Prevention of Orthopedic Prosthetic Infections Using Evidence-Based Surgical Site Infection Care Bundles: A Narrative Review

Pre-operative bathing/cleansing

Topical antiseptic skin preparation reduces the burden of skin flora at the incisional site. Standardized regimens of two pre-admission showers, or skin cleansing, with either 4% aqueous or 2% chlorhexidine gluconate (CHG) impregnated polyester cloths, results in skin surface concentrations of CHG exceeding the minimum inhibitory concentration (MIC) required to inhibit or kill skin staphylococci, including methicillin-resistant Staphylococcus aureus (MRSA) (Table 1) [29,30]. Although CHG pre-operative bathing/cleansing is controversial, a reduction in SSIs has been reported after hip arthroplasty when patients completed a standardized pre-operative CHG skin cleansing regimen [31]. Three systematic reviews have found that pre-operative CHG shower/cleansing, with or without administration of mupirocin nasal ointment, results in a reduction of SSI [32–34].

Staphylococcal screening and nasal decolonization

Staphylococcus aureus is the most common microbial pathogen responsible for incisional SSIs and approximately 20% of Staphylococcus aureus isolates harvested from SSIs are MRSA, according to the National Health and Safety Network (NHSN) surveillance system of the CDC [35]. Between 15% and 30% of healthy adults are nasally colonized by methicillin-susceptible Staphylococcus aureus (MSSA), whereas 1% to 3% are nasally colonized with MRSA [36].

Screening patients for both MSSA and MRSA colonization in the pre-admission period, particularly in high-risk surgical procedures such as implantation of an orthopedic device, followed by a prescribed nasal and skin surface decolonization protocol prior to surgery has been shown to be effective. Three nasal decolonization regimens are available: mupirocin ointment, nasal application of 5% or 10% povidone iodine, or application of an alcohol-based nasal antiseptic.

Nasal mupirocin is the most widely used topical antibacterial agent. An RCT that compared mupirocin to placebo found that 83% of the patients in the mupirocin group had staphylococcal suppression, compared with only 27% in the placebo group (p = 0.001). Two systematic reviews and meta-analyses showed a protective effect of mupirocin decolonization, especially among cardiac, orthopedic, and neurosurgical patients [33,37]. It has been suggested that continued use of mupirocin leads to the emergence of resistant organisms, but short-term use does not appear to be associated with a significant risk of mupirocin resistance [38]. The typical decolonization protocol for patients who are positive for MSSA or MRSA involves receiving intranasal mupirocin ointment twice daily for five days prior to surgery in combination with a minimum of two standardized showers, on the night before and morning of surgery.

Povidone iodine (PI) has been widely used as a peri-operative antiseptic on skin, open wounds, and mucous membranes. Povidone iodine is broad-spectrum and active against both gram-positive and gram-negative wound pathogens. A randomized, placebo-controlled, involving 429 patients undergoing primary or revision total joint arthroplasty (TJA), found that a 5% PI-based nasal antiseptic was more effective at nasal decolonization of Staphylococcus aureus four hours after treatment (p = 0.003) compared with 10% PI swab sticks or saline, but there was no statistically significant difference in Staphylococcus aureus colonization at 24 hours [39]. Another trial of twice-daily nasal mupirocin for five days before surgery, compared with two applications of a 5% nasal PI solution within two hours of surgical incision, in patients undergoing arthroplasty or spine fusion surgery found that Staphylococcus aureus-related deep SSIs developed in five patients (0.66%) who received mupirocin and in no patients who received PI (p = 0.03). However, the proportion of negative nasal cultures at 24-hours postoperative was 92% (78 of 85 patients) for those assigned to mupirocin compared with 54% (45 of 84 patients) for those assigned to PI. This study suggests that mupirocin may suppress Staphylococcus aureus colonization over a 24-hour period, whereas PI likely suppresses Staphylococcus aureus during the peri-operative period [40]. However, mupirocin is taken BID over a five-day period that may adversely impact patient compliance compared with the application of PI in the pre-operative holding area prior to surgery.

Alcohols such as isopropanol or ethanol are active against most gram-positive and gram-negative vegetative bacteria, viruses, or fungi, with optimal antimicrobial activity at concentrations between 60% and 90% [41]. A formulation of intra-nasal antiseptic consisting of 70% ethanol was shown to be effective in reducing Staphylococcus aureus nasal colonization among health-care workers in an RCT [42]. At present, studies documenting the benefit of this pre-operative decolonization strategy are limited; one involves a single-center quasi-experimental study of patients undergoing spine surgery [43]. Alcohol-based intra-nasal antiseptic was applied immediately pre-operatively, and then three times a day during hospitalization, and for five to seven days after discharge from hospital. The study reported an 81% decrease in mean SSI rates for the post-implementation nine-month period. The study limitations were a small number of enrolled spine patients and failure to document which type of SSIs was prevented (superficial or deep incisional). A recent systematic review and meta-analysis has found that regardless of the nasal decolonization strategy, pre-operative Staphylococcus aureus screening/decolonization protocols lowers the risk of infection after elective orthopedic and TJA procedures [44]. Further studies are warranted to determine the optimal clinical and cost-effective methodology.

Smoking cessation

Smoking cessation a minimum of four to eight weeks prior to surgery reduces the risk of SSI. Compared with never smokers, current smokers (smoking within one year of surgery) have higher odds of developing a superficial (odds ratio [OR], 1.30; 95% confidence interval [CI], 1.20–1.42) or deep (OR, 1.42; 95% CI,1.21–1.68) incisional SSI [45]. Nicotine and toxins in cigarette smoke adversely impact angiogenesis by impairing oxygen delivery to tissues, interfering with the reparative process at each stage of wound healing. Tissue hypoxia causes poor wound healing and a higher risk of infection [46,47]. The adverse impact of smoking on post-operative outcome in both current and former smokers was reported in a large national database study [48]. Current smokers have a higher rate of complications than former smokers [49,50].

Normothermia

Peri-operative hypothermia, defined as a core body temperature <36°C, can occur at any stage in the peri-operative period and is observed in one-third of patients despite the use of active warming techniques [51,52]. The consequences of peri-operative hypothermia include increased intra-operative blood loss, increased risk of SSI and pressure injury, increased length of ICU and overall hospital stay, decreased patient comfort, and increased rates of cardiac events. Prolonged hypothermia impacts macrophage and phagocytic cell function and diminishes the cell-mediated immune response [53]. Patients who are most susceptible to heat loss include the elderly, those with a high anesthetic risk (American Society of Anesthesiologists [ASA] grade 3 to 4), cachexia, burns, hypothyroidism or cortico-adrenal insufficiency.

A prospective review of 493 patients found that hypothermia occurred in 105 patients (21.3%) prior to the induction of anesthesia and, as a consequence, measurement of core temperature should be mandatory 60 to 120 minutes before induction [54]. Intra-operative hypothermia was found to occur in 17% of patients undergoing operative intervention for hip fractures [55]. It was suggested that pre-warming patients 10–30 minutes prior to surgery may reduce the risk of hypothermia [56]. Monitoring of core temperature and maintenance of normothermia peri-operatively should be standard surgical care.

Strategies to maintain normothermia include passive and active patient warming. Passive systems involve layers of insulation, such as the use of blankets or other types of surgical draping. There are many active warming systems available, which include circulating water mattresses, forced air, resistive polymer heating, radiant warmers, airway heating and humidification, and internal warming methods. The 2017 CDC Guideline for the Prevention of Surgical Site Infection recommends maintaining perioperative normothermia (Category IA–strong recommendation; high to moderate–quality evidence) [23].

Antimicrobial prophylaxis

The choice of an antibiotic for surgical prophylaxis should reduce the risk of post-operative SSI, SSI-related morbidity and mortality; reduce the duration and cost of health care; produce no adverse effects; and have no adverse consequences for the patient's endogenous microbiome or the hospital [57,58]. To achieve these goals, the antibiotic should be active against those pathogens most likely to contaminate the surgical site, given in an appropriate dosage and at a time that ensures adequate serum and tissue concentrations during the intra-operative period, and administered for the shortest effective period to minimize development of microbial resistance.

Antibiotic agents should provide serum and tissue concentrations exceeding the minimum inhibitory concentration (MIC) for organisms encountered during the procedure, at the time of incision, and for the duration of the procedure. The majority of orthopedic SSIs are caused by gram-positive cocci and the predominant responsible organisms are normal skin flora, including Staphylococcus aureus and coagulase-negative staphylococci.⁵⁸ The optimal timing for administration of antibiotics is within 30–60 minutes of surgical incision. However, some agents, such as fluoroquinolones and vancomycin, require administration over a one- to two-hour period; therefore, administration of these agents should begin within 120 minutes of the surgical incision [57].

To ensure that adequate serum and tissue concentrations of antimicrobial agents for prophylaxis of SSIs are achieved, antibiotic-specific pharmacokinetic and pharmacodynamic properties and patient factors need consideration when selecting an appropriate dose. In obese patients (body mass index [BMI] > 30), and morbid obesity (BMI >40), antibiotic serum and tissue concentrations of some drugs may differ from those in non-obese patients because of pharmacokinetic alterations that depend on the lipophilicity of the drug and other factors [59,60]. The pharmacokinetics of drugs may be altered in obese patients, so dosage adjustments based on body weight may be warranted in these patients. Intra-operative re-dosing is needed to ensure adequate serum and tissue concentrations of the antibiotic if the duration of the surgical procedure exceeds two half-lives of the antibiotic or there is excessive blood loss (i.e., >1,500 mL) [57,60]. The re-dosing interval should be measured from the time of administration of the initial pre-operative dose, not from the beginning of the procedure. Of note, the WHO and CDC guidelines all strongly recommend against continuing prophylactic antibiotic dosing beyond incision closure even in the presence of surgical drains [23,24,61–63].

Glycemic control

Diabetic hyperglycemia at the time of TJA is a significant risk factor for post-operative infection [64,65]. Diabetes-induced hyperglycemia, or stress (non-diabetic)-induced hyperglycemia, impairs leucocyte function causing patients to be immunocompromised and increasing the risk for superficial and deep incisional SSIs as well as overall mortality [66]. A meta-analysis has reported an odds ratio of 2.04 with a 95% confidence interval 1.69 to 2.46 of increased infection rates among patients with diabetes mellitus [67]. It is recommended that hemoglobin (HbA_1c) values be measured pre-operatively since this accurately documents the level of glycemic control over a three-month interval. Patients who present with a HbA_1c value of <7.0 have a lower risk of infection compared with patients who have values >7 [68]. However, blood glucose measurement in the pre-operative period is a valid determinant of glycemic control in patients with diabetes mellitus. Patients with uncontrolled diabetes mellitus exhibit increased odds of surgical and systemic complications, higher mortality, and increased length of stay during the index hospitalization following lower extremity TJA [65]. Although the primary focus has been on reducing the risk of infection in the patient with diabetes mellitus, stress-induced post-operative hyperglycemia can be problematic in patients without diabetes mellitus undergoing lower limb arthroplasty [69].

Peri-operative skin antisepsis

The purpose of topical antiseptic preparation of the skin is to reduce the burden of skin flora at and adjacent to the incisional site, thereby reducing the likelihood of SSI caused by contaminating skin flora. There are two major classes of skin antiseptic agents commonly used: chlorhexidine-based agents and iodophor-based agents [70]. These two classes are further divided into agents that include alcohol—typically isopropyl alcohol (IPA)—and those that do not. A systematic review of chlorhexidine-based antisepsis compared with iodophor-based antisepsis found CHG to be the superior agent [71]. A meta-analysis from the Cochrane Collaboration analyzed 13 RCTs with 2,632 total patients that compared pre-operative skin preparation agents in elective clean surgery. Eleven different comparisons were made in these 13 studies; five studies compared iodine-containing preparations with chlorhexidine-containing preparations. A mixed-treatment comparison-analysis concluded that alcohol-containing preparations had the highest effectiveness in reduction of the risk of SSI. There was a 78% probability that 4% chlorhexidine scrub-70% IPA was the best preparation for preventing SSIs, followed by PI plus IPA [72]. Analysis has shown that alcohol-based preparations have the highest likelihood of being most effective at reducing SSIs, whether combined with chlorhexidien or with PI [73]. Based on a recent RCT, some investigators have considered dual preparation of the skin before and after surgical draping because contamination can occur after this procedure [74,75].

Hyperoxygenation

Peri-operative supplemental oxygen or hyperoxia increases tissue oxygen tension, which may lead to an increase in oxidative killing of surgical pathogens and a reduction in SSIs [76,77]. Initially, incised tissue is hypoxic, a condition that acts as a stimulus for repair. The production of reactive oxygen species (ROS) and correction of hypoxia with increased oxygen availability achieves optimum wound repair. During the healing process, resistance to microbial pathogens depends largely on oxygen as a substrate for production of ROS. In vitro data have suggested that hyperoxia has additional cellular and immunologic effects such as enhancement of intra-cellular killing by increased production of ROS and attenuation of the proinflammatory cytokine response [78]. However, the potential clinical harm that is suggested to be associated with hyperoxia, includes increased pulmonary complications, impairment of glucose regulation, and increased systemic vascular resistance resulting in a decreased cardiac output [79].

Regardless of the level of controversy surrounding this risk-reduction strategy, the recommendation for hyperoxia was given a boost when the CDC recommended supplemental oxygen to reduce infection risk in the 2017 Guidelines for the Prevention of Surgical Site Infection [22]. However, a recent large robust British clinical trial found that despite recent World Health Organization and CDC guidance, clinicians should not give routinely give supplemental oxygen, to prevent infections and healing-related complications after a major intestinal surgery [24,80]. There is currently no consensus for the administration of enhanced intra-operative/post-operative oxygen therapy in patients undergoing hip or knee arthroplasty, but the consideration warrants further study.

Wound irrigation

The cleansing of wounds to prevent SSI and promote wound healing can be traced back to 2200 BC [81]. Surgical irrigation is widely used to maintain tissue moisture, remove cellular debris, and to render the wound free of foreign materials. Intra-operative irrigation may be used independent of wound classification but there is little agreement among surgical specialties on the selection of irrigation solution, what additives be used (e.g., antibiotic agents or antiseptics), the volume necessary to irrigate the wound, and what is the safest method of delivery. The American College of Surgeon (ACS) defines low-pressure delivery devices as those that deliver solution in the 1–15 psi range, and high-pressure devices as those that can deliver from 15–35 psi [82]. Most studies have demonstrated that higher pressure irrigation is more effective in removing foreign materials and bacteria from the wound bed but may cause tissue damage and spread of bacteria deeper into bone and tissue [83,84].

Normal saline is the most common wound irrigation fluid used in surgical procedures [85]. Clinicians have added antibiotic agents, antiseptics, or surfactant to the solution to enhance the effect of irrigation. There are few well-designed clinical trials that have documented the benefit and efficacy of additives [82,86]. The practice of using additives is based on clinician preference and specialty. Bacitracin was used off-label until 2019 when the Food and Drug Administration (FDA) issued a request to manufacturers to withdraw it from the market because if its risks (nephrotoxicity and anaphylactic reactions) [87]. Antibiotic agents are often used off-label but the World Health Organization recommends against using antibiotic agents for incisional irrigation because of the low-level quality of evidence and the potential for the development of antibiotic resistance [24]. Furthermore, the effectiveness of an antibiotic against bacterial contamination requires a prolonged exposure time, which is counterintuitive to the pragmatic nature of intra-operative irrigation.

Alternatively, antiseptics have a rapid mechanism of action, requiring a minimal contact time. Povidone iodine is the most commonly used antiseptic additive for irrigation. A meta-analysis, including 24 RCTs (5,004 patients), compared intra-operative PI lavage with no PI in patients who were undergoing surgery. The SSI rate was 8.0% in the PI group and 13.4% in the control group (relative risk [RR], 0.58; 95% CI, 0.40–0.83; p = 0.003) [88]. Povidone iodine at a concentration of 0.85% is recommended as an additive to intra-operative irrigation fluid and often use in the orthopedic practice prior to wound closure.²³ An FDA-approved alternative for wound cleansing is 0.05% CHG. CHG at a concentration of 0.05% has emerged as a possible antiseptic alternative to PI as an intraoperative irrigation agent. Chlorhexidine gluconate possesses several positive characteristics suggesting that it is an ideal irrigation additive. Chlorhexidine gluconate has broad-spectrum activity against both gram-positive and gram-negative aerobic and anaerobic surgical wound pathogens. Unlike PI, CHG is not inactivated in the presence of blood or tissue protein, and it is non-toxic to granulation tissue and wound healing at low concentrations. Furthermore, CHG binds to tissues, retaining its bioactivity long after initial application [89,90].

Antimicrobial suture wound closure

The intrinsic benefit of antimicrobial suture wound closure for all layers of an incision relates to the documented antimicrobial activity against both gram-positive and gram-negative surgical wound pathogens [91,92] When considering the benefits of antimicrobial suture wound closure, two questions need answering. First, are sutures placed in the surgical site a potential nidus for infection? Traditional (non-antimicrobial) braided or monofilament sutures, excised from the infected wounds of surgical patients, were shown to have an established microbial biofilm in 100% of cases. Implanted sutures, and other biomedical devices, are at high risk for microbial biofilm formation, and subsequent risk of SSI [93]. Second, does the level of evidence for antimicrobial sutures justify their inclusion in current, evidence-based surgical care bundles? Numerous clinical studies including multiple RCTs, systematic reviews and meta-analyses, have shown that antimicrobial sutures compared with traditional, non-antimicrobial absorbable braided or monofilament wound closure technology for fascia, muscle, subcutaneous tissues, and skin reduces the risk of SSI in all surgical disciplines [94–96] A double-blinded, randomized trial that compared triclosan-coated sutures with non-antimicrobial sutures found a 35% reduction in superficial/deep incisional SSIs in patients undergoing TJA [97]. The findings were not deemed significant by the authors because they had set an SSI target reduction of 60%. Critical review of this finding suggests that the study was underpowered to achieve a 60% reduction. A sensitivity analysis published in 2019 documented a significant SSI reduction benefit for clean, clean-contaminated, and contaminated surgical procedures [98].

Two studies that estimated the economic benefits of using antimicrobial sutures for wound closure documented a reduction in cost for superficial and deep incisional SSIs when antimicrobial (triclosan-coated or -impregnated) sutures were used [99,100]. The use of antimicrobial sutures for wound fascia and subcuticular wound closure is supported by level 1A clinical evidence and recommended by multiple national, international, and societal SSI prevention guidelines [22–24,101–103].

Mechanistically, what evidence exist to document that a suture can be a nidus for infection? The classic study by Elek and Cohen [104] demonstrated that when a surgical incision is closed in the presence of a foreign body such as a suture, the presence of that device can reduce the inoculum burden required to produce a wound infection by a factor of 1,000. Over a period of days, weeks, and even months, microbial surface contamination produces a biofilm on the suture that can then detach, spreading bacterial contamination across multiple tissue planes, including superficial, deep incisional, organ/space, and in the case of an orthopedic implant, leading to a peri-prosthetic (device-related) infection [105]. Antimicrobial (triclosan) sutures reduce the risk of bacterial colonization on the surface of the device, regardless of inoculum burden, preventing biofilm formation and further microbial dissemination within the wound bed.

Post-operative incision care

Presently, there are many surgical dressings available, with supportive publications that document benefit. Unfortunately, many of these studies are small, retrospective, and with less than 100 patients and although individual dressings may promote selected clinical benefits, no single dressing provides all of the desired clinical benefits. Most decision-making by healthcare professionals on the type and use of post-operative surgical dressing relies on local expertise as the scientific evidence for their use is limited. Despite there being a lack of adequately powered RCTs, the 2008 UK National Institute for Health and Clinical Excellence (NICE) guidelines advocated that post-operative surgical incisions should be covered with an appropriate interactive dressing depending on physician opinion [106].

The 2016 WHO guideline has advocated that simple, or advanced, surgical dressings should not be used on primarily closed surgical incisions with the aim of reducing the risk of SSI [24]. Their recommendation was conditional, with low quality of evidence, based on ten RCTs. The Cochrane Collaboration found little evidence for any value of post-operative surgical dressings to reduce the risk of SSI [107,108].

The range of dressings available to cover surgical incisions includes low adherence dressings such as absorbent cotton pads placed in contact within the incision. More advanced dressings, such as vapor-permeable adhesive polyurethane film dressings, allow incision inspection and the passage of water vapor and oxygen but are impermeable to water and micro-organisms. Hydrocolloids and foam dressings acquire the shape of the incision and can be left in place for several days. Silicone polymers, adherent or non-adherent, are often used for burn care, but hydrogels, alginates, and other dressings have a wider application for long-term open incisions that are healing by secondary intention or for incisions that have dehisced.

The inclusion of antiseptics in surgical dressings is a theoretical and logical method to prevent early biofilm formation in incisional wounds but well-designed, well-powered RCTs are required. Many peer-reviewed publications exist that support the benefits of using a post-operative antiseptic dressing: these include silver, honey, CHG, PI, and polyhexamethylene biguanide. However, few of these publications are suitable for inclusion into systematic reviews and meta-analyses because they are non-comparative, retrospective, underpowered, biased, or not blinded. Lower rates of acute PJI have been demonstrated in two retrospective analyses of patients receiving silver-impregnated gauze dressings compared with standard gauze dressings (OR, 0.165 and 0.092) [109,110]. The WHO recommends not using any type of advanced dressing but advocates the use of negative-pressure wound therapy (NPWT) in high-risk wounds based upon a systematic review and meta-analysis that shows benefit [24].

The use of prophylactic NPWT as an incisional device/dressing to help prevent SSI or wound dehiscence is a new risk-reduction strategy that has been supported by an increasing number of RCTs and systematic review and meta-analyses (SR&Ms) [111–114]. Incisional NPWT has broad application across the surgical spectrum and may suppress or delay formation of biofilm, which may very well be the basis for its success [111]. NPWT should be considered to prevent SSIs in incisional wounds, which are at particular risk, such as contaminated wounds after colorectal surgery and where the risk of SSI is high; open heart surgery, after which sternal dehiscence can be a life threatening disaster; cesarean delivery after which dehiscence is common despite being a clean-contaminated procedure; breast surgery in which SSIs or dehiscence may incur delays for radiotherapy or chemotherapy; and orthopedic prosthetic surgery, when joint salvage can be a prolonged and expensive complication.

Antibiotic-loaded cement

The practice of putting antibiotic agents into bone cement is now approaching almost 40 years of use. However, the practice is not without controversy and no national or international SSI prevention guidelines recommend its use. Antibiotic-loaded bone cement (ALBC) is widely used to treat orthopedic infections based on the rationale that high-dose local delivery is essential to eradicate biofilm-associated bacteria. However, ALBC formulations are empirically based on drug susceptibility from routine laboratory testing, which is known to have limited clinical relevance for biofilm eradication [115]. Several organizations and investigators have suggested that the use of antibiotic-loaded bone cement results in a lower risk of infection following total hip arthroplasty [116–118]. However, more recent data from two national joint registries, and published findings from an institutional database, document no substantial benefit associated after the use of antibiotic-loaded bone cement compared with plain cement [119,120]. A significant issue that has not been adequately addressed following the use of antibiotic-loaded bone cement is the role it may have on the emergence of antibiotic resistance [121].

Protective suits

From a historical perspective, the Charnley-type negative pressure, exhaust suit has been shown to reduce the risk of infection after arthroplasty compared with modern exhaust suits [122]. The primary concern associated with the modern positive pressure, exhaust suit involves the interface connections such as the gown–glove interface or the hood–gown interface that can result in the release of contamination within the surgical field [123.124]. In an International Consensus on Orthopedic Infections, the authors concluded that the use of personal protection suits does not reduce the rate of infection in patients undergoing TJA [125].

Topical vancomycin powder

The rationale for the use of topical vancomycin powder prior to incision closure is to provide a high concentration of an antibiotic that is active against gram-positive cocci (Staphylococcus aureus and coagulase-negative staphylococci). Although systematic reviews and meta-analyses appear to support the use of intra-incisional vancomycin powder, it is controversial [126–129]. There are several concerns associated with the routine use of intra-wound vancomycin powder that include anaphylaxis, potential renal failure and, more specifically, the emergence of resistance of gram-negative and resistant polymicrobial SSIs [130,131]. However, two recent meta-analyses have suggested that intra-wound vancomycin can provide a substantial benefit in reducing the risk of post-arthroplasty infection [132,133]. Although the favorable use of this strategy appears to be gaining support within both the total joint and spinal disciplines, further studies are warranted to assess unintended consequences that may be associated with this clinical practice.