The Infected Total Knee Arthroplasty: Prevention, Diagnosis, and Treatment

186 Pages·2018·6.043 MB·English

by  E. Carlos Rodríguez-Merchán,Sam Oussedik (eds.)

E. Carlos Rodríguez-Merchán
Sam Oussedik
Editors
T he Infected Total
Knee Arthroplasty
Prevention, Diagnosis,
and Treatment
123
The Infected Total Knee Arthroplasty
E. Carlos Rodríguez-Merchán
Sam Oussedik
Editors
The Infected Total
Knee Arthroplasty
Prevention, Diagnosis,
and Treatment
Editors
E. Carlos Rodríguez-Merchán Sam Oussedik
Department of Orthopaedic Surgery University College London Hospitals
La Paz University Hospital London
Madrid UK
Spain
ISBN 978-3-319-66729-4 ISBN 978-3-319-66730-0 (eBook)
https://doi.org/10.1007/978-3-319-66730-0
Library of Congress Control Number: 2017959149
© Springer International Publishing AG 2018
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, express or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Printed on acid-free paper
This Springer imprint is published by Springer Nature
The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
Patients presenting for total knee arthroplasty (TKA) have one goal in mind—
to return to “normal” knee function, however they choose to define this.
Postoperative complications are understood to be a risk, but one worth taking
for all patients who consent to undergo the procedure.
The diagnosis of infection following TKA is devastating and represents
the most cruel betrayal of this dream of normal knee function. Both patients
and clinicians alike may experience the classical grief response, journeying
from denial through anger and guilt to acceptance.
Differentiating periprosthetic joint infection (PJI) from other causes of
postoperative pain is the surgeon’s first challenge. Infection does not
always declare itself overtly and so a knowledge of the latest diagnostic
strategies is vital.
From this diagnosis flows the appropriate treatment strategy. Here again,
risk of failure must be navigated, both in terms of recurrent infection and
poor knee function. Unfortunately, while offering the possibility of retaining
good function, less invasive procedures also run a greater risk of failing to
clear the infection.
As with many medical conditions, the key to obtaining a good outcome is
rapid diagnosis which then enables the timely deployment of the correct
treatment strategy.
In the chapters of this book we have collected the current evidence regard-
ing the best management of PJI, drawing on the most experienced clinicians
in the field from Europe and the United States. It is our hope that our readers
will gain the information they need to help their patients achieve their goals.
Madrid, Spain E. Carlos Rodríguez-Merchán
London, UK Sam Oussedik
v
Contents
1 Epidemiology of the Infected Total Knee Arthroplasty:
Incidence, Causes, and the Burden of Disease . . . . . . . . . . . . . . 1
E. Carlos Rodríguez-Merchán and Alexander D. Liddle
2 Microbiological Concepts of the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
E. Carlos Rodríguez-Merchán and Alexander D. Liddle
3 Preoperative Optimization to Prevent an Infected
Total Knee Arthroplasty: Host Factors . . . . . . . . . . . . . . . . . . . . 19
Sven E. Putnis and Sam Oussedik
4 Depilation and Skin Preparation to Prevent an Infected
Total Knee Arthroplasty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Carlos Kalbakdij-Sánchez, Gregorio Arroyo- Salcedo,
and E. Carlos Rodríguez-Merchán
5 Antibiotic Prophylaxis to Prevent Infection in Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Alfonso Vaquero-Picado and E. Carlos Rodríguez-Merchán
6 Preoperative Screening and Eradication of Infection . . . . . . . . 47
Alexander D. Liddle and E. Carlos Rodríguez-Merchán
7 Clinical Diagnosis of the Infected Total Knee Arthroplasty . . . 55
Stephen M. Petis and Matthew P. Abdel
8 Diagnosis by Imaging of the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Carmen Martín-Hervás and E. Carlos Rodríguez-Merchán
9 Serological Markers of Infection in the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Alexander J. Rondon, Timothy L. Tan, and Javad Parvizi
10 The Role of Knee Aspiration in the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Juan S. Ruiz-Pérez, Mercedes Agüera-Gavaldá,
and E. Carlos Rodríguez-Merchán
vii
viii Contents
11 Polymerase Chain Reaction (PCR) in the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Andrea Volpin and Sujith Konan
12 Histological Diagnosis in the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Paddy Subramanian and Rahul Patel
13 Intraoperative Cultures for the Suspected Total Knee
Arthroplasty Infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Antony C. Raymond and Sam Oussedik
14 Sonication of Removed Implants in the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Enrique Gómez-Barrena and Eduardo García-Rey
15 Antibiotic Suppression in the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
José A. García-Ramos García, Alicia Rico-Nieto,
and E. Carlos Rodríguez-Merchán
16 Arthroscopic Debridement of Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Jonathan Miles and Michael T. Parratt
17 Open Debridement and Polyethylene Exchange (ODPE)
in the Infected Total Knee Arthroplasty . . . . . . . . . . . . . . . . . . . 133
Carlos A. Encinas-Ullán, Ángel Martínez-Lloreda,
and E. Carlos Rodríguez-Merchán
18 One-Stage Revision Arthroplasty in the Infected Total
Knee Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Jurek R.T. Pietrzak, David A. George, and Fares S. Haddad
19 Two-Stage Revision of Infected Total Knee Arthroplasty . . . . . 151
Agustín Garabito-Cociña, Primitivo Gómez-Cardero,
and E. Carlos Rodríguez-Merchán
20 Knee Arthrodesis in the Infected Total Knee Arthroplasty . . . 165
Nima Razii, Rahul Kakar, and Rhidian Morgan-Jones
21 Above Knee Amputation in the Treatment of Failed
Septic Total Knee Arthroplasty . . . . . . . . . . . . . . . . . . . . . . . . . . 181
E. Carlos Rodríguez-Merchán, Hortensia de la
Corte-Rodriguez, and Juan M. Román-Belmonte
1
Epidemiology of the Infected Total
Knee Arthroplasty: Incidence,
Causes, and the Burden of Disease
E. Carlos Rodríguez-Merchán
and Alexander D. Liddle
Abstract
Periprosthetic joint infection (PJI) after total knee arthroplasty (TKA) is a
severe complication. The purpose of this chapter is to review the inci-
dence, causes, and burden of PJI after TKA. At 30 days, the overall rate of
surgical site infection (SSI) is 1.1%, while the reported rate of deep infec-
tion is 0.1%. The lifetime incidence of PJI after TKA ranges from 0.7 to
4.6%. Many related and predisposing factors have been identified. These
can be classified as preoperative, intraoperative, postoperative, and late
infections. The preoperative factors are previous knee surgery, inflamma-
tory arthritis, and the use of glucocorticoids and immunosuppressants. The
intraoperative factors are prolonged surgical time, inadequate antibiotic
prophylaxis, and intraoperative fractures. The postoperative factors are
wound drainage for longer than 10 days, reoperation and deep venous
thrombosis. Factors related to late infections include cutaneous infections,
urinary tract infections, lower respiratory tract infections, abdominal
infections, and generalized sepsis. Patients with PJIs have significantly
longer hospitalizations (5.3 vs. 3 days), more readmissions (3.6 vs. 0.1),
and more clinic visits (6.5 vs. 1.3) when compared to a matched control
group. The mean annual cost is significantly higher in patients who have
PJIs ($116,383 on average) when compared to the matched control group
($28,249 on average). Hospital costs are between 2- and 24-fold higher in
patients with PJI than in those without PJI. PJIs following TKA represent
a huge burden for the patient, for the surgeon, and for the health-care
economy.
E.C. Rodríguez-Merchán (*) A.D. Liddle
Department of Orthopaedic Surgery, University College London Institute of Orthopaedics
“La Paz” University Hospital-IdiPaz, and Musculoskeletal Science, Royal National
Paseo de la Castellana 261, 28046 Orthopaedic Hospital, Brockley Hill, Stanmore,
Madrid, Spain Middlesex HA7 4LP, UK
e-mail: [email protected] e-mail: [email protected]
© Springer International Publishing AG 2018 1
E.C. Rodriguez-Merchan, S. Oussedik (eds.), The Infected Total Knee Arthroplasty,
https://doi.org/10.1007/978-3-319-66730-0_1
2 E.C. Rodríguez-Merchán and A.D. Liddle
1.1 Introduction reported rate of 0.67%, meaning that the rate was
underestimated by over a third [5]. Similar find-
Periprosthetic joint infection (PJI) after total ings were reported by audits of the Danish and
knee arthroplasty (TKA) is a severe complication Swedish joint registries with the rates being
which has significant personal and financial costs underestimated by 40% and 33%, respectively [6,
[1]. In this chapter, we discuss the burden of PJI 7]. Reasons for underreporting may include the
worldwide and aim to define the risk factors for fact that some reoperations do not class as revi-
the development of PJI following TKA. sion surgery (such as debridement and exchange
of modular components) or that revisions for
infection are not recognized and are therefore not
1.2 Incidence reported.
The most common causative organisms for PJI
The development of institutional and national are staphylococci [8, 9], although gram-n egative
joint registries has allowed us to define the inci- organisms are becoming more common as are
dence and prevalence of PJI with greater accu- multidrug resistant organisms. The incidence of
racy than was previously possible. Pugely et al. infection with lower virulence organisms such as
[2], in a study of 23,128 joint replacements (pri- propionibacteria may be underreported as they
mary and revision total knee and hip arthro- require prolonged incubation periods for cultures
plasty), estimated that the rate of surgical site to be isolated [10, 11].
infection (SSI) at 30 days was 1.1%, with the rate
of deep infection being 0.1% at the same time
point. Infection is now the most common reason 1.3 Risk Factors for Prosthetic
for revision in the National Joint Registry for Joint Infection Following
England, Wales, and Northern Ireland, reporting TKA
a patient-time incidence rate (PTIR) of 1.05 revi-
sions for infection per 1000 patient years [3]. Several authors have attempted to delineate the
While improvements in implant design, materi- principal risk factors for PJI after TKA. Kunutsor
als, and instrumentation have reduced the rate of et al. [12] conducted a meta-analysis examining
aseptic loosening and other “technical” compli- risk factors for infection after primary hip and
cations, no such improvement has been demon- knee replacement and reported a higher risk of
strated in the rate of infection. As a result, the rate PJI in men compared to women and in smokers
of revision for infection is increasing relative to compared to nonsmokers. Diabetes, rheumatoid
other reasons for revision. It is not clear whether arthritis, corticosteroids, and previous surgery to
the absolute rate of PJI is increasing; Dale et al. the joint in question were all reported to increase
studied the rate of revision in four Nordic arthro- the risk of PJI as did frailty, but alcohol intake,
plasty registers and reported that the risk of PJI age, hypertension, or previous intra-articular ste-
increased between 1995 and 2009 in all countries roid injection was not found to be associated
[4]. There were no significant changes in risk fac- with PJI.
tors during that time period, but the increase may Tayton et al. analyzed data on 64,566 cases
simply indicate improvements in diagnosis and from the New Zealand Joint Registry, using revi-
reporting. sion surgery for PJI at 6 and 12 months after sur-
Overall, however, it is likely that the reported gery as primary outcome measures [13]. Again,
rates of infection represent an underestimate. Zhu male gender was a significant risk factor for
et al. cross-referenced 4009 records from the infection. Other factors included previous sur-
New Zealand Joint Registry with records from gery (osteotomy, ligament reconstruction), the
three tertiary hospitals, finding that the rate of use of laminar flow, and the use of antibiotic-
revision for infection was 1.1%, compared to the laden cement. There was a trend toward signifi-
1 Epidemiology of the Infected Total Knee Arthroplasty: Incidence, Causes, and the Burden of Disease 3
cance at 6 months with the use of surgical helmet deep infections, the remaining 45 were superficial
systems. These findings showed that patient fac- and were treated successfully with antibiotics
tors remain the most important in terms of pre- with or without surgical debridement. Six patients
dicting early PJI following TKA and suggested required superficial debridement (in all cases the
that some factors previously identified as being infection was superficial to the fascia); the mean
protective (such as laminar flow and exhaust sys- duration of antibiotic treatment was 16.5 days. At
tems) may in fact increase the risk of PJI. Jamsen almost 6 years, 6 patients had died of unrelated
et al. [14] analyzed 7181 TKAs and total hip causes, and 3 were revised for other causes; none
arthroplasties (THAs) from a single center, find- of the 45 patients had a deep infection [16].
ing an increased risk of PJI with diabetes (odds
ratio, OR, 2.3 compared to nondiabetics) and
obesity (OR 6.4 in the morbidly obese). Patients 1.3.2 Previous Arthroscopy
with both diabetes and morbid obesity had a 10% and Intra-articular Injection
rate of PJI. In the study of Pugely et al., risk fac-
tors were body mass index (BMI) > 40, hyperten- Werner et al. [17] found that the incidence of
sion, prolonged operative time, electrolyte infection was higher in patients who underwent
disturbance, and previous infection [2]. TKA within 6 months after knee arthroscopy
A number of smaller studies have supported compared to controls. There was no increase in
these findings and identified further risk factors. infection when TKA was performed more than 6
De Dios and Cordero [15] performed a case- months after knee arthroscopy.
control study comparing 32 consecutive knee Cancienne et al. [18] investigated the associa-
infections with 100 matched controls. The signifi- tion between intra-articular injection of glucocor-
cant factors were classified as pre-, intra-, or post- ticoids and infection in subsequent ipsilateral
operative. Preoperative factors included previous TKA. In patients who had undergone steroid
knee surgery, use of glucocorticoids, immunosup- injections less than 3 months prior to surgery, the
pressants, and a diagnosis of inflammatory rate of infection was higher than controls at both
arthropathy. Intraoperative factors included pro- 3 and 6 months post-TKA (with infection being
longed surgical time, inadequate antibiotic pro- twice as likely as controls at 3 months and 50%
phylaxis, and intraoperative fracture. Postoperative more likely than in controls at 6). If more than 3
factors included wound drainage longer than months had elapsed between injection and TKA,
10 days, the presence of a hematoma, the need for there was no increased risk of infection compared
early reoperations, a diagnosis of deep vein throm- to controls.
bosis (DVT), and the presence of distant infections This association was supported by a study of
in the skin, respiratory, or urinary tract. In the 83,684 patients, of whom 35% had previously
cohort study of Lee et al. [9], significant risk fac- had an ipsilateral injection of corticosteroid [15].
tors included young age and comorbidities such as Those patients who had undergone injection were
diabetes, anemia, thyroid disease, heart disease, stratified into 12 groups on the basis of the num-
lung disease, and prolonged operating time. ber of months prior to TKA that the injection was
performed. The proportion of TKAs who went on
to develop PJI was higher in knees that received
1.3.1 Superficial Wound Infection an injection before TKA than in controls (4.4%
vs. 3.6%; OR = 1.2, p < 0.001), as were the pro-
Properly treated, a superficial postoperative infec- portion who required further surgery for infec-
tion should not lead to later deep PJI. Guirro et al. tion (1.49% vs. 1.04%; OR 1.4, p < 0.001). Time
[16] reviewed a cohort of 3000 TKAs, 63 of to TKA analysis suggested that the effect of
whom were diagnosed with an acute infection. injection remained significant up to 6 months
While 18 of these were considered to be acute with respect to any infection and 7 months with