Critical Care Medicine

Intra-abdominal infection

Intra-abdominal infection

Synonyms

Peritonitis, intra-abdominal abscess, intra-abdominal sepsis, complicated- and uncomplicated intra-abdominal infection

Related conditions

Cholecystitis, pancreatitis, appendicitis, diverticulitis, colitis, perforated viscous, bowel obstruction, clostiridium difficile colitis, abdominal compartment syndrome, hepatic abscess, mesenteric ischemia, postoperative peritonitis, primary peritonitis, secondary peritonitis, tertiary peritonitis, peritoneal dialysis-associated peritonitis, spontaneous bacterial peritonitis

1. Description of the problem

What every clinician needs to know

Intra-abdominal infection is fairly common in both the community and hospitalized settings, carries significant associated morbidity and mortality, and is second only to respiratory tract infection as a cause of sepsis, severe sepsis and septic shock. Intra-abdominal infection encompasses a diverse spectrum of processes (sterile inflammation, obstruction, perforation, necrosis, primary contamination from trauma or surgery, failed immune system or recent antimicrobial therapy) at a broad range of sites (peritoneal space, retroperitoneum, any abdominal viscous). High risk populations include the elderly, prior abdominal surgery and trauma patients, chronic abdominal inflammatory conditions , and immunocompromised hosts.

Clinical features

Clinical features of intra-abdominal infection include fever; tachycardia; tachypnea; hypotension; local, referred, generalized or absent abdominal pain; anorexia; nausea and vomiting; diarrhea; abdominal fullness; distension; obstipation; shock; acidosis; and extra-abdominal organ failure.

Presentations are most often acute, although subacute progression over days to weeks may be observed. Specific areas of abdominal discomfort may be suggestive of certain foci of primary inflammation or infection (but not diagnostic). For example:

  • Right upper quadrant pain (cholecystitis, cholangitis, hepatic abscess).

  • Right lower quadrant (appendicitis and peri-appendiceal abscess).

  • Left lower quadrant (diverticulitis).

  • Epigastric pain radiating to the back (posterior penetrating ulcer, pancreatitis, costoverteral-flank pain (pyelonephritis, perinephric abscess).

Key management points

Any idealized management scheme has to be senstive to the level of patient acuity, rapidity of the process and estimation of the patient's tolerance for any delay in more invasive definitive management (i.e. percutaneous drainage/decompression or exploratory laparotomy) for diagnosis or therapeutic intent. Thus concurrent diagnostic, stabilization and source control are indicated in sicker patients.

  1. Stabilize the patient (see "Emergency management" section below). Urgent surgical intervention is indicated in diffuse peritonitis and should be considered as an early stabilizing procedure, even with ongoing resuscitation.

  2. Cognitive recognition (i.e. consider the diagnosis of intra-abdominal infection). This is easier for isolated typical abdominal signs and symptoms but more difficult when signs and symptoms are atypical, muted or absent due to an altered mental status, analgesics, corticosteroid or other immunosuppression, chronic liver disease, or a dominant concurrent extraabdominal condition (respiratory failure, myocardial infarction, stroke, diabetic ketoacidosis) . The diagnosis of abdominal compartment syndrome should be considered in patients with tense abdominal distension with monitoring of bladder pressure and either operative or non-operative decompression.

  3. Prompt patient assessment. Thorough intake of relevant history, risk factors, physical findings.

  4. Diagnostic laboratory studies. Generally all patients require a complete blood count and differential, chemistry, blood culture. Some studies are ordered based on pre-test clinical suspicion. For example right upper quadrant pain and other symptoms suggestive of hepatobiliary infection should prompt obtaining liver-associated enzymes, lipase/amylase determination. If the patient has clinically evident ascites, a paracentesis performed under ultrasound guidance for a cell count, differential, gram's stain and culture

  5. Diagnostic radiographic studies.Plain abdominal films may demonstrate "free air," an abnormal gas pattern indicative of partial or complete obstruction, enlarged colon luminal diameter, thickened loops or bowel, stones, or vascular calcifications suggestive of mesenteric ischemia. A normal or non-specific finding on an abdominal film may occur in peritonitis, perforated viscous, intra-abdominal abscess, or obstructed viscus. Abdominal CT imaging provides vastly superior data with higher sensitivity and specificity.

  6. Surgical consultation.

  7. Empiric antimicrobial therapy tailored to likely site and microbial etiology(s). Recent antimicrobial exposure, duration of hospitalization of more than 5 days and severity of illness are three risk factors associated with high incidence of multidrug-resistant pathogens which should be assessed to enhance the accuracy of empiric therapy. When definitive cultures and susceptibilities, antimicrobials should be subsequently narrowed if possible. The most common targets include enteric facultative gram negative baccilli (e.g. E. coli, Klebsiella, Enterobacter, Serratia, Proteus), enterococci, and obligate anaerobes (Bacteroides fragilis and non-fragilis spp., Peptostreptococci, Clostridium spp., Fusobacteria spp.).

  8. Source control.Most critical intervention for favorable outcome. Aims are to control, reduce or eliminate the focus of infection, which can include drainage of infected-contaminated fluid, debridement of necrotic or devitalized tissue and definitive management of precipitating condition. Patients with diffuse peritonitis require immediate surgery even if other diagnostic and stabilization interventions are ongoing. Two- or multistage interventions should be employed in very unstable patients who are poor surgical and healing candidates (i.e. intra-abdominal hypertension), with damage control initially followed by definitive correction.

2. Emergency Management

Stabilizing the patient

Septic manifesations, if present, need to be managed early and aggressively to reduce the risk of cumulative organ failure(s) and to optimize the patient for source control interventions.

  1. Transfer to an intensive care unit for appropriate nursing care, monitoring and resuscitation in rapidly deteriorating patients.

  2. Placement of large bore peripheral intravenous catheter(s) or central venous access, arterial line and Foley catheter. Nasograstric intubation may be indicated for nausea, vomiting and distension.

  3. Hemodynamic monitoring and support. The vast majority of patients who exhibit either SIRS, severe sepsis (organ hypoperfusion) or shock will require 2-4+ liters of crystalloid fluid support best titrated to urine flow rate, correction of acidosis, heart rate and blood pressure, and central venous pressure greater than 12-15 mm Hg (see Surviving Sepsis Guidelines in Dellinger RP reference below). Pressor support titrated to a mean arterial blood pressure 60-65 mm Hg. Central venous access both for volume and pressor infusion and CVP monitoring is the optimal standard; however, if central access cannot be quickly achieved, fluid resuscitation should not be delayed and can be given by large bore peripheral access. Valid measurable resuscitation endpoints include mixed venous saturation of more than 65%, serum lactate and correction of base deficit.

  4. Closely monitor respiratory status including respiratory rate, use of accessory muscles, oxygen requirement and hypercapnia. Patients with progressive metabolic acidosis, underlying chronic pulmonary disease or ARDS in evolution may require early endotracheal intubation and mechanical ventilator support. A relative indication for intubation also includes transport to radiology for diagnostic studies or percutaneous drainage in the borderline patient.

  5. Surgical or interventional radiology consultation for source control plan and enactment.

  6. Antimicrobial therapy. Ideally initiated prior to drawing blood cultures.

  7. Optimize the patient for radiologic or surgical intervention. This can include correction of coagul0pathy, electrolye and acid-base abnormalities, significant anemia and fluid deficits, and address respiratory deterioration.

  8. Consider adjunctive therapies. Patients who have had a steroid - dependent condition (COPD, organ transplant recipients) should receive "preemptive" stress-dose steroids i.e. hydrocortisone 300 mg/day. Critical illness relative corticosteroid insufficiency (CIRCI) should be considered in any patient with refractory shock after adequate restoration of circulatory volume has been achieved.

Management points not to be missed

  1. Triage patient to appropriate hospital setting.

  2. Place appropriate monitoring devices.

  3. Establish laboratory baseline (CBC with differential, coagulation studies, serum lactate, ABG, blood cultures, dirarrhea.

  4. Hemodynamic and organ assessments and fluid-pressor resuscitation.

  5. Consider and evaluate for abdominal compartment syndrome with direct or indirect manometry. Clinically significant intraabdominal hypertension requires urgent non-operative decompression (ascites drainage) or laparotomy.

  6. Inadequate empiric antimicrobial therapy is associated with worse outcomes and is most often due to an underestimation of the antimicrobial resistance pattern of multidrug resistant pathogen(s). Common examples include coverage of multidrug resistant enterococci in hospital acquired peritonitis; coverage of Candida, coagulase-negative staphylococci and enterococci in tertiary peritonitis patients with recent heavy antimicrobial exposure; and appropriate coverage for Clostridium difficile in patients with colitis symptoms or radiographic features of colitis on plain film or CT imaging.

  7. Consideration of adjunctive treatment (corticosteroids, activated protein C).

  8. Appropriate imaging studies.

  9. Source control assessment and plan (surgical, radiologic consultation).

  10. Narrow antimicrobial spectrum if possible and determine treatment duration. Early inflammatory conditions are often sterile and do not require prolonged antimicrobial therapy (uncomplicated appendicitis, cholecystitis, pancreatitis).

3. Diagnosis

Diagnostic criteria and tests

Localized intraluminal inflammatory-infectious conditions (appendiicitis, cholesycstics, pancreatitis, diverticulitis, C. difficile colitis) may culminate in either peritonitis or abscess formation due to micro- or macro- perforation and spillage of enteric contents.

Peritonitis may be classified as primary, secondary or tertiary. Primary peritonitis states include CAPD peritonitis and spontaneous bacterial peritonitis in chronic liver disease or nephritic syndrome. Secondary peritonitis occurs from teh spillage of intraluminal contents from a perforated hollow viscous, usually due to trauma, devitalization or necrosis. Tertiary peritonitis occurs in patients with recent abdominal surgery(ies) and is characterized by diffuse infection with a poor inflammatory response due to compromised host defences.

Diagnosis of peritonitis includes classic abdominal symptoms (diffuse pain, guarding, rebound tenderness, rigidity, usually accompanied by "toxic" systemic signs (fever, tachycardia, tachypnea, leukocytosis or left shift SIRS). Laboratory, microbiologic and radiologic tests are complementary in value to complete the overall patient assessment.

There are specific diagnostic criteria for several categories of peritonitis. Spontaneous bacterial peritonitis is confirmed by the presence of more than 250 neutrophils/mm3 ascitic fluid (also termed neutrocytic ascites) with or without a positive ascites culture. Another variant is termed bacterascites (organisms seen on ascites Gram's stain).

CAPD associated peritonitis is present when there is a cell count greater than 100/mm3 with more than 50% neutrophils usually accompanied by cloudy fluid return and monomicrobial growth of either coagulase-negative staphylococci or S. aureus in more than 90% of cases. It is critical to keep in mind that secondary peritonitis due to occult perforation may occur either in liver failure or in CAPD patients. A polymicrobial Gram's stain or growth of any anaerobic organism is highly suggestive of a secondary peritonitis process.

Normal lab values and abnormalities

Laboratory abnormalities are common in intra-abdominal infection. However, all should be considered non-specific and secondary in value to the clinical assessment.

Hematologic - leukocytosis and/or a left shift. Overwhelming sepsis can present with leukopenia. In the absence of concurrent hemorrhage hemoconcentration (rising hematocrit) may be seen due to third space fluid losses.

Liver enzymes - cholestatic pattern (hyperbilirubinemia ± elevated alkaline phosphatase) may be observed in sepsis or due to a local hepatobiliary infection (cholangitis, calculous or acalculous cholecystitis, pancreatitis). Elevated transaminases is most suggestive of "shock liver."

Microbiologic - the diagnosis of intra-abdominal infection-sepsis does not require a "positive culture." Properly obtained specimens of ascites should normally be sterile. Specimens which have visible organisms on Gram's stain but don't grow in culture may occur due to suppressive effect of antibiotics, non-viable organisms, or fastidious, slow growing species (i.e. anaerobes). The overall incidence of bacteremia is low (<5%) but does rise with increasing illness severity.

Plain films - free abdominal air on an upright chest film in a patient who isn't early post-laparotomy is almost always significant and an early prompt for surgical exploration (Figure 1). Low sensitivity limits the value of plain films for both localized abscess and peritonitis. Common abdominal radiographic findings which may accompany intra-abdominal infection (but are not pathognomonic for infection) include obstructive gas pattern (Figure 2), colonic dilation and thickening in colitis (Figure 3). A general rule is that if clinical suspicion is high, evaluation should not end with an abdominal film only.

Figure 1.

Subdiaphagmatic free air

Figure 2.

Distal small bowel obstruction

Figure 3.

Coloic dilatation and mucosal thickening in patient with C. dificile colitis

CT scan - contrast enhanced, helical methods offer the best sensitivity and specificity, are less operator-dependent than ultrasound, and may be naturally coupled to determining if there is a "window" to perform percutaneous drainage. Figure 4 shows a large amount of anterior free air in a patient with a perforated gastric ulcer and early peritonitis. Figure 5shows a large, anterior, homogeneous, low attenuation fluid collection with a contrast enhancing capsule (abscess) in a postoperative patient who subsequently underwent percutaneous drainage.

Free intraperitoneal air in septic patient with perforated gastric ulcer

Large anterior rim -enhancing abscess in postoperative patient

Figure 6 demonstrates a peri-appendiceal abscess prior to appendectomy and drainage. Sometimes a CT finding reveals a secondary manifestation of intra-abdominal infection but doesn't define the primary source such as portal venous gas (Figure 7).

Figure 6.

Peri-appendiceal abscess (arrow)

Figure 7.

Gas in the portal venous system of the liver

Ultrasonography - diagnostic modality of choice for gallbladder, biliary and renal obstructive processes. When avoidance of ionizing radiation is a priority (pregnancy), ultrasonography may be attempted as the first line diagnostic modality.

Establishing the diagnosis

The weight of diagnostic certainty has to rest primarily on the patient history and physical findings. A period of serial observations in a relatively stable patient may be indicated to rule out other diagnoses in the differential. Other than direct patient assessment, findings which provide the most diagnostic certainty include the presence of free intra-abdominal air on imaging indicative of visceral perforation, rim-enhancing collections which may or may not contain gas and laparotomy findings of peritoneal inflammation or focal intra-abdominal abscess. Percutaneous drainage of purulent material is also reliably predictive that an intra-abdominal infection is present.

Other possible diagnoses

Confounding conditions which may masquerade as serious abdominal infection include conditions where abdominal pain is a prominent early feature. These include obstipation or ileus causing luminal distension, uncomplicated bacterial, viral or parasitic gastroenteritis, mesenteric ischemia, exacerbations of inflammatory bowel disease, non-perforated gastroduodenal ulcer disease, simple pancreatic phlegmon, abdominal dermatomal Herpes zoster prior to the appearance of a characteristic rash, ileus, incarcerated hernia, other mechanical bowel conditions such as volvulus, intussuception, diabetic metabolic disorders (diabetic ketoacidosis, acute porphyria), lumbosacral disk disease, pleuropulmonary infection contiguous with the diaphragm, splenic infarction, abdominal aortic aneurysm, and occult tumors causing obstruction or carcinomatosis.

However, it is important to recognize that some of these conditions may progress to true intra-abdominal infection if the anatomic integrity of the intestines or other viscous structures is breached and that some of these conditions may co-exist with an established or evolving infection.

Confirmatory tests

As a single study, CT imaging of the abdomen has a very high positive predictive and negative predictive value for both true infection and the presence of many of the confounding conditions listed above. Utilization of oral and intravenous contrast is the optimal technique to demarcate parenchymal or peritoneal abscesses and/or devitalized necrotic tissue. A major exception is the superiority of ultrasonography to define the gall bladder dimension, wall thickness, gas, and pericholecystic fluid in suspected cholecystitis and biliary tract dilatation in suspected choangitis.

4. Specific Treatment

Excluding systemic supportive measures outlined above, the core of specific treatment for abdominal infection is empiric and definitive antimicrobial therapy, and, if indicated, source control. Discussion of treatment is best sequenced along the paradigm of community acquired peritonitis, hospital acquired peritonitis including postoperative and tertiary peritonitis, special peritonitis states such as CAPD and spontaneous peritonitis, abscess, and other focal processes such as cholangitis, diverticulitis, etc.

With the exception of CAPD peritonitis, regimens that provide broad polymicrobial coverage are indicated. Aminoglycosides are no longer considered first line treatment due to diminished efficacy in abdominal sites and high toxicity. Empiric guidelines and suggestions are shown below. Tailoring of the empiric regimen is usually based upon the results of peri- and intra-operative cultures, remembering that some organisms may not grow in culture even when present in vivo.

Antimicrobial therapy

Community-acquired peritonitis and intra-abdominal abscess:

  1. Majority of randomized controlled trials comparing well chosen single or combination antimicrobial regimens in community acquired peritonitis have shown theapeutic equivalence with respect to time-to-improvement, clinical and bacteriologic cure, superinfection and relapse rates.

  2. Polymicrobial infection is the norm in secondary peritonitis and intra-abdominal abscess, with facultative enteric gram negative bacilli and anaerobes comprising the most important pathogens.

  3. Recent healthcare exposure (antimicrobials, hospitalization, nursing homes) may lead to carriage and infection with multidrug resistant strains, even in community settings.

  4. Enterococcal coverage is usually not needed in community acquired settings.

Reasonable single agent regimens include a beta-lactam-beta lactamase inhibitor combination (ampicillin-sulbactam, piperacillin-tazobactam, ticarcillin-clavulinic acid), carbapenem, cefotetan, moxifloxacin or tigecycline. Alternatively, a combination system of a cephalosporin or aztreonam or a fluoroquinolone + metronidazole also provide adequate coverage.

Hospital, healthcare-associated intra-abdominal infection:

  1. The adequacy of empiric antimicrobial coverage prior to the availability of cultures and susceptibilities has been associated with clinical outcome in postoperative peritonitis. Important datapoints to enhance the accuracy of empiric antibiotic selection include antimicrobial exposure, known colonization or prior infection with multidrug resistant species (ESBL strains of Klebsiella, E. coli, Enterobacter spp., carbapenamase producing strains of Klebsiella, MRSA, VRE, and multi-drug resistant P. aerusinosa and A. baumanii)

  2. Breakthrough infection with susceptible or multidrug resistant enterococci (vancomycin-resistant E. faecium) has become much more prevalent in regimens that lack anti-enterococcal activity. This is especially common in solid organ abdominal recipients, hepatobiliary infection and settings with endemic VRE issues.

  3. Coverage of Candida species with fluconazole is indicated in high risk patients with tertiary peritonitis, surgically treated pancreatic infection, immunocompromised patients, protracted anti-bacterial coverage and isolation of Candida from operative cultures.

  4. Core antimicrobials (listed above) modified with the addition of agent with activity vs. multidrug resistant species e.g. VRE-linezolid, daptomycin, tigecycline.

Spontaneous bacterial peritonitis:

Efficacious agents include cefotaxime or other third generation cephalosporins, ampicillin-sulbactam or other beta lactam/beta-lactamase inhibitors, and fluoroquinolone. Anti-anaerobic activity not indicated.

CAPD peritonitis:

  • Therapy is tailored to the effluent centrifuged Gram's stain and final culture and susceptibilities.

  • Patients with known MRSA colonization require intraperitoneal vancomycin. Otherwise intraperitoneal ampicillin (for enterococcus) or 1st generation cephalosporin is most commonly used.

  • Recurrent cases are more likely to be drug-resistant (Pseudomonas, VRE).

  • Tunnel- or exit site infections and bacteremia require appropriate parenteral antimicrobials.

Clostridium difficile colitis:

First line management includes either oral metronidazole or oral vancomycin coupled with cessation of unecessary antibiotics. Patients with ileus or gastroparesis may require parenteral metronidazole. Toxic megacolon with or without systemic signs of sepsis or serious relapses may necessitate surgical consultation for total colectomy to prevent perforation and extra-colonic abdominal infection.

Source control

Two main approaches to achieve source of infection control are surgical and percutaneous drainage. There are clear indications for each approach but some specific infections might be managed with either approach each supported by some evidence (diverticular abscess, appendicitis). Smaller parenchymal abscesses (<2 cm) may respond only to a prolonged trial of antimicrobial therapy and not need formal source control.

Although controversial, some advocate initial management of appendicitis with only antimicrobial treatment and supportive therapy although there is a significant rate of recurrence.

Surgical source control is clearly indicated for those infections where percutaneous drainage would not achieve the desired result. These include:

  1. Diffuse peritonitis.

  2. Enteric leak requiring resection, patch and/or proximal diversion of the enteral stream.

  3. Necrotic, gangrenous organ or tissue.

  4. Decompression of abdominal compartment syndrome.

  5. Cases where percutaneous drainage has been ineffective or cannot be performed safely.

  6. Exploration where cause of abdominal sepsis is unknown.

Percutaneous drainagehas become the standard for management of most intra-abdominal abscesses previously managed by open surgical methods. Percutaneous drain(s) can be placed with either ultrasound or CT scan guidance depending on anatomic location. Case selection factors should contain at least one of the following:

  1. Single or multiple abscesses with a radiographic approach window unimpeded by bowel or vascular structures.

  2. Obstructed viscous (e.g. percutaneous transhepatic cholangiogram guided biliary drain with obstructed proximal biliary system).

  3. Patients who are unsuitably high risk for surgical intervention.

  4. Management of abdominal compartment syndrome if due to an easily drainable fluid (ascites).

Drugs and dosages

Table I. Common parenteral antimicrobials for intra-abdominal infection

Table I.

Common parenteral antimicrobials for intra-abdominal infection
Antimicrobial Dosing Comments
Ampicillin-sulbactam 1.5- 3 g q6h AR
Piperacillin-tazobactam 4.5 g q6- 8h AR
Ticarcillin-clavulanate 3.1 g q 4-6h AR
Cefazolin 500 mg - 1 g q 8h AR
Ceftriazone 1-2 g q 12h AR
Cefepime 2 g q8h AR
Cefotetan 2-3g q 12h AR
Meropenem 1 g q 6-8h AR
Imipenem-cilastatin 500 mg q6h AR
Ertapenem 1 g q 24h AR
Ciprofloxacin 400 mg q 12h AR
Moxifloxacin 400 mg q 24h
Aztreonam 1-2 g q 8h AR
Tigecuycline 50 mg q12h 75 mg loading dose
Metronidazole 500 mg q6h 500 mg q6-8h
Clindamycin 900 mg q8h
Vancomycin 1 g q 12h Trough target of 15-20 mg/LAR
Linezolid 600 mg q 12h
Daptomycin 6 mg/kg q 24 h AR
Colistin 1.5 mg/kg q 12h AR
Flluconazole 400-800 mg/d 1. 400 mg loading dose2.High dose in dose-dependent Candida spp.3. AR
Micafungin 100 mg q24h For use with suspected/documented azole resistant Candida

Duration of antimicrobial treatment

There is a paucity of data that define the optimal duration of therapy in intra-abdominal infection. However, consensus now favors shorter courses of treatment. Resolution of the original local and systemic signs and symptoms (fever, tachycardia, shock), normalization of leukocyte count and left shift, and clearance of the bloodstream in patients who were bacteremic at baseline, are core expectations for terminating treatment.

Consideration to short course treatment(<2-3days) for localized uncomplicated infection (appendicitis, cholecystitis, diverticulitis) in which source control is rapidly achieved is appropriate. Alternatively, extended courses of treatment are indicated when there is late or inadequate source control or for multidrug resistant organisms. The collateral benefits of shorter course therapy include decreased selection of drug-resistant strains, decreased fungal superinfection, and diminished costs and toxicity.

Refractory cases

In patients who do not respond to the initial therapeutic regimen, several important considerations should be entertained:

1. Does the patient have a continuing source of peritoneal contamination, retained infected foci, or obstructed viscous despite initially effective source control and well chosen antimicrobial therapy? This may require repeat imaging appropriate to the site, such as right upper quadrant ultrasonography to detect biliary tract obstruction or CT imaging to rule out retained foci. However, a caveat is that CT imaging in the recent post-laparotomy patient often demonstrates non-specific postoperative changes (free air, fluid, inflammatory changes) that make interpretation difficult.

2. Has the patient developed a new infectious complication either related to the initial invasive therapy, such as an incidental bowel perforation, or associated with critical illness?

3. Has a superinfection with a new organism(s) or a drug-resistant microbial strain developed that impairs effective micro-eradication?Breakthrough infection with either enterococci including vancomycin-resistant strains and Candida species is well described when the original antimicobial regimen has no activity against these organisms. At centers with endemic strains of multi-drug resistant gram negative organisms (MDR Acinetobacter baumanii and Pseudomonas aeruginosa, extended spectrum beta-lactamase strains of Klebsiella, E coli and Enterobacter, and most recently, carbapenase-producing Klebsiella) refractory or relapsing abdominal infection may occur at the original site of infection or at a non-contiguous site.

4. Does the patient have antimicrobial associated colitis? Clostridia difficile colitis is a separate category of superinfection exclusively related to recent use of almost any antimicrobial(s). Atypical presentations with ileus and no diarrhea accompanied by severe leucocytosis or even leukemoid reactions (> 30,000 wbc/ / cu mm) are more common in critically ill and postoperative patients. Recent hyper-virulent strains which elaborate greater quantities of cytotoxin (NAP-1 strains) and are refractory to metronidazole and enteral vancomycin have become much more prevalent over the past decade.

5. Is there a new extra-abdominal infection? Common examples include device-related infections (catheter-related bloodstream infection, urinary tract infection), ventilator-associated pneumonia and surgical wound infection.

5. Disease monitoring, follow-up and disposition

Expected response to treatment

The clinical response is highly dependent upon the baseline severity of illness (chronic and acute co-morbidity, APACHE score), type of infection (uncomplicated vs. complicated, localized abscess vs. diffuse peritonitis) and adequacy of source control and antimicrobial therapy. Less favorable prognoses, including diffuse peritonitis, tertiary peritonitis or retained-unresectable devitalized tissue are more likely to continue to exhibit persistent abdominal signs and symptoms and systemic sequelae of sepsis-infection (fever, tachycardia, acidosis, extra-abdominal organ failure).

Incorrect diagnosis

An incorrect diagnosis may be confirmed at the time of laparotomy (the "gold standard" for confirmation). For instance, a negative laparotomy rate of 15-25% is reported in patients with a pre-operative diagnosis of acute appendicitis. More liberal use of abdominal imaging, such as ultrasonography and CT, coupled with delayed intervention approaches have reduced this rate in recent years. Similarly, negative laparotomy rates are well described for pre-operative diagnoses such as cholecystitis and diverticulitis. It is critical to emphasize that negative laparotomy does not appear to contribute to excess mortality and surgical exploration may uncover unsuspected infectious processes.

Follow-up

In the early treatment period, the clinician should closely observe the patient for the persistence of signs of residual infection. Such signs can include fever trend, tachycardia, persistent fluid/pressor requirements, objective abdominal signs/symptoms exclusive of incisional pain, leukocyte count and differential, quality and quantity of wound or percutaneous drainage.

Early repeat CT radiologic studies, blood cultures and cultures through existing indwelling drains are rarely helpful and may offer only confusing data. However patients who undergo percutaneous drainage may need serial CT scan or ultrasounds to ensure there is not any significant residual undrained focus.

Pathophysiology

Infection may occur within any abdominal organ and extend to adjacent or distant spaces within the abdominal cavity as either formed abscess or peritonitis (see next paragraph). More localized but serious infection can occur within an organ and cause both localized and systemic symptoms. Common examples include compromised excretory outflow (cystic or biliary duct obstruction leading to cholecystitis or cholangitis, or ureteral obstruction or reflux resulting in "stasis, microbial overgrowth and upper urinary tract infection).

Dissemination of hematogenous organisms from a distant source can result in metastatic infection, usually to organs with high blood flow (liver, kidney, spleen), that may lead to multiple parenchymal micro-abscesses. Endovascular infection causing high grade bacteremia due to staphylococci, streptococci and enterococci is a common cause of such hematogenous dissemination.

Secondary peritonitis and intra-abdominal abscess are the most common processes leading to intra-abdominal sepsis. The inciting process begins as the micro- or macro-perforation of a hollow viscus due to inflammation, mechanical obstruction, ischemia or trauma, which leads to the spillage of either septic or aseptic (chemical) contents. The magnitude and location of the spillage, coupled with the adequacy of the host repsonse and iatrogenic intervention, influence whether the process localizes to a discrete abscess or spreads to become a diffuse peritonitis (Figure 1).

Early tissue response includes the release of vasoactive humoral mediators (histamine, kinins, leukotriene, prostacyclines) that increase vascular and peritonieal surface permeability, allowing influx of complement, opsonins and coagulation factors. Activated complement is chemotactic and promotes the egress of neutrophils which act to promote further inflammation with cytokine (IL-1,IL-6, tumor necrosis factor) release, engulfment and destruction of bacteria (phagocytosis).

Tissue factor release also activates the coagulation cascade, leading to fibrin formation and promoting local containment. This host response may be sufficient to reduce the local inoculum of viable microorganisms and create a local adherence boundaried by bowel, omentum, mesentery or abdominal wall.

The amplification of inflammatory mediators at the site of abdominal tissue injury are absorbed into the peripheral circulation or lymphatic drainage portals culminating in collateral systemic effects. These include vascular smooth muscle relaxation, vascular hyper-permeability, systemic activation of the coagulation cascade and synthesis of acute phase proteins, which account for

the SIRS response, severe sepsis and septic shock manifestations.

Epidemiology

NA

Prognosis

Prognosis can be measured as time to recovery, incidence of complications including organ failure and surgical site infection, length of stay, need for repeat laparotomy, and, of course, in-hospital mortality. Mortality rates can be characterized as low (below 5%), moderate (5-15%) and high (15-30% or higher).

Factors prognostic of mortality include:

  1. Extent of infection at presentation - localized controlled infection such as perforated appendicitis carries a better prognosis than diffuse peritonitis.

  2. Severity of illness - APACHE II score greater than 15 is the most validated scoring system predicting mortality.

  3. Ineffective antimicrobial therapy.

  4. Inability to achieve source control.

  5. Other mortality risk factors include hypoalbuminemia, elderly, previous antimicrobial therapy, prolonged hospitalization.

Special considerations for nursing and allied health professionals.

NA

What's the evidence?

Description of the problem

Roehrborn, A, Thomas, L, Potreck, O. "The microbiology of postoperative peritonitis". Clin Infect Dis. vol. 33. 2001. pp. 1513-9.

(Prospective study of hospital acquired postoperative peritonitis vs. community acquired peritonitis over a 6-year period which demonstrates a higher incidence of enterococci and Enterobacter species in the former and its treatment implications.)

Mazuski, JE, Solomkin, JS. "Intra-abdominal infections". Surg Clin N Amer. vol. 89. 2009. pp. 421-37.

(Broad review of etiology, presentation, diagnosis and management of peritonitis and intraabdominal abscess.)

Lopez, N, Kobayashi, L, Coimbra. "A comprehensive review of abdominal infections". World J Emerg Surg. vol. 6. 2011. pp. 7.

(Another contemporaneous review of intraabdominal infection diagnosis and management.)

Solomkin, JS, Maszuski, JE, Bradley, JS. "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". Surg Infect. vol. 11. 2010. pp. 79-109.

(Contemporary evidence based recommendations from SIS and IDSA on all aspects of intra-abdominal infection.)

Nathens, AB, Rotstein, OD, Marshall, JC. "Tertiary peritoitis: clinical features of a complex nosocomial infection". World J Surg. vol. 22. 1998. pp. 158-63.

(Retrospective review of a subset of 44 patients with tertiary peritonitis requiring ICU admissions demonstrating the unique demographics, co-morbidities, microbiology and outcome compared to non-tertiary peritonitis.)

Emergency management

Dellinger, RP, Levy, MM, Carlet, JM. "Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock 2008". Crit Care Med. vol. 36. 2008. pp. 296-327.

(Comprehensive multi-society endorsed evidence based recommendations of all clinical aspects of severe sepsis and septic shock.)

Vincent, JL, Gerlach, H. "Fluid resuscitation in severe sepsis and septic shock: an evidence-based review". Crit Care Med. vol. 32. 2004. pp. S451-64.

(Evidence based review of literature relevant to the fluid quantity and composition for the resuscitation of septic shock by 11 experts representing 11 professional societies.)

Diagnosis

Solomkin, JS, Maszuski, JE, Bradley, JS. "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". Surg Infect. vol. 11. 2010. pp. 79-109.

(See previous Solomkin et al reference above.)

Specific treatment

Solomkin, JS, Maszuski, JE, Baron, EJ. "Guidelines for the selection of anti-infective agents for complicated intra-abdominal infections". Clin Infect Dis. vol. 37. 2003. pp. 997-1005.

(Evidence based , graded, recommendations for the empiric and definitive antimicrobial treatment for community and hospital acquired intraabdominal infection by experts from four professional societies.)

Christou, NV, Turgeon, P, Wassef, R. "Management of intra-abdominal infections. The case for intraoperative cultures and comprehensive broad spectrum antibiotic coverage. The Canadian Intra-abdominal infection study group". Arch Surg. vol. 131. 1996. pp. 1193-201.

(Interesting study that highlights the improvement in treatment accuracy and outcome benefit when guided by obtaining intraabdominal cultures at the time of repeat laparotomy.)

Chow, AW, Evans, GA, Nathens, AB. "Canadian practice guidelines for surgical intra-abdominal infections". Can J Infect Dis. vol. 21. 2010. pp. 11-37.

(Comprehensive updated evidence-weighted recommendations for surgical intraabdominal infection from multiple Canadian medical and surgical societies.)

Krobot, K, Yin, D, Zhand, Q. "Effect of inappropriate initial empiric antibiotic therapy on outcome of patients with community-acquired intra-abdominal infections requiring surgery". Eur J Clin Microbiol Infect Dis. vol. 23. 2004. pp. 682-7.

(German study of 425 patients with community acquired secondary peritonitis demonstrating greater clinical success in those patients begun on appropriate empiric antimicrobial therapy.)

Cinat, ME, Wilson, SE, Din, AM. "Determinants for successful image-guided drainage of intra-abdominal abscess". Arch Surg. vol. 137. 2002. pp. 845-9.

(Study of 96 patients with intraabdominal abscesses managed with initial CT or US guided percutaneous drainage demonstrating that postoperative etiology and lack of a yeast isolate were strongest predictors of successful outcome.)

Walker, JJ, Criddle, LM. "Pathophysiology and management of abdominal compartment syndrome". Am J Crit Care. vol. 12. 2003. pp. 367-71.

(Very practical review of pathophysiology, diagnosis and treatment of elevated abdominal pressure.)

Montravers, P, Gauzit, R, Muller, C. "Emergence of antibiotic-resistant bacteria in cases of peritonitis after intra-abdominal surgery affects the efficacy of empirical antimicrobial therapy". Clin Infect Dis. vol. 23. 1996. pp. 486-94.

(Excellent prospective study of 100 patients with peritonitis that demonstrates the emergence of multidrug resistant organisms and the negative impact on outcome when such pathogens are not adequately treated at the time of re-exploration.)

Disease monitoring, followup and disposition

Pacelli, F, Doglietto, GB, Alfieri, S. "Prognosis in intra-abdominal infection: Multivariate analysis of 604 patients". Arch Surg. vol. 131. 1996. pp. 641-5.

(12-year large study of surgical intraabdominal infection with multivariate logistic regression demonstrating APACHE II score, the Mannheim peritonitis index, hypoalbuminemia, hypocholesterolemia and preoperative organ impairment were independent predictors of death.)

Kulkarni, SV, Naik, AS, Subramanian, N. "APACHE II scoring system in perforative peritonitis". Am J Surg. vol. 194. 2007. pp. 549-52.

(Prospective study of 50 patients with secondary peritonitis showing that an APACHE II score in the 10-20 range at the time of surgical intervention showed the best correlation between predicted and observed hospital mortality.)
You must be a registered member of Clinical Advisor to post a comment.
close

Next Article in Critical Care Medicine

Sign Up for Free e-newsletters