Effect of surgical approach on bone vascularisation, fracture and soft tissue healing: comparison of less invasive to open approach Dr Martin Eduard Wullschleger MBBS University of Zurich MD University of Zurich FMH Surgery EBSQ Trauma Submitted for the award of degree of Doctor of Philosophy in The Faculty of Built Environment and Engineering, Queensland University of Technology Brisbane, Australia 2010
193
Embed
Effect of surgical approach on bone vascularisation, fracture ......Effect of surgical approach on bone vascularisation, fracture and soft tissue healing: comparison of less invasive
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
and Lactate Dehydrogenase values in the ORIF group compared to MIPO.
After four weeks, the torsional stiffness was significantly higher in the MIPO group
(p=0.018) compared to the ORIF group. The torsional strength also showed increased
values for the MIPO technique (p=0.11). The measured total mineralised callus
volumes were slightly higher in the ORIF group. However, a newly developed
morphological callus bridging score showed significantly higher values for the MIPO
technique (p=0.007), with a high correlation to the mechanical properties (R2
After eight weeks, the same trends continued, but without statistical significance.
=0.79).
In summary, this clinically relevant study, using the newly developed severe trauma
model in sheep, clearly demonstrates that the minimally invasive technique minimises
additional soft tissue damage and improves fracture healing in the early stage compared
to the open surgical approach method.
iv
TTaabbllee ooff CCoonntteennttss Keywords………………………………………………………………………...……ii Abstract……………………………………………………………………………….iii Table of Contents……………………………………………………………………..iv List of Figures………………………………………………………………………..viii List of Tables……………………………………………………………………..…xviii List of Abbreviations………………………………………………………………..xviii Statement of Original Authorship………………………………….…………………xx Acknowledgments……………………………………………………………………xxi CHAPTER 1: INTRODUCTION……………………………………………………1 1.1 Background………………………………………………………………….....1 1.2 Aims and Objectives…………………………………………………………..2 CHAPTER 2: LITERATURE REVIEW……………………………………………3 2.1 High energy injuries…………………………………………………….…......3 2.2 Fracture healing……………………………………………………….……….5 2.2.1 Types and phases of bone healing…………………………………………....…5 2.2.2 Role of the soft tissues in the process of fracture healing………………….…...8 2.3 Fixation strategies ..............…………………………………………………...9 2.3.1 Plate osteosynthesis……………………………………………………...…….10 2.3.1.1 Open reduction and internal fixation (ORIF)………………………………….12 2.3.1.2 Minimally invasive plate osteosynthesis (MIPO)……………………………...15 2.3.1.3 Clinical evidence of MIPO………………………………………………...…..17 2.3.1.4 Experimental evidence of MIPO…………………………………...………….18
v
2.4 Animal trauma models………………………………………………………19 2.4.1 Soft tissue trauma models.…………………………………………………….20 2.4.2 Fracture models ……………………………………………………………….21 2.4.3 Combined animal trauma models……………………………………………...25 2.5 Methods to monitoring soft tissue damage………………………………….28 2.6 Methods to assess fracture healing…………………………………………..30 CHAPTER 3: RESEARCH DESIGN………………………………………………34 3.1 General Research Plan……………………………………………………….34 3.2 Development of trauma model……………………………………………….35 3.2.1 Development and validation of soft tissue trauma model……………...………36 3.2.1.1 Requirements on the degree and extent of the soft tissue damage………….....36 3.2.1.2 Development of the device………………………………………...…………..36 3.2.1.3 Description of the soft tissue trauma (STT) device……………………...…….39 3.2.1.4 Soft tissue trauma device: tests and validation…………………...……………40 3.2.2 Development and validation of fracture creation model…...…...……….…….50 3.2.2.1 Requirements of the fracture model………………………...…………………50 3.2.2.2 Development of the fracture model……………………………………………51 3.2.3 Validation of combined trauma model…..…………………………………….62 3.2.4 Development and validation of ‘Supporting Trolley System’…...…………….63 3.3 Study design……………………………………………….…………..….…...66 3.3.1 Establishment of fracture fixation model…………………………………...….66 3.3.2 Experimental design and description of operative techniques…...…………….71 3.3.3 Instruments and implants…………..……………………………………….….79
vi
3.4 Animal care and welfare…………………………………….……………….81 3.4.1 Sheep data……………………………………………..……………………….81 3.4.2 Perioperative management ………………………………………………..…...82 3.4.3 Postoperative management and care…………………………………...………83 3.4.4 Euthanasia…………………………………..………………………….....……84 3.5 Methods to monitor the soft tissue damage……………………………..…..84 3.5.1 Local soft tissue assessments…………………………………………………..84 3.5.2 Systemic investigation (blood serum tests)…………………………………….86 3.6 Methods to assess fracture healing…………………………………………..87 3.6.1 Conventional radiographs………..……………………………………...……..87 3.6.2 Computer tomography (CT)…………………………………...………………90 3.6.3 Mechanical testing……………………………………...……………………...96 3.7 Statistical analysis………………………………………...…………………..98 3.7.1 Estimation of sample size and power analysis…………………………......…..98 3.8 Animal ethics documentation…………………………...……………………98 3.9 Procedures and Timeline……………………………………...……………...99 3.9.1 Pilot study…………………………………………..………………...…...….100 3.9.1.1 Methodology and timeline………………………………………........………100 3.9.1.2 Results……………………………………………..………………………….100 3.9.1.3 Discussion and Conclusion………………………………………………..….108 3.9.2 Main test series…………………………………………………..………...…109 CHAPTER 4: RESULTS…………………………………………..………………110 4.1 Characterisation of Trauma………….………………...………..…………110
vii
4.1.1 Soft tissue trauma recovery...…………………………………………..…….110 4.1.2 Fracture model………….…………………………………………..….…….112 4.2 Minimally invasive versus open plate osteosynthesis………………….…114 4.2.1 General results………………………………….………………………...….114 4.2.2 Soft tissue recovery………………………………………………………….116 4.2.3 Fracture healing………..………………………………………………...….118 CHAPTER 5: DISCUSSION……………………………………………………..124 5.1 Trauma model evaluation……………………………………………….…124 5.2 Minimally invasive versus open plate osteosynthesis……………………128 CHAPTER 6: CONCLUSION AND FUTURE DIRECTIONS……...………...134 BIBLIOGRAPHY………………………………………………………………….138 APPENDICES………………………………………………………………...……149 Appendix A: List of publications (2005 – 2009)
Appendix B: List of conference papers (2005 – 2009)
Appendix C: AO Research Grant approval document 05-W17 (20 June 2005)
In the last century, at the same time as numerous improvements and developments
across all medical specialties took place, advances in the field of fracture fixation also
occurred, improving the outcome for function and quality of life of patients with limb
fractures. While the healing progress of certain bone fractures clearly benefits by the
method of using a plate and screw fixation construct, in 1995 a new application of plate
fixation started. After a small surgical approach remote to the fracture zone, the fracture
fixation plate was applied by sliding it in a percutaneous fashion over the bone, passing
the fracture area and, finally, fixing the plate to the bone with screws. Prior to the
definitive fixation, the crucial realignment of the main fragments was carried out, if
possible, with closed, indirect reduction manoeuvres. This technically demanding
procedure was used instead of the standard method with a large surgical open approach
(open reduction and internal fixation: ORIF) including soft tissue dissection and
denudation of the periosteum to visualise the fracture fragments prior to the plate
fixation. Nowadays, this technique is widely accepted and more and more commonly
used.
This so called ‘minimally invasive plate osteosynthesis’ (MIPO) approach, was
developed to minimise the additional damage to the surrounding soft tissues of the bone
during the surgical procedure in order to optimise the biological conditions of the
fracture healing process. Clinically, this effect gains even more importance in critical
situations with severe soft tissue trauma and might be crucial in some anatomical areas
with thin soft tissue coverage of the bone, e.g. medial shin bone region.
However, the MIPO technique faces some significant drawbacks. Because of its
closed fashion without direct sight and overview of the fracture and the implant
manipulations, it requires training, manual skills and surgical experience to adequately
perform the procedure. In order to overcome the lack of visualisation, intra-operative
fluoroscopy with an image intensifier is used, but that inflicts the patients, surgeons and
nurses to a significantly higher exposure of radiation. If the operation including the
fracture reduction is not appropriately managed, complications may result, such as mal-
2
alignment and mal-rotation problems, as well as infections and delayed or non-unions.
Consequently, further secondary operations may be required to correct those conditions
and ensure the appropriate healing progress.
Clinically, many case collection studies indicate that the minimally invasive plate
osteosynthesis is a reasonable but serious treatment option; however, no randomised,
prospective study has proved this so far. Other evidence comes from the only
experimental animal study, performed by Schuetz et al. (1999) 1
Therefore, without clear evidence and in light of the increasing popularity of the
minimally invasive procedures in clinical practice, this study aims to answer the clinical
and scientific question: Does the surgical approach influence the fracture healing
process? Or, in other words: Is it worth undertaking a more challenging, complex
minimally invasive technique to perform plate fixation of bone fractures or should we
still rely on the simpler, open surgical method?
, which directly
compared the two different surgical approaches, minimally invasive versus open
technique. It did not show any significant differences in the mechanical integrity of the
healing fractures.
1.2 AIMS AND OBJECTIVES
In order to answer this question and to assess the effect of the surgical approach on
soft tissue recovery and the progress of fracture healing, the goal of this present study is
to develop and validate a comprehensive and suitable animal trauma model. Following
the development of this model the second step is to use it in an in-vivo animal
experiment to investigate, whether the healing progress of a defined, severe injury (soft
tissue damage and fracture) is better using a minimally invasive technique for the plate
stabilisation of the fracture in comparison with the standard, conventional open
approach for plate fixation.
• Development of severe trauma model to the distal femur and thigh in sheep:
Part I of the project:
o Development of an apparatus to inflict a severe soft tissue injury
(Tscherne III) 2
3
o Development of a mechanism to create a multi-fragmentary shaft
fracture (AO C-type 3
• Validation of the newly developed devices:
)
o In-vitro validation of the soft tissue injury apparatus
o In-vivo validation of the degree and extent of the soft tissue damage
o In-vitro validation of fracture creation on sheep bones
o In-vivo validation of fracture pattern
o In-vivo validation of combined trauma model
• Experimental study: (MIPO versus ORIF)
Part II of the project:
o Planning and performance of pilot experiments with 2 sheep (one each)
o Pilot study evaluation and final adjustments for main study
o Planning and performance of main series with 24 sheep:
• In-vivo monitoring (early soft tissue and fracture healing)
• Post-mortem testing (fracture healing progress)
• CT scanning
• Mechanical testing
• Histology
o Evaluation and analysis:
• Final Trauma model evaluation
• Soft tissue recovery evaluation
• Fracture healing evaluation
CHAPTER 2: LITERATURE REVIEW
2.1 HIGH ENERGY INJURIES
Data from the Australian Institute of Health and Welfare (AIHW) describes the
epidemiology of traumatic fractures in Australia for 2001/02 4 with forearm fractures
(21,931 patients per annum), fractures of the femur including neck of femur (21,353)
and fracture of the lower leg including ankle fractures (19,109), as the three most
common injuries for hospital admissions. Statistics shows an increasing trend for these
4
injuries, which leads to a substantial economic burden to the health system. Fifty four
percent of all traumatic fractures occur by falls, about 8% of them by high falls (> 5 m)
and 19% result from road traffic accidents. High-energy or high-velocity mechanisms
are mainly caused by those road traffic accidents and high falls.
Open or compound long bone fractures occur with a frequency of about 11.5 per
100,000 persons per year; the most common area is the tibial shaft (Figure 1), but
femoral shaft, distal femur and proximal tibial fractures tend to show up in the more
seriously injured patients and the soft tissue envelope is moderately to severely
damaged (Gustilo type II and III 5) in 68.6% of all those fractures 6.
Figure 1: Clinical picture of a severe lower leg injury sustained by a motorbike crash: large wound with a substantial, soft tissue damage (Gustilo type IIIB) and multiple bone fragments of the complex tibial shaft fracture.
Combined, severe bone and soft-tissue injuries challenge orthopaedic surgeons and
need to be treated as surgical emergencies. The damage to the morphological structures
caused at the time of trauma is set and further damage has to be prevented or
diminished through strict treatment modalities. The body itself responds to the injury in
a fundamental reaction to protect the tissue integrity, and the response aims to replace
lost cells and restore compound tissues by the physiological and biochemical processes
of wound healing 7. This wound healing sequence of overlapping reactions starts with
the initial coagulation phase, which takes some minutes and consists of two
mechanisms: vasoconstriction and haemostasis. The inflammatory phase (exudative
period) follows on with an active vasodilatation and increased blood flow to the injury
5
site. The first cells to move into the damaged tissues are neutrophilic granulocytes
(unspecific infection-control) and macrophages (removal of necrotic tissue and micro-
organisms as well as production and secretion of cytokines). The next stage is the
granulation phase with fibroplasia, angiogenesis, and epithelialisation and wound
contraction and takes a few days. After the establishment of a newly formed connective
tissue with rapid capillary proliferation and the contraction of the wound edges, the last
stage of the tissue repair, the scar formation phase occurs. This stage is of greatest
importance as it requires a balance of synthesis and degradation of collagen to re-
establish the tissue integrity and strength.
Severe injuries require a comprehensive understanding of those pathophysiological
mechanisms, clinical judgement skills as well as experience to adequately manage the
fracture and soft-tissue treatment in a timely and procedural manner. As part of this,
significant complications such as wound infections, soft-tissue loss, compartment
syndrome, non-union, chronic pain and stiffness, joint contracture, chronic
osteomyelitis, reflex dystrophy and even amputations have to be prevented 7
.
2.2 FRACTURE HEALING
Fracture healing is a complex, however, well-orchestrated, regenerative process that
is initiated in response to injury. During the repair process, the pathway of normal
embryonic development is repeated with coordinated participation from several cell
types. There are four components to the injury site: the cortex of the bone, the
periosteum, the bone marrow and the external soft tissues, all of which contribute to the
healing process. The extent to which each component is involved depends on the
conditions present at the injured tissue, such as the level of growth factors, hormones,
nutrients, pH-level, oxygen tension, the electrical environment and the mechanical
stability of the fracture 8
.
2.2.1 Types and phases of bone healing
Histologically, fracture healing has been divided into direct (primary) and indirect
(secondary) healing. Much of the current knowledge regarding cellular events and their
6
temporal and spatial characteristics has been elucidated from investigating fracture
healing in rats 8; 9
- Direct / primary cortical fracture healing:
:
Direct cortical fracture healing involves a direct attempt by the cortex to re-
establish the integrity of the bone by forming discrete remodelling units known as
‘cutting cones’, a process aimed at restoring mechanical continuity. Vascular
endothelial cells and perivascular mesenchymal cells produce the osteoprogenitor cells
that differentiate into osteoblasts. During this process little or no periosteal response is
noted (no callus formation) 10. Perren (2008) 11 even describes direct fracture healing as
an internal remodelling process and, therefore, not healing in a strict sense of the word,
but a side effect of internal removal of necrotic bone in the ‘fracture gap’. However,
this direct ‘remodelling’ of the broken cortex requires anatomical reduction of the
fracture fragments and almost complete stability (‘absolute stability’12) as well as
minimisation of the inter-fragmentary strains 13
- Indirect / secondary fracture healing:
. This is rarely seen in clinical practice.
Secondary bone healing occurs in the vast majority of bony injuries. It involves a
combination of intramembraneous and endochondral ossification with the subsequent
formation of callus. This type of healing benefits from motion 14 and is inhibited by
rigid fixation 13
Intramembraneous ossification involves the formation of bone directly, without
first forming cartilage. Committed osteoprogenitor and undifferentiated mesenchymal
cells that reside in the periosteum produce histological ‘hard callus’ within a few
millimetres from the fracture site
.
9. In this type of healing, the bone marrow with a high
cellular density contributes to the formation of bone during the early phase of healing,
when those endothelial cells transform into polymorphic cells, which subsequently
express an osteoblastic phenotype 15
Endochondral ossification consists of the recruitment, proliferation and
differentiation of undifferentiated mesenchymal cells into cartilage, which becomes
calcified and eventually is replaced by bone.
.
7
- Four stages of endochondral fracture healing:
- haematoma formation, inflammation and angiogenesis
This stage is characterised by formation of the fracture haematoma, which
resolves into granulation tissue with the typical inflammatory cascade. This
cascade is driven by signalling molecules: pro-inflammatory cytokines
(Interleukin-1, Interleukin-6 and tumour necrosis factor alpha (TNF-α); growth
factors such as transforming growth factor beta (TGF-β), bone morphogenetic
proteins (BMPs) and growth differentiation factors (GDFs); and the third group
of angiogenic factors (platelet-derived growth factor (PDGF), fibroblast growth
intra-medullary ‘nails’ as pins, un-reamed or reamed metallic rods, and several
10
combinations across those implants. Certain injuries may even be managed with a
combination of those implants.
High energy injuries with compound, contaminated, severely damaged soft tissue
and fracture conditions with defect wounds or loss of soft tissues or bony fragments, are
differently managed in a sequential manner: a step by step management is crucial for
the successful limb recovery. Initially, as soon as possible, a thorough washout and
debridement for decontamination, a temporary fixation (mostly with an external fixator)
and, if required, an urgent re-vascularisation with vascular reconstruction of the
damaged limb has to be carried out. This is called ‘damage control orthopaedics’. Then,
on a regular basis for the following days, or sometimes even a couple of weeks,
repetitive washouts and debridements are performed to achieve clean and early healing
soft tissue conditions. At a third stage, the definitive fracture fixation is defined and
performed, as well as the soft tissue management finalised by coverage of the bone and
the implant, if necessary by a plastic procedure with local or free soft tissue flaps. The
definitive fracture treatment includes joint reconstructions and in complex bone defect
situations, using special options such as limb shortening (temporary) or bone grafts
(autograft, allograft, bone substitutes), either primary or in a delayed timely fashion.
2.3.1 Plate osteosynthesis
The plate osteosynthesis was established at the end of the 19th century by the
surgeon C. Hansmann from Hamburg, Germany. He started to treat tibial shaft fractures
with a fixation method like a plate. He used a metal strip and fixed it with steel screws
to the bone fragments. In his publication 27 ‘A new possibility to fix bone fragments of
complicated fractures’ he presented the result of 21 cases treated with this method and
that was the foundation for all later techniques of plate osteosynthesis.
11
Figure 2: C. Hansmann worked in the middle of the 19th century in Hamburg (Germany) as a surgeon and is nowadays considered as the pioneer of plate osteosynthesis. In his publication 27
he presented his developed method for fracture fixation by plate.
An important step for surgical procedures in general and for the operative treatment
of fractures, in particular, was the development of asepsis which led to further
developments of fixation methods. Albin Lambotte was substantially involved in this
process 28. This Belgian surgeon created the term ‘osteosynthesis’ and brought forward
the concept of internal and external ‘splinting’. Today, these principles are still used in
almost all modern stabilisation methods. Apart from the external fixator, he developed
many different plate and screw designs, which made anatomical reconstruction and
early mobilisation of the limb and the patient possible.
Figure 3: Albin Lambotte (1866-1955) 28 was a Belgian surgeon, who defined the term: osteosynthesis. To the right side: A drawing of the treatment of a non reduced tibial shaft fracture by using a plate.
12
In the same period of time William A. Lane (1856-1943) performed surgery in
London and operated on fractures, first of all with cerclage wires, then after 1883 also
with screws, and finally at the end of the 19th century using plates. He published
extensively his experiences about operative fracture treatment in 1914.29 Another
Belgian surgeon, who is considered as the ‘Father of the modern osteosynthesis’, was
Robert Danis. In 1947, he published the development of a special compression plate
which permitted immediate mobilisation after fracture stabilisation 30
. With this kind of
fracture fixation with axial compression and rigid fixation the fractures healed without
radiological signs of callus formation. He described this finding as ‘primary fracture
healing’, in contrast to the ‘secondary fracture healing’ with callus formation by
conservative treatment. This conclusion led to the opinion that callus healing was
connected to instability with the associated tendency to develop delayed or non-unions
or risk of implant failure.
2.3.1.1 Open reduction and internal fixation (ORIF)
In the 1950s, the founders of the Swiss Association for the Study of Internal
Fixation (AO/ASIF) 31 standardised the use of plate systems. They described the main
goals of fracture treatment in the first edition of the ‘AO Manual of Internal Fixation’ in
1965 21 as the restoration of the function of the injured limb. Through performing a
stable osteosynthesis, the bone should get the primary strength to recover by early
functional aftercare. This could be achieved by a conventional, open surgical approach
for visualisation of the fracture site, open reduction of the fragments and stabilisation of
the reduced fracture with a plate, so called: open reduction and internal fixation
(ORIF). Complications, such as wound and bone infections (by large approaches and
wide dissections of the bone), mal-alignments and fracture disease caused by long-term
immobilisation of the limb and the patient, should be avoided. To reach this goal, four
fundamentals were set in the ‘AO Manual’ 21
1. anatomical reduction
:
2. absolute stability with inter-fragmentary compression
3. preserving blood supply throughout atraumatic operation technique
4. avoiding additional damage by immobilisation.
13
The original AO technique was carried out using lag screws or with a special
tension device. Later on the so called ‘dynamic compression plate (DCP)’ (Synthes,
Switzerland) made axial compression possible through eccentric screw insertion in the
special designed plate holes. Another concept to protect the lag screws was developed
using ‘neutralisation or protection plates’. All these plate techniques were implemented
to get primary fracture healing without callus formation. But these operation techniques
with sophisticated anatomical reduction required large open approaches and wide
dissections with additional damage and oppression of the blood supply to the fractured
bone fragments. As a consequence of such traumatic operation techniques, delayed
healings, pseudarthrosis and increased infection rates could be seen.
Parallel to this progress in the field of plate osteosynthesis, the intra-medullary
nailing procedure was initiated by Küntscher in the 1940.32 One important stage of
these further developments was the invention of locking nails in the 1970s. The
findings of this closed reduction and fixation technique with secondary callus healing,
even of large comminuted fractures, led to a change in the concept of fracture fixation
from the anatomical reduction and fixation with absolute stability to a fracture
reduction with correct axes, length and rotation of the bone and then a fixation
technique with relative stability. Consequently, bridging plates were performed without
touching or fixing of the multi-fragmentary fracture area. These plates were only
stabilised from one main fragment to the other main fragment (proximal and distal).
The advantage of this technique was to avoid further devascularisation of the small
fragments by the surgical procedure. Long plates were used to obtain optimal stress
distribution of the plates and implant failures could be reduced. The callus healing was
not an unwanted side effect anymore, but the aim of fracture treatment was a safe and
secure fracture healing with callus formation. Ganz suggested the term of a biological
plate osteosynthesis 33, which was characterised by an atraumatic surgical performance
without dissection of the fracture zone, indirect reduction techniques and stabilisation
with any bridging fixation device 34
The development of the plate design went parallel to the above mentioned stages.
The properties and shapes of the plates changed from simple, straight plate designs to
many different anatomically pre-shaped plates with different sizes adapted to the
.
14
different skeletal regions. These plates were all fixed and pressed to the bones to
overcome the friction forces, which were produced by loading of the limbs.
Experimentally, Perren and his group (1988) 35 found areas of osteonecrosis and
osteoporosis underneath the compression plates. If those cortical changes were purely
caused by biological reasons with reduced periosteal perfusion, or whether the
mechanical conditions changed to bone remodelling by stress shielding, is still
unproven. Clinically, the use of these plates was associated with higher risk for
infection and delayed or non unions in fracture healing 36-38
A first step to reduce the contact area between the plate and the bone was via plate
adaptation with undercutting of its under-surface with lower contact or point contact
design (LC-DCP and PC-Fix, Synthes, Switzerland). These constructs caused less
periosteal damage
.
39. Finally, with the development of the so called ‘internal fixator’,
the plates are positioned without any contact to the bone. To achieve this periosteal
perfusion preserving procedure, the screw-hole interface has been changed. Instead of
normal screws, which were inserted in the plate holes and worked only with
compression, the screw heads lock into the plate hole and function as an angular stable
device. This locking mechanism is achieved by a threaded screw head, which fits and
locks into the thread of the hole in the plate 40; 41. This improvement in terms of the
biological environment also changed the plate mechanics. In conventional compression
plates, the loads were transmitted by the friction between the implant under-surface and
the bone surface, and were therefore dependent upon the screw anchorage and the bone
quality. With this new angular stable screw-plate system, the forces are totally held
from the screws in one main fragment, transmitted by the screw-plate construct, over
the other main fragment. Because this involves the same concept as for an external
fixator, but its application is inside the body with coverage of soft tissues, it is called
‘internal fixator’. This technique of the internal fixator was realised with the so-called
PC-Fix 42, which was shown in animal studies to cause better cortical perfusion and less
cortical necrosis 41; 43; 44, advanced fracture healing in terms of mechanical conditions 41;
45 and a reduction in the incidence of infection 46. A further development of the internal
fixator technique was a new plate design including special instrumentation tools (such
as an aiming arm), named as Less Invasive Stabilisation System (LISS). This plate-
15
screw construct was developed to fix complex distal femur and proximal tibial fractures 47; 48. Clinical prospective collection studies 49-53
The last evolutionary development of plate design is the combination of both screw-
plate interfaces in one system. The LCP (Locking Compression Plate, Figure 4)
using the LISS method for distal femur
fractures showed high union rates of the fractures, relatively low complication rates
even in cases of severe injury and lower rates of bone grafts.
54 is
designed with special plate-holes: the configuration of the holes correspond to a figure
‘eight’-shape: on one side the conventional compression hole and, on the other side, the
threaded hole for angular stable screw insertion. Thus, the LCP can be used as a
conventional compression plate, as an internal fixator or for certain applications as a
combination of both techniques 55. The LCP system is regarded as technically mature
and offers numerous fixation possibilities and has proven its worth in complex fracture
and revision situations 56
.
Figure 4: The design of the Locking Compression Plate (LCP) 54
with the ‘eight-shaped combi’ hole, the left part for conventional, compression screw insertion and the right part for head locking screw insertion.
(Dr Roland Steck) was performed to determine the impactor movement (distance/time)
and the final speed defined with plotting of displacement versus time curves (Microsoft
Excel, Redmond, WA, USA, Figure 12 and 13).
Figure 9: Picturial overview of Figure 10: Ruler fixed perpendicular to the high-speed camera setup. the impactor and counterpart. The foam noodle is positioned to dampen the hit.
42
Figure 11: Illustration of frame-by-frame evaluation with x-vision software.
Figure 12: Displacement vs time curve 1: Figure 13: Displacement vs time curve 2: Linear portion of the curve Determination of the velocity representing the velocity of the impactor. Slope of the before the impact. linear portion = Impact velocity
Results: Two trials with the high-speed films were evaluated. The first impact
velocity was 7.40 m/s, the second impact velocity was 7.40 m/s. Therefore, the average
impact velocity was consistent at 7.4 m/s (26.6 km/h).
Conclusion: The validation of the pendulum device in terms of the reliability of the
impact speed shows that both trials with high-speed camera measurements present
exactly the same values of 7.4 m/s impact speed. Hence, with the given impactor
weight and the constant impact speed, the energy of the impact is consistent and very
reliable.
43
Report of in vivo tests
Following the foam noodle and cadaver leg tests, the new pendulum device was
tested and validated in vivo.
Four sheep (Merino ewes, approximately seven years old, slightly underweight
(<30 kg), but in good general health condition) were anaesthetised and later sacrificed
for other scientific purposes. At the end of the general anaesthesia and with an adequate
analgesic dose, the sheep were positioned on the pendulum trestle with the distal thigh
(5 cm proximally from the knee joint) within the groove of the impact counterpart and
the soft tissue trauma is induced. The subsequent hit was carried out with the sheep
turned around on the device and re-positioned to fit the contra-lateral thigh in the
counterpart groove. In order to create and detect several degrees of the severity of the
soft tissue injury, different impactor weights were chosen and monitored. The following
are the denominations of how the experiments were performed with the additional
weights: Sheep 1 - No additional weight;
Sheep 2 – 2 kg additional weight (refer to Figure 14),
Sheep 3 - 0.5 kg additional weight
Sheep 4 - No additional weight.
The second traumatic blow to the leg was timed to be 15 minutes before sacrifice,
so that the there was sufficient interval to determine the ‘full’ extension of the soft
tissue damage, as well as the borders between the damaged and the undamaged tissues.
Immediately after the sacrifice, the regions of interest (distal part of the thigh on both
sides) were dissected by scalpel to evaluate the macroscopic extend and depth of the
emerged damage. Sheep 3 alone was managed differently: After euthanasia, both hind
limbs were amputated (hip exarticulation) and a native post-mortem CT scan (Philips,
Brilliance 64, Netherlands) was performed in the Department of Radiology at The
Prince Charles Hospital. Then, the legs were transported back to the animal laboratory
and macroscopically dissected, in the same manner as the other six legs.
The above-mentioned test series helped us to validate the exact position of the lever
arm distance as well as the best angle of the impactor and its counterpart (see Technical
data in Chapter 3.2.2 and Figure 15).
44
Figure 14: Outer side of impactor Figure 15: Final position of the counter- showing 2 kg additional weight. part block for blow against the right thigh.
All eight trial blows were successfully performed (Figure 16). During the trial
blows, the sheep did not show any sign of pain or discomfort. The first hit included an
‘accidental’ femur fracture (Figure 17), which occurred, because of the mal-positioning
of the sheep’s leg between the grooves of the impact blocks.
45
Figure 16: (Left) In vivo trial of soft tissue trauma device: The pendulum is moving towards the point of impact on the distal thigh. The ‚marked cross’ on the right picture shows the point of contact of the pendulum with the sheep.
Figure 17: Surgical dissection to visualise the damage incurred. This figure also shows incidental femur shaft fracture caused due to the mal-positioning of the sheep’s leg.
46
Figure 18: The team involved in the first in vivo trial of the soft tissue trauma device in the animal operating theatre of The Prince Charles Hospital, Brisbane.
Dissection Results
The soft tissue trauma was in vivo tested on five sheep, under general anaesthesia
and 15 minutes later the sheep were sacrificed. Straight after their sacrifice a careful
dissection in layers, from skin to bone, was carried out. The macroscopic assessment
included the detailed description of the skin, subcutaneous and muscle conditions such
as bruises, contusion marks, signs of ruptures and haematomas as well as the conditions
of the nerve and vascular structures at the postero-medial localisation of the distal thigh
(refer to Appendix F). This assessment was performed by the candidate and confirmed
by Dr Kathleen Wilson and Dr Roland Steck.
47
Summary of the assessment of macroscopic dissections:
• The soft tissue trauma device was highly reproducible in terms of its impact speed
and impact energy.
• The soft tissue trauma model was able to produce moderate (Figure 19) to very
severe soft tissue damage (Figure 20). In one particular scenario, even a ‘crush’
fracture was produced (Figure 17).
• The impact weight and the severity of the soft tissue damage correlated with each
other; 2 kg additional impact weight caused significantly more destructive damage
(Figure 20) compared to the trials without any additional weight to the pendulum
(Figure 19).
• The slight modification to the counterpart block to the pendulum impactor during
the last trial protected the postero-medially located vital structures, such as the
sciatic nerve and the femoral vessels.
• The leg preparation with circumferential shearing and shaving at the impact
location, as well as a relaxed muscle tone by the chosen leg position was important
considering the reproducibility of the soft tissue injury.
Figure 19: This macroscopic dissection of the left picture shows a moderate muscle contusion to the lateral vastus muscle and the right picture shows an intact sciatic nerve with superficial perineural haematomas. Both the pictures represent the damage resulted from a pendulum hit without any additional weights.
48
Figure 20: These pictures represent the extensive damage after the pendulum hit with 2 kg additional weights. In the left picture, the lateral vastus muscle is severely damaged, partially disruptured, the subcutanous layers degloved with small bruises and haematomas. The right picture shows the sciatic nerve substantially contused with epineural haematomas, but longitudinally intact. The deep soft tissues are bruised with underlining haematomas and edematous.
Evaluation of computer tomography:
For further evaluation and validation of the soft tissue trauma device and the
damage it inflicts, a post-mortem CT scan was performed 15 minutes after the soft
tissue injury. For this purpose, one of the sheep (no 336) was euthanized and its right
leg was amputated at the level of the hip joint. Then, the amputation site was wrapped
in sterile, moist gauzes and the entire leg was transported to the Radiology Department
for imaging with the high-resolution CT scanner (Philips, Brilliance 64, Netherlands:
slice thickness 0.67 mm, pixel size 0.625 mm, 120 kV, 250 mAs, rotation time 1 s).
After the scan, the leg was brought back to the animal laboratory for macroscopic
dissection analysis.
Results: The native scan without intravenous contrast of the entire right leg is
reconstructed in all axial, sagittal and frontal planes. The native scan in sagittal plane is
shown in Figure 21. The areas of damaged muscles are visible as slightly darker tissue
spots within the homogenous normal muscle tissues. The muscles are all in anatomical
place and there is no evident disruption in their positioning. The anterior part of the
rectus femoris is severely bruised and some areas contain intramuscular haematomas.
49
The exact quantification of the extent of the damage is difficult because of the limited
resolution of this non-contrast scan.
Figure 21: High-resolution CT scan of a right sheep leg after soft tissue trauma trial showing sagittal plane of the entire leg (Sheep 336).
Figure 22: CT scan with mid-sagittal reconstruction of Figure 23: Axial view of the right thigh. Yellow spots show: intramuscular bruises the distal thigh region. and haematomas.
Although the images could not quantitatively describe the damage inflicted, the
images acquired by post-mortem CT scan without intravenous contrast qualitatively
50
indicates significant muscle damage. This finding corresponds to the macroscopic
dissection findings that shows severe amount of bruising in all muscles around the
femur including some intramuscular haematomas. Further analysis of the exact extent
of the damage is difficult because of the non-contrast scan that does not clearly
distinguish between the normal and damaged tissues. Fifteen minutes time interval
between the soft tissue trauma and euthanasia of the sheep was sufficient to show the
muscle damage, but for a more precise evaluation of the extent of the soft tissue injury,
a longer period of time would be necessary.
Conclusions of the soft tissue trauma trials:
This newly developed soft tissue trauma device has demonstrated its ability to
create moderate to very severe soft tissue injury to a well-defined area of the sheep’s
leg. The severity of the damage depends on the amount of additional weights loaded on
the pendulum impactor. The clinically comparable Tscherne III soft tissue damage 2
This successful setup of the soft tissue trauma device will be applied in the pilot
study on two animals as part of the combined trauma model. This model will be
employed to study the effect of the surgical approaches on soft tissue and fracture
healing. The soft tissue damage will be monitored and re-validated prior to the main
series of the animal experiments.
was obtained by the impactor weight alone. The results show a significant soft tissue
bruise of all layers including subcutaneous and intramuscular haematomas as well as
disruption of the shear or degloving injury mechanisms. The vital anatomical structures
including the sciatic nerve and femoral vessels are still preserved.
3.2.2 Development and validation of fracture creation model
3.2.2.1 Requirements of the fracture model
In real clinical circumstances, the previously described soft tissue injury model will
ideally inflict a comminuted fracture as a result of the high-energy trauma that is
involved. Therefore, a fracture model is required to create a multi-fragmentary fracture.
This fracture pattern will include two main fragments, proximal and distal, and at least
51
one intermediate fragment. The intermediate fragment splits the two main fragments
without any direct contact with either of the main fragments. The fracture created will
correspond to an AO C-type fracture as classified by the Association for the Study of
Internal Fixation (AO/ASIF). The fracture will be around 3 cm in length and will be
located at the distal end of the femur shaft, in the transitional area of the metaphyseal to
the diaphyseal region and 5 cm proximal to the condylar line (Figure 24). All these
fracture properties fulfil the requirements of a defined, complex fracture and ensure an
accurate and consistent fixation construct. The following are the requirements to satisfy
such a fracture type:
• a reproducible fracture pattern in terms of fracture length and numbers of
fragments created
• a less invasive and constant approach to create the fracture, and minimize
iatrogenic soft tissue as well as bone damage
• simple tools applied in a well-defined sequence to cause breakage of the
bone
Figure 24: Multi-fragmentary distal femur fracture – a butterfly type, AO C-type fracture 3
of a cadaver sheep bone.
3.2.2.2 Development of the fracture model
The required fracture pattern emulating an AO C-type fracture was constructed in
collaboration with Romano Matthys (AO Development and Research Institute, Davos,
Switzerland) (See Appendix E).
52
The fracture pattern constituted:
• two transverse fracture lines (distal and proximal)
• one bicortical, longitudinal fracture line to separate the middle fragment in two
small intermediate fragments (Figure 24)
The resulting “H” shaped fracture pattern evidently meets the criteria of an AO C-
type fracture 3
The fracture pattern was created via the following steps:
.
• Step I
•
: Weakening of the cortical bone with total or partial osteotomies and by
drilling holes into the bone.
Step II
o Mediating direct impact to fracture area by either
: Initiating the final fracture by
A blunt blow with a heavy weight – hammer, pendulum, etc.,
A sharp hit with a chisel or
With an expanding device inserted in drill holes or osteotomy
gaps
o Mediating indirect impact to leg or bone by:
Applying external force either manually or with a torque
machine
The analysis of the merits and demerits of the fracture model enabled us to infer the
following:
1) The use of osteotomies was not preferred, because of its potential to cause
‘unnatural’ fracture gaps with smooth fragment ends and heat necrosis.
2) Creating the fracture with a ‘blunt blow’ method was discarded due to its inability to
create a reproducible fracture pattern (Figure 17).
3) Any technique requiring a larger surgical approach or a technically complex
apparatus was also abandoned.
Out of the remaining possibilities the challenge was to find the optimal combination
and sequence of the following tools (instruments): oscillating saw, drill, blade bar and
chisel. Three different blade bars with two, three and four blades (‘wings’, Figure 25),
53
were designed and produced in the AO Development Institute, Davos (Switzerland) and
sent to Brisbane. The other instruments were delivered from Synthes, Australia and
Switzerland.
Figure 25: Custom made blade bars with two, three and four blades.
To start with and measure of the forces to break a sheep femur in the shaft region,
three femur bones (supplied from the local butcher, Westfield Mall, Chermside) were
tested with the Instron machine, applying a three-point bending moment. The bones
fractured at an average force of 3493±629 N (SD) as ultimate bending failure. Figures
26, 27 and 28 show the bending test, the degenerated bending to failure curve and the
resulted fracture respectively.
Figure 26: Right femur bone tested with the Instron machine for three-point bending properties.
54
Three point bending test of sheep femur
0
500
1000
1500
2000
2500
3000
3500
0 1 2 3 4 5 6 7distance (mm)
Forc
e (N
)
Figure 27: Test curve showing breakage of an ovine femur on application of three-point bending forces by the Instron machine (Figure 26).
Figure 28: Fractured femur resulted after the application of three-point bending forces. The image shows a random fracture configuration with a posterior intermediate split fragment.
All the in vitro fracture model tests were carried out with cadaver femurs of sheep
supplied from a local butcher (Westfield Mall, Chermside) and tested on the vice of the
biological bench in the Biomechanical Engineering laboratory, Gardens Point,
Queensland University of Technology.
As a part of the validation process, an ‘H’ fracture was created on several femur
bones using different numbers, sizes and distribution of drill holes. The femur was
fractured by turning the proximal and distal external fixator frames towards each other
(Figure 29 left). Examination of the fractured femur showed that the fracture lines were
randomly distributed, sometimes not even through the weakening drill holes (Figure 29
right).
55
Figure 29 (left): Cadaver femur bone with drill holes in H shape. Two external fixator frames, proximal and distal to the fracture, are inserted for fracture initiation. Figure 29 (right): Fractured femur with random fracture lines. There is no evident fracture seen through some of the drill holes.
A subsequent optimisation included the creation of two transverse fracture planes
aimed to crack the thick and strong anterior cortex without any longitudinal splits. In
order to achieve that, it was decided to perform two partial, anterior osteotomies, at the
expense of a small surgical approach. Further tests on many cadaver bones to refine the
model were carried out, including implementation of chisel hits to initiate the fracture
planes more precisely. Hence, the fracture creation model was further optimised and the
following sequence of fracture creation pattern was tested on 16 femur bones (Figure
30 and Appendix G):
• Partial anterior osteotomy was performed (1/5 of circumference) 50 mm
proximal from knee joint line with a sagittal saw. The sagittal saw blade used
was narrow and thin (Stryker, 2108-150, USA) (Figure 30).
• Partial anterior osteotomy was performed (1/5 of circumference) 80 mm
proximal from knee joint line with a sagittal saw. The same blade was used as
before.
• Two holes of 2.5 mm were drilled through each of the anterior osteotomies.
Four other holes of 2.5 mm were drilled in the remote cortex in posterior and
postero-lateral directions. The drill was purchased from Synthes, 310.240,
Switzerland (Figure 30).
56
• One hole of 3.5 mm was bicortically drilled in antero-posterior direction 65 mm
proximal of the knee joint line with the 3.5 mm drill bit from Synthes, 310.360,
Switzerland (Figure 30).
• Two, unicortical holes of 3.5 mm were drilled in the lateral cortex 50 and 80
mm proximal to the knee joint line.
• A 2.5 mm drill bit was inserted in each of the lateral 3.5 mm drill holes. Then,
the remote, medial cortex was perforated with 2.5 mm drill holes in two
different directions (medial and postero-medial).
• Blade bars with two sharpened wings were inserted bicortically into the 3.5 mm
hole in antero-posterior direction with the support of a slotted hammer. Slotted
hammer was purchased from Synthes, 332.200, Switzerland (Figure 30).
• Chisels of 8mm width were inserted into the two anterior osteotomies with the
help of the slotted hammer used previously. This initiated a transverse fracture
that manifested as a crack in the remote cortex (transverse fracture initiation).
Chisels were flat and straight and were purchased from Synthes, 397.970,
Switzerland.
• The longitudinal fracture line was created with the rotation and removal of the
blade bar.
Figure 30: All the instruments used on the fracture model. From left, saw blade, drill bit 3.5 mm, drill bit 2.5 mm, blade bar with two wings and chisel respectively.
Results: As mentioned before, the fracture creation model was tested on the sixteen
distal femur bones of sheep. The average bone length was 18.2 cm. The proportion of
the partial osteotomies to the entire circumference was 19% for the proximal osteotomy
57
and 20% for the distal osteotomy. The longitudinal fracture initiation with the blade bar
and transverse fracture initiation with the two chisels was incomplete in three cases. All
fractures were C-type fractures according to the AO classification. Except for one case
of three-fragment fracture, in all the other fifteen cases, the fractures consisted of four
main fragments (Figure 31). The measured maximal fracture length was 4.3 cm in
average. In four cases (25%) the fractures were more than 5.0 cm long due to a
longitudinal split at the proximal lateral side (Figure 31).
Figure 31: C-type fracture with long lateral fragment of 66 mm.
Conclusion
: This newly applied fracture model with partial anterior osteotomies
and several drill holes creates reproducible C-type fractures. The longitudinal splitting
that produced longer intermediate fracture fragments in 25% of all cases is not ideal,
and would have caused inconsistency for fracture fixation (with the same plate
construct), for mechanical testing and histological analysis. Therefore, two steps of the
fracture sequence had to be adjusted: Firstly, the partial osteotomies will be slightly
deeper, especially on the lateral proximal aspect to avoid the longitudinal fracture
splitting. Secondly, to initiate the longitudinal fracture line with the blade bar, the
insertion will be performed only unicortical. Both the above-mentioned changes should
ensure that the fracture length will not be longer than 4.0 cm in further experiments.
Second Test Series:
These minor changes of the previous fracture series were implemented and the
second in vitro series were performed with 20 bones.
58
Results
: The average bone length of 20 femurs was 17.9 cm. The osteotomies were
slightly deeper with proximally 22.6% and distally 22.9% of the whole circumference
(Figures 32 and 33). The fracture initiation was constantly done by unicortical blade bar
insertion and chisel hits. In all 20 cases, the fracture configuration resulted in an AO C-
type butterfly fracture with four main fragments (Figures 34 and 35). In one case, a
small fifth fragment was produced. The maximal fracture length averaged at 3.5 cm. In
two cases, a short proximal lateral split towards the middle shaft region was seen. In
five cases, the longitudinal fracture line was created by bicortical blade bar hits without
any complications.
Figure 32: Proportion of partial proximal Figure 33: Proportion of partial distal osteotomy to entire femur circumference. osteotomy to entire femur circumference.
Figure 34: Fracture configuration with intact Figure 35: Fracture pattern (periosteum periosteum and fragments in place. removed and fragments laid to sides).
Proximal Osteotomy Femur
22.6%
77.4%
Distal Osteotomy Femur
22.9%
77.1%
59
Conclusion
: The outcome of this second series clearly shows that the developed
fracture model creates highly reproducible fractures of consistent type and length. The
technique employed is reliable and easy to perform without complications. Therefore,
this combination of fracture model with two partial anterior osteotomies, antero-
posterior and latero-medial drill holes, unicortical blade bar insertion and chisel hit
initiation, will be continued to be employed for further validation.
Cadaver leg tests
To further validate the fracture model, in vitro tests were carried out with the entire
sheep’s leg instead of just the sheep’s femur bone. This validation process using a
standard lateral approach would test the fracture model’s performance and its reliability
through the soft tissues. The entire fracture model was performed on six cadaver hind
limbs of Merino ewe’s sheep (Figure 36).
Figure 36: Ovine cadaver leg fixed in the vice of the bench: After the fracture model was performed, the outcome was checked by a standard lateral approach to the femur shaft.
60
In all of those six cadaver leg experiments, the fracture model was successfully
carried out through an antero-lateral, trans-muscular approach with a skin incision of 3
cm (Figure 37 left). No complications occurred and the fracture configuration was
consistent (AO type C3) with two intermediate fragments (Figure 37 right) and fracture
length shorter than 4 cm. To minimise tension forces during the procedure, an antero-
lateral external fixator frame was inserted prior to the testing of the fracture model
(Figure 38).
Figure 37 (left): 2.5 cm skin incision required to perform the fracture model. Figure 37 (right): Standard lateral surgical approach to visualise and evaluate the fracture outcome.
Figure 38: Picture with external fixator frame in situ and the fracture model performed by blade bar and chisel insertion through 3 cm skin incision.
Conclusion: The newly developed fracture model was successfully validated to
create a multi-fragmentary fracture of the distal femur in sheep. This model generated
fracture patterns of consistent and reproducible configurations and lengths. The final
sequence of the fracture model (Figure 39) consists of two partial anterior osteotomies,
antero-posterior and latero-medial drill holes. Also, the longitudinal split was initiated
61
between the two osteotomies using a sharpened blade bar inserted unicortically into the
anterior middle drill hole. To generate the ultimate butterfly fracture, slightly oblique
fracture lines were created from transverse osteotomies using two chisel hits. This
sequence is easy and reproducible, even through a small (< 3 cm) skin incision and a
trans-muscular approach to the antero-lateral distal aspect of the femur shaft.
In all the in vitro trials, ovine femur bones were used to test the fracture model and
the outcome was very successful. Further experiments required the testing of the
combination of the fracture model and the two step fixation method (see Chapter 3.3.1).
Female sheep bones would not be ideal for these experiments due to their thin cortical
bone quality and hence, the probability of further bone splitting it presents. Therefore,
male sheep (Merino wethers) with stronger and slightly larger bone properties were
chosen for future experiments.
Figure 39: Fracture creation sequence: 1) Two partial transverse anterior osteotomies performed at 30 mm apart from each other. 2) In between the osteotomies, an antero-posterior (a-p), bicortical, 3.5 mm drill hole was placed. 3) Through the anterior osteotomies, four bicortical, 2.5 mm, oblique holes were drilled at 30 degrees to each other. 4) In the latero-medial direction: four bicortical, 2.5mm diameter holes were
62
drilled at 30 degrees to each other. 5) To split the middle segment longitudinally into two fragments, a sharpened blade bar was inserted into the a-p drill hole by soft hits with a hammer. 6) Finally, two chisels were used to initiate the two transverse fractures by a strong hit with the hammer.
3.2.3 Validation of combined trauma model
The soft tissue trauma model as well as the fracture creation model were tested and
validated as separate applications so far. To perform both models as an ‘all-in-one
impact’ procedure would reflect the real-life circumstance better. Such an experimental
scenario is unrealistic in terms of reproducibility and reliability, mainly because of the
fracture configuration (as discussed in previous chapters). Therefore, the combined
trauma model was decided to be setup as a two step concept: firstly, the production of
the soft tissue damage on the pendulum trestle, and secondly, the creation of the desired
fracture on the operating theatre table. To assess and validate the entire course of the
two steps p, an in vivo setup was required. A female Merino sheep, 8 years of age and
weighing 30Kgs was sacrificed for other research purposes. Before euthanasia and
under general anaesthesia, the combined trauma model with two steps was tried out.
The soft tissue trauma was produced at the distal thigh of the right hind limb without
any additional impactor weight. This caused moderate to severe soft tissue damage as
assessed by macroscopic dissection. Immediately after, the fracture model was
performed at the ‘pre-injured’ right leg. The fracture model sequence resulted in an AO
C-type fracture of 3.8 cm with four main fragments and one smaller intermediate
cortical piece. After the combined trauma model was successfully completed, the
fracture fixation model with an external fixator frame was tested (see chapter 3.3.1 and
Figure 44). Although a longitudinal split occurred in the femur shaft during the
insertion of the proximal Schanz’ screws, the implant was positioned successfully. In
the end, a stable fixation construct was created over the distal femur fracture. To
prevent the occurrence of longitudinal splits in future experiments, the external fixator
pins have to be inserted prior to the fracture creation and the Schanz’ screw should be
positioned as far away from the fracture region as possible. In addition, stronger bones
with thicker cortical properties such as those of male sheep (Merino wethers) will be
employed.
63
Conclusion of the combined trauma model:
This newly developed trauma model represents the first combined severe trauma
model in large animals. Because of its safe and reliable performance with consistent
soft tissue injury and fracture configuration, this trauma model is considered as a
valuable tool for experimental, in vivo studies.
3.2.4 Development and validation of ‘Supporting Trolley System’
Outlining the problem
During the early stage of recovery from anaesthesia and injury, the animals require
clinical observation and support for their impaired leg. Research shows, that other
animal laboratories in Australia (Adelaide and Sydney), Europe (Davos, Switzerland
and Berlin, Germany) and the United States (Cleveland) use supporting sling systems
or similar adapted devices to aid in recovery. These devices have been used for several
years yielding optimal recovery outcomes, low rates of complications and reduced loss
of animals. The main goal of these devices is to protect the sheep and the operated leg
from strain that is caused during lying down and getting up. These movements usually
lead to high shear forces at the fracture site and potentially could cause implant failure.
In our experimental setup, these devices would also stabilise the external fixator during
the first five days and also aid to regularly monitor the soft tissue conditions. If not for
the devices, the sheep would have to be caught at least twice a day for regular
monitoring. This poses a high risk of additional injuries, dislocation or breakage of the
soft tissue measurement probes.
Dr Roland Steck designed a new Supporting Trolley System after collating various
possibilities with the currently available animal supporting systems (see Figure 40).
Four supporting trolley system was constructed at the mechanical workshop in O block,
Gardens Point, Queensland University of Technology.
64
Figure 40: Supporting Trolley System with the first sheep in trial based at the animal housing on The Prince Charles Hospital campus.
Description of the Supporting Trolley System
The Supporting Trolley System consists of a stainless steel trolley frame supported
by four wheels capable of 360 degrees of rotation. The trolley frame is provided with
two hinges at the front and rear end. These hinges allow the trolley frame to be folded
inwards or outwards and hence, help in the stability as well as transport of the trolley.
At the top of the metal frame six ring screws are fixed for hanging up six chains. At the
bottom end, a shade cloth is suspended through these six chains (Figure 41). The shade
cloth were designed by the candidate and produced by Advanced Shade Systems,
Capalaba, Australia. The shade cloth bears five holes, four for the legs and the middle
one for excretion purposes. The length of the chains can be adjusted to position the
animal depending on its size and shape. The sheep in the trolley would be able to move
and walk around, if the wheels are unlocked.
65
Figure 41: Shade cloth with five holes.
Sheep trials
The trolleys were transported to the animal facility of The Prince Charles Hospital.
The competency of the complete trolley frame was tested by placing a healthy sheep
overnight under the supervision of Dr Kathy Wilson, Michael Lindeberg and William
Sommers. The behaviour of the sheep was recorded to be normal with no evident
adverse event. For further validation, a second trial with another healthy sheep was
carried out for five days.
Results: The sheep showed normal behaviour with regular food and water
consumption during the entire experimental duration. The daily clinical assessments of
the general condition as well as the state of the skin and wool around the legs
(especially shoulders and thighs) showed a completely healthy animal without any
signs of pressure or abrasions. A weight loss of nearly 600 g was observed, which was
within the expected range, due to the change of environment with a certain amount of
minor stress.
Conclusion
For future experiments, the two pilot sheep will stay for one night in the trolleys
prior to the experiment. This will allow the sheep to get used to the trolleys and their
environment. After the pilot test experiments, the trolley trial will be re-assessed for its
use and efficiency over the main series of experiments. Any necessary refinements or
changes will be applied accordingly.
: This newly designed ‘Supporting Trolley System’ is well tolerated and
efficient to keep the sheep for at least five to seven days regularly monitored, supported
and protected.
66
In general, sheep are shy animals and they normally live in groups. Therefore, we
will keep at least two sheep together in one cage, but only one of them will be
supported in the trolley (Figure 42).
Figure 42: Trolley trial with another ‘free’ sheep in the same cage.
3.3 STUDY DESIGN
The newly developed and validated combined trauma model with the established
soft tissue injury and the multi-fragmentary fracture will now be applied to study the
soft tissue recovery and fracture healing process within a specific area of clinical and
experimental interest. This study will focus on the differences in surgical approach
techniques, and whether the operative procedures influence the healing outcome. To
assess and compare the different surgical approaches such as the minimally invasive
technique and standard open approach, a realistic fracture fixation model is required.
3.3.1 Establishment of fracture fixation model
This trauma model that simulates a high-velocity injury is capable of causing a
significant damage distal femur of the animal leg. In a clinical situation, the treatment
67
regime for equivalent injury to the leg is commonly a two stage procedure: an external
fixator is temporarily applied at the initial acute phase. After a certain period of time,
the definitive fixation procedure is performed by replacing the external fixator with an
internal fixation device to aid the soft tissue recovery. In the case of distal femur
damage, often a plate-screw construct is chosen as definitive fixation method. This
treatment strategy with proven successful clinical outcome will be modified and applied
to the sheep model.
To start with, the right pin and frame configuration was developed and established
to secure a successful early fracture stabilisation. The anatomy of the ovine femur was
studied and landmarks such as the condylar block, femoro-patellar notch and the
concave ‘groove’ between the lateral condyle and the anterior part of the femoro-
patellar notch were defined. From the mechanical point of view, two 4.5mm Schanz’
screws were fixed in each of the two main fragments, mid-proximal shaft and condylar
block. The pins were inserted in latero-medial direction and a unilateral external fixator
construct was created such that all the pins were parallel to each other along with two
longitudinal bars attached with clamps. With lamb femurs and sheep cadaver legs,
several trials were carried out to test the practicability of the external fixator
configuration in relation to the fracture model and the secondary plate fixation
procedures. The following pictures (Figures 43, 44, 45 and 46) illustrate the four steps
in the optimisation process for configuration of the external fixator in a cadaver leg test:
Figure 43: First step
: The four Schanz’ screws were placed in relation to the approach and application of the fracture model. In this case, the figure illustrates a small anterior approach marked in bold blue in longitudinal direction and fracture zone is marked with transverse blue lines. The proximal two pins were positioned parallel and in longitudinal direction and the distal two pins were positioned in a perpendicular plane, slightly convergent to each other.
68
Figure 44: Second step
: After attaching the ‘unilateral external fixator’ to the pins, the fracture creation model was performed via the small anterior approach. The figure also shows the chisel and blade bar being inserted.
Figure 45: Third step
: The percutaneous plate insertion was tested with the support of the locked drill guide at the end of the plate. This external fixator configuration clearly shows, that without removal of the frame, the plate can not be fully inserted and that the longitudinal rod crosses the percutaneous screw insertion. Therefore, the distal condylar block pins have to be re-arranged and the rod must be moved to the anterior region of the pins.
Figure 46: Fourth step
: The last part of the trial concludes with the performance of the open, lateral surgical approach, but as stated in the previous step, the rods and clamps of the external fixator have to be re-positioned or even removed to proceed with the open approach. Distally, this pin configuration is too close to each other and must also be altered, otherwise the plate can not be fixed properly.
69
The final configuration of the external fixator (Figure 47):
• The two distal pins in the condylar block had to be changed into the
perpendicular plane of the shaft axis, because the proximal end of the two pins
was placed too close to the fracture area. This increased the risk for a fracture
extension, and would have compromised the lateral plate fixation of the
secondary, definitive fixation procedure. Therefore, the new position of the two
distal pins was defined as, the anterior one positioned dorsally and parallel to
the anterior femoro-patellar notch in antero-lateral to postero-medial direction,
and the second pin from the lateral condyle into the medial condyle in ‘true’
latero-medial direction.
• The proximal two pins had to be inserted from antero-lateral to postero-medial
direction so that they do not interfere with the plate fixation of the second stage
procedure.
• Because of the new positions of the four pins the configuration of the bars had
to be adjusted. One of the bars was longitudinally fixed with clamps to the three
antero-lateral pins. The bars were placed at the anterior side of the pins.
Through this plating procedure, it is ensured that the minimally invasive
technique is not compromised, while the external fixator frame remains in situ,
until the plate fixation is finalised. The second rod is fixed in oblique direction
to connect the two proximal pins and the posterior lateral condylar pin. This
external fixator frame is constructed in a V-shape and provides sufficient
mechanical stability.
70
Figure 47: Lateral view (left) and axial distal view (right) of the external fixator frame (’V-shape’ type) and the locking compression plate (LCP) fixed onto the antero-lateral distal femur of a plastic ovine bone model.
The second part of the fixation model includes the definitive fracture fixation. This
is executed by the application of the ‘internal fixator’. This type of fixation device
consists of a metal plate with at least two screws on each side of the fracture. The
specialty of such an internal fixator is the locking mechanism between the screw heads
and the plate, which leads to a different construct mechanics with transferring the loads
from one bone fragment, onto the screws, then through the screw-head – plate-hole
interface on the plate. The load crosses the fracture area and gap within the plate and
again, is transferred via the interface between the plate hole and the screw head onto the
screws and finally to the opposite bone fragment. This kind of fixation construct, which
does not directly fix the intermediate fracture zone, is called ‘bridging plate’
configuration.
The requirements of the internal fixator are:
• Appropriate mechanical properties: The construct has to resist all loads of a
sheep hind limb, without risking an implant failure. The chosen implant is the
4.5mm (narrow) stainless steel locking compression plate (LCP) used with
5.0mm head locking screws (HLS).
• Correct length of the plate: The created fracture is approximately 3.5 to 4cm
long, and the screw insertion points have to be in a safe distance from the
fracture site, not to risk further fracture extension with bone splitting. Therefore,
71
the shortest possible plate is a 7-hole 4.5mm LCP. Additionally, this plate is not
too long and will not potentially irritate the lateral soft tissues along the femur.
• Optimal plate-screw configuration: The 7-hole LCP is fitted with two, bicortical
head locking screws (HLS, 5.0mm) in the proximal and the distal main
fragment and three plate holes in the middle are empty (Figure 48). This
configuration satisfies all the above mentioned requirements and provides a safe
and stable implant position.
Figure 48: The internal fixator configuration with a 7-hole narrow 4.5mm, stainless steel LCP with four 5.0mm head locking screws.
3.3.2 Experimental design and description of operative techniques
The research question is to compare the minimally invasive plate osteosynthesis
versus open reduction and internal plate fixation on the soft tissue recovery and fracture
healing. We aimed to address and study this question in an experimental animal model
with sheep. The previously developed and validated trauma model with a multi-
fragmentary distal femur fracture with concomitant severe soft tissue damage is applied
for this purpose
Initially, the external fixator as discussed in the previous chapter is fixed. Within
the following five days, the animals were clinically observed and their soft tissue
conditions were monitored. Following which the sheep undergo the definitive fracture
fixation with an internal fixator (Figure 48). This plate fixation procedure is performed
either with the standard, open surgical approach (ORIF) or by minimally invasive
technique (MIPO). Following this, the external fixator was removed and the sheep were
clinically observed and monitored for nine days. The fracture healing progress was
regularly assessed with biweekly conventional radiographs. Two groups with twelve
sheep each were sacrificed at four and eight weeks, respectively. Post-mortem analysis
72
was carried out by CT scan to assess the callus morphology followed by mechanical
testing to investigate the mechanical integrity of the healing femur bone.
In the following paragraphs, each of the above mentioned procedures is described in
detail, and the following chapters will detail the entire course of the animal experiments
(pilot study and main series), including the aspects of animal care and welfare, as well
as post-mortem investigations and evaluations with all specific aspects and features.
Description of operative techniques:
1) Application of external fixator
The application of external fixator was performed during the fracture model
procedure (see also Chapter 3.2.2). While the sheep was still under induced general
anaesthesia, the sheep was positioned in left lateral recumbent position with the right
leg on top of the operating theatre at the animal laboratory. The leg has been entirely
shorn and shaved before the soft tissue trauma. After cleaning with 0.5% red
Chlorhexidine in 70% Ethanol supplied by the pharmacy of The Prince Charles
Hospital and draping the right leg accessible from the greater trochanter to the tibia
head, the four external fixator pins were inserted by mini-incisions: the two proximal
pins in antero-lateral position parallel to each other, the two distal pins in antero-lateral
position just behind the femoro-patellar notch and in postero-lateral direction into the
lateral condyle of the femur (Figure 47 and 49). Then, the fracture model was applied
over the antero-lateral small approach to the distal, metaphyseal part of the femur.
Before fracture initiation, the longitudinal external fixator rod was fixed to the three
antero-lateral pins and only the distal clamp was loosened. Afterwards, the fracture was
initiated by blade bar insertion and two chisel hits, and the leg was rotated in the
fracture plane at 90° ante flexed and 90° retro flexed positions to complete the
disruption of the periosteum. After this final manoeuvre of the fracture model, the
external fixator frame is applied. Initially, the distally loosened, longitudinal rod was
clicked in the distal clamp again at the same position. After preliminary fracture
reduction by manual adjustment of the two intermediate fragments through the small
antero-lateral approach, the second oblique rod is fixed to the pins (Figure 51). Finally,
:
73
the wounds were washed out and the fracture model approach was closed with sutures
in layers and the pin sites were dressed.
Figure 49: Figure shows the insertion of the four external fixator pins. The two proximal pins at the left are positioned longitudinal to the femur shaft and the two distal pins at the right are placed perpendicular to the proximal once.
Figure 50: Figure showing the small antero-lateral approach employed for the application of the fracture creation model.The oscillating saw was used to carried out the partial anterior osteotomies .
Figure 51: External fixator frame (two carbon rods) on the upper part of the right hind limb of a sheep. The three stitches of the fracture model approach are seen in front of the longitudinal rod. The fixation tool of the probe to monitor the muscle condition such as compartment pressure and partial oxygen tension is placed on top of the proximal clamps.
74
Five days after the trauma model and the external fixator was applied, the sheep
was brought back to the animal laboratory for applying the internal fixator. Internal
fixator is going to be applied by either the open surgical approach or the minimally
invasive method.
2) Operation of the open reduction and internal plate fixation (ORIF)
The sheep was anaesthetised and positioned in left lateral recumbent position on the
operating theatre table. After cleaning with 0.5% red Chlorhexidine and draping the
right upper part of the hind limb, including the external fixator, the rods and clamps of
the external fixator were removed. Then, a standard lateral surgical approach of 15 cm
length to the right femur was performed. After the skin incision, the subcutaneous
layers (Figure 52) and the fascia of the ileo-tibial tract were longitudinally split. The
lateral vastus muscle was mobilised into anterior direction and the perforating blood
vessels were dissected between ligatures. Behind the lateral vastus muscle, the lateral
aspect of the femoral shaft was prepared from the subtrochanteric to condylar region
including visualisation of the fracture area (Figure 53). Initially, the fracture
haematoma was removed and the gaps and cavities were washed out with sterile saline.
Subsequently, the fracture lines -two transverse planes and one longitudinal split were
identified and the fracture margins were cleaned including removal of 5 mm adjacent
periosteum. Under slight axial traction on the pins of the external fixator, the fracture
fragments were anatomically reduced and temporarily held in place with pointed
reduction forceps and small K-wires of 1.25 mm. If the accurate fracture position was
achieved, the internal fixator (7-hole 4.5mm narrow stainless steel Locking
Compression Plate (LCP, Synthes, Switzerland) was then placed on the lateral side of
the femur and the distal end of the plate was centred in the groove anteriorly to the
lateral condyle. Four threaded drill guides were turned into each of the screw holes
before drilling two distal holes followed with two bicortical head locking screws (HLS)
with the correct length inserted (Figure 54). If necessary, any final adjustments to the
fracture position were made. Following which two proximal screw holes were drilled
and two bicortical HLS inserted. All screw heads were locked in the plate with the
torque limited screw driver of 4 Nm. All temporary fixation devices and the external
:
75
fixator pins were removed (Figure 55). The entire wound was thoroughly washed out.
The wound closure was performed in layers, with a continuous suture of the fascia
(Monocryl 2-0), with interrupted sutures of the subcutaneous layers (Vicryl 2-0) and
with interrupted ‘Donati’ stitches of the skin (Novafil 2-0) (Figure 56). After cleaning
thoroughly, the wound was disinfected with a moisture vapour permeable spray
dressing. The spray dressing was bought from Opsite, Smith & Nephew Pty. Ltd.,
Australia.
Figure 52: Conventional lateral open surgical approach: The dissection of the subcutaneous layers is being performed with the ultrasound knife. The pins of the external fixator are still in situ and might be supportive during the fracture reduction process.
Figure 53: The lateral aspect of the femur shaft is being prepared for visualisation by retracting the lateral vastus muscle anteriorly using a metal retractor ‘Langenbeck hook’.
76
Figure 54: LCP is being placed on the lateral aspect of the femur and fixed with screws. The two proximally applied drill guides act as a guide to drill the holes in exact, perpendicular direction to the plate.
Figure 55: The LCP was fixed with four screws (HLS), two proximal and two distal. The two intermediate fracture fragments are shown to be anatomically reduced, but not fixed as supposed to with the bridging plate technique.
Figure 56: The final result of the open surgical approach with many skin sutures in place. The soft tissue monitoring probe shown as the orange cable fixed with the blue caped device is still in situ for post-operative measurements.
77
3) Operation of the minimally invasive plate osteosynthesis
The sheep was anaesthetised and positioned in left lateral recumbent position on the
operating theatre table. After prepping with red Chlorhexidine 0.5% and draping the
right upper part of the hind limb, including the external fixator, a short (2 to 3 cm)
longitudinal skin incision was made between the distal pins of the external fixator at the
lateral femur condyle. The subcutaneous layers and the fascia were longitudinally split
and the periosteum in the ‘lateral groove’ anteriorly on the lateral condyle prepared.
The image intensifier (Philips, BV 25, Netherlands), with a sterile plastic bag covering
the top of the C-arm, was used in the operation field to check the fracture reduction and
control the implant position during the procedure (Figure 60). With a blunt bony
elevator, the epi-periosteal tunnel was prepared without damaging the periosteum. The
fracture region was not visualised during the entire procedure. Further, the 7-hole
4.5mm narrow stainless steel LCP with one threaded drill guide locked in the most
distal plate hole was inserted in a percutaneous fashion (Figure 57 - 61). A second
threaded drill guide was percutaneously inserted through a stab incision of the skin in
lateral projection to the planned plate hole. This drill guide was fixed in the most
proximal plate hole. With the support of those two drill guides and under image
intensifier control, the plate was placed on the femur bone in the correct position in
both radiographic planes (Figure 60). Before fixing the plate to the bone, the fracture
reduction was checked. If the fracture position had to be adjusted, the external fixator
frame was slightly loosened and changed into a better position. Once the fracture and
implant position was correctly established, four bicortical head locking screws were
percutaneously inserted. The screw configuration is the same as with the open
technique. After the final control with the image intensifier, the external fixator was
removed and the distal wound was closed in layers, with a suture of the fascia
(Monocryl 2-0) and a couple of skin stitches (Novafil 2-0). The three stab incisions
were closed with only a skin suture (Novafil 2-0) (Figure 61). The operated area was
cleaned and the wound finally disinfected with a moisture vapour permeable spray
dressing (Opsite, Smith & Nephew Pty. Ltd., Australia).
:
78
Figure 57: The LCP with a threaded drill guide fixed at the most distal plate hole is being slid into the small incision.
Figure 58: LCP is percutaneously advanced passing the fracture region.
Figure 59: LCP is positioned with the distal plate end in the ‘condylar groove’ between the two distal pins of the external fixator.
79
Figure 60: The image intensifier (Philips, BV 25, Netherlands) is used to check and control the fracture reduction as well as the implant position during the whole procedure.
Figure 61: The postoperative clinical picture showing few skin stitches and the soft tissue monitoring probe in situ.
3.3.3 Instruments and implants
The following instruments and implants were used for this study:
• Instruments for the fracture creation model:
o general tools and instruments: scalpel, surgical forceps, scissors and
wound retractors including small ‘Hohmann’ hooks, rulers
o sagittal oscillating saw (Stryker, Stryker 5 system, USA) with oscillating
blade (Stryker, 2108-150, USA) (Figure 30)
o drill (Stryker, Stryker 5 system, USA) with drill bits (2.5 mm and 3.5
mm) and drill guides (Figure 30 and 62)
o blade bar (designed and made by AO Development Institute, Davos,
Switzerland) (Figure 25 and 30) and sliding hammer
80
o surgical chisel (Synthes, 397.970, Switzerland) (Figure 30)
o Kirschner wires (Synthes, 1.25 mm, 292.120, Switzerland)
• Surgical instruments for the operations (Figure 62):
o general surgical tools (scalpel, forceps, scissors, towel clamps, arteries,
retractors, needle holders, suction device)
o drill (Stryker, Stryker 5 system, USA) with 4.3 mm drill bits and
threaded drill guides (4.3 mm)
o screw drivers (large fragment and 4.0 Nm torque limiting device,
Synthes, 324.052, Switzerland)
o pointed reduction forceps
o bony elevator
o Kirschner wires (Synthes, 1.25 mm, 292.120, Switzerland)
• Implants (Figure 47 and 48):
o external fixator (Synthes, large AO external fixator, Switzerland)
carbon fibre rods (Synthes, 11.0 mm diameter, 150 mm length,
394.820, Switzerland)
clamps (Synthes, 393.978, Switzerland)
self-drilling Schanz’ screws (Synthes, 4.0 mm x 100 mm,
294.776 (Titanium), Switzerland)
o internal fixator
7-hole 4.5/5.0 mm narrow, stainless steel LCP (Synthes,
224.571, Switzerland)
5.0 mm self-tapping HLS (Synthes, 213.328-213.346,
Switzerland).
81
Figure 62: General surgical instruments used for the fracture creation model and external fixator application except the implants, the sagittal saw and the drill.
3.4 ANIMAL CARE AND WELFARE
This chapter outlines the exact description of the animals used, their handling and
care as well as the details of the peri-operative and post-operative management.
3.4.1 Sheep data
In this experimental study, male sheep (ovis aries, Merino species) were used.
These sheep were supplied from a local breeder at Lyle Nation, Gympie. These adult
wethers were fully grown with a strong and mature skeleton, weighed around 40 kg and
were about four to six years. Male sheep were chosen because of their bigger size,
heavier weight and stronger bones than the ewes. The sheep were housed in a separate
yard at the Biological Research Facility of The Prince Charles Hospital, Chermside
(Figure 63). The sheep were tested for Q-fever, vaccinated with Glanvac 6 (100 ml im),
treated for worms every six weeks and ear tagged for identification. These protocols
followed the regulations of the Queensland Health and the Department of Primary
Industries (DPI).
82
Figure 63: Sheep flock in the yard at the Biological Research Facility of the Prince Charles Hospital; nowadays, Medical Engineering Research Facility (MERF), Chermside.
3.4.2 Perioperative management
The day before the experiments were conducted, the animals were moved to a
separate area within the animal house including daylight access, for preparation and
adaptation to this new environment. The next morning, the sheep were weighed on the
animal scale for baseline weight measurement and later, transported to the animal
laboratory of Biological Research Facility at The Prince Charles Hospital with a custom
made, large animal trailer. The animals were kept in small in-house cages until the
procedure started.
The animals were caught in the cage by two animal nurses, put into a special trolley and
then, the entire right hind limb from the heel up to greater trochanter and the iliac crest
as well as the neck from the jugular notch to the jaw angle was appropriately shorn and
shaved. After the sheep were moved to the operating theatre room, Dr Kathleen Wilson
or Dr Sadahiro Sugiyama managed the peri-operative care. Under sterile conditions,
one tri-luminal central venous line was inserted into right external jugular vein by
Seldinger technique. Before inducing the general anaesthesia, 10 ml blood was
aspirated for examination of serum concentrations of Creatine Kinase and Lactate
83
Hydrogenase as pre-operative baseline parameters. The general anaesthesia was
induced with propofol (10 mg/ml) 0.5 mg/kg body weight as bolus intravenous
injection. Then, the sheep were endotracheally intubated and ventilated with 100%
oxygen on the ULCO Campbell ventilator (MK.5PC, ULCO Engineering Pty. Ltd.,
Australia). The anaesthesia was maintained with propofol at 4 mg/kg body weight/hour
and buprenorphine (0.01 mg/kg body weight) was also administered intravenously to
provide sufficient analgesia. Before the experimental procedures began, two
intravenous antibiotics (gentamycin 80 mg and cephalothin 1 g) and intra-muscular iron
supplementation (Ferrosig 2 ml) were injected, too. The animals were then carefully
positioned on the pendulum trestle for application of soft tissue trauma to the right
femur after the supra-patellar thigh circumference was measured. Afterwards, the sheep
were re-positioned onto an operating table in left lateral recumbent position and the
multi-fragmentary fracture created and the external fixator applied as described
previously.
Post-operatively, conventional radiographs were taken (see chapter 3.6.1) and then
the anaesthesia discontinued. The animals were extubated and were kept under close
clinical observation until they were directly transported in a special trailer) to the
animal facility. While still in the post-operative recovery stage, the sheep were safely
put into the supporting trolleys and the height of the shade cloths adjusted according to
the size of the animals.
3.4.3 Post-operative management and care
The sheep were looked after and fed by the animal nurses according to the protocol
and standards of the animal facility. Twice a day, the investigators supervised by the
veterinary surgeon carried out clinical assessments of general medical condition as well
as local wound care for the sheep. The pain management included a second dose of
buprenorphine (0.01 mg/kg body weight), and further doses were given if required.
The sheep spent four days closely monitored in the supporting trolleys and on day
5, the animals were again transferred to the animal laboratory for the second procedure.
The induction and maintenance of the general anaesthesia was performed in the same
manner as for the first operation. After the x-rays were taken at the end of the definitive
84
plate fixation procedure, the general anaesthesia was terminated and the sheep were
transported back to the animal facility and put into the supporting trolleys for another
night to fully recover from the operation and anaesthesia. Following which the sheep
were safely released into a separated outside paddock and closely observed by the
animal nurses. Within the following week, further blood samples were taken from the
central venous line, and after following the complete blood sample collection
procedure, the lines were removed on day 9 after the second procedure. The skin
stitches were taken out 14 days post-operatively.
As a precaution, the sheep spent at least two weeks post-operatively in this
separated paddock in a small group of only four to six sheep. After this period of time,
the animals were released back to the flock into the main yard of the animal facility.
3.4.4 Euthanasia
After the defined recovery time (six weeks for the two pilot sheep, and four to eight
weeks for the main series animals, respectively), the sheep were finally weighed and
transported to the animal laboratory. The animals were laid on the special laboratory
trolley and euthanized by an intravenous injection (external jugular vein) with
pentobarbitone sodium (Lethabarb, 325 mg/ml) 0.5 ml/kg body weight.
3.5 METHODS TO MONITOR THE SOFT TISSUE DAMAGE
In this chapter, the methods of measuring the extent of the soft tissue damage and
its recovery are described. The soft tissue damage is monitored in two different ways:
locally via an in situ probe to measure compartment pressure and partial oxygen
pressure and systemically by collecting blood to assess blood serum markers for cell
and soft tissue damage.
3.5.1 Local soft tissue assessments
Compartment pressure, partial oxygen pressure and temperature
To monitor the soft tissue damage and recovery over time, special catheters were
used to measure the compartment pressure and the oxygen tension (partial tissue
oxygen pressure) within a defined area of the soft tissues. The probe (Neurovent PTO,
85
No. 095 008, Raumedic AG, Germany) was inserted in the rectus femoris muscle of the
right hind limb (Figure 64) at the end of the first operation, after the fracture was
created and the temporary external fixation. The probe was percutaneously inserted
from the proximal anterior aspect of the thigh and tunnelled into the central part of the
rectus femoris muscle. The location of the probe was manually controlled from the
small antero-lateral fracture creation approach. The cannulated needle with a
longitudinal split on one side is removed and the catheter at the skin entry point fixed
compartment measurement probe in rectus femoris muscle
lateral surgical approach for open fracture fixation method (ORIF)
Figure 64: This graph shows the localisation and direction of the surgical approaches and the inserted probe in a cross-section through the middle part of the ovine thigh. The red lines indicate the planes of the two surgical approaches and the bold black line presents the direction of the inserted probe.
86
Figure 65: Clinical post-operative site of the proximal right hind limb with the external fixator frame. The figure also shows the sutures, the inserted soft tissue monitoring probe as well as the cables secured in a circle.
The cables of this soft tissue measurement probe are connected to the monitor
(datalogger MPR 2 logO, No. 095 254, Raumedic AG, Germany). Three parameters are
collected twice a day at 7am and 5pm) for the first 14 days. The parameters recorded
are compartment pressure and partial oxygen pressure. To ensure accurate
measurements especially of the compartment pressures, the animals were either
monitored in a calm standing position in the trolleys or in a laying, lateral position on
the special animal trolley under serene circumstances.
Data were collected on paper and later transferred to a Microsoft Excel
database/worksheet.
3.5.2 Systemic investigation:
Blood serum tests
Within the first two weeks after the soft tissue trauma, blood samples were taken to
assess the serum concentration (Units per liter: U/l) of specific cell damage markers,
such as Creatine Kinase (CK) and Lactate Dehydrogenase (LDH). Five ml of blood was
collected through the central venous catheter and directly transported to the Queensland
Health Pathology Laboratory at The Prince Charles Hospital for analysis. The blood
samples were collected before induction of the general anaesthesia as well as post-
operatively, before discontinuing the general anaesthesia. On day 1, 2, 3 and 4 samples
were taken at 7 am. On the day of the second procedure (day 5), pre-operative and post-
87
operative samples were collected just as before. The same procedure for blood sample
collection was repeated on day 6, 7, 8, 9, 11 and 14. All values were copied from the
Pathology Laboratory sheets into a Microsoft Excel table and analysed thereafter.
3.6 METHODS TO ASSESS FRACTURE HEALING
3.6.1 Conventional radiographs
To monitor the progress of the fracture healing over time, conventional radiographs
(x-rays) were taken every fortnight. The first two x-rays were taken post-operatively
after the first and the second operation in the operating theatre (Shimadzu MU 125,
Japan). The latero-medial x-ray was taken with 60 kV and 6 mAs, the antero-posterior
picture with 80 kV and 10 mAs. The x-rays after two, four and six weeks were acquired
in the animal facility with an image intensifier (fluoroscopy; Philips, BV 25,
Netherlands). The latero-medial x-ray (Figure 66) was shot with 40 kV and 3.2 mAs
and the antero-posterior (Figure 67) with 40 kV and 4.0 mAs. The final x-rays after
four and eight weeks, respectively, were captured directly post sacrifice in the animal
laboratory with the Shimadzu device in the same manner as described above. All the
radiographic plates were delivered to the Department of Radiology of The Prince
Charles Hospital and digitally developed and saved as bitmap files (Digitizer, ADC
Compact plus, AGFA, Type 5146, Belgium).
88
Figure 66: Latero-medial x-ray taken using the image intensifier in the animal facility. The sheep positioned in lateral recumbent position on the special large animal trolley. The ‘black’ radiographic plate is placed on the medial, inner side of the femur and the x-ray is taken from the top.
Figure 67: Antero-posterior x-ray taken with the BV 25. The radiographic plate is placed on the back side (posterior) of the leg and the x-ray shot taken from the front.
The first post-operative, digitized radiographs were evaluated qualitatively, with
respect to fracture pattern according to the AO classification of long bone fractures 3
and quantitatively with the computer software AMIRA (Visage Imaging GmbH, Berlin,
Germany). This software was used to determine the mean length of the fracture zone in
the anterior, lateral (middle) and posterior aspect of the lateral view radiographs (Figure
69) and the medial, anterior (middle) and lateral in the a-p view (Figure 68).
89
Figure 68: This screenshot of the AMIRA programme that illustrates the measurements of the fracture length at three areas (left to right: lateral, middle, medial) from the antero-posterior view of the post-operative x-ray after the fracture creation and external fixation procedure.
Figure 69: This cropped AMIRA screenshot from the lateral view of a post-operative x-ray shows posterior, middle and anterior (left to right) fracture length measurements.
To monitor the fracture healing course and determine potential problems, such as
change of fracture, implant failure or re-fractures biweekly x-rays of all sheep were
compared and evaluated. The fractures and implant positions were measured and the
callus morphology was assessed by determining the number of bridging cortices (in
quarters) (Figure 70 and 71).
90
Figure 70: X-rays taken at 4 weeks of Figure 71: X-rays taken at 4 weeks of MIPO group with medially and posteriorly ORIF group with medial and posterior complete callus bridges (2 bridges). callus, but not completely bridged (0). 3.6.2 Computer tomography (CT)
Computer tomograms of all femur bones were taken post-mortem. After euthanasia
of the sheep in the animal laboratory, the entire right femur bone was carefully
explanted (Figure 72) and the implants (LCP plus four screws) removed (Figure 73).
The femur bones were wrapped in saline soaked gauzes and put in two plastic bags.
Afterwards, these bones were transported to the Department of Radiology at The Prince
Charles Hospital and a high-resolution computer tomography scan was performed
(Philips, Brilliance 64, Netherlands: slice thickness 0.67 mm, pixel size 0.625 mm, 120
kV, 250 mAs, rotation time 1 s). After scanning the femora were stored frozen at -20
°C.
91
Figure 72: Right explanted femur without soft tissues, but implant still attached.
Figure 73: Right femur after the implant removal.
The CT data sets of all femur bones were evaluated with the software AMIRA
(Visage Imaging GmbH, Berlin, Germany) and 3D reconstructions generated (Figure
74 left). After determining the density threshold between the cortical bone and the
newly developed, mineralised callus, the callus was visualised and evaluated for its
total callus volume in mm3 and callus length in mm (Figure 74 right). Further
measurements from this dataset included the cross-sectional surfaces in mm2 of the
callus formation in the proximal and the distal transverse fracture planes (Figure 75).
To analyse the fracture reduction and fracture configuration, the anterior fracture
length, the maximal fracture length as well as the maximal distance between the most
displaced fragments were assessed (Figure 76). All the three measurements were
performed in 3D.
92
Figure 74: With the AMIRA software, (left) a three-dimensional histogram is generated and the proximal and distal callus planes are defined. After cropping this middle callus field, (right) the total mineralised callus volume is visualised and measured (mm3
). The longitudinal length of that callus volume defines the total callus length.
Figure 75: This picture, gained with the AMIRA software, shows the areas of the callus formation in the proximal and the distal, transverse fracture planes. Example from the eight week ORIF group.
93
Figure 76: This three-dimensional reconstruction of the cortical bone fragments shows the two intermediate fragments (lateral (left) aligned, medial (right) distally displaced). The anterior fracture length (23.27 mm), the maximal fracture length (25.77 mm) and the maximal distance of fragment displacement (6.15 mm) is measured and displayed.
To more specifically investigate the callus morphology of this relatively complex
fracture pattern, a novel approach was developed to semi-quantitatively analyse the
callus morphology. With a scoring system, the callus morphology, especially the callus
bridges were assessed. This score consists of two different measurements: firstly, we
determined the number of callus bridges that were linking the fracture gaps, and
secondly, the thickness of the callus bridges was measured. A score was given for each
bone according to the following criteria:
Number of callus bridges
• Proximal transverse fracture plane divided into four areas or quarters: anterior,
lateral, posterior and medial areas. Each quarter gets 1 point (maximal 4 points)
:
• Distal transverse fracture plane divided into four areas or quarters: anterior,
lateral, posterior and medial areas. Each quarter gets 1 point (maximal 4 points)
• The proximal and distal fracture quarters are added together
• The score for the number of callus bridges will range from 0 – 8
94
Size of callus bridges
• Proximal and distal callus bridges are counted separately
:
Scoring pattern for callus bridges:
• No callus bridge at all = 0
• All callus bridges with callus thickness < cortical thickness = 1
• Some callus bridges (partially) with callus thickness > cortical thickness = 2
• All callus bridges (complete) with callus thickness > cortical thickness = 3
• Both fracture planes are added together: maximal 6 points
• Callus bridge reaches over proximal and distal fracture plane + 1
• The score for the size of the callus will range from 0 – 7
Total score of callus morphology
• maximal number of bridges x maximal size of callus bridges
:
• The total score will range from 0 – 56
This scoring system was applied into a two- and three-dimensional analysis pattern:
For the two-dimensional method, two longitudinal planes were defined in antero-
posterior and latero-medial direction (Figure 77) and those two slices were generated
and extracted from the 3D-CT dataset (similar to virtual histology slices). Then, the
callus morphology was assessed according to the callus bridging score (Figure 78).
95
Figure 77: With the AMIRA software, two perpendicular planes cut the 3D CT model (red - callus, blue - cortical bone).
Figure 78: (left) Cross-sectional surface in medio-lateral direction and (right) cross-sectional cut in antero-posterior direction. Numbers and quality of callus bridges are assessed according to the morphologic scoring system. This example belongs to one of the four weeks ORIF group with medial and posterior callus formation bridging the fracture gap distally and proximally, respectively.
The three-dimensional method was performed on the entire 3D-CT dataset. In the
AMIRA programme evaluation of numbers and sizes of the callus bridges was carried
out dynamically by rotating and moving the dataset throughout the assessment. The
entire circumference of the callus dataset was divided into four parts: anterior, lateral,
posterior and medial (Figure 79).
96
Figure 79: Clockwise from left top: lateral, anterior, medial and posterior view of a 3D CT-reconstruction. A representative picture of the four weeks MIPO group.
Blinded test approach was used to evaluate all CT scans and the results were
collected and analysed with Microsoft Excel software.
3.6.3 Mechanical testing
The frozen femora were slowly thawed over a period of two days at room
temperature in the biomechanical laboratory at the Institute of Health and Biomedical
Innovation. The proximal and distal ends of the bones were thoroughly dissected to the
cortical surfaces and embedded with polyurethane in two specially designed stainless
steel cups (Figure 80). The left and right femurs were then appropriately positioned and
tested mechanically in the Instron machine (Instron 8874). The testing algorithm
included a first trial run with minimal torsional load. This was followed by the two
defined runs to assess the torsional stiffness and finally, the maximum torque at failure
of the healed femurs (Figure 81). The results were reported as absolute values (Nm) and
as a percentage of the values from the intact contra-lateral side, as determined
according to a procedure previously described 124.
97
Figure 80: Dissected femur proximally and distally embedded and fixed into metallic cups. It was attached to the Instron testing machine to perform the torsional test of stiffness and strength.
Figure 81: This picture shows the spiral fracture in the distal part of the femur shaft (right femur of sheep 418) after performing the mechanical test run to assess the ultimate torque.
98
3.7 STATISTICAL ANALYSIS
All experimental data was saved electronically and analysed using Microsoft Excel
software, and the statistical analysis done using the SPSS (Statistical Package for Social
Sciences) software with the assistance from a bio-statistician (Dr Sarah Whitehouse).
Inter-approach variances in mechanical testing which is normalised to the intact contra-
lateral femur, computer tomography and general sheep data were compared using
independent t-test or Mann-Whitney test. Serum marker levels which were normalised
to pre-operative were compared between approaches using a repeated regression model
analysis with time as repeat parameter. Because of normal distribution of those
parameters, ANOVA test was performed with a statistical significance of p<0.05.
3.7.1 Estimation of sample size and power analysis
In a previous study 1, the tibia treated with the ORIF approach had bending stiffness
of approximately 11 Nm/deg with a standard deviation of 1.0 Nm/deg. Based on these
values, a sample size of 6 in each group (MIPO/ORIF for each survival duration) will
have 80% power to detect a difference in means of 14% (MIPO-ORIF / ORIF) using an
independent t-test with a 0.05 one-sided significance level. In the study of Tepic et al.41
,
the mechanical parameters were normalised to the contra-lateral intact tibia similar to
the present animal model. The tibial fractures treated with an internal fixator in ORIF
manner obtained 64.5% (standard deviation 5.1%) of the intact tibial strength after 12
weeks. Based on these normalised values, a sample size of 6 in each group will have
80% power to detect a difference in means of 12% (MIPO-ORIF / ORIF) using an
independent t-test with a 0.05 one-sided significance level. Although a larger sample
size would have a similar 80% power to detect even smaller differences, the clinical
relevance of such small differences between approaches may be questionable.
3.8 ANIMAL ETHICS DOCUMENTATION
To perform animal studies at Queensland University of Technology, animal ethics
application was submitted to the QUT ethics committee prior to the start of
experiments. On the 13th of September 2005, the animal ethics application was
conditionally granted by the University Animal Ethics Committee (UAEC). Further
99
information and more detailed descriptions of the project were submitted and on the
26th
• Species: Sheep (Ovis aries)
of October 2005, the UAEC decided to grant the project: "Effect of surgical
approach on bone vascularisation, fracture and soft tissue healing by internal fixation:
comparison of less invasive to open approach" the approval to perform animal studies.
Reference number for this application was: QUT Ref No 4222A .The following species
and number of animals were approved:
• Numbers: 2 pilot animals + 36 sheep (main series)
The researchers complied with the policies and guidelines issued by the NHMRC
and AVCC (including the Australian Code of Practice for the care and use of animals
for scientific purposes (7th
Through the stage of further developments within the project, three minor changes
to the approved version of the animal ethics protocol were proposed and therefore, two
additional animal ethics applications were submitted and approved:
Edition 2004).
• intramuscular measurements for monitoring the soft tissue injury and changes in
the postoperative care with the use of special designed trolleys (28th
• soft tissue monitoring using the method of microdialysis (7
February
2006) th
Annual animal ethics progress reports (2005, 2006, 2007, and 2008) were submitted
to the UAEC as part of the standard regulations after which the project was allowed to
be further continued.
July 2006)
3.9 PROCEDURES AND TIMELINE
This chapter summarised the compilation of previously described developments and
methods into one pilot project. The detailed construction and course taken is outlined
with the results presented and discussed. Conclusions were drawn to optimise the main
series of the experimental study and its outcome. Furthermore, both parts of the project
include comprehensive project plans and timelines.
100
3.9.1 Pilot study
The purpose and the aim of the pilot study were to try and assess:
• The newly developed animal trauma model in vivo and throughout the entire
time of soft tissue recovery, which had to be evaluated and defined.
• The application of the animal trauma model to study the effects of surgical
approaches on the soft tissue and fracture healing processes.
• The performance of the experimental study and evaluate the course and
outcomes towards further refinements and adjustments for the main study.
• The overall performance of the animal experiments to report back to the animal
ethics committee.
3.9.1.1 Methodology and timeline
The pilot tests were carried out with two adult, male sheep (Merino species) and the
entire experimental setup was run and tested. As described in the experimental design
of this study, the initial procedures, namely, combined trauma model (soft tissue injury
and fracture creation) and external fixator application were the same for both pilot
sheep. The subsequently operations were split into two experimental methods: the first
pilot sheep underwent the internal plate fixation via the standard open surgical
approach, while the second pilot sheep was fixed via the minimally invasive plate
osteosynthesis technique. After thorough planning and preparation of every step and
part of the experimental setup, the pilot experiments for the first and second sheep were
started on the 13th and 20th
September 2006 respectively. The detailed pilot study
timeline is attached in Appendix H.
3.9.1.2 Results
General data:
Both pilot sheep (Merino wethers) had highly similar characteristics: 7-8 years of
age, and 44.9 and 45.0 kg of body weight. The four general anaesthetics went smoothly
with the postoperative clinical observations performed by the candidate, the veterinary
surgeon and the animal nurses demonstrated no signs of any adverse effects. The day
after the second surgery, the animals were released from the trolleys and kept in the
101
facility for further clinically assessments such as blood sample collections and soft
tissue monitoring. Slippery concrete floor on which both sheep slipped and fell over a
couple of times posed to be the only problem during this protocol. After 14 days of
close observation and monitoring, the sheep were released outside into the paddock of
the animal facility.
Times of the procedures:
The times of the four general anaesthesia and the operations are listed in Table 1.
first procedure second procedure anaesthesia operation anaesthesia operation Sheep 1 (ORIF) 155 min 95 min 130 min 80 min Sheep 2 (MIPO) 135 min 60 min 130 min 50 min
Table 1: General anaesthesia and operation times
Soft tissue trauma evaluation:
The condition of the muscle damage in the anterior compartment of the distal thigh
(rectus femoris muscle) was locally measured and evaluated:
For the first sheep, the compartment pressure measurement (Figure 82) after the
second surgery (ORIF) demonstrated normal values of 0 to 6 mmHg with a peak of 12
mmHg. In the second sheep, the compartment pressures started with higher values of
12-14 mmHg, which decreased to normal values (around 6 mmHg) after the second
intervention.
Compartment pressure measurement
02468
10121416
post 1 2 3 4post 6 7 8 9 10 11 12 13 14
time (days)
mmHg pilot sheep 1pilot sheep 2
Figure 82: This graph presents the compartment pressure values of the two pilot sheep from the rectus femoris muscle compartment measured twice a day over a period of 14 days post trauma.
The second parameter measured in the anterior thigh compartment was the partial
oxygen pressure (Figure 83). In the first pilot sheep, the partial oxygen pressure
102
increased to 80 mmHg within the first 24 hours post-operation and steadily diminished
to about 45 mmHg before the second operation. After the second intervention, the
pressure peaked to 160 mmHg at day 7 and fell back to a constant level of around 70 to
80 mmHg over time. In the second pilot sheep, the starting curve was similar to the
curve for the first sheep, with an increase of up to 70 mmHg and then a steady decrease
again. The values of the second sheep showed only a small second increase (60 mmHg)
after the second operation and distinct fall to about 10 mmHg from day 9 onwards.
Partial oxygen pressure (pO2) measurement
020406080
100120140160180
post 1 2 3 4post 6 7 8 9 10 11 12 13 14
time (days)
mmHg pilot sheep 1pilot sheep 2
Figure 83: This graph shows the measurements of the partial oxygen pressures of two pilot sheep in the anterior muscle compartment of the thigh.
On the other hand, to monitor the soft tissue damage systemically, daily blood
samples were collected and assessed for Creatine Kinase (CK) and Lactate
Dehydrogenase (LDH) serum levels:
The CK levels of pilot sheep 1 (Figure 84) showed a distinct peak (~1550 U/l) on
the first postoperative day and a recovery to normal levels within the following two
days. After the second operation, there was another increase (~450 U/l) on day 6 with a
full recovery over the next day. Pilot sheep 2 presented a similar pattern, but both peaks
on day 1 and day 6 were lower than those of pilot sheep 1. An additional short peak
occurred on day 4, with the level reaching normal the day after.
103
Creatine Kinase (CK) serum levels
0200400600800
1000120014001600
pre post 1 2 3 4pre post 6 7 8 9 11 14
time (days)
U/L pilot sheep 1pilot sheep 2
Figure 84: This graph shows the serum Creatine Kinase levels of both pilot sheep over the first two weeks of the experiments.
The pattern of the LDH serum concentration curve was similar to those of CK
curves (Figures 84 and 85). The postoperative LDH concentration (day 1) of pilot sheep
1 showed a higher peak than pilot sheep 2. Over the following days, the LDH levels
recovered and presented a second smaller peak (day 6) after the second procedures (day
5) with a quicker recovery of the LDH levels by pilot sheep 2.
Lactate Dehydrogenase (LDH) serum levels
300
400
500
600
700
800
900
pre post 1 2 3 4pre post 6 7 8 9 11 14
days
U/L pilot sheep 1pilot sheep 2
Figure 85: This graph presents the serum Lactate Dehydrogenase levels of both pilot sheep over the first 14 days of the experiments.
Fracture healing evaluation:
Conventional radiographs:
The post-operative x-rays for Pilot sheep 1 (Fig. 86/87) showed a good implant
position which was slightly anterior in the distal condylar fragment. However, the
implant was not perpendicular to the distal screws, which is a non-ideal screw position,
resulting in potentially non-locked screw heads in the two distal plate holes. The
fracture reduction was acceptable, with a slight varus (3°) and ante curved (2°) position.
104
Despite the successful open fracture reduction, the intermediate fragments were not
reduced. Within the first four weeks of recovery, the fracture position changed with an
increased varus (12°) and ante curve (20°) misalignment, due to the non-locked
application of the distal screws. However, between Week 2 and Week 4, the visible,
mineralised callus formation progressed distinctly and by Week 6, the medial and
posterior hard callus was bridging the fracture area.
Post-op 1 Post-op 2 2 weeks 4 weeks 6 weeks
Figure 86: Conventional radiograph series of recovering fracture in Pilot sheep 1 (antero-posterior view).
Postop 1 Postop 2 2 weeks 4 weeks 6 weeks
Figure 87: Conventional radiograph series of recovering fracture in Pilot sheep 1 (lateral view).
The postoperative x-rays following the first and second procedures in Pilot sheep 2
(Fig. 88 and 89) presented an almost anatomical reduction of the fracture. The internal
fixator consisting of a bridging plate was placed at a distance of approximately 1.5 cm
105
away from the femur. The most proximal screw was slightly angled, but otherwise the
positioning of the fixation device was appropriate. After two weeks of recovery,
beginning callus formation was already visible both medially and on the posterior. At
Week 2, the position of the plate and the fracture also remained in good alignment. On
the other hand, a new ‘cortical shadow’ was detected retrospectively, proximal to the
internal fixator at this time point. Secondly, a proximal, subtrochanteric fracture was
identified. Over the following four weeks, the callus formation progressed in size and
density, and after six weeks of recovery, the fracture area appeared to be bridged on the
medial, lateral and posterior side.
Post-op 1 Post-op 2 2 weeks 4 weeks 6 weeks
Figure 88: Conventional radiograph series of Pilot sheep 2 (antero-posterior view).
Postop 1 Postop 2 2 weeks 4 weeks 6 weeks
Figure 89: Conventional radiograph (x-ray) series of Pilot sheep 2 (lateral view).
106
Computer tomography
After six weeks of recovery, the two pilot animals were sacrificed following the
removal of the femur explants containing the plate and screw implants. These were
wrapped in soaked saline gauzes, and directly transported to the Department of
Radiology of The Prince Charles Hospital where they were scanned using a high-
The Pilot sheep 1 (Fig. 90) which was operated with the standard open surgical
approach, presented a large callus mass and the integration of the two intermediate
cortical fragments together with what appears to be the beginning of a callus bridging.
The regions with the most callus formation were posterior, medial and lateral of the
femur. The least amount of callus formation was found on the antero-medial side.
Anterior Middle Posterior
Figure 90: Coronal reconstruction of CT scan of Pilot sheep 1.
Pilot sheep 2 (Fig. 91) which was fixed with the minimally invasive technique,
showed less callus formation and a reduced fracture position, especially in terms of the
two intermediate fragments as compared to Pilot sheep 1. In terms of the fracture
107
healing progress, some bridging callus was detected in Pilot sheep 2 as well. The
second proximal femur fracture which was initially suspected was confirmed, and
appeared clearly on the sagittal reconstruction of the CT scan (Fig. 91). Also around
this fracture site, callus formation was apparent on the scan indicating that this fracture
occurred at the earlier stage of the trial (prior to Week 2).
Figure 91: Sagittal reconstruction of CT scan of Pilot sheep 2. The accidental, second displaced fracture is clearly visible proximal to the most proximal screw hole.
Mechanical testing:
After six weeks of recovery, the relative torsional stiffness which is defined as the
percentage of the intact contra-lateral femur was smaller in Pilot sheep 1 (ORIF;
36.0%) than in Pilot sheep 2 (MIPO; 60.6%). The relative ultimate torque was also
smaller in Sheep 1 (22.6%) than in Sheep 2 (56.2%). The absolute values of both pilot
sheep are listed in Table 2.
Torsional stiffness Nm/Deg Ultimate torque Nm Femur side: Right (fractured) Left (intact) Right (fractured) Left (intact) Sheep 1 (ORIF) 0.18 0.51 14.2 62.7 Sheep 2 (MIPO) 0.39 0.65 35.4 63.1
Table 2: Mechanical test results of Pilot sheep 1 and 2.
108
3.9.1.3 Discussion and Conclusion
The purpose of the pilot study which used two sheep as experimental models was to
examine the entire experimental setup and course of recovery of fractures created in the
femurs of sheep. In terms of animal recovery, the pilot tests were successful as both
sheep underwent the two-stage procedure and recovered successfully afterwards. The
animal behaviour appeared normal and any weight lost within the first week post-
procedure was subsequently regained. After the first two to three weeks, the gait of the
sheep returned to an almost normal pattern with the right hind limb able to support the
full weight bearing on it. At the site of the procedure, no adverse events occurred and
the wounds appeared to be healing well.
The use of sheep femur fractures as a model to study trauma and fixation
demonstrated an overall good outcome in terms of its ability to examine the process of
recovery. However, there was a mix of both constants and fluctuations among the
parameters measured. All anaesthetic and surgical operations were performed safely
and as planned. The soft tissue trauma showed elevated values of compartment
pressures but without any signs of compartment syndrome. High post-traumatic peaks
of the serum markers CK and LDH as an indication of severe soft tissue damage were
also not observed. The fracture model was very consistent with an AO C-type fracture
with two intermediate fragments in both animals. After the second operation,
monitoring of the soft tissue showed alterations from the first procedure. However, this
data was limited after the damage of one of the compartment probes. These events
demonstrate the importance of careful positioning and handling of these delicate and
fragile probes and cables. The external fixator application worked well as a temporary
fixation method. The position of the Schanz’ pins proved to be crucial for the second
procedure as the location of the pins may affect the positioning of the plate screws.
Another point of consideration in this minimally invasive procedure is to ensure the
complete insertion of the external fixator frame, and the positioning of the internal
fixator. Also, leaving the frame in place until the end of the plate fixation is vital. The
second operations which were either open or minimally invasive were successfully
performed without wound or fracture healing complications.
109
The position and the reduction of the fracture fragments were different, which is
due to the bridging osteosynthesis. From the clinical perspective, this difference is
acceptable. However, as part of the experimental model, the reduction of the axes and
rotation should be improved. Another important factor of the minimally invasive plate
fixation procedure is the plate position. It should be as closely as possible located to the
bone as when done using the open technique.
The recovery of the soft tissue damage was, as expected, occurred within the first
10 to 14 days of the experiments. This indicated that no changes to the original methods
used were necessary. The fortnightly conventional radiographs presented a useful visual
sequence of events occurring during the fracture healing process. These radiographs
indicated significant callus formation from week 4 onwards. The results of the
computer tomography and the mechanical testing confirmed these findings by showing
a similar progress in fracture healing. The computer tomography revealed a
morphologically clear callus formation while experimental data on the mechanical
integrity of the callus indicated the strength of the healed fracture.
These results were presented and discussed in a video-conference meeting with our
international collaborators, Prof Keita Ito, Romano Matthys, Peter Toggwiler and Dr
Simon Pearce from the AO Research and Development Institute, Davos, Switzerland.
From this meeting, it was concluded that the trauma model produces a reproducible
severe soft tissue damage and multi-fragmentary distal femur shaft fracture. To treat the
damage and examine the processes involved in healing, the two-stage fixation model
was deemed suitable. Due to the advanced fracture healing progress seen after six
weeks of recovery time and to investigate differences in the early stages of recovery
which may lead to different healing outcomes, the recovery times of the two
experimental groups was changed from six to four weeks and from twelve to eight
weeks.
3.9.2 Main test series
After thorough planning and preparation including alterations to the main study, the
test series was initiated in November, 2006. The entire timetable of the main series is
shown (Appendix I). The first experiments were conducted on groups of two or four
110
animals over a course of eight weeks and were completed in April, 2007. The next set
of experiments was carried out on the animals of in the four weeks recovery group and
data was collected by October, 2007.
CCHHAAPPTTEERR 44:: RREESSUULLTTSS
4.1 CHARACTERISATION OF TRAUMA
4.1.1 Soft tissue trauma recovery
Results of local, intra-compartmental soft tissue monitoring
In a group of 16 sheep, the intra-compartmental probe was routinely inserted at the
end of the first procedure after the external fixator application was completed. The
immediately following the operation of 6.2 ± 6.0 mmHg (mean value ± SD) within the
rectus femoris muscle. These values remained below the threshold of a pressure reading
indicating the onset of a compartment syndrome. The elevated pressure decreased over
the next five days with the lowest value measured on day 5 of 1.2 ± 1.6 mmHg.
Compartment pressure (cp) measurements (n=16)
0
2
4
6
8
10
12
14
2 16 26 40 50 64 74 88 98 112
time (hours postop)
mmHg
Figure 92: Mean values of the intra-compartmental pressures of 16 animals over a time span of five days. Critical compartment pressures (> 20 mmHg) were not reached.
Partial oxygen pressures (Fig. 93) could only be measured in seven of the 16
animals, due to defects in the probe catheter or cables which were inserted in these
animals. The results showed slightly elevated partial oxygen pressure within the first
two days after trauma, with the highest value of 69.8 ± 24.3 mmHg (mean ± SD)
111
measured after 48 hours. After this peak, the partial oxygen pressure decreased
continuously to reach final values of 44.8 ± 27.9 mmHg five days after the procedure.
Partial oxygen pressure (pO2) measurements (n=7)
0102030405060708090
100
2 16 26 40 50 64 74 88 98 112time (hours postop)
mmHg
Figure 93: Partial oxygen pressure monitoring within the rectus femoris muscle of the right thigh following operations. Due to damage to measuring equipment, only seven animals remained for evaluation. The mean values (n=7) and the standard deviations are listed.
Results of systemic soft tissue monitoring by serum markers
In all the 24 animals included in this study, daily blood samples were taken from
the central venous line over five days. The evaluation of the serum levels of CK and
LDH, as markers of cellular break down, yielded peak values of both CK and LDH one
day after trauma and surgery, with 1140.2 ± 672.0 U/L and 627.9 ± 130.6 U/L,
respectively. These values decreased over the following two days back to
physiologically normal serum levels (Fig. 94).
112
Creatine Kinase (CK) and Lactate Hydrogenase (LDH) serum levels (n=24)
0
500
1000
1500
2000
pre-op post-op 1 2 3 4 5time (days)
U/L
LDH
CK
Figure 94: Mean values (n=24) and their standard deviations of the CK and LDH serum levels over a period of five days.
4.1.2 Fracture model
The fracture creation technique was successfully performed in vivo in 24 sheep
through a small (3 cm) antero-lateral, trans-muscular approach. The conventional post-
operative x-ray radiographs showing both the antero-posterior and latero-medial aspects
of the fracture as well as the evaluation of computer tomography data confirmed the
AO C3-type fracture pattern in all of the 24 fractures (Fig. 95/96). Further analysis
based on 3D-CT scan data showed 21 fractures (87.5%), where the middle segment was
divided into two fragments as intended (Fig. 95) and only three fractures (12.5%),
where the middle segment was divided into three fragments (Fig. 96). The average
length of the fracture zone, measured on the post-operative x-rays in both planes (Fig.
68/69) was 30.1 ± 2.7 mm. Fracture lengths (Appendix J) showed consistency across all
fractures. The measurement of the reduction angles of the bone axes in both planes
(Fig. 97) showed slightly varus position of 3.2 ± 5.5 degrees and a minor ante-curved
position of 2.6 ± 7.0 degrees. All 24 distal femur fractures were stabilised using the
same V-shaped external fixator construct (Fig. 51) without any complications observed
clinically during the five days.
113
Figure 95: These post-operative x-rays show the fracture configuration (AO C-type with two intermediate fragments), the aligned fracture reduction as well as the appropriate positioning of the external fixator.
Figure 96: The three-dimensional CT reconstructions of the cortical fragments illustrates the multi-fragmentary fracture pattern (AO C-type) with three intermediate fragments, the smallest fragment located in the proximal posterior region. The images are from the antero-lateral and medial perspectives respectively.
114
Figure 97: These two post-operative x-ray radiographs processed with AMIRA show the angles of alignment in both axes. On the left, a varus position of 7 degrees, and the x-ray on the right shows a slight retro-curved position of 3 degrees.
4.2 MINIMALLY INVASIVE VERSUS OPEN PLATE OSTEOSYNTHESIS
4.2.1 General results
The 24 sheep used in this experimental study had a mean age of 5.4 years, a mean
starting body weight of 39.1 kg and a mean final body weight of 37.9 kg. The thigh
circumference of the completely shorn and shaved right hind limbs was an average of
32.0 cm. The detailed physical database is listed in Appendix K.
All the procedures conducted had no adverse effects on the animals. Also, during
the soft tissue trauma and the postoperative aftercare, there were no clinical signs of
any distress or discomfort caused by pain. All surgical interventions were performed by
the same surgeon. The operation time for the minimally invasive technique of 51 ± 16
minutes was significantly (p<0.005) shorter compared to the open surgical method of
76 ± 16 minutes. For the first six days, the sheep remained in the supporting trolleys,
which caused some distress and sleeping problems in the animals. Despite an
appropriate food and water intake during the initial post-operative period, all the
animals lost approximately 10% of their body weight. Overall however, every sheep
stayed in general healthy condition. Regular observations of the animals’ behaviour in
115
the outside paddock showed that most of the animals started to use the operated hind
limb within the first two weeks with touch or even partial weight bearing. After
approximately four weeks, the sheep appeared to be fully weight bearing and the
injured leg fully functional.
From the experience of the pilot study with a fall and the occurrence of the
proximal femur fracture, precautions were undertaken to prevent the animals from
slipping and falling on the slippery concrete floor of the animal facility. Nevertheless,
four animals had significant falls and sustained fractures and implant failures, which
were confirmed radiologically (Fig. 98). These animals were sacrificed immediately
and replaced. One animal from the 4 weeks group, which had been operated with the
minimally invasive method, developed a wound infection at one of the proximal pin
sites after the second operation. After regular washing and disinfection of the wound,
the infection appeared to have healed. The post-mortem examination via dissection and
explanation of the femur revealed no clinical signs of any ongoing soft tissue
infections. However, osteolysis around the proximal holes of the external fixator as
well as of the internal fixator was seen. The decision was made to exclude this sheep
from further evaluations.
Figure 98: X-rays of a sheep which sustained a fall on the first post-operative day confirmed a suspected fracture extension with proximal implant failure.
116
4.2.2 Soft tissue recovery
For the first 14 days of the experiments, the effect of the soft tissue injury and its
recovery were monitored locally using the compartment measurement probes and
systemically via the analysis of serum markers. Blood for the serum marker analysis
was collected from all sheep for the entire duration of the experimentation period. All
16 compartment measurement sensors were operational after insertion during surgery.
Within the first seven days, two of the minimally invasive group and five of the open
group lost the partial oxygen pressure probes or cables due to technical failure. Due to
issues concerning reproducibility and small sample sizes, analysis of the partial oxygen
pressure measurement was therefore not performed.
The compartment pressures recovered to normal levels after peaking following the
first procedure and presented a rise after the second operation (Fig. 99). The second
peak of the compartment pressure tended to be higher in the ORIF group (9.8 ± 10.0
mmHg, p = 0.08) than in the MIPO group (3.3 ± 4.5 mmHg). Over the next one and
half days the compartment pressure fell to normal levels below 5 mmHg.
Intra-compartmental pressure measurement (n=8)
0
2
4
6
8
10
postop
11 /
am1 /
pm2 /
am2 /
pm3 /
am3 /
pm4 /
am4 /
pm
preop 2
postop
26 /
am6 /
pm7 /
am7 /
pm8 /
am8 /
pm9 /
am9 /
pm
10 / a
m
10 / p
m
11 / a
m
11 / p
m
12 / a
m
12 / p
m
13 / a
m
13 / p
m
14 / a
m
14 / p
m
time (postop days)
mmHg
MIPO
ORIFpost 1st procedure: trauma and external fixation
post 2nd operation: plate fixation
Figure 99: Intra-compartmental pressures over the entire soft tissue monitoring time of 14 days. After the second, randomised operation (MIPO and ORIF), the peak of the ORIF group is higher (p=0.08) than the MIPO group.
The evaluation of the systemic soft tissue monitoring by serum markers during the
first 14 days following the procedures showed two peaks, one following the initial
operation and creation of the fracture and soft tissue trauma, the second occurring
following the second operation to replace the external fixator with the internal fixation
device (Fig. 100/101).
117
The serum Creatine Kinase (CK) levels on the first post-operative day after the
second procedure were significantly higher (p = 0.01) in the sheep operated with ORIF
approach (1029 ± 827 U/L) compared to MIPO (286 ± 260 U/L). Serum levels of CK
returned to normal levels (<200 U/L) on the 9th
Creatine Kinase serum level
0
500
1000
1500
pre post 1 2 3 4 pre post 6 7 8 9 11 14
Time (days)
Cre
atin
e K
inas
e co
ncen
trat
ion
(U/L
)
ORIF
MIPO
*p < 0.05
post 2nd operation: plate fixation
post 1st procedure: trauma and external fixation
post-operative day in both groups.
Figure 100: The Creatine Kinase serum levels (n=12) show two peaks, after the trauma and external fixator application as well as after the second operation for definitive plate fixation. The second peak is significantly higher with the ORIF group compared to the MIPO group.
A similar trend was observed for the serum LDH levels. On the first (ORIF 617 ±
179 vs. MIPO 484 ± 91 U/L), second (ORIF 559 ± 197 vs. MIPO 408 ± 81 U/L) and
third (ORIF 483 ± 98 vs. MIPO 397 ± 78 U/L) postoperative days following the second
operation the LDH levels were significantly higher (p < 0.05) in the ORIF group.
post 2nd operation: plate fixationpost 1st procedure: trauma and external fixation
Figure 101: Serum LDH concentrations (n=12) over a period of the first 14 days. The first peak indicates the soft tissue damage after the first procedure (trauma and external fixator application). The second peak follows the second operation of the randomised plate fixation: the first three days post-operatively (day 6, 7 and 8) the LDH levels are significantly higher in the ORIF group than in the MIPO group.
4.2.3 Fracture healing
Conventional radiography
The progress of the fracture healing was monitored with bi-weekly conventional
radiographs taken in two planes, perpendicular to each other (antero-posterior and
latero-medial). The postoperative x-rays after the second operation showed good
reduction and alignment of the two main fragments with appropriate implant positions.
The fracture reduction of the middle fragments was better aligned with the open method
(ORIF) as compared to the minimally invasive technique. The analysis of the
fortnightly x-rays over the eight week period of time showed the beginning of callus
formation at week 2. Further expansion of the calcified callus areas including the
formation of callus bridges was apparent at week 4. At week 6, further calcification
with more callus formation was evident and finally at week 8, callus formation had
progressed and showed clear signs of callus bridges. These observations are confirmed
by the assessment of the number of callus bridges on radiographs (Fig. 102). At all time
points there is the trend of more progressed callus formation across the fracture gaps for
the MIPO group at 4 weeks. The difference of callus formation in this group as
119
compared to the ORIF group is statistically significant (p = 0.006). Two representative
x-rays from each experimental group are attached in Appendix L.
Number of callus bridges from conventional radiographs
0
1
2
3
4
2 weeks 4 weeks 6 weeks 8 weeks
MIPO
ORIF
Figure 102: Callus bridges at each of the fortnightly time points from the conventional radiographs in both planes. At four weeks recovery time, MIPO group presents significantly higher callus bridges than the ORIF group (p = 0.006).
Computer tomography (CT scan)
The 3D-reconstructions of the post-mortem CT scans were analysed for differences
in the fracture configuration and the callus morphology. The analysis of the callus
morphology was divided into three parts: the callus volumes and lengths, the cross-
sectional callus areas through the proximal and distal fracture planes as well as the
morphological callus bridging score measured in two and three dimensions.
The evaluation of the fracture configuration, in terms of differences in the fracture
reduction, showed no significant differences in the fracture length, neither the maximal,
3D fracture length (ORIF 35.8 ± 9.1 mm (mean ± SD) and MIPO 31.1 ± 2.8 mm) nor
the anterior fracture length from the proximal to the distal anterior osteotomy (ORIF
27.6 ± 4.9 mm and MIPO 25.7 ± 2.7 mm). The measurements of the maximal 3D
fracture displacement, however, highlighted a significantly larger distance in the
minimally invasive group with 9.8 ± 2.2 mm (p=0.04) as compared to the open
reduction method with 7.5 ± 2.9 mm.
Callus morphology between both surgical approaches was measured using the total
mineralised callus volume and the total callus length (Table 3). At both time points, 4
p<0.05
120
and 8 weeks, similar mineralised callus volumes were determined in both groups.
Similarly, no significant differences in callus length were observed at either time point.
Callus volume (mm3 ORIF ) MIPO P Value
4 weeks 19100 ± 8500 17400 ± 4800 p = 0.70
8 weeks 24400 ± 8500 20700 ± 5700 p = 0.39
Callus length (mm)
4 weeks 89.7 ± 14.0 80.5 ± 15.5 p = 0.33
8 weeks 72.9 ± 10.1 76.7 ± 7.2 p = 0.47
Table 3: Total mineralised callus volumes and callus lengths determined by 3D-reconstructions of the images from the CT scans at 4 and 8 weeks post-operatively.
The measurements of the cross-sectional callus areas were made in the proximal as
well as in the distal fracture plane. At 4 weeks, the callus areas of the MIPO group were
higher than those at 8 weeks for the ORIF group (Table 4) but none of those results
were statistically significant (p>0.05).
Callus area proximal (mm2 ORIF ) MIPO P Value
4 weeks 256.5 ± 151.9 337.3 ± 142.9 p = 0.39
8 weeks 538.6 ± 92.8 490.4 ± 140.7 p = 0.50
Callus area distal (mm2 )
4 weeks 390.0 ± 139.7 437.6 ± 106.2 p = 0.54
8 weeks 559.3 ± 159.7 508.2 ± 77.2 p = 0.50
Table 4: Cross-sectional callus area in the proximal and distal fracture plane of both groups and at both time points.
The semi-quantitative callus bridging score indicated statistically significantly
higher callus bridging rates at the 4 weeks time for the MIPO group (38.2 ± 8.9 (SD);
p=0.007 with the 3D method and 27.6 ± 15.3; p=0.039 with the 2D method) compared
to the ORIF group (20.0 ± 8.6 with the 3D technique and 10.7 ± 7.4 with the 2D
technique). At 8 weeks of recovery, this difference of the callus bridging score was not
statistically significant (p>0.05).
121
Mechanical Testing (Figure 103)
Mechanical tests were conducted to obtain absolute and relative measures of the
mechanical properties including torsional rigidity and ultimate torque at failure. After 4
weeks, the absolute torsional stiffness was significantly higher in the MIPO group (0.21
± 0.06 Nm (mean ± SD) with p=0.006) as compared to the ORIF group (0.06 ± 0.07
Nm). The absolute ultimate torque also presented higher values for the MIPO group
(12.7 ± 8.1 Nm; p=0.15) than for the ORIF group (5.8 ± 6.3 Nm).
After 8 weeks, the absolute torsional stiffness showed similar values of both groups,
slightly higher for the minimally invasive approach (MIPO 0.62 ± 0.12 Nm/deg (mean
± SD; p=0.46) and ORIF 0.56 ± 0.15 Nm/deg). The absolute torsional strength was 37.6
± 10.0 Nm/deg for MIPO group and 37.2 ± 11.9 Nm/deg for the ORIF group.
After 4 weeks, the relative torsional rigidity of the fractured femur as a percentage
of the intact contra-lateral bone was significantly higher (p=0.018) in the MIPO group
(30.1 ± 10.6%) compared to the ORIF group (9.8 ± 12.4%). The mean torsional
moment at failure was higher (p=0.11) in the MIPO group (20.7 ± 11.0% versus 9.6 ±
9.8%).
After 8 weeks, the relative torsional rigidity tended to be higher (p=0.36) in the
MIPO group (130.4 ± 61.9%) compared to the ORIF group (100.4 ± 41.4%). The
torsional moment at failure after 8 weeks also tended to be higher (p=0.26) in the MIPO
group (79.0 ± 23.6 vs. 64.4 ± 18.3%).
122
Mechanical test results
64.4
9.6
100.4
9.830.1
20.7
129.4
79.0
4 weeks torsional rigidity
8 weeks torsional rigidity
4 weeks ultimate torque
8 weeks ultimate torque
% o
f int
act c
ontr
a-la
tera
l fem
ur ORIF
MIPO
Figure 103: The mechanical properties of torsional rigidity and ultimate torque in relation (percentages) to the intact, contra-lateral femur are higher in both timelines with MIPO than in ORIF technique. After four weeks, the MIPO group shows significantly stiffer mechanical properties than the ORIF group (p=0.018).
Correlation analysis
Any correlations between the mechanical test results and the morphological
findings of the callus formation from the CT scans were then analysed. Firstly, the
relationship between the results of the absolute torsional stiffness and the proximal
cross-sectional callus area at 4 weeks was found to be reasonable with R²=0.43 (Figure
104). The correlation between the distal cross-sectional callus area however was
considerably smaller with R²=0.27.
Correlation between mechanical property and proximal callus area at 4 weeks
y = 960.37x + 156.54R2 = 0.4254
0100
200300400
500600
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
torsional stiffness (Nm/deg)
prox
imal
cal
lus
area
(m
m2 )
Figure 104: This graph shows the good correlation (R²=0.43) between the absolute torsional stiffness (Nm/deg) and the cross-sectional callus area of the proximal fracture zone of all eleven animals at four weeks.
p<0.05
123
Secondly, the relationship between the mechanical properties and the
morphological callus bridging score (3D data) showed a high correlation at 4 weeks
with R²=0.79 (Fig. 105) and a reasonable correlation at 8 weeks with R²=0.35 (Fig.
106).
Correlation between mechanical property and callus morphology score at 4 weeks
Figure 106: Correlation between the relative ultimate torque at failure and the 3D callus bridging score of all twelve animals at eight weeks observation time.
124
CCHHAAPPTTEERR 55:: DDIISSCCUUSSSSIIOONN
5.1 TRAUMA MODEL
The aim of the first part of this project was to create a suitable and reliable trauma
model to enable us to perform experimental studies on large animals (in this case
sheep) and to examine and answer clinically relevant research questions in the field of
orthopaedic trauma. A specific trauma model provided the inspiration towards this
doctoral project. A car bumper hit to a pedestrian leg causes a severe injury to the limb.
This type of injury contains severe soft tissue damage as well as a complex fracture.
Therefore, to investigate the trauma model further, the region of interest was defined to
be the distal part of the thigh and femur. Based on the information from previous
research studies and pilot experiments, an all-in-one trauma model to create such a
complex injury was abandoned. Consequently, the trauma model was developed,
validated and implemented in two separate parts: the severe soft tissue damage model
classified according to Tscherne 2 to be of grade III and the multi-fragmentary fracture
model classified according as AO type C-fracture 3
Many challenges had to be overcome throughout the development stage of both the
models. The validation process involved the replicating each newly developed step
several times to ensure reliability and consistency of the results. Further adjustments to
both the components of the trauma model were made as required. This finally led to the
following optimised protocol. Under general anaesthesia, the combined trauma was
achieved in one single session, with approximately 20 minutes required for the entire
procedure. The pendulum impact device created the desired degree of severity of soft
tissue injury in a reproducible manner. The consistency of the experiments was
confirmed by post-mortem macroscopic dissection and post-mortem CT scans, as well
as the consistent increase in serum breakdown markers (CK and LDH). The appropriate
positioning of the sheep’s femur in the device is simple and guarantees that the injury is
limited to the soft tissues only. The steps for creating the fracture consisted of partial
osteotomies, drill holes, blade bar insertion and chisel hits to initiate the fracture lines.
. Finally, in the in vivo experiments,
both parts were carried out in sequence and hence, collectively turned out to be the
novel, combined trauma model.
125
All these procedures were carried out with considerable consistency and created AO
C3-type fractures 3
. All these steps could be performed through a small, trans-muscular
approach, thereby minimising further iatrogenic damage to the soft tissues. Finally, all
fractures could be stabilised subsequently with the same configuration of the external
fixator frame, as well as internal fixator construct, resulting in a comparable mechanical
environment for all animals, which further minimises variability during the course of
fracture healing.
Soft tissue damage and recovery
Local monitoring of the soft tissue damage and its recovery with the compartment
probe clearly showed changes of the compartment pressure in the traumatised rectus
femoris muscle. The compartment pressure peaked post-operatively within a few hours
up to 6 mmHg (maximally 12 mmHg), despite a ‘mini’-fasciotomy performed during
the small approach for the fracture creation. This corresponds with a moderate damage
and muscle swelling. Over the following two days the compartment pressure recovered,
while the partial oxygen pressure reached its highest value in this period. These
physiological changes may be explained by the hyper-perfusion reaction of the
inflammation phase during the early recovery of the muscle damage. This observation
is consistent with the measurements by Epari et al. (2008) 114, the only published report
on partial oxygen pressure in a fracture healing study to date. They used the same
sensor to monitor hydrostatic pressure, partial oxygen pressure and temperature, in a
simple, tibia osteotomy model in sheep. In their experiments, measurements could only
be taken from seven of the 16 implanted sensors, whilst the other sensors were
damaged during the experimental course. These circumstances including the small
sample size and the high variations of the values from day three onwards, did not allow
them to draw further conclusions. The challenges of using this highly sensitive sensor
in sheep have been discussed by Epari et al. (2008) 114
The systemic soft tissue monitoring with serum markers for tissue breakdown
started with pre-operative baseline values. Some of those values were above the normal
level, which may be explained by pre-operative physical stress and soft tissue
contusions caused by falls during the preparation stage. All animals showed the highest
.
126
CK and LDH levels on the first post-operative day. These peak values reached 2610 U/l
for CK (normal range from 50 to 180 U/l) and 1010 U/l for LDH (normal range from
150 to 400 U/l), which reflects significant, severe soft tissue and muscle damage,
especially with this localised injury. The analysis of daily blood tests showed steadily
decreasing values and all animals including those with high serum marker
concentrations recovered within three to four days. This short recovery time highlights
the healthy general condition of the sheep. However, as the macroscopic dissection in
the ex vivo experiments and the visual functional observation of the animals in the in
vivo experiments confirmed, the injury was limited to the anatomical structure of the
muscles only. Gait analysis or weight bearing studies were not performed, however, the
limb function recovered well from the initial reduced weight bearing and limping. At
any time, all injured limbs were adequately perfused and moved, ruling out any damage
to the vital structures, such as the main blood vessels and the nerves.
The amount of muscle mass in the area of trauma which was measured by the
circumference of the distal thigh, did not vary significantly between the individual
animals (32 ± 1.1 cm (mean ± SD), range from 30.5 to 34.5 cm) or between the
experimental groups. The serum markers, LDH and CK, to measure the muscle damage
presented a relatively wide distribution of the values. This outlines a certain limitation
for the use of these serum markers to quantify muscle damage. These findings are
supported by the clinical experience with the inter-individual variability of those
markers. To determine and validate the amount of muscle damage morphological
assessments such as in vivo CT scans with intravenous contrast, in vivo MRI and early
histological investigations, could have been included. The use of these methods was
evaluated, but unfortunately they were not accessible for this study for logistical
reasons.
The newly developed soft tissue trauma device was successfully implemented and
used in this in vivo ovine study. Severe soft tissue injury was achieved without any
local or general complication, such as nerve damage, main blood vessel damage, the
development of a compartment syndrome or a crush syndrome with renal failure. The
severity of the damage was consistent within a range of physiological and inter-
individual variability and reflected quite realistically a clinical scenario.
127
Fracture creation model
The primary goal was to create a realistic and natural multi-fragmentary fracture
with a consistent pattern. In order to strike a balance between reproducibility and close
similarity to clinical, high impact fractures with multiple fragments, the fracture
configuration was defined to be an AO C-type 3 pattern with two intermediate
fragments. High forces are required to break cortical bone in sheep, especially in certain
anatomical areas, such as the anterior cortex of the distal femur. On the other hand,
ovine bones tend to split, mainly in longitudinal direction, which can cause
inconsistency in the length of the fractures. This could potentially result in differences
in mechanical conditions of the fracture site as well as manifest as problems to stabilise
them with the same mechanical construct. These challenges were overcome by a few
methods as detailed below. Firstly, bones were weakened in pre-determined areas. Then
the fracture lines were initiated in a sequential manner as published by Baumgaertel et
al. (1994) 39. Partial anterior osteotomies were required to weaken the strong anterior
cortical bone of the distal femur. This procedure was performed carefully by cooling
the blade and the bone edges with continuous rinsing with sterile saline. This
precaution was undertaken due to the possibility of the development of heat necrosis at
the adjacent bone fragments. However, a disadvantage of this fracture model was the
creation of smooth fracture lines at the osteotomy sites. These smooth osteotomy sites
behaved mechanically different to the spiky fragments observed in clinical fractures
with more contact surfaces. All procedures carried out towards bone weakening and
fracture initiation were performed through a small, consistent trans-muscular surgical
approach. This procedure did not replicate the most common clinical scenario with
either an open compound or a closed blunt fracture mechanism. The open approach
thereby influenced the development of the compartment pressure by releasing, at least
partially, the intra-muscular pressure by performing a ‘mini-fasciotomy’. However,
high-impact injuries often result in lacerations of the soft tissue envelope which would
have the same effect 6
The analysis of the fracture configuration of the 24 in vivo fractures showed a high
consistency in fracture classification with 100% AO C3-type fractures, in fracture
pattern and fracture length. 87.5% of fractures were with two intermediate cortical
.
128
fragments. The fracture lengths were with constant values of 30.1 ± 2.7 mm (mean ±
SD). Additionally all fractures could be stabilised with the same implant configuration
(external fixator and internal fixator). This provided comparable mechanical conditions
between the specimens and emphasized the reliability of the achieved fracture model.
5.2 MINIMALLY INVASIVE VERSUS OPEN PLATE OSTEOSYNTHESIS
Recently, there has been a trend in clinical practice towards less invasive surgical
procedures that preserves the integrity of the tissues. This way, the biological potential
for optimal healing conditions is maximised. This ‘damage reduction’ concept has also
been applied to the operative treatment of fractures. The aim of minimally invasive
plate osteosynthesis is to minimise the iatrogenic damage including any disturbance to
the vascularity of the fracture zone and the surrounding soft tissues. From the clinical
experience, this new technique seems to contribute to an advanced fracture healing in
terms of shorter recovery times with less pain and discomfort for the patient. There are
also fewer complications that occur such as infections and delayed-unions. Despite
these promising experiences, there is a lack of solid clinical and experimental evidence
supporting the advantages for MIPO 1; 76. Furthermore, there are concerns that the
perceived benefits of MIPO may be outweighed by the demanding surgical technique,
the possibility for mal-reduction and mal-alignment as well as the increased exposure to
radiation during visualisation 63
This animal model was designed to replicate a clinical situation of a multi-
fragmentary fracture and associated severe soft tissue injury that typically results from
a high-energy accident. The treatment modalities were chosen according to clinical
practices. The initial treatment of this severe combined injury involved the preliminary
fracture reduction and stabilisation with the application of an external fixator frame.
The external fixator immobilised the injured area and aided the recovery process of the
soft tissues. In our study, this recovery could be demonstrated by the restoration of
normal levels of local and systemic parameters after an initial increase. After this
. The goal of the present study was therefore to compare
the healing course of fractures treated either with a standard open surgical approach or
by the minimally invasive plate osteosynthesis technique, using the newly developed
trauma model in sheep.
129
recovery period of four days, the second surgical procedure with fracture reduction,
randomised, definitive internal plate fixation, as well as removal of the external fixator
was carried out safely. This minimised the risk of reaching critical inflammation levels
of the soft tissues.
The demographic data of the sheep were normally distributed in terms of
differences in age, body weight and muscle mass of the right thigh. All operative
procedures were performed without any intra-operative complications. Despite the
more demanding percutaneous technique, the duration of the minimally invasive
surgery was significantly shorter. This difference in operation time could be even larger
in the clinical environment, where the image intensifier is also routinely used in the
ORIF method and not just the MIPO technique and further prolongs the surgery time.
Shorter procedure times could potentially impact the patient’s morbidity in terms of a
slightly lower peri-operative complication risk and lower rates of wound infections,
especially if the operation exceeds the two hour time span. Furthermore, shorter use of
the operating theatre could have a positive logistical and health-economical impact.
All except one sheep presented uneventful recoveries with adequate wound healing
and timely return of limb function and full weight-bearing status. The infected sheep
presented a wound infection at one of the proximal external fixator pin sites. The site of
infection was not related to either of the randomised procedures and the sheep was
consequently excluded from the study. Such a pin track infection especially during the
hot and humid weather conditions of Brisbane was not a surprise. However of
particular relevance, this event did not influence the outcome of the study.
Following the second operation, the enzyme levels (CK and LDH) were distinctly
different between the two groups. In the ORIF group, both enzyme levels peaked at a
similar magnitude to that caused by the soft tissue injury. In contrast, only a small
increase was determined in the MIPO group. In addition, higher values of compartment
pressure occurred following internal fixation with the open approach, although in both
groups the measured hydrostatic pressures in the tissue were below those indicating in a
potential compartment syndrome (< 20 mmHg). These in vivo measurements of intra-
muscular pressure and the serum markers associated with tissue breakdown clearly
demonstrate that a higher degree of tissue injury is coupled to the open surgical
130
approach. This finding was certainly expected from the clinical experience, however,
the question that remains to be answered is whether these ‘biological’ differences
around the fractured bone significantly influence the fracture healing outcome or not.
The course and outcome of fracture healing was assessed with monitoring the callus
progress by fortnightly x-rays, post-mortem morphological evaluations based on CT
scans as well as post-mortem mechanical testing to determine the mechanical properties
of the healed femurs. The healing of the femur was monitored at two defined time
points, four and eight weeks post-operatively. To sum it up, in this severe animal
trauma model, the fracture healing showed advanced progress with significant
differences in the early stage (four weeks) and smaller differences at the later stage
(eight weeks) for the minimally invasive plate osteosynthesis.
The evaluation of the biweekly conventional radiographs (x-rays) taken in two
perpendicular planes, presented at all time points further progressed callus healing with
more connecting callus bridges (significantly higher value after four weeks) in the
MIPO group compared to the ORIF group. A limitation of the analysis and
interpretation of conventional radiographs originates from the two-dimensional
projections of the three-dimensional fracture callus. This represents a summation of the
mineralised areas, which may potentially cover and hide non-calcified regions.
Furthermore, the analysis is highly dependent on the quality of the pictures and the
exact leg position during the radiographic procedure. In the presented series, the quality
of the radiographs was assured with constant radiation levels and a standardised
positioning of the image intensifier and the sheep hind limb. Nevertheless, some intra-
and inter-individual differences with respect to rotation and angulation in both planes
(antero-posterior and latero-medial) could not be avoided.
Quantitative analysis of the CT data was performed to specifically investigate the
callus morphology. No significant differences were found in the total mineralised callus
volume, total callus length and the proximal and distal cross-sectional callus areas
through the two transverse fracture planes. This may be partly explained by the large
variability in these measures, which is common and attributable to inter-individual
differences, as well as due to the complex fracture pattern used in this trauma model.
However, the total mineralised callus volume and length tended to be larger in the
131
ORIF group, while both cross-sectional callus areas showed higher values after four
weeks for the MIPO group and after eight weeks for the ORIF group. These results
indicate that the quality of callus, in terms of callus distribution, location of new bone
formation, mineral density of the callus organisation of the callus and possibly, even
early signs of callus remodelling must have been superior in the MIPO group to explain
the greater mechanical competence determined in that group. The results of the semi-
quantitative callus morphology analysis with the callus bridging score attempted to
capture some of those previously mentioned characteristics. The significantly higher
bridging score at four weeks for the MIPO technique with high correlation to the
mechanical properties (torsional stiffness) supported the advanced functional integrity
of the new bone formation from the morphological side.
The fracture healing was further quantified by determining the mechanical
competence assessed by torsional stiffness and torsional moment at failure of the
healing femurs. At the early time point of four weeks, a significantly higher torsional
stiffness was found in the group that underwent the minimally invasive plating
procedure. The torsional strength was also higher in this group, but not statistically
significant. Chehade et al. (1997) 125 investigated the relationship between stiffness and
strength changes during fracture healing and its clinical implications. They concluded
that the testing of stiffness is valuable primarily in assessing the progress towards
fracture union, while the ultimate strength more accurately describes the recovery of the
mechanical integrity until the fracture is clinically healed and the bone is fully
functional. Taking these findings into account, the significant difference in torsional
stiffness in the early healing phase gives evidence of a more advanced fracture healing
progress with MIPO. This enhanced mechanical competence could be explained by the
callus which has matured with more areas of highly mineralised callus bridges and
better positioned callus sites linking the fracture gaps. This observation is supported by
the higher callus bridging score of the CT scan evaluation. Especially the three-
dimensional analysis showed a more homogenous callus and more mineralised callus
bridges linking the cortical fragments for the bones of the MIPO group. From the
clinical point of view, at this stage of fracture healing the focus lies on the
demonstration of callus formation and maturation, indicating the successful progression
132
of the fracture healing towards union of the fracture fragments. The strength of the
fracture callus at this stage is secondary, as the bone is mechanically protected by the
internal fixator. Therefore, in light of the consistent mechanical conditions for the
bones of both groups as designed in this study, the increased stiffness in the healing
bones of the MIPO group at four weeks indicates an advanced healing rate that may
lead to an earlier fracture union. However, at eight weeks, the torsional stiffness and the
ultimate torque at failure tended to be higher in the MIPO group, but were no longer
significant. It has been demonstrated that once the fracture fragments are linked
together, the differences in the mechanical integrity are expected to be smaller (Schell
et al. (2005) 122
While the superior mechanical competence of the femurs that were stabilised by
means of MIPO can be explained partly through morphological advantages of the callus
of the healing bones, there are several possible explanations as to why the healing was
improved in this group. From the experimental design with exactly the same implant
construct, no difference in the overall mechanical stability was introduced. However,
differences were found in the degree of the reduction of the intermediate bone
fragments, with larger maximal displacement in the MIPO group. This may have
caused more micro-movements within the fracture zone and potentially increased the
mechanical stimuli for callus induction. Furthermore, the accelerated fracture healing
associated with the minimally invasive procedures may be attributed to a better
biological environment with less disruption of the bone vascularity and less irritation of
the healing process. As part of the open surgical approach, the fracture fragments were
cleaned and the fracture haematoma was washed out. It is possible that this procedure
caused a setback to the fracture healing process through possible renewal of the fracture
haematoma formation and inflammatory processes. This could have delayed the healing
process in comparison to the minimally invasive technique.
. At eight weeks, all bones in our study showed solid bridging of the
fracture fragments, which might explain the smaller differences of the mechanical
properties observed at that time.
In the single previous experimental study 1 performed to compare the less invasive
technique with the open approach, no differences in healing were found. That study was
carried out on sheep tibiae and the difference in the anatomical locations of the two
133
studies may have been the cause for the inability to detect a difference in healing.
Because the tibiae have minimal soft tissue coverage on the antero-medial aspect, the
bone can be approached just by a simple longitudinal, sharp dissection of the skin and
the subcutaneous layers without the need for a trans-muscular approach. Hence, even
the open approach would have caused minimal damage to the surrounding soft tissues.
To address this clinically relevant problem, it was necessary to use a suitable
experimental animal model. Hence, this study was developed in a novel trauma model
to mimic the conditions of a significant impact causing both a severe soft tissue injury
and a multi-fragmentary fracture. Whilst every effort was made to perform this in a
reproducible manner, there is more inherent variability in such a model as compared to
a simple osteotomy model studying bone healing. This variability made the analysis of
callus morphology more challenging. However, the benefits of a clinically relevant
trauma model outweighed this disadvantage.
Before discussing the clinical relevance of these results, it is pertinent to firstly
highlight the following exposition: The surgical procedures carried out in the present
study were performed on healthy, middle-aged sheep. Healing in sheep, in spite of their
early and occasionally excessive weight bearing, is observed to be very robust as
indicated by the extent that one must go to in creating models of non-union 126
This experimental, animal study and the previously discussed outcomes can lead
into the following possible outcomes in relation to clinical perspective and practice: In
general, soft tissue management and the biological handling of fracture care influences
the fracture healing process. Consequently, the soft tissues adjacent to and around the
fracture areas are to be treated cautiously to minimise additional, iatrogenic damage in
both, open and minimally invasive, techniques of fracture management. Fractures fixed
by MIPO may unite in a shorter timeframe than the ORIF, especially in anatomical
regions with larger surrounding soft tissues, such as the femur or the humerus. A faster
.
Therefore, it may be interpreted that the deficit of the blood supply to surrounding
tissues caused by open surgical approaches may be more pronounced in the clinical
situation than observed here, particularly in cases where a deficit in the biological
healing potential already exists through a pre-existing medical condition, such as
diabetes, smoking, vascular diseases or just through old age.
134
recovery from these injuries would definitely benefit the individual patients, but also,
when viewed at a larger scale, have further socioeconomic implications, with faster
interfragmentary movements and promotes healing after unreamed nailing. Study of
a displaced osteotomy model in sheep tibiae. J Bone Joint Surg Am 87: 2028-2037.
122. Schell, H, Epari, DR, Kassi, JP, et al. 2005. The course of bone healing is
influenced by the initial shear fixation stability. J Orthop Res 23: 1022-1028.
123. Hente, R, Cordey, J, Rahn, BA, et al. 1999. Fracture healing of the sheep tibia
treated using a unilateral external fixator. Comparison of static and dynamic
fixation. Injury 30 Suppl 1: A44-51.
124. Klein, P, Schell, H, Streitparth, F, et al. 2003. The initial phase of fracture healing
is specifically sensitive to mechanical conditions. J Orthop Res 21: 662-669.
125. Chehade, MJ, Pohl, AP, Pearcy, MJ, Nawana, N. 1997. Clinical implications of
stiffness and strength changes in fracture healing. J Bone Joint Surg Br 79: 9-12.
126. Schell, H, Thompson, MS, Bail, HJ, et al. 2008. Mechanical induction of critically
delayed bone healing in sheep: radiological and biomechanical results. J Biomech
41: 3066-3072.
149
AAPPPPEENNDDIICCEESS
APPENDIX A:
List of publications (2005 – 2009)
Schmutz B, Wullschleger ME, Kim H, Noser H, Schuetz MA: Fit assessment of anatomic plates for the distal medial tibia. J Orthop Trauma, 22:258-263, 2008. Reichert JC, Saifzadeh S, Wullschleger ME, Epari DR, Schuetz MA, Duda GN, Schell H, van Griensven M, Redl H, Hutmacher DW: The challenge of large animal models for segmental bone defect research. Biomaterials 30:2149-2163, 2009. Lutton C, Sugiyama S, Wullschleger ME, Williams R, Cambpell J, Crawford R, Goss B: Transplanted abdominal granulation tissue induced bone formation – an in vivo study in sheep. Connective Tissue Research, accepted in February 2009. Nijboer JMM, Wullschleger ME, Nielsen SE, McNamee AM, Lefering R, ten Duis HJ, Schuetz MA: A comparison of severely injured trauma patients admitted to level one trauma centers in Queensland and Germany. ANZ J Surg, accepted in March 2009. Reichert JC, Epari DR, Wullschleger ME, Saifzadeh S, Steck R, Lienau J, Schuetz MA, Duda GN, Hutmacher DW: Establishment of a Preclinical Ovine Model for Tibial Segmental Bone Defect Repair by Applying Bone Tissue Engineering Strategies. Tissue Engineering, accepted in September 2009. Chen G, Schmutz B, Wullschleger ME, Pearcy M, Schuetz MA: Computational Investigations of Mechanical Failures of Internal Plate Fixation. J Engineering in Medicine, accepted in October 2009. Schmutz B, Rathnayaka K, Wullschleger ME, Meek J, Schuetz MA: Quantitative fit assessment of tibial nail designs using 3D computer modeling. Injury, accepted in October 2009. Schmutz B, Wullschleger ME, Meek J, Noser H, Barry M, Schuetz MA: Fit optimization of a Distal Medial Tibia Plate, submitted to Computer Methods in Biomechanics and Biomedical Engineering in March 2009. Thomas C, Athanasiov A, Wullschleger ME, Schuetz MA: Current concepts in tibial plateau fractures, submitted to Acta Chir Orthop Trauma Cechoslovaca in August 2009.
150
Wullschleger ME, Steck R, Matthys R, Webster J, Wilson K, Ito K, Schuetz MA: Ovine model allowing the study of bone fracture healing with closed, severe soft tissue trauma, rejected by J Orthop Res in September 2009, in revision process.
APPENDIX B:
List of conference papers (2005 - 2009)
Sommer C, Jenni R, Bircher HP, Wullschleger ME: Minimally invasive plate osteosynthesis (MIPO) with locking compression plate (LCP) in proximal diaphyseal humerus fractures. Swiss knife (special edition):10, 2005. Wullschleger ME, Sommer C: Experience with the locking compression plate (LCP) in fracture treatment of osteoporotic bone. The bone and joint decade, Multidisciplinary Research Day, Brisbane, November 2005. Wilson CJ, Pettet GJ, Chen G, Mishra SK, Steck R, Wullschleger ME, Schuetz MA: Is callus formation optimised for fracture stability? A computational study. The 10th
Conference of the International Society for Fracture Repair, Adelaide, May 2006.
Wullschleger ME, Jenni R, Walliser M, Schuetz MA, Sommer C: Minimally invasive plate osteosynthesis (MIPO) with locking compression plate (LCP) in proximal diaphyseal humerus fractures. The 10th
Conference of the International Society for Fracture Repair, Adelaide, May 2006.
Wullschleger ME, Walliser M, Jenni R, Schuetz MA, Sommer C: Distal tibia shaft fractures: First results of minimally invasive plate osteosynthesis. The 10th
Conference of the International Society for Fracture Repair, Adelaide, May 2006.
Sommer C, Walliser M, Jenni R, Wullschleger ME: Distal tibia shaft fractures: 5 years follow-up of minimally invasive plate osteosynthesis. Annual Conference of Swiss Surgical Society, Lugano, June 2006. Wullschleger ME, Schmeling A, Steck R, Ito K, Schuetz M: Comparison of fracture healing treated with an internal fixator or plate in an ovine tibial model, Australian & New Zealand Orthopaedic Research Society (ANZORS), 12th
Annual Scientific Meeting, Canberra, October 2006.
Gongfa C, Wilson CJ, Wullschleger ME, McElwain DLS, Mishra SK, Pearcy MJ, Perren SM, Pettet GJ, Steck R, Schuetz, MA: Modelling the effects of bone fragment contact in fracture healing, ANZORS, 12th
Annual Scientific Meeting, Canberra, October 2006.
Schmutz B, Wullschleger ME, Schuetz MA: The effect of CT slice spacing on the geometry of 3D models, 6th
Biennial Australasian Biomechanics Conference, Auckland, February 2007.
151
Goss B, Sugiyama S, Wullschleger ME, Lutton C: Transplanted abdominal granulation tissue induced bone formation – an in vivo study in sheep, Spine Society of Australia Conference, Hobart, April 2007. Stillhardt P, Wullschleger ME, Furrer M, Sommer C: Management of peripheral arterial injuries, EATES conference, Graz, Austria, May 2007 (Poster). Nielsen S, Lefering R, Wullschleger ME, McNamee A, Sarai H, Carter L, Davies T, Aitkin L, Schuetz MA: Benchmarking the Trauma care in an Australian Hospital against an European Trauma Registry, Swan XY Trauma Conference, Sydney, July 2007. Wullschleger ME, Steck R, Matthys R, Toggwiler P, Wilson K, Ito K, Schuetz MA: A novel sheep model for the experimental study of severe trauma to the distal femur, 67th
AOA Scientific Meeting, Gold Coast, October 2007.
Tung JP, Fung L, Wullschleger ME, Wood P, Wilson K, Fraser J: A comparison of the oxidative responses of ovine and human neutrophils, HAA 2007 Conference, combined annual scientific meeting, Gold Coast, October 2007. Wullschleger ME, Webster J, Freeman A, Sugiyama S, Steck R, Schuetz MA: Soft tissue evaluation in a sheep model comparing minimally invasive versus open plate osteosynthesis, ANZORS, 13th Annual Scientific Meeting, Auckland, October 2007. Schmutz B, Wullschleger ME, Steck R, Schuetz MA: Optimising a periarticular fracture fixation plate: Does one shape fit all? ANZORS, 13th Annual Scientific Meeting, Auckland, October 2007. Mishra S, Collier L, Chen G, Steck R, Schmutz B, Wullschleger ME, Schuetz MA: The effect of distraction vector orientation on the regenerate tissue during bilateral mandibular distraction, ANZORS, 13th Annual Scientific Meeting, Auckland, October 2007. Klaus A, Schmutz B, Wullschleger ME, Schuetz MA, Steck R: Quantification of fracture callus volume from CT scans, ANZORS, 13th Annual Scientific Meeting, Auckland, October 2007. Schmutz B, Wullschleger ME, Schuetz MA: Optimising a periarticular fracture fixation plate: Does one shape fit all?, 6th
Combined Meeting of the Orthopaedic Research Societies, Hawaii, October 2007.
Schmutz B, Wullschleger ME, Schuetz MA: Optimising the fit of a periarticular fixation plate, Third Asian Pacific Conference on Biomechanics, AP Biomech 2007, Tokyo, November 2007.
152
Wullschleger ME, Webster J, Freeman A, Steck R, Schuetz MA: Soft tissue evaluation in a sheep model comparing minimally invasive versus open plate osteosynthesis, IHBI Postgraduate Student Research Conference, Brisbane, November 2007. Nijboer A, Wullschleger ME, Nielsen S, ten Duis HJ, Schuetz MA: Benchmarking trauma care performance of a tertiary hospital in Queensland to European Trauma Centres: Using the German Trauma Registry as a model, Austrauma, Sydney, February 2007. Lutton C, Sadahiro S, Wullschleger ME, Goss B: Transplanted abdominal granulation tissue induced bone formation – an in vivo study in sheep, Orthopaedic Research Society, San Francisco, March 2008. Gongfa C, Schmutz B, Steck R, Pearcy M, Wullschleger ME, Wilson C, Schuetz MA: Predicting the fatigue life of internal fracture fixation plates, 16th
Congress of European Society of Biomechanics, Luzern, July 2008.
Steck R, Gregory L, Minehara H, Wullschleger ME, Schuetz MA: Three-dimensional visualisation of callus geometry in new murine fracture model. ISFR’s 11th
Biennial Conference, Lake Tahoe, USA, July 2008 (poster).
Wullschleger ME, Nielsen SE, Nijboer A, McNamee A, Lefering R, Schuetz MA: Secondary referred trauma patients: Comparison to those with primary admissions to a tertiary hospital in South-East Queensland, SWAN conference, Sydney, July 2008. Schmutz B, Wullschleger ME, Steck R, Schuetz MA: Periarticular implants – optimisation of plate design with computer modelling, 68th
Annual scientific meeting of Australian Orthopaedic Association, Hobart, October 2008.
Wullschleger ME, Steck R, Wilson K, Ito K, Schuetz MA: Minimally invasive plate osteosynthesis: advanced early fracture healing in an ovine trauma model, 68th
Annual scientific meeting of Australian Orthopaedic Association, Hobart, October 2008.
Schuetz MA, Nijboer A, McNamee A, Wullschleger ME, Nielsen S: Development of a comprehensive trauma system in Queensland – Benchmarking trauma care performance in a tertiary hospital in Queensland to European trauma centres, 68th
Annual scientific meeting of Australian Orthopaedic Association, Hobart, October 2008.
Nijboer A, Wullschleger ME, Nielsen S, Duis ten HJ, Schuetz MA: Benchmarking trauma care performance of a tertiary hospital in Queensland to European Trauma Centres: using the German Trauma Registry as a model, German Orthopaedic and Trauma Conference (DGU), Berlin, October 2008. Sugiyama S, Goss B, Wullschleger ME, Wilson K, Williams R: Reliability of clinical measurement for assessing spinal fusion: an experimental sheep study, ANZORS, 14th Annual Scientific Meeting, Brisbane, November 2008 (poster).
153
Wullschleger ME, Schmutz B, Ito K, Steck R, Schuetz MA: Minimally invasive versus open plate osteosynthesis: Quantitative radiographic analysis of callus morphology of distal femur fractures in sheep using computed tomography imaging, ANZORS, 14th Annual Scientific Meeting, Brisbane, November 2008 (poster). Steck R, Gregory L, Minihara H, Wullschleger ME, Schuetz MA: Three-dimensional visualisation of callus geometry in new murine fracture model, ANZORS, 14th
Annual Scientific Meeting, Brisbane, November 2008.
King B, Wullschleger ME, Mishra S, Schuetz MA, Schmutz B: Volumetric analysis of the space between bone surfaces and fracture fixation plates using 3D computer models, ANZORS, 14th Annual Scientific Meeting, Brisbane, November 2008 (poster). Wullschleger ME, Steck R, Webster J, Freeman A, Ito K, Schuetz MA: Minimally invasive plate fixation leads to advanced early fracture healing in a sheep trauma model, IHBIinspires Conference, Gold Coast, December 2008. Schuetz MA, Wullschleger ME: Major Pelvic Trauma, Austrauma, Sydney, February 2009. Schmutz B, Wullschleger ME, Meek J, Schuetz MA: Optimized Fitting of Precontoured Implants for Fracture Care, 55th
Annual Meeting of Orthopaedic Research Society (ORS), Las Vegas, USA, February 2009.
Wullschleger ME, Steck R, Webster J, Wilson K, Ito K, Schuetz: Minimally invasive plate osteosynthesis accelerates early fracture healing in an ovine trauma model, 55th
Annual Meeting of Orthopaedic Research Society (ORS), Las Vegas, USA, February 2009.
Wullschleger ME, Steck R, Webster J, Wilson K, Ito K, Schuetz MA: Minimally invasive plate osteosynthesis: advanced early fracture healing in a sheep trauma model, 78th Annual Scientific Congress of Royal Australasian College of Surgeons, Brisbane, May 2009; published in ANZ J Surg 79.1:86, 2009. 1st
prize for free paper of Trauma sessions.
Louie B, Wullschleger ME, Mundy J, Schuetz MA, O’Rourke TR: Outcomes of emergency thoracotomies in an Australian Trauma Centre, 78th
Annual Scientific Congress of Royal Australasian College of Surgeons, Brisbane, May 2009.
Broszczak D, Wullschleger ME, Steck R, Upton Z, Parker T: A novel method to assess extra-cellular fluid using SELDI-TOF-MS, ASMR Queensland Postgraduate Student Conference / IHBI inspires conference, Brisbane, May / November 2009.
154
Wullschleger ME, Steck R, Webster J, Wilson K, Ito K, Schuetz MA: Minimally invasive plate osteosynthesis: advanced early fracture healing in a sheep trauma model, Queensland AOA Annual Meeting, Noosa, June 2009. Schmutz B, Rathnayaka K, Wullschleger ME, Meek J, Schuetz MA: Quantitative fit assessment of tibial nail designs using 3D computer modeling, Queensland AOA Annual Meeting, Noosa, June 2009. Wullschleger ME, Steck R, Webster J, Wilson K, Ito K, Schuetz MA: Minimally invasive plate osteosynthesis: advanced early fracture healing in a sheep trauma model, 96th
Annual scientific meeting of Swiss Society of Surgery (SGC), Montreux, Switzerland, June 2009 (Swiss knife, special edition, p25, 2009).
Sugiyama S, Goss B, Wullschleger ME, Wilson K, Williams R: Reliability of clinical measurement for assessing spinal fusion: an experimental sheep study, Global Spine Congress, San Francisco, USA, June 2009. Wullschleger ME, Steck R, Webster J, Wilson K, Ito K, Schuetz MA: Minimally invasive plate osteosynthesis: advanced early fracture healing in a sheep trauma model, 73rd
Annual scientific meeting of German Orthopaedic & Trauma Association (DGU), Berlin, Germany, October 2009 (poster).
Schmutz B, Rathnayaka K, Wullschleger ME, Meek J, Schuetz MA: Quantitative fit assessment of tibial nail designs using 3D computer modeling, 73rd
Annual scientific meeting of German Orthopaedic & Trauma Association (DGU), Berlin, Germany, October 2009.
Steck R, Ueno M, Gregory L, Wullschleger ME, Schuetz MA, Itoman M: Histological characterization of healing of murine femoral fractures stabilized with locking plates with two different stiffnesses, 73rd
Annual scientific meeting of German Orthopaedic & Trauma Association (DGU), Berlin, Germany, October 2009 (poster).
Wullschleger ME, Steck R, Webster J, Wilson K, Ito K, Schuetz MA: Minimally invasive plate osteosynthesis: advanced early fracture healing in a sheep trauma model. 25th
Annual Meeting of Orthopaedic Trauma Association (OTA), San Diego, USA, October 2009.
Schmutz B, Rathnayaka K, Wullschleger ME, Meek J, Schuetz MA: Quantitative fit assessment of tibial nail designs using 3D computer modeling, accepted for 69th
Annual Scientific meeting of Australian Orthopaedic Association, Cairns, October 2009.
Steck R, Gregory L, Minehara H, Wullschleger ME, Schuetz MA: Histological characterisation of a new murine fracture model with locking plate stabilisation, 69th
Annual Scientific meeting of Australian Orthopaedic Association, Cairns, October 2009.
155
APPENDIX C:
AO Research Grant approval document 05-W17 (20 June 2005)
156
APPENDIX D:
QUT Animal project approval certificate 4222A (26 October 2005)
157
APPENDIX E:
Flowchart of trauma model algorithm (June 2005)
158
APPENDIX F:
Detailed results of macroscopic dissection from trials of the soft tissue trauma
model with the pendulum device:
Sheep 346:
Right leg (comminuted distal femur fracture, Figure 1):