CHAPTER 64 Deep Sternal Wound Infection
Most cardiac surgery procedures are performed through a median sternotomy, an approach pioneered by Milton in 1897.1 Although fairly uncommon, infective complications for this type of incision remain a difficult challenge for cardiac surgeons.
DEFINITION
Sternal wound complications have been classified by El Oakley and Wright2 as follows:
It is the deep wound infection subtype that is the focus of this chapter.
According to the guidelines from the Centers for Disease Control and Prevention in the United States,3 deep sternal wound infection (DSWI) can be defined by one of the following: (1) the presence of an organism isolated from culture of mediastinal tissue or fluid; (2) evidence of mediastinitis seen during operation; and (3) one of the following conditions: chest pain, sternal instability, or fever (>38° C) in combination with either purulent discharge from the mediastinum or an organism isolated from blood culture or culture of mediastinal drainage.
INCIDENCE AND ETIOLOGY
The incidence of DSWI in most recent series ranges from 0.75% to 2.4%.4–7 In our institution, prospectively collected results for 23,499 sternotomies performed between 1992 and 2007 were retrospectively reviewed. A total of 267 patients presented with DSWI, accounting for 1.1% of the surgical population. Staphylococcus aureus and coagulase-negative staphylococci are the most commonly found microorganisms.8 In their review of 30,102 consecutive patients undergoing sternotomy for cardiac surgery between 1990 and 2003, Tang and collaborators9 found them to be respectively responsible for 42% and 24% of DSWI. Gram-negative bacteria and fungi are less commonly encountered. These offending organisms have been associated with different predisposing factors and modes of presentation.10,11 Coagulase-negative staphylococci colonize the wound from the normal skin flora and proliferate within a self-contained pocket protected by an extracellular polysaccharide biofilm. The infection is mostly seen in obese patients and those affected by chronic obstructive pulmonary disease; it is associated with a slow and late onset with resulting sternal instability but fewer systemic signs. S. aureus infections are more aggressive in nature and more often associated with classic systemic signs and bacteremia. Perioperative contamination but also nasopharyngeal colonization are important sources for this type of infection. Gram-negative bacteria are more commonly associated with a more complicated postoperative course and concomitant nosocomial infections such as pneumonia, urinary tract infections, and abdominal sepsis.
RISK FACTORS
Host predispositions as well as a large number of perioperative environmental and technical aspects can play a significant role in the development of DSWI. There is a vast body of literature concerning the most important of these factors that have commonly been associated with a higher incidence of DSWI.4,5,12–22 Obesity, diabetes mellitus, chronic obstructive pulmonary disease, heart failure, renal failure, smoking, older age, and male gender have consistently been identified in patients more prone to DSWI. Preoperative considerations include prolonged hospital stay and the use of an intra-aortic balloon pump. The use of both internal thoracic arteries, re-do surgeries, excessive use of bone wax, and prolonged procedural times have all been described as perioperative contributing factors. Reexploration for bleeding, transfusions, prolonged intensive care unit stay, and prolonged intubation time are commonly associated in the postoperative period. Specific mechanisms have been proposed to account for the increased risk seen in patients affected by these conditions. Poor distribution of antibiotics in adipose tissue and inadequate skin preparation in obese patients, impaired wound healing by elevated blood glucose concentration in diabetes mellitus, improper sternal vascularization associated with the use of both internal thoracic arteries, and heart failure–induced low cardiac output are common examples. In our series, prolonged intubation was the strongest predictor of DSWI with an odds ratio of 5.7, probably reflecting composite indices of risk and patient fragility.
DIAGNOSIS
DSWI diagnosis should be suspected whenever sternal tenderness or instability, erythema, fluid collection, wound dehiscence, or purulent discharge is found, especially in the presence of fever or leukocytosis. The diagnosis is essentially clinical and based on the criteria listed in the Definition section. Blood cultures should be done when patients present with pyrexia. Any fluid discharge from the wound should also be cultured. Chest radiography is of limited interest for diagnostic purposes but may reveal ruptured or malpositioned wires as well as sternal fractures or dehiscence as indirect signs of DSWI. Computed tomography can help determine the extent, depth, and localization of the infectious process. It has been associated with 95.3% sensitivity and 81.7% specificity.23 Computed tomography is also useful to guide needle aspiration and culture. Nuclear imaging has not been used extensively, but labeling of leukocytes with 99mTc-HMPAO has shown promise as a reliable method for the early diagnosis of sternal infections in a small cohort of 41 patients.24
PREVENTION
The incidence of nasal colonization with S. aureus ranges from 10% to 15% in the normal population and increases the risk of sternal wound infection.11 Perioperative intranasal application of mupirocin is a safe and inexpensive method to significantly reduce DSWI.25 Skin preparation should be performed immediately before surgery, and hair is removed preferably by clipping rather than by shaving.3 Prophylactic antibiotics before surgery are recommended and should be administered to reach therapeutic concentrations before the initial incision and maintained throughout the procedure and for several hours after closure. The intravenous administration of cefazolin (1 g) or cefuroxime (1.5 g) is recommended within 30 to 60 minutes before the procedure; vancomycin should be used, in a dose adjusted to renal function, for patients with a history of penicillin allergy or at risk for methicillin-resistant S. aureus.3 Unless there is evidence of ongoing sepsis, prophylactic use of antibiotics should not be extended beyond 36 to 48 hours.
Blood glucose levels should be controlled, as maintenance at or below 110 mg/dL has been shown to decrease morbidity and mortality among critically ill patients, regardless of a presurgical history of diabetes.26 Furnary and coworkers27 have demonstrated that maintenance of blood glucose concentration between 150 and 200 mg/dL through continuous insulin infusion compared with intermittent subcutaneous insulin led to a significant reduction in the incidence of DSWI (0.8% versus 2%).
During surgery, caution to perform a precise midline sternotomy and discriminate use of the electrocautery and bone wax are simple methods that can contribute to DSWI prevention. The use of both internal thoracic arteries should also be weighed against potential risks of infection and tailored for each patient. Diabetic patients are particularly at risk, especially when other factors such as obesity and chronic obstructive pulmonary disease are present.28 Skeletonized bilateral internal thoracic artery grafting can contribute to lowering of the risk of DSWI in these patients.29 A closure technique that affords sternal stability should be adopted.30 A review of a variety of primary closure techniques designed to further stabilize the sternum and to lower the risks of dehiscence and infection is provided by Losanoff and coworkers.31
SURGICAL MANAGEMENT
Like any surgical wound infection, DSWI requires appropriate assessment of the underlying causes. This includes identification of the causative pathogens and usually involves surgical revision, thorough débridement, and wound preparation for reconstruction, with or without soft tissue flaps or plate fixation. Poststernotomy mediastinitis was initially approached by surgical revision with frequent dressing changes to promote granulation leading to sternal rewiring or secondary wound closure. The method was flawed by long hospital stays and mortality rates as high as 45%.32
