Intra-abdominal abscesses

 

MARK S. PASTERNACK AND MORTON N. SWARTZ

 

 

PATHOGENESIS OF INTRA-ABDOMINAL ABSCESSES

Generalized bacterial peritonitis and focal intra-abdominal abscesses are extremes in the spectrum of intra-abdominal infection. Primary bacterial peritonitis (not secondary to a perforated viscus), particularly when due to a single species of aerobic or facultative bacteria, can generally be treated successfully with medical therapy alone. In contrast, the intra-abdominal abscesses require drainage in addition to antibiotic therapy.

 

Investigations in rodent models of intra-abdominal infection have identified polymicrobial infection, specifically with a mixture of aerobic and anaerobic bacteria, and foreign matter as key requisites for the development of intra-abdominal abscesses. When Enterobacteriaceae such as E. coli are the sole pathogens, the acute mortality rate due to bacteraemia is high, but survivors do not develop late intra-abdominal abscesses. Surprisingly, intra-abdominal infection with the anaerobic Gram-negative Bacteroides fragilis does not lead to early death or subsequent abscess development. However, an infection with the combination of these pathogens is remarkably effective in generating abscesses in surviving animals, even when antibiotic therapy effective against E. coli is administered acutely, preventing early deaths from bacteraemia. The presence of foreign matter facilitates the development of progressive infection. These requisites are readily met in clinical practice, where a perforated viscus is the most common risk factor for the development of intra-abdominal abscess, where polymicrobial infection is the rule, and when luminal contents or devitalized tissue provide a suitable nidus for infection.

 

The acute inflammatory response is ineffective in overcoming such mixed infections, and may even promote the development of abscesses. The presence of bacteria within the normally sterile peritoneal cavity leads to an influx of acute inflammatory cells and the ingress of plasma across the inflamed vascular bed. Omentum and fibrinous adhesions localize the site of infection, producing a phlegmon, but the hypoxic environment characteristic of such mixed infections facilitates the growth of anaerobes, and impairs the bactericidal activity of granulocytes. The degradation of cellular and bacterial debris by granulocytic hydrolases creates a hypertonic fluid that expands in response to osmotic forces, and the abscess cavity enlarges. This process can continue until a lethal complication such as bacteraemia, or erosion of a viscus or blood vessel develops, or until the host succumbs to the debilitating chronic inflammatory process.

 

ANATOMICAL CONSIDERATIONS

Intra-abdominal abscesses tend to develop in one or more discrete locations. The subphrenic spaces (in relation to both the right and left hemidiaphragms), subhepatic space (particularly the posterior aspect of the subhepatic space), the parocolic gutters located posterolaterally in the concavities between the spine and the lateral abdominal wall, and the pelvic peritoneum are the most commonly encountered sites of abscess formation. In addition to pooling in these dependent spaces, the respiratory movement of the diaphragms produces negative intra-abdominal forces that draw fluid to the subphrenic spaces. Abscesses in the lesser sac generally develop as a consequence of severe pancreatitis or perforated gastric or duodenal ulcers, or as complications of gallbladder or gastric surgery. Localized abscesses may develop in relation to a gangrenous viscus, resulting in periappendiceal, subhepatic, pericholecystic, or interloop abscesses, localized by the gut mesentery and adjacent loops of bowel.

 

MICROBIAL AETIOLOGY

The microbial flora of the digestive tract shift from small numbers of aerobic streptococci (including enterococci) and facultative Gram-negative bacilli in the stomach, duodenum, and jejunum to larger numbers of these species together with a great excess of anaerobic Gram-negative bacilli (particularly Bacteroides spp.) and anaerobic Gram-positive flora (streptococci and clostridia) in the distal ileum and colon. Intra-abdominal abscesses usually contain multiple pathogens, and almost always a mixture of aerobic and anaerobic organisms. The yield obtained from abscess cultures depends on the prior or concurrent administration of antibiotics, the care with which cultures are transported in a truly anaerobic environment (preferably a capped syringe of abscess contents, cultured promptly), and the diligence with which cultures are processed both aerobically and anaerobically.

 

Occasionally, yeasts such as Candida species may be recovered from abscess fluid. Patients who develop intra-abdominal abscesses after prolonged hospital admission or prior antibiotic therapy may be colonized by a variety of nosocomial pathogens. Abscesses that follow penetrating trauma may also include skin flora such as staphylococci, or non-fermenting Gram-negative bacilli such as Pseudomonas species acquired from the environment.

 

UNDERLYING CAUSES

Perforation of a viscus (complicating peptic ulcer disease, appendicitis, diverticulitis, cholecystitis, Crohn's disease, adenocarcinoma of the colon); progressive ischaemia and infarction of the gallbladder, pancreas, or intestine, including typhlitis in the setting of profound neutropenia; penetrating injuries to the abdominal viscera; or leakage of intestinal contents during abdominal surgery are the usual antecedent events which lead to the development of intra-abdominal abscesses. Although abscesses may localize in close relation to the site of initial peritoneal contamination, the anatomical considerations described above regularly lead to the development of subphrenic or pelvic abscesses distant from the primary lesion.

 

CLINICAL FEATURES

An abscess is readily suspected in a patient with a predisposing primary intra-abdominal disease or whose postoperative course is complicated by abdominal pain, focal tenderness, progressive fever, leucocytosis, and intermittent polymicrobial bacteraemia. However, many of these classic features may be absent, and patients with an intra-abdominal abscess may present with persistent ileus, abdominal distension, mild liver function abnormalities, an unexplained and persistent pleural effusion, or an entirely non-localizing debilitating illness with or without fever. Pelvic abscesses may be associated with urinary frequency, diarrhoea, or tenesmus. Fever, malaise, and associated symptoms may emerge only after the completion of antibiotic therapy for a postoperative pneumonia or superficial wound infection. Occasionally, symptoms may develop indolently months after an initial presentation, and the relationship between the current illness and prior abdominal surgery may be overlooked. Rarely, the symptoms may suggest a primarily intrathoracic process (empyema, purulent pericarditis, or pneumonia) as a result of extension of a subphrenic abscess through the diaphragm. A pneumonia-like presentation may result from slow spread through the diaphragm with necessitation directly into a bronchus rather than the more frequent extension into the pleural space.

 

DIAGNOSIS

Abnormal haematological parameters accompanying progressive inflammation such as leucocytosis, anaemia, and thrombocytosis are frequently present, but debilitated and elderly patients often fail to mount a reactive leucocytosis. Polymicrobial bacteraemia strongly implicates the presence of an intra-abdominal abscess. In many patients receiving antibiotic therapy, only a single species may be recovered on blood culture, but other foci of infections (urinary tract, intravascular catheter, pneumonia) should also be considered.

 

Modern cross-sectional imaging techniques have greatly facilitated the diagnosis of intra-abdominal abscesses, and have revolutionized their therapy (Fig. 1) 1191: simply considering the possibility of such a complication and initiating the appropriate imaging studies become crucial factors in confirming the diagnosis. Conventional abdominal plain films may indicate the presence of an abscess by demonstrating abnormal (extraluminal) collections of gas, air–fluid levels, or mottled material. More subtle findings may include pleural effusion, elevation of the diaphragm, or mass effect displacing the stomach or other viscera. The use of iodinated soluble oral contrast agents with CT scanning provides information on displacement or obstruction of viscera by an inflammatory mass or abscess, and has largely supplanted barium contrast studies in such evaluations.

 

Ultrasound and CT scanning provide independent approaches for the detection of intra-abdominal abscesses. Both show established abscesses to have an ellipsoid shape and to displace adjacent viscera, and some have a discernible abscess wall or rind. Volumetric estimates can be extrapolated directly from the images. The sequence of imaging studies depends largely on an individual institution's resources and experience. Ultrasound instruments are mobile and may be transported to the bedside of a critically ill patient, allowing rapid imaging without X-irradiation. The quality of such studies is somewhat dependent on the skill of the operator, the sensitivity of the instrument used, and the interpretation skills of the consulting radiologist. The most important limitations, however, are the need for a sonolucent ‘window’ to image the region of interest, and the interference created by overlying bowel gas, dressings, and drains. Pelvic, subphrenic, and relatively superficial intra-abdominal collections abutting the abdominal wall are readily visualized by ultrasound. The ultrasound appearance of intra-abdominal abscesses varies from homogeneous hypoechoic fluid to more complex echogenic collections. A visualized collection may be sterile (bile, haematoma) or infected, and aspiration is usually required to document the presence of an abscess.

 

CT scanning requires transport of the patient to the scanner, and is routinely performed after the administration of both oral and intravenous contrast material; this limits its utility in patients with profound ileus or renal insufficiency. Serial images are obtained from the level of the diaphragm to the pelvis; these images are particularly helpful in identifying an occult abscess, and in excluding multiple abscesses. In addition to identifying a low-attenuation extraluminal mass, CT scans can document inflammatory oedema in the adjoining fat (obliteration of fat planes) and hyperaemia in the abscess capsule (enhancement). CT imaging is particularly well-suited for investigation of small or deep intra-abdominal collections.

 

Ultrasound and CT imaging techniques have limited the role of radionuclide scanning in the diagnosis of intra-abdominal abscesses. Gallium-67 citrate is administered intravenously, and binds to lactoferrin released by leucocytes at sites of inflammation. Indium-111 oxine is used to label the patient's granulocytes ex vivo, which are then re-infused intravenously. Although each method is effective at identifying intra-abdominal abscesses, the images provide insufficient resolution to guide needle drainage. Gallium-67 scanning is limited by its excretion by the colon, delaying scanning for 1 to 3 days before the background level is satisfactory to image intra-abdominal collections. It also lacks specificity, since Gallium-67 is concentrated by neoplasms and sterile inflammatory processes. Indium-111 labelling of granulocytes is laborious, the labelling and scanning process requires 6 to 12 h, and leucocyte uptake may also lack specificity. False-positive findings suggesting an abdominal abscess can result from a pulmonary infection from which indium-111-labelled leucocytes have been coughed up in sputum, swallowed, passed through the gastrointestinal tract and pooled temporarily in the caecal area. The use of such techniques is now largely limited to patients in whom intra-abdominal abscesses are strongly suspected but whose ultrasound or CT images fail to provide adequate diagnostic information.

 

THERAPY

Once an intra-abdominal abscess has been defined radiologically, a drainage procedure is necessary. Over the past decade percutaneous catheter drainage procedures based on cross-sectional imaging techniques have become increasingly important. Ultrasound or CT study localizes the collection and defines safe access for catheter placement, avoiding adjoining viscera and blood vessels, and facilitating dependent drainage (Fig. 2) 1192,1193,1194. An initial diagnostic needle aspiration is frequently performed to document access to the suspected abscess, to provide fluid for Gram stain and culture, and to facilitate placement of a larger bore drainage catheter. The highest rates of successful catheter drainage have been associated with single, unilocular abscesses. Multiple or complex abscesses, such as those which are poorly defined, multilocular, or septated containing necrotic material, or associated with a visceral fistula, have lower cure rates (about 50 per cent) when managed by catheter drainage alone. Initial percutaneous drainage often offers important benefits to the critically ill patient, even when the abscess is complex. Such manoeuvres may control sepsis, improve haemodynamics, and stabilize the patient prior to definitive surgical treatment. Initial catheter drainage may also drain peridiverticular or other local abscesses sufficiently to make single-stage resection and internal anastomosis possible, rather than traditional multiple-stage procedures.

 

After placement of a suitable percutaneous catheter and gentle aspiration of abscess contents, repeat imaging should be performed to confirm catheter position and to estimate the size of the residual cavity. Subsequent cessation of drainage may be due to effective obliteration of the cavity, or occlusion or dislodgement of the catheter. Catheters may be gently irrigated with 5 ml of sterile saline; if this procedure is unsuccessful in re-establishing drainage, catheters may be manipulated or replaced during repeat imaging. The duration of drainage is usually 1 to 2 weeks, and is accompanied by resolution of fever, signs of toxicity, and leucocytosis. Persistent drainage usually reflects the presence of an enteric fistula—this can be documented by sinography under fluoroscopic control.

 

Surgical exploration is largely restricted to patients who have failed to improve despite percutaneous drainage, or in whom collections are not amenable to catheter drainage. Ideally, direct surgical drainage should be performed via an extraperitoneal approach, to limit the risk of further contamination of the peritoneal cavity. After thorough debridement and irrigation of the abscess, multiple soft drains should be placed, facilitating dependent drainage. In some instances transabdominal exploration is necessary. When pancreatic abscesses or large collections develop in association with extensive necrotizing pancreatitis, particularly involving the lesser sac, surgical exploration is inevitable, and in some centres surgical drainage is undertaken directly. Interloop abscesses are also difficult to drain percutaneously and may be multiple; they generally require transabdominal drainage. Pelvic abscesses are often palpable as tender bulging masses impinging on the vagina or rectum. Drainage is achieved transvaginally or transrectally, with incision, digital exploration, and thorough irrigation.

 

Antibiotic therapy plays an adjunctive role in the management of abscesses, but medical treatment alone never represents definitive therapy. Since abscess fluid generally contains a mixture of aerobic or facultative organisms as well as anaerobes, initial empirical therapy must be effective against a broad range of pathogens. Simple and effective therapy against all of the recovered pathogens is sometimes problematic, and with adequate abscess drainage it is not usually necessary to treat every organism with maximally potent antibiotics. Fortunately, currently available antibiotics offer several options for empiric therapy.

 

Clindamycin and gentamicin have become a ‘gold standard’ of empirical therapy for intra-abdominal infections. However, the combination of ampicillin, gentamicin, and metronidazole is a more cost-effective regimen that provides therapy against enterococci as well as Gram-negative bacilli and anaerobes. In selected circumstances, such as in critically ill patients with bacteraemia due to ampicillin-resistant Gram-negative bacilli, or in neutropenic patients, a variety of newer &bgr;-lactam agents (piperacillin; second generation cephalosporins such as cefotetan or cefoxitin; third-generation agents such as cefotaxime, ceftriaxone, or ceftazidime; or imipenem) or agents combined with &bgr;-lactamase inhibitors (ampicillin-sulbactam or ticarcillin–clavulanic acid) may be used in place of ampicillin. When one of these agents provides adequate activity against anaerobic bacteria, metronidazole or clindamycin may be omitted. Careful monitoring of blood levels of gentamicin is important to minimize the risk of nephrotoxicity. When Candida species appear to play an important aetiological role, based on its presence in abundance on Gram stain, subsequent candidaemia, and the patient's immune status, therapy with amphotericin B is indicated, although its use may be supplanted in selected patients by the less toxic antifungal agent fluconazole.

 

Special considerations in the management of subphrenic abscesses

The importance of intra-abdominal catastrophes such as suppurative appendicitis and perforation of the stomach and duodenum requiring emergency laparotomy as antecedent causes for the development of subphrenic abscesses has waned; prior surgical procedures, particularly of the biliary tract, stomach, and colon are now the most frequent causes of subphrenic abscess. The growing frequency of left subphrenic abscesses, as high as 40 per cent in some series, has reflected this change in aetiology. The falciform ligament divides the left and right subphrenic spaces and limits the frequency of bilateral subphrenic abscesses. However, multiple synchronous intra-abdominal abscesses including subphrenic collections are rather common, and failure to identify and drain these additional abscesses is frequently responsible for adverse outcomes. The coexistence of infected and non-infected loculated collections adds to the complexity of this situation. Occasionally, right subphrenic abscesses develop in relation to a hepatic visceral abscess, particularly amoebic abscess. Since amoebic disease is cosmopolitan and ubiquitous, this diagnosis should always be considered in acutely ill individuals with parenchymal hepatic abscesses and subphrenic collections.

 

The deep location of these infections leads to a paucity of localizing signs and symptoms beyond the systemic findings discussed above. Ultrasound readily displays the diaphragm and is particularly valuable in distinguishing between pleural effusions and subphrenic collections. CT scanning is helpful in defining the precise localization of a subphrenic abscess and identifying the presence of additional intra-abdominal collections. Traditional open drainage of subphrenic abscesses through the bed of the twelfth rib used extraperitoneal and extrapleural approaches whenever possible, to minimize the risk of spreading infection within the abdomen and certainly within the pleural space: complicating empyemas were often fatal. Percutaneous catheter drainage is effective in draining most subphrenic abscesses, and a transperitoneal approach is most commonly used, reducing the risk of pneumothorax associated with transthoracic approaches. The risks associated with open transabdominal drainage of such abscesses have fallen greatly, and this approach allows synchronous collections to be treated at the same time. The management of individual patients depends on the antecedent causes of abscess formation, the precise anatomical location of the abscess, the availability of skilled interventional radiologists, and the presence of additional lesions.

 

Special considerations in the management of pelvic abscesses

Women can also develop abscesses following infection of the pelvic viscera as part of the spectrum of pelvic inflammatory disease. Neisseria gonorrhoeae and C. trachomatis are the primary pathogens responsible for salpingitis in sexually active women. These organisms are frequently recovered by endocervical culture or rapid diagnostic methods, but progression to tubo-ovarian abscess is often accompanied by polymicrobial infection with aerobic and anaerobic bacteria. Intrauterine contraceptive devices and altered local resistance factors accompanying menses are considered to be risk factors for the development of salpingitis. The spectrum of clinical presentation can be broad, but tubo-ovarian abscesses should be considered in patients with persistent fever and pelvic pain and tenderness despite appropriate initial intensive parenteral antibiotic therapy, such as cefoxitin and doxycycline (Fig. 3) 1195. Pelvic ultrasound is valuable in corroborating the clinical diagnosis, and distinguishing frank abscess from phlegmon; laparoscopy can be performed if there is diagnostic uncertainty. Surgical drainage (usually via colpotomy) is indicated in patients who fail to respond to initial medical therapy. In patients with a ruptured tubo-ovarian abscess, laparotomy with unilateral salpingo-oophorectomy is required. In some centres, percutaneous catheter drainage is attempted prior to surgical exploration.

 

In addition to polymicrobial tubo-ovarian abscesses, intrauterine devices have also been associated with pelvic actinomycosis. This indolent process often produces few clinical manifestations until a pelvic mass is appreciated. The extensive inflammatory mass is usually associated with tubo-ovarian abscess formation and surrounding woody induration, and can produce ureteral and/or bowel obstruction. There is obliteration of normal tissue planes, and the lesion may be mistaken for widespread malignancy. Initial resection is not feasible, and patients require prolonged parenteral therapy (6–12 weeks of penicillin) prior to definitive resection.

 

FURTHER READING

Altemeier WA, Culbertson WR, Fullen WD, Shook CD. Intra-abdominal abscesses. Am J Surg, 1973; 125: 70–9.

Boyd DP. The subphrenic spaces and the emperor's new robes. N Engl J Med, 1966; 275: 911–7.

Centers for Disease Control. Pelvic inflammatory disease: guidelines for prevention and management. MMWR 1991; 40: 1–25.

Clark RA, Towbin R. Abscess drainage with CT and ultrasound guidance. Radiol Clin N Am, 1983; 21: 445–59.

Halasz NA. Subphrenic abscess: myths and facts. JAMA, 1970; 214: 724–6.

Mueller PR, Simeone JF. Intra-abdominal abscesses: diagnosis by sonography and computed tomography. Radiol Clin N Am, 1983; 21: 425–43.

Mueller PR, et al. Percutaneous drainage of subphrenic abscess: a review of 62 patients. Am J Roentgenol, 1986; 147: 1237–40.

Mueller PR, vanSonnenberg E. Interventional radiology in the chest and abdomen. N Engl J Med, 1990; 322: 1364–74.

Ochsner A, Graves AM. Subphrenic abscess: an analysis of 3,372 collected and personal cases. Ann Surg, 1933; 98: 961–90.

Onderdonk AB, Bartlett JG, Louie T, Sullivan-Seigler N, Gorbach SL. Microbial synergy in experimental intra-abdominal abscess. Infect Immun, 1976; 13: 22–6.

Papanicolaou N, et al. Abscess-fistula association: radiologic recognition and percutaneous management. Am J Roentgenol, 1984; 143: 811–5.

Rotstein OD, Pruett TL, Simmons RL. Mechanisms of microbial synergy in polymicrobial surgical infections. Rev Infect Dis, 1985; 7: 151–70.

Sanders RC. The changing epidemiology of subphrenic abscess and its clinical and radiological consequences. Br J Surg, 1970; 57: 449–55.

vanSonnenberg E, et al. Temporizing effect of percutaneous drainage of complicated abscesses in critically ill patients. Am J Roentgenol, 1984; 142: 821–6.

Wang SMS, Wilson SE. Subphrenic abscess: the new epidemiology. Arch Surg, 1977; 112: 934–6.

Хостинг от uCoz