Perspectives: The Best Prophylaxis for Primary Arthroplasty

, Christian Carulli⁴, Gabriele Ristori⁴, Fabrizio Matassi⁴ and Roberto Civinini⁴

(4)

Orthopaedic Clinic, University of Florence, Largo P. Palagi 1, Florence, 50139, Italy

Christian Carulli

23.2.1 Introduction

An orthopaedic surgery is a “clean” and successful surgery, as widely demonstrated by several long-term series particularly for hip and knee procedures (Sharkey et al. 2002; Ritter 2009; Corbett et al. 2010; Carulli et al. 2012; Toossi et al. 2013). However, the annual worldwide increasing of the number of hip and knee arthroplasties is leading to a proportional rise of the complications: among these, infections to date represent one of the major risks of failure, reaching up to 15 and 25 % of the indications to revision in hip and knee surgery, respectively (Babkin et al. 2007). The implant of a prosthetic knee or hip induces a susceptibility condition of the host to infections. This situation may be very hard to manage, both for patients and surgeons: moreover, these clinical issues are related to high healthcare and social costs, given the necessity of several medical and surgical procedures. This complication may occur with different clinical patterns: an acute infection, within a few days or weeks from surgery, insidious pain, swelling, rash and possibly with a sinus tract or fistula generally present; a chronic postoperative infection, that realizes from months to years after the implant. In this clinical setting, symptoms are usually nonspecific: it is often difficult to distinguish an infection from an aseptic loosening or a case of hypersensitivity to metals or cement (Carulli et al. 2011; Villano et al. 2011; Innocenti et al. 2014). Finally, the hematogenous seeding type of infection of knee and hip implants usually show as a late infection from index surgery, although as acute clinical presentation: this is not directly correlated to the orthopaedic procedure and may also occur decades after surgery. Fortunately, the infection rates in joint arthroplasties are relatively low and decreasing with respect to the last 30 years: the estimated current incidence is 0.5–1.5 % and 1 % for hip and knee implants, respectively (Engesaeter et al. 2003; Blom et al. 2004; Hamilton and Jamieson 2008; Willis-Owen et al. 2010). These improvements have been achieved by advanced prophylactic perioperative strategies and modern antibiotic therapies. Particularly, the advantages of specific antibiotics has been proved by meticulous randomized trials and demonstrated by important reviews comparing antibiotics with placebo (Heath 1991; Al-Buhairan et al. 2008). Despite this evidence, there is still debate on what should be the best antibiotic to be used as prophylaxis in arthroplasty. Recently, two interesting surveys on the clinical practice of large groups of American and Canadian orthopaedic surgeons have shown a significant variability in the choice of the specific antibiotic, timing and duration of the prophylaxis (Berry and Bozic 2010; de Beer et al. 2009). However, there is no possibility to confirm the superiority of a drug over the others. Given the small percentages of infected implants over the million procedures yearly performed, no randomized trial or head-to-head clinical study may reach a statistical power to assign the greater efficacy of an antibiotic with respect to others. The unique reliability is related to date to the results obtained by the best available evidence in the orthopaedic practice (Parvizi and Gehrke 2013).

23.2.2 Analysis of the Literature and Critical Discussion

The basic requirement to reduce the risk of infection in orthopaedic surgery is to perform the procedures in complete respect of a rigorous sterility: several pre- and intraoperative strategies have been proposed, but only few are supported by scientific evidence (Matar et al. 2010). Preoperative hair removal; the use of a double pair of gloves, surgical gowns and body exhaust suits; laminar flow in the operating room; sterile field by a bacteriostatic and/or bactericidal disinfectant; the use of antibiotic-loaded cement; and antibiotic prophylaxis are now established (Jamsen et al. 2010). Other procedures, such as preoperative shower with antiseptic agents and the use of disposable impervious drapes instead of reusable cotton ones, are still now debated (Merollini et al. 2013).

The main established basic strategy is however the antibiotic prophylaxis. The first choice to be addressed is the specific drug, selected on the basis of the potential agents of infection. In orthopaedic surgery, generally the most involved bacteria are represented by Staphylococci (Staphylococcus aureus and epidermidis), followed by Streptococci (Al-Buhairan et al. 2008): specifically, their antibiotic-resistant variants seem to be involved in the septic failures. Several case studies of infected hip and knee implants have been reported in the literature confirming the high rates of bacterial resistance. Phillips et al. reported 4 % of methicillin-resistant Staphylococcus aureus (MRSA) and 25 % of methicillin-sensitive Staphylococcus aureus (MSSA) in his series (Phillips et al. 2006). Pulido et al. (2001) reported a percentage of MRSA and MSSA of 20 % (Pulido et al. 2008), as similar results have been recently reported (Stefánsdóttir et al. 2013).

Carrega et al. (2008) proposed a hypothesis regarding the type of microbial agent and the timing of occurrence of the infections: early infections are usually induced by non-resistant species of Staphylococcus aureus (<3 months), while late infections (>1 year) seem to be correlated to low-virulence bacteria, such as coagulase-negative staphylococci (CoNS). However, the increased number of infections in the interval between 3 months and 1 year is related to resistant species:

Methicillin-resistant CoNS were identified in 30 % of early infections, in 24 % of delayed infections and in 17 % of late infections.
MRSA were isolated in 13 % of early infections, in 22 % of delayed infections and in 15 % of late infections.

Therefore, the antibiotic prophylaxis must be essentially focused on Staphylococci.

As mentioned, given the current low rates of infection (nowadays decreasing with respect to the 1970s), it is not feasible to conduct a randomized controlled trial that may directly compare the efficacy of two or more antibiotics: thus, there is no evidence of a superiority of an antibiotic over another (Glenny and Song 1999; Al-Buhairan et al. 2008).

The most commonly used and studied antibiotics are first- and second-generation cephalosporins (cefazolin and cefonicid, respectively). They belong to the group of β-lactam molecules with a time-dependent bactericidal action, acting by an irreversible inhibition of the enzyme responsible of the bacterial cell wall integrity. Their pharmacokinetic features as the high water solubility, the elevated tissue penetration, the renal elimination, the absent or negligible toxicity and the half-life (0.5–1.5 h) allow a flexible use in this type of surgery. On the other hand, they have a narrow spectrum of action, mainly active on Gram-positive bacteria (except enterococcus), and in the case of the second-generation cephalosporin, some Gram-negative bacteria (Escherichia coli, Klebsiella pneumoniae, Proteus mirabilis). Recent studies revealed a significant increase of antibiotic resistance particularly MRSA but also Staphylococcus epidermidis and other methicillin-resistant CoNS (Al-Maiyah et al. 2005; Fulkerson et al. 2006). For these reasons, the use of drugs with a broader spectrum of action as the third-generation cephalosporins (ceftriaxone), extending an action on difficult Gram-negative bacteria, such as Enterobacter and Pseudomonas species, has been proposed. Similarly, the fourth-generation cephalosporins (Cefepima) showed an enhanced activity on Gram-negative bacteria, maintaining their action on Gram-positive species. These drugs have comparable pharmacokinetic features, but present a longest half-life (ceftriaxone, 4 h; Cefepima, 2 h) with respect to first generation. However, they present an increased risk of colitis by Clostridium difficile, associated with a higher mortality risk (from 3.5 to 15.3 %) with respect to the past (Crabtree et al. 1999). For this reason, it is mandatory to reasonably administer the antibiotic prophylaxis in hip and knee surgery to avoid this dangerous complication (Jenkins et al. 2010).

The other major class of antibiotics proposed for the orthopaedic prophylaxis is represented by glycopeptides (vancomycin and teicoplanin). The management of these drugs is absolutely more complex, given the high risk of adverse reactions and limitations. They are almost completely eliminated by the kidneys, and this explains the high rate of (reversible) nephrotoxicity (5 %), and the contraindications in patients with renal insufficiency or in combination with other nephrotoxic drugs. Moreover, vancomycin (but not teicoplanin) may induce a non-immune-mediated release of histamine (5–10 %) in cases of too quick intravenous administration: tachycardia, flushing, tingling, itching, erythema, deafness, tinnitus and in exceptional cases severe arterial hypotension up to shock are reported. Despite these limitations, they are characterized by good pharmacokinetic properties, with long half-life (about 6 h for vancomycin, even more for teicoplanin) and good tissue penetration. Their action is held on all Gram-positive bacteria (except teicoplanin that is not active on enterococcus), and Gram-negative bacteria, acting by an irreversible block of the synthesis of bacterial cell wall. Their spectrum of activity also includes MRSA and methicillin-resistant CoNS, so they are considered “second-line” drugs, reserved for cases of β-lactam resistance or β-lactam allergy. In local ecosystems with specific risks for MRSA infections, it is reasonable to adopt an antibiotic prophylaxis with vancomycin (teicoplanin is more expensive) as “first-line” drugs (Meehan et al. 2009 Smith et al. 2012) with very satisfactory results. Some authors found a recent rise of MRSA and CoNS species with a reduced susceptibility to vancomycin, especially vancomycin-resistant enterococci (VRE) with potential serious consequences (Nixon et al. 2006; Gemmel et al. 2006; Kock et al. 2010). A recent review has concluded that there is insufficient evidence to identify a threshold beyond which it is advisable to switch to routine prophylaxis with glycopeptides (Cranny et al. 2008). The reasonable conclusion is that the analysis of the environmental risks and the patients’ features are the main factors to be considered in the choice of the type of drug (Furuno et al. 2006). To date, it has to be considered an increased risk of:

A previous MRSA infection
Local environments with a high risk of contamination (elderly communities, hospitalizations in the previous 6–12 months)
Antibiotic therapy in the previous 30 days, in particular oral fluoroquinolones
Diabetes mellitus
Haemodialysis
Ulcers or unhealed surgical wounds
Healthcare workers

The first topic (MRSA healthy carriers or shedders) has been the subject of many studies and investigations: in particular, the presence of Staphylococcus aureus in nasal colonies in patients undergoing a knee or hip arthroplasty has been evaluated in several papers. It is estimated that 95 million people in the USA are carriers of bacterial colonies inside their nostrils, and in particular, 2.5 million are represented by MRSA species. In these patients, a risk of infection of 2–9 times higher compared to patients without this contamination has been calculated. A dedicated study showed that a prophylaxis with nasal mupirocin 5 days prior the intervention in patients with a positive nasal swab was not correlated to a reduction of postoperative infections (Kalmeijer et al. 2002). On the other hand, two papers showed that this type of prophylaxis in association with a preoperative intravenous prophylaxis showed a reduction of infection rate (Rao et al. 2008; Hacek et al. 2008). Currently, nasal decontamination in patients with positive nasal swabs is considered as a good practice (Goyal et al. 2013).

Another debate is on the usefulness of a second antibiotic in the prophylaxis (Sewick et al. 2012). reported on the efficacy of the association of vancomycin and cefazolin: the result was a reduction of the rates of infection by MRSA, while the overall incidence of infections was similar to the single-antibiotic protocol. However, given the large number of patients to be treated to prevent a single case of infection, it is not recommended to extend the indications of a double-antibiotic prophylaxis.

The current trend is to provide the ordinary administration of a first- or second-generation cephalosporin. In selected cases of high risk of MRSA infection as provided by the international guidelines, the choice of a glycopeptide is indicated, as demonstrated by a recent comparative study with cefuroxime, fusidic acid and vancomycin: no superiority of an antibiotic over the others was found, except for vancomycin and MRSA infection (Tyllianakis et al. 2010) (Table 23.1).

Table 23.1

International guidelines for the antibiotic prophylaxis in orthopaedics

	AAOS	SIGN	SIOT
1° choice	Cefazolin Cefuroxime	Ceftriaxone (or a low-cost equivalent antibiotic)	Cefazolin
2° choice	Vancomycin Clindamycin (in case of allergy to cephalosporin or of risk for MRSA infection)	Glycopeptides (in case of MRSA infections)	Vancomycin (in case of allergy to cephalosporin or of high risk for MRSA infection)

AAOS American Academy of Orthopaedic Surgeons. Recommendations for the use of intravenous antibiotic prophylaxis in primary total joint arthroplasty (2004)

SIGN Scottish Intercollegiate Guidelines Network. Antibiotic prophylaxis in surgery: a national clinical guideline (2008)

SIOT Società Italiana di Ortopedia e Traumatologia. Profilassi antibiotica perioperatoria nella chirurgia protesica di anca e ginocchio (2011)

Regarding the timing of administration, it is well known that it is fundamental to reach the maximum blood concentration before surgery. Provided that the intravenous administration is the most efficient and useful, the ideal concentration should be achieved before the incision or inflation of the tourniquet, depending on the specific pharmacokinetics of the chosen antibiotic. There is a universal agreement on the fact that it is a crucial step of the procedure (Table 23.2), given the evidence that an inadequate timing of administration is correlated to a rise of the risk for infections (Classen et al. 1992; van Kasteren et al. 2007a, b).

Table 23.2

International guidelines for the timing of the antibiotic prophylaxis in orthopaedics

Only gold members can continue reading. Log In or Register to continue