr Microsurg 2023; 39 (07) 502-507

Publication History

Received: 25 November 2023

Accepted: 05 May 2024

Accepted Manuscript online:
23 May 2024

Article published online:
21 June 2024