Chapter 8: Inflammation, Infection, & Antimicrobial Therapy in Surgery

TERMINOLOGY

Surgery and infection are unfortunately intimately intertwined. For the purpose of this chapter, we will differentiate between infections resulting from surgery (surgical site infections) and those resulting from other disease processes but requiring surgical management. Surgical incisions involve a breach of the skin and immune barriers, and can thus be complicated by infection. The term “wound infection” has been replaced by the more accurate term “surgical site infection” (SSI), to emphasize that the infection can occur anywhere within any of the areas accessed surgically (not exclusively at the skin level) and to differentiate it from “traumatic wound infection.”

As opposed to SSI, the term “surgical infection” is used to indicate infections that are unlikely to respond to medical and antimicrobial treatments, and require surgical intervention or management. Common examples include abscesses, empyema, intra-abdominal infections, and necrotizing skin and soft tissue infections. Surgical decision making involves, at its core, the knowledge and experience to determine the timing of surgery and a right balance between surgery and other adjunct therapy such as antibiotic therapy, resuscitative efforts, and nutritional optimization, in order to provide the patient with the best chances to cure the infection with the best overall outcomes.

Pathogenesis

The development of a surgical infection involves a close interplay between three elements:

1. A susceptible host

2. An infectious agent

3. A suitable medium or environment (Figure 8–1).

The degree of contribution of each of these three factors to the eventual occurrence of the infection depends on the individual patient and the specific nature and site of the infection. Whether a given inoculum of bacteria results in an established infection or not depends on the virulence of the bacteria, the strength of the immune and inflammatory host response (eg, chemotaxis, phagocytosis, B- and T-lymphocyte activation), and the amount of blood perfusion and oxygen tension in the medium where the inoculum resides.

A. The Susceptible Host

Many surgical infections occur in patients with no evidence of decreased immune defenses. However, with the major advances in medicine and health care services over the last century, more immunocompromised patients (eg, transplant, HIV, and diabetic patients) are presenting with infections requiring surgical management or with SSIs following surgical interventions. Table 8–1 delineates a list of patient-related conditions associated with decreased immunity and potential predisposition to surgical infections. The mechanisms by which the listed conditions affect a host’s immunity are diverse in nature. For example, diabetic patients have suboptimal neutrophil adherence, migration and anti-bacterial functions, making them less able to fight an occult infection. Such effect of diabetes on immunity is much more pronounced in patients whose glucose levels are poorly controlled than in those with appropriate and consistent management of their diabetes.

B. The Infectious Agent

Identification of the organism causing the infection is crucial. This is often achieved through the use of Gram stains and cultures of the tissue or purulent material at the site of infection. Staphylococcus aureus is the most common pathogen in SSIs, and its ability to develop resistance to antimicrobial therapy continues to be one of the biggest challenges currently faced by the medical community. Staphylococcus epidermidis is one of the other 33 known species of the genus Staphylococcus, and is usually a skin and mucous membrane nonpathogenic colonizer. Nonetheless, in immunosuppressed individuals and those with indwelling surgical catheters or implants, S epidermidis can cause serious infections requiring treatment. In necrotizing skin and soft tissue infections, the etiology is often polymicrobial with gram-positive, gram-negative, and anaerobic bacteria involved. Monobacterial necrotizing infections can result from either clostridial species (especially Clostridium perfringens) or Streptococcus pyogenes. Streptococci are well known for their ability to invade even minor breaks in the skin and to cause superficial skin and skin structure infections by spreading through connective tissue planes and lymphatic channels. When surgical procedures involve the small or large intestines, gram-negative and anaerobic bacteria are implicated in a significant proportion of the SSIs in addition to the regular gram-positive colonizers. Among these, Escherichia coli, Klebsiella pneumoniae, Enterobacter, and Bacteroides species are the most commonly isolated pathogens. In immunocompromised or critically ill patients, fungi (eg, Candida, Aspergillus, Histoplasma) may result in localized or systemic life-threatening infections. Parasites such as amebas and echinococcus may also cause abscesses in internal organs, especially the liver.

C. The Suitable Medium

A suitable medium for bacteria consists of a closed space with poor vascular perfusion, low oxygen tension, and low pH. Ischemic and necrotic tissues are perfect examples of such a medium where microorganisms can thrive. The appendix, with its narrow orifice presents a classical example. When an appendicolith blocks the orifice of the appendix, the intraluminal pressure increases and eventually blocks lymphatic and venous outflow. As the vascular perfusion of the appendix diminishes, the tissue oxygen tension decreases, the milieu becomes acidotic, and the appendix becomes ischemic and necrotic. Unless removed, the inflamed appendix and its blocked lumen contain colonic bacteria that will result in an appendiceal infection followed by perforation and a periappendiceal abscess and/or peritonitis.

A foreign body such as a joint prosthesis can become seeded in the event of a systemic infection. As with necrotic tissue, the lack of vascularity decreases the formation of free oxygen radicals within prosthesis and prevents the immune system from readily fighting microorganisms in that area.

SURGICAL SITE INFECTION

Definition

The Centers for Disease Control and Prevention (CDC) define SSI as an infection that occurs at or near the surgical incision within 30 postoperative days of the surgical procedure, or within 1 year if an implant is left in place (eg, mesh, heart valve [www.cdc.gov]). The CDC further classifies SSI as

1. Superficial incisional

2. Deep incisional

3. Organ/space SSI

Table 8–2 illustrates the criteria that define each of these three classes of SSIs. It is noteworthy that a skin infection at or around the site of a traumatic wound is classified as a skin and skin structure infection.

Epidemiology

At least 234 million surgical procedures are performed globally each year, including more than 16 million in the United States alone; SSIs develop in 2%-5% of these patients. In this surgical patient population, SSI accounts for up to 38% of nosocomial infections, and as such is considered the most common nosocomial infection in surgical patients. If asymptomatic bacteriuria is excluded, SSI is arguably the most common nosocomial infection overall. The rate of SSI depends on the nature of the surgical procedure performed and the extent of concomitant intraoperative contamination. In an attempt to quantify the inoculum of bacteria typical of certain surgical procedures, a wound classification was developed in 1964 by the National Academy of Sciences (Table 8–3). This classification divides wounds into clean (eg, inguinal hernia repair), clean-contaminated (eg, right hemicolectomy), contaminated (eg, laparotomy for penetrating injury to small intestines), and dirty (eg, laparotomy for peritonitis and intra-abdominal abscesses) wounds. Several studies have emerged since the conception of the classification confirming that, with reasonable risk adjustment, the classes of contamination correlate well with the incidence of postoperative SSI. A more recent 2012 American College of Surgeons-National Surgical Quality Improvement (ACS-NSQIP) study of more than 600,000 patients, suggested that the rates of SSI increases as the degree of contamination increases: 2.58% for clean wounds, 6.67% for clean-contaminated wounds, 8.61% for contaminated wounds, and 11.80% for dirty wounds.

Attributable Cost & Impact

In addition to the morbidity incurred by patients, the health care and economic burden of SSIs is significant. Multiple studies have consistently documented that SSIs lead to a considerable increase in length of hospital stay (LOS), hospital charges, and societal health services cost. In a 2012 study, SSI occurring after colorectal surgery increased the LOS by 2.8-23.9 days, depending on the nature of the procedure (colon vs rectum resection), the operative approach (open vs laparoscopic), and the depth of the SSI. In another study of cardiac surgery, it was estimated that the total excess cost attributable to SSI is more than $12,000 per patient. This excess cost resulted from the additional procedures (eg, incision and drainage, wound washout, skin grafts), antibiotics and increased hospital LOS. Another study in general and vascular surgery patients estimated the excess cost and LOS at $10,497 and 4.3 days, respectively. In addition, even when SSI patients are successfully discharged, their rate of hospital readmission is at least doubled. Another study using patient-centered surveys and large administrative databases suggested almost a threefold increase in the total cost for patients diagnosed with SSI following discharge from the hospital compared to those who do not develop SSI. The increased costs were accounted for by a significant increase in the use of visiting nurses for wound care, diagnostic imaging, as well as an increase in readmissions to emergency departments and to the hospital.

Pathophysiology

The etiology of SSI is multifactorial; patient’s comorbidities, the nature of the surgical procedure and technique, and the perioperative environment including infrastructure and processes of care all interact in SSI. It is thought that most SSIs are directly caused by the patient’s endogenous flora at the time of surgery. This is further supported by the fact that, in clean wounds, SSIs are most commonly caused by S aureus or coagulase negative Staphylococcus, both abundantly present on patient’s skin. On the other hand, SSIs following bowel surgery are usually caused by endogenous intestinal flora. When a surgical device or implant is left in place, bacteria can secrete special glycocalyx-based biofilms that shield the offending bacteria from the body’s immune defenses. One of the immediate sequelae of such a phenomenon is the fact that the inoculums needed to cause a SSI in the presence of a foreign body such as a surgical implant or device is typically smaller than the inoculum needed to cause the same infection in the absence of a foreign body.

Despite the belief that most SSIs originate from endogenous microbes, there is convincing literature suggesting that nonendogenous sources are implicated in the occurrence of SSIs as well. Breach of the strict sterile surgical field, undetected surgical gloves’ perforations, increased personnel traffic through the operating room, suboptimal air flow and ventilation in the operating room, and poor surgical technique with excessive tissue injury or tension are some of the many factors implicated in SSI that will be discussed further in the next few sections.

Risk Factors

Many risk factors have been implicated as predictors of SSI. Some are directly related to the patient’s health status, immune system, and comorbidities, while others are inherent to the nature of the procedure being performed and therefore are not easily modifiable. A few risk factors reflect suboptimal health care systems’ design and less than reliable processes of care delivery, and their importance relies in the possibility of subjecting them to improvement and optimization strategies.

A. Patient-Related Risk Factors

Several studies have shown that advanced age, diabetes mellitus, obesity, smoking, malnutrition, and immunosuppression (eg, chronic steroid use, HIV/AIDS, posttransplant) are all risk factors for the development of SSI (Table 8–1). Most of these factors are not amenable to immediate preoperative modification or prevention, although long-term control of glucose levels (reflected in better HBA1C levels), smoking cessation, and attempts to improve nutrition may help in the overall performance of the patient perioperatively and may also decrease complications, including SSI.

B. Procedure-Related Risk Factors

The nature of the procedure, especially the wound classification (Table 8–3) and the expected inoculum of bacteria into the surgical site are key factors that influence the risk of developing a SSI. For example, the rate of SSIs complicating eye surgery, where the inoculum of bacterial contamination is minimal, is almost non-existent. In comparison, the rate of SSI complicating colorectal surgery remains closer to 20% despite strategies and efforts aimed at prevention. In the early 1990s, the National Healthcare Safety Network (NHSN) risk index, previously called the National Nosocomial Infections Surveillance System (NNIS), was developed in an attempt to better define a priori, a patient’s risk of developing SSI. This index considered the following three factors:

1. An American Society of Anesthesiologists (ASA) score ≥ 3

2. A surgical wound classification as either contaminated or dirty

3. A prolonged duration of procedure over a certain number of hours T, where T depends on the nature of the procedure being performed

The analysis of the CDC data by Culver and colleagues revealed that the rates of SSIs are 1.5%, 2.9%, 6.8%, and 13% in the presence of none, 1, 2, or 3 of the above risk index factors, respectively. More recently, The NHSN has introduced improved SSI risk adjustment models for specific operations based on easily tractable patient risk factors.

C. Process- and System-Related Risk Factors

Several process- and system-related risk factors have been identified over the years as potential factors contributing for a higher rate of SSI.

Suboptimal choice and timing of perioperative antibiotics

In a systematic review of more than 30,000 Medicare patients, Bratzler and colleagues found that less than 56% of patients undergoing surgery received their perioperative antibiotics within the recommended 60-minutes window before incision. In the same study, up to 10% of patients did not receive the appropriate antibiotic regimen that would cover the type of pathogens specific to the procedures the patients were undergoing. The two main components of prophylactic perioperative antibiotics, namely, choice and timing, will be discussed in further detail in the SSI prevention section below.

Unrecognized breach of asepsis

Operating room staff, including nursing, anesthesia and surgical personnel, are strongly encouraged to immediately report to the surgeon when suspecting a potential breach of the sterile field (eg, inadequate skin preparation, soiled clothes or equipment, glove perforation). In a study of gynecological laparotomies, glove perforations were found to occur in 27 out of 29 laparotomies. The relationship between the rates of SSIs and glove perforation is less than evident when the data is analyzed. In a study of more than 4000 procedures, the rates of SSIs increased with glove perforation only in the absence of appropriate antibiotic prophylaxis.

Preoperative hair removal

Preoperative hair removal has been correlated in several studies with a higher rate of SSI, even when the procedure involves the scalp or the patient has abundant hair at the surgical site. A 2006 Cochrane database systematic review and meta-analysis concluded that preoperative shaving increases the rate of SSI by at least twofold; it is noteworthy that there was no difference in the rates of SSI when hair clipping was compared to no hair removal. Therefore, if hair removal is deemed necessary, preoperative hair clipping is preferred to shaving.

Surgical technique

The use of “rough” surgical techniques with unnecessary or excessive tissue disruption and trauma, whether resulting from electrocautery, traction or blunt dissection, is believed to lead to higher rates of SSI. Even though most surgeons and common sense suggest the above correlation, strong evidence supporting a cause-effect relationship has not been demonstrated.

Operating room traffic

It is well established that the burden of airborne bacteria in a closed space is directly related to the number of people in the space, as well as to the number of people entering and exiting a closed space such as the OR. The link between increased traffic through the OR doors and SSIs has been studied mostly in the orthopedic literature, but has not been clearly established. Nonetheless, the CDC has issued guidelines that include a decrease in OR traffic.

Perioperative hypothermia

Even though hypothermia may protect tissue from ischemia and necrosis by decreasing cellular oxygen consumption, perioperative hypothermia may cause vasoconstriction at the skin level, and therefore decrease oxygen delivery. Two randomized trials attempted to study the effect of hypothermia on the incidence of SSI. The first one found an increased SSI risk with hypothermia in colorectal surgery, while the second failed to find any correlation in cardiac surgery. At present, maintenance of perioperative normothermia is recommended.

Prevention Strategies

Prevention of SSI relies primarily on optimizing the patient’s comorbidities and creating a systematic method to ensure that all modifiable risk factors are addressed. All processes of care proven or arguably thought to prevent SSIs should be executed in all patients at all times. Table 8–4 lists potential measures to minimize the risk of SSI. Although level-one evidence is lacking for many of those measures, inclusion of some or all in a bundle might prove to be a beneficial strategy. The surgical improvement project (SIP) established in 2004 and folded within the surgical care improvement program in 2006 (SCIP) created performance measures out of several of these preventative measures based on best evidence in literature (Table 8–5). A detailed discussion of SCIP is beyond the scope of this text and the reader is referred to the CDC and CMS websites for more details. Perioperative antibiotics and the choice of the skin preparation solution are discussed below.

A. Perioperative Antibiotics

The timing of prophylactic antibiotic administration (SCIP 1) and the choice of the appropriate perioperative antibiotic (SCIP 2) are key components in the strategy to prevent SSI. Prophylaxis implies discontinuation of the antibiotic within 24 hours in noncardiac surgery and within 48 hours in cardiac surgery (SCIP 3); this will also prevent side effects of the antibiotic and the emergence of drug resistance. Proper antibiotic dosing based on the patient’s weight and manufacturer’s recommendation for the antibiotic is an important consideration in prophylaxis. In addition, redosing of the antibiotic during surgery based on the antibiotic half-life, amount of fluid administration and blood loss is more important than the continuation of the antibiotic after the wound is closed. Table 8–6 is a practical guide to dosing and time for redosing for commonly used prophylactic antibiotics.

B. Antiseptic Skin Preparation

Based on a prospective randomized trial and an observational study, the National Quality Forum has recommended the use of alcohol-containing solutions for the preparation of patient’s skin prior to surgery.

C. Additional Preventative Measures

Maintenance of perioperative normoglycemia (SCIP 4), normothermia (SCIP 7), and hair clipping or avoiding hair removal (SCIP 6) are definitely advocated and monitored as performance measures for certain operations. The use of preoperative antibacterial soap showers, perioperative oxygen administration, the use of antibacterial coated sutures, wound barriers, antimicrobial irrigants, and antibacterial dressings are additional measures that remain controversial in the prevention of SSI (Table 8–4). Mechanical bowel preparation and oral antibiotics in patients undergoing colorectal surgery are the subjects of renewed controversy and discussion but remain an important strategy in the prevention of SSI in this field.

Treatment

The treatment of an established SSI consists of opening the incision, draining the purulent material, and debridement of any necrotic tissue. Adjunctive antimicrobial treatment is given when the patient has systemic symptoms, associated cellulitis or a deep SSI with risk for fascial and subcutaneous spread. Cultures are important to track emergence of resistant organisms, to de-escalate from broad-spectrum empiric antibiotics, and for epidemiological surveillance. In the case of an organ/space SSI, percutaneous drainage using radiological guidance, along with adequate antibiotic coverage, is the preferred approach.

INFECTIONS NEEDING SURGICAL MANAGEMENT

When infections fail to respond or are deemed unlikely to respond to medical and antimicrobial treatments alone, surgical intervention might be needed. The list of surgical infections necessitating surgical management is long, but the most common ones include abscesses, empyema, necrotizing skin and soft tissue infections (NSSTIs), intra-abdominal infections, and Clostridium difficile (C difficile) colitis. Many of these surgical infections can become life threatening by causing sepsis, septic shock, and multiple organ dysfunction syndrome (MODS) if not controlled and treated promptly. Spreading can occur through tissue planes (necrotizing infections), abscess/fistulae formation, or through the lymphatic system and the bloodstream. Spread of an infection through the bloodstream (bacteremia) can lead to distant seeding of bacteria and subsequent abscess formation (eg, brain, liver, adrenal glands, heart valves).

INFLAMMATION, SEPSIS, THE SYSTEMIC INFLAMMATORY RESPONSE SYNDROME, BACTEREMIA, & MULTIPLE ORGAN DYSFUNCTION SYNDROME

Microbial infection results in tissue damage that leads to an inflammatory response at the local tissue level in an effort by the patient’s immune system to overcome the infection. At the site of injury, endothelial cells and leukocytes coordinate the local release of mediators of the inflammatory response, including cytokines (tumor necrosis factor-α), interleukins, interferons, leukotrienes, prostaglandins, nitric oxide, reactive oxygen species, and products of the classic inflammatory pathway (complement, histamine, and bradykinin) (Table 8–7). Once these mediators reach the infected site, they are extremely effective at recruiting and priming cells of both the innate and adaptive immune systems to identify and attack invading pathogens. In addition, these inflammatory mediators initiate the damaged tissue healing processes. However, if the degree of the infectious or traumatic insult overwhelms the ability of the body to control it, the inflammatory mediators might trigger a systemic inflammatory reaction of serious damaging consequences. Such a quasi-maladaptive systemic response can disrupt normal cellular metabolism and jeopardize tissue perfusion at the cellular level. From a terminology perspective, when the systemic response occurs in the absence of a documented infection (eg, burns, trauma, pancreatitis), it is called systemic inflammatory response syndrome (SIRS). The term sepsis is used when the systemic response results from a documented infection rather than systemic inflammation alone. Severe sepsis is reported when sepsis is associated with at least one sign of tissue or organ hypoperfusion. Septic shock occurs when severe sepsis is associated with hypotension or hemodynamic instability. When septic shock leads to progressive dysfunction of multiple organs such as the brain (delirium), lungs (hypoxia and respiratory failure), heart (hypotension, pulmonary edema), and kidneys (oliguria and renal failure), it is referred to as multiple organ dysfunction syndrome (MODS). Bacteremia is the presence of viable bacteria in blood, often documented through blood cultures. Transient bacteremia (eg, following dental work) is common, benign, and self-limited. It usually has no clinical implications except in patients with damaged heart valves; cardiac, vascular, or orthopedic implants; or impaired immunity. Bacteremia happening in a patient with an uncontrolled source of infection can be devastating and may result in severe sepsis, septic shock, and MODS. The interrelationships among infection, bacteremia, sepsis, and SIRS are depicted in Figure 8–2, and the detailed clinical definitions of sepsis, severe sepsis, septic shock, SIRS, and MODS are presented in Table 8–8.

Diagnosis

Localizing the source of the suspected infection or sepsis is crucial, and a good and thorough physical examination is indispensable. Laboratory evaluation often shows leukocytosis, but leukopenia is not infrequent in cases of severe infection or sepsis. Acidosis is occasionally present and might be helpful in making the diagnosis. Radiologic examinations, including plain films for suspected pneumonia, computed tomography or ultrasound for suspected intra-abdominal or intrathoracic abscesses, and bone scans or magnetic resonance imaging (MRI) for suspected osteomyelitis, are often essential for making the diagnosis and establishing the source of infection.

Identifying the causative organism by culturing the infected site, followed by testing the implicated pathogen’s sensitivity to diverse antimicrobial agents are the two next key steps. If infection is suspected but the source remains unclear, cultures of blood, sputum, and urine should be performed. Culturing other body fluids (eg, cerebrospinal fluid, pleural and joint effusions, ascites) can be performed if warranted based on the patient’s history, physical examination, and the degree of clinical suspicion.

Treatment

Infectious source control and antibiotics constitute the mainstay of surgical infection treatment. When patients are septic with or without shock, “goal-directed” management of the sepsis with prompt patient resuscitation and immediate initiation of the appropriate empiric antibiotics are of ultimate importance, and have been shown by high-level evidence to improve patient survival. The Surviving Sepsis Guidelines emphasize the systematic and evidence-based approach to treating sepsis and septic shock patients. These guidelines are available at www.survivingsepsis.org.

Surgical cutaneous infections span the spectrum of the superficial skin abscesses such as furuncles and carbuncles to the devastating and often fatal necrotizing deeper soft tissue infections.

FURUNCLE

Pathophysiology

A furuncle (boil) is a superficial skin abscess usually caused by S aureus infection at the level of the hair follicle. Bacteria asymptomatically inhabit most hair follicles, but it is thought that obstruction of the pilosebaceous apparatus is the triggering event for the formation of furuncles.

Risk Factors

Some of the identified risk factors include puberty, male gender, obesity, diabetes, poor hygiene, as well as living in weathers characterized by a high level of humidity.

Causative Organisms

Even though staphylococci are the most common causative organisms, streptococci, gram-negative organisms, and anaerobic diphtheroids can also cause furuncles.

Clinical Presentation

Furuncles present as itchy indurated small abscesses, surrounded by skin erythema. A small white area of skin necrosis is often notable at the top of the abscess.

Treatment

Most furuncles resolve spontaneously, although larger lesions require incision and drainage. The use of antibiotics following incision and drainage is not routinely needed and is controversial. We recommend antibiotics only in the presence of concomitant extensive surrounding cellulitis. When antibiotics are prescribed, they should be stopped with resolution of the local and systemic symptoms. Antibacterial soap showers might be helpful in both treatment and prevention of recurrence.

CARBUNCLE

Pathophysiology

Carbuncles are rare and result when several furuncles coalesce, extend to the subcutaneous tissue, and/or form a “network” of multilocular interconnected abscesses and tracts.

Risk Factors

Carbuncles on the back of the neck are seen almost exclusively in diabetic or relatively immunocompromised patients. In addition to obesity and diabetes, chronic steroid intake and malnutrition are potential risk factors for carbuncles.

Causative Organism

Similar to furuncles, S aureus remains the most common organism involved in carbuncles.

Clinical Presentation

Carbuncles often have the appearance of multiple large furuncles with several openings draining pus. As carbuncles enlarge, the blood supply to the skin is destroyed and the tissue over the top of the abscess becomes white in color and necrotic in appearance. Patients might show some systemic signs such as fever and malaise.

Treatment

Carbuncles often require incision and drainage; occasionally, a more extensive excision is required. When appropriate, the excision is continued until the many associated deep sinus tracts are removed. Antibiotics are usually needed in view of the associated extensive skin cellulitis and induration.

HIDRADINITIS SUPPURATIVA

Pathophysiology

Hidradenitis suppurativa is a serious skin condition characterized by blockage and infection of the apocrine sweat glands. Hidradenitis most commonly involves the axilla or the groin. Since apocrine glands develop postpuberty, prepubertal disease is very rare. Even though hidradenitis often becomes chronic with serious patient morbidity and disability, systemic complications, and constitutional signs and symptoms are uncommon.

Risk Factors

Obesity, puberty, female gender, and smoking are some of the identified risk factors associated with hidradenitis.

Causative Agents

Hidradenitis is a polymicrobial infection involving gram-positive, gram-negative, and anaerobic organisms. Superinfection with fungal organisms is not uncommon, especially in the patient with chronic or recurrent hidradenitis.

Clinical Presentation

Hidradenitis can affect any area where apocrine glands exist, most commonly the axilla and groin, but also the perineum, inframammary folds, gluteal folds, areola, and scrotum. On examination, erythematous tender skin nodules characterize early stages. This can quickly progress to deeper indurated abscesses with interconnected sinuses, purulent drainage, and associated regional lymphadenopathy.

Treatment

Drainage of the individual abscesses and their associated sinuses is needed. If healing is delayed, fungal super-infections should be suspected and, if confirmed, should be treated as well. With chronic and recurrent infections, the involved skin and apocrine glands can be excised, and the underlying soft tissue left to heal by secondary intention. Skin flaps and grafts might be needed in later stages, as the resultant scarring can be severe. Topical clindamycin, systemic tetracycline, and isotretinoin are potential adjunctive therapies. Weight loss and hygiene-focused interventions should be encouraged to decrease the risk of recurrence.

NECROTIZING SKIN & SKIN STRUCTURE INFECTIONS

Introduction

Accurate data on the incidence of necrotizing skin and skin structure infections (NSSSIs) is lacking, and most existing data are based on single institutions’ experiences, suggesting a large variability in clinical presentation, severity as well as outcome. NSSSIs are skin and soft tissue infections characterized by widespread and severe tissue necrosis resulting from aggressive and life-threatening bacteria that are often able to secrete toxins. NSSSIs are also known as gas gangrene, necrotizing fasciitis (when involving fascial layers), Fournier gangrene (when involving the perineum and genitalia), and Ludwig angina (when involving the floor of the mouth).

Pathophysiology

NSSSIs can be caused by single agents such as clostridia or streptococci species, but most are polymicrobial in etiology. An inciting site (eg, puncture wound, insect bite) can occasionally be discovered by history or physical examination. In the perineum, manipulation of the urogenital tract and diabetes are known to be associated with NSSTI. The infection is classically one of sudden onset and rapid progression through ischemic tissue planes, occasionally with air formation at deeper layers. Small vessel thrombosis occurs along the infection progression pathway, and thus, the underlying deeper tissue damage is almost always much more pronounced than what the overlying skin appearance suggests.

Causative Agents

By definition, type I NSSSIs are polymicrobial (eg, streptococci, clostridia), while type II NSSSIs are monomicrobial. C perfringens secretes exotoxins (eg, lecithinases, collagenases, proteases, hemolysins) that lead to small vessel thrombosis and allow for rapid progression and aggressive invasion of fascial and muscle planes. These toxins contribute to the serious deep tissue destruction and the overlying skin grayish discoloration. The alpha toxin (a lecithinase) is believed to be an essential contributor to clostridial virulence. In addition to their local effect, these toxins may lead to severe sepsis, pronounced SIRS, and not uncommonly MODS. Streptococci can also secrete several exotoxins with similar virulence leading to fascial and deeper plane disruptions. S pyogenes (group A strep), although rare, is classically described to secrete a superantigen toxin that can result in toxic shock syndrome with resultant organ failure. NSSSIs are often caused by mixed nonclostridial, nonstreptococcal bacteria, including staphylococci, gram negatives, and anaerobes.

Risk Factors

Immunocompromised (eg, HIV, diabetes), intravenous drug abuse, and cancer patients are at a particularly higher risk of developing NSSSIs, as well as malnourished and obese patients.

Clinical Presentation

Prompt suspicion and early diagnosis of NSSSIs are essential. In early stages, patients often have fever and pain out of proportion to physical examination findings. Erythema, induration, hemorrhagic skin bulla, blisters, and vesicles are occasionally present. Grayish discoloration of the underlying skin and pain beyond the affected skin area are especially suggestive of deeper underlying tissue involvement. “Dishwasher-like” gray discharge might be seen and should raise suspicion of NSSSIs. Patients can quickly develop systemic signs of toxicity with hypotension, tachycardia, electrolyte imbalances, lethargy, and even organ failure. Laboratory workup typically shows leukocytosis (with or without bandemia) or leucopenia, hyponatremia, and acidosis. Several models have been developed over the years in an attempt to predict the probability of a skin infection being a NSSTI versus a more benign superficial skin infection. The Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC) score assigns different point scores to each of six laboratory values (Table 8–9). A LRINEC score of six or more was found in a study of 89 NSSSIs to have a positive predictive value of 92% and a negative predictive of 96% for NSSTI. Radiologically, plain films showing deep soft tissue air are pathognomonic. Computed tomography (CT) scan is the standard of care when difficulty arises in differentiating between “benign” skin or soft tissue infection and NSSSIs. CT images often show inflammation and edema at the deeper fascial or muscular layers, and occasionally will show gas between these tissue layers. Although crepitus as felt on physical examination and/or deep tissue air as diagnosed by plain radiography or CT imaging have been classically described with several (but not all) clostridial species, they may occur with nonclostridial infections as well. When the clinical picture and the radiological imaging fail to differentiate between benign soft tissue infection and NSSSIs, diagnostic surgical exploration of the area in question is needed. Operative findings suggestive of NSSSIs include pale and/or necrotic deep tissue, “dishwasher” nonpurulent gray discharge, nonbleeding soft tissue, and thrombosed microvessels. Easy separation with unopposed finger “sliding” between the tissue planes (the “finger test”) is highly suggestive of NSSSIs. Pathologic examination will reveal severe inflammation and tissue necrosis.

Treatment

A NSSSI is a surgical emergency and requires immediate

1. Wide surgical debridement

2. Broad-spectrum antibiotics

3. Hemodynamic and fluid resuscitation

Wide Surgical Debridement

All tissue that appears pale, ischemic, or necrotic, “lifts off” easily, or does not bleed appropriately needs to be removed. This should include the overlying skin, even when it appears viable, as studies have documented severe vasculitis and microvessel thrombosis that eventually lead to loss of skin, if not adequately debrided. Extensive debridement is usually needed, with one or more repeat trips to the operating room, until viable tissue is ensured. When an extremity is involved, amputation might be required, although attempts at debridement and limb salvage, with a planned “second look” are reasonable and might be preferred. Perirectal NSSSI and Fournier gangrene occasionally necessitate creation of an ostomy and/or urinary diversion. Depending on the extent and location of debridement, many patients, if they survive and recover, are disfigured and will need reconstructive surgery ranging from simple split thickness skin grafts to major flap reconstructions.

A. Antibiotics

As soon as a NSSSI is suspected, broad-spectrum antibiotics should be initiated. Gram-positive, gram-negative, and anaerobic organisms need to be covered. A combination of a penicillin-based agent or cephalosporin with an aminoglycoside or fluoroquinolone and an anti-anaerobic agent (eg, metronidazole or clindamycin) is a reasonable starting empiric regimen. Clindamycin has been shown by in vitro studies to have anti-inflammatory and toxin-neutralizing effects, in addition to its antibacterial effects. If methicillin-resistant S aureus is suspected, penicillins can be replaced by vancomycin or linezolid. Aminoglycosides can be replaced by third- or fourth-generation cephalosporins (eg, cefepime, cefotaxime) or by carbapenems (eg, imipenem, meropenem, ertapenem), if indicated (eg, in patients with acute renal failure). The Infectious Diseases Society of America (IDSA) published guidelines for the antibiotic choices in NSSSIs are reported in Table 8–10.

B. Hyperbaric Oxygen

The usefulness of hyperbaric oxygen in the treatment of NSSTI is controversial. Several small retrospective studies have suggested decreased mortality with hyperbaric therapy, but the absence of supportive level one evidence, combined with the logistical difficulties in critically ill patients remain the main obstacles to recommend its routine utilization.

Prognosis

NSSSI has a poor prognosis with reported mortality as high as 20%-50%, and an even higher morbidity rate. Diabetic patients and/or those presenting with septic shock and/or multisystem organ failure are at a particularly higher risk of mortality. Shorter time from onset of infection to operative debridement has been shown consistently to correlate with better survival.

ORGAN/SPACE INFECTIONS

Organ/space infections are often deep infections resulting from uncontrolled infection or perforation of an internal organ. Internal postoperative abscesses are termed organ/space SSIs, and have been discussed previously. In addition to antibiotic treatment, many of these organ/space infections will need drainage.

Intra-Abdominal Abscesses and Infections

In the era of high-resolution CT imaging, diseases like acute diverticulitis and acute perforated appendicitis are more likely to be successfully managed nonoperatively in the acute clinical setting. When patients have no signs of diffuse peritonitis, periappendiceal, pericolonic abscesses, and other contained intraperitoneal abscesses may be drained percutaneously under radiological guidance when accessible. With the combination of percutaneous drainage and adequate antibiotic coverage, patients’ intra-abdominal infections can be treated adequately, allowing for the inflammation to resolve with potentially one-stage and less complicated surgical intervention at a later time to address the diseased organ. When percutaneous drainage is not possible, usually because of the abscess location and its inaccessibility to radiological interventions, surgical drainage with removal of the infected organ (eg, appendectomy, colectomy) is warranted. If the abscess is less than 3-4 cm in diameter, an attempt at treatment with antibiotics alone without percutaneous or surgical drainage is a reasonable option in the stable patient without peritoneal signs. The management of complicated intra-abdominal infections, whether community acquired or health care associated, whether originating from the hepatobiliary system or the gastrointestinal system, is beyond the scope and goal of this chapter. The IDSA has recently published useful guidelines for diagnosis and treatment of intra-abdominal infections that emphasize

1. Source control (surgical or percutaneous drainage)

2. Early and appropriate antibiotics coverage

3. Fluid resuscitation

4. Microbiological evaluation, when feasible

These guidelines can be accessed at www.idsociety.org.

Empyema

Empyema is a collection of pus in the pleural cavity, most commonly related to bacterial pneumonia and resulting parapneumonic effusions. Occasionally, empyemas occur postprocedurally, such as following thoracentesis, chest tube placement, or lung resection. When empyema occurs, drainage is needed. In addition to antibiotics, tube thoracostomy is the recommended initial management. Unfortunately, chest tubes lumens often get clogged with thick pus, leading to placement of additional tubes, and not infrequently, failure to completely drain the pleural cavity and re-expand the lung. When that occurs, surgical drainage with decortication using a video-assisted thoracoscopic approach or an open thoracotomy is indicated. In cases of recurrent or persistent empyema despite surgical management, creation of an Eloesser flap or window with an open track between the pleural cavity and the outside skin might be needed to allow adequate and definitive drainage of the empyema.

CLOSTRIDIUM DIFFICILE COLITIS

Clostridium difficile as an organism can be recovered in the feces of 5% of healthy individuals. The use of antibiotics can alter the colonic flora allowing for C difficile overgrowth and colonization of the colon. The result is C difficile colitis with severity ranging from simple watery diarrhea to life-threatening sepsis. C difficile is currently considered one of the most common and most serious health care–associated infections.

Risk Factors

Antibiotic intake is the single most important risk factor for C difficile colitis and has been described even following a single antibiotic dose administration as with SSI prophylaxis. Although all antibiotics can be implicated, fluoroquinolones, clindamycin, penicillins, and cephalosporins are the most frequently encountered culprits, partly because of their widespread utilization. A Canadian study conducted during a relatively recent C difficile outbreak in Quebec strongly suggested that fluoroquinolones are currently the antibiotics most commonly associated with C difficile infections rather than clindamycin as classically and historically described. Table 8–11 lists the different antibiotics classified into those with high, moderate, and small risk of C difficile infection.

Transmission of C difficile by health care providers caring for patients with C difficile colitis has resulted in outbreaks of C difficile among patients within hospitals and the emergence of multidrug-resistant organisms. These epidemics are addressed and prevented by proper hand hygiene among health care providers and the isolation of patients with C difficile colitis.

Clinical Presentation

The presentation of C difficile colitis can range from the asymptomatic carrier state that requires no treatment, and the mild diarrhea state that can easily be treated with oral antibiotics as an outpatient, to the severely toxic, septic, and MODS patient presentation. Across all the severity range, watery diarrhea, and abdominal distention remain the sine-qua-non of C difficile colitis. Lower abdominal pain may also be present, and is usually of a crampy nonspecific nature. Constitutional symptoms such as fever and malaise are also frequently reported. The presence of peritoneal signs such as rebound tenderness or guarding may be indicative of colonic perforation (rare) or severe and/or fulminant C difficile colitis. Leukocytosis is often present and not infrequently elevated above 20K cells/mcL. In fact, it is not unreasonable to test the hospitalized patient with unclear etiology of leukocytosis for occult C difficile infection, even in the absence of diarrhea. In a small study of inpatients with white blood cell counts above 15K cells/mcL, and without a clear etiology of the leukocytosis, Wanahita et al found that 58% of the patients tested positive for C difficile. Fecal leukocytes are also often present, but nonspecific. On colonoscopy, when performed, patients with C difficile colitis frequently show mucosal ulcerations with the pathognomonic pseudomembranes, often described as raised yellow plaques composed of fibrinous exudates. Abdominal x-rays may show “thumb printing” suggestive of colonic inflammation, but are often nonspecific. CT examination of the abdomen shows thickening of the colonic wall. In fulminant C difficile colitis, the abdominal x-rays and CT may show toxic megacolon with a diffusely dilated colon.

Diagnosis

In addition to the clinical patient presentation described above, diagnostic tests for C difficile currently consist of either C difficile organism detection (eg, anaerobic stool culture, antigen testing) or toxin A or B detection tools (eg, cytotoxin assays, enzyme immunoassays, or polymerase chain reaction [PCR] testing). The best approach to diagnose C difficile remains controversial and institution-dependent. The stool culture is a sensitive test, but has a long turnover time (2-3 days) that is suboptimal in the critically ill patient who needs a fast and accurate diagnosis. Antigen detection is a reasonable screening test, but needs further confirmation when positive. The most sensitive test (considered standard by many) is the cytotoxin assay. Its sensitivity is above 95%, and its specificity approaches 100%, but it is an expensive test with 24-48 hours turnover time.

Treatment

Treatment of asymptomatic patients is not indicated. On the other hand, the clinical, laboratory, and radiological presentation of the patient might carry a sufficient suspicion for C difficile infection that empiric antibiotics are justified pending C difficile organism or toxin detection tests. For patients with mild disease, level 1 evidence suggests the safety and equivalency of single-agent oral antibiotic treatment with PO metronidazole (500 mg every 8 hours) or PO vancomycin (125 mg every 6 hours). PO vancomycin is not systemically absorbed and is thus wholly available in the colonic lumen where the C difficile infection exists. A 2-week course of either antibiotics is recommended. Patients might still test positive for C difficile for many weeks despite treatment, and therefore retesting these patients for C difficile when the antibiotic course is done and the patient’s signs and symptoms resolved, is not recommended. Adjunct therapies for C difficile such as probiotics, toxin-binding resins, and immunoglobulins have been used, but the evidence for their utility is still controversial.

Treatment of Severe C difficile Colitis

The definition of severe C difficile infection varies, but the IDSA 2010 guidelines define it as leukocytosis above 15K cells/mcL or a creatinine 1.5 times higher than the patient’s predisease baseline. For severe C difficile colitis, these same guidelines recommend PO vancomycin rather than metronidazole as a first-line agent. Combination therapy with PO vancomycin, intravenous metronidazole +/– vancomycin enemas might be indicated in case of failure of single-agent therapy. Vancomycin enemas are particularly useful in patients with ileus (questionable adequate intracolonic concentration from PO vancomycin alone), those with PO intolerance, and possibly those with severe distal colonic inflammation (rectosigmoid colon). Fidaxomicin (200 mg every 12 hours) is a relatively new bactericidal antibiotic with promising results as an alternative therapy. In a randomized clinical trial comparing fidaxomicin and vancomycin, the cure rates were equivalent. Patients with refractory severe C difficile colitis, those who progress to toxic megacolon, or those with hemodynamic instability or showing signs of early MODS should be considered for prompt surgical intervention. The timing and decision to operate are controversial, but there is certainly value in earlier intervention in critically ill patients, as the mortality of fulminant C difficile infection remains as high as 30%-50%. Independent predictors of mortality include age more than 70 years, leukocytosis more than 35K cells/mcL, leucopenia less than 4K cells/mcL, hemodynamic instability and respiratory failure. When surgery is indicated, the procedure of choice is a subtotal colectomy with end ileostomy. A single-center study has recently demonstrated improved outcomes in severe C difficile colitis using an alternative approach where a loop ileostomy is created and subsequent intraoperative colonic washout was performed with polyethelene glycol solution. In this small study of fewer than 50 patients, the new approach was associated with decreased mortality and more than 90% colon preservation, when compared to the traditional subtotal colectomy approach.

ANTIMICROBIAL MANAGEMENT

General Principles

Antibiotic optimal management includes the following guiding principles:

1. Establishing a clinical diagnosis of infection and/or sepsis

2. Establishing an “evidence-based-guess” of the likely nature of the involved pathogen(s)

3. Attempting to obtain microbiological data (Gram stain and culture), when possible

4. Prompt initiation of broad-spectrum empiric antibiotics likely to cover the culprit pathogen(s)

5. Selectively narrowing the antibiotic regimen to cover the pathogen recovered in microbiological testing

The decisions to initiate, continue, or tailor the antibiotic choices for each patient should be carefully balanced against the increasingly serious challenge facing health care, namely, the epidemic emergence of multidrug-resistant organisms such as methicillin-resistant S aureus (MRSA), vancomycin-resistant Enterococcus (VRE), and multidrug-resistant gram-negative organisms.

EMPIRIC & PATHOGEN-DIRECTED ANTIBIOTIC SELECTION

Selecting the appropriate empiric antibiotics entails an understanding of the nature of the infectious process affecting the patient, the most likely organisms causing the infection, as well as awareness of the antibiotic resistance map at the local health care institution level. For example, simple skin infections are most likely caused by skin flora such as staphylococci and streptococci, and therefore an appropriate empiric treatment should at least cover gram-positive organisms. Acute diverticulitis with a pericolonic abscess is more likely to be caused by colonic flora, such as gram-negative and anaerobic bacteria, and therefore empiric treatment should focus on covering these pathogens. Once a specific organism has been recovered from the tested microbiological specimen, it is often possible to narrow the antibiotic regimen, even before the specific data on antibacterial susceptibilities is available using the microbiological properties of antibiotics and the currently available clinical experience. The use of local antibiograms is crucial and highly encouraged, as the patterns of bacterial susceptibilities and resistance vary not only from a region to the other but also from one institution to the other. Susceptibility testing methods include disk diffusion and dilution (broth microdilution, plates, and E-tests) procedures. Disk diffusion tests indicate whether a microbial culture is susceptible or resistant to serum-achievable, in vivo drug concentrations with conventional dosage regimens. In contrast, dilution procedures allow report of the minimal inhibitory concentration (MIC) and minimal bactericidal concentration (MBC). The MIC is the lowest concentration of a specific antimicrobial agent that inhibits the test organism, while the MBC is the lowest concentration of a specific antimicrobial agent that kills the test organism. Tailoring therapy based on reported of susceptibility via disk diffusion or MIC is recommended.

Antibiotic Duration of Therapy

If all the sent cultures were found to be negative, and the suspicion for continuing infection is low, empiric antibiotics should be discontinued, as unnecessary antibiotic therapy clearly increases the risk of multidrug-resistant bacteria, C difficile colitis, and other side effects. The duration of antibiotic treatment depends on the nature of the infection and the severity of the clinical presentation. Treatment of acute uncomplicated infections should be continued at least until the patient is afebrile, the white blood cell count is normal, and the patient looks clinically well. Evidence-based guidelines for type of antibiotics and duration in the treatment of various infections have been developed and should be followed when possible. Most antibiotic regimens are now limited, although more difficult infections such as liver abscesses, brain abscesses, endocarditis, septic arthritis, or osteomyelitis require prolonged antibiotic therapy.

Failure of Clinical Response

Upon failure of clinical response to initial surgical intervention and antibiotic therapy, the clinician is encouraged to think of all the possible following explanations of failure of therapy:

1. The existence of another organism

2. The presence of an uncontrolled source of infection needing additional interventions

3. The presence of a concomitant super-infection (eg, with fungal organisms)

4. The development of a new drug resistance by the same isolated organism

5. Failure of the chosen antibiotic medication (despite favorable susceptibility data) to penetrate the site of infection

The latter may be due to the properties of the antibiotic itself, the chosen route of administration, or the specific physiology and morbidity of the patients themselves.

Antibiotics’ Side Effects

Adverse antibiotic reactions may occasionally mimic infection by causing fever, skin rashes, and mental status changes. Clinicians should consider this possibility in patients with persistent fevers despite appropriate antibiotic coverage and seeming resolution of the infectious process. In addition, antibiotics can result in microbial superinfections (eg, fungi, C difficile), emergence of new bacterial drug resistance, or organ toxicity.

Antibiotics in Renal or Hepatic Failure

Several antibiotics may induce or exacerbate existing organ dysfunction, most commonly renal or hepatic, and thus require intermittent assessment of kidney and liver function. Aminoglycosides and renal failure is a classic example.

In addition, decreased creatinine clearance at baseline or due to critical illness has an important influence on antimicrobial drug dosage, since most of these drugs are excreted, at least partially, by the kidneys. Therefore, many of these medications, such as vancomycin, penicillins, and aminoglycosides require adjustments in dosage or frequency of administration in the presence of renal insufficiency to prevent toxicity. Table 8–12 provides practical guidance into adjusting diverse antibiotic dosage and frequency for patients with hepatic or renal insufficiency, although additional assistance from the pharmacy department in these cases is universally advisable.

CONCLUSION

In conclusion, surgery can result in infection (such as SSI), but may also be the adjunct treatment to obtain source control in many infectious processes (ie, surgical infections). In the former, prevention by implementing processes of care (eg, timely provision of perioperative antibiotics, maintenance of strict sterile techniques) and patient risk factors optimization prior to surgery should be the goal of every surgeon, as SSI remains one of the biggest health care-associated perioperative challenges. For the latter (surgical infections), source control is the cornerstone in the prevention of systemic progression of the infection into sepsis, septic shock, and/or MODS. Alongside surgical treatment of infections, antibiotic stewardship with appropriate timing, duration, and choice of antibiotics are needed now more than ever to prevent the emergence of resistant organisms as well as C difficile infection.

References