The majority of coronary artery bypass grafting (CABG) and valvular repair or replacement operations are performed through a median sternotomy incision. This procedure was first reported in the journal Lancet in 1897¹ by the British surgeon Herbert Meyrick Nelson Milton at the Kasr El Aini Hospital in Cairo for the treatment of tuberculous lymphadenopathy; later, it was reintroduced by Ormand Julian in 1957 to replace the more painful and disfiguring bilateral anterior thoracotomy for open heart surgery and then in 1964 for CABG.² The two most catastrophic complications of the median sternotomy soon followed—dehiscence and deep sternal wound infection (DSWI) with mediastinitis.³ Indeed, Milton’s second patient suffered fatal sternotomy dehiscence. The overall incidence of deep sternotomy wound infection or dehiscence is relatively uncommon, ranging from 0.1% to 5%. Oakley et al provided a review in 1996 in which centers reported incidences from 0.1% to 2%. The National Nosocomial Infection Surveillance (NNIS) system of the Centers for Disease Control and Prevention (CDC) in 1997 reported a sternal infection rate of 3.7%. More recently, the Society of Thoracic Surgeons (STS) Databank, a prospective registry, which has recorded outcome events of over 6 million cardiac surgical procedures since 1989, reports a DSWI incidence of 0.4%.⁴

The morbidity and mortality rate of DSWI is substantial. Mortality rates for DSWI as high as 30% have been reported⁵ and are dependent upon the virulence of the infecting agent, patient comorbidities, and treatment modality used to effect healing of such an infection. DSWI morbidity includes suffering, chronic debility, reoperation, prolonged hospital stay (9-20 days), and cost.⁶ Patients with DSWI experience on average an additional hospital cost of $20,000 in the first year in 1999 dollars; infected patients who died experienced an additional $60,547 more than those who survived. In 2008, an effort to control hospital costs and compel physicians to prevent DSWI was initiated by the Centers for Medicare & Medicaid Services (CMMS); acting in concert with the Deficit Reduction Act of 2006, CMMS established a list of hospital-acquired conditions (HAC) which, when acquired during hospitalization (ie, not present on admission), do not qualify the facility for additional payment (http://www.cms.gov/HospitalAcqCond/downloads/HACFactsheet.pdf).

Mediastinitis following CABG is one of the 12 listed HACs in this government action.

Risk factors associated with a deep sternotomy wound infection can be categorized as occurring in the preoperative,⁷,⁸ intraoperative, and postoperative phase of care; risk factors can also be considered modifiable and unmodifiable (Table 94.1).

Preoperative risk factors are primarily those related to patient conditions and comorbidities. Operative factors that contribute to wound infection are listed in Table 94.1. Antibiotic prophylaxis is proven to prevent surgical site infections. However, the timing of antibiotic prophylaxis more than 60 minutes before incision or more than 30 minutes after incision and variance of antibiotic choice to a nonrecommended prophylaxis agent is associated with DSWI. Use of bone wax as a topical hemostatic agent on the sternal bone marrow, intra-aortic balloon pump use, indiscriminate use of the Bovie electrocautery device, undrained fluid collections (especially blood), cardiopulmonary bypass time, total operative time, and operation in addition to CABG—such as a valve-CABG procedure—also contribute to the risk of DSWI.⁸ Postoperative factors include the following: shock, need for a postop tracheostomy, poorly controlled hyperglycemia.

Interest was generated from the observation that patients who were on statin agents experienced fewer wound infections. Although the preoperative administration of statins has been demonstrated in cohort studies to reduce the incidence of all infections, they were not specifically linked to a statistically significant reduction in specific infections, such as DSWI.⁹

Blood transfusion following cardiac surgery has also been associated with increased infection risk,¹⁰ which was not mitigated by the use of leukocyte-depleted blood products.¹¹ A tiered-level effect was observed in a study published by Murphy et al¹²; in a stepwise fashion, the infection rate was 3.9% in patients receiving up to 2 units of packed red blood cells, 6.9% for those receiving 3 to 5 units, and as high as 22% in those receiving 6 units of blood or more. Logistic regression analysis showed that the most significant predictor of infection was transfusion amount; the increased risk of infection was postulated to be related to transfusion-mediated immunomodulation. Relevant to blood transfusions, the independent effect of preoperative anemia has been evaluated but not demonstrated to directly affect the rate of sternal wound complications.¹³

Hemoglobin A1c influences the risk of developing a wound infection. A single-center study from UT Southwestern Medical Center demonstrated that every 1% increase in preoperative hemoglobin A1c is associated with a 13% relative risk increase for a wound infection.¹⁴ In addition, perioperative glycemic control—independent of hemoglobin A1c—influences the incidence of sternotomy wound infections and mediastinitis.¹⁵

The effect of single versus bilateral internal thoracic artery harvesting on the incidence of severe sternal wound complications has been extensively studied. In the Arterial Revascularization Trial,¹⁶ patients were randomly assigned to undergo single (N = 1554) versus bilateral (N = 1548) internal thoracic artery grafting during CABG procedures; 23% of patients enrolled were diabetic. Although not a primary end-point of this trial, sternal wound complications (1.9% vs 3.5% in the single versus bilateral graft group, respectively) and sternal wound reconstruction (0.6% vs 2.0% in the single versus bilateral graft group, respectively) were not statistically different.