Surgical Peritonitis and Other Diseases of the Peritoneum, Mesentery, Omentum, and Diaphragm

Published on 16/05/2015 by admin

Filed under Gastroenterology and Hepatology

Last modified 16/05/2015

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 3575 times

CHAPTER 37 Surgical Peritonitis and Other Diseases of the Peritoneum, Mesentery, Omentum, and Diaphragm

Secondary peritonitis is often referred to as surgical peritonitis because the many and varied disease processes which present with peritonitis frequently require procedural intervention for treatment. Primary or spontaneous bacterial peritonitis is discussed in Chapter 91.

In addition, this chapter discusses the primary disease processes affecting the peritoneum, mesentery, omentum, and diaphragm. Primary disease processes of these structures are often diagnosed late due to the often nonspecific and vague symptoms related to them.



The peritoneum is a membrane covered by a single sheet of mesothelial cells, with an estimated area of 1.7 m2, similar to the total body surface area. The structure of the peritoneum is sealed in men and open to the exterior via the ostia of fallopian tubes in women. Usually the peritoneal space contains a few milliliters of sterile peritoneal fluid that may act as part of the local defense against bacteria, as well as a lubricant.

The peritoneum is divided into parietal and visceral components. The parietal peritoneum covers the anterior, lateral, and posterior abdominal walls; the inferior surface of the diaphragm; and the pelvis. A large portion of the surface of the intraperitoneal organs (stomach, jejunum, ileum, transverse colon, liver, and spleen) is covered by visceral peritoneum, whereas only the anterior aspect of the retroperitoneal organs (duodenum, left and right colon, pancreas, kidneys, and adrenals) is covered by visceral peritoneum. The intraperitoneal organs are suspended by thickened bands of peritoneum, or abdominal ligaments. The nine ligaments and two mesenteries identified by Meyers and colleagues are the coronary, gastrohepatic, hepatoduodenal, falciform, gastrocolic, duodenocolic, gastrosplenic, splenorenal, and phrenicocolic ligaments, the transverse mesocolon, and the small bowel mesentery.1 These ligamentous structures, which are apparent at laparotomy, as well as on computed tomography (CT), subdivide the abdomen into interconnected compartments. Familiarity with the anatomy can be used to predict the route of spread of disease; for example, the gastrohepatic and gastrocolic ligaments allow a gastric tumor to spread to the liver and colon. The spread of infection within the peritoneal cavity is governed by the site of infection, the sites of fibrinous and fibrous adhesions, intraperitoneal pressure gradients, and the position of the patient. After leakage of visceral contents, dependent recesses (e.g., paracolic gutters, pelvis, lesser sac, and subhepatic and subphrenic spaces) tend to become sites of abscess formation. For instance, patients with perforated peptic ulcer disease may present with right lower quadrant pain secondary to the dependent nature of the right lower quadrant and the right paracolic gutter. A common practice before modern imaging and percutaneous drainage methods was to place the patient in a semirecumbent position (Fowler’s position) to encourage pooling of contaminated fluids within the pelvis, in order to palpate the resultant abscess and drain it through the rectum.

The mesentery is defined as a membranous bilayer of peritoneum that attaches an organ to the body wall. An omentum is a fold of peritoneum that connects the stomach with adjacent organs of the peritoneal cavity. The greater omentum spreads from the greater curvature of the stomach to the transverse colon. The lesser omentum, which joins the lesser curvature of the stomach to the liver, is called the gastrohepatic omentum. The right edge of the lesser omentum is the hepatoduodenal ligament, and the opening posterior to this (the epiploic foramen of Winslow) is the only connection between the greater and lesser peritoneal sacs.


The visceral peritoneum is supplied by the splanchnic blood vessels, and the parietal peritoneum by intercostal, subcostal, lumbar, and iliac vessels. The visceral peritoneum is supplied by nonsomatic nerves, whereas the parietal peritoneum is supplied by somatic nerves. Therefore, visceral pain is poorly localized, diffuse, and vague (see Chapter 10). Visceral pain is caused by stretching, distention, torsion, and twisting. The visceral peritoneum does not produce pain when it is cut or burned. When visceral pain fibers of midgut structures are stimulated, a vague periumbilical discomfort results because the visceral pain fibers enter the spinal cord at the same level as the T10 dermatome somatic fibers (see Chapters 10 and 11). This sensation is, therefore, experienced as discomfort in the dermatomal distribution. Likewise, visceral stimulation from foregut structures produces epigastric (T8 distribution) discomfort, and visceral stimulation in the hindgut produces suprapubic (T12) discomfort. Parietal (somatic) pain fibers are activated by such stimuli as cutting, burning, and inflammation. This type of pain is sharply localized. A good example of this process is appendicitis. Early in the disease process the patient experiences periumbilical discomfort secondary to distention of the appendiceal lumen, and this progresses to localized right lower quadrant pain and tenderness as the inflammation becomes transmural and stimulates the parietal peritoneum.


Particles, solutes, and fluids are absorbed from the peritoneal cavity by two different routes. Substances smaller than 2 kd may be absorbed through peritoneal mesothelial venous pores and are directed to the portal circulation.3 Particles larger than 3 kd are absorbed through peritoneal mesothelial lymphatics, entering the lymphatic thoracic duct and from there the systemic circulation.4 This last route of absorption plays an important role in controlling abdominal infections because it has a huge capacity for absorption. The anatomic structure of these large channels between the peritoneal cavity and the diaphragmatic vessels and the negative pressure of the thorax during inspiration make this mechanism extremely effective in the removal of bacteria and cells. The large surface area and semipermeability of the peritoneal membrane can be exploited therapeutically in peritoneal dialysis of patients with kidney failure and in rewarming hypothermic patients with peritoneal lavage.


Secondary (surgical) peritonitis is a result of an inflammatory process in the peritoneal cavity secondary to inflammation, perforation, or gangrene of an intra-abdominal or retroperitoneal structure. Surgical intervention is typically required to treat these processes, although antibiotics often are useful while the process resolves (e.g., uncomplicated diverticulitis). If untreated, secondary peritonitis will, in most cases, lead to septic shock and death.


Secondary peritonitis has numerous causes. The diagnosis is based on history, physical examination, radiographic studies, and operative exploration. History and physical examination are very important in secondary peritonitis, and a good history and physical examination can often obviate further studies. Some of the more common causes of secondary peritonitis include perforated peptic ulcer disease, appendicitis, diverticulitis, acute cholecystitis, and postsurgical complications.

Other nonbacterial causes of peritonitis include leakage of blood into the peritoneal cavity due to rupture of a tubal pregnancy, ovarian cyst, or aneurysmal vessel. Blood is highly irritating to the peritoneum and may cause abdominal pain similar to that found in septic peritonitis. Bile leakage into the peritoneal cavity also can cause signs and symptoms of peritonitis, especially when there is also bacterial contamination of the bilious contents. However, pure bile in the abdomen can be surprisingly asymptomatic. Large bilomas may have minimal symptoms.

Bacteria reach the peritoneal cavity by a variety of pathologic processes: transmural inflammation with luminal obstruction (see Chapter 119), perforation of the gastrointestinal (GI) tract, and ischemia (see Chapter 114). The initial inoculum of bacteria is determined by the normal flora in the involved portion of the GI tract (see Chapter 102).


Although the flora of the gut, especially of the large bowel, is diverse and extensive, the numbers of types of organisms rapidly decrease after leakage of gut contents into the peritoneal cavity.5 Aerobes such as Escherichia coli and enterococci and anaerobes such as Bacteroides fragilis and Clostridium organisms predominate. A recent study of infections associated with ruptured colonic diverticulitis reported anaerobes only in 15% of cases, aerobic bacteria only in 11%, and mixed aerobic and anaerobic flora in 74%; cultures from peritoneal abscesses detected anaerobic bacteria in 18%, aerobes alone in 5%, and mixed aerobic and anaerobic flora in 77%.6 In addition to bacteria, the presence of fungi in intra-abdominal infection has recently been more frequently recognized and may have clinical significance. For instance, a positive fungal culture is quite common in perforated peptic ulcer disease and may adversely affect outcome.7

On the basis of an animal model of monomicrobial and polymicrobial peritonitis with various combinations of bacteria, it is apparent that (a) E. coli is the organism most often responsible for death from this form of iatrogenic peritonitis, at least in part because of its ability to cause bacteremia, and (b) that combinations of anaerobes and facultative organisms lead to abscess formation.8 As stated, 77% of bacterial cultures from peritoneal abscesses are polymicrobial.6 Other adjuvant substances, such as devitalized tissue, mucus, bile, hemoglobin, and barium, can act synergistically with microorganisms to increase mortality in surgical peritonitis through their ability to interfere with phagocytosis and killing of bacteria. These considerations form the basis for the treatment of surgical peritonitis, which is described later.

The peritoneal cavity possesses several lines of defense against bacterial infection (Table 37-1). Peritonitis results when these are overwhelmed.

Table 37-1 Peritoneal Defense Mechanisms Against Bacteria

Removal Mechanisms

Leukocyte-Attracting Mechanisms

Killing Mechanisms

Sequestration Mechanisms

ICAM, intercellular adhesion molecule; VCAM, vascular cell adhesion molecule.

Peritoneal Clearance of Bacteria

Once bacteria enter the peritoneal cavity, clearance of the offending microorganisms begins immediately. Within 6 minutes of intraperitoneal inoculation of bacteria in dogs, bacteria can be cultured in thoracic lymph, indicating passage of organisms through the diaphragm. Twelve minutes later, bacteremia may be evident. This clearance mechanism is probably important in survival because blockade of the thoracic duct in an animal model of peritonitis decreases bacteremia episodes4 but increases mortality and induces liver necrosis. This appears to be directly related to the amount of endotoxin to which the liver is exposed.9 Decades before it was known that the diaphragm was the predominant site of clearance of bacteria, Fowler, in 1900, proposed his head-up, pelvis-down position for prevention of absorption of toxins from infected peritoneal cavities. In the preantibiotic era, documentation of the delayed clearance of bacteria from experiments in infected dogs in the head-down position confirms the wisdom of this positioning for patients with peritonitis.

Killing Mechanisms

In addition to mechanisms of bacterial clearance through the diaphragm, intraperitoneal defense mechanisms include cellular and humoral responses (see Chapter 2). Macrophages and neutrophils are attracted to the peritoneal cavity, and in this setting, microvilli of the mesothelial cells play a significant role in leukocyte migration into the peritoneal cavity by providing the needed substrates for their adhesion, namely intercellular adhesion molecule-1 (ICAM-1, or CD 54), and vascular cell adhesion molecule-1 (VCAM-1, or CD 106).10

The degree of cellular recruitment may be a key factor in a patient’s survival because a prolonged peritoneal inflammatory response has been observed to be adversely correlated with survival in an animal model of peritonitis.11 Humoral antibacterial agents, such as complement factors, fibronectin, and globulins, are released into the peritoneal cavity. These opsonins coat bacteria and render them recognizable as foreign; then they are entrapped and killed by phagocytes.12

Sequestration Mechanisms

Sequestration mechanisms include fibrin trapping of bacteria, fibrinous adhesions, and omental loculation of foci of infection (see Table 37-1).13 It has been known since 1950 that bacteria are more readily destroyed on a surface than in a liquid medium. The microscopic and macroscopic networks of surfaces provided by fibrin and the omentum assist phagocytes in locating, trapping, ingesting, and killing bacteria. The volume of peritoneal fluid in which infection develops has a remarkable effect on mortality; 20% of rats inoculated with E. coli diluted in 1 mL of saline die, whereas 75% of rats inoculated with the same number of viable bacteria but diluted in 30 mL of saline die.14 This phenomenon explains in part the risk of development of spontaneous bacterial peritonitis in relation to the ascitic fluid total protein concentration.15 The more voluminous the ascitic fluid, the lower the concentration of proteins and opsonins, the less efficient the trapping of bacteria, and the higher the risk of an uncontrolled infection (see Chapter 91). Patients undergoing chronic ambulatory peritoneal dialysis may be vulnerable to peritonitis because of dilution of opsonins by dialysis fluids.

Bacterial contamination in the peritoneal cavity and the subsequent response of immune cells such as neutrophils and macrophages lead to an inflammatory response including the release of cytokines. The systemic inflammatory response syndrome (SIRS) is marked by fever, a hyperdynamic cardiovascular response, muscle protein breakdown,16 and respiratory failure. If the underlying cause is treated by surgical intervention, antibiotics, or the body’s own defense mechanisms, these processes can be thwarted or reversed. However, if the process goes unchecked, multisystem organ failure and death will result. In addition, even if the underlying cause is treated, the inflammatory response can lead to multisystem organ failure and death if the treatment is delayed or the inflammatory response is particularly vigorous.

Patients with severe peritonitis may have a higher mortality from a shift from type 1 to type 2 T-helper cells leading to greater immunosuppression.17 When treating peritonitis or operating within the abdomen, the clinician’s goal is to minimize or eliminate inflammation. For instance, laparoscopic operations may induce less of a systemic inflammatory response than their open counterparts.18,19 In addition, laparoscopy differs from laparotomy in regard to peritoneal macrophage response,20 less cortisol release,21 and less reduction in natural killer (NK) cell subsets.22 Laparoscopic operations may well confer an immunologic advantage over conventional open operations.23 The additional benefits of smaller incisions, less tissue trauma, decreased postoperative pain, and shorter recovery are driving a trend to laparoscopic operations over open operations even in acute settings.


Clinical history and careful physical examination are the key factors in making a timely diagnosis of surgical peritonitis. In general, the sooner the diagnosis is made, the better the prognosis. Abdominal pain is the hallmark of peritonitis. The exact details of the onset of pain can be helpful in drawing attention to the affected organ (see Chapter 10). The pain’s character, location, area of radiation, change over time, and provocative and palliative factors are key pieces of information in assisting with the diagnosis. Peritoneal inflammation is typically associated with ileus, and therefore nausea and vomiting are common symptoms.

The ability of the clinician to elicit an accurate history of abdominal pain and peritoneal signs is limited in patients with neurologic and immunologic compromise. The pain of peritonitis can be reduced or even absent in older adult patients. Infants and children may be incapable of furnishing any history or cooperating with the physical examination. Notoriously difficult patients to assess for secondary peritonitis include emergency room patients under the influence of alcohol or illicit drugs, trauma patients with central nervous system or spinal cord injuries, and sedated and ventilated intensive care unit (ICU) patients. Analgesics typically will not relieve the pain of peritonitis on examination but may relieve some discomfort as related to the history of present illness. In fact, it has been shown that early provision of analgesia to patients with undifferentiated abdominal pain does not affect diagnostic accuracy.24 Diabetic patients have deficits in neurologic and immune function. Patients receiving immunosuppressive and anti-inflammatory drugs, such as glucocorticoids and chemotherapeutic drugs, may have blunted perception of pain and minimal signs of peritoneal irritation. Patients with cirrhosis and ascites may show no pain during episodes of spontaneous bacterial peritonitis unless the parietal peritoneum becomes involved with the inflammatory process (see Chapter 91).

On examination, the patient with surgical peritonitis is usually immobile because any movement acutely worsens the pain. Fever of 100° F or higher is typical, as is tachycardia, which may be in part secondary to pain. Hypotension is usually a late finding accompanying sepsis. Fever is a basic endogenous mechanism to help fight infection. In fact, the increase in body temperature that is usually found during bacterial infections, including peritonitis, seems to be essential for optimal host defense against bacteria.25 The absence of percussible hepatic dullness suggests the presence of free air in the peritoneal cavity. Exquisite tenderness to percussion should lead to very gentle palpation. Overly vigorous palpation of a very tender abdomen may cause patients such pain that they are subsequently unable to cooperate for the remainder of the examination.

Palpation should begin farthest from the area that the patient identifies as the source of the most pain. Palpation of a truly boardlike abdomen is so impressive to the examiner that it cannot be forgotten. Lesser degrees of rigidity must be compared with this extreme end of the spectrum. Voluntary guarding in the presence of mild tenderness may be misinterpreted as rigidity by the inexperienced examiner if the patient is anxious and palpation too vigorous. It is usually not necessary to check for rebound tenderness to palpation if rebound tenderness is noted during auscultation or percussion. Often, the presence of rebound tenderness can be inferred if the patient’s pain is exacerbated when the bed or stretcher is jarred.

Peritoneal signs signify inflammation of the parietal peritoneum secondary to an intra-abdominal process. Peritoneal signs consist of rebound tenderness, involuntary guarding and extreme tenderness on palpation. Peritonitis can be diffuse, such as that associated with perforated ulcer, or localized, such as that of diverticulitis confined to the left lower quadrant. Significant septic processes may be confined to the pelvis by overlying bowel and omentum with a resulting absence of peritoneal signs in the anterior abdominal wall. Therefore, careful rectal and pelvic exams are essential in order to detect pelvic peritonitis. The presence of iliopsoas and obturator signs (described in Chapter 116) can be helpful in detecting retroperitoneal or pelvic inflammation and abscesses.

Repeated physical examinations by the same examiner will provide evidence of progressive peritoneal irritation. The evolution of the physical exam over time provides additional information for diagnosis and evaluation of response to initial conservative therapy. This, together with laboratory tests and imaging procedures described below, will indicate the need for surgical intervention.


The most common laboratory sign of peritonitis in an immunocompetent patient is an increased white blood cell count with left shift. The presence of circulating juvenile forms (e.g., bands) is a reflection of an increasing demand of white cells from the bone marrow. A low white blood cell count in the course of a bacterial infection associated at times with gram-negative septicemia may indicate the presence of an exhausted bone marrow, with a poorer prognosis. In addition, metabolic acidosis, hemoconcentration, and prerenal azotemia may be present.

Free air may be detected on upright chest radiograph or on upright or decubitus abdominal films, but this finding may be only 60% sensitive in detecting gut perforation.26 The absence of free air should not delay surgical intervention in an otherwise appropriate clinical setting. Ultrasonography can be helpful in demonstrating abscesses, bile duct dilatation, and large fluid collections. CT scan of the abdomen and pelvis, generally with both oral (occasionally rectal) and intravenous contrast, is increasingly preferred as the most sensitive and specific imaging modality for acute abdominal pain. Multidetector CT scanners are capable of imaging the entire abdomen and pelvis in a single breath-hold. The axial images are of extremely high resolution and can be reconstructed in coronal, sagittal, and three dimensional sets of images.27 CT is much more sensitive than plain films for the detection of free air, and with multidetector CT it is possible to visualize the actual site of perforation.28 Although CT images are increasingly accurate and the images compelling, they should not delay surgical consultation, resuscitation, and operation in a patient with suspected peritonitis.


The following two principles in the management of surgical peritonitis cannot be overemphasized. First, not all patients with peritonitis require surgery. For example, a patient with localized left lower quadrant peritonitis secondary to diverticulitis can be managed with bowel rest and intravenous antibiotics alone. Another patient with the same clinical presentation and findings of a diverticular abscess on CT scan can be successfully treated with antibiotics and percutaneous drainage (see Chapter 26). Second, the absence of peritonitis does not exclude the possibility of surgical emergency. The classic example of this clinical situation is early acute mesenteric ischemia with abdominal pain out of proportion to findings on physical examination findings. Likewise, a complete mechanical small bowel obstruction without peritoneal signs, an indication of perforation or vascular compromise, still requires operation.

For most cases of secondary peritonitis fluid resuscitation and antibiotic therapy followed by urgent laparotomy or laparoscopy are the mainstays of treatment. Fluid resuscitation is guided by frequent monitoring of physiologic parameters in an ICU, including blood pressure (by arterial line if shock is present), heart rate, central venous pressure or pulmonary capillary wedge pressure, and urine output. Hematocrit, WBC, electrolytes, glucose, creatinine, and blood gases should also be monitored. Hypovolemia, hypotension, metabolic acidosis, hypoxia, and hemoconcentration from loss of plasma into the peritoneal cavity are expected. Glucocorticoids have been shown not to provide benefit in the setting of septic shock.31

The patient should be aggressively fluid resuscitated to treat intravascular fluid depletion secondary to movement of fluid out of the vascular space. Pressors are generally to be avoided, if possible, and surgical intervention should be pursued when indicated as soon as the patient is hemodynamically stable for operation.


Antibiotic therapy is required before, during, and after surgical intervention. The type of bacteria causing secondary peritonitis depends in part on the normal flora of the part of the GI tract that is the source of sepsis and in part on the clinical setting. In community-acquired peritonitis, susceptible gram-negative bacilli, strict anaerobic bacteria, and enterococci are typically found. In general, antibiotics directed against the most likely pathogens should be chosen. For instance, colonic processes require coverage for gram-negative aerobes and anaerobes. In animal models, antibiotics directed against gram-negative enteric aerobic organisms minimize mortality, and drugs effective against anaerobes prevent abscess formation.32 It has been shown in experimental models of peritonitis that there is synergism between aerobic and anaerobic bacteria.33 The coverage of all potential organisms is not necessary.34 The flora of surgical peritonitis simplifies with time, even before initiation of antibiotics. Killing certain key species may change the microenvironment sufficiently to prevent growth and allow killing of other flora. If a Candida species is cultured from the peritoneal cavity in a patient with secondary peritonitis, this organism should be treated if the patient is in septic shock or is immunocompromised despite being hemodynamically stable. Hemodynamically stable immunocompetent patients with secondary peritonitis do not need treatment for Candida.35

A variety of antibiotic regimens have been proposed using the following classes of antibiotics alone or in combination: second-generation cephalosporins, third-generation cephalosporins, broad-spectrum beta-lactams, fluoroquinolones and metronidazole, and aminoglycosides with clindamycin or metronidazole. Many controlled trials of antibiotic regimens show equivalency. For example, it has been shown that monotherapy with a broad-spectrum beta-lactam is as effective as combination therapy with a beta-lactam and an aminoglycoside.36

Data-supported guidelines regarding optimal treatment have been hampered by suboptimal study design and nonuniform efficacy criteria in the controlled trials that have been performed. A recent Cochrane review of 40 randomized trials involving 16 different regimens showed no difference in mortality.37 The specific antibiotics chosen should take into account other considerations such as the avoidance of toxicities, the sensitivity profile of cultured organisms, the ease and route of administration, and cost. The availability of broad-spectrum antibiotics, including beta-lactams, fluoroquinolones, and third- and fourth-generation cephalosporins, makes it unnecessary to use aminoglycosides with their potential nephrotoxicity in patients with compromised renal function.36

The failure to clear secondary peritonitis after an appropriate course of antibiotic therapy or the recurrence of peritonitis is termed tertiary peritonitis. Nosocomial infections occurring in patients after long periods of hospitalization may include infections with multiresistant Pseudomonas, Enterobacter, Enterococcus, Staphylococcus, and Candida species. The development of multiple organ dysfunction syndrome (MODS) after an initial operation should prompt an aggressive search for inadequate source control and abscesses, involving repeat CT scans, percutaneous or operative drainage, and culture of persistent fluid collections, in addition to antimicrobial therapy.38