https://doi.org/10.1177/1071100717750837

Foot & Ankle International®2018, Vol. 39(5) 573 –584© The Author(s) 2018Reprints and permissions: sagepub.com/journalsPermissions.navDOI: 10.1177/1071100717750837journals.sagepub.com/home/fai

Article

Introduction

Lisfranc injuries involve a disruption of the tarsometatarsal joint (TMTJ) and typically occur following high-energy trauma.20 Without surgery, these injuries can lead to consid-erable pain and disability.23 Operative fixation for unstable fractures usually involves reduction followed by fixation using plates, screws, or wires. This can be performed percu-taneously, via a mini-open approach or via a formal arthrot-omy and open soft tissue dissection. Anatomic reduction can best be achieved via an open approach but may incur wound dehiscence, skin necrosis, soft tissue infection, neu-roma formation, and development of complex regional pain syndrome (CRPS).7,9,13,19,22,25,31-33

Traditionally, the approach to the midfoot has involved multiple longitudinal incisions over the dorsum of the foot separated by skin bridges.2,19 However, the risks of this approach include poor visualization of the transversely ori-ented TMTJ, as well as the potential for skin necrosis between the narrow skin bridges. In contrast, transverse approaches to the midfoot have also been described. This

750837 FAIXXX10.1177/1071100717750837Foot & Ankle InternationalPhilpott et alresearch-article2018

1Royal Melbourne Hospital, Melbourne, Australia

Corresponding Author:Andrew Philpott, MBBS, BMSci, Royal Melbourne Hospital, 64 Raleigh Street, Footscray, Melbourne, Victoria, Australia. Email: Andrew@andrewphilpott.com

Modified Dorsal Approach in the Management of Lisfranc Injuries

Andrew Philpott, MBBS, BMSci1 , Callum Lawford, MBBS1, Simon C. Lau, MBBS, BMSci, MSc1, Simon Chambers, MBBS, RCS1, Michael Bozin, MBBS1, and Andrew Oppy, MBBS, FRACS1

AbstractBackground: Open reduction and internal fixation of Lisfranc injuries has typically used multiple longitudinal incisions or a single transverse incision to approach the tarso-metatarsal joint (TMTJ). The incidence of wound-related complications is considerable. We describe a novel single-incision approach that utilizes subcutaneous windows to the medial TMTJ.Methods: A retrospective review identified 150 patients who underwent open reduction and internal fixation for Lisfranc injuries, via the modified dorsal approach, at our center between January 2011 and June 2016. Removal of hardware (ROH) was routinely undertaken in 105 patients at a median of 210 days postoperatively. Medical records were reviewed to record patient demographics, mechanism of injury, and operative details. Outpatient notes were reviewed to identify wound-related complications, including delayed wound healing, superficial infection, wound dehiscence, deep infection, complex regional pain syndrome (CRPS), neuroma, and impaired sensation. Median age was 37 years (range, 19-78 years). Seventy-three percent of patients (110) were male. Most frequent mechanisms of injury were motor vehicle accident (MVA), 39%; motorbike accident (MBA), 19%; and fall, 18%. Sixteen percent (24) of injuries were open. Five patients required soft tissue reconstruction at the primary operation. Median follow-up was 144 (range, 27-306) weeks.Results: Following the primary procedure, 14% of patients experienced wound-related complications including delayed healing (3%), superficial infection (5%), dehiscence (3%), complex regional pain syndrome (CRPS) (1%), and impaired sensation (1%). MBA injuries were at 15.1 times odds of superficial infection (P =.01) than were MVA injuries. Following ROH, 13% of patients experienced wound-related complications, including delayed healing (2%), superficial infection (8%), dehiscence (1%), CRPS (2%), and neuroma (1%). Overall, 5 patients returned to surgery for soft tissue reconstruction for wound dehiscence.Conclusion: The modified dorsal approach using intervals to the midfoot offers a viable alternative with comparable wound complication rates to existing midfoot approaches.Level of Evidence: Level IV, case series.

Keywords: midfoot, Lisfranc, operative, approach, complications

574 Foot & Ankle International 39(5)

incision is advantageous as it parallels lines of skin tension, but runs perpendicular to the dorsal neurovascular bundle, theoretically putting these structures at risk. Wound compli-cations have also been documented with this approach.13,33 There is much literature comparing the outcomes of various fixation methods and arthrodesis; however, few describe approach-related complications, and choice of approach remains largely surgeon dependent. In this study, we describe a modified dorsal approach that utilizes a single longitudinal incision with multiple intervals or windows to the midfoot and quantify the incidence of complications when used in a large consecutive series at our center. We expected a reduced rate of wound-related complications with this approach compared to existing approaches.

Methods

We retrospectively reviewed the hospital records of 179 patients who underwent open reduction and internal fixa-tion (ORIF) of Lisfranc injuries at our center between January 2011 and June 2016. These patients were identi-fied via a search of the hospital’s Medical Information System. Twenty-nine patients were operated via alternate approaches and were excluded from the cohort. The remaining 150 patients were included. The majority of patients had their primary surgery performed by the same surgeon, although 5 patients had surgery with other trauma surgeons at our institution. Operations were delayed until more life-threatening injuries were managed, the patient’s condition had been optimized, and soft tissue swelling had adequately subsided, as determined by the primary sur-geon, in order to safely undertake the procedure. Median age was 37 (range, 19-78) years, and 73% (110) of patients were male (Table 1). Delay between injury and surgery was a median of 14 (range, 0-31) days. Open injuries were sus-tained in 22% (13) of MVA injuries, 18% (2) of crush inju-ries, 28% (8) of MBA injuries, and 11% (1) of other injuries. There were no open injuries as a consequence of falls and sporting or biking accidents. Follow-up was to a median of 144 (range, 27-306) weeks postoperatively.

We examined medical records to record data such as demographics, mechanism of injury, operative approach, and fixation method. Comorbid medical conditions were grouped into 3 categories: mental illness or drug abuse, smoking, and systemic medical conditions (including isch-emic heart disease, peripheral vascular disease, cardiac fail-ure, diabetes mellitus, and cancer)

A review of outpatient notes was undertaken to identify complications. We grouped these into separate categories including delayed wound healing, superficial infection, wound dehiscence, CRPS, neuroma formation and impaired sensation. Delayed wound healing was defined as healing taking longer than the expected postoperative course and was managed with regular dressings until skin union

occurred. Superficial infection was defined as the presence of clinical signs of infection (pain, swelling, redness, heat or purulent drainage) within 30 days of surgery as previ-ously described by Horan et al8 and was managed with oral antibiotics. Wound dehiscence was defined as breakdown of the wound requiring further operative intervention as previously described by Wiewiorski et al.36 All cases of CRPS were referred to our specialist pain clinic. Complications were classified on a scale of 1 to 5 using the adapted Clavien-Dindo classification system, which reflects the impact upon the patient and the treatment required (Table 2).30

At our institution, it was protocol for patients who under-went plate fixation of the midfoot to have their bridging plates removed at the 6-month postoperative mark. We believe this encourages movement at the TMTJ and prevents the development of excessive arthrosis. The procedure was

Table 1. Patient, Injury, and Operative Characteristics (Median, Range).

Patients (n) 150Age, y 37 (19-78)Sex, male:female 110:40Comorbidity Mental illness or drug abuse 17 (11%) Smoking 22 (15%) Systemic medical condition 17 (11%)Mechanism of injury MVA 59 (39%) Fall 27 (18%) Crush 11 (7%) Sport 7 (5%) Bicycle 8 (5%) MBA 29 (19%) Other 9 (6%)Closed/open Closed 126 (84%) Open 24 (16%)Delay to fixation, d 14 (0-31)Columns fixed, n (%) 1 TMTJ 12 (8%) 1/2 TMTJ 13 (9%) 1/2/3 TMTJ 102 (68%) 2/3 TMTJ 15 (1%) Other 8 (5%)Soft tissue reconstruction No 145 Yes 5Split thickness skin graft 1Free flap 4ROH postoperative, wk 30 (8-84)Length of follow-up, wk 144 (27-306)

Abbreviations: ROH, removal of hardware; MBA, motorbike accident; MVA, motor vehicle accident; TMTJ, tarsometatarsal joint.

Philpott et al 575

performed via the same approach, typically in a day operat-ing suite, by a junior registrar who performed the case with only limited supervision. Tourniquet time was only intermit-tently recorded and so was excluded as one of our outcome measures.

We utilized descriptive statistics to compare complica-tion rates and type across each patient including predictors of complications in this study. Univariate analyses were performed to assess which of our patients were more likely to have complications. We also developed odds ratios to help describe the likelihood of various complications. P values are 2-sided and were calculated using Fisher exact test. This statistical analysis was performed using SPSS, Mac Version 24 (IBM SPSS Statistics, Armonk, NY).

Operative Technique

The patient was positioned supine with a thigh or calf tour-niquet. The knee was flexed over an appropriately sized support such as the obtuse angle bone foam small (Figure 1B) with the sole of the foot flat on the operating table. Image intensifier (OEC Elite C arm) was available and introduced perpendicular to the leg to provide foot radio-graphs as required (Figure 1C). The incision was made overlying the second metatarsal starting from the TMTJ and extending distally to the metatarsophalangeal joint (MTPJ) (Figure 1D). Further proximal extension was occasionally required, particularly to access the third metatarsal (MT) base. Cutaneous branches from the deep peroneal nerve were often encountered and were preserved where possible. Three windows were created to assess and manage involved TMTJs (Figure 2).

Medial Interval

A window was made medial to extensor hallucis longus (EHL), which allowed assessment to the first TMTJ (Figure 1F). If damage to the joint was in doubt, the dorsal capsule was opened and the joint was probed for stability. It is well known that the strong plantar structures are the most important for stability of the Lisfranc ligament complex.14 The dorsal structures have only a minimal contribution to joint stability, and we suspect opening them for assessment

is unlikely to induce instability. In our experience, the first TMTJ was frequently involved in the Lisfranc injury (Figure 1F).

Middle Interval

The second interval allowed access to the second TMTJ (Figure 1G). The neurovascular bundle lies closely in rela-tion to this joint, and the interval was developed most safely by starting on the MT shaft distally, and dissecting proxi-mally while staying close to the bone. The extensor digito-rum brevis (EDB) muscle belly crosses from laterally to medially at the proximal end and was retracted medially, protecting the neurovascular bundle without formally iden-tifying it.

Lateral Interval

A third interval was developed between the third and fourth tendon slips of extensor digitorum longus (EDL) (Figure 1H). There is a slightly wider interval between these 2 ten-dons. We have found it preferable to develop this separate interval rather than simply elevate all of the structures later-ally from the second interval because it was difficult to achieve a satisfactory drilling angle for applying the bridg-ing plate (Figure 1K).

Each TMTJ was assessed for injury and instability. Fixation was performed where necessary, but sparing unin-jured joints where possible. The most common pattern, in our experience, required plating of the first, second, and third TMTJs and a transarticular “Lisfranc screw” (Figure 3A). This was placed retrograde in an oblique orientation from the base of the second metatarsal, to the medial cunei-form. The starting point was lateral to the bridging plate, usually just distal to the third screw. A retrograde orienta-tion was preferable as it allowed the use of an image intensi-fier to efficiently position the screw at the apex of the second metatarsal base.

The Lisfranc injury is not one consistent fracture or dis-location, and common variants include extension proxi-mally to involve the intercuneiform joints (Figure 3B) or a less severe injury, sparing the medial column. Our approach allowed direct assessment of all of the medial joints that may have been involved. By proximal extension, it was even possible to place a dorsal bridging plate from the talar neck down to the second MT shaft, which has proven use-ful in cases with involvement of the naviculocuneiform joints, and concomitant comminuted fractures of the navic-ular itself.

Addressing the Lateral Column

Our preferred approach for the lateral column fourth/fifth rays was nonoperative if the length of the cuboid was

Table 2. Adapted Clavien-Dindo Classification of Complications.

Grade Post Primary Post Removal of Hardware Total

1 2 0 22 15 13 283 4 1 54 0 0 05 0 0 0

576 Foot & Ankle International 39(5)

Figure 1. (A, B) Setup and positioning; (C) image intensifier setup; (D, E) incision overlying second metatarsal from TMTJ to MTPJ; (F) creation of medial window medial to EHL; (G) middle window with EDB belly retracted medially in order to protect neurovascular bundle; (H) lateral window between the third and fourth tendon slips of EDL; (I) plate fixation of first TMTJ, (J) second TMTJ, and (K) third TMTJ. EDL, extensor digitorum longus; EHL, extensor hallucis longus; MTPJ, metatarsophalangeal joint; TMTJ, tarsometatarsal joint.

Philpott et al 577

unaffected. The joints between the fourth/fifth metatarsals and the cuboid are naturally mobile. We fear that transarticu-lar fixation is likely to lead to stiffness. Some patients did require a small external fixator from the calcaneus to fourth/fifth metatarsal bases or Kirschner wires to restore lateral column anatomy. These were typically removed at 6 weeks.

The tourniquet was deflated prior to skin closure to ensure optimal hemostasis. Skin closure was in a single layer with interrupted nylon. Adhesive dressings were applied, including a short leg splint.

Postoperative Protocol

Following 6 weeks, non–weight bearing radiographs were taken. Weight bearing was commenced at this stage with the protection of a CAM walker boot and a medial arch support. From 12 weeks until 6 months, a normal shoe was permit-ted, but only with the use of a medial arch support.

Hardware was removed electively at 6 months. Occasionally some of the screws had broken prior to removal, but one of the advantages of our operative tactic was that there were no transarticular screws (aside from the Lisfranc screw). Any screw fragments were safely buried within bone and could usually be left.

Results

Internal fixation was achieved using a combination of trans-articular Lisfranc screw, bridging plate, and Kirschner wires as required. Eight percent (12) of patients underwent inter-nal fixation of the first TMTJ only, 9% (13) fixation of the first/second TMTJs, 68% (102) fixation of the first/second/third TMTJs, 10% (15) fixation of the second/third TMTJs, and 6% (8) fixation of other patterns. Fourth/fifth TMTJs were unstable in 33% of patients (49): 23% (35) underwent percutaneous Kirschner wiring and 9% (14) external-fixa-tion. Four patients required free flap reconstruction and 1 required split thickness skin graft at the time of primary operation for management of soft tissue defects. Seventy percent (105) patients underwent subsequent removal of hardware (ROH) through the same incision at median 30 (range, 8-84) weeks postoperatively.

Complications After Primary Procedure

Following their primary procedure 14% (21) patients expe-rienced wound-related complications (Table 3). Three per-cent (4) patients required return to surgery for management of wound dehiscence (Table 4). Postoperative complica-tions following primary surgery were greater in high-energy (OR 1.65, 95% CI 0.52-5.22, P = .40) and open injuries (OR 2.47, 95% CI 0.85-7.19, P = .10) and in those who sustained crush injuries (OR 3.31, 95% CI 0.69-15.97, P = .14) or MBA injuries (OR 2.81, 95% CI 0.85-9.32, P = .09)

Figure 2. Diagram showing (A) medial, (B) middle, and (C) lateral windows to the tarsometatarsal joint.

Figure 3. (A) Postoperative radiographs: Fixation of first/second/third TMTJs with dorsal bridging plates and transarticular “Lisfranc” screw and (B) fixation of second/third TMTJs with dorsal bridging plates and screw fixation of first TMTJ/intercuneiform joints/navicular.

578 Foot & Ankle International 39(5)

(Table 5). MBA injuries were at 15.1 times odds of superfi-cial infection (95% CI 1.72-132.71, P = .01) compared to MVA injuries. When open injuries were excluded, MBA injuries were at 10.6 times odds of superficial infection (95% CI 1.10-101.60, P = .04) compared to MVA injuries. Patients requiring soft tissue reconstruction had 4.4 greater odds of complication (95% CI 0.69-28.20, P = .12) than those closed primarily.

Complications after ROH

Following removal of hardware 13% (14) patients experi-enced wound-related complications (Table 3). One percent (1) patients required return to surgery for management of wound dehiscence (Table 4). Postoperative complications following ROH were higher in patients whose initial injury was a result of high-energy trauma (OR 1.55, 95% CI 0.40-5.99, P = .53) (Table 6). Medical comorbidities were associ-ated with increased risk of complication following ROH: mental illness and drug abuse (OR 5.01, 95% CI 1.04-24.83, P = .05), smoking (OR 4.93, 95% CI 1.30-18.74, P = .02), and systemic medical condition (OR 4.44, 95% CI 0.93-21.26, P =.06). Soft tissue reconstruction at primary proce-dure was associated with increased incidence of complications following ROH (OR 2.26, 95% CI 0.22-23.35, P = .49).

Discussion

This study presents a large series of operatively managed Lisfranc injuries. Our key finding is that a modified dorsal approach based on intervals to the TMTJ offers a viable alternative to existing approaches. This study is the first to clearly document the results of longitudinal and dorsal sur-gery to the foot in such extensive numbers.

Complication rates in foot surgery are higher than the proximal lower limb, likely due to poorer perfusion pres-sures and difficulty in thorough disinfection of the foot.36,38 Furthermore, the approach to the midfoot risks damage to

the dorsalis pedis artery and the peroneal nerves, which may impede wound healing and contribute to impaired sen-sation and neuroma formation. Cadaveric studies reveal substantial variability in the origin, path, and branches of this neurovascular bundle.26,34 Vijayalakshmi et al found variant origin, course, or branches of the dorsalis pedis artery in 46% of cadavers.34 This poses a challenge in the approach to the midfoot.

The traditional approach utilizes multiple longitudinal incisions to access the TMTJs.2 Reported complication rates vary widely (Table 7). Henning et al reported a 3% incidence of cellulitis, with no wound breakdown and no neurologic complications in a prospective cohort of 32 patients undergoing ORIF or arthrodesis7; Mann et al reported 1 case of delayed healing, 1 case of dehiscence, and 3 cases of neuroma in 40 patients who underwent mid-foot arthrodesis19; Mulier et al reported sympathetic dystro-phy in 28% of 18 patients who underwent ORIF or arthrodesis22; and van Koperen et al reported 9% infection in 34 patients who underwent fixation of Lisfranc injuries.32 The variety of reported complication rates is likely owing to differences in patient cohorts and variability in the reporting of complications. Some authors required positive microbio-logical cultures prior to classifying a wound as infected,39 others required clinical signs alone,8 and many failed to report their criteria for identifying complications.

In the present cohort, 14% of patients experienced wound-related complications following their primary pro-cedure. The most frequent complications were superficial infection (5%), delayed healing (3%), and dehiscence (3%). These rates are comparable to those in the literature of 0% to 9% superficial infection and 0% to 13% delayed healing.7,12,13,17,19,25,29,31-33 The majority of patients under-went a subsequent removal of hardware procedure, and 13% of these patients experienced complications after this procedure. Our findings suggest that the use of a single incision with intervals to the TMTJ is a viable way of approaching and visualizing the medial Lisfranc injury, while minimizing soft tissue compromise.

A suggested alternate approach utilizes a transverse inci-sion, superficial to the neurovascular bundle, with multiple windows created to access each TMTJ. Vertullo et al expe-rienced 25% impaired sensation and 8% wound necrosis in a small series of predominantly elective midfoot opera-tions.33 Kupcha et al reported a series of 60 traumatic Lisfranc injuries with an 8% rate of impaired sensation and 8% rate of delayed wound healing postoperatively.13 The authors attributed these complications in part to the high-impact mechanism of injury, with potential neurologic dam-age occurring at the time of injury.

Wound complication rates in the present series should be appreciated in the context of the patient cohort; 75% experi-enced their injury as a result of high-energy trauma, many of which were open injuries. Furthermore, a high proportion

Table 3. Postoperative Wound-Related Complications.

Post PrimaryPost Removal of

Hardware

No 129 86% 91 87%Yes 21 14% 14 13% Delayed healing 5 3% 2 2% Superficial infection 8 5% 8 8% Dehiscence 4 3% 1 1% CRPS 2 1% 2 2% Neuroma 0 0% 1 1% Impaired sensation 2 1% 0 0%

Abbreviation: CPRS, complex regional pain syndrome.

579

Tab

le 4

. C

ases

Req

uiri

ng R

eope

ratio

n.

Age

, ySe

xM

echa

nism

Ope

n/C

lose

dC

omor

bidi

ties

Del

ay t

o Su

rger

y, d

Fixa

tion

of T

MT

JsC

losu

reC

ompl

icat

ion

Occ

urre

nce

Day

s

Post

Pri

mar

yT

reat

men

tO

utco

me

(at

Fina

l Fo

llow

-up)

28F

Fall

Clo

sed

Nil

81/

2/3/

4/5

Prim

ary

Deh

isce

nce

Post

Pri

mar

y56

SSG

Hea

lthy

wou

nd32

FM

VA

Ope

nN

il10

1/2/

4/5

Flap

Deh

isce

nce

+ c

hron

ic

infe

ctio

n (M

RSA

, Ps

eudo

mon

as)

Post

Pri

mar

y17

1SS

GH

ealth

y w

ound

36M

MV

AO

pen

Nil

141/

2/3/

4/5

Prim

ary

Deh

isce

nce

Post

Pri

mar

y62

SSG

Hea

lthy

wou

nd29

MSp

ort

Clo

sed

Smok

ing

121/

2Pr

imar

yD

ehis

cenc

e fo

llow

ing

infe

cted

hem

atom

a (M

RSA

+

Sten

otro

phom

onas

)

Post

Pri

mar

y29

Flap

Hea

lthy

wou

nd

earl

y po

stop

(t

hen

LTF)

34M

MV

AC

lose

dN

il19

1/2/

3Pr

imar

yD

ehis

cenc

ePo

st R

OH

291

SSG

Hea

lthy

wou

nd

Abb

revi

atio

ns: L

TF,

lost

to

follo

w-u

p; M

RSA

, met

hici

llin-

resi

stan

t St

aphy

loco

ccus

aur

eus;

MV

A, m

otor

veh

icle

acc

iden

t; R

OH

, rem

oval

of h

ardw

are;

SSG

, spl

it-th

ickn

ess

skin

gra

ft; T

MT

Js,

tars

omet

atar

sal j

oint

s.

580 Foot & Ankle International 39(5)

had medical comorbidities and that have been associated with higher complication rates.3,10,18,27,35,36 These factors, as well as poorly standardized reporting of wound complica-tions in the literature, limit the extent of comparison to pre-vious studies.

Patient, injury, and operative factors are likely to impact wound complication rates. Patient factors cited to increase orthopedic foot and ankle wound complication rates are advanced age,36 smoking,15,35,36 peripheral vascular dis-ease,35,36 and prolonged tourniquet time.36 The mechanism

of injury as well as severity of skeletal and soft tissue injury are also likely to impact postoperative wound healing. As early as 1954, Posch and Weller et al identified that force and contact were more predictive of morbidity than appear-ance following crush injuries.24 Schaser et al identified fail-ure of microvascular perfusion, endothelial leakage, and enhanced leukocyte-endothelium activation following closed crush injuries in an animal model.28 MBA injuries experienced 10.6 times greater odds of superficial infec-tions than their MVA counterparts when controlled for open

Table 5. Predictors of Complication Following Primary Operation.

n

Delayed Wound Healing

Superficial Infection Dehiscence CRPS Neuroma

Impaired Sensation

Total (n)

Total (%) OR P Value

Age, y <40 85 2 4 4 1 0 1 12 14 1 ≥40 65 3 4 0 1 0 1 9 14 0.98 .96Gender Male 110 4 6 2 1 0 2 15 14 1 Female 40 1 2 2 1 0 0 6 15 1.06 .92Mechanism MVA 59 2 1 2 0 0 1 6 10 1 Fall 27 0 1 0 1 0 0 2 7 0.71 .68 Crush 11 2 0 0 1 0 0 3 27 3.31 .14 Sport 7 0 0 1 0 0 0 1 14 1.47 .74 Bicycle accident 8 1 0 0 0 0 0 1 13 1.26 .84 MBA 29 0 6 0 0 0 1 7 24 2.81 .09 Other 9 0 0 1 0 0 0 1 11 1.1 .93Mechanism energy level Low 40 0 1 2 1 0 0 4 10 1 High 110 5 7 2 1 0 2 17 15 1.65 .4Open/closed Closed 126 4 6 2 1 0 2 15 12 1 Open 24 1 2 2 1 0 0 6 25 2.47 .1Comorbidity None 108 5 5 3 0 0 2 15 14 1 Mental illness/drug

abuse17 0 1 0 1 0 0 2 12 0.83 .81

Smoking 22 0 2 1 1 0 0 4 18 1.38 .6 Systemic medical

condition17 0 1 0 0 0 0 1 6 0.39 .37

Joints ORIF 1 TMTJ 12 0 1 0 0 0 0 1 8 0.49 .51 1/2 TMTJ 13 0 0 2 0 0 0 2 15 0.98 .98 1/2/3 TMTJ 102 5 5 2 2 0 2 16 16 1 2/3 TMTJ 15 0 1 0 0 0 0 1 7 0.38 .37 Other 8 0 1 0 0 0 0 1 13 0.77 .81Soft tissue reconstruction None 145 5 8 3 1 0 2 19 13 1 Yes 5 0 0 1 1 0 0 2 40 4.42 .12

Abbreviations: CRPS, complex regional pain syndrome; MBA, motorbike accident; MVA, motor vehicle accident; OR, odds ratio; ORIF, open reduction internal fixation; TMTJs, tarsometatarsal joints.

Philpott et al 581

injuries. We attribute this to the degree of force applied to the foot at time of injury and its considerable effect upon soft tissues, regardless of whether the soft tissue envelope is broken. We anticipated more patients with more extensive skeletal injuries to experience greater wound-related com-plication rates owing to higher force mechanisms, more extensive fixation, and longer tourniquet times and their associated ischemia,37 elevated proteolytic enzyme activ-ity,11 and potential reperfusion injury.4,21 Lau et al recently demonstrated impaired functional outcomes in patients with injury to all 3 columns of the midfoot in comparison with less severe injuries.16 The present series was inadequately

powered to confirm these associations, and further research is needed to help to confirm influential factors.

The present study has several limitations. Twenty-nine patients were operated via alternate approaches and were excluded, and this is a potential source of selection bias. We relied on reporting of complications during outpatient clinic visits, and this reporting may be somewhat subjec-tive, and may have introduced a collection bias. Tourniquet time was not routinely recorded and was therefore was excluded from analysis. And finally, owing to the limited incidence of complications in the cohort, we were unable to demonstrate statistically significant associations

Table 6. Predictors of Complication Following ROH Operation.

n

Delayed Wound Healing

Superficial Infection Dehiscence CRPS Neuroma

Impaired Sensation

Total (n)

Total (%) OR P Value

Age, y <40 59 1 1 1 1 1 0 5 8 1.00 ≥40 46 1 7 0 1 0 0 9 20 2.63 .11Gender Male 77 2 5 1 1 1 0 10 13 1.00 Female 28 0 3 0 1 0 0 4 14 1.12 .86Mechanism MVA 37 0 3 1 1 0 0 5 14 1.00 Fall 21 0 3 0 0 0 0 3 14 1.07 .93 Crush 6 1 0 0 0 0 0 1 17 1.28 .84 Sport 7 0 0 0 0 1 0 1 14 1.07 .96 Bicycle accident 5 0 0 0 0 0 0 0 0 0.54 .69 MBA 23 1 2 0 1 0 0 4 17 1.35 .68 Other 6 0 0 0 0 0 0 0 0 0.45 .61Mechanism energy level Low 30 0 2 0 0 1 0 3 10 1.00 High 75 2 6 1 2 0 0 11 15 1.55 .53Open/closed Closed 92 1 8 1 1 1 0 12 13 1.00 Open 13 1 0 0 1 0 0 2 15 1.21 .82Comorbidity None 77 0 3 1 1 1 0 6 8 1.00 Mental illness/drug abuse 10 0 2 0 1 0 0 3 30 5.07 .045 Smoking 17 2 3 0 0 0 0 5 29 4.93 .02 Systemic medical condition 11 0 3 0 0 0 0 3 27 4.44 .06Joints ORIF 1 TMTJ 8 0 0 0 0 0 0 0 0 0.30 .42 1/2 TMTJ 8 0 0 0 1 0 0 1 13 0.75 .80 1/2/3 TMTJ 75 2 7 1 1 1 0 12 16 1.00 2/3 TMTJ 9 0 0 0 0 0 0 0 0 0.27 .37 Other 5 0 1 0 0 0 0 1 20 1.31 .81Soft tissue reconstruction None 101 1 8 1 2 1 0 13 13 1.00 Yes 4 1 0 0 0 0 0 1 25 2.26 .49

Abbreviations: CRPS, complex regional pain syndrome; MBA, motorbike accident; MVA, motor vehicle accident; OR, odds ratio; ORIF, open reduction internal fixation; TMTJs, tarsometatarsal joints.

582 Foot & Ankle International 39(5)

between patient, injury, and operative factors and postop-erative wound-related complications. Nonetheless, this Lisfranc cohort is one of the largest in the literature and represents a series of consecutive 150 Lisfranc injuries managed via a novel approach.

Conclusion

We have described a modified dorsal approach using win-dows to the midfoot, which can be used to safely achieve reduction and fixation of Lisfranc injuries. We feel this

Table 7. Review of Reported Wound Complication Rates in Existing Studies.

Study Design Cohort n Approach FixationMean

Follow-up Complications (Treatment)

Vertullo et al 200233

Prospective cohort

Elective and traumatic Lisfranc injuries

12 Transverse approach ± medial T

Arthrodesis vs ORIF

12 mo Wound necrosis 1/12Impaired sensation 3/12

Kupcha et al 201513

Prospective + retrospective cohort

Traumatic Lisfranc injury

60 Transverse approach ± medial T

Arthrodesis vs ORIF

9 mo Delayed wound healing 5/60 (3 antibiotics, 1 debridement, 1 skin graft)

Impaired sensation 5/60Arntz et al

19882Prospective

cohortLisfranc injury 34 Multiple longitudinal

incisions (first webspace, third webspace)

Screw 3.4 y No wound-related complications described

Mann et al 199619

Cohort Osteoarthritic / rheumatoid / Lisfranc midfoot injuries

41 Multiple longitudinal incisions (dorsal to midline, second webspace, fourth webspace)

Screw vs plate

70 mo Skin slough 2/41 (1 wound care, 1 debridement)

Neuroma 3/41

Del Vecchio et al 20166

Prospective cohort

Medial column Lisfranc energy from low-energy trauma

5 Dorsomedial minimally invasive

Plate + Screw

19 mo Did not report wound-related complications

Coetzee et al 20075

Prospective randomized control trial

Lisfranc injury (ligamentous)

41 Multiple longitudinal incisions (first webspace, fourth metatarsal)

Arthrodesis vs ORIF

42 mo Did not report wound-related complications

Henning et al 20097

Prospective randomized control trial

Lisfranc injury 32 Longitudinal (first webspace ± fourth metatarsal)

Arthrodesis vs ORIF

24 mo Cellulitis (antibiotics) 1/32Wound breakdown 0/32Neuroma 0/32

Mulier et al 200222

Retrospective cohort

Lisfranc injury 18 Multiple longitudinal incisions (first webspace, fourth webspace)

Arthrodesis vs ORIF

30 mo Sympathetic dystrophy 5/18

Abassian et al 20151

Retrospective case-control

Lisfranc injury (ligamentous and osseous)

58 Longitudinal incision first webspace + stab incision over third metatarsal

ORIF 8 y No wound-related complications described

Li et al 201517 Retrospective cohort

Lisfranc injury 10 Longitudinal incision first webspace

ORIF (miniplate)

20 mo Infection 0/10

Schepers et al 201329

Retrospective case series

Lisfranc injury 28 Longitudinal incision first webspace

ORIF vs Kirschner wire

>3 mo Infection 2/28

Hu et al 20149 Prospective cohort

Lisfranc injury 60 Not stated Plate vs screw

31 mo Infection 8% Delayed healing 12%

Kuo et al 200012 Retrospective cohort

Lisfranc injury (ligamentous and osseous)

48 Longitudinal incision first webspace ± lateral incision

Screw 52 mo Neuroma/CRPS/infection 0/48

Abbreviations: CRPS, complex regional pain syndrome; MBA, motorbike accident; MVA, motor vehicle accident; OR, odds ratio; ORIF, open reduction internal fixation; TMTJs, tarsometatarsal joints.

Philpott et al 583

approach offers superior exposure to the TMTJ with compa-rable rates of complication to existing approaches.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. ICMJE forms for all authors are available online.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Andrew Philpott, MBBS, BMSci, http://orcid.org/0000-0001 -9914-1011

Supplemental Material

Supplementary video is available online with this article.

References

1. Abbasian MR, Paradies F, Weber M, Krause F. Temporary internal fixation for ligamentous and osseous lisfranc injuries: outcome and technical tip. Foot Ankle Int. 2015;36(8):976-983.

2. Arntz CT, Veith RG, Hansen ST Jr. Fractures and fracture-dislocations of the tarsometatarsal joint. J Bone Joint Surg Am. 1988;70(2):173-181.

3. Chen S, Anderson MV, Cheng WK, Wongworawat MD. Diabetes associated with increased surgical site infections in spinal arthrodesis. Clin Orthop Relat Res. 2009;467(7):1670-1673.

4. Cheng YJ, Chien CT, Chen CF. Oxidative stress in bilateral total knee replacement, under ischaemic tourniquet. J Bone Joint Surg Br. 2003;85(5):679-682.

5. Coetzee JC, Ly TV. Treatment of primarily ligamentous Lisfranc joint injuries: primary arthrodesis compared with open reduction and internal fixation. Surgical technique. J Bone Joint Surg Am. 2007;89(suppl 2 pt 1):122-127.

6. Del Vecchio JJ, Ghioldi M, Raimondi N, De Elias M. Minimally invasive medial plating of low-energy lisfranc injuries: preliminary experience with five cases. Adv Orthop. 2016;2016:4861260.

7. Henning JA, Jones CB, Sietsema DL, Bohay DR, Anderson JG. Open reduction internal fixation versus primary arthrode-sis for lisfranc injuries: a prospective randomized study. Foot Ankle Int. 2009;30(10):913-922.

8. Horan TC, Gaynes RP, Martone WJ, Harvis WR, Emori TG. CDC definitions of nonsocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infec-tions. Infect Control Hosp Epidemiol. 1992;13(10):606-608.

9. Hu SJ, Chang SM, Li XH, Yu GR. Outcome comparison of Lisfranc injuries treated through dorsal plate fixation versus screw fixation. Acta Ortop Bras. 2014;22(6):315-320.

10. Jain RK, Shukla R, Singh P, Kumar R. Epidemiology and risk factors for surgical site infections in patients requir-ing orthopedic surgery. Eur J Orthop Surg Traumatol. 2015;25(2):251-254.

11. Jawhar A, Hermanns S, Ponelies N, Obertacke U, Roehl H. Tourniquet-induced ischaemia during total knee arthroplasty results in higher proteolytic activities within vastus medialis cells: a randomized clinical trial. Knee Surg Sports Traumatol Arthrosc. 2016;24(10):3313-3321.

12. Kuo RS, Tejwani NC, Digiovanni CW, et al. Outcome after open reduction and internal fixation of Lisfranc joint injuries. J Bone Joint Surg Am. 2000;82-A(11):1609-1618.

13. Kupcha P, Gradisek B. Transverse dorsal approach to the midfoot joint in acute traumatic injury. Tech Foot Ankle Surg. 2015;14(4):164-170.

14. Kura H, Luo ZP, Kitaoka HB, Smutz WP, An KN. Mechanical behavior of the Lisfranc and dorsal cuneometatarsal liga-ments: in vitro biomechanical study. J Orthop Trauma. 2001;15(2):107-110.

15. Lampley A, Gross CE, Green CL, et al. Association of ciga-rette use and complication rates and outcomes following total ankle arthroplasty. Foot Ankle Int. 2016;37(10):1052-1059.

16. Lau SC, Guest C, Hall M, et al. Do columns or sagittal dis-placement matter in the assessment and management of Lisfranc fracture dislocation? An alternate approach to classi-fication of the Lisfranc injury. Injury. 2017;48(7):1689-1695.

17. Li B, Zhao W, Liu L, et al. Efficacy of open reduction and internal fixation with a miniplate and hollow screw in the treatment of Lisfranc injury. Chin J Traumatol. 2015; 18(1):18-20.

18. Li GQ, Guo FF, Ou Y, Dong GW, Zhou W. Epidemiology and outcomes of surgical site infections following orthopedic surgery. Am J Infect Control. 2013;41(12):1268-1271.

19. Mann RA, Prieskorn D, Sobel M. Mid-tarsal and tarso-metatarsal arthrodesis for primary degenerative osteoarthro-sis or osteoarthrosis after trauma. J Bone Joint Surg Am. 1996;78(9):1376-1385.

20. Mantas JP, Burks RT. Lisfranc injuries in the athlete. Clin Sports Med. 1994;13(4):719-730.

21. Mathru M, Dries DJ, Barnes L, et al. Tourniquet-induced exsanguination in patients requiring lower limb surgery. An ischemia-reperfusion model of oxidant and antioxidant metabolism. Anesthesiology. 1996;84(1):14-22.

22. Mulier T, Reynders P, Dereymaeker G, Broos P. Severe Lisfrancs injuries: primary arthrodesis or ORIF? Foot Ankle Int. 2002;23(10):902-905.

23. Myerson MS, McGarvey WC, Henderson MR, Hakim J. Morbidity after crush injuries to the foot. J Orthop Trauma. 1994;8(4):343-349.

24. Posch JL, Weller CN. Mangle and severe wringer injuries of the hand in children. J Bone Joint Surg Am. 1954;36-A(1):57-63; passim.

25. Qu W, Ni S, Wang Z, et al. Severe open Lisfranc injuries: one-stage operation through internal fixation associated with vacuum sealing drainage. J Orthop Surg Res. 2016;11(1):134.

26. Ranade AV, Rajanigandha V, Rai R, Ebenezer DA. Relationship between the deep peroneal nerve and dor-salis pedis artery in the foot: a cadaveric study. Clin Anat. 2008;21(7):705-712.

27. Richards JE, Kauffmann RM, Zuckerman SL, Obremskey WT, May AK. Relationship of hyperglycemia and surgical-site infection in orthopaedic surgery. J Bone Joint Surg Am. 2012;94(13):1181-1186.

584 Foot & Ankle International 39(5)

28. Schaser KD, Vollmar B, Menger MD, et al. In vivo analy-sis of microcirculation following closed soft-tissue injury. J Orthop Res. 1999;17(5):678-685.

29. Schepers T, Oprel PP, Van Lieshout EM. Influence of approach and implant on reduction accuracy and stability in lisfranc fracture-dislocation at the tarsometatarsal joint. Foot Ankle Int. 2013;34(5):705-710.

30. Sink EL, Leunig M, Zaltz I, et al. Reliability of a complication classification system for orthopaedic surgery. Clin Orthop Relat Res. 2012;470(8):2220-2226.

31. Stavlas P, Roberts CS, Xypnitos FN, Giannoudis PV. The role of reduction and internal fixation of Lisfranc fracture-dislocations: a systematic review of the literature. Int Orthop. 2010;34(8):1083-1091.

32. van Koperen PJ, de Jong VM, Luitse JS, Schepers T. Functional outcomes after temporary bridging with locking plates in Lisfranc injuries. J Foot Ankle Surg. 2016;55(5): 922-926.

33. Vertullo CJ, Easley ME, Nunley JA. The transverse dorsal approach to the Lisfranc joint. Foot Ankle Int. 2002;23(5):420-426.

34. Vijayalakshmi S, Raghunath G., Shenoy V. Anatomical study of dorsalis pedis artery and its clinical correlations. J Clin Diagn Res. 2011;5(2):287-290.

35. Whalen JL, Spelsberg SC, Murray P. Wound breakdown after total ankle arthroplasty. Foot Ankle Int. 2010;31(4):301-305.

36. Wiewiorski M, Barg A, Hoerterer H, et al. Risk factors for wound complications in patients after elective orthopedic foot and ankle surgery. Foot Ankle Int. 2015;36(5):479-487.

37. Wilgis EF. Observations on the effects of tourniquet isch-emia. J Bone Joint Surg Am. 1971;53(7):1343-1346.

38. Zacharias J, Largen PS, Crosby LA. Results of preprocedure and postprocedure toe cultures in orthopaedic surgery. Foot Ankle Int. 1998;19(3):166-168.

39. Zgonis T, Jolly GP, Garbalosa JC. The efficacy of prophylac-tic intravenous antibiotics in elective foot and ankle surgery. J Foot Ankle Surg. 2004;43(2):97-103.