Experimental
		
31. Double-dose cefuroxime concentrations in bone, knee joint, and subcutaneous adipose tissue – A randomized porcine microdialysis study 
Andrea René Jørgensen , Pelle  Hanberg , Mats Bue, Maja Thomassen, Nis  Pedersen Jørgensen , Maiken  Stilling 
Orthopaedic Research Unit , Aarhus University Hospital ; Department of Orthopaedic Surgery , Horsens Regional Hospital ; Department of Orthopaedic Surgery , Aarhus University Hospital ; Orthopaedic Research Unit , Aarhus University Hospital ; Department of Infectious Diseases, Aarhus University Hospital ; Department of Orthopaedic Surgery , Aarhus University Hospital 
Background: Surgical site infection is a severe 
complication to orthopaedic surgery, which 
can prolong admission and increase cost. 
Optimal perioperative antimicrobial 
prophylactic treatment is a key factor in 
preventing surgically related infections. 
Purpose / Aim of Study: To evaluate the time with concentrations 
above relevant minimal inhibitory 
concentrations (fT>MIC) of 4 μg/mL in bone, 
knee joint, and subcutaneous adipose 
tissue after double dose of cefuroxime given 
as either one bolus administration (1x3,000 
mg) or two single doses with a four-hour 
interval (2x1,500 mg) 
Materials and Methods: Sixteen female pigs (Danish Landrace 
breed, weight 73-77 kg) were randomized 
into two groups of eight: Group 1 received a 
bolus administration of 3,000 mg 
cefuroxime. Group 2 received two single 
doses of 1,500 mg administered with a four-
hour interval. Microdialysis was applied for 
sampling in cortical and cancellous bone, 
knee joint, and subcutaneous adipose 
tissue. Plasma samples were collected as 
reference. Sampling was performed for 
eight hours. 
Findings / Results: During an 8 h sampling interval, the mean 
percentage fT>MIC (4 μg/mL) across 
compartments was longer for Group 2 (298 
– 422 min) compared to Group 1 (221 – 269 
min) (p<0.01). In cortical bone, there was a 
tendency towards longer fT>MIC (4 μg/mL) 
in Group 2 (298 min) compared to Group 1 
(221 min) (p=0.053). Within 50 min after 
administration, the mean concentration of 4 
μg/mL was reached for both groups in all 
compartments. In tissues the mean 
concentrations decreased below 4 μg/mL 
after approximately 4 h (Group 1) and 3 h 
(Group 2) from initiation of administration 
(time zero). 
Conclusions: A delayed tissue penetration was found in 
all tissues, where a mean concentration of 4 
μg/mL was reached within 50 min for both 
groups in all compartments. During an 8 h 
interval, double dose cefuroxime 
administered as bolus 2x1,500 mg with a 4 
h interval provides longer time above MIC 
breakpoint for S. aureus (4 μg/mL) than a 
single bolus of 3,000 mg cefuroxim. 
However, to maintain sufficient tissue 
concentrations during longer surgeries, re-
administration of cefuroxime (1,500 mg) 
should be considered already 3 h after the 
first administration.
32. What is so special about the myotendinous junction ?– a RNA-sequencing study
Jens Rithamer Jakobsen, Peter Schjerling, Michael Kjær, Abigail Mackey, Michael Rindom Krogsgaard
Department of sports traumatology M51, Copenhagen University Hospital, Bispebjerg and Frederiksberg; Institute of Sports Medicine, Department of Orthopaedic Surgery M, Copenhagen University Hospital, Bispebjerg and Frederiksberg; Institute of Sports Medicine, Department of Orthopaedic Surgery M, Copenhagen University Hospital, Bispebjerg and Frederiksberg; Center for Healthy Aging, Department of Biomedical Sciences, Faculty of Health and Medical Science, University of Copenhagen; Department of sports traumatology M51, Copenhagen University Hospital, Bispebjerg and Frederiksberg
Background: The connection between the muscle fibers and 
the tendon, name the myotendinous junction 
(MTJ), is architecturally constructed to transmit 
force between muscle and tendon, but at the 
same time it is vulnerable to strain injury. In order 
to explain why these injuries occur and suggest 
how they can be prevented, a better 
understanding of the composition and cellular 
components of the MTJ is needed. Previous 
studies have shown the presence of an unique 
collagen type at the MTJ, Collagen XXII, which is 
not demonstrated elsewhere in the skeletal 
muscle system.
Purpose / Aim of Study: The purpose was to evaluate the gene expression of 
the MTJ and compare it to the adjacent muscle and 
tendon. We aimed to find new targets that are 
unique to the MTJ and of importance for the strength 
or recovery of the tissue. In addition, we wanted to 
identify targets that are higher expressed at the MTJ 
compared to the neighboring muscle and tendon.
Materials and Methods: Samples were collected from the superficial 
digitorum flexor muscle from 20 horses, frozen 
and sliced into sections containing muscle, MTJ 
and tendon tissue before preparation for RT-
PCR. Based on the mRNA results a t-stochastic 
neighboring embedded plot (t-SNE) was made 
and sets of samples from 5 horses with the 
clearest separation between tissues were 
chosen for RNA sequencing. An expected 
contribution of muscle and tendon was 
calculated for all targets based on the known 
expression of 2-300 of the most selective muscle 
and tendon genes. Any variation between the 
expected and measured gene expression was 
regarded as expressed by the MTJ.
Findings / Results: No targets were found to be uniquely expressed at 
the MTJ. Collagen XXIIα1 was expressed 17-fold 
higher compared to the expected value. Generally, 
genes involved in remodeling and reformation of 
skeletal muscle fibers and extracellular matrix were 
expressed to a larger extent at the MTJ.
Conclusions: Despite  the MTJ being a region specialized in force 
transmission with a highly specialized morphology 
no genes could be demonstrated as being unique to 
this region. The genes expressed higher in the MTJ 
compared to muscle and tendon were related to 
remodeling activities, and this confirms the previous 
finding of high rates of remodeling at the MTJ. 
33. ELECTRICAL IMPEDANCE CHARACTERIZATION OF BONE FRACTURE 
Markus Frost, Ming Shen, Stanislav  Zhekov, Ben Krøyer, Gert Frølund  Pedersen, Søren Kold
Dept. of Orthopaedics and Interdisciplinary Orthopaedics, Aalborg University Hospital, Denmark; Dept. of Antennas, Propagation and Millimetre-wave Systems , Aalborg University, Denmark; Dept. of Antennas, Propagation and Millimetre-wave Systems , Aalborg University, Denmark; Dept. of Antennas, Propagation and Millimetre-wave Systems , Aalborg University, Denmark; Dept. of Antennas, Propagation and Millimetre-wave Systems , Aalborg University, Denmark; Dept. of Orthopaedics and Interdisciplinary Orthopaedics, Aalborg University Hospital, Denmark
Background: Differences in electrical characteristics of different 
tissues can provide information of the tissue 
composition at the fracture region. Real-time 
monitoring of bone healing with electrical 
impedance spectroscopy might provide information 
for an individualized treatment of fracture patients. 
However, electrodes must be placed at a distance to 
the fracture site in order not to interfere with bone 
healing.
Purpose / Aim of Study: We investigated whether electrical impedance 
measurements from electrodes placed at a distance 
to a bone defect can detect differences between 
intact bone and bone defects in vivo.
Materials and Methods: Approval was granted from the Inspectorate of 
the Animal Experimentation of Danish Ministry of 
Justice. 6 rabbits were anaesthetized, and a 
fracture protocol was subjected to both tibias. 
Electrical impedance was measured in 
frequencies from 10 Hz to 1 MHz at each step: 1) 
intact bone, 2) medial defect, 3) medial and 
lateral defect, 4) complete 2 mm bone defect. 
One electrode was placed in the medullary canal 
and two electrodes extracortical (lateral and 
posterior) 2 mm from the defect. For each rabbit, 
one tibia had measurements with a free inner 
electrode and the other tibia had measurements 
both with a nail and an isolated nail. 
Findings / Results: For all tibias, the intact bone resulted in higher 
impedance compared with the complete defect, and 
this difference was most pronounced in the 
frequency range of 1kHz to 100 kHz. This applied 
for all types of electrodes including electrode, nail, 
isolated nail. The isolated nail showed the biggest 
impedance difference between the intact bone and 
the complete defect. Incomplete bone defects had 
lower impedance compared with intact bone, but no 
consistent pattern for differences in impedance was 
observed between the different applied defects.
Conclusions: Consistent impedance differences between intact 
bone and complete defects were detected in-vivo in 
rabbits. Further research is needed to explore 
whether the presented method can be used to 
characterize bone healing over time.
34. The ability of comorbidity indices to predict mortality in an orthopedic setting: A systematic review and meta-analysis
Mari Jørstad, Per Gundtoft, Julie Erichsen, Hagen Schmal, Bjarke Viberg
Orthopaedic Surgery and Traumatology, Lillebaelt Hospital, University Hospital of Southern Denmark; Orthopaedic Surgery and Traumatology, Lillebaelt Hospital, University Hospital of Southern Denmark; Orthopaedic Surgery and Traumatology, Lillebaelt Hospital, University Hospital of Southern Denmark; Clinic of Orthopaedic Surgery Medical Center, Faculty of Medicine, University of Freiburg, Germany; Orthopaedic Surgery and Traumatology, Lillebaelt Hospital, University Hospital of Southern Denmark
Background: Several comorbidity indices have been created to 
measure and adjust for the estimated burden of 
comorbidity. 
Purpose / Aim of Study: The objective of this systematic review was to 
evaluate and compare the ability of different 
comorbidity indices to predict mortality in an 
orthopedic setting.
Materials and Methods: A search string was developed in collaboration 
with a scientific librarian. The search string was 
used to extract possible studies from Embase, 
Medline, and the Cochrane Library. Two 
reviewers independently screened the studies 
using Covidence. This study was registered in 
Prospero and ROBINS-I assessment tool was 
used to assess the risk of bias. The area under 
the curve (AUC) was chosen as the primary 
effect estimate. An exploratory meta-analysis 
was made comparing the ability of the Elixhauser 
Comorbidity Index (ECI) and the Charlson 
Comorbidity Index (CCI) to predict in-hospital 
and 1-year mortality.
Findings / Results: Of the 5338 studies identified, 16 met the 
eligibility criteria. Most studies included patients 
with either hip fractures (7 studies) or 
arthroplasties (5 studies). The risk of bias was 
serious for 2 studies and moderate for the 
remaining.
Overall the predictive ability of the different 
comorbidity indices ranged from poor (i.e. AUC 
<0.70) to excellent (AUC >0.89).  
The majority of the included studies only 
compared the ECI and CCI. In-hospital mortality 
for ECI and CCI was reported in 8 studies 
yielding an overall effect size of 0.84 (CI 0.81-
0.88) for ECI and 0.83 (CI 0.79-0.86) for CCI. 
The AUC values were generally lower for all 
other time points ranging from 0.67 to 0.78. For 
1-year mortality the overall effect size was 0.69 
(CI 0.66-0.72) for ECI and 0.70 (CI 0.67-0.74) for 
CCI. 
Conclusions: ECI and CCI can equally be used to adjust for 
comorbidities when analyzing in-hospital mortality in 
an orthopedic setting. However, in general, both 
indices have poor to fair AUC values for 30-day and 
1-year mortality, where other indices might perform 
better.
35. Sampling of the myotendinous junction – how can we do it?
Jens Rithamer Jakobsen, Peter Schjerling, Michael Kjær, Abigail Mackey, Michael Rindom Krogsgaard
Department of sports traumatology M51, Copenhagen University Hospital, Bispebjerg and Frederiksberg; Institute of Sports Medicine, Department of Orthopaedic Surgery M, , Copenhagen University Hospital, Bispebjerg and Frederiksberg; Institute of Sports Medicine, Department of Orthopaedic Surgery M, , Copenhagen University Hospital, Bispebjerg and Frederiksberg; Center for Healthy Aging, Department of Biomedical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen; Department of sports traumatology M51, Copenhagen University Hospital, Bispebjerg and Frederiksberg
Background: The myotendinous junction (MTJ) is the region 
where strain injuries most often occur. Clinically, 
the risk for these injuries can be reduced through 
specific resistance training. This positive effect 
may be caused by changes in the concentration 
of structural proteins, leading to a strengthening 
of the MTJ. However, specific knowledge about 
the effect on the structure and tissue composition 
of the MTJ of resistance exercise is sparse. In 
order to study changes in protein content or gene 
expression at the MTJ it is necessary to isolate 
MTJ from the skeletal muscle and tendon to 
avoid that results from different tissues are 
mixed. 
Purpose / Aim of Study: We aimed to develop a method to divide a sample 
taken from the MTJ into its three components: 
muscle, tendon and MTJ.
Materials and Methods: Samples were collected from the superficial 
digitorum flexor muscle from 20 horses and 
frozen routinely for immunohistochemistry. In 
frozen form each specimen was manually sliced 
parallel to MTJ into 10 µm thick sections and 
sampled for further processing. By controlling 
every 20th section visually it was noted whether 
the section contained muscle, MTJ or tendon. 
RT-PCR was performed on the collected 
sections identifying mRNA targets regularly used 
in the study of skeletal muscle and tendon. A 
Principle Component Analysis (PCA) and a t-
distributed stochastic neighboring plot (t-SNE) 
were made on all the results to evaluate how well 
the different tissue regions had been isolated. 
Findings / Results: It was possible visually to group the samples 
according to the three tissues. The t-SNE plot 
confirmed that the MTJ samples grouped specifically 
and were very similar in relation to their expression 
of the mRNA targets. 
Conclusions: It was possible by this method to divide a specimen 
from the MTJ into muscle-, tendon- and MTJ-tissue. 
It is a cheap and specific method which is useful in 
studies looking into changes introduced at the MTJ 
following resistance exercise and experimental set-
ups. 
36. Manipulating the journal impact factor? The use of Journal-Self-Citations among orthopedic journals to boost journal rankings
Dorte Drongstrup, Søren Overgaard, David Minguillo
Research and Analysis Department, University Library of Southern Denmark; Orthopaedic research unit, Department of Orthopaedic Surgery and Traumatology, Odense University Hospital, Department of Clinical Research, University of Southern Denmark; , KTH Royal Institute of Technology
Background: The Journal Impact Factor (JIF) is often used 
as an indicator of research quality by tenure, 
promotion, and funding assessment 
committees. Thus, a higher JIF could lead to 
increased visibility for journals and more 
publication submissions. This provides 
incentives for journal editors to optimize in 
accordance with the JIF formula; the number 
of citations received in a given year to a 
journal’s publications from previous two 
years divided by the number of only articles 
and reviews from previous two years.
However, the use of JIF to assess research 
quality is highly problematic, since it can 
easily be manipulated. A strategy to boost 
the JIF-score is by increasing the rate of 
Journal-Self-Citations (JSC) to the two 
previous years (JIF-years), which increases 
the number of citations (size of the 
numerator). 
Purpose / Aim of Study: The aim is to investigate to what extent 
Orthopedic journals might use different 
strategies to influence and increase their JIF-
scores.
Materials and Methods: All journals indexed in the subject category 
Orthopedics by the Journal Citation Report 
between 1997 and 2018 were analyzed. The 
data source was the in-house database 
version of Web of Science owned by the 
Royal Institute of Technology (KTH). The 
study covers 95 journals, 210,528 
publications, and 3,990,809 citations. We 
analyze the publishing and citation patterns 
of these journals and apply different 
measures to identify which strategies might 
be the most frequent in the field to optimize 
the impact factors and which journals might 
take most advantage of these strategies to 
boost their JIF and ranking. 
Findings / Results: Our first results show that the rate of JSC to JIF-
years tend to be as almost double as high than 
usual. Still, there are large variance in the JSC 
intensity among journals. If the JSC to the JIF-
years are excluded, the impact factor on 
average decreases 15%. For the 2018 JIF 
ranking, four journals in the top10 changes 
position when JCS are excluded.  
Conclusions: The study finds a strong tendency for JSC in the 
JIF-years. It suggests that the inclusion of JSC 
in the calculation influences the JIF-scores and 
ranking of journals.
37. Bacteriophage treatment against Staphyloccus aureus periprosthetic osteomyelitis in rats
Britt  Siesing-Mejer, Mogens  Kilstrup, Steen  Sejer Poulsen, Kristine  Stub Rønø, Ulla Rydahl Johansen, Niels Henrik Søe
Department of Orthopaedic Surgery Handsection, Hillerød University Hospital; DTU Bioengineering, Technical University Lyngby; Endocrinology and metabolism, The Panum Institute, KU; DTU Bioengineering, Technical University Lyngby; Department of Microbiology, , Rigshospitalet, KU; Handsection, Department of Orthopaedic, , Herlev and Gentofte university Hospital, Denmark
Background: Bacteriophages (phages) are virus-like entities 
that only target bacteria and are composed of 
protein-encapsulated phage chromosomes. 
Using recognition proteins attached to the protein 
capsule, phages bind to the surface of specific 
bacteria and inject their chromosome into the 
cytoplasm. Inside the cell the bacterial gene 
expression machinery is hijacked by the phage 
chromosome, to produce a large litter of 
phage progeny, eventually killing the bacterium 
and releasing the phages into the environment.
Purpose / Aim of Study: To investigate the ability of Bacteriophages to 
eradicate S. aureus, in a knee prosthesis model of
osteomyelitis in rats.
Materials and Methods: Ten Sprague-Dawley rats had prosthesis 
inserted in their left knee, and were divided into 
three groups.
103 S. aureus MN8, ica+ was inserted into the 
femoral and tibial bone marrow of the knee, 
before insertion of the prosthesis. The study 
included two bacteriophage groups with 4 rats in 
each group and one control group with two rats: 
One Bacteriophage group was given high dose 
(106), and one low dose (103) of 
bacteriophages, 
before insertion of the prosthesis. Control rats 
were given sterile saline (0.1 ml) and S. aureus 
bacteria.
After two weeks the rats were sacrificed, and all 
specimens were analyzed clinically, 
radiographically, microbiologically and 
histologically.
Findings / Results: In the group with high dose of bacteriophage, two 
rats died of allergic reaction or cytokine storm.
The remaining two rats showed no reaction to 
treatment. In the group with low dose of 
bacteriophage, one rat died. One rat had a nearly 
total
clearing of infection. The other two rats, showed a 
significate reduction of infection in nearly all 
parameters. One rat in the control group died of 
unknown causes.
Conclusions: Bacteriophage treatment against S. aureus 
osteomyelitis, reduced the infection in the low 
dose group,
but no effect was seen in the group given high 
dose of bacteriophage in any parameters. Two 
rats died in the group given high dose of 
bacteriophages because of cytokine storm.
Lower dose than 103 bacteriophages might 
reduce the infection around the prosthesis, as 
judged by all parameters.