G Model

JINJ 7763 No. of Pages 10

Injury, Int. J. Care Injured xxx (2018) xxx–xxx

Contents lists available at ScienceDirect

Injury
journal homepage: www.elsevier.com/locate/injury

Technical Note

Antibiotic coated hinged threaded rods in the treatment of infected

nonunions and intramedullary long bone infections
Jae-Woo Choa , Jinil Kima , Won-Tae Choa , William T. Kentb , Hyung-Jin Kimc,** ,
Jong-Keon Oha,*
a
Department of Orthopaedic Surgery, Guro Hospital, Korea University Medical Center, Seoul, South Korea
b
Department of Orthopaedic Surgery, UC Sandiego Medical Center, Sandiego, CA, USA
c
Department of Orthopedic Surgery Dongtan Sacred Heart Hospital, Hallym University College of Medicine, 40, Seokwoo-dong, Hwaseong, Gyeonggi-do,
South Korea

A R T I C L E I N F O A B S T R A C T

Introduction: Local delivery of high dose antibiotics in the form of antibiotic impregnated polymethyl
Keywords: methacrylate (PMMA) cement beads or coated rods is commonly used in the management of long bone
Intramedullary infection
infections. The downsides of antibiotic cement beads for intramedullary long bone infections are
Infected nonunion
associated with difficulty in removal from the medullary canal, bead breakage, and lack of stability.
Osteomyelitis
Antibiotic rod Antibiotic cement-coated smooth flexible guide wires, rods and nails can have complications such as
Threaded rod delamination or debonding of the cement. In addition, the current techniques for cement rod insertion
have a risk of iatrogenic joint contamination.
To improve upon this technique and decrease potential complications, we propose the use of an antibiotic
cement-coated hinged threaded rod as a temporary intramedullary spacer. This technique utilizes both
an antegrade and retrograde insertion of the threaded rod into the medullary canal through the bony
defect site with connection at the hinge to treat intramedullary long bone infections and infected
nonunions.
Material and Methods: A total of 40 patients were included in the study. The details in making the cement
rod were well documented. The shape of cement rod and the integrity of the cement at the time of rod
insertion and rod removal were compared to identify any cement debonding or delamination. Potential
postoperative complications including iatrogenic joint infection, displacement or breakage of the
threaded cement rods, and fracture displacement were all carefully documented. The preliminary
biological effect of the initial debridement and antibiotic cement rod placement was determined using
the negative conversion rate of intraoperative cultures.
Results: A single antibiotic coated threaded rod was inserted in 18 cases. Two separate antibiotic coated
threaded rods were inserted and connected via hinge in 22 cases. There were zero cases of rod breakage
and no secondary loss of reduction from antibiotic rod placement to the definitive staged operation.
There were zero iatrogenic joint infections. There were zero cases of cement debonding or delamination
from the rod. The conversion rate to a negative culture after initial debridement and antibiotic rod
placement was 85% (34/40 cases).
Conclusions: The use of an antibiotic coated cement threaded rod with a hinge as an intramedullary
spacer provides the benefits of local antibiotic delivery, offers improved construct stability, makes
implant removal easier without delamination of the cement mantle, and utilizes the versatility of a hinge
to prevent violation of native joints when treating infected nonunions and intramedullary long bone
infections.
© 2018 Elsevier Ltd. All rights reserved.

Introduction

* Corresponding author at: Department of Orthopaedic Surgery, Guro Hospital,
Korea University Medical Center, 148, Gurodong-ro, Guro-gu, Seoul, South Korea.
** Corresponding author.
E-mail addresses: oskimhyungjin@gmail.com (H.-J. Kim), jkoh@korea.ac.kr
(J.-K. Oh).
Chronic osteomyelitis and infected nonunions are a challenge for
both patients and surgeons. Complex cases require staged proce-
dures, oftentimes with repeated irrigation and debridements. In
cases of chronic osteomyelitis or infected nonunion of the femur or

https://doi.org/10.1016/j.injury.2018.07.016
0020-1383/© 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
G Model
JINJ 7763 No. of Pages 10
2 J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx
tibia with a prolonged history of recurrence, staged reconstruction
can be a safer approach for the eradication of infection [1,2]. Current
treatment modalities have two main objectives: infection control,
which is achieved by radical debridement, implant removal,
irrigation, local delivery of antibiotics by polymethyl methacrylate
(PMMA)-coated antibiotic bead chains or PMMA-coated rods or
nails; and stabilization, which can be achieved by external fixation or
definitive fixation with an antibiotic-coated nail [3–5].
PMMA antibiotic beads are potential difficultly with removal
due to the ingrowth of granulation tissue, their lack of mechanical
stability, and they may prevent the application of an external
fixator [5–7]. In addition, beads cannot be inserted into the
intramedullary space very easily and are difficult to apply to in the
setting of intramedullary osteomyelitis. Smooth wires or nails
coated with PMMA are a good alternative to cement beads, fill the
intramedullary canal, and provide mechanical stability [3,8,9,10].
However, these smooth wires or nails have complications and
delamination or debonding of the cement during insertion or
removal (Fig. 1). There are reports of debonding of the cement
during insertion and removal with early definitive fixation with
antibiotic coated nails [11,12]. Another concern with the current
technique of antibiotic coated wires or nails is with the insertion
site. There is a risk of iatrogenic joint contamination when placing
the coated implant in patients with an intramedullary infection
without a previous intramedullary implant [12].
To address these issues, we propose a technique for the use of an
antibiotic cement-coated threaded rod with a hinge connection that
can be inserted through the bony defect and debridement site.
Theoretically, cement binding to a threaded rod would provide
increased resistance to debonding, thus lessening the risk of
delamination. In addition, the threaded cement rods can also be
connected via the hinge to make intramedullary splinter without an
additional incision for insertion or create a 130-degree angle for
providing antibiotic effect and stability to the head and neck of
proximal femur. The purpose of this study was to introduce and
evaluate the effectiveness of antibiotic cement coated threaded rods
with a hinge and dual hinged cement coated threaded rod constructs.
Patients and methods
This retrospective case series was approved by our institutional
review board. Between January 1, 2013 and January 1, 2015, 183
Fig. 1. Cement breakage of the smooth, flexible nail. a) Radical debridement of femur diaphysis osteomyelitis and implantation with smooth flexible nail and antibiotic
cement bead chain. b and c) Coated cement breakage of the nail during removal.
Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
patients were treated with a diagnosis of post-traumatic osteo-
myelitis and infected nonunion at our level 1 trauma center.
Diagnoses of post-traumatic osteomyelitis and infected nonunion
were determined by clinical examination, laboratory evaluation
(white blood cell count, C Reactive Protein, and Erythrocyte
Sedimentation Rate), imaging (radiographs, CT scan, MRI) and
intraoperative tissue cultures. Classification of infected nonunion
and post-traumatic osteomyelitis based on the status of bony
continuity. Acute infected nonunion case which could be treated
with implant removal and debridement were excluded. Among
183 patients, 73 patients who were proven to intramedullary
osteomyelitis by intraoperative site specific culture (refered to as
Intramedullary-, Center) and were required radical resection of
dead bone which resulted in critical sized bone defect were
included in this cohort. Among 73 patients, 33 patients who
treated by other modality (with cement spacer or bead alone) or
had insufficient follow up data at least one year were excluded.
Finally, 40 patients who treated with an antibiotic cement coated
hinged threaded rod were enrolled in final cohort.
Staged reconstruction protocol
All patients were treated with a 3-staged surgical protocol. The
first stage is a radical debridement with application of an antibiotic
coated cement hinged threaded rod, cement spacer insertion and
temporary stabilization. The second stage was removal of the
antibiotic coated hinged threaded rod, conversion to the definitive
fixation construct with placement of an antibiotic-loaded PMMA
cement spacer. The final stage is removal of the cement spacer and
autogenous bone grafting.
During the staged surgical protocol, site-specific tissue cultures
were performed intraoperatively for identifying the microorgan-
ism and confirming the presence of intramedullary osteomyelitis.
Site-specific intraoperative cultures were performed in a system-
atic fashion. At least five tissue cultures were obtained, beginning
at the center of the osteomyelitis lesion and expanding to the
periphery during the debridement portion of the procedure
(referred to as: center, intramedullary proximal-(IMP), intra-
medullary distal-(IMD), extramedullary proximal-(EMP), and
extramedullary distal-(EMD)). Copious irrigation with normal
saline followed the debridement. After these portions of the
procedure, all drapes were replaced and contaminated
G Model
JINJ 7763 No. of Pages 10
J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx
instruments were changed. These steps were then repeated after
changing drapes and instruments to obtain a set of post-
debridement cultures. After obtaining post-debridement cultures,
the remaining portions of the procedure were carried out
according to the stage. During this staged surgical protocol,
negative post-debridement cultures were required to in order
proceeding to the next stage. (Fig. 2) Targeted intravenous
antibiotics were administered with consultation with an infectious
disease specialist after identification of microorganism during
whole reconstruction process. Intravenous antibiotics were
continued at least 4 weeks after bone grafting with consultation.
Patient demographics, operation records, and clinical follow-up
data were collected prospectively and analyzed in a retrospective
fashion. Details for making the cement rod construct were
recorded (time to make, instruments, and any complications
encountered while making the rod) in the operation records. The
shape of the cement rod was also recorded with photos at both
insertion and removal to compare the shape and identify any
debonding or delamination of the cement from the threaded rod.
Serial radiographs were compared to identify any significant rod
displacement, rod breakage, change in limb alignment, or fracture
displacement. Follow-up complications related to the cement rods
also were also documented and evaluated. The preliminary
biologic effect of the first debridement and antibiotic-coated
cement rod insertion was evaluated using the negative conversion
rate of intraoperative culture results. The intraoperative cultures of
the 1st and 2nd stages of the protocol for each patient were
compared to determine the biologic effect.
Surgical technique
Selection of rod construct
For these intramedullary osteomyelitis and infected nonunion
cases, we designed two types of intramedullary threaded rods, a
single hinge threaded rod and a dual hinged threaded rod. The type
of rod used was determined preoperatively taking multiple factors
into consideration. The anatomy of the involved long bone, the
location and extent of infection, the presence or previous existence
of an intramedullary implant, previous surgical approaches and
Fig. 2. Staged protocol of infected nonunion and osteomyelitis treatment.
Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
3
incisions, and extent of resection of necrotic or infected bone were
all considered. In brief, a single hinge threaded rod was utilized
when a previously placed infected intramedullary tibial or femoral
nail needed removal. This single hinged rod could then be inserted
in an anterograde fashion through the same start site as the
removed infected nail. A dual hinged threaded rod was utilized in
all other cases. This avoided creating an additional incision around
the hip and knee, which might help to avoid iatrogenic spread of
the medullary infection to a native knee joint or the hip capsule.
This type dual hinged threaded rod requires sufficient medullary
space to allow the connection between the hinges of two rods with
one rod inserted in a retrograde fashion and another rod inserted in
an anterograde fashion both through the through the resected
bone site (Fig. 3).
Method of preparation of the cement rod
A single threaded rod construct was used in patients with an
existing intramedullary infection with an intramedullary nail
present or previous nail removal for infection. A threaded rod of
the appropriate size and length was selected from the Ilizarov
system and was pre-bent in the shape of the intramedullary nail of
the tibia (with Herzog bend proximally) or femur (with appropri-
ate anterior bow), to accommodate patient anatomy. A chest tube
with one-millimeter smaller inner diameter from the largest sized
reamer used to ream the canal was then prepared. As a general
reference, a 38Fr chest tube will create a 9 mm nail and an increase
in 1 Fr increases the nail by 0.33mm [13]. The proximal end of the
rod is fixed with one connecting hinge for easy removal later. Next,
40 mg of bone cement (Antibiotic Simplex, Stryker USA) was mixed
with 4-g vancomycin and 4-g tobramycin. Liquid monomer was
then added and the cement was prepared. The threaded rod is next
positioned inside the tube, keeping the rod centered within the
tube with the connecting hinge outside the tube. The chest tube is
then filled with cement using an enema syringe. The chest tube is
then removed using a 10-blade to make a longitudinal incision
along the length of the chest tube after the cement hardened. Next,
the cooled cement rod was inserted into the canal in a manner
similar to the anterograde nailing procedure for the tibia or femur
(Fig. 4).
The dual hinged rod construct was used in patients who had no
preexisting intramedullary implant. This included patients with a
localized infection and a bone defect either from a previous open
fracture, or a segmental bone defect after debridement of necrotic
infected bone. After debridement, these bony defects had no
cortical contact and both the proximal and distal ends of bone were
viable and free of infection. Two rods, two hinges, a connecting bolt
and nut were needed to make the hinged construct from the
Ilizarov system. An appropriate construct was planned preopera-
tively, and the appropriate length rods were selected intra-
operatively. Each hinge was connected to a rod. As in the previous
method, PMMA cement and antibiotics were mixed. Two rods, for
both the proximal and distal parts of the bone, were prepared and
coated with antibiotic cement as described above. Each individual
rod and hinge construct were inserted through the bony defect.
The rod entering the proximal portion was inserted in a retrograde
fashion through the bony defect. The rod entering the distal
portion was inserted in an anterograde fashion. The hinges were
then connected with a bolt and nut at the defect site. The
remaining defect cavity was then filled with antibiotic cement,
thus eliminating the dead space (Fig. 5).
Choice the overall temporary construct
The torsional, bending, and compressive forces the across the
nonunion and resected bony site makes it difficult for a single rod
alone to stabilize the femur or tibia. In these cases of infection, we
tried to minimize additional stabilization to avoid pin tract
G Model
JINJ 7763 No. of Pages 10

4 J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx

Fig. 3. Diagrams for selection of rod construct.

infections and seeding of the implants. We thus tried to balance the
need for a stable environment to help eradicate the infection with
the addition of unnecessary hardware. Three main factors were
considered in construct stability. These factors included: the
location of the defect, the size and extent of the defect, and the
amount of cortical contact after the debridement. For a defect
located in the diaphyseal region, if more than 3/4 of the cortex
remained intact after debridement, it was stabilized with an
intramedullary rod alone or stabilized with an intramedullary rod
augmented with a cement spacer. Any other wedge type defect of
the diaphysis was stabilized with a rod and augmented with a
cement spacer. A critical sized segmental bone defect in the
diaphysis was filled with a cement spacer and stabilized with a
cement coated rod and an external fixator or with the temporary
use of a low profile plate. Additional temporary stabilization was
required in the metaphyseal area. A single planar external fixator
(EF) or temporary plating using a low profile plate was added to a
cement intramedullary rod and spacer.
Results
Between January 1, 2013 and January 1, 2015, 40 patients with
intramedullary osteomyelitis or an infected nonunion were treated
with antibiotic cement-coated threaded rods from the Ilizarov
system using our surgical protocol for the first stage treatment of
osteomyelitis. The mean patient age was 49.6 years (range, 19–81

Fig. 4. Method of making a single threaded rod. a) Single hinged threaded rod was pre-bent according to the entry point and alignment with the single hinge at the proximal
end. b) Antibiotics (Vancomycin + Tobramycin) mixed polymethyl methacrylate (PMMA) cement was inserted into the chest tube by an enema syringe. c) The tube was
removed after hardening of the cement. d) The single rod was inserted in an anterograde fashion.

Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
G Model
JINJ 7763 No. of Pages 10
J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx
Fig. 5. Method of making a Dual hinged threaded rod. a and b) Two threaded rods, two hinges, and a connecting bolt and nut for making the hinged rod. c) Rods were inserted
without an additional joint incision and connected inside the segmental bone defect. d) A bridging external fixator was applied for temporary stabilization. e) The segmental
bone defect was filled with a cement spacer.
years). Twenty-eight patients had infections of the tibia and 12
patients had infections of the femur. Twenty-one patients had
chronic post-traumatic osteomyelitis and 17 patients had infected
nonunions after previous fracture surgery. Of these 17 infected
nonunions, 2 had an open fracture as their initial injury. The
specific level of long bone involvement was classified using the AO/
OTA classification [14]. The extent of infection and host type were
classified using the Cierny-Mader classification [15]. The average
duration of infection was 13.5 months.
At the time of our initial debridement surgery, 7 of the 40
patients presented with external fixators in place. Fifteen patients
had their previously implanted plate in place. A previously placed
intramedullary nail was present in 14 cases. One patient presented
with combined fixation of an intramedullary nail and a plate. Three
patients had undergone previous implant removal so that no
implant existed at the time of our initial staged surgery.
Among the 7 patients presenting with external fixators in place
at the time of initial debridement, a dual hinged threaded rod was
inserted in 6 cases, and a single threaded rod was inserted in one
case in a retrograde fashion up the tibia through the debridement
site at the first stage procedure. No additional incisions were made
for rod placement through the knee or hip. Twelve of the 14 cases
with presenting with an infected intramedullary nail, a single
threaded rod was inserted through the previous nail entry site after
implant removal. In the remaining two cases, dual hinged threaded
rods were inserted to span the entire length of the long bone.
Among the 15 cases with an infection and previously placed plate,
a dual hinged threaded rod was inserted in 12 cases. A single
threaded rod was used in the remaining three cases. None of these
cases required an incision through the knee or hip for cement rod
insertion. All of the cement rods were inserted through the bony
defect after the debridement of infected nonviable bone. A single
threaded rod was used in three cases via retrograde insertion
through the defect site. In the three cases with no existing implant,
a dual hinged threaded rod was used in two cases and a retrograde
single threaded rod in one case. These cases all avoided an incision
throughout the knee or hip for cement rod placement and instead
utilized placement of the rod through the bony debridement site.
In the patient with an existing plate nail construct, both implants
were removed and a single threaded rod was inserted in an
anterograde fashion through the nail removal and previous nail
entry site. In summary, a single threaded rod was inserted in 18
cases (13 cases in an anterograde fashion, 5 cases in a retrograde
Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
5
fashion). A dual hinged threaded rod was utilized in 22 cases. The
cement intramedullary rods were combined with a cement spacer
in 11 cases, an external fixator in 23 cases, and a low profile plate in
6 cases for additional temporary fixation. The cement rod
remained in place for an average of 23.6 days (range, 5–60 days)
until secondary debridement. The effect of the first debridement
and the antibiotic-coated cement rod was evaluated by the
negative conversion rate of the intraoperative culture results. The
rate of conversion to a negative intramedullary culture after the
first debridement and cement rod placement was 85% (34/40
cases). The six cases that had positive intramedullary cultures after
the first procedure all had negative culture results after the repeat
debridement and cement rod placement. An average of 1.5
debridement surgeries were needed to obtain a negative culture
result before the definitive fixation procedure (Table 1).
The average time spent making a single threaded rod was
12.5 min. A slightly longer average time of 15.2 min were needed to
construct the dual hinged threaded rod. There were no major
complications encountered when making the rod (cement
breakage, failure due to premature hardening of PMMA, or
adhesion of the cement to the chest tube). There were no cases
of rod breakage or failure or loss of reduction during follow up.
There were no instances of iatrogenic joint infections of the hip or
knee. Moreover, cement debonding or delamination were not
observed during cement rod removal in any of the cases (Table 2).
Discussion
Cement beads are used to provide higher antibiotic concen-
trations locally in the treatment of osteomyelitis. The drawbacks
for the use of antibiotic beads are the difficulty in placement in the
medullary canal and the lack of stability they provide at the bony
defect, fracture, or nonunion site. Other downsides include the
inability to completely fill the dead space and the difficulty in
external fixator placement after medullary canal bead placement
[2,16]. The antibiotic-impregnated cement rod with a metal core
and a cement mantle helps to overcome these obstacles by
providing high concentration local antibiotic therapy with added
stability to the area with easier removal from the medullary canal
at a later procedure [3,7,10,17,18].
Ball-tipped guide-wires, Ender’s nails, and Kirshner wires are
also commonly used to create intramedullary antibiotic cement
spacers. The problems with using these implants to create a
 6

JINJ 7763 No. of Pages 10

G Model
Table 1
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and

Patient’s demographics and surgical details.

Demographics Surgical details

Case Age Sex Host Ana. Duration of Diagnosis Location Previous Defect type Rod type Incision for rod Temp. Culture result at first Duration Culture result No. of debridement
typea typeb infection (AO- implant after radical insertion fixation debridementd of Rod at 2nd procedures before
(month) OTA) debridement through the constructc (day) debridement definitive fixation
joint
1 48 M A III 22 Infected 32, Nail Segmental Single, Y R+C MRSA 17 Negative 1
nonunion subtroce Anterograde
2 33 M A III 6 Infected 32 Nail Segmental Dual, Y R+C MRSA Enterococcus 18 Negative 2
nonunion Anterograde faecalis
3 49 F A III 12 Infected 32, Nail Segmental Single, Y R + EF + C MRSE 10 Negative 2
nonunion subtroc Anterograde
4 81 F B III 11 Infected 32, Nail Segmental Dual, N R + EF + C MRSA 11 Negative 1
nonunion distal retrograde
5 64 M A IV 10 PTOM 32, EF Segmental Dual, N R + P+C Pseudomonas 14 Negative 1
subtroc Retrograde
6 60 M A II 20 PTOM 32 Nail Wedge long Single, Y R+C MRSA 28 Negative 1

J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx
Anterograde
7 52 M A III 4 Infected 32, Nail Segmental Single, Y R + P+C MRSA 42 Negative 1
nonunion subtroc Anterograde
8 47 M A IV 24 PTOM 32, EF Segmental Dual, N R + P+C MRSE 21 Negative 1
subtroc Retrograde
9 74 F B IV 12 PTOM 33,41 EF Segmental Dual, N R+C MRSA 16 Negative 1
(fused Retrograde
knee)
10 51 M B IV 5 Infected 33 Plating Segmental Dual, N R + EF + C MRSA 5 MRSA 2
nonunion Retrograde
11 77 F B III 12 Infected 33,41 Plating Segmental Dual, N R+C MRSE 14 Negative 1
nonunion (fused Retrograde
knee)
12 47 M A IV 5 Infected 41 EF Segmental Dual, N R + EF + C MRSA 5 Negative 1
nonunion Ilizarov Retrograde
13 43 M B IV 7 Infected 42 Nail Segmental Single, Y R + P+C Acinetobacter 60 Negative 1
nonunion d/t Anterograde baumannii
open fracture
14 42 F A IV 20 PTOM 42 Plating Segmental Dual, N R + EF + C Enterobacter 45 MRSA 2
Retrograde cancerogenus MRSA
15 55 M B III 6 PTOM 42 Nail Segmental Single, Y R + EF + C Enterococcus faecium 56 Negative 1
Anterograde MRSE
16 38 M A III 11 Infected 42 Nail Segmental Single, Y R + EF + C MRSA 9 Negative 1
nonunion Anterograde
17 51 M B IV 42 PTOM 42 Nail Segmental Single, Y R + EF + C MRSA, MRSE, 11 Negative 1
Anterograde Enterococcus faecalis
Acinetobacter
baumannii
18 49 M B III 35 PTOM 42 Nail Segmental Single, Y R + EF + C MRSE 14 Negative 2
asthma Plating Anterograde
19 54 M A III 6 Infected 42 Plating Segmental Dual, N R+C MRSE 12 Negative 2
nonunion Retrograde
20 19 F A IV 11 Infected 42 EF Segmental Single, N R + EF + C MRSE Pseudomonas 21 MRSE 2
nonunion Retrograde
21 53 M A I 12 PTOM 42 None Wedge long Dual, N R+C MSSA 58 MSSA 2
Retrograde
22 62 M B IV 15 PTOM 42 Nail Segmental Single, Y R + P+C MSSA 14 MSSA 3
Anterograde
23 74 M A III 9 Infected 42 Plating Segmental Dual, N R + EF + C Staphylococcus 21 Negative 1
nonunion Retrograde caprae
 JINJ 7763 No. of Pages 10
G Model
24 22 M A IV 36 PTOM 42 Nail Segmental Single, Y R + EF + C MSSE 55 Negative 2
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and

Anterograde
25 48 M B III 14 PTOM 42 Nail Segmental Single, Y R + EF + C Pseudomonas 19 Negative 2
DM Anterograde
26 29 M A IV 18 Infected 42 Plating Segmental Dual, N R+C Pseudomonas 20 Pseudomonas 3
nonunion Retrograde
27 41 M A IV 6 Infected 42 Nail Segmental Single, Y R + EF + C Serratia marcescens 36 Negative 2
nonunion d/t Anterograde
open fracture
28 58 M A III 14 PTOM 42 None Segmental Dual, N R + EF + C MRSE 14 Negative 1
Retrograde
29 55 M A III 9 PTOM 42 EF Segmental Dual, N R + EF + C MRSE 18 Negative 1
Ilizarov Retrograde
30 51 M A III 16 Infected 42 Plating Segmental Dual, N R + EF + C MRSA 42 Negative 1
nonunion Retrograde
31 62 M B III 10 PTOM 42 Plating Segmental Dual, N R + EF + C MRSA 29 Negative 1
Retrograde
32 42 M A III 8 Infected 42 Plating Wedge Dual, N R+C MRSA 15 Negative 1
nonunion Retrograde
33 23 M A I 15 PTOM 42 None Wedge Single, N R+C MRSA 14 Negative 1

J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx
Retrograde
34 48 M A III 13 Infected 43 Plating Segmental Single, N R + EF + C MRSA 28 Negative 1
nonunion Retrograde
35 51 M B III 6 PTOM 43 Plating Segmental Dual, N R + EF + C MRSA 23 Negative 1
(fused Retrograde
ankle)
36 27 M A IV 7 PTOM, 43 Plating Segmental Single, N R + EF + C MRSA 23 MRSA 3
Retrograde
37 26 M A IV 22 PTOM 43 Plating Segmental Single, N R + EF + C MRSE, MRSA 23 MRSA 3
(fused Retrograde
ankle)
38 74 M B III 7 PTOM 43 Plating Segmental Dual, N R + EF + C Pseudomonas 21 Negative 2
CRF (fused Retrograde Staphylococcus
ankle) lugdunensis
39 57 M A III 8 PTOM 43 EF Segmental Dual, N R+C MSSA 33 Negative 1
Ilizarov Retrograde
40 53 F B CRF III 14 Infected 32, Plating Segmental Dual, N R + P+C MRSA 10 Negative 1
nonunion subtroc Retrograde
a,b
: Cierny-Mader classification.
a
Host type, A: Good immune system, B: Compromised locally BL or systemically BS, C: Required suppression or no treatment.
b
Anatomical type, I: Medullary, II: Superficial, III: Localized, IV: Diffuse osteomyelitis.
c
Temp. fixation construct, Temporary fixation construct.
R + C: Rod + Cement spacer.
R + EF + C: Rod + External fixator + Cement spacer.
R + P+C: Rod + plating + Cement spacer.
d
Microorganisms which were grown up from initial intraoperative culture.
MRSA: Methicillin Resistant Staphylococcus Aureus.
MRSE: Methicillin Resistant Staphylococcus Epidermis.
MSSE: Methicillin susceptible Staphylococcus Epidermis.
MSSA: Methicillin susceptible Staphylococcus Aureus.
e
subtroc, Subtrochanteric lesion.

7
G Model
JINJ 7763 No. of Pages 10

8 J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx

Table 2
Complications related to antibiotic coated threaded rods.

Result
Intraoperative complications Time for making rod Single rod: 12.5 min. (range, 8.2-14.5)
Hinged rod: 15.2 min. (range, 13.7-16.8)
Breakage of cement 0/40
Adhesion of cement with tube 0/40
Follow-up complications Cement breakage 0/40
Loss of reduction 0/40
Iatrogenic joint infection 0/40
Debonding and delamination 0/40
Discomfort due to protruded tip of rod Single rod: 2/18 cases
Hinged rod: 0/22 cases

cement antibiotic rod are cement rod breakage and the potential
problems encountered with implant removal. Breakage of these
rods has been documented especially in cases where additional
stability constructs were not [10,12]. There is also a risk of
delamination of the cement mantle during both insertion or
removal of the smooth flexible rods. These rods must be introduced
in a similar fashion as an intramedullary nail, which may lead to
unnecessary and unwanted exposure to other uninfected areas of
the long bone including the start sites for tibial and femoral nailing
such as the knee joint and the proximal femur. With this insertion
technique, not only the start sites are exposed to infection, but also
other areas of the long bone previously unaffected by infection are
contaminated. Qiang et al. reported that during the management of
19 patients, they experienced difficulty with rod removal in two
cases and the spread of the infection to the knee joint in one case
[3]. Thonse and Conway [11] used antibiotic-impregnated nails in a
recent report with a 95% success rate. However, in their study, 15%
of the patients required exchange nailing, and 20% showed
delamination of the cement during removal and insertion.
The advantages of this technique include the threaded rod to
help with cement bonding, the increased rod diameter when
compared to the more traditional antibiotic nail using a guide wire.
A rod from the Ilizarov system is 6 mm in diameter compared with
the standard flexible nails, which are typically 3.5 and 4.5 mm in
diameter. This larger sized diameter creates a stiffer more stable
construct, which in turn helps to prevent translation at the bony
defect or fracture site and maintain better alignment. These
threaded rods have threads with a 1 mm pitch, which provides
better adherence to the cement mantle, making implant removal
easier without cement breakage or delamination even after
remaining in place for a prolonged duration. In our series there
were no instances of delamination or debonding of the cement
mantle during either implant insertion or implant removal for any
patient. Second advantage is flexibility to create the appropriate

Fig. 6. Peri-prosthetic fracture post-TKA with infected nonunion (Patient number 4).
a) Implant failure with infected nonunion.
b) Stage 1, Radical debridement with temporary fixation with an external fixator; an antibiotic cement-coated dual hinged threaded rod was inserted through the segmental
defect. Hip and knee joints were not opened.
c) Stage 2, Definitive fixation 11 days after initial debridement. The defect was filled with a cement spacer.
d) Stage 3, Reconstruction with autogenous bone graft using the induced membrane technique.
e) 12 months after bone grafting, bony consolidation could be achieved.

Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
G Model
JINJ 7763 No. of Pages 10

J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx 9

Fig. 7. Fixative failure with infected nonunion (Patient number 40).

a) Implant failure with infected nonunion.
b) After radical debridement, hinged rods were inserted and temporary augmentation plating was done.
c and d) Dual hinged threaded rods with plating using a low profile plate for temporary stabilization and metaphyseal angular stability.
e) Definitive fixation using an intramedullary nail after initial debridement.
f) After 7 months after bone grafting, bony consolidation is progressing but not completely.

length implant using the hinge connection. Thrid advantage is the
ability to insert the rods in both a retrograde and anterograde
fashion through the bony defect to avoid potential spread into the
knee joint or around the capsule of the proximal femur (Fig. 6).
Another advantage is the ability to create various angles at the
hinge connection between the two rods (Fig. 7).
The limitation of this antibiotic coated threaded rod system is
that rod itself cannot be used for definitive fixation. In addition, it
often requires some form of additional temporary stabilization
such as a cement spacer, temporary external fixator, or low profile
plate. As our study shows, all of these adjuncts can be used with a
threaded rod for a long duration with few complications. In our
case series, 8–12 weeks was required for maturation of the pedicle
flap for soft tissue coverage before proceeding to the next stage
after initial debridement. Our temporary stabilization was suffi-
cient for this staged time frame without any evidence of further
displacement or implant failure. This staged protocol ensured the
next phase of the operation could be performed with a stable flap
and blood supply. The second limitation of this study is that this
study was conducted under modified bi-Masquelet’s technique.
Compared to original Masquelet’s technique, we divided second
procedure of internal fixation and bone graft into separate
procedures with additional debridement. Our aggressive protocol
of debridement was based on the reason behind additional
debridement and conversion of internal fixation without bone
grafting (second stage) could reduce the possibility of recurrence
of infection. But, a comparative study verifying the effectiveness of
three-stage reconstruction is still needed.
In this study, we have focused on the surgical technique, its
clinical application, and the effectiveness of the antibiotic coated
threaded rod. The high rate of recurrence in osteomyelitis
treatment could be attributed to various factors including host
factors, the microorganisms causing the infection, the antibiotic
treatment, the surgical treatment, and others [19,20]. We still need
future controlled trials with a larger number of patients to help
validate this treatment protocol. Despite these limitations, we
believe that an antibiotic cement-coated threaded rod can be
widely used for the management of intramedullary long bone
infections.
Chronic osteomyelitis and infected nonunions are difficult to
treat and often require multiple procedures. The use of the
threaded rod from the Ilizarov system as a metal core for the
antibiotic cement provides a high concentration of local antibiotics
to the area, offers improved stability at the bony defect or fracture
site, and makes implant removal easier without delamination of
the cement mantle. In addition, this technique utilizes the
versatility of a hinge connection to avoid joint contamination,
prevent potential start site issues, and create an intramedullary
implant to match the dimensions of the proximal femur.
Conflict of interest
The authors declare that they have no conflict of interest.
Funding
This work was supported by the Technology Innovation
Program (10077279) funded by the Ministry of Trade, Industry &
Energy (MOTIE, Korea).
Ethical approval
This article does not contain any studies with human
participants or animals performed by any of the authors.
Informed consent
Informed consent was obtained from all individual participants
included in the study.
References
[1] Kempf I., Grosse A, Rigaut P. The treatment of noninfected pseudarthrosis of
the femur and tibia with locked intramedullary nailing. Clin Orthop Relat Res
1986;142–54.
[2] Klemm KW, Borner M. Interlocking nailing of complex fractures of the femur
and tibia. Clin Orthop Relat Res 1986;89–100.
[3] Qiang Z, Jun PZ, Jie XJ, Hang L, Bing LJ, Cai LF. Use of antibiotic cement rod to
treat intramedullary infection after nailing: preliminary study in 19 patients.
Arch Orthop Trauma Surg 2007;127:945–51.

Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016
G Model
JINJ 7763 No. of Pages 10

10 J.-W. Cho et al. / Injury, Int. J. Care Injured xxx (2018) xxx–xxx

[4] Klemm KW. Antibiotic bead chains. Clin Orthop Relat Res 1993;63–76.
[5] Patzakis MJ, Wilkins J, Wiss DA. Infection following intramedullary nailing of
long bones. Diagnosis and management. Clin Orthop Relat Res 1986;182–91.
[6] Seligson D, Popham GJ, Voos K, Henry SL, Faghri M. Antibiotic-leaching from
polymethylmethacrylate beads. J Bone Joint Surg Am 1993;75:714–20.
[7] Miller ME, Ada JR, Webb LX. Treatment of infected nonunion and delayed
union of tibia fractures with locking intramedullary nails. Clin Orthop Relat
Res 1989;233–8.
[8] Bar-On E, Weigl DM, Bor N, Becker T, Katz K, Mercado E, et al. Chronic
osteomyelitis in children: treatment by intramedullary reaming and
antibiotic-impregnated cement rods. J Pediatr Orthop 2010;30:508–13.
[9] Sancineto CF, Barla JD. Treatment of long bone osteomyelitis with a
mechanically stable intramedullar antibiotic dispenser: nineteen
consecutive cases with a minimum of 12 months follow-up. J Trauma
2008;65:1416–20.
[10] Paley D, Herzenberg JE. Intramedullary infections treated with antibiotic cement
rods: preliminary results in nine cases. J Orthop Trauma 2002;16:723–9.
[11] Thonse R, Conway J. Antibiotic cement-coated interlocking nail for the
treatment of infected nonunions and segmental bone defects. J Orthop Trauma
2007;21:258–68.
[12] Selhi HS, Mahindra P, Yamin M, Jain D, De Long [9_TD$DIFF][1_TD$DIFF]Jr. WG,
Singh J. Outcome in patients with an infected nonunion of the long bones
treated with a reinforced antibiotic bone cement rod. J Orthop Trauma
2012;26:184–8.
[13] Hake ME, Young H, Hak DJ, Stahel PF, Hammerberg EM, Mauffrey C. Local
antibiotic therapy strategies in orthopaedic trauma: practical tips and tricks
and review of the literature. Injury 2015;46:1447–56.
[14] Fracture and dislocation compendium. Orthopaedic trauma association
committee for coding and classification. J Orthop Trauma 1996;10(Suppl. 1)
1–154 v-ix.
[15] Cierny Iii G, Mader JT, Penninck JJ. The classic: a clinical staging system for
adult osteomyelitis. Clin Orthop Relat Res 2003;414:7–24.
[16] Klemm K. The use of antibiotic-containing bead chains in the treatment of
chronic bone infections. Clin Microbiol Infect 2001;7:28–31.
[17] Smeltzer MS, Thomas JR, Hickmon SG, Skinner RA, Nelson CL, Griffith D, et al.
Characterization of a rabbit model of staphylococcal osteomyelitis. J Orthop
Res 1997;15:414–21.
[18] Nelson CL, Hickmon SG, Harrison BH. Elution characteristics of gentamicin-
PMMA beads after implantation in humans. Orthopedics 1994;17:415–6.
[19] Simpson A, Deakin M, Latham J. Chronic osteomyelitis. Bone Joint J
2001;83:403–7.
[20] Patzakis MJ, Zalavras CG. Chronic posttraumatic osteomyelitis and infected
nonunion of the tibia: current management concepts. J Am Acad Orthop Surg
2005;13:417–27.

Please cite this article in press as: J.-W. Cho, et al., Antibiotic coated hinged threaded rods in the treatment of infected nonunions and
intramedullary long bone infections, Injury (2018), https://doi.org/10.1016/j.injury.2018.07.016