Pathophysiology of surgical site infections

Even with strict aseptic technique and a clean surgical wound, some degree of bacterial contamination undoubtedly occurs in every operation. It has been well documented that the first 4 to 6 h after contamination determine whether the body will clear the bacteria or an established infection will occur. Thus, preventing SSIs relies in large part on optimizing the immune system during the perioperative period.

The body’s first defense against bacterial contamination is the neutrophil. Neutrophils are critically dependent on adequate O₂ stores to maintain their oxidative killing capacity. In one study, subcutaneous O₂ tension was measured in operative patients considered at elevated risk for developing SSI. In patients with subcutaneous O₂ tensions greater than 90 mm Hg, no SSIs developed, whereas patients with a subcutaneous O₂ tension of 40 to 50 mm Hg had an infection rate of 43%. As will be detailed in the following discussion, anesthesiologists may play an important role in the maintenance of immune and, in particular, neutrophil function.

Hypothermia

Without active efforts to warm patients, mild perioperative hypothermia (core body temperature 34° C-36° C) is commonly observed. In a landmark study, 200 patients undergoing colorectal operations were randomly assigned to a mild hypothermia group (34.4° C ± 0.4° C) or a normothermia group (37° C ± 0.3° C). This trial was stopped early because of the high rates of SSIs in the hypothermia group; the incidence of SSIs in the hypothermia group was 18.8%, versus 5.8% in the normothermia group. Patients who developed SSIs stayed nearly 1 week longer in the hospital. In addition, patients who were maintained at normothermia had evidence of increased wound healing and tolerated oral intake sooner. In a subgroup analysis, 74% of the hypothermic patients had evidence of intraoperative vasoconstriction, versus 6% of the normothermic patients.

The effect of hypothermia on SSIs is multifactorial. In the aforementioned study, the high incidence of vasoconstriction in the hypothermic group likely means a decrease in blood flow and O₂ delivery to the surgical site and, thus, impairment in the oxidative killing capacity of neutrophils. In addition, animal models have shown that hypothermia induces an anti inflammatory T-cell cytokine profile similar to that seen in patients with thermal injuries. Lastly, irrespective of the effect on blood flow and O₂ delivery to the surgical wound, hypothermia decreases the neutrophils’ production of superoxide radicals for any given O₂ tension. Indeed, bacterial killing by neutrophils is reduced in the face of hypothermia. With all that is known regarding the complications from mild perioperative hypothermia, including an increased risk of SSIs, it should be every clinician’s goal to maintain patients’ normothermia unless contraindicated.

Hyperoxia

In most clinical scenarios, O₂ delivery to end organs is vastly more dependent on the amount of O₂ bound to hemoglobin than the amount of O₂ dissolved in the blood. However, the subcutaneous tissue uses very little O₂, compared with the rest of the body. In addition, the mean partial pressure of O₂ in the subcutaneous tissue is approximately 60 mm Hg, a level above the range in which O₂ readily dissociates from hemoglobin. Lastly, when the microvasculature is traumatized at the site of the wound, the diffusion distance for O₂ is increased. These facts likely combine to decrease the importance of hemoglobin-bound O₂ on O₂ tension at the site of the wound and increase the importance of O₂ dissolved in the bloodstream (Figure 162-1).

To date, two randomized trials involving 800 patients undergoing colorectal operations have evaluated the effects of 80% inspired O₂ versus 30% inspired O₂ administered intraoperatively and for 2 h (500 patients) or 6 h (300 patients) postoperatively. Both studies found significantly decreased rates of SSIs in the patients receiving 80% inspired O₂. When data from the two studies are pooled, the relative risk reduction for developing an SSI is 45% (p = 0.02) in patients treated with hyperoxia. A subgroup analysis did not show any detrimental effects of 80% inspired O₂ during the study period when patients were evaluated with computed tomography scanning and pulmonary function tests.

Providing 80% O₂ in the operating suite is likely helpful in preventing SSIs and does not seem to be associated with any significant risk, but the continuation of 80% O₂ for 2 to 6 h postoperatively does present some potential procedural complications. It remains to be seen whether or not hyperoxia in the operating room without continuation in the postoperative period decreases SSIs.

Hyperglycemia

Hyperglycemia has been shown to have numerous deleterious effects on immune function in both in vitro and human models (Figure 162-2). A glucose challenge in healthy subjects induces a transient reduction in leukocyte counts. Hyperglycemia also deactivates immunoglobulins by nonenzymatic glycosylation and glycosylation of the C3 component of complement blocks binding to bacterial surfaces. The importance of neutrophils in preventing SSIs has already been emphasized, and the neutrophils of diabetic patients have numerous functional deficits, including impaired chemotaxis, decreased phagocytic ability, and lower bactericidal capacity. If these dysfunctional neutrophils are placed in a normoglycemic environment, their function can be at least partially restored in a short period of time. Lastly, it has been shown that, in patients undergoing cardiac surgery, glucose control with continuous insulin infusions improves the phagocytic function of neutrophils when compared with the use of intermittent insulin boluses to treat hyperglycemia.

What does the preceding discussion mean for the management of perioperative hyperglycemia with the goal of preventing SSIs? Surprisingly few studies have been performed with the goal of answering this question. In a frequently cited retrospective study of diabetic patients undergoing cardiac operations, it was shown that continuous insulin infusions to maintain blood glucose levels between 150 and 200 mg/dL decreased the incidence of sternal wound infections by 66% versus historical control subjects who were treated with sliding-scale insulin with the goal of maintaining blood glucose levels less than 200 mg/dL. To date, only one study has evaluated tight glycemic control (goal 80-110 mg/dL) in the operating room. In patients undergoing cardiac operations, tight glucose control has not been shown to improve outcomes but did lead to a higher incidence of stroke.

There is certainly significant in vitro data to suggest that hyperglycemia adversely affects the immune system, but the optimal goal for blood glucose levels is currently unknown. It appears that the historical threshold for treating blood glucose levels greater than 200 mg/dL is too high. However, very tight glycemic control is currently not supported by the literature and may, in fact, be dangerous.

Antibiotic prophylaxis

A detailed discussion of perioperative antimicrobial prophylaxis is beyond the scope of this chapter, but some salient points warrant discussion. The goal of perioperative antibiotic administration is to obtain blood and tissue drug levels that exceed the minimum inhibitory concentration of the organisms likely to be encountered. Adequate blood and tissue levels of the antibiotic must be obtained before incision, and current recommendations state that the infusion should begin within 60 min of incision. This period can be lengthened to 120 min for drugs, such as vancomycin, in which rapid administration may have adverse effects.

The choice of antibiotic is as important as the timing. Prophylaxis should target the most commonly encountered organisms and should not be administered with the goal of covering all possible pathogens. For most operations that do not violate chronically colonized organs, such as the bowel, the most common pathogens will be skin flora microbes, specifically Streptococcus and Staphylococcus. A first-generation cephalosporin (cefazolin) adequately covers these organisms. Procedures involving the bowel necessitate gram-negative coverage as well. Recently there has been discussion that vancomycin may be the antibiotic of choice for prophylaxis in institutions with a high incidence of methicillin-resistant Staphylococcus infections. It should be noted that there is no evidence to support this practice, and it is not recommended by any national organization. First-generation cephalosporins have a wider spectrum of coverage and are superior bactericidal agents, as compared with vancomycin, and they should be considered the first-line agents for surgical prophylaxis.

Conclusions

SSIs continue to be a major source of morbidity and death after surgical procedures. The critical window for preventing SSIs is the first few hours after the inevitable bacterial contamination; during this time, the anesthesiologist has the opportunity to help optimize the patient’s immune function so that these infections can be prevented.