Keith M. Olsen, Alan E. Gross, and Joseph T. DiPiro
Most intraabdominal infections are “secondary” infections that are polymicrobial and are caused by a defect in the GI tract that must be treated by surgical drainage, resection, and/or repair.
Primary peritonitis is generally caused by a single organism (Staphylococcus aureus in patients undergoing chronic ambulatory peritoneal dialysis [CAPD] or Escherichia coli in patients with cirrhosis).
Secondary intraabdominal infections are usually caused by a mixture of bacteria, including enteric Gram-negative bacilli and anaerobes, which enhance the pathogenic potential of the bacteria.
For peritonitis, early and aggressive IV fluid resuscitation and electrolyte replacement therapy are essential. A common cause of early death is hypovolemic shock caused by inadequate intravascular volume and tissue perfusion.
Treatment is generally initiated on a “presumptive” or empirical basis and should be based on the likely pathogen(s) and local resistance patterns.
Antimicrobial regimens for secondary intraabdominal infections should include coverage for enteric Gram-negative bacilli and anaerobes. Antimicrobials that may be used for the treatment of secondary intraabdominal infections depending on severity of illness and microbiology data include (a) third-generation cephalosporin (ceftriaxone or cefuroxime) with metronidazole, (b) ticarcillin–clavulanate or piperacillin–tazobactam, (c) a carbapenem (imipenem, meropenem, doripenem, and ertapenem), and (d) quinolone (levofloxacin or ciprofloxacin) plus metronidazole or moxifloxacin alone.
Treatment of patients with peritoneal dialysis-associated peritonitis should include an antistaphylococcal antimicrobial such as a first-generation cephalosporin (cefazolin) or vancomycin (intraperitoneal administration is preferred).
The duration of antimicrobial treatment should be for 4 to 7 days for most secondary intraabdominal infections.
Patients treated for intraabdominal infections should be assessed for the occurrence of drug-related adverse effects, particularly hypersensitivity reactions (β-lactam antimicrobials), diarrhea (most agents), fungal infections (most agents), and nephrotoxicity (aminoglycosides).
Intraabdominal infections are those contained within the peritoneal cavity or retroperitoneal space. The peritoneal cavity extends from the undersurface of the diaphragm to the floor of the pelvis and contains the stomach, small bowel, large bowel, liver, gallbladder, and spleen. The duodenum, pancreas, kidneys, adrenal glands, great vessels (aorta and vena cava), and most mesenteric vascular structures reside in the retroperitoneum. Intraabdominal infections may be generalized or localized, complicated or uncomplicated, and community or healthcare-associated. Uncomplicated intraabdominal infections are confined within visceral structures, such as the liver, gallbladder, spleen, pancreas, kidney, or female reproductive organs while complicated intraabdominal infections involve anatomical disruption, extend beyond a single organ, and yield peritonitis and/or abscess. Peritonitis is defined as the acute inflammatory response of the peritoneal lining to microorganisms, chemicals, irradiation, or foreign-body injury. This chapter deals only with peritonitis of infectious origin.
An abscess is a purulent collection of fluid separated from surrounding tissue by a wall consisting of inflammatory cells and adjacent organs. It usually contains necrotic debris, bacteria, and inflammatory cells. These processes differ considerably in presentation and approach to treatment.
Peritonitis may be classified as primary, secondary, or tertiary.1–5 Primary peritonitis, also called spontaneous bacterial peritonitis, is an infection of the peritoneal cavity without an evident source in the abdomen.6 Bacteria may be transported from the bloodstream to the peritoneal cavity, where the inflammatory process begins. In secondary peritonitis, a focal disease process is evident within the abdomen. Secondary peritonitis may involve perforation of the GI tract (possibly because of ulceration, ischemia, or obstruction), postoperative peritonitis, or posttraumatic peritonitis (blunt or penetrating trauma). Tertiary peritonitis occurs in critically ill patients and is infection that persists or recurs at least 48 hours after apparently adequate management of primary or secondary peritonitis.
Primary peritonitis occurs in both children and adults, although the rates in children have been declining.4 Primary peritonitis develops in up to 10% to 30% of patients with alcoholic cirrhosis.4–7 Patients undergoing chronic ambulatory peritoneal dialysis (CAPD) average one episode of peritonitis every 33 months.8 Epidemiologic data for secondary and tertiary intraabdominal infections are less understood. Secondary peritonitis may be caused by perforation of a peptic ulcer; traumatic perforation of the stomach, small or large bowel, uterus, or urinary bladder; appendicitis; pancreatitis; diverticulitis; bowel infarction; inflammatory bowel disease; cholecystitis; operative contamination of the peritoneum; or diseases of the female genital tract, such as septic abortion, postoperative uterine infection, endometritis, and salpingitis. Appendicitis is one of the most common causes of intraabdominal infection. In 2006, 353,000 appendectomies were performed in the United States for suspected appendicitis.9 Most healthcare-associated intraabdominal infections occur as complications following intraabdominal surgeries.
Primary peritonitis in adults occurs most commonly in association with alcoholic cirrhosis, especially in its end stage, or with ascites caused by postnecrotic cirrhosis, chronic active hepatitis, acute viral hepatitis, congestive heart failure, malignancy, systemic lupus erythematosus, or nephritic syndrome. It may also result from the use of a peritoneal catheter for dialysis or CNS ventriculoperitoneal shunting for hydrocephalus. Rarely, primary peritonitis occurs without apparent underlying disease.
Table 92–1 summarizes many of the potential causes of bacterial peritonitis. Causes include inflammatory processes of the GI tract or abdominal organs, bowel obstruction, vascular occlusions that may lead to gangrene of the intestines, and neoplasia that may cause intestinal perforation or obstruction. Other possible causes include those resulting from traumatic injuries, postoperative infections, or solid organ transplant in the abdomen.
TABLE 92-1 Causes of Bacterial Peritonitis
Abscesses are the result of chronic inflammation and may occur without preceding generalized peritonitis. They may be located within one of the spaces of the peritoneal cavity or within one of the visceral organs, and may range from a few milliliters to a liter or more in volume. These collections often have a fibrinous capsule and may take from a few weeks to years to form.
The causes of intraabdominal abscess overlap those of peritonitis and, in fact, may occur sequentially or simultaneously. Appendicitis is the most frequent cause of abscess. Other potential causes of intraabdominal abscess include pancreatitis, diverticulitis, lesions of the biliary tract, genitourinary tract infections, perforation in the abdomen, trauma, and leaking intestinal anastomoses. In addition, pelvic inflammatory disease in women may lead to tuboovarian abscess. For some diseases, such as appendicitis and diverticulitis, abscesses occur more frequently than generalized peritonitis.
Microflora of the Gastrointestinal Tract and Female Genital Tract
A full appreciation of intraabdominal infection requires an understanding of the normal microflora within the GI tract. There are striking differences in bacterial species and concentrations of flora within the various segments of the GI tract (Table 92–2), and this bacterial environment usually determines the severity of infectious processes in the abdomen. Generally, the low gastric pH eradicates bacteria that enter the stomach. With achlorhydria, bacterial counts may rise to 105 to 107 organisms/mL (108 to 1010/L). The normally low bacterial count may also increase by 1,000- or 10,000-fold with gastric outlet obstruction, hemorrhage, gastric cancer, and in patients receiving histamine 2 (H2)-receptor antagonists, proton pump inhibitors, or antacids.
TABLE 92-2 Usual Microflora of the GI Tract
The biliary tract (gallbladder and bile ducts) is sterile in most healthy individuals, but in people older than 70 years of age, those with acute cholecystitis, jaundice, or common bile duct stones, it is likely to be colonized by aerobic Gram-negative bacilli (particularly Escherichia coli and Klebsiella spp.) and enterococci.10 Patients with biliary tract bacterial colonization are at greater risk of intraabdominal infection.
In the distal ileum, bacterial counts of aerobes and anaerobes are quite high. In the colon, there may be 500 to 600 different types of bacteria in stool, with concentrations often reaching 1011 organisms/mL (1014/L) and anaerobic bacteria outnumbering aerobic bacteria by more than 1,000 to 1.2,11 In fact, up to 50% of the dry mass of stool is Bacteroides spp. Fortunately, most colonic bacteria are not pathogens because they cannot survive in environments outside the colon. Perforation of the colon results in the release of large numbers of anaerobic and aerobic bacteria into the peritoneum.
The colonic flora are generally consistent unless broad-spectrum antimicrobials have been used. Depending on the type of antibiotic and spectrum, the duration of use, route of administration, and the pharmacokinetic and pharmacodynamic properties, antibiotics can cause shifts in the normal GI microflora including causing increased drug resistance.12
The lower female genital tract is generally colonized by a large number of aerobic and anaerobic bacteria. Anaerobes may number 109 organisms per milliliter (1012/L) and often include lactobacilli, eubacteria, clostridia, anaerobic streptococci, and, less frequently, Bacteroides fragilis. Aerobic bacteria most often are streptococci and Staphylococcus epidermidis, and these may number 108 organisms per milliliter (1011/L).
Intraabdominal infection results from bacterial entry into the peritoneal or retroperitoneal spaces or from bacterial collections within intraabdominal organs. In primary peritonitis, bacteria may enter the abdomen via the bloodstream or the lymphatic system by transmigration through the bowel wall, through an indwelling peritoneal dialysis catheter, or via the fallopian tubes in females. Hematogenous bacterial spread (through the bloodstream) occurs more frequently with tuberculosis peritonitis or peritonitis associated with cirrhotic ascites. When peritonitis results from peritoneal dialysis, skin surface flora are introduced via the peritoneal catheter. In secondary peritonitis, bacteria most often enter the peritoneum or retroperitoneum as a result of perforation of the GI or female genital tracts caused by diseases or traumatic injuries. In addition, peritonitis or abscess may result from contamination of the peritoneum during a surgical procedure or following anastomotic leak.
The physiologic characteristics of the peritoneal cavity determine the nature of the response to infection or inflammation within it.1,4 The peritoneum is lined by a highly permeable serous membrane with a surface area approximately that of skin. The peritoneal cavity is lubricated with less than 100 mL of sterile, clear yellow fluid, normally with fewer than 250 cells/mm3 (250 × 106/L), a specific gravity below 1.016, and protein content below 3 g/dL (30 g/L). These conditions change drastically with peritoneal infection or inflammation, as described below.
After bacteria are introduced into the peritoneal cavity, there is an immediate response to contain the insult. Humoral and cellular defenses respond first; then the omentum adheres to the affected area. A limited bacterial inoculum is handled rapidly by defense mechanisms, including complement activation and a leukocyte response. Under certain conditions, the bacterial insult is not contained, and bacteria disseminate throughout the peritoneal cavity, resulting in peritonitis. This is more likely to occur in the presence of a foreign body, hematoma, dead tissue, a large bacterial inoculum, continuing bacterial contamination, and contamination involving a mixture of synergistic organisms. Protein–calorie malnutrition, antecedent steroid therapy, and diabetes mellitus may also contribute to the formation of an intraabdominal abscess.
When bacteria become dispersed throughout the peritoneum, the inflammatory process involves most of the peritoneal lining. There is an outpouring into the peritoneum of fluid containing leukocytes, fibrin, and other proteins that form exudates on the inflamed peritoneal surfaces and begin to form adhesions between peritoneal structures. This process, combined with a paralysis of the intestines (ileus), may result in confinement of the contamination to one or more locations within the peritoneum. Fluid also begins to collect in the bowel lumen and wall, and distension may result.
The fluid and protein shift into the abdomen (called third-spacing) may be so dramatic that circulating blood volume is decreased, which may cause decreased cardiac output and hypovolemic shock. Accompanying fever, vomiting, or diarrhea may worsen the fluid imbalance. A reflex sympathetic response, manifested by sweating, tachycardia, and vasoconstriction, may be evident. With an inflamed peritoneum, bacteria and endotoxins are absorbed easily into the bloodstream (translocation), and this may result in septic shock.1,4,5 Other foreign substances present in the peritoneal cavity potentiate peritonitis. These adjuvants, notably feces, dead tissues, barium, mucus, bile, and blood, have detrimental effects on host defense mechanisms, particularly on bacterial phagocytosis.
Many of the manifestations of intraabdominal infections, particularly peritonitis, result from cytokine activity. Inflammatory cytokines, such as tumor necrosis factor-α (TNF-α), interleukin (IL) 1, IL-6, IL-8, and interferon γ(INF-γ), are produced by macrophages and neutrophils in response to bacteria and bacterial products or in response to tissue injury resulting from the surgical incision.1,4 These cytokines produce wide-ranging effects on the vascular endothelium of organs, particularly the liver, lungs, kidneys, and heart. With uncontrolled activation of these mediators, sepsis may result (see Chap. 97, Sepsis and Septic Shock).13,14
Peritonitis may result in death because of the effects on major organ systems. Fluid shifts, cytokines and endotoxin may result in hypovolemia, hypoperfusion, and shock. Hypoalbuminemia may result from protein loss into the peritoneum exacerbating intravascular volume loss. Pulmonary function may be compromised by the inflamed peritoneum, producing splinting (muscle rigidity caused by pain) that inhibits adequate diaphragmatic movement leading to atelectasis and pneumonia. Increased lung vascular permeability and resulting shunting of blood may induce onset of the respiratory distress syndrome and associated hypoxemia and hypercarbia. With fluid loss and hypotension, renal and hepatic perfusion may be compromised, and acute renal and hepatic failure are potential threats.
If peritoneal contamination is localized but bacterial elimination is incomplete, an abscess results. This collection of necrotic tissue, bacteria, and white blood cells may be at single or multiple sites and may be within one of the spaces of the peritoneal cavity or in one of the visceral organs. The location of the abscess is often related to the site of primary disease. For example, abscesses resulting from appendicitis tend to appear in the right lower quadrant or the pelvis; those resulting from diverticulitis tend to appear in the left lower quadrant or pelvis.
An abscess begins by the combined action of inflammatory cells (such as neutrophils), bacteria, fibrin, and other inflammatory mediators. Bacteria may release heparinases that cause local thrombosis and tissue necrosis or fibrinolysins, collagenases, or other enzymes that allow extension of the process into surrounding tissues. Neutrophils gathered in the abscess cavity die in 3 to 5 days, releasing lysosomal enzymes that liquefy the core of the abscess. A mature abscess may have a fibrinous capsule that isolates bacteria and the liquid core from antimicrobials and immunologic defenses.
Within the abscess, the oxygen tension is low and anaerobic bacteria thrive; thus, the size of the abscess may increase because it is hypertonic, resulting in an additional influx of fluid. Hypertonicity promotes the formation of bacterial L forms, which are resistant to antimicrobial agents that disrupt cell walls. Abscess formation may continue and mature for long periods of time and may not be readily evident to either patient or physician. In some instances, the abscess may resolve spontaneously, and, infrequently, it may erode into adjacent organs or rupture and cause diffuse peritonitis. If the abscess erodes through the skin, it may result in an enterocutaneous fistula, connecting bowel to skin, or in a draining sinus tract.
The overall outcome from an intraabdominal infection depends on key factors: inoculum size, virulence of the contaminating organisms, the presence of adjuvants within the peritoneal cavity that facilitate infection, the adequacy of host defenses, source control, and the adequacy of initial treatment.15,16
Microbiology of Intraabdominal Infection
Primary bacterial peritonitis is often caused by a single organism. In children, the pathogen is usually group A Streptococcus, E. coli, Streptococcus pneumoniae, or Bacteroides species.4,17–20 When peritonitis occurs in association with cirrhotic ascites, E. coli is isolated most frequently. Other potential pathogens are: Haemophilus influenzae, Klebsiella spp., Pseudomonas spp., anaerobes, and S. pneumoniae.21 Occasionally, primary peritonitis may be caused by Mycobacterium tuberculosis. Peritonitis in patients undergoing peritoneal dialysis is caused most often by common skin organisms, such as coagulase-negative staphylococci, Staphylococcus aureus, streptococci, and enterococci. Gram-negative bacteria associated with peritoneal dialysis infections include E. coli, Klebsiella spp., and Pseudomonasspp.6 Mortality from primary peritonitis caused by Gram-negative bacteria is much greater than that from Gram-positive bacteria.4,5
Because of the diverse bacteria present in the GI tract, secondary intraabdominal infections are often polymicrobial.2 The mean number of different bacterial species isolated from infected intraabdominal sites ranged from 2.9 to 3.7, including an average of 1.3 to 1.6 aerobes and 1.7 to 2.1 anaerobes.21,22 With proper anaerobic specimen collection, anaerobic organisms are isolated in most patients. In one report of patients with gangrenous and perforated appendicitis, an average of 10.2 different organisms was isolated from each patient, including 2.7 aerobes and 7.5 anaerobes.23 Purely aerobic or anaerobic infections are uncommon, as are infections caused by fungi. Table 92–3 gives the frequencies with which specific bacteria were isolated from patients with peritonitis and other intraabdominal infections.3,24 Nosocomial infections tend to have a more diverse array of pathogens and higher likelihood of multidrug resistance compared with isolates from community-acquired infections.25 E. coli, Streptococcus spp., and Bacteroides spp. were isolated most often from the infection site, as well as from blood cultures. In patients diagnosed with severe infections, the pattern of bacterial isolates may change and commonly includes Candida spp., enterococci, Enterobacteriaceae, and S. epidermidis.
TABLE 92-3 Pathogens Isolated from Patients with Intraabdominal Infection
Visceral organ abscesses differ in character from the typical intraabdominal abscess. Hepatic abscesses may be polymicrobial (involving E. coli, Klebsiella spp., and anaerobes) or occasionally may be caused by amoeba.11Pancreatic abscesses are often polymicrobial, involving enteric bacteria that ascend through the biliary system. Splenic abscesses usually result from hematogenous dissemination of bacteria, such as E. coli, S. aureus, Proteus mirabilis, Enterococcus spp., and K. pneumoniae, as well as anaerobes.11 Pelvic inflammatory disease is associated initially with Neisseria gonorrhoeae or Chlamydia trachomatis. However, tuboovarian abscesses are usually polymicrobial, having a mix of Gram-positive and Gram-negative aerobes and anaerobes.
The size of the bacterial inoculum and the number and types of bacterial species present in intraabdominal infections influence patient outcome. The combination of aerobic and anaerobic organisms appears to greatly increase the severity of infection. In animal studies, combinations of aerobic and anaerobic bacteria were much more lethal than infections caused by aerobes or anaerobes alone.
Facultative bacteria may provide an environment conducive to the growth of anaerobic bacteria.2 Although many bacteria isolated in mixed infections are nonpathogenic by themselves, their presence may be essential for the pathogenicity of the bacterial mixture.7 The role of facultative bacteria in mixed infections can include (a) promotion of an appropriate environment for anaerobic bacterial growth through oxygen consumption, (b) production of nutrients necessary for anaerobes, and (c) production of extracellular enzymes that promote tissue invasion by anaerobes.
Rat models of intraabdominal infection demonstrate that uncontrolled infection with an implanted mix of aerobes and anaerobes leads to a two-stage (biphasic) infectious process. There is an early peritonitis phase with a high mortality rate and isolation of E. coli from blood and a late abscess formation phase in all survivors with isolation of anaerobes such as B. fragilis and Fusobacterium varium. These experiments and others support the concept that aerobic enteric organisms and anaerobes are pathogens in intraabdominal infection. Aerobic bacteria, particularly E. coli, appear responsible for the early mortality from peritonitis, whereas anaerobic bacteria are major pathogens in abscesses, with B. fragilis predominating.26
Enterococcus can be isolated from many intraabdominal infections in humans, but its role as a pathogen is not clear. Enterococcal infection occurs more commonly in postoperative peritonitis, in the presence of specific risk factors indicating failure of the host’s defenses (immunocompromised patients), or with the use of broad-spectrum antibiotics.27,28
Intraabdominal infections have a wide spectrum of clinical features often depending on the specific disease process, the location and magnitude of bacterial contamination, and concurrent host factors. Peritonitis is usually recognized easily, but intraabdominal abscess may often continue for considerable periods of time, either going unrecognized or being attributed to an unrelated disease process. Patients with primary and secondary peritonitis present quite differently (Table 92–4).1,4,5
TABLE 92-4 Clinical Presentation of Peritonitis
Primary peritonitis can develop over a period of days to weeks and is usually a more indolent process than secondary peritonitis. The first sign of peritonitis may be a cloudy dialysate in patients undergoing peritoneal dialysis or worsening encephalopathy in a cirrhotic patient.
The patient with generalized bacterial peritonitis presents most often in acute distress. The patient lies still, usually on his or her back, possibly with the hips slightly flexed. Any movement of the patient, including rocking the bed or breathing, worsens the generalized abdominal pain.
If peritonitis continues untreated, the patient may experience hypovolemic shock from third-space fluid loss into the peritoneum, bowel wall, and lumen. This may be accompanied by sepsis because the inflamed peritoneum absorbs bacteria and toxins into mesenteric blood vessels and lymph nodes, initiating production of inflammatory cytokines. Hypovolemic shock is the major factor contributing to mortality in the early stage of peritonitis.
Intraabdominal abscess may pose a difficult diagnostic challenge because the symptoms are neither specific nor dramatic. The patient may complain of abdominal pain or discomfort, but these symptoms are not reliable. Fever is usually present; often it is low grade, but it may be high, with a spiking pattern. The patient may have a paralytic ileus and abdominal distension. The abdominal examination is unreliable; tenderness and pain may be present, and a mass may be palpated.
Peritonitis may result from an abscess that ruptures, spreading bacteria and toxins throughout the peritoneum. In other patients, the entry of bacterial toxins into the systemic circulation from the abscess may lead to sepsis and progressive multisystem organ failure (e.g., renal, hepatic, pulmonary, or cardiovascular).
Laboratory studies are not generally helpful in the diagnosis of intraabdominal abscess, although most patients will have leukocytosis. Some patients may have positive blood cultures, whereas others, particularly diabetics, may have hyperglycemia. The finding of Bacteroides or any two enteric bacteria in the bloodstream is often indicative of an intraabdominal infectious process.
Radiographic methods are used to make the diagnosis of an intraabdominal abscess. Plain radiographs may show air–fluid levels or a shift of normal intraabdominal contents by the abscess mass. GI contrast studies may also demonstrate this displacement of abdominal structures. Both of these modalities provide indirect evidence of abscess presence but are not generally helpful in precisely locating the abscess.
Ultrasound is a frequent first diagnostic method used when an intraabdominal abscess is suspected. The procedure may be done at the bedside, which is particularly helpful when the patient is in the intensive care unit.
Computed tomographic (CT) scanning is the preferred modality used to evaluate the abdomen for the presence of an abscess and is the imaging study of greatest value. If not contraindicated, an oral radiocontrast agent should be given to allow differentiation of the abscess from the bowel. IV radiocontrast material will be taken up preferentially in the wall of the abscess, creating a unique radiographic appearance, so-called rim enhancement. Magnetic resonance imaging offers no significant advantage when compared with CT scanning.
Intraabdominal infection caused by disease processes at specific sites often produces characteristic manifestations that are helpful in diagnosis. For example, a patient with diverticulitis may exhibit stabbing left-lower-quadrant abdominal pain and constipation. Fever and leukocytosis are frequently present, and a tender mass is sometimes palpable. With appendicitis, the findings may be inconsistent, but many patients have a sudden onset of periumbilical or epigastric pain that is usually colicky and later shifts to the right lower quadrant. The location of pain may vary because the appendix can be in many locations (e.g., retrocecal or pelvic) in the abdomen. A mass may be palpable on abdominal, pelvic, or rectal examination. The patient’s temperature is generally mildly elevated early and then increases. If perforation and peritonitis occur, findings would include diffuse abdominal pain, rigidity, and sustained fever. More often, however, appendiceal perforation results in a local abscess.
The primary goals of treatment are correction of the intraabdominal disease processes or injuries that have caused infection and the drainage of purulent collections (abscesses). A secondary objective is to achieve a resolution of infection without major organ system complications (pulmonary, hepatic, cardiovascular, or renal failure) or adverse drug effects. Ideally, the patient should be discharged from the hospital after treatment with full function for self-care and routine daily activities.
General Approach to Treatment
The treatment of intraabdominal infection most often requires hospitalization and the coordinated use of three major modalities: (a) prompt drainage of the infected site, (b) hemodynamic resuscitation and support of vital organ functions, and (c) early administration of appropriate antimicrobial therapy to treat infection not eradicated by surgery.2
Antimicrobials are an important adjunct to drainage procedures in the treatment of secondary intraabdominal infections; however, the use of antimicrobial agents without surgical intervention is usually inadequate. For most cases of primary peritonitis, drainage procedures may not be required, and antimicrobial agents become the mainstay of therapy.
In the early phase of serious intraabdominal infections, attention should be given to the maintenance of organ system functions. With generalized peritonitis, large volumes of IV fluids are required to restore vascular volume, to improve cardiovascular function, and to maintain adequate tissue perfusion and oxygenation. Adequate urine output should be maintained to ensure adequate resuscitation and proper renal function. Respiratory function can be assisted by a variety of methods, including oxygen therapy, pulmonary physiotherapy, and ventilatory support in severely ill patients. Often the critically ill patient with intraabdominal infection will require intensive care management, particularly if there is cardiovascular or respiratory instability. In addition, isolation procedures may be required if the infectious process poses a threat to other hospitalized patients.
An additional important component of therapy is nutrition. Intraabdominal infections often directly involve the GI tract or disrupt its function (paralytic ileus). The return of GI motility may take days, weeks, and, occasionally, months. In the interim, enteral or parenteral nutrition as indicated facilitates improved immune function and wound healing to ensure recovery.
Primary peritonitis is treated with antimicrobials and rarely requires drainage. Secondary peritonitis requires surgical correction of the underlying pathology. The drainage of the purulent material is the critical component of management of an intraabdominal abscess. Without adequate drainage of the abscess, antimicrobial therapy and fluid resuscitation can be expected to fail.
Secondary peritonitis is treated surgically; this is often called source control, which refers to all the physical measures undertaken to eradicate the focus of infection.2,5 At the time of laparotomy (surgical opening and exploration of the abdomen), attempts are made to correct the cause of the peritonitis. This may include patching a perforated ulcer with omentum, removal of a segment of perforated colon, or excision of a portion of gangrenous small intestine. In addition, the surgeon may elect to leave the abdomen open after the laparotomy, plan a re-laparotomy at a later time regardless of the patient’s condition, or, perform re-laparotomy if the patient develops reinfection.5 The goal of all these procedures is to repair or remove the inflamed or gangrenous viscus and to prevent further bacterial contamination. The presence of active inflammation increases the difficulty of the surgical procedure, which results in a higher morbidity and mortality rate than if the same procedures were performed in an elective setting without inflammation.
The presence of active inflammation may make it technically impossible to perform the definitive surgical procedure. In this situation, attempts are made to provide drainage of the infected or gangrenous structures. If an intraabdominal abscess, separate from any intraabdominal organ, is discovered during an exploratory laparotomy, it may be debrided, excised, or drained. If the intraabdominal abscess involves an abdominal structure, then a resection of part or of the entire organ may be required. An example of this situation is an abscess associated with diverticular disease of the colon. Management may include drainage of the abscess and resection of the involved part of the colon. All foreign material, necrotic tissue, feces, blood, or pus should be removed from the operative field, and the peritoneum should be copiously irrigated with 0.9% sodium chloride to decrease the concentrations of bacteria or other noxious substances.
After an abscess is located, it must be drained. This may be performed surgically or with percutaneous, image-guided techniques.5,29 Typically, image-guided techniques employ ultrasonography or CT scanning. The management of an intraabdominal abscess with percutaneous catheter drainage may be sufficient to resolve the infection. Some patients may require a subsequent procedure to treat the underlying GI conditions; however, a significant advantage is obtained by first draining the abscess percutaneously. This allows the surgical procedure to be performed on a patient who is no longer suffering the systemic manifestations of uncontrolled infection. Drainage techniques may be performed using endoscopy or laparoscopy. These minimal-access techniques may offer advantages when compared with traditional surgery but will probably be used less often than radiologically assisted percutaneous drainage techniques.
The most valuable microbiologic information may be obtained at the time of percutaneous or operative abscess drainage. If pus or fluid is found that is believed to be infected, it is best to aspirate 2 to 3 mL into a syringe, remove any air, and tightly cap the syringe. The specimen should be taken promptly to the microbiology laboratory, where a Gram stain should be performed immediately and cultures prepared for identification of aerobic and anaerobic bacteria. If no fluid is available for collection, culture swab devices may be applied to the infected area; however, anaerobic organisms often are not isolated from swabs.
Patients should be evaluated for signs of hypovolemia, hypoperfusion, and shock. Aggressive fluid repletion and management are required for successful treatment of intraabdominal infections. The Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock recommend treatment goals during the first 6 hours or resuscitation: (a) central venous pressure (CVP) 8 to 12 mm Hg, (b) mean arterial pressure (MAP) ≥65 mm Hg, and (c) maintain urine output ≥0.5 mL/kg/h.30,31 If the patient is mechanically ventilated a target CVP 12 to 15 mm Hg should be achieved.30 Fluid therapy is instituted for the purposes of achieving or maintaining proper intravascular volume to ensure adequate cardiac output, tissue perfusion, and correction of acidosis. Loss of fluid through vomiting, diarrhea, or nasogastric suction contributes to dehydration. Intravascular volume can be assessed by blood pressure and heart rate but more accurately by measurement of CVP or urinary output. When a contracted vascular volume is accompanied by hemorrhage, the initial hematocrit may be normal, but if there is no associated hemorrhage, the hematocrit is usually elevated as an indication of hemoconcentration. Urine output should be monitored continuously in severely ill patients by use of a urinary bladder catheter, quantitated hourly, and should equal or exceed 0.5 mL/kg of body weight per hour.
In patients with peritonitis, hypovolemia is often accompanied by metabolic acidosis. IV fluids should consist of a bolus of crystalloids or colloids with additional fluids targeting predefined therapeutic goals.30,31 In the initial hour of treatment, large volumes of solution may be required to restore intravascular volume. Thereafter, fluids may be required at a rate of 1 L/h or higher. Once targeted therapeutic goals are reached, maintenance fluids should be instituted with 0.9% sodium chloride and potassium chloride (20 mEq/L [20 mmol/L]) or 5% dextrose and 0.45% sodium chloride with potassium chloride (20 mEq/L [20 mmol/L]). The administration rate should be based on estimated daily fluid loss through urine and nasogastric suction, including 0.5 to 1 L for insensible fluid loss. Potassium would not be included routinely if the patient is hyperkalemic or has renal insufficiency. If appropriate fluid management fails to restore target goals of perfusion, vasopressor therapy should be initiated.30 A more thorough discussion of fluid and vasopressor therapy are presented elsewhere in this text (Chaps. 13, 14, and 96).
In patients with significant blood loss, blood transfusion may be indicated. This is generally in the form of packed red blood cells. The criteria for blood transfusion are controversial, but a hematocrit of 25% is generally accepted. In the individual patient, the decision is often determined by the overall clinical status and the ability of the patient to compensate for the reduction in oxygen-carrying capacity associated with an acute anemia. Additional blood component therapy with fresh-frozen plasma or platelets is also based on the needs of the individual patient. Aggressive fluid therapy must often be continued in the postoperative period because fluid will continue to sequester in the peritoneal cavity, bowel wall, and lumen.
The goals of antimicrobial therapy are (a) to control bacteremia and prevent the establishment of metastatic foci of infection, (b) to reduce suppurative complications after bacterial contamination, and (c) to prevent local spread of existing infection. After suppuration has occurred (e.g., an abscess has formed), a cure by antibiotic therapy alone is very difficult to achieve; antimicrobials may serve to improve the results obtained with surgery.
An empirical antimicrobial regimen should be started as soon as the presence of intraabdominal infection is suspected. Therefore, antibiotics are usually initiated after culture specimens are collected but before identification of the infecting organisms is complete. Therapy must be initiated based on the likely pathogens. Increasing resistance to fluoroquinolones, ampicillin–sulbactam, and clindamycin emphasize the importance of utilizing local susceptibility data for empiric therapy and tailoring the antibiotic regimen based on susceptibility results. Predominant pathogens, as discussed in the preceding section, vary depending on the site of intraabdominal infection and the underlying disease process. Table 92–5 lists the likely pathogens against which antimicrobial agents should be directed.
TABLE 92-5 Likely Intraabdominal Pathogens
Antimicrobial Experience Many studies have been conducted evaluating or comparing the effectiveness of antimicrobials for the treatment of intraabdominal infections. Substantial differences in patient outcomes from treatment with a variety of agents have not generally been demonstrated.32
Important findings from over 20 years of clinical trials regarding selection of antimicrobials for intraabdominal infections are the following:
1. Antimicrobial regimens used for secondary infections should cover a broad spectrum of aerobic and anaerobic bacteria from the GI tract. Empiric treatment should be guided by the local epidemiology of resistant pathogens and patient-specific risk factors for acquisition pathogens of concern.
2. Single-agent regimens (such as antianaerobic cephalosporins, extended-spectrum penicillins with β-lactamase inhibitors, and carbapenems) are as effective but have the benefit of being less nephrotoxic compared to combinations of aminoglycosides with antianaerobic agents. This is also true for antimicrobial treatment of acute bacterial contamination from penetrating abdominal trauma.33,34
3. Resistance is now prevalent among B. fragilis to clindamycin and cefotetan and E. coli to ampicillin–sulbactam and quinolones and therefore these agents should not be routinely used empirically for complicated intraabdominal infections.35,36
4. If susceptible, antimicrobial treatment can be completed orally with amoxicillin–clavulanate, metronidazole with either ciprofloxacin or levofloxacin, or moxifloxacin.37
5. Four to seven days of antimicrobial treatment is sufficient for most intraabdominal infections with adequate source control.39
Intraabdominal infection presents in many different ways and with a wide spectrum of severity. The regimen employed and duration of treatment depends on the specific clinical circumstances (i.e., the nature of the underlying disease process, severity of illness, and risk of resistant pathogens).
Recommendations For most intraabdominal infections, the antimicrobial regimen should be effective against both aerobic and anaerobic bacteria.38,39 When initial antimicrobial therapy is inactive, morbidity and mortality rates are higher than when initially active therapy is used.39 Although it is impossible to provide antimicrobial activity against every possible pathogen, agents with activity against enteric Gram-negative bacilli such as E. coli and Klebsiella spp., and anaerobes such as B. fragilis should be administered. If most of the organisms can be eliminated through drainage or antimicrobials, the synergistic effect may be removed, and the patient’s defenses may be able to resolve the remaining infection.
Table 92–6 presents the recommended agents for treatment of community-acquired complicated intraabdominal infections from the Infectious Diseases Society of America and the Surgical Infection Society.39These recommendations were formulated using an evidence-based approach. Table 92–7 lists additional evidence-based recommendations for the treatment of complicated intraabdominal infections. Most community-acquired infections are of mild-to-moderate severity whereas healthcare-associated infections tend to be more severe, more difficult to treat, and more commonly due to resistant pathogens. Table 92–8 presents guidelines for treatment and alternative regimens for specific situations. These are general guidelines; there are many factors that cannot be incorporated into such a table including local resistance patterns to commonly used agents such as quinolones.
TABLE 92-6 Recommended Agents for the Treatment of Community-Acquired Complicated Intraabdominal Infections in Adults
TABLE 92-7 Evidence-Based Recommendations for Treatment of Complicated Intraabdominal Infections
TABLE 92-8 Guidelines for Empiric Antimicrobial Agents for Intraabdominal Infections39,49
Most patients with severe intraabdominal infection, sepsis of intraabdominal source, or healthcare-associated infection should be placed on piperacillin–tazobactam, cefepime with metronidazole, or a carbapenem such as imipenem, doripenem, or meropenem. In patients with IgE-mediated allergic reactions to β-lactams (hives/urticaria, bronchospasm, angioedema, or anaphylaxis), combination therapy with aztreonam–vancomycin and metronidazole may be used. The benefits of systemic empiric antifungal (with fluconazole) have not been established for intraabdominal infection and should not be used routinely.
Aminoglycoside-based treatment regimens are no longer routinely recommended due to their narrow therapeutic index (nephrotoxicity, ototoxicity) relative to the recommended agents such as β-lactams. Aminoglycosides are typically reserved for use in patients with IgE-mediated allergic reactions to alternative agents or as dictated by the susceptibility of the presumed or proven infecting pathogen(s).32,39
The initial dosage for aminoglycosides should be determined based on the patient’s weight and renal function. Traditionally, gentamicin and tobramycin were administered multiple times daily with specific peak (6 to 10 mcg/mL [mg/L; 13 to 21 μmol/L]) and trough (<1 to 2 mcg/mL [mg/L; <2 to 4 μmol/L]) concentration targets. Because aminoglycosides have concentration-dependent killing and have a relatively long postantibiotic effect for aerobic Gram-negative bacilli, extended-interval dosing of aminoglycosides is possible. For most patients and indications, extended-interval aminoglycoside dosing (i.e., 5 to 7 mg/kg once daily for tobramycin or gentamicin) has replaced traditional dosing given equivalent efficacy and decreased nephrotoxicity.40–42
Antimicrobial resistance continues to rise worldwide while at the same time few new agents, particularly for multidrug-resistant Gram-negative pathogens, are being brought to market.43 These problematic multidrug-resistant bacteria include enteric pathogens producing extended-spectrum β-lactamases (ESBL) which have been increasingly isolated from intraabdominal cultures.35 For patients with ESBL-producing pathogens, carbapenems (such as imipenem–cilastatin, meropenem, or ertapenem) are typically the drugs of choice. With increased use of carbapenems, pathogens continue to evolve with the development of β-lactamases that hydrolyze carbapenems (e.g., Klebsiella pneumoniae carbapenemase [KPC]), multidrug-resistant Pseudomonas spp., and carbapenem-resistant Acinetobacter spp. Especially in patients with healthcare-associated intraabdominal infections, these multidrug-resistant pathogens have caused clinicians to use more toxic and potentially less effective agents such as the polymyxins/colistin, tigecycline, and aminoglycosides. This increasing resistance highlights the need, from an individual patient and public health standpoint, to ensure that antimicrobials are selected appropriately, at the right dose, and for the right duration.
Due to recent reports of increased risk of mortality associated with tigecycline, some clinicians recommend the drug be reserved for the treatment of infections due to multidrug-resistant pathogens that are sensitive to tigecycline.44,45Others cite the lack of clinical trial evidence in seriously ill patients or those with multidrug-resistant pathogens as a reason for hesitation in using tigecycline in these patient populations.46
With intraabdominal contamination from the upper GI tract (perforation of a peptic ulcer or biliary tract disease), B. fragilis is an uncommon pathogen, and other agents therefore may be substituted for metronidazole. Alternatives include ampicillin, penicillin, or first-generation cephalosporins.
Coverage for enterococci for mild-to-moderate community-acquired intraabdominal infections is not recommended.39 The failure of host defenses may be a critical factor in the pathogenicity of enterococci. In patients with severe community-acquired intraabdominal infection or patients with healthcare-associated infection, it is recommended to include coverage of enterococcus in the initial regimen.39 Ampicillin remains the drug of choice for this indication because it is most active in vitro against enterococcus. Vancomycin is active against most enterococci; however, rates of vancomycin-resistant enterococci are increasing, particularly in select patient populations (e.g., liver transplantation, immunocompromised patients).47 Agents including linezolid or daptomycin are commonly utilized for vancomycin-resistant enterococcus infections. Table 92–7 lists additional evidence-based recommendations for Enterococcus spp. coverage.
Enterococci are often isolated from intraabdominal infections, and many antimicrobials are ineffective against enterococci (such as cephalosporins and quinolones). Regimens without activity against enterococci (e.g., cefepime with metronidazole) are generally effective in treating intraabdominal infections; however, there are numerous reports of enterococcal superinfection in immunocompromised patients, particularly after broad-spectrum antimicrobial use. The Infectious Disease Society of America and Surgical Infection Society guidelines39 recommend empiric coverage of Enterococcus for severe or high-risk community-acquired infection and healthcare-associated infection. The guidelines indicate, “Empiric coverage of Enterococcus is not necessary in patients with (mild-moderate) community-acquired intraabdominal infection.”
Intraperitoneal administration of antibiotics is preferred over IV therapy in the treatment of peritonitis that occurs in patients undergoing CAPD.48 The International Society of Peritoneal Dialysis guidelines for the diagnosis and pharmacotherapy of peritoneal dialysis-associated infections provide dosing recommendations for intermittent and continuous therapy based on the modality of dialysis (CAPD or automated peritoneal dialysis) and the extent of the patient’s residual renal function.49 Third-generation cephalosporins, such as ceftriaxone, remain the treatments of choice for primary peritonitis associated with cirrhosis.50
Antimicrobial agents effective against both Gram-positive (including Staphylococcus aureus) and Gram-negative organisms should be used for initial intraperitoneal empiric therapy for peritonitis in peritoneal dialysis patients. The most important factors to take into consideration for initial antimicrobial selection are the dialysis center’s and the patient’s history of infecting organisms and their sensitivities. For empiric intraperitoneal therapy, cefazolin (loading dose [LD] 500 mg/L; maintenance dose [MD] 125 mg/L) or vancomycin (LD 1,000 mg/L; MD 25 mg/L) in cases of high prevalence of methicillin-resistant Staphylococcus aureus (MRSA) or β-lactam allergy may be utilized for Gram-positive coverage. One of these Gram-positive agents should be combined with a Gram-negative agent such as ceftazidime (LD 500 mg/L; MD 125 mg/L) or cefepime (LD 500 mg/L; MD 125 mg/L) or an aminoglycoside (gentamicin or tobramycin LD 8 mg/L; MD 4 mg/L). Another option is monotherapy with cefepime or imipenem–cilastatin (LD 250 mg/L; MD 50 mg/L). Antimicrobial doses should empirically be increased by 25% in patients with residual renal function (more than 100 mL/day urine output).49 Antimicrobial therapy should be continued for at least 1 week after the dialysate fluid is clear and for a total of at least 14 days. The reader is referred to these guidelines for additional information.49
After acute bacterial contamination, such as with abdominal trauma where GI contents spill into the peritoneum, antibiotics should be administered. If the patient is seen soon after injury (within 2 hours) and surgical measures are instituted promptly, antianaerobic cephalosporins (such as cefoxitin), a third-generation cephalosporin (such as ceftriaxone or cefuroxime) with metronidazole, or piperacillin/tazobactam are effective in preventing most infectious complications. Antimicrobials should be administered as soon as possible after injury.
For appendicitis, the antimicrobial regimen used should depend on the appearance of the appendix at the time of operation, which may be normal, inflamed, gangrenous, or perforated. Because the condition of the appendix is unknown preoperatively, it is advisable to begin antimicrobial agents before the appendectomy is performed. Reasonable regimens would be antianaerobic cephalosporins or, if the patient is seriously ill, piperacillin–tazobactam or a carbapenem (such as imipenem–cilastatin or meropenem). If, at operation, the appendix is normal or inflamed, postoperative antimicrobials are not required. If the appendix is gangrenous or perforated, a treatment course of 4 to 7 days with the agents listed in Table 92–6 is appropriate.
The necessary duration of treatment for intraabdominal infections is not clearly defined. Acute intraabdominal contamination, such as after a traumatic injury, may be treated with a very short course (24 hours).51 For established infections (i.e., peritonitis or intraabdominal abscess), an antimicrobial course limited to 4 to 7 days is justified. This allows eradication of bacteria remaining in the peritoneum after a surgical procedure that may enter the peritoneum through healing suture lines. Under certain conditions, therapy for longer than 7 days would be justified (e.g., if the patient remains febrile or is in poor general condition, or when a focus of infection in the abdomen is still present). For some abscesses, such as pyogenic liver abscess, antimicrobials may be required for a month or longer.
Intraperitoneal irrigation of antimicrobial agents for treatment of intraabdominal infection has been studied, often with conflicting results.52 Intraoperative antimicrobial irrigation does not improve patient outcomes in comparison with copious intraoperative irrigation with normal saline. Possibly the most important aspect of peritoneal irrigation is the dilutional effect on bacteria and adjuvants that promotes infection (intestinal contents and hemoglobin). Most systemically administered antimicrobials easily cross the peritoneal membrane so that peritoneal fluid concentrations are similar to serum. Confined areas, such as an abscess, can be expected to attain much lower antimicrobial concentrations.
EVALUATION OF THERAPEUTIC OUTCOMES
Whichever antimicrobial regimen is chosen, the patient should be reassessed continually to determine the success or failure of therapies. The clinician should recognize that there are many reasons for poor patient outcome with intraabdominal infection; improper antimicrobial administration is only one. The patient may be immunocompromised, which decreases the likelihood of successful outcome with any regimen. It is impossible for antimicrobials to compensate for a nonfunctioning immune system. There may be surgical reasons for poor patient outcome. Failure to identify all intraabdominal foci of infection or leaks from a GI anastomosis may cause continued intraabdominal infection. Even when intraabdominal infection is controlled, accompanying organ system failure, most often renal or respiratory, may lead to patient demise. Finally, antimicrobial resistance may relate to treatment failure as isolates from intraabdominal infections are increasingly drug resistant.53
The outcome from intraabdominal infection is not determined solely by what transpires in the abdomen. Unsatisfactory outcomes in patients with intraabdominal infections may result from complications that arise in other organ systems. Infectious complications commonly associated with mortality after intraabdominal infection are urinary tract infections and pneumonia.54 A high APACHE (Acute Physiology and Chronic Health Evaluation) II score, low serum albumin concentration, and high New York Heart Association cardiac function status were significantly and independently associated with increased mortality from intraabdominal infection.55
Once antimicrobials are initiated and the other important therapies described earlier are used, most patients should show improvement within 2 to 3 days. Usually, temperature will return to near normal, vital signs should stabilize, and the patient should not appear in distress, with the exception of recognized discomfort and pain from incisions, drains, and the nasogastric tube. At 24 to 48 hours, aerobic bacterial culture results should return. If a suspected pathogen is not sensitive to the antimicrobial agents being given, the regimen should be changed if the patient has not shown sufficient improvement. If the isolated pathogen is susceptible to one antimicrobial and the patient is progressing well, antimicrobial therapy may often be deescalated.
With anaerobic culturing techniques and the slow growth of these organisms, anaerobes are often not identified until 4 to 7 days after culture, and sensitivity information is difficult to obtain. For this reason, there are usually few data with which to alter the antianaerobic component of the antimicrobial regimen. A report indicating that anaerobes were not isolated should not be the sole justification for discontinuing antianaerobic drugs because anaerobic bacteria that were present in the infectious process may not have been transported properly to the microbiology laboratory, or other problems may have led to cell death in vitro.
Although some investigators suggest that routine culturing of patients with community-acquired intraabdominal infections contributes little to their management,56 other investigators suggest that antimicrobial therapy should be based on susceptibility of the bacteria collected from the operative site because this correlates with clinical outcome.57
Reasons for antimicrobial failure may not always be apparent. Even when antimicrobial susceptibility tests indicate that an organism is susceptible in vitro to the antimicrobial agent, therapeutic failures may occur. Possibly there is poor penetration of the antimicrobial agent into the focus of infection, or bacterial resistance may develop after initiation of antimicrobial therapy. In addition, it is possible that an antimicrobial regimen may encourage the development of infection by organisms not susceptible to the regimen being used. Superinfection in patients being treated for intraabdominal infection can be caused by Candida; however, enterococci or opportunistic Gram-negative bacilli such as Pseudomonas or Serratia may be involved.
Treatment regimens for intraabdominal infection can be judged as successful if the patient recovers from the infection without recurrent peritonitis or intraabdominal abscess and without the need for additional antimicrobials. A regimen can be considered unsuccessful if a significant adverse drug reaction occurs, reoperation or percutaneous drainage is necessary, or patient improvement is delayed beyond 1 or 2 weeks. The costs of treatment can be significantly reduced if parenteral antimicrobials can be switched to oral agents for completion of therapy.58
1. Ordonez CA, Puyana JC. Management of peritonitis in the critically ill patient. Surg Clin North Am 2006;86: 1323–1349.
2. Marshall JC. Intra-abdominal infections. Microbes Infect 2004;6:1015–1025.
3. Marshall JC, Innes M. Intensive care unit management of intraabdominal infection. Crit Care Med 2003;31: 2228–2237.
4. Levison ME, Bush LM. Peritonitis and intraperitoneal abscesses. In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 7th ed. Philadelphia: Saunders, 2010:1011–1034 [chapter 71].
5. Sartelli M, Viale P, Koike K et al. WSES consensus conference: Guidelines for the first-line management of intra-abdominal infections. World J Emerg Surg 2011;6:1–29.
6. Wiest R, Krag A, Gerbes A. Spontaneous bacterial peritonitis: Recent guidelines and beyond. Gut 2012;61: 297–310.
7. Mowat C, Stanley AJ. Spontaneous bacterial peritonitis—Diagnosis, treatment, and prevention. Aliment Pharmacol Ther 2001;15:1851–1859.
8. Mujais S. Microbiology and outcomes of peritonitis in North America. Kidney Int 2006;70:555–562.
9. Buie VC, Owings MF, DeFrances CJ, Golosinskiy A. National Hospital Discharge Survey: 2006 summary. National Center for Health Statistics. Vital Health Stat 2010;13(168):22–53.
10. Toloza EM, Wilson SE. Cholecystitis and cholangitis. In: Fry DE, ed. Surgical Infections. Boston: Little, Brown, 1995:254–263.
11. Brook I. Microbiology and management of abdominal infections. Dig Dis Sci 2008;53:2585–2591.
12. Jernberg C, Lofmark S, Edlund C, Jansson JK. Long-term impacts of antibiotic exposure on the human intestinal microbiota. Microbiology 2010;156:3216–3223.
13. Riche FC, Cholley BP, Panis YH, et al. Inflammatory cytokine response in patients with septic shock secondary to generalized peritonitis. Crit Care Med 2000;28:433–437.
14. Solomkin JS, Mazuski J. Intra-abdominal sepsis: Newer interventional and antimicrobial therapies. Infect Dis Clin North Am 2009;23:593–608.
15. Malangoni MA. Contributions to the management of intraabdominal infection. Am J Surg 2005;190:255–259.
16. Herzog T, Chromic, Uhl W. Treatment of complicated intra-abdominal infections in the era of multidrug resistant bacteria. Eur J Med Res 2010;15:525–532.
17. Thompson AE, Marshall JC, Opal SM. Intraabdominal infections in infants and children: Descriptions and definitions. Pediatr Crit Care Med 2005;6:S30–S35.
18. Rice-Townsend SE, Lawrence Moss R, Rangel SJ. Peritonitis. In: Long SS, Pickering LK, Prober CG, eds. Principles and Practice of Pediatric Infectious Diseases, 4th ed. Elsevier Churchill Livingstone; 2012.
19. Guillet-Caruba C, Cheikhelard A, Guillet M, et al. Bacteriologic epidemiology and empirical treatment of pediatric complicated appendicitis. Diagn Microbiol Infect Dis 2011;69:376–381.
20. Lee SL, Islam S, Cassidy LD, et al. Antibiotics and appendicitis in the pediatric population: An American Pediatric Surgical Association Outcomes and Clinical Trials Committee systematic review. J Pediatr Surg 2010;45: 2181–2185.
21. Johnson DH, Cuhna BA. Infections in cirrhosis. Infect Dis Clin North Am 2001;15:363–371.
22. Brook I, Frazier EH. Aerobic and anaerobic microbiology of retroperitoneal abscesses. Clin Infect Dis 1998;26:938–941.
23. Bennion RS, Baron EJ, Thompson JE, et al. The bacteriology of gangrenous and perforated appendicitis—Revisited. Ann Surg 1990;211:165–171.
24. Sawyer RG, Rosenlof LK, Adams RB, et al. Peritonitis into the 1990s: Changing pathogens and changing strategies in the critically ill. Am Surg 1992;58:82–87.
25. Montravers P, Lepape A, Dubreuil L, et al. Clinical and microbiological profiles of community-acquired and nosocomial infections: Results of the French prospective, observational EBIIA study. J Antimicrob Chemother 2009;63:785–794.
26. Onderdonk AB, Bartlett JG, Louie T, et al. Microbial synergy in experimental intraabdominal abscess. Infect Immun 1997;13:22–26.
27. Donskey CJ, Chowdhry TK, Hecker MT, et al. Effect of antibiotic therapy on the density of vancomycin-resistant enterococci in the stool of colonized patients. Ann Surg 2000;343:1925–1932.
28. Sitges-Serra A, Lopez MJ, Girvent M, et al. Postoperative enterococcal infection after treatment of complicated intraabdominal sepsis. Br J Surg 2002;89:361–367.
29. Jaffe TA, Nelson RC, Delong DM, Paulson EK. Practice patterns in percutaneous image-guided intraabdominal abscess drainage: Survey of academic and private practice centers. Radiology 2004;233:750–756.
30. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013; 41:580–637.
31. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345:1368–1377.
32. Wong PF, Gilliam AD, Kumar S, et al. Antibiotic regimens for secondary peritonitis of gastrointestinal origin in adults. Cochrane Database Syst Rev 2007;2:CD004539.
33. Hooker KD, DiPiro JT, Wynn JJ. Aminoglycoside combinations versus single β-lactams for penetrating abdominal trauma: A meta-analysis. J Trauma 1991;31: 1155–1160.
34. Solomkin JS, Dellinger EP, Christou NV, et al. Results of a multicenter trial comparing imipenem/cilastatin to tobramycin/clindamycin for intraabdominal infections. Ann Surg 1990;212:581–591.
35. Hoban DJ, Bouchillon SK, Hawser SP, Badal RE, Labombardi VJ, DiPersio J. Susceptibility of gram-negative pathogens isolated from patients with complicated intra-abdominal infections in the United States, 2007–2008: Results of the Study for Monitoring Antimicrobial Resistance Trends (SMART). Antimicrob Agents Chemother 2010;54:3031–3034.
36. Snydman DR, Jacobus NV, McDermott LA, et al. Lessons learned from the anaerobe survey: Historical perspective and review of the most recent data (2005–2007). Clin Infect Dis 2010;50(Suppl 1):S26–S33.
37. Hawser SP, Bouchillon SK, Hoban DJ, Badal RE. In vitro susceptibilities of aerobic and facultative anaerobic Gram-negative bacilli from patients with intra-abdominal infections worldwide from 2005–2007: Results from the SMART study. Int J Antimicrob Agents 2009;34: 585–588.
38. Gauzit R, Pean Y, Mistretta F, Lalaude O. Epidemiology, management, and prognosis of secondary non-postoperative peritonitis: A French prospective observational multicenter study. Surg Infect 2009;10:119–127.
39. Solomkin JS, Mazuski JE, Bradley JS, et al. Diagnosis and management of complicated intra-abdominal infection in adults and children: Guidelines by the Surgical Infection Society and the Infectious Diseases society of America. Clin Inf Dis 2010;50:133–164.
40. Nicolau DP, Freeman CD, Belliveau PP, Nightingale CH, Ross JW, Quintiliani R. Experience with a once-daily aminoglycoside program administered to 2184 adult patients. Antimicrob Agents Chemother 1995;39:650–655.
41. Rybak MJ, Abate BJ, Kang SL, Ruffing MJ, Lerner SA, Drusano GL. Prospective evaluation of the effect of an aminoglycoside dosing regimen on rates of observed nephrotoxicity and ototoxicity. Antimicrob Agents Chemother 1999;43:1549–1555.
42. Olsen KM, Rudis MA, Rebuck JA, Gelmont D, Mehdian R, Nelson C, Rupp ME. Effect of single daily dose vs. multiple daily dosing of tobramycin on alanine aminopeptidase and N-acetyl-β-D-glucosaminidase excretion. Crit Care Med 2004;32:1678–1682.
43. Boucher HW, Talbot GH, Bradley JS, et al. Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America. Clin Infect Dis 2009;48: 1–12.
44. Mutnick AH, Biedenbach DJ, Jones RN. Geographic variations and trends in antimicrobial resistance among Enterococcus faecalis and Enterococcus faecium in the SENTRY Antimicrobial Surveillance Program (1997–2000). Diagn Microbiol Infect Dis 2003;46:63–68.
45. Prasad P, Sun J, Danner R, Natanson C. Excess deaths associated with tigecycline after approval based on non-inferiority trials. Clin Infect Dis 2012;54:1699–1709.
46. US Food and Drug Administration. FDA Drug Safety Communication: Increased risk of death with Tygacil (tigecycline) compared to other antibiotics used to treat similar infections. US Food and Drug Administration. 1 September 2010. http://www.fda.gov/Drugs/DrugSafety/ucm224370.htm.
47. Powers JH. Asking the right questions: Morbidity, mortality, and measuring what’s important in unbiased evaluations of antimicrobials. Clin Infect Dis 2012;54:1710–1713.
48. Wiggins KJ, Craig JC, Johnson DW, Strippoli GF. Treatment for peritoneal dialysis-associated peritonitis. Cochrane Database Syst Rev 2008;(1):CD005284.
49. Li PK, Szeto CC, Piraino B, et al. International Society for Peritoneal Dialysis Peritoneal dialysis-related infections recommendations: 2010 update. Perit Dial Int 2010;30: 393–423.
50. Runyon BA. Management of adult patients with ascites due to cirrhosis: An update. Hepatology 2009;49:2087–2107.
51. Bozorgzadeh A, Pizzi WF, Barie PS, et al. The duration of antibiotic administration in predicting abdominal trauma. Am J Surg 1999;172:125–135.
52. Schein M, Gecelter G, Freinkel W, et al. Peritoneal lavage in abdominal sepsis: A controlled clinical study. Arch Surg 1990;125:1132–1135.
53. Baquero F, Hsueh P, Paterson DL, et al. In vitro susceptibilities of aerobic and facultatively anaerobic gram-negative bacilli isolated from patients with intra-abdominal infections worldwide: 2005 results from study for monitoring antimicrobial resistance trends. Surg Infect 2009;10: 99–104.
54. Merlino JI, Yowler CJ, Malangoni MA. Nosocomial infections adversely affect the outcomes of patients with serious intraabdominal infections. Surg Infect (Larchmt) 2004;5:21–27.
55. Christou NV, Barie PS, Dellinger EP, et al. Surgical infection society intraabdominal infection study. Arch Surg 1993;128:193–199.
56. Dougherty SH. Antimicrobial culture and susceptibility testing has little value for routine management of secondary bacterial peritonitis. Clin Infect Dis 1997;25(Suppl 2): S258–S261.
57. Montravers P, Gauzit R, Muller C, Marmuse JP, Fichelle A, Desmonts JM. Emergence of antibiotic-resistant bacteria in cases of peritonitis after intraabdominal surgery affects the efficacy of empirical antimicrobial therapy. Clin Infect Dis 1996;23(3):486–494.
58. Paladino JA, Gilliland-Johnson KK, Adelman MH, Coohn SM. Pharmacoeconomics of ciprofloxacin plus metronidazole vs. piperacillin–tazobactam for complicated intra-abdominal infections. Surg Infect 2008;9:325.