Peritoneal infections are classified as primary, secondary, or tertiary. Primary peritoneal infections arise from hematogenous dissemination, usually in the setting of an immunocompromised state, while secondary infections are related to a pathologic process in a visceral organ, such as perforation, trauma, or a postoperative problem. Tertiary peritoneal infection is a persistent or recurrent infection that exists after an adequate initial therapy.
Antibiotic therapy is used to prevent local and hematogenous spread of an intra-abdominal infection and to reduce late complications.  Several different antibiotic regimens are available for the treatment of intra-abdominal infections (see Table 1 below). 
Single-agent, broad-spectrum therapy and combination therapies have been used against these infections, although no specific therapy has been found to be superior to another.
Infection of the abdominal cavity requires coverage for gram-positive and gram-negative bacteria, as well as for anaerobes. Antipseudomonal coverage is recommended for patients who have had previous treatment with antibiotics or who have had a prolonged hospitalization. 
Usually, in patients with intra-abdominal infection who have been treated with proper source control and prompt surgical intervention, antibacterial therapy is given for 5-7 days, but this regimen may need to be extended, depending on the clinical situation.  Shorter courses have been used successfully.
Antibiotics can be discontinued once clinical signs of infection have resolved. Recurrence is a concern with certain infections, such as those from Candida and Staphylococcus aureus, and treatment should be continued for 2-3 weeks.
Table 1. Proposed Empirical Antimicrobial Therapy (Open Table in a new window)
|Amoxicillin/clavulanic acid||Cefuroxime + metronidazole|
|Piperacillin/tazobactam||Third- or fourth-generation cephalosporin + metronidazole|
|Ertapenem||Ciprofloxacin + metronidazole|
|Aminoglycoside + clindamycin|
|Tigecycline||Aztreonam + metronidazole|
In patients with community-acquired peritonitis, coverage for Enterococcus is not recommended. Enterococcal coverage may be warranted in patients with septic shock who have received prolonged cephalosporin therapy, in patients who are immunosuppressed and are at risk for bacteremia, in patients with prosthetic heart valves, and in patients with recurrent intra-abdominal infections accompanied by severe sepsis. 
Patients with intra-abdominal contamination are at a high risk for candidiasis, and this has led to the increased use of antifungal prophylaxis. Patients who are immunocompromised or who have received long-term, broad-spectrum antibiotic therapy or steroid therapy are predisposed to candidal infections.
Other predisposing factors include gastric acid suppressive therapy, central venous catheterization and intravenous hyperalimentation, malnutrition, and diabetes.
Infection in the Critically Ill
Candida albicans is most commonly isolated from the peritoneum in critically ill patients with culture-proven intra-abdominal infections and preoperative Acute Physiology and Chronic Health Evaluation II (APACHE II) scores of greater than or equal to 15.
Additional common peritoneal organisms in this patient population are Enterococcus and Enterobacter species and Staphylococcus epidermidis. These data suggest that the microbiology of intra-abdominal infections may be inherently different in severely ill patients and that broader antimicrobial, and possibly antifungal, coverage may be warranted in these cases.
Spontaneous Bacterial Peritonitis
Spontaneous bacterial peritonitis (SBP) resulting from chronic liver disease is the most common etiology of primary peritonitis. Untreated SBP has a mortality rate of up to 50%, but with prompt diagnosis and treatment of the condition, this figure may be reduced to 20%. Empiric therapy with a third-generation cephalosporin must be started promptly. 
The patient with SBP is also likely to require attention to changes in hemodynamic function related to inflammatory pathways, as well as resultant renal function impairment, although a discussion of this is beyond the scope of this article.
Cefotaxime is considered the treatment drug of choice. Initial coverage should include gram-negative enteric bacteria and gram-positive cocci, which are responsible for 90% of infections.  Cefotaxime is effective against 98% of causative organisms. Anaerobic, pseudomonal, and staphylococcal coverage is not needed.
Cefotaxime (2 g IV q8h) has been shown to achieve excellent ascitic fluid levels. The dosing interval may need to be reduced in patients with renal insufficiency.
Amoxicillin-clavulanic acid has been shown to be as effective as cefotaxime; however, a parenteral formulation is not available in the United States. 
Oral ofloxacin has been reported to be as effective as cefotaxime in the treatment of SBP. Ofloxacin should not be given to patients who are vomiting, in shock, bleeding, or in renal failure.
Alternatively, intravenous ciprofloxacin (200 mg q12h for 2 d), followed by oral ciprofloxacin (500 mg q12h for 5 d), has been used successfully. To prevent fluoroquinolone resistance, these antibiotics should not be used empirically to treat SBP.
When cultures identify a particular pathogen, susceptibility testing allows the clinician to narrow the spectrum of the antibiotic.
Duration of therapy
The optimal duration of therapy is not known. In patients without shock, ileus, hepatic coma, and/or renal failure, SBP usually resolves within 2-5 days of starting cefotaxime therapy. Traditionally, a course of 10 days is recommended, although studies have suggested that 5 days of therapy (with documentation of a decrease of peritoneal fluid WBC count to < 250 cells/μL) may be sufficient in most cases.
Avoid aminoglycosides in patients with liver disease, because these patients are at an increased risk for nephrotoxicity.
The risk of relapse after SBP is high (40-70% in 12 mo); various prophylactic antibiotic regimens are available. A preliminary study found that long-term norfloxacin (400 mg/d) was effective for secondary prevention of SBP. 
Secondary and Tertiary Peritonitis
In secondary and tertiary peritonitis, systemic antibiotic therapy is the second mainstay of treatment. [8, 9, 10] Several studies suggest that antibiotic therapy is not as effective in the later stages of infection and that early (preoperative) systemic antibiotic therapy can significantly reduce the concentration and growth rates of viable bacteria in the peritoneal fluid.
Antibiotic therapy begins with empiric coverage (effective against common gram-negative and anaerobic pathogens), which should be initiated as soon as possible, with a transition made to narrower-spectrum agents as culture results become available.
Specific conditions and effective agents
Perforations of upper GI tract organs are associated with gram-positive bacteria, whereas distal small bowel and colon perforations involve polymicrobial aerobic and anaerobic species.
Antibiotic therapy appears to be less effective in tertiary peritonitis than in other forms of peritonitis. Resistant and unusual organisms (eg, Enterococcus, Candida, Staphylococcus, Enterobacter, Pseudomonas species) are found in a significant proportion of cases.
Culture results may be especially important in tertiary peritonitis, which is more likely to involve gram-positive bacteria (enterococci); antibiotic-resistant, gram-negative bacteria; and yeast. In community-acquired infections, a second- or third-generation cephalosporin or a quinolone with or without metronidazole provides adequate coverage, as do broad-spectrum penicillins with anaerobic activity (ie, ampicillin/sulbactam) and newer quinolones (ie, trovafloxacin, clinafloxacin).
Most studies suggest that single-drug therapy is as effective as dual- or triple-combination therapy in mild to moderate abdominal infections.
For peritoneal dialysis–associated infections, a Cochrane review of all published randomized controlled trials did not find significant differences between antimicrobial agents or combinations, with similar response and relapse rates for glycopeptide regimens and first-generation cephalosporins.  Intraperitoneal antibiotics had a lower failure rate than intravenous regimens. Risk for early peritonitis is reduced with perioperative intravenous antibiotics; other prophylactic approaches are not yet proven.
In severe and hospital-acquired intra-abdominal infections, imipenem, piperacillin/tazobactam, and a combination of aminoglycosides and metronidazole are often effective.
A study of nearly 400 patients documented that ertapenem, a novel carbapenem with a half-life that allows once-a-day dosing, was more effective (86.7% success rate) than piperacillin/tazobactam (81.2% success rate) in the treatment of complicated intra-abdominal infection and was well tolerated. Additional clinical antimicrobial studies are underway to investigate the efficacy of new quinolones in the treatment of intra-abdominal infection.
Considerations for spectrum testing
With persistence of the infection (ie, tertiary peritonitis) and prolonged critical illness, obtaining peritoneal fluid and/or abscess cultures with sensitivities at operation or drainage is important to properly treat unusual (eg, gram-positive organisms, fungi) and resistant organisms (eg, Enterococcus, Staphylococcus, Pseudomonas, resistant Bacteroides, Candida species).
Certain preexisting conditions, immunocompromised state, gastric acid suppression therapy, and recent antibiotic use may also influence the spectrum of microorganisms. Consultation with infectious disease specialists is warranted in these cases.
Duration of therapy
The optimal duration of antibiotic therapy must be individualized and depends on the underlying pathology, severity of infection, speed and effectiveness of source control, and patient response to therapy. In uncomplicated peritonitis in which there is early, adequate source control, a course of 5-7 days of antibiotic therapy is adequate in most cases. Mild cases (eg, early appendicitis, cholecystitis) may not need more than 24-72 hours of postoperative therapy.
Inadequate initial therapy has been linked to worse outcomes, and these outcomes could not be significantly changed by later specific or prolonged therapy. Antimicrobial therapy should continue until signs of infection (eg, fever, leukocytosis) have resolved; if signs of infection continue, persistent infection or the presence of a nosocomial infection should be investigated.
Complicated persistent infections and infections in immunocompromised patients may warrant a prolonged course of antibiotic therapy. In these cases, continuously seeking and aggressively treating all new extraperitoneal and new or persistent intra-abdominal sources is important. The length of the individual course of treatment is variable and is often linked to signs of resolution of the inflammatory process (eg, lack of fever for >24-48 h, return of the white blood cell [WBC] count to the reference range levels).
Dangers of prolonged therapy
Some patients demonstrate persistent signs of inflammation without a defined infectious focus. In these patients, continued broad-spectrum antibiotic therapy may be more harmful than beneficial (eg, may promote emergence of resistant organisms or Clostridium difficile colitis), and a trial of antibiotic therapy cessation under close surveillance may be warranted.
Antibiotics alone are seldom sufficient to treat intra-abdominal abscesses, and adequate drainage of the abscess is of paramount importance. For most of the commonly used antibiotics, drug levels achieved in the abscess fluid are generally below the minimum inhibitory concentration–90 (MIC90) for Bacteroides fragilis and Escherichia coli, and repeated dosing or high-dose therapy does not improve penetration significantly.