Osteoarthritis and Cartilage 25 (2017) 685e693

One-year incidence of prosthetic joint infection in total hip

arthroplasty: a cohort study with linkage of the Danish Hip
Arthroplasty Register and Danish Microbiology Databases
P.H. Gundtoft y z x *, A.B. Pedersen k, H.C. Schønheyder ¶ #, J.K. Møller yy zz,
S. Overgaard z x xx
y Department of Orthopaedics, Kolding Hospital e a Part of Lillebaelt Hospital, Denmark
z Department of Orthopaedic Surgery and Traumatology, Odense University Hospital, Denmark
x Institute of Clinical Research, University of Southern Denmark, Denmark
k Clinical Epidemiology, Aarhus University Hospital, Denmark
¶ Department of Clinical Microbiology, Aalborg University Hospital, Aalborg, Denmark
# Department of Clinical Medicine, Aalborg University, Aalborg, Denmark
yy Department of Clinical Microbiology, Vejle Hospital e a part of Lillebaelt Hospital, Vejle, Denmark
zz Institute of Regional Health Research, University of Southern Denmark, Denmark
xx Danish Hip Arthroplasty Registera, Denmark

a r t i c l e i n f o s u m m a r y

Article history: Objective: To examine the trend of Prosthetic Joint Infections (PJI) following primary total hip arthro-
Received 22 June 2016 plasty (THA) and the antimicrobial resistance of the bacteria causing these infections.
Accepted 6 December 2016 Materials and methods: We identified a population-based cohort of patients in the Danish Hip Arthro-
plasty Register (DHR) who had primary THA and received their surgery in Jutland or Funen between
Keywords: 2005 and 2014. We followed the patients until revision, emigration, death, or up to 1-year of follow-up.
Prosthetic joint infection
Data from the DHR were combined with those from microbiology databases, the National Register of
Epidemiology
Patients, and the Civil Registration System. We estimated the cumulative 1-year incidence of PJI for two
Incidence
Microbiology
5-year periods; 2005e2009 and 2010e2014. The hazard ratio of PJI as a measure of relative risk after
Antimicrobial resistance adjusting for multiple risk factors was calculated.
Results: Of 48,867 primary THAs identified, 1120 underwent revision within 1 year. Of these, 271 were
due to PJI. The incidence of PJI was 0.53% (95% confidence interval (CI): 0.44; 0.63) during 2005e2009
and 0.57% (95% CI: 0.49; 0.67) during 2010e2014. The adjusted relative risk was 1.05 (95% CI: 0.82; 1.34)
for the 2010e2014 period vs the 2005e2009 period.
The most common micro-organisms identified in the 271 PJI were Staphylococcus aureus (36%) and
coagulase-negative staphylococci (CoNS) (33%); others commonly identified included Enterobacteri-
aceae, enterococci, and streptococci. Antimicrobial resistance to beta-lactams and gentamicin did not
change during the study period.
Conclusion: The risk of PJI within 1-year after primary THA and the antimicrobial resistance of the most
prevalent bacteria remained unchanged during the 2005e2014 study period.
© 2016 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Introduction
According to several studies from national arthroplasty regis-
ters, the risk of prosthetic joint infection (PJI) after total hip
* Address correspondence and reprint requests to: P.H. Gundtoft, Orthopaedic
Department, Kolding Hospital, Skovvangen 6-8, 6000 Kolding, Denmark.
E-mail addresses: per.hviid.gundtoft@rsyd.dk (P.H. Gundtoft), abp@dce.au.dk
(A.B. Pedersen), hcs@rn.dk (H.C. Schønheyder), Jens.Kjoelseth.Moeller@rsyd.dk
(J.K. Møller), Soeren.Overgaard@rsyd.dk (S. Overgaard).
a
www.dhr.dk
arthroplasty (THA) has been increasing1e3, and the trend for the
increase has remained robust after adjusting for known risk factors
in large international arthroplasty register studies4. This has
resulted in growing concern about the potential causes of this in-
crease because none of the risk factors explored in the previous
studies were able to explain this increase4. Several explanations for
the increasing incidence have been suggested including more pa-
tients with comorbidity having operations and enhanced antimi-
crobial resistance among bacteria causing foreign body infections5.

http://dx.doi.org/10.1016/j.joca.2016.12.010
1063-4584/© 2016 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
686 P.H. Gundtoft et al. / Osteoarthritis and Cartilage 25 (2017) 685e693
Microbiology studies of intraoperative cultures from infected
arthroplasties have reported an increase in antimicrobial resistance
to beta-lactams, especially among staphylococci over the past two
decades6e8. However, none of the national arthroplasty register
studies on the incidence of PJI included data from microbiology
databases. Furthermore, relying solely on national arthroplasty
registers in the analysis of PJI trends may be questionable because
studies have shown that these registers underestimate the inci-
dence of PJI by up to 40%9,10.
In a recent study, we showed that a combination of data from
the national Danish Hip Arthroplasty Register (DHR) and microbi-
ology databases made it possible to achieve data on PJI of very high
validity; i.e., with a sensitivity of 90% and specificity of 100%11. In
addition, this combination of databases allows for analysis of the
micro-organisms causing PJI and their antimicrobial resistance. By
combining data from the DHR and the Danish National Register of
Patients, we can further adjust for comorbidity when estimating
the relative risk of PJI for different time periods.
The purpose of this study was to evaluate the trend in PJI inci-
dence within 1 year of implantation of a primary THA by linking the
DHR with the Civil Registration System, the National Register of
Patients, and microbiology databases during the 10-year period
from 2005 to 2014. Our hypothesis was that the risk of PJI within
1 year would increase during the study period, which could be
related to an increase in antimicrobial resistance among prevalent
bacterial isolates.
Methods
Setting
In this Danish, population-based, cohort study, we identified
patients with primary THA performed in Jutland or Funen (3.0
million out of the 5.7 million inhabitants of Denmark) that were
reported to the DHR between 1 January 2005 and 31 December
2014. We used the civil register number to perform linkage at the
individual-level between the DHR, Civil Registration System, Na-
tional Register of Patients, and microbiology databases.
This study is reported according to the RECORD guidelines12.
Data sources
DHR
This register was established in 1995. The purpose of the register
is to improve the quality and outcomes of THA by examining the
epidemiology and identifying risk factors for revision, death, and
complications. Primary THA and subsequent revisions can be linked
directly using the civil register number. Reporting is compulsory for
all private and public orthopedic departments, which results in a
completeness of 97.9% for primary THAs and 96.0% for revisions13.
As of 31 December 2014, 139,525 primary THA and 22,118 revisions
were reported to the register. Data from the register were validated
in 200414. The following data were extracted from the register: date
of primary and revision surgery, side of the operation, primary
diagnosis, duration of surgery, type of fixation, preoperative anti-
biotic prophylaxis (none or generic name and duration of treat-
ment), and type of operating theater (conventional or laminar
airflow ventilation).
Civil Registration System
All Danish citizens are assigned a unique and unchangeable civil
register number upon birth or immigration15. The Civil Registration
System contains information on gender, vital status, and date of
death or migration. The civil register number is recorded for all
Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
health care contacts (public and private) and is included in all
medical databases in Denmark.
National Register of Patients
The register contains information on all discharges from public
and private hospitals since 1977, and emergency room and outpa-
tient visits at hospitals since 199516. Each discharge or outpatient
visit is recorded with one primary diagnosis and one or more
secondary diagnoses. Diagnoses in the register are classified ac-
cording to the International Classification of Diseases Tenth Revi-
sion (ICD-10). The ICD-10 codes used to calculation the Charlson
comorbidity index have been validated in the register17.
Microbiology databases
For revisions prior to 2010, information on intraoperative cul-
tures was extracted from the two electronic Laboratory Information
Systems (MADS and ADBakt) used by the Departments of Clinical
Microbiology in the regions of Jutland or Funen18,19. Only microbi-
ology databases in Jutland and Funen were available for electronic
extracting of data prior to 2010, which is why primary THA in the
rest of Denmark was excluded. If a revision of a primary THA origi-
nally operated in Jutland or Funen was performed outside this area
prior to 2010, information on intraoperative cultures was extracted
manually. After January 2010, an electronic copy of microbiology
reports was automatically sent from all Departments of Clinical
Microbiology in Denmark to the national Danish Microbiology
Database (acronym, MiBa)20. Information on intraoperative cultures
for revisions after 2010 was extracted from this database. Informa-
tion on intraoperative cultures detailing specimen type, bacterial
isolate(s), and antimicrobial antibiogram(s) was extracted. During
the study period, antibiotic susceptibility testing underwent a
transition from Nordic to European technical and interpretive
standards; i.e., from the Swedish Reference Group for Antibiotics
(SRGA) to the European Committee on Antimicrobial Susceptibility
Testing (EUCAST). The transition took place independently for each
department (n ¼ 13) during the first 5-year period. The majority of
Departments of Clinical Microbiology categorized antimicrobial
resistance as S for ‘susceptible’, I for ‘intermediate’, or R for ‘resistant’
(SIR), and other categorizations were also transformed to SIR.
Participants
Patients in our study population were excluded if an incorrect
civil register number was registered in the DHR, more than one
primary THA was registered for the same side of the body, or in-
formation was missing regarding the operation date or side of the
body (Fig. 1). If a revision was registered in the DHR with matching
civil register numbers but on a different side of the body than the
primary THA, review of medical records was conducted in order
determine the correct side of the operation.
Data from the DHR were linked with the Civil Registration
System thereby allowing for complete follow-up of all patients
from the study population until first revision surgery, death,
emigration, or after 1 year of follow-up. Furthermore, linkage of
the DHR with the National Register of Patients was conducted on
an individual-level for all THA in the study population to deter-
mine the level of comorbidity (Fig. 1). Comorbidity was assessed
using the Charlson comorbidity index21, which has been adapted
and validated for use with hospital discharge data22 (Appendix
Comorbidity). In accordance with recent studies23, all available
data from 1977 forward from the National Register of Patients were
used to estimate the Charlson comorbidity index.
Data on revisions registered in the DHR were linked through the
civil registration number with the microbiology databases to obtain
data on intraoperative cultures (Fig. 1).
 P.H. Gundtoft et al. / Osteoarthritis and Cartilage 25 (2017) 685e693
Fig. 1. Flowchart of study population and linkage of registers.
Definition of PJI
A revision was classified as due to PJI if three or more intra-
operative cultures taken during a revision showed growth of the
same micro-organism or if ‘deep infection’ was reported by the
surgeon to the DHR as the indication for revision. A revision was
classified as a case of non-PJI if all of five or more intraoperative
cultures from the revision were negative, regardless of the regis-
tered indication in the DHR. This definition of PJI has been validated
in a previous study11.
Micro-organisms from intraoperative cultures were grouped as
A (virulent pathogen), B (opportunistic pathogen), and C (spore-
forming or less virulent pathogen of questionable significance)
based on their perceived virulence. Only micro-organisms in
groups A and B were used in the definition of PJI (Appendix Micro-
organisms).
In addition to using the above mentioned validated definition,
the relative risk of PJI in the 2010e2014 period compared to the
2005e2009 period was further estimated using the two following
definitions:
DHR definition: only data from the DHR were used; i.e., a revi-
sion was defined as PJI if ‘deep infection’ was reported by the
surgeon to the DHR as an indication for revision.
Microbiological definition: a revision was defined as PJI if three
or more intraoperative cultures showed growth of the same
pathogen from groups A or B, regardless of the indication for
revision registered in the DHR.
Antibiotic prophylaxis and antimicrobial resistance
Preoperative prophylaxis with dicloxacillin or cefuroxime was
used in 97.8% of the primary THA. For 0.5%, a combination of
dicloxacillin or cefuroxime and a third antibiotic was used; in 0.8%,
a third antibiotic or a combination of other antibiotics was used;
and for 0.9%, information was missing. Dicloxacillin is an iso-
xazolylpenicillin with a high degree of stability to beta-lactamases
Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
687
and is very similar to cloxacillin. In staphylococci, resistance to
dicloxacillin or cefuroxime is synonymous with methicillin resis-
tance. We focused on antibiotics used as preoperative prophylaxis
because we wanted to analyze whether changes in antimicrobial
resistance of bacteria introduced at the primary THA surgery might
be related to a presumed increasing trend in PJI. In addition to
dicloxacillin and cefuroxime, gentamicin was included in the
analysis of resistance because it was added to the cement in 98.4%
of the cemented primary THA.
A PJI revision was defined as being infected with a bacterium
resistant to dicloxacillin, cefuroxime, or gentamicin if a bacterium
was found in three or more of the intraoperative cultures and
antimicrobial resistance was reported as either ‘resistant’ or ‘in-
termediate’ according to SIR. Both intrinsic and acquired resistance
were included in the analysis (Appendix, Micro-organisms).
Screening for dicloxacillin- and cefuroxime-resistant Staphylo-
coccus aureus and coagulase-negative staphylococci (CoNS) was
undertaken with cefoxitin throughout the study period24.
We analyzed whether or not there was a difference in antimi-
crobial resistance between the period from 2005 to 2009 compared
to the period from 2010 to 2014 by calculating the number of PJI
revisions performed during each period that were infected with a
bacterium that was resistant to dicloxacillin, cefuroxime, or
gentamicin, respectively. This analysis was performed for the five
most common bacteria.
Additionally, we estimated the percentage of PJI revisions that
were infected with a bacterium that was resistant to the antibiotics
(dicloxacillin or cefuroxime) used as preoperative prophylaxis at
implantation of the primary THA.
Statistical methods
The study population was divided into two 5-year periods (1
January 2005e31 December 2009 and 1 January 2010e31
December 2014). The cumulative incidence was estimated using
competing risk analysis considering death, emigration, and revision
due to causes other than PJI as competing risks; 95% confidence
intervals (CI) were calculated.
688 P.H. Gundtoft et al. / Osteoarthritis and Cartilage 25 (2017) 685e693

FineeGray proportional hazard was used to estimate the hazard The study was approved by the Danish Data Protection Agency
rate ratio as a measure of relative risk, considering PJI to be a rare (journal no. 15/23504).
event. The proportionality assumption was verified by a log minus
Results
log plot and was not violated. The 95% CIs were used for the esti-
mated hazard rate ratio. Because none of the 6657 patients with
Study population
bilateral THA had an infection in both hips, bilateral THA was
treated as an independent observation25.
We identified 48,867 primary THA in 42,210 patients in our
The following known risk factors were used in the adjusted
study population. Only 15 patients (16 primary THA) emigrated
relative risk estimate: comorbidity3,26 (Charlson comorbidity index
within 1 year (Fig. 1). Patients' mean age was 68.85 (95% CI: 68.8;
score: low 0e1, medium 1e2, high 3) (Appendix Comorbidity),
68.9), and 55.1% were female. The majority of patients were oper-
age27,28; gender26,29; indication for primary THA29,30; primary oste-
ated due to osteoarthritis (Table I).
oarthritis, traumatic (acute proximal femoral fracture and sequelae
Within 1-year of follow-up, 1120 patients had a revision per-
from fracture) or non-traumatic avascular femoral head necrosis,
formed. PJI was reported to the DHR as the indication for revision in
inflammatory arthritis, and congenital hip diseases; duration of
279 cases, but 70 cases were shown to be false-positives after linkage
surgery31,32 (less or more than 60 min); fixation technique27,33;
with microbiology databases because all of five or more intra-
cemented, cementless, hybrid (one component being cemented
operative cultures were without growth. An additional 62 cases
and the other being cementless); operating theater29 (conventional
were identified as PJI because three or more intraoperative cultures
or laminar airflow ventilation); and preoperative antibiotics pro-
showed growth of the same pathogen from group A or B, which
phylaxis prior to the primary THA (cefuroxime or dicloxacillin).
resulted in a total of 271 validated cases of surgically-treated PJI.
Fisher's exact test was used to evaluate changes in the distri-
bution of antimicrobial resistance between the two 5-year periods Incidence and relative risk
of 2005e2009 and 2010e2014. The level of statistical significance
was set to P ¼ 0.05 (two-sided). The cumulative 1-year incidence of PJI was 0.53% (95% CI: 0.44;
All statistical analyses were performed with STATA 14.0 statis- 0.63) for the 2005e2009 period and 0.57% (95% CI: 0.49; 0.67) for
tical software (StataCorp, College Station, TX, USA). the 2010e2014 period.

Table I
Patients characteristics in the two study periods

Patients characteristics 2005e2009 2010e2014 Total

No. of primary THA 22,956 25,911 48,867

Age at time of primary THA 68.4 (CI: 68.3; 68.6) 69.2 (CI: 69.1; 69.4) 68.85 CI: 68.8; 68.9)
Gender (% female) 55.1% 55.1% 55.1%
Charlson comorbidity index score:
Low 0e1 69.4% 67.7% 68.5%
Medium 1e2 24.4% 25.5% 25.0%
High 3 6.2% 6.8% 6.5%
Indications for primary THA:
Primary osteoarthritis 79.2% 77.3% 78.2%
Traumatic 12.2% 14.0% 13.1%
Non-traumatic avascular femoral head necrosis 2.3% 2.4% 2.3%
Inflammatory arthritis 1.5% 1.2% 1.3%
Congenital hip diseases 3.7% 3.9% 3.8%
Missing 1.2% 1.3% 1.3%
Duration of primary THA surgery:
<60 min 31.3% 50.1% 41.2%
60 min 68.5% 49.5% 58.4%
Missing 0.2% 0.5% 0.4%
Fixation technique:
Cemented 23.3% 11.1% 16.9%
Cementless 57.6% 66.6% 62.4%
Hybrid 18.0% 21.1% 19.7%
Missing 1.0% 1.2% 1.1%
Operating theater
Conventional 98.6% 98.2% 98.4%
Laminar airflow ventilation 1.4% 1.8% 1.6%
Missing 0.1% 0.1% 0.1%
Preoperative antibiotics prophylaxis
Cefuroxime 57.1% 56.5% 56.3%
Dicloxacillin 41.9% 41.2% 41.4%
Other 1.0% 2.3% 2.3%
Indication for revision
No. of revisions 511 609 1120
Aseptic loosening 40 (7.8%) 58 (9.5%) 98 (8.8%)
Osteolysis without loosening 0 (0%) 1 (0.2%) (0.1%)
Femoral fracture 97 (19.0%) 123 (20.2%) 220 (19.6%)
Dislocation 175 (34.3%) 190 (31.2%) 365 (32.6%)
Component failure 20 (3.9%) 24 (3.9%) 44 (3.9%)
Pain 9 (1.8%) 8 (1.3%) 17 (1.5%)
Other 39 (7.6%) 37 (6.1%) 76 (6.8%)
Reported PJI to DHR 121 (23.7%) 158 (25.9%) 279 (24.9%)
Validated PJI 123 (24.1%) 148 (24.3%) 271 (24.2%)
False-negative PJI 33 29 62
False-positive PJI 31 39 70

Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 P.H. Gundtoft et al. / Osteoarthritis and Cartilage 25 (2017) 685e693
Compared to the period from 2005 to 2009, the crude relative
risk of PJI was 1.08 (95% CI: 0.82; 1.38) in the 2010e2014 period.
Adjustment for risk factors, including comorbidity, did not change
this notably (adjusted relative risk 1.05 (95% CI: 0.82; 1.35)).
When we used DHR data separately, without linkage to the
microbiology databases (DHR definition), the adjusted relative risk
for the 2010e2015 period was 1.16 (95% CI: 0.91; 1.49) when
compared to the 2005e2009 period. Using three or more intra-
operative cultures of the same pathogen from group A or B as the
definition for PJI regardless of the registered classification in the
DHR (microbiology definition), the adjusted relative risk was 1.01
(95% CI: 0.77; 1.32).
Bacteria and antimicrobial resistance
Forty-eight different micro-organisms were identified from all
revisions performed and 42 different micro-organisms from the
revisions performed due to PJI (Appendix, Micro-organisms). The
most commonly identified micro-organisms were S. aureus and
CoNS for both all revisions and revisions due to PJI (Table II).
For the five most common groups of bacteria (Table II), we could
not detect a change in antimicrobial resistance to dicloxacillin,
cefuroxime, or gentamicin (Table III).
Of 97 PJI revisions with dicloxacillin used as the prophylaxis
prior to implantation of the primary THA, the bacteria from 70
revisions had a susceptibility test report; in 18 (25.7%) of these,
bacteria had acquired resistance to dicloxacillin. In addition to the
70 revisions, 13 PJI revisions were infected with bacteria that had
intrinsic resistance to dicloxacillin.
Of the 164 PJI revisions with cefuroxime used as the prophylaxis
prior to primary THA surgery, bacteria from 116 revisions had
susceptibility test reports; in 43 (37.1%) of these, bacteria had ac-
quired resistance to cefuroxime. Additionally, 14 revisions were
infected with bacteria that had intrinsic resistance to cefuroxime.
Other notable changes during the two study periods
The overall incidence of revisions due to all causes within 1-year
of implantation was 2.20% (95% CI: 2.02; 2.40) during the
2005e2009 period and 2.33% (95% CI: 2.15; 2.52) during the
2010e2014 period.
There was an increase in the number of revisions in which
intraoperative cultures had been taken during surgery from 72.4%
in the 2005e2009 period to 79.0% in 2010e2014 period (P ¼ 0.01).
This increase was primary due to a more frequent collection of
intraoperative cultures from revisions, which were reported as
being due to mechanical problems with the prosthesis, pain, or
‘other’ indications. For revisions reported to the DHR as either PJI,
aseptic loosening, osteolysis, or pain, 89.1% had intraoperative
cultures taken, which remained unchanged throughout the study
period (P ¼ 0.75).
Table II
The five most common micro-organisms in the two study periods. More than one type of micro-organisssm was identified in 48 (19%) of the revisions
Most common identified Study period n ¼ 27 1
micro-organism in PJI
No. of revisions with 3
of same bacteria
Staphylococcus aureus 85 (31%)
CoNS 70 (26%)
Enterobacteriaceae* 31 (11%)
Enterococcus spp. 25 (9%)
Streptococcus spp. 23 (8%)
*
Escherichia coli: 18, Klebsiella spp.: 5, Proteus: 5, Enterobacter spp.: 6.
Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
689
Discussion
This study shows that the risk of PJI was stable during the entire
2005e2014 study period when assessed using validated data on PJI
and adjusting for surgical and patient-related risk factors including
comorbidity. The 1-year incidence of PJI following primary THA was
estimated as 0.53% (95% CI: 0.44; 0.63) for the 2005e2009 period,
which was very similar to the incidence of 0.57% (95% CI: 0.49; 0.67)
for the 2010e2014 period. The adjusted relative risk of PJI in
2010e2014 compared to 2005e2009 was 1.05 (95% CI: 0.82; 1.35).
The frequency of antimicrobial resistance to dicloxacillin, cefurox-
ime, and gentamicin did not change in bacterial isolates from
revision surgeries during the study period.
The strength of this study is the linkage between the DHR and
several national registers including the Civil Registration System,
which allowed complete follow-up of all 42,210 patients (except for
15 patients who emigrated); National Register of Patients, which
allowed us to adjust for comorbidities; and the microbiology da-
tabases that enabled us to use validated data from PJI cases and
identify the micro-organisms and their antimicrobial resistance to
prophylactic antibiotics used during the primary THA procedure.
This study has several limitations. First, intraoperative cultures
were only obtained from 76.0% of the revisions (89.1% of those that
could possibly be infected; i.e., revisions due to PJI, aseptic loos-
ening, pain, or osteolysis), which might result in an underestima-
tion of PJI if some of the revisions with missing cultures were due to
PJI. However, the relative risk should not be affected if the under-
estimation is consistent throughout the study period. Secondly, we
only included surgically-treated PJI in our analysis of incidence; a
shift to a more conservative and non-surgical management would
therefore affect our results. Thirdly, the generalizability of our study
may be reduced because the prevalence and patterns of antimi-
crobial resistance probably differ between countries. Nevertheless,
the incidence of PJI and the micro-organisms that were isolated
were similar to those found in other studies1,4,34,35. Finally,
although we could adjust for a number of known risk factors
through linkage with other registers, there were a number of risk
factors not recorded in the registers; e.g., weight/body mass in-
dex36, smoking37, and alcohol consumption38. These variables
represent individual risk factors and they are only accounted for if
the comorbidity is implemented in the Charlson comorbidity index.
A suggested explanation for the increasing incidence of PJI re-
ported from the 1990's to the beginning of the twenty first century
is an increase in antimicrobial resistance5. Several studies, without
linkage with arthroplasty registers, have identified an increase in
beta-lactam-resistant CoNS during this period6,7. We did not detect
a change in antimicrobial resistance in our study e despite a similar
study design and size. This unaltered antimicrobial resistance over
the study period corresponds well with the unaltered incidence of
PJI. However, it is noteworthy that the proportion of beta-lactam-
resistant CoNS increased from 67% (24/36) in the period
2005e2009 to 75% (24/32) in period 2010e2014. Other studies of
2005e2009 n ¼ 123 2010e2014 n ¼ 148 P
No. of revisions with 3 No. of revisions with 3
of same bacteria of same bacteria
35 (28%) 50 (34%) 0.36
37 (30%) 33 (22%) 0.16
13 (11%) 18 (12%) 0.71
14 (11%) 11 (7%) 0.30
13 (11%) 10 (7%) 0.28
690 P.H. Gundtoft et al. / Osteoarthritis and Cartilage 25 (2017) 685e693

Table III using validated data in the continuous monitoring of PJI trends.
Resistance to preoperative antibiotic prophylactic.*n: number of revision THA This study underlines the utility of validating the diagnosis used
infected resistant micro-organism/N: number of THA in which resistance was
reported
for surveillance of PJI by a combination of an arthroplasty register
and microbiology databases11. This was instrumental for achieving
Resistance to antibiotics in revisions due to PJI valid estimates of PJI trend, the microbiological spectrum of PJI,
Microorganisms 2005e2009 2010e2014 Difference and the antimicrobial resistance pattern. However, validated data
n/N* n/N* P: on PJI can be obtained in different ways, e.g., by combination of
other registers, such as prescription registers9 or medical records
Dicloxacillin
Staphylococcus aureus 0/33 1/50 1.00 review45.
CoNS 24/36 24/32 0.45 There may be several explanations for the unchanged incidence
Enterobacteriaceae 15/15 19/19 1.00 compared to earlier studies, which showed an increase incidence.
Enterococcus spp. 14/14 11/11 1.00 Most importantly the study population and periods are not the
Streptococcus spp. 0/2 e/e e
Cefuroxime
same. Moreover, the above mentioned awareness of PJI in the later
Staphylococcus aureus 0/33 1/50 1.00 years of our study period might explain why we did not find the
CoNS 24/36 24/32 0.45 increasing incidence of PJI that we hypothesized, as this awareness
Enterobacteriaceae 7/11 5/18 0.12 might result in initiation of a number of precautions and preventive
Enterococcus spp. 14/14 11/11 1.00
measures that we cannot account for, as they are not recorded in
Streptococcus spp. e/e e/e e
Gentamicin any of the registers, such as e.g., the timing of the administration of
Staphylococcus aureus 0/23 0/5 1.00 perioperative antibiotics46,47 and behavior and number of persons
CoNS 12/25 0/4 0.12 in the operating theater.
Enterobacteriaceae 1/7 0/18 0.28 An increase in PJI incidence might eventually lead to a reduction
Enterococcus spp. 8/12 0/1 0.39
in the number of patients who can undergo this otherwise suc-
Streptococcus spp. 0/2 e/e e
cessful operation because the risk of PJI becomes too high for pa-
tients with known risk factors48. Therefore, a major concern, as
bacteria identified from arthroplasty revisions performed in Scan- shown by a number of previous studies (Dale et al. 1995e20094,
dinavia have reported similar proportions of beta-lactam-resistant Pedersen et al. 1995e20083, Lindgreen et al. 2005e20081, and Kurtz
CoNS from 71% to 84% in the late 1990s and beginning of the twenty et al. 2001e20092), has been an increase in PJI from 1990s to the
first century6,7. A possible explanation of this high prevalence of 2000s, and only a few studies have questioned this reported in-
beta-lactam-resistant CoNS is that beta-lactam susceptible CoNS crease49. In the present study, we did not find a continuous increase
are eliminated by the preoperative prophylactic antibiotics. in PJI incidence during the 10-year study period between 2005 and
Furthermore, previous studies have indicated that CoNS transform 2014, but further studies and continued monitoring of this PJI trend
from susceptible to resistant strains following admission to hospital is mandatory.
and selected by preoperative prophylaxis prior to primary THA
surgery6,39,40. Currently, the most common used preoperative Conclusion
prophylaxis antibiotic prior to revision surgery is cefuroxime and
dicloxacillin41, but the high prevalence of beta-lactam-resistant The relative risk of PJI within the first year following a primary
CoNS in the later study period might result in an increased risk of THA implantation did not increase during the 2005e2014 study
incomplete antibiotic coverage and infection clearance. period. Antimicrobial resistance to beta-lactams and gentamicin
Furthermore, it must be emphasized that resistance to preop- did not change during the study period.
erative antibiotic prophylaxis is a challenge since a high percentage
of the revisions due to PJI were infections with bacteria with either Author contributions
intrinsic or acquired resistance to the antibiotics used as prophy- P. H. Gundtoft: responsible for design of the study, acquisition
laxis prior to implantation of the primary THA. and analysis of data, statistical test and writing the article.
Only one of the PJI caused by S. aureus was found to be a A. B. Pedersen: responsible for design of the study, the analysis
methicillin-resistant S. aureus (MRSA). This correlates with previ- of data and revision of article.
ous reports from Denmark, which found that the incidence of H. C. Schønheyder and J.K.Møller: gave input to the design,
MRSA in Denmark remains low despite an increase in recent contributed substantially to the acquisition and analysis of data and
years42 (approximately 1.7% of tested S. aureus being MRSA in revision of the article.
Denmark compared with 18% in Europe as a whole43). Other S. Overgaard: responsible for designing of the study, data anal-
studies of bacteria identified from arthroplasty revisions in Scan- ysis and revision of article.
dinavia have found a similarly low prevalence of MRSA6,7,39. The All authors have made approved the final version of the article.
discrepancy between the relative risk of PJI incidence estimated by Declaration: P.H. Gundtoft, A.B. Pedersen and S. Overgaard take
DHR data separately and the relative risk estimated by microbi- full responsibility for the integrity of this study, including design,
ology separately was surprising. Especially since the chance of study material, assemble and analysis of data and statistical
diagnosing PJI by microbiology was higher in the 2010e2014 methods and final approval of article.
period than in the 2005e2010 period because the percentage of
revisions in which intraoperative cultures were taken was higher Conflict of interests
in the later period. This possible increase in reporting PJI to the Benefits to one of the authors have been received, but have been
DHRdwithout a corresponding increase in positive culturesdas directed solely to a research fund and educational institution, and
well as the increasing trend toward obtaining intraoperative cul- are not related directly or indirectly to the subject of this article.
tures might be the result of growing attention to PJI in recent years.
This is reflected in a 100-times increase in publications related to Role of funding source
PJI44. In an analysis using register studies, researchers and clini- Region of Southern Denmark and Lillebaelt Hospitals supported
cians must remain aware that the increasing focus on PJI in later this study but had no involvement in the design, acquisition and
years may lead to a risk of bias, which stresses the importance of analysis of data or writing the article.

Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 P.H. Gundtoft et al. / Osteoarthritis and Cartilage 25 (2017) 685e693 691

Acknowledgements Malignancies including lymphoma and leukemia except malignant

neoplasm of skin, Metastatic solid tumor, Mild liver disease, Moder-
None. ate or severe liver disease, Myocardial infarction, Peptic ulcer disease,
Peripheral vascular disease, Renal disease, and Rheumatic disease.
Appendix A We computed the Charlson comorbidity index score for each
patient in the study population based on National Register of Pa-
Charlson comorbidity index tients records of outpatient visits and inpatient hospitalization
discharge.
Charlson comorbidity index include the following disease cate-
gories: HIV/AIDS, cerebrovascular disease, Chronic pulmonary dis- Appendix B
ease, Dementia, Diabetes with chronic complication, Diabetes
without chronic complication, Hemiplegia or paraplegia,

Tabel 1A
Charlson comorbidity index, International Classification of Diseases (ICD) and point

Category ICD-codes Point

AIDS B20.xeB22.x, B24.x 6

Cerebrovascular disease G45.x, G46.x, H34.0, I60.xe169.x 1
Chronic pulmonary disease I27.8, I27.9, J40.xeJ47.x, J60.xeJ67.x, J68.4, J70.1, J70.3 1
Congestive heart failure I09.9, I11.0, I13.0, I13.2, I25.2, I42.0, I42.5eI42.9, I43.x, I50.x, P29.0 1
Dementia F00.xeF03.x, F05.1, G30.X, G31.1 1
Diabetes with chronic complication E10.2eE10.5, E10.7, E11.2eE11.5, E11.7, E12.2eE12.5, E12.7, E13.2eE13.5, E13.7, E14.2eE14.5, E14.7 2
Diabetes without chronic complication E10.0, E10.1, E10.6, E10.8eE11.1, E11.6, E11.8eE12.1, E12.6, E12.8eE13.1, E13.6, E13.8eE14.1, 1
E14.6, E14.8, E14.9
Hemiplegia or paraplegia G04.1, G11.4, G80.2, G81.x, G82.x, G83.0eG83.4, G83.9 2
Malignancies including lymphoma C00.xeC26.x, C30.xeC34.x, C37.xeC41.x, C43.x, C45.xeC58.x, C60.xeC76.x, C81.xeC85.x, 2
and leukemia except malignant C88.x, C90.xeC97.x
neoplasm of skin
Metastatic solid tumor C77.xeC80.x 6
Mild liver disease B18.x, K70.0eK70.3, K70.9, K71.3eK71.5, K71.7, K73.x, K74.x, K76.0, K76.2eK76.4, K76.8, K76.9, Z94.4 1
Moderate or severe liver disease I85.0, I85.9, I86.4, I98.2, K70.4, K71.1, K72.1, K72.9, K76.5, K76.6, K76.7 3
Myocardial infarction I21.x, I22.x, I25.2 1
Peptic ulcer disease K25.xeK28.x 1
Peripheral vascular disease I70.x, I71.x, I73.1, I73.8, I73.9, I77.1, I79.0, I79.2, K55.1. K55.8, K55.9, Z95.8, K95.9 1
Renal disease I12.0, I13.1, N03.2eN03.7, N05.2eN05.7, N18.x, N19.x, N25.0, Z49.0eZ49.2, Z94.0, Z99.2 2
Rheumatic disease M05.x, M06.x, M31.5, M32.xeM34.x, M35.1 1

Intrinsic Intrinsic Intrinsic No. of all revisions in which % of all revisions in which No. of PJI revision in which
Group A,
Micro-organism idenƟfied in revisions resistance to resistance to resistance to Gram Anae rob/A e rob Fami l y micro-organism were microorganism were micro-organism were % of PJI revisions
B, C
dicloxacillin cefuroxime gentamicin idenƟfied idenƟfied idenƟfied

1 Bacillus spe ci e s
2 Bacteroides spe ci e s
3 Clostridium difficile
4 Clostridium perfringens
5 Clostridium spe ci e s
6 Corynebactererium amycolatum
7 Corynebactererium spe ci e s
8 Corynebactererium striatum
9 Corynebacterium simulans
10 Corynebacterium tuberculosteriaricum
11 Candida albicans
12 Enterobacter cloacae
13 Escherichia coli
14 Klebsiella oxytoca
15 Klebsiella pneumoniae
16 Morganella morganii
17 Pantoea spe ci e s
18 Proteus mirabilis
19 Proteus vulgaris
20 Serra a marcescens
21 Enterococcus faecalis
22 Enterococcus faecium
23 Enterococcus spe ci e s
24 Micrococcus spe ci e s
25 Acinetobacter lwoffii
26 Acinetobacter spe ci e s
27 Peptoniphilus asaccharoly cus
28 Peptostreptococcus spe ci e s
29 Prevotella spe ci e s
30 Propionibacterium acnes
31 Propionibacterium spe ci e s
32 Pseudomonas aeruginosa
33 Pseudomonas pu da
34 Pseudomonas stutzeri
35 Coagulase-nega ve staphylococcus species*
36 Staphylococcus aureus
37 Staphylococcus capi s
38 Staphylococcus cohnii
39 Staphylococcus epidermidis
40 Staphylococcus haemoly cus
41 Staphylococcus lugdunensis
42 Staphylococcus saprophy cus
43 Staphylococcus spe ci e s
44 Staphylococcus warneri
45 Group A streptococcus/S. pyogenes
46 Group B streptococcus/S. agalac ae
47 Group C stre ptococcus
48 Group D stre ptococcus
49 Group G stre ptococcus
50 Non-he mol yti c stre ptococcus
51 Streptococcus anginosus
52 Streptococcus oralis
53 Streptococcus pneumoniae
54 Streptococcus species
C + obl i gate ae robe s/facul tati ve anae robe s Baci l l ace ae 4 1, 3% 2 0, 7%
B x x - obl i gate anae robe s Bacte roi dace ae 1 0, 3% 1 0, 4%
C x + obl i gate anae robi c Cl ostri di ace ae 1 0, 3% 1 0, 4%
C x + obl i gate anae robi c Cl ostri di ace ae 2 0, 7% 2 0, 7%
C x + obl igate anae robi c Cl ostri di ace ae 1 0, 3% 1 0, 4%
B + ae robi c/f acul tati ve anae robi c Coryne bacte ri ace ae 1 0, 3% 1 0, 4%
B + ae robi c/f acul tati ve anae robi c Coryne bacte ri ace ae 1 0, 3% 1 0, 4%
B + ae robi c/f acul tati ve anae robi c Coryne bacte ri ace ae 3 1, 0% 2 0, 7%
B + ae robi c/f acul tati ve anae robi c Coryne bacte ri ace ae 1 0, 3% 1 0, 4%
B + ae robi c/f acul tati ve anae robi c Coryne bacte ri ace ae 1 0, 3%
C Debaryomycetaceae 1 0, 3% 1 0, 4%
B x x - f acul tati ve anae robe s Ente robacte ri ace ae 10 3, 3% 10 3, 7%
B x - f acul tati ve anae robe s Ente robacte ri ace ae 20 6, 6% 18 6, 6%
B x - f acul tati ve anae robe s Ente robacte ri ace ae 2 0, 7% 2 0, 7%
B x - f acul tati ve anae robe s Ente robacte ri ace ae 3 1, 0% 3 1, 1%
B x - f acul tati ve anae robe s Ente robacte ri ace ae 1 0, 3% 1 0, 4%
B x - facultaƟve anaerobes Enterobacteriaceae 1 0, 3%
B x - f acul tati ve anae robe s Ente robacte ri ace ae 1 0, 3% 1 0,4%
B x - f acul tati ve anae robe s Ente robacte ri ace ae 4 1, 3% 4 1, 5%
B x - f acul tati ve anae robe s Ente robacte ri ace ae 1 0, 3% 1 0, 4%
B x x + f acul tati ve anae robe s Ente rococcace ae 23 7, 5% 23 8, 5%
B x x + f acul tati ve anae robe s Ente rococcace ae 2 0, 7% 2 0, 7%
B x x + f acul tati ve anae robe s Ente rococcace ae 8 2, 6% 8 3, 0%
C + obl i gate ae robi c Mi crococcace ae 1 0, 3% 1 0, 4%
C x - obligate aerobic Moraxellaceae 1 0, 3%
C x - obligate aerobic Moraxellaceae 1 0, 3% 1 0, 4%
B x + obl igate anae robi c Pe ptoni phi l ace ae 2 0, 7% 2 0,7%
C x + obl igate anae robi c Pe ptostre ptococcace ae 1 0, 3% 1 0,4%
B x x - obl i gate anae robi c Pre vote l l ace ae 1 0, 3% 1 0,4%
B x + obl igate anae robi c Propi onibacte ri ace ae 3 1, 0% 2 0,7%
B x + obl igate anae robi c Propi onibacte ri ace ae 1 0, 3%
B x x - obl i gate ae robi c Pse udomonadace ae 8 2, 6% 7 2, 6%
B x x - obl i gate ae robi c Pse udomonadace ae 1 0, 3%
B x x - obl i gate ae robi c Pse udomonadace ae 1 0, 3% 1 0, 4%
B + obl i gate ae robi c Staphyl ococcace ae 52 17,0% 31 11, 4%
A + f acul tati ve anae robe s Staphyl ococcace ae 108 35,4% 97 35, 8%
B + f acul tati ve anae robe s Staphyl ococcace ae 6 2, 0% 5 1, 8%
B + facultaƟve anaerobes Staphylococcaceae 1 0, 3% 1 0, 4%
B + f acul tati ve anae robe s Staphyl ococcace ae 62 20,3% 54 19, 9%
A + facultaƟve anaerobes Staphylococcaceae 1 0, 3% 1 0, 4%
A + f acul tati ve anae robe s Staphyl ococcace ae 2 0, 7% 2 0, 7%
B + facultaƟve anaerobes Staphylococcaceae 1 0, 3%
B + f acul tati ve anae robe s Staphyl ococcace ae 11 3, 6% 8 3, 0%
B + f acul tati ve anae robe s Staphyl ococcace ae 4 1, 3% 2 0, 7%
A + f acul tati ve anae robe s Stre ptococcace ae 1 0, 3% 1 0, 4%
A + f acul tati ve anae robe s Stre ptococcace ae 13 4, 3% 13 4, 8%
A + f acul tati ve anae robe s Stre ptococcace ae 1 0, 3% 1 0, 4%
A + f acul tati ve anae robe s Stre ptococcace ae 5 1, 6% 5 1, 8%
A + f acul tati ve anae robe s Stre ptococcace ae 3 1, 0% 3 1, 1%
B + f acul tati ve anae robe s Stre ptococcace ae 3 1, 0% 2 0, 7%
A + f acul tati ve anae robe s Stre ptococcace ae 1 0, 3% 1 0, 4%
A + f acul tati ve anae robe s Stre ptococcace ae 1 0, 3% 1 0, 4%
A + f acul tati ve anae robe s Stre ptococcace ae 1 0, 3% 1 0, 4%
A + facultaƟve anaerobes Streptococcaceae 1 0, 3% 1 0, 4%

Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
692 P.H. Gundtoft et al. / Osteoarthritis and Cartilage 25 (2017) 685e693
References
1. Lindgren V, Gordon M, Wretenberg P, Karrholm J, Garellick G.
Deep infection after total hip replacement: a method for na-
tional incidence surveillance. Infect Control Hosp Epidemiol
2014;35(12):1491e6.
2. Kurtz SM, Lau E, Watson H, Schmier JK, Parvizi J. Economic
burden of periprosthetic joint infection in the United States.
J Arthroplasty 2012;27(8 Suppl). 61e65 e1.
3. Pedersen AB, Svendsson JE, Johnsen SP, Riis A, Overgaard S.
Risk factors for revision due to infection after primary total hip
arthroplasty. A population-based study of 80,756 primary
procedures in the Danish Hip Arthroplasty Registry. Acta
Orthop 2010;81(5):542e7.
4. Dale H, Fenstad AM, Hallan G, Havelin LI, Furnes O,
Overgaard S, et al. Increasing risk of prosthetic joint infection
after total hip arthroplasty. Acta Orthop 2012;83(5):449e58.
5. Kapadia BH, Berg RA, Daley JA, Fritz J, Bhave A, Mont MA.
Periprosthetic joint infection. Lancet 2016;387(10016):
386e94.
6. Lutro O, Langvatn H, Dale H, Schrama JC, Hallan G, Espehaug B,
et al. Increasing resistance of coagulase-negative Staphylococci
in total hip arthroplasty infections: 278 THA-revisions due to
infection reported to the Norwegian Arthroplasty Register
from 1993 to 2007. Adv Orthop 2014;2014:580359.
7. Stefansdottir A, Johansson D, Knutson K, Lidgren L,
Robertsson O. Microbiology of the infected knee arthroplasty:
report from the Swedish Knee Arthroplasty Register on 426
surgically revised cases. Scand J Infect Dis 2009;41(11e12):
831e40.
8. Malhas AM, Lawton R, Reidy M, Nathwani D, Clift BA. Causative
organisms in revision total hip & knee arthroplasty for infec-
tion: increasing multi-antibiotic resistance in coagulase-
negative Staphylococcus and the implications for antibiotic
prophylaxis. Surgeon 2015;13(5):250e5.
9. Lindgren JV, Gordon M, Wretenberg P, Karrholm J, Garellick G.
Validation of reoperations due to infection in the Swedish Hip
Arthroplasty Register. BMC Musculoskelet Disord 2014;15:
384.
10. Gundtoft PH, Overgaard S, Schonheyder HC, Moller JK,
Kjaersgaard-Andersen P, Pedersen AB. The “true” incidence of
surgically treated deep prosthetic joint infection after 32,896
primary total hip arthroplasties: a prospective cohort study.
Acta Orthop 2015;86(3):326e34.
11. Gundtoft PH, Pedersen AB, Schonheyder HC, Overgaard S.
Validation of the diagnosis ‘prosthetic joint infection’ in the
Danish Hip Arthroplasty Register. Bone & Joint J 2016;98-b(3):
320e5.
12. Benchimol EI, Smeeth L, Guttmann A, Harron K, Moher D,
Petersen I, et al. The REporting of studies Conducted using
Observational Routinely-collected health Data (RECORD)
statement. PLoS Med 2015;12(10):e1001885.
13. Danish Hip Arthroplasty Register: Annual Report 2015 2015.
14. Pedersen A, Johnsen S, Overgaard S, Soballe K, Sorensen HT,
Lucht U. Registration in the Danish Hip Arthroplasty Registry:
completeness of total hip arthroplasties and positive predic-
tive value of registered diagnosis and postoperative compli-
cations. Acta Orthop Scand 2004;75(4):434e41.
15. Schmidt M, Pedersen L, Sorensen HT. The Danish Civil Regis-
tration System as a tool in epidemiology. Eur J Epidemiol
2014;29(8):541e9.
16. Schmidt M, Schmidt SA, Sandegaard JL, Ehrenstein V,
Pedersen L, Sorensen HT. The Danish National Patient Registry:
a review of content, data quality, and research potential. Clin
Epidemiol 2015;7:449e90.
Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
17. Thygesen SK, Christiansen CF, Christensen S, Lash TL,
Sorensen HT. The predictive value of ICD-10 diagnostic coding
used to assess Charlson comorbidity index conditions in the
population-based Danish National Registry of Patients. BMC
Med Res Methodol 2011;11:83.
18. [cited 2016, 5 April]. Available from: http://www.madsonline.
dk/.
19. autonik ab [cited 2016, 5 April]. Available from: http://www.
autonik.dk/.
20. Voldstedlund M, Haarh M, Molbak K. The Danish Microbiology
Database (MiBa) 2010 to 2013. Euro Surveill 2014;19(1).
21. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method
of classifying prognostic comorbidity in longitudinal studies:
development and validation. J Chronic Dis 1987;40(5):373e83.
22. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC,
et al. Coding algorithms for defining comorbidities in ICD-9-
CM and ICD-10 administrative data. Med Care 2005;43(11):
1130e9.
23. Brunelli SM, Gagne JJ, Huybrechts KF, Wang SV, Patrick AR,
Rothman KJ, et al. Estimation using all available covariate in-
formation versus a fixed look-back window for dichotomous
covariates. Pharmacoepidemiol Drug Saf 2013;22(5):542e50.
24. DANMAP 2010-Use of Antimicrobial Agents and Occurrence of
Antimicrobial Resistance in Bacteria From Food Animals, Food
and Humans in Denmark [cited 2016, 5 April ]. Available from:
http://www.danmap.org/~/media/Projekt%20sites/Danmap/
DANMAP%20reports/Danmap_2010.ashx.
25. Ranstam J, Karrholm J, Pulkkinen P, Makela K, Espehaug B,
Pedersen AB, et al. Statistical analysis of arthroplasty data. I.
Introduction and background. Acta Orthop 2011;82(3):253e7.
26. Rasouli MR, Restrepo C, Maltenfort MG, Purtill JJ, Parvizi J. Risk
factors for surgical site infection following total joint arthro-
plasty. J Bone Joint Surg Am 2014;96(18):e158.
27. Dale H, Skramm I, Lower HL, Eriksen HM, Espehaug B,
Furnes O, et al. Infection after primary hip arthroplasty: a
comparison of 3 Norwegian Health Registers. Acta Orthop
2011;82(6):646e54.
28. Ridgeway S, Wilson J, Charlet A, Kafatos G, Pearson A, Coello R.
Infection of the surgical site after arthroplasty of the hip.
J Bone Joint Surg Br 2005;87(6):844e50.
29. Dale H, Hallan G, Hallan G, Espehaug B, Havelin LI,
Engesaeter LB. Increasing risk of revision due to deep infection
after hip arthroplasty. Acta Orthop 2009;80(6):639e45.
30. Reina N, Delaunay C, Chiron P, Ramdane N, Hamadouche M.
Infection as a cause of primary total hip arthroplasty revision
and its predictive factors. Orthop Traumatol Surg Res
2013;99(5):555e61.
31. Pugely AJ, Martin CT, Gao Y, Schweizer ML, Callaghan JJ. The
incidence of and risk factors for 30-day surgical site infections
following primary and revision total joint arthroplasty.
J Arthroplasty 2015;30(9 Suppl):47e50.
32. Ong KL, Kurtz SM, Lau E, Bozic KJ, Berry DJ, Parvizi J. Prosthetic
joint infection risk after total hip arthroplasty in the Medicare
population. J Arthroplasty 2009;24(6 Suppl):105e9.
33. Hooper GJ, Rothwell AG, Stringer M, Frampton C. Revision
following cemented and uncemented primary total hip
replacement: a seven-year analysis from the New Zealand
Joint Registry. J Bone Joint Surg Br 2009;91(4):451e8.
34. Kapadia BH, Berg RA, Daley JA, Fritz J, Bhave A, Mont MA.
Periprosthetic joint infection. Lancet 2016 Jan 23;387(10016):
386e94.
35. Lindeque B, Hartman Z, Noshchenko A, Cruse M. Infection after
primary total hip arthroplasty. Orthopedics 2014;37(4):257e65.
36. Lubbeke A, Zingg M, Vu D, Miozzari HH, Christofilopoulos P,
Uckay I, et al. Body mass and weight thresholds for increased
 P.H. Gundtoft et al. / Osteoarthritis and Cartilage 25 (2017) 685e693 693

prosthetic joint infection rates after primary total joint
arthroplasty. Acta Orthop 2016;87(2):132e8.
37. Debreuve-Theresette A, Diallo S, Siboni R, Ohl X, Dehoux E,
Bajolet O. Infections in total hip and total knee arthroplasty:
development of a score to assess endogenous risk of surgical
site infections. Surg Infect (Larchmt) 2015;16(6):794e8.
38. Grammatico-Guillon L, Baron S, Rosset P, Gaborit C, Bernard L,
Rusch E, et al. Surgical site infection after primary hip and knee
arthroplasty: a cohort study using a hospital database. Infect
Control Hosp Epidemiol 2015;36(10):1198e207.
39. Stefansdottir A, Johansson A, Lidgren L, Wagner P, W-Dahl A.
Bacterial colonization and resistance patterns in 133 patients
undergoing a primary hip- or knee replacement in Southern
Sweden. Acta Orthop 2013;84(1):87e91.
40. Zingg W, Demartines N, Imhof A, Senn G, Ruef C. Rapid colo-
nization with methicillin-resistant coagulase-negative staph-
ylococci after surgery. World J Surg 2009;33(10):2058e62.
41. Danish Hip Arthroplasty Register: Annual Report 2016 [cited
2016, 1 July]. Available from: https://www.sundhed.dk/
content/cms/98/4698_dhr-%C3%A5rsrapport-2016.pdf; 2016.
42. DANMAP. DANMAP 2013-use of Antimicrobial Agents and
Occurrence of Antimicrobial Resistance in Bacteria from Food
Animals, Food and Humans in Denmark. Statens Serum Insti-
tut, National Veterinary Institute Technical University of
Denmark, National Food Institute Technical University of
Denmark; 2013.
43. European Centre for Disease Prevention and Control. Antimi-
crobial resistance surveillance in Europe e annual report of
the European Antimicrobial Resistance Surveillance Network
(EARS-Net). Eur Centre Dis Prev Control 2013:1e37.
44. Parvizi J, Haddad FS. Periprosthetic joint infection: the last
frontier. Bone & Joint J 2015;97-b(9):1157e8.
45. Zhu M, Ravi S, Frampton C, Luey C, Young S. New Zealand Joint
Registry data underestimates the rate of prosthetic joint
infection. Acta Orthop 2016;87(4):346e50.
46. van Kasteren ME, Mannien J, Ott A, Kullberg BJ, de Boer AS,
Gyssens IC. Antibiotic prophylaxis and the risk of surgical site
infections following total hip arthroplasty: timely adminis-
tration is the most important factor. Clin Infect Dis 2007;44(7):
921e7.
47. Dhammi IK, Ul Haq R, Kumar S. Prophylactic antibiotics in
orthopedic surgery: controversial issues in its use. Indian J
Orthop 2015;49(4):373e6.
48. Teillant A, Gandra S, Barter D, Morgan DJ, Laxminarayan R.
Potential burden of antibiotic resistance on surgery and cancer
chemotherapy antibiotic prophylaxis in the USA: a literature
review and modelling study. Lancet Infect Dis 2015;15(12):
1429e37.
49. Rasouli MR, Maltenfort MG, Purtill JJ, Hozack WJ, Parvizi J. Has
the rate of in-hospital infections after total joint arthroplasty
decreased? Clin Orthop Relat Res 2013;471(10):3102e11.

Downloaded for FK UMI 3 Makassar (fkumimks3@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on November 07, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.