​Antimicrobial Therapy in Cardiac Surgical Patients
Mary Elizabeth Sexton, MD, Abd Al Rahman Anani, Nadine Rouphael, MD

Introduction
Post-operative infections are an important consideration in cardiac surgery, given the potential for patient morbidity and mortality, increased length of stay, and increased hospital costs [1]. 
 
The importance of preventing post-operative infections when possible and addressing them quickly when they do occur was underscored by a multicenter study conducted by the Cardiothoracic Surgery Clinical Trials Network in 2010. Researchers followed a cohort of more than 4000 patients to determine the prevalence of post-operative infections and the associated cost, and found that almost 3% of the patients had an infection following surgery that was considered healthcare-associated (including but not limited to surgical site infections). These infections were more likely to occur in patients who underwent more complex surgeries, particularly cardiac transplantation and LVAD placement. Patients who developed an infection had their mean length of stay increase by 14 days, and had an adjusted mean hospitalization cost that was almost $40,000 higher [1].
 
There has therefore been significant focus on preventing these infections via preoperative, intraoperative, and postoperative interventions.  In this chapter we will discuss recommendations for selection, timing, and duration of perioperative antibiotic therapy, the use of topical cleansing agents and antibiotics including for Staphylococcus aureus decolonization, and the importance of adjunctive measures such as maintaining normothermia, adequate oxygenation, and blood glucose control. Despite best practices, however, infections still occur in a small percentage of patients, and so this chapter will also discuss management of sternal wound infections, mediastinal infections, and cardiac device infections (including pacemakers and automated defibrillators). 
 
Prevention of Post-Surgical Infections
Evidence Behind Perioperative Antibiotic Use
Multiple studies have demonstrated that administration of antibiotics in the perioperative period is beneficial in decreasing the risk of surgical site infections[2, 3]. A 2002 study collected data on numbers of surgical site infections over the course of a year at 44 hospitals, during which time the hospitals devised programs to improve their compliance with guidelines for perioperative antibiotic administration. At these hospitals, there was a decrease in infection risk from 2.28% to 1.65% seen in association with improved compliance [4]. 
 
This benefit has also been seen in less invasive procedures, such as pacemaker or cardiac defibrillator insertion.  Permanent pacemaker insertion can be complicated by infection of the device pocket, which has been documented in up to 5% of cases. A meta-analysis was therefore performed in 1998 to evaluate whether administration of peri-procedural antibiotics was effective.  Seven randomized trials were included, and there was a statistically-significant decrease in infections, including wound infections at the implantation site and erosion of the device leads, when antibiotics were given at the time of device placement [5]. A larger randomized trial of 649 patients who had either a defibrillator or a pacemaker inserted was then performed from 2003-2005, with patients receiving 1 gram of cefazolin or a placebo dose of saline prior to the procedure. The study was stopped prior to reaching the planned enrollment of 1000 patients because the use of cefazolin appeared to significantly decrease the risk of infection, from 3.28% in the group who did not receive cefazolin to 0.63% in the group who received prophylaxis [6].
 
Antibiotic Selection
For most cardiac procedures, use of a first or second generation cephalosporin is recommended (typically cefazolin or cefuroxime), with cefazolin as the sole first-line recommendation for cardiac transplant patients [3].
 
Investigations have been performed to determine whether antibiotic selection should differ for different cardiac procedures, and have supported following these general recommendations for all types of cardiac surgery. For example, a review of 10 studies of peri-operative antibiotic use in LVAD placement found that multiple different antibiotics have been tested for prophylactic use, including cephalosporins, vancomycin, and fluroquinolones.  The authors found relatively similar rates of post-operative infections (although they note this was difficult to determine given differences both in criteria for defining infection and in how patients were monitored after their procedures), and they determined that the majority of the infections that occurred were secondary to gram-positive bacteria. The authors therefore recommended utilizing a first-generation cephalosporin in most cases, with substitution of vancomycin in patients with an increased likelihood of MRSA (see below), which is in line with the overall cardiac surgery guidelines [7].
 
 
When to Consider Methicillin-Resistant Staph Aureus (MRSA) Coverage
Vancomcyin remains a second-line choice for surgical prophylaxis in the most recent guidelines.  However, the 2013 guidelines published by the Infectious Diseases Society of America (IDSA), the Society for Healthcare Epidemiology of America (SHEA), the American Society of Health-System Pharmacists, and the Surgical Infection Society note that there are select situations in which it might be considered as part of first-line therapy. These include when patients have a documented history of MRSA infection or colonization, when there is a known MRSA outbreak at the facility where the surgery is taking place, or when patients would be high-risk for MRSA colonization. There are no definitive criteria for defining a high-risk patient, although patients with significant healthcare exposures would be of concern [3, 8].
The guidelines also stress that even if vancomycin is indicated, it should usually be given in addition to a cephalosporin, rather than as a replacement for one. They note that cefazolin may have higher efficacy in decreasing infections secondary to methicillin-susceptible Staph aureus (MSSA), as well as infections secondary to gram-negative pathogens [3].
 
Antibiotic Choice in Patients with Allergies
The 2013 guidelines for surgical prophylaxis also address the issue of patients who have a history of an allergy to b-lactam antibiotics, as these patients may have a contraindication to cephalosporin use. Patients who have a history of severe reactions to either a cephalosporin or a pencillin, including anaphylaxis, other airway compromise, toxic epidermal necrolysis or Stevens-Johnson syndrome, or urticaria, should not be given any of the cephalosporins or a carbapenem.  In these cases, either vancomycin or clindamycin can be considered as an alternate option [3]. 
 
 
Antibiotic Timing and Duration
SHEA published additional guidelines for hospital use in preventing surgical site infections in 2014 in collaboration with the American Hospital Association, IDSA, the Joint Commission, and the Association for Professionals in Infection Control and Epidemiology.  These guidelines supported prior recommendations [2, 3] for administration of prophylactic antibiotics within the hour prior to surgery [8]. If either a fluoroquinolone or vancomycin is given, it is acceptable to start the antibiotic infusion two hours prior to surgery in order to allow adequate time for it to finish [3].
 
 These guidelines also recommend cessation of antibiotics by 24 hours post-surgery.  The authors note that on review of multiple prior studies, there is limited evidence that antibiotics given post-operatively have benefit [8]. They do, however, recommend redosing antibiotics during the procedure if the surgical length is more than two half-lives of the antibiotic, or if the patient has significant blood loss in surgery [3, 8]. 
 
The guideline authors do note that they consider 48 hours of post-surgical antibiotic administration reasonable in cardiac surgery patients, although this has not been proven to be essential [3, 8]. Several studies have attempted to investigate use of even more extended courses, but have concluded that longer antibiotic administration does not provide significant benefits.  For example, a review of head-to-head trials of different regimens performed in the 1980s and 1990s found no difference in the risk of surgical site infections or mortality in patients who received four days of cefazolin, two days of cefuroxime, one day of cefazolin, two doses of ceftriaxone, or a single preoperative dose of ceftriaxone [9].
 
A large cohort study done at Beth Israel Deaconess Medical Center compared patients who received prophylactic antibiotics for less than 48 hours and more than 48 hours after CABG to evaluate both the risk of developing a surgical site infection and the risk of having a culture positive for a resistant organism (either vancomycin-resistant enterococcus (VRE) or a gram-negative organism with cephalosporin resistance). The researchers found that the group who received antibiotics for more than 48 hours post-operatively had increased odds of having a culture positive for a resistant organism, but no decrease in their odds of developing a surgical site infection [10].
 
Another study conducted from 2007-2011 at the Mayo Clinic and the University of Michigan followed patients after LVAD placement to determine whether there was any benefit seen with extending antibiotics beyond 24 hours after implantation to prevent infections of the driveline site. There was no statistically-significant difference in the number of infections between the two groups [11]. 
 
An additional study in England compared cardiothoracic surgery patients who received only perioperative cefuroxime and patients in whom cefuroxime was continued post-operatively until all central lines had been removed.  All central line tips were cultured after removal, and the number of positive cultures was not significantly different between the two groups, suggesting that a longer antibiotic course would not have a benefit in preventing post-operative bacteremia in patients who have an indwelling line [12].
 
These studies therefore all support the current guidelines for limiting postoperative antibiotic administration to 24-48 hours, as there is not evidence for extending the course even in patients at high risk of infection, including those with indwelling lines or devices. 
 
Topical Antibiotic Use
Several trials in Europe have investigated the use of a biodegradable gentamicin-collagen sponge that can be placed in the sternal wound prior to closure as an adjunct to IV perioperative antibiotic therapy [13].  A study of 2000 patients undergoing cardiac surgery in Sweden found that use of these sponges decreased sternal wound infections by approximately fifty-percent [14], and so a phase III trial was performed in the United States from 2007-2009.  The trial enrolled patients who were considered to be at increased risk for infection as a result of obesity and/or diabetes, but did not find the same significant benefit to gentamicin sponge use [13]. Given conflicting trial results, use of these sponges is considered an “unresolved issue” in the 2014 guidelines, and is not FDA-approved at this time [8].
 
Staph Aureus Decolonization
Testing patients preoperatively for Staph aureus colonization, with subsequent decolonization in patients who test positive, is recommended in specific situations in the 2014 guidelines. The authors note that this issue is complex, as study results have been mixed regarding both whether screening improves outcomes and what type of decolonization (nasal mupirocin, topical chlorhexidine, or both) is most beneficial [8].  However, they do recommend screening prior to “high-risk” cardiothoracic surgery, particularly if there is a high prevalence of infections secondary to Staph aureus in the treating institution and/or if the patient is considered to be at increased risk [8]. If patients are colonized, nasal mupirocin use is recommended, typically for five days prior to surgery [3, 8]. Mupirocin should not be applied in the absence of a positive screening result, given the risk of increasing resistance with use of this topical antibiotic [8]. 
 
Adjunctive Measures for Infection Prevention
Several adjunctive perioperative and intraoperative interventions may also help to decrease infection risk.  SHEA’s 2014 guidelines make recommendations regarding these interventions that include findings of the Centers for Medicare and Medicaid Services Surgical Improvement Project (SIP) and the subsequent Surgical Care Improvement Project (SCIP):

• Hair removal: The 2014 guideline authors noted that removing the patient’s hair is typically not indicated, and that if it is necessary for some reason it should be done with clippers rather than a razor to avoid breaks in the skin [8].

•  Skin cleansing:  The authors recommended cleaning the surgical site with both an alcohol-containing solution and a disinfectant (either chlorhexidine or iodine) prior to the incision. They discuss that studies have had contradictory results regarding whether chlorhexidine or povidone-iodine performs better, although a 2010 study of more than 800 patients showed a benefit of chlorhexidine use with decreased infection rates [8, 15].

• Intraoperative temperature control:  They also stressed the importance of sustaining normothermia (defined as a temperature 35.5°C) during any surgeries where the patient is under anesthesia for more than an hour [8]. 

• Maintaining normal oxygenation: In the post-operative period, the guidelines recommend giving the patient supplemental oxygen if they required mechanical ventilation during surgery, based on the results of several abdominal surgical studies. The authors cite five different studies that demonstrated a significant decrease in surgical site infections in patients who were given 80% FiO2 via non-rebreather compared to patients who were given 30-35% FiO2 [8].

• Glucose control: The authors noted that blood glucose should be monitored closely following cardiac surgery. They discuss that the prior recommendation by the SCIP had been to keep morning glucose levels <200 mg/dL in the first 2 days after surgery. However, they report that starting in 2014, the recommendation is instead to check a blood glucose level 18-24 hours post-operatively with a goal glucose £180 mg/dL[8].

 
 
Antibiotic Management of Post-Operative Infections
Sternal Wound Infections
A study of 18,460 patients who had cardiac surgery with a sternotomy in Turkey from 2005-2012 found that sternal wound infections were predominantly secondary to gram-positive organisms -- 39% and 24% were caused by coagulase-negative staphylococci and Staph aureus, respectively [16].  
 
When a sternal wound infection is suspected, the most critical intervention is to open the surgical site and drain any purulent material [17]. The IDSA guidelines for treatment of skin and soft tissue infections also recommend a short course of IV antibiotic therapy directed at the gram-positive organisms that have been shown most likely to be responsible. This could be achieved using a penicillin with activity against staphylococci or a cephalosporin, but broadening to antibiotics with coverage against MRSA (such as vancomycin or daptomycin) is reasonable if the patient has risk factors such as a prolonged hospitalization, administration of other antibiotics, or a history of MRSA colonization or infection [17].  
 
Mediastinitis
The same study from Turkey referenced above also evaluated patients who developed mediastinitis, and found that the most common causal organism was MRSA [16]. Results were similar in a study of 55 patients with mediastinitis in Switzerland, where ~42% had cultures positive for Staph aureus and another ~33% had cultures positive for coagulase-negative staphylococci [18].  The recommended empiric antibiotic therapy would therefore be very similar to that discussed for sternal wound infections, with treatment adjusted based on results of deep cultures. The most critical treatment, however, is again surgical with aggressive debridement [18].
 
Pacemaker and Implantable Defibrillator Device Infections
Pacemaker and implantable defibrillator (AICD) placement can be complicated by several different types of infections, including superficial skin infection at the incision site, infection involving the device pocket, and deeper infection involving the intravascular device leads with the potential for associated endocarditis. The type of infection suspected determines what evaluation should be performed, whether the device needs to be removed, when a new device can be inserted, and the length of the antibiotic course [19]. 
 
The American Heart Association (AHA) published guidelines regarding pacemaker and AICD device infections in 2010 that include an algorithm for diagnosis and management. According to these guidelines, the first step when there is concern for a device infection should be to obtain two sets of blood cultures. A transesophageal echo (TEE) should be performed to look for valvular or device lead vegetations if these blood cultures return positive, or if the patient had recent antibiotic exposure that could have resulted in culture sterilization [19]. 
Based on these test results and examination of the pocket site, the extent of the infection can then be assessed to determine whether the device should be removed.  If only the overlying skin is involved (for example, if there is a cellulitis at the site of the surgical incision), then the device can remain in place and the patient can be treated with 7-10 days of antibiotics.  If, however, there is erosion of the device leads or the generator, if there is drainage from the pocket or a pocket abscess, if there is a vegetation seen on a device lead or a heart valve, if the patients is septic, or if a patient has staphylococcal bacteremia with no clear source in the presence of a device, then the device should be removed [19]. 
 
Patients should also be started on empiric antibiotic therapy, which can be narrowed based on blood culture results or results of cultures of the pocket site and device leads sent during device removal [19]. A review of 189 device infections at the Mayo Clinic from 1991-2003 found that the majority were caused by staphylococcus species (29% Staph aureus and 42% coagulase-negative staph) [20]. Therefore, the AHA guidelines recommend initiation of empiric vancomycin, with subsequent narrowing as appropriate [19].
 
Duration of antibiotic therapy depends on the type of infection present [19]. If the patient has a pocket infection with negative blood cultures, the AHA guidelines recommend treating for 10-14 days after the device is removed. This can be decreased to 7-10 days if they had erosion of the device leads or generator without associated erythema, swelling, or drainage from the pocket.   If the patient has positive blood cultures but no evidence of a vegetation on TEE, the recommendation is that they receive 4 weeks of antibiotic therapy if cultures were positive for Staph aureus, and 2 weeks in other cases. These would also be the treatment durations for a lead vegetation with no associated valvular vegetations or embolic phenomena. If, however, a lead vegetation is seen in association with septic emboli, the patient should receive 4-6 weeks of therapy, and if a valvular vegetation is seen the patient should be managed under the endocarditis guidelines, which are also published by the AHA [19].
 
The length of time that it is necessary to wait prior to replacing the device is also dependent on the extent of the infection [19].  Patients with a pocket infection that has been debrided and blood cultures that are negative for at least three days can have a new device placed at that time under the AHA guidelines. Patients with positive blood cultures but no evidence of vegetations on TEE can have a new device placed when repeat blood cultures have been negative for at least three days.  Patients who had a vegetation on a device lead without valvular involvement should have blood cultures repeated after the device has been extracted, and then can have a new device placed when those cultures have been negative at least three days. Finally, patients who have endocarditis should wait at least two weeks after blood cultures are negative to place a new device [19].  

Key Points

• Perioperative antibiotic prophylaxis has significant benefits in decreasing the morbidity and mortality associated with surgical site infections.
• Use of a first or second generation cephalosporin is recommended prior to cardiac surgical procedures.
• The first dose should be given within an hour of starting surgery, and antibiotics should not be continued for longer than 24-48 hours post-operatively.
• Vancomycin is not first-line for perioperative prophylaxis, but can be used in cases of cephalosporin allergy, or can be added to the cephalosporin if the patient is at high risk for MRSA infection or has a prior history.
• Adjunctive measures including pre-operative MRSA decolonization for patients known to be colonized, skin disinfection prior to surgery, maintenance of intraoperative temperature control, and post-operative maintenance of normoglycemia and normal oxygenation can all also have benefit in decreasing post-operative infections.
• The primary treatment for sternal wound infections and mediastinitis is surgical debridement, but adjunctive antibiotics should also be given. These infections are most commonly caused by staphylococcal species, and so use of an empiric antibiotic with MRSA coverage is likely appropriate while awaiting culture results. 
• Management of suspected pacemaker and defibrillator infections depends upon an assessment of the extent of the infection.  Superficial skin infections overlying the pocket site do not require device removal, but pocket drainage, abscesses, device erosion, and device lead or valvular vegetations all require that the device be extracted. 
• Blood cultures should be done in all cases of suspected intracardiac device infection, and a TEE should be performed if they are positive.
• Empiric antibiotic coverage for device infections should include MRSA coverage, with antibiotics then narrowed as appropriate based on blood culture results and cultures of the device pocket and device leads following removal.  Length of therapy depends upon the extent of the infection and blood culture clearance.  

References
1. Greco G, Shi W, Michler RE, et al., Costs associated with health care-associated infections in cardiac surgery. J Am Coll Cardiol, 2015. 65(1): p. 15-23.
2. Bratzler, D.W.a.H., P.M., Antimicrobial Prophlaxis for Surgery: An Advisory Statement from the National Surgical Infection Prevention Project. Clin Infect Dis, 2004. 38: p. 1706-15.
3.Bratzler DW, Dellinger EP, Olsen KM, et al., Clinical practice guidelines for antimicrobial prophylaxis in surgery. Am J Health Syst Pharm, 2013. 70(3): p. 195-283.
4. Dellinger EP, Hausmann SM, Bratzler DW, et al., Hospitals collaborate to decrease surgical site infections. Am J Surg, 2005. 190(1): p. 9-15.
5. Da Costa A, Kirkorian G, Cucherat M, et al., Antibiotic prophylaxis for permanent pacemaker implantation: a meta-analysis. Circulation, 1998. 97(18): p. 1796-801.
6. de Oliveira JC, Martinelli M, Nishioka SA, et al., Efficacy of antibiotic prophylaxis before the implantation of pacemakers and cardioverter-defibrillators: results of a large, prospective, randomized, double-blinded, placebo-controlled trial. Circ Arrhythm Electrophysiol, 2009. 2(1): p. 29-34.
7. Acharya, M.N., R. Som, and S. Tsui, What is the optimum antibiotic prophylaxis in patients undergoing implantation of a left ventricular assist device? Interact Cardiovasc Thorac Surg, 2012. 14(2): p. 209-14.
8. Anderson DJ, Podgorny K, Berríos-Torres SI, et al., Strategies to prevent surgical site infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol, 2014. 35(6): p. 605-27.
9. Kriaras I, Michalopoulos A, Turina M, et al., Evolution of antimicrobial prophylaxis in cardiovascular surgery. European Journal of Cardiothoracic Surgery, 2000: p. 440-446.
10. Harbarth S, Samore MH, Lichtenberg D, et al., Prolonged Antibiotic Prophylaxis After Cardiovascular Surgery and Its Effect on Surgical Site Infections and Antimicrobial Resistance. Circulation, 2000. 101: p. 2916-2921.
11. Stulak JM, Maltais S, Cowger J, et al., Prevention of percutaneous driveline infection after left ventricular assist device implantation: prophylactic antibiotics are not necessary. ASAIO J, 2013. 59(6): p. 570-4.
12. Sandoe JA, Kumar B, Stoddart B, et al., Effect of extended perioperative antibiotic prophylaxis on intravascular catheter colonization and infection in cardiothoracic surgery patients. J Antimicrob Chemother, 2003. 52(5): p. 877-9.
13. Bennett-Guerrero E, Ferguson TB Jr, Lin M, et al., Effect of an Implantable Gentamicin-Collagen Sponge on Sternal Wound Infections Following Cardiac Surgery. JAMA, 2010. 304(7): p. 755-762.
14. Friberg O, Svedjeholm R, Söderquist B, et al., Local gentamicin reduces sternal wound infections after cardiac surgery: a randomized controlled trial. Ann Thorac Surg, 2005. 79(1): p. 153-61; discussion 161-2.
15. Darouiche RO, Wall MJ Jr, Itani KM, et al., Chlorhexidine-Alcohol versus Povidone-Iodine for Surgical-Site Antisepsis. N Engl J Med, 2010. 362(1): p. 18-26.
16. Yavuz SŞ, Tarçın O, Ada S, et al., Incidence, aetiology, and control of sternal surgical site infections. J Hosp Infect, 2013. 85(3): p. 206-12.
17. Stevens DL, Bisno AL, Chambers HF, et al., Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the infectious diseases society of America. Clin Infect Dis, 2014. 59(2): p. 147-59.
18. Immer FF, Durrer M, Mühlemann KS, et al., Deep sternal wound infection after cardiac surgery: modality of treatment and outcome. Ann Thorac Surg, 2005. 80(3): p. 957-61.
19. Baddour LM, Epstein AE, Erickson CC, et al., Update on cardiovascular implantable electronic device infections and their management: a scientific statement from the American Heart Association. Circulation, 2010. 121(3): p. 458-77.
20. Sohail MR, Uslan DZ, Khan AH, et al., Management and outcome of permanent pacemaker and implantable cardioverter-defibrillator infections. J Am Coll Cardiol, 2007. 49(18): p. 1851-9.