Injury, Int. J.

Care Injured 47S6 (2016) S53–S61

Contents lists available at ScienceDirect

Injury
j o u r n a l h o m e p a g e : w w w. e l s e v i e r . c o m / l o c at e / I n j u r y

Restoration of long bone defects treated with the induced membrane technique:
protocol and outcomes
Peter V. Giannoudisa,b, *, Paul J. Harwooda, Theodoros Tosounidisa,b, Nikolaos K. Kanakarisa
a
Academic Department of Trauma and Orthopaedics, Leeds Teaching Hospitals, School of Medicine, University of Leeds, Leeds, UK
b
NIHR Leeds Biomedical Research Unit, Chapel Allerton Hospital, Leeds, UK

K E Y W O R D S A B S T R A C T

Fracture This prospective study was undertaken at a regional tertiary referral centre to evaluate the results of treatment of
bone defect bone defects managed with the induced membrane (IM) technique. Inclusion criteria were patients with bone
induced membrane defects secondary to septic non-union, chronic osteomyelitis and acute fracture with bone loss. Pathological
Masquelet fractures with bone loss were excluded. Data collection included patient demographics, pathology, previous
surgical intervention, size of bone defect, type of graft implanted, time-to-union and complications/re-
interventions. The minimum time of follow up was 12 months. Forty-three patients (32 males) met the inclusion
criteria with a mean age of 47.9 years (range 18–80 years). 22 patients had an acute traumatic bone loss associated
with open fracture and 21 presented with an infected non-union or underlying osteomyelitis requiring bone
excision. The most common microorganisms grown were staphylcoccous aureus and coagulase negative
staphylococcous. The mean length of the bone defect area was 4.2 cm (range 2–12 cm). All patients were managed
with the two stage technique receiving composited grafting (Autologous bone graft (Iliac crest/RIA), graft
expander as required, osteoprogenitor cells, growth factor) during the second stage. There was one failure
(humeral infected non-union) in a previous background of bone radiation that necessitated reconstruction with a
free fibula vascularized graft. One patient had a fall and sustained implant failure (humeral defect) 3 months after
reconstruction and following re-plating progressed to union 4 months later. Two patients required re-grafting due
to failure of healing in one of the defect sides. One patient presented with a discharging sinus 2 years after
successful healing of a tibial defect that was treated successfully with soft tissue and bone debridement without
necessitating further interventions. One patient despite union (distal 1/3 tibia) underwent a below knee
amputation due to a dysfunctional ankle/foot ( previous foot compartment syndrome- regional pain syndrome).
Of those patients, with lower limb injuries, 4 patients had leg length discrepancies of 1 cm, 1.5 cm, 2 cm (two
patients) respectively. The mean time to radiological union was 5.4 months (range 2–12 months). The average
time of healing of 1 cm bone defect was 1.24 months. Patients with upper limb reconstruction recovered earlier
than those with lower limb injuries. At the latest follow up all patients were able to mobilize full weight bearing
without residual pain.
The induced membrane technique appears to be an alternative good option for the management of large bone
defects secondary to acute bone loss or infected non-unions. The incidence of re-interventions was low in this
challenging cohort of patients. The technique should be considered in the surgeon’s armamentarium as it is
effective and is associated with a low rate of complications.
© 2016 Elsevier Ltd. All rights reserved.

Introduction continues to be challenging despite the latest advances in surgical

Critical size bone defects are defined as defects exceeding 2–2.5
times the diameter of the affected bone [1,2]. Their treatment
* Corresponding author at: Peter V Giannoudis BSc, MD, FRCS, FACS, Professor and
Chairman, Academic Department of Trauma and Orthopaedics, School of Medicine, University
of Leeds, Leeds General Infirmary, Clarendon wing, Level D, LS13EX, Leeds, West Yorkshire,
United Kingdom. Tel: 0044-113-3922750; Fax: 0044-113-3923290
E-mail address: p.giannoudi@aol.com (Peter V Giannoudis).
techniques and tissue regeneration procedures [3]. Their incidence is
variable ranging between 0.4% and 11.4% [4].
Treatment modalities include such techniques as bone transport
(distraction osteogenesis), vascularised bone free transfer, massive
cancellous autograft or allograft, titanium cages, replacement of the
defect with megaprosthesis and the last resort being amputation of
the affected extremity [5–14]. Each one of the above techniques has
limitations and the results of treatment are variable in regards to each
technique’s effectiveness.

0020-1383 / © 2016 Elsevier Ltd. All rights reserved.

S54 P. V. Giannoudis et al. / Injury, Int. J. Care Injured 47S6 (2016) S53–S61
During the last 2 decades, the induced membrane (Masquelet)
technique was introduced as a complementary option for the
management of bone defects [15]. The technique consists of two
stages. During the first stage a PMMA cement is implanted at the bone
defect area and acts not only as a spacer to maintain the length of the
affected limb but also facilitates the production of a pseudo membrane
which has been shown to contain osteoprogenitor cells and has the
ability to release inductive molecules (growth factors) capable of
stimulating osteogenesis [16,17]. At the second stage, 6–8 weeks later,
after incision of the membrane, the cement spacer is removed and
autologous bone grafting is implanted in the defect area facilitating a
positive osseous healing response. Healing of defects as lengthy as
25 cm have been treated effectively with this technique [18].
The aim of the current study is to present our institutional
experience of applying this technique in a series of patients that
were managed in our institution with bone defects. We wished to
determine the incidence of success rate (osseous healing) and the
number of complications encountered.
Patients and methods
Between January 2008 and December 2014, all consecutive patients
who presented in our unit with bone defects either as a result of
traumatic injury or developed bone loss as a result of infection
requiring radical bone debridement were eligible to participate. Main
inclusion criteria were: acute fracture with bone loss, septic nonunion
and chronic osteomyelitis associated with bone loss secondary to
radical bone debridement. Patients with pathological fracture resulting
in bone loss or patients treated with another surgical technique (i.e.
bone transport) were excluded.
Prospective data documented included patient demographics,
mechanism of injury, type of injury and presence of associated
injuries; open or closed injury, anatomical region involved, type of
surgery, time elapsed between the 2 stages of the technique, graft
material implanted, type of pathogen isolated in the cases where
infection was the causative factor, complications and mortality. All
patients were managed by fellowship trained consultant grade
surgeons according the protocol designed by the senior author.
Assessment and management
Patients were assessed and managed as the previously developed
algorithm by the senior author (Figure 1). This was designed with
an aim to standardize the assessment and treatment of patients.
The surgical management consisted of 2 stages as previously des-
cribed [18]. In summary, stage 1 surgery consisted of the following
standardized surgical steps in each case:
1. Debridement of the soft tissue in case of open fracture (acute
setting) or sinus secondary to chronic infection, debridement of
the devascularised or infected bony segment.
2. Insertion of a PMMA cement spacer (PALACOS® Bone Cements
mixed with 2grams of Vancomycin)
3. Reconstruction of the soft tissue envelope as needed (rotational
muscle flap, free muscle flap, skin grafting)
4. Temporary (external fixator) or permanent skeletal stabilization
(open fractures) depending on the skeletal and local soft tissue
conditions.
5. Prescription of appropriate pathogen specific antibiotics when
necessary (infected cases) for a period of minimum of 6 weeks as
guided by the local microbiology guidelines.
Stage one lasted between 6 and 8 weeks, (average 7 weeks). Stage 2
consisted of the following standardized steps:
1. Autologous bone graft was harvested from intramedullary cavity
of femur/iliac crest as indicated, pending on the volume of graft
needed. Concentrated bone marrow aspirate to increase the
cellularity of the graft material was aspirated from the iliac crest.
Bone harvesting from the femoral canal was carried out using the
RIA device – Reamer Irrigator Aspirator (Depuy- Synthes, West
Chester, PA, USA) either from the contralateral and/or ipsilateral
femur as it was indicated. In cases that the volume of graft was not
adequate, Orthos scaffold (Geistlich, Wolhusen, Switzerland) was
used as a graft expander.
2. Removal of external fixator – pin site irrigation /debridement.
3. Cement spacer removal following incision and preservation of the
membrane formed.
4. Further inspection of the bone defect area. Tissue biopsies for
microbiology. If suspicion that any bone area is non-viable, tissue
removed.
5. Definitive surgical stabilization of the fracture with plate fixation
or intramedullary nail as indicated.
6. Implantation in the defect area of the composite graft (autologous
bone graft, BMP-7 (Olympus Biotek) and BMA (bone marrow
aspirate- concentrated osteoprogenitor cells)
7. Closure of the membrane, creating a closed compartment.
8. Wound closure without using a surgical drain.
Post-operatively all patients except those with upper limb surgery
received thromboprophylaxis (low molecular weight heparin subcuta-
neously (Tinzaparin 4.500 IU)) for 6 weeks. Patients with lower limb
injuries were mobilized toe touch weight bearing initially using either
a zimmer frame or crutches for 6–8 weeks and thereafter progressed
from partial to full weight bearing. Outpatient follow-up with both
clinical and radiographic assessment was carried out at 2 weeks for
wound inspection and at 6 weeks, followed by a 3, 4, 5, 6, 8, 12 months
or until radiological signs of union and pain free mobilization.
Radiologically union was defined when 3 out of 4 cortices showed
callus formation (bridging).
Broad spectrum empirical antibiotic therapy (Augmentin 1.2 g) was
given after obtaining tissue samples at the time of debridement. The
empirical antimicrobial regimen was either continued or modified
according to the culture results and local microbiology guidance. The
duration and the route of administration of the antibiotics were
dependent on the patients systemic and local wound response.
All infected patients were treated systemically with a course of
antibiotics based on the local microbiology tissue sensitivities for a
minimum period of 6 weeks and were discontinued only after the
haematological biomarkers were normalised.
This study was approved by the institutional review board. The
minimum follow-up period was 12 months.
Results
Forty-three patients (32 males) with bone loss after debridement
of a septic non-union or after an acute fracture were eligible to
participate during the specified period of the study. The mean age was
47.9 years (range 18–80 years). Patient’s characteristics are shown in
Table 1.
Overall, 22 patients had an acute traumatic bone loss associated
with open fracture. Out of the twenty-one patients with infected non-
union, 11 had a previous open fracture. One patient developed
infection as a result of wound contamination following leaking of the
ileostomy on to the hip wound after gamma nail fixation of the
subtrochanteric femur fracture. A patient with Lisfranc injury had a
successful open reduction internal fixation but subsequently devel-
oped wound infection and deep sepsis requiring removal of metal
work and bone debridement leading to a bone defect of the medial
cuneiform. The most common microorganisms grown were staphyl-
coccous aureus and coagulase negative staphylococcous and all
patients had 6–8 weeks of systemic antibiotic except of one patient
who continued for 3 months, Table 2. The mean length of the bone
 P. V. Giannoudis et al. / Injury, Int. J. Care Injured 47S6 (2016) S53–S61 S55

Bone defect Aetiology

Traumatic Infective

pre operative assessment

clinical Investigations

History, clinical examination of the limb: AP and Lateral radiographs of the

Assess: neurovascular status of the limb, involved bone with a joint above and
state of surrounding soft tissue. below, full length films and weight
bearing x rays if possible of both limbs
Any signs of infection?
CT scan/ MRI or bone scan according to
Previous surgery? the cause of bone defect.
Range of movements? Bio chemical investigations (FBC,
Limb length discrepancies? Electrolytes, CRP, ESR, bone chemistry,
Vitamin D).
Patient Comorbidities?

Diagnosis; Is the limb salvageable? Discuss with patient and family

Options of treatment available? Decision to proceed with IM technique

Patient’s expectations? Multidisciplinary group meeting

-Plan/proceed 1 stage (debride the infected bone or devascularised bone to

healthy/vascular bone, cement spacer, fracture stabilization (ex-fix, (nail/plate acute cases)
soft tissue cover, microbiology, antibiotics).

-Plan/proceed with 2nd stage after 6-8 weeks (remove spacer, send tissues to microbiology,
implant bone graft and definitive fracture stabilization in previous infected cases, antibiotic
treatment if infection is grown from tissues). If during 2nd stage evidence/suspicion of
infection, re-debride go back to stage one again.

Post op management

Antibiotics as indicated, protected weight bearing, physiotherapy, thromboprophylaxis,

regular clinic follow up for both radiological and clinical assessment.

Fig. 1. Proposed algorithm for management of bone defect treatment.

defect area was 4.2 cm (range 2–12 cm). There was one failure
(humeral infected non-union) in a previous background of bone
radiation that necessitated reconstruction with a free fibula vascular-
ized graft. One patient had a fall and sustained implant failure
(humeral defect) 3 months after reconstruction and following re-
plating progressed to union 4 months later. Two patients required re-
grafting due to failure of healing in one of the defect sides. One patient
presented with a discharging sinus 2 years after successful healing of a
tibial defect that was treated successfully with soft tissue and bone
debridement without necessitating further interventions. One patient
despite union (distal 1/3 tibia) underwent a below knee amputation
due to a dysfunctional ankle/foot ( previous foot compartment
syndrome- regional pain syndrome). Of those patients, with lower
limb injuries, 4 patients had leg length discrepancies of 1 cm, 1.5 cm,
2 cm (two patients) respectively. The mean time to radiological union
was 5.4 months (range 2–12 months), (Figures 2–4). Two patients
required free flap cover for soft tissue defect while another patient
required split skin graft for successful wound closure before he
developed infected no union and successful management. There were
no reported cases of thromboembolism.
The average time of healing of 1 cm bone defect was 1.24 months.
Patients with upper limb reconstruction recovered earlier than those
with lower limb injuries. At the latest follow up all patients were able to
mobilize full weight bearing without residual pain.
S56 P. V. Giannoudis et al. / Injury, Int. J. Care Injured 47S6 (2016) S53–S61

Table 1.
Patient’s characteristics-defect size- graft material used- complications-time to union

Traumatic Time to
Anatomical or Size of union
Patients Gender Age region Fixation Method Infective defect Complications Graft material used (months)

1 M 73 Forearm Plate T 5 cm RIA + BMP-7 + BMA 6

2 F 28 Femur IM nail T 9 cm Regrafting distal non-union RIA + BMP-7 + BMA 12
site due to slow
mineralization/osseous
healing
3 M 37 Forearm Plate T 2 cm ICBG + Cells + BMP-7 3
4 M 71 Forearm Plate T 4 cm RIA + BMP-7 + BMA 5
5 M 70 Forearm Plate T 2 cm ICBG 2
6 M 53 MT Plate T 5 cm RIA + BMP-7 + BMA + orthos 6
7 M 44 Femur IM nail T 7 cm RIA + BMP-7 + BMA + orthos 9
8 M 66 Radius Plate I 4.8 cm RIA + BMP-7+BMA 4
9 M 18 Radius Plate I 5 cm RIA + BMP-7 + BMA 5
10 M 29 Femur Plate I 12 cm RIA + BMP-7 + BMA + orthos 11
11 M 29 Femur Blade plate I 5 cm RIA + BMP-7 + BMA 6
12 M 80 Femur IM nail I 5 cm RIA + BMP-7 + BMA+orthos 7
13 F 60 Tibia Plate I 5 cm Amputation due to RIA+BMP-7+BMA 8
dysfunctional foot from
original injury
14 F 45 Tibia Plate I 4 cm RIA + BMP-7 + BMA + Orthos 6
15 M 18 Tibia Plate I 5 cm RIA + BMP-7 + BMA + orthos 7
16 M 63 Tibia Plate I 5 cm RIA + BMP-7 + BMA 5
17 M 22 Tibia Plate I 7.5 cm Presented with discharging RIA + BMP-7 + BMA 7
sinus/excision of sinus and
curettage of bone. No
functional problems.
18 M 19 MT* No plate used due I 2 cm ICBG + BMA 3
to previous
infection/only
bone graft
19 F 47 Femur IM nail T 4 cm RIA + BMP-7 + BMA 5
20 M 77 Radius Plate T 2 cm ICBG 2.5
21 M 55 Femur Plate I 4 cm RIA + BMP-7 + BMA 4
22 M 58 Femur IM nail I 4.5 cm RIA + BMP-7 + BMA + orthos 6
23 M 78 Femur IM nail T 3 cm Deceased (unrelated cause) Allograft + BMP-7 + BMA 6.5
24 M 23 Radius Plate I 2.5 cm ICBG 3
25 M 32 Ulna Plate I 2.5 cm RIA + BMP-7 + BMA 2
26 F 57 Femur Plate T 4 cm RIA + BMP-7 + BMA 6.5
27 F 74 Radius Plate T 2.5 cm RIA + BMP-7 + BMA 3
28 M 20 Ulna Plate T 2 cm RIA + BMP-7 + BMA 3.5
29 F 47 Tibia Plate I 3.5 cm RIA + BMP-7 + BMA 4.5
30 M 33 Femur IM nail I 4.5 cm RIA + BMP-7 + BMA 5
31 M 20 Tibia Plate T 3.5 cm RIA + BMP-7 + BMA 4
32 M 36 Femur IM nail T 5.5 cm RIA + BMP-7 + BMA + orthos 11.5
33 M 57 Femur IM nail T 4 cm RIA +BMP-7 + BMA 5.5
34 M 23 Radius Plate T 2.8 cm RIA + BMP-7 + BMA 4
35 M 40 Radius Plate T 2.5 cm ICBG 3.5
36 F 74 Ulna Plate I 2 cm ICBG 3.5
37 M 45 Tibia Tibia T 3.8 cm RIA + BMP-7 + BMA 4.5
38 M 54 Tibia IM nail I 3.6 cm RIA + BMP-7 + BMA 5
39 M 32 tibia Ilizarov T 4 cm RIA + BMA 5
40 F 74 Femur IM nail I 5 cm Regrafting after 5 months RIA + BMP-7 + BMA 10
united 4 months later
41 F 31 Tibia Plate T 4.5 cm RIA + BMP-7 + BMA + orthos 6.5
42 M 19 2nd metatarsal Plate T 2 cm ICBG 2
43 F 57 Humerus Plate I 6.5 cm Background of previous RIA + BMA + orthos Failure
radiation of bone due to
myosarcoma,
Reintervention with free
fibula

T = traumatic, I = infectious, ICBG = iliac crest bone grafting, BMA = bone marrow aspirate, BMP-7 = Bone morphogenetic protein 7, MT = metatarsal.
*Previous Lisfranc injury.
& Iliostomy wound contamination subtrochanteric area.

Discussion
Bone defects managed in the herein study were secondary to acute
bone loss or bone excision (debridement) due to infection post fracture
fixation. Both situations represent challenging scenarios for the
surgeon and the patient. In the acute bone loss group, since most of
the injuries were open fractures, the objective was to use the cement
spacer as a space occupying device to allow resuscitation of the soft
tissues, support the restoration of length and mechanical axis of the
affected extremity and to deliver locally antibiotics, thus reducing the
risk of infection. Soft tissue reconstruction was carried out as it was
necessary to convert an open fracture to a closed one as soon as it was
appropriate. Such strategy provided a “stable” local environment
facilitating the formation of the induced membrane which after the
removal of the cement spacer and the implantation of the bone grafting
contributed to the healing process. Such a strategy proved to be safe
 P. V. Giannoudis et al. / Injury, Int. J. Care Injured 47S6 (2016) S53–S61 S57

Table 2.
Common microorganisms and antibiotic treatment

Cases Age Organisms Antibiotics Duration

1 29 Candida + Coag-ve staph Amoxycillin+flucanozole 12/52 3 months

Diphtheroid+Streptococcus
2 66 Staph aureus + group C strept IV – 2 weeks, Oral 6/52 - flucloxacillin+clindamycin 8 weeks
3 18 coag neg staph Clindamycin-6/52 6 weeks
4 63 Gram -ve bacilli Iv flu -2weeks and oral flucloxacillin- 4 weeks 6 weeks
5 18 coliforms+ gram -ve bacillus +coag -ve satph Ciprofloxacin +flucloxacillin for 6/52w 6 weeks
6 19 Staph aureus Clindamycin 6/52 6 weeks
7 68 pseudomonas aeruginosa Ciprofloxacin-6 weeks 6 weeks
8 29 Staph aureus IV + oral flucloxacillin 6 weeks 6 weeks
9 45 Staph aureus Vancomycin 6/52 6 weeks
Coag neg staph
10 52 Mixed growth Vancomycin+ rifampicin 6 weeks
11 60 Coag –ve staph Vancomycin 6/52 6 weeks
12 MRSA Vancomycin+ rifampicin 12 weeks
13 Staph aureus IV + oral flucloxacillin 6 weeks 6 weeks
14 Staph aureus IV + oral flucloxacillin 6 weeks 6 weeks
15 coag neg staph Clindamycin-6/52 6 weeks
16 Staph aureus Clindamycin 6/52 6 weeks
17 pseudomonas aeruginosa Ciprofloxacin-6 weeks 6 weeks
18 Staph aureus Clindamycin 6/52 6 weeks
19 Mixed growth Vancomycin+ rifampicin 6 weeks
20 MRSA Vancomycin+ rifampicin 12 weeks
21 Staph aureus IV + oral flucloxacillin 6 weeks 6 weeks

supporting the formation of a sterile environment creating optimum
conditions for the subsequent bone grafting of the defect area.
In the group of patients with infected non-union, the objective was
eradication of the infection and the creation of an aseptic environment
along with the formation of the induced membrane. Here, an initial
radical bone debridement is essential for success [19]. Moreover,
vigilance in regards to the state of the bone during the second stage and
the need for further bone debridement is of paramount importance.
Careful inspection of the bone edges to assess their vitality is necessary
to avoid the risk of redevelopment of infection and failure of the
bone grafting procedure. Bone biopsies should be send routinely to
microbiology for the confirmation of a sterile bed. If cultures are proven
positive, then a course of antibiotics is recommended. If there is
suspicion of ongoing infection, then further bone debridement and
re-implantation of a new cement spacer (repetition of stage 1) is
recommended. In the herein case series we did not encounter such
situation.
Initial stabilisation of the affected extremity in the infected non-
union cases was achieved with external fixator. Such a strategy was
found to be safe providing adequate mechanical stability and allowing
daily inspection of the soft tissue envelope. Until the timing of
execution of the second stage of the procedure, a pin site care protocol
is essential to minimise the risk of pin site sepsis. In the acute cases, the
fracture was stabilised with either a nail, a plate or an external fixator
(open fractures) as it was appropriate pending on the location of the
fracture, and state of the soft tissues.
The cement spacer implanted in each case was loaded with 2 g of
vancomycin for local delivery of antibiotics. We are aware that there is a
difference of opinion by other authors whether this is necessary and
whether it should be carried out routinely. Except in one case, we did
not encounter episodes of re-infections to support the view that such
practice could contribute to the development or reoccurrence of
infection.
For the second stage, we routinely planned the procedure within
6–8 weeks following the first stage. We did not confront any issues
with the quality of membrane formation. It has been suggested that
the second stage should be carried out between 4 and 6 weeks as
during this period of time secretion of inductive molecules by the
membrane reaches a peak declining thereafter. In our experience, even
a delay of up to 8 weeks does not appear to have a negative impact
on the quality of the membrane structure and handling. With regard to
the issue of reduced/diminished secretion of growth factors by that
time period, it was not an issue as we implanted routinely BMP-7.
The autologous bone graft material used for the second stage was
most frequently harvested from the femoral intramedullary cavity
using the RIA device. We found this approach useful since a large
volume was harvested in each case. When more graft was necessary
(large defects more that 6 cm), we mixed the RIA graft with the orthos
bovine scaffold as a graft expander. Moreover, we enhanced the
osteogenecity of the graft material with concentrated bone marrow
aspirate. Our strategy for optimum biological enhancement was based
on the “diamond concept” for bone repair [14]. Such an approach we
feel increases the chances of revascularisation of the graft material for
faster integration with the host and assists in the subsequent
mineralisation process [20–22].
Overall, the union rate in this series of patients was 93% (one failure
and two patients required re-grafting procedures). More over the mean
time to union was 5.4 months (2–12 months). The average time of
healing of 1 cm bone defect being 1.24 months is slightly faster
compare to bone transport. This can be attributed to the powerful
biological enhancement delivered during the second stage with the
diamond conceptual framework for bone healing.
Masquelet et al. in his first series of 35 patients reported
amputation, infection and stress fracture of the reconstructed skeleton
as 3 main complications [15]. Pre surgical assessment for the suitability
of the reconstructive surgery is of paramount importance as 2 of
the patients in the above series who required amputation following
reconstructive surgery could have been prevented as they had
associated compromised vascular conditions (thrombosis of the
arterial bypass). Pelissier et al. have also reported a case of amputation
for the similar reasons [23]. The above reported case series have
strengthened the importance of thorough pre surgical assessment
for the selection of suitable patients. The second complication in
Masquelet et al. original series was Infection [15]. He reported
reoccurrence of infection in 5 patients, which necessitated further
surgical debridement for successful management. Herein, we had a
delayed presentation of a discharging sinus reaching down to bone,
two years after a tibial defect was managed successfully with complete
osseous healing. Local soft tissue and bone debridement was adequate
to cure the localised infection. We speculate that this event was
secondary to failure of a small graft area to intergrade with the host’s
bone and remained avascular thus contributing to the development of
S58 P. V. Giannoudis et al. / Injury, Int. J. Care Injured 47S6 (2016) S53–S61

Fig. 2. (a,b) AP and lateral x-ray view of left tibia demonstrating a distal 1/3 closed fracture. (c,d) AP and lateral radiographs after stabilization of the fracture with reamed intra-
medullary nailing. (e-f) Lateral and AP x-ray view of left tibia showing metal work failure with no progress to fracture healing. (g-h) Photographs demonstrating the presence of
a discharging sinus over the non-union site.

infection/sinus formation. Noteworthy, we didn’t experience any
serious untoward incidents of graft resorption as it has been reported
previously [24]. This could be attributed to the success of eradication of
the infection, the appropriate closure of the membrane creating locally
the concept of a “bioreactor” and the appropriate combination of the
graft materials used in terms of their volume and ratio.
There are limitations to the study since we did not routinely
assess functional scores with specific generic instruments. Our
objective was to evaluate the results of treatment and to document
the perioperative and post-operative complications. Other limitations
include the small sample size, the non-availability of a comparable
group or statistical analysis and the lack of randomization process.
However, such a study will be difficult to carry out due to the
complexity and heterogeneity of these cases. Nonetheless, as the
literature regarding the effectiveness of this technique remains poor,
we feel that our series representing the experience of one large center
will contribute further to the better understanding of this technique
for bone regeneration.
 P. V. Giannoudis et al. / Injury, Int. J. Care Injured 47S6 (2016) S53–S61 S59

Fig. 2. (Continued) (i-m) Intraoperative images demonstrating debridement of the sinus, removal of broken nail, and size of bone defect created after the completion of debride-
ment. (n) stabilization of tibia with external fixator and insertion of cement spacer; (o) wound closure; ( p) wound healing 2 weeks later; (q,r) AP lateral x-rays 3 weeks after
cement insertion (s,t) Eight weeks after 1st stage, removal of external fixator prior to initiation of 2nd stage; (x) incision of induced membrane for removal of cement spacer;
(y) RIA graft was harvested from ipsilateral femur and implanted into the defect area along with osteoprogenitor cells and BMP-7; (z) after graft implantation and closure of the
IM closure, a locking plate was used for definitive stabilization.
S60 P. V. Giannoudis et al. / Injury, Int. J. Care Injured 47S6 (2016) S53–S61

Fig. 2. (Continued) (i-v) Left tibia radiographs at 5 months follow up demonstrating osseous healing of the 5 cm defect.

Fig. 3. Anteroposterior (A) and Lateral (B) x-rays of Open grade IIIA left distal tibial fracture (C and D) of a 44 year old male patient. 3D CT (E) and 2D (coronal F, and sagittal G)
reconstruction of the left distal tibia post initial inadequate debridement, wound closure and bridging external fixation to another unit. First stage induced membrane with thor-
ough debridement of the distal metaphysis – 5 by 4 cm defect filled with palacos cement and stabilization with a circular fixator (lateral H view). Second stage induced mem-
brane with removal of cement and insertion of RIA graft with BMAC (anteroposterior J and lateral K views). X-rays (N anteroposterior and O lateral views), post removal of the
frame (6months post second stage). Final follow up views at 12months post second stage (P anteroposterior and Q lateral ankle views).
 P. V. Giannoudis et al. / Injury, Int. J. Care Injured 47S6 (2016) S53–S61
Fig. 4. Forearm fracture of a 72 year old patient (anteroposterior a and lateral b views). Initial fixation failure and infection at 2-months post orif (anteroposterior c and lateral d
views). First stage induced membrane follow up x-rays (anteroposterior f and lateral e views). Second stage induced membrane follow up x-rays (anteroposterior h and lateral g
views) at 4 months post cement removal and grafting with demineralized bone matrix and osteoprogenitor cells.
In conclusion, the induced membrane technique appears to be an
alternative good option for the treatment of large bone defects
secondary to acute bone loss or as a result of chronic infected non-
unions as seen in this series of patients. This procedure should be
considered in the surgeon’s armamentarium as it is effective and is
associated with a low incidence of complications.
Conflict of Interest
No benefits in any form have been received or will be received
from a commercial party related directly or indirectly to the subject of
this article.
Acknowledgements
The authors express their appreciation to Mr Suri Gudipati for assisting
in the collection of the data of some of the cases.
References
[1] Gugala Z, Gogolewski S. Regeneration of segmental diaphyseal defects in sheep tibiae
using resorbable polymeric membranes: a preliminary study. J Orthop Trauma
1999;13:187–95.
[2] Key AJ. The effect of a local calcium depot on osteogenesis and healing of fractures. J Bone
Joint Surg 1934;16:176–184.
[3] Dimitriou R, Jones E, McGonagle D, Giannoudis PV. Bone regeneration: Current concepts
and future directions. BMC Medicine 2011;9:66.
[4] Keating JF, Simpson AH, Robinson CM. The management of fractures with bone loss.
J Bone Joint Surg Br 2005;87:142–50.
[5] Calori GM, Colombo M, Malagoli E, Mazzola S, Bucci M, Mazza E. Megaprosthesis in post-
traumatic and periprosthetic large bone defects: Issues to consider. Injury 2014;45(Suppl
6):S105–10.
[6] Giannoudis PV, Calori GM, Bégué T, Schmidmaier G. Tissue loss and bone repair: Time to
develop an international strategy? Injury 2015;46(Suppl 8):S1–2.
[7] Giannoudis PV. Treatment of bone defects: Bone transport or the induced membrane
technique? Injury 2016;47:291–2.
[8] Shanmuganathan R, Chandra Mohan AK, Agraharam D, Perumal R, Jayaramaraju D,
Kulkarni S. Successful reimplantation of extruded long bone segments in open fractures
of lower limb—a report of 3 cases. Injury 2015;46(7):1389–92.
S61
[9] Lindsey RW, et al. The efficacy of cylindrical titanium mesh cage for the reconstruction
of a critical-size canine segmental femoral diaphyseal defect. J Orthop Res 2006;24:
1438–53.
[10] Campanacci DA, Puccini S, Caff G, Beltrami G, Piccioli A, Innocenti M, et al. Vascularised
fibular grafts as a salvage procedure in failed intercalary reconstructions after bone
tumour resection of the femur. Injury 2014;45:399–404.
[11] DeCoster T, Gehlert R, Mikola E, Pirela-Cruz M. Management of posttraumatic segmental
bone defects. J Am Acad Orthop Surg 2004;12:28–38.
[12] Chmell MJ, McAndrew MP, Thomas R, Schwartz HS. Structural allografts for reconstruc-
tion of lower extremity open fractures with 10 centimeters or more of acute segmental
defects. J Orthop Trauma 1995;9:222–26.
[13] Watanabe Y, Harada N, Sato K, Abe S, Yamanaka K, Matushita T. Stem cell therapy:
is there a future for reconstruction of large bone defects? Injury 2016;47(Suppl 1):
S47–51.
[14] Giannoudis PV, Gudipati S, Harwood P, Kanakaris NK. Long bone non-unions treated
with the diamond concept: a case series of 64 patients. Injury 2015;46(Suppl 8):
S48–54.
[15] Masquelet AC, Fitoussi F, Begue T, Muller GP. Reconstruction of the long bones
by the induced membrane and spongy autograft. Ann Chir Plast Esthet 2000;45:
346–53.
[16] Pelissier P, Masquelet AC, Bareille R, Pelissier SM, Amedee J. Induced membranes secrete
growth factors including vascular and osteoinductive factors and could stimulate bone
regeneration. J Orthop Res 2004;22:73–9.
[17] Cuthbert RJ, Churchman SM, Tan HB, McGonagle D, Jones E, Giannoudis PV. Induced
periosteum a complex cellular scaffold for the treatment of large bone defects. Bone
2013;57:484–92.
[18] Giannoudis PV, Faour O, Goff T, Kanakaris N, Dimitriou R. Masquelet technique for the
treatment of bone defects: tips-tricks and future directions. Injury 2011;42:591–8.
[19] Kanakaris NK, Tosounidis TH, Giannoudis PV. Surgical management of infected non-
unions: an update. Injury 2015;46(Suppl 5):S25–32.
[20] Olesen UK, Eckardt H, Bosemark P, Paulsen AW, Dahl B, Hede A. The Masquelet technique
of induced membrane for healing of bone defects. A review of 8 cases. Injury 2015;46
(Suppl 8):S44–7.
[21] Moghaddam A, Zietzschmann S, Bruckner T, Schmidmaier G. Treatment of atrophic tibia
non-unions according to “diamond concept”: Results of one- and two-step treatment.
Injury 2015;46(Suppl 4):S39–50.
[22] Ollivier M, Gay AM, Cerlier A, Lunebourg A, Argenson JN, Parratte S. Can we achieve bone
healing using the diamond concept without bone grafting for recalcitrant tibial
nonunions? Injury 2015;46:1383–8.
[23] Pelissier P, Boireau P, Martin D, Baudet J. Bone reconstruction of the lower extremity:
complications and outcomes. Plast Reconstr Surg 2003;111:2223–9.
[24] Accadbled F, Mazeau P, Chotel F, Cottalorda J, Sales de Gauzy J, Kohler R. Induced-
membrane femur reconstruction after resection of bone malignancies: three cases
of massive graft resorption in children. Orthop Traumatol Surg Res 2013;99:479–83.