NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Holzheimer RG, Mannick JA, editors. Surgical Treatment: Evidence-Based and Problem-Oriented. Munich: Zuckschwerdt; 2001.

Cover of Surgical Treatment

Surgical Treatment: Evidence-Based and Problem-Oriented.

Show details

Management of intra-abdominal abscesses


Author Information

“‘Signs of pus somewhere, signs of pus nowhere else, signs of pus there - under the diaphragm' - this was 100% true when I was a student, 50% true when I was a resident. Today it is irrelevant …”

The contents of this chapter could have been summarized only in a sentence: an abscess is a pus-containing, confined structure, which requires to be drained by whichever means available at your disposition. The Editors, however, commanded us to further elaborate. Abscesses may develop anywhere within the abdomen, resulting from a myriad of conditions. Specific types of abscesses are covered elsewhere in this book; this chapter will introduce you with general concepts.

Evidence: the vast majority of publications dealing with this topic represent level V-retrospective case series; a few include historical controls (level IV), even fewer publications use non-randomized concurrent controls (level III). Thus, almost all recommendations given by the publications quoted in this chapter are grade C. As usual - when high level of evidence is not available - the clinician has to use an individual approach and common sense-which will be herein emphasized. Whenever higher level evidence and recommendation are available it will be mentioned.

Definition and significance

Erroneously, the term intra-abdominal abscess has been used - and still is - as a synonym with secondary peritonitis. This is obviously not true as abscesses develop thanks to effective host defenses and represent a relatively successful outcome of peritonitis (1). To be termed an abscess the confined structure has to be walled off by an inflammatory wall and possess a viscous interior. In contrary, free-flowing contaminated or infected peritoneal fluid, or loculated collections, which are deprived of a wall, represent a phase in the spectrum-continuum of peritoneal contamination/infection and not an abscess (2).

Classification & pathogenesis

The immense spectrum of intra-abdominal abscesses makes its classification complex (table I). But practically, abscesses are visceral (e.g. hepatic or splenic) or non-visceral (e.g. subphrenic, pelvic), intraperitoneal or extra-peritoneal. Non-visceral abscesses arise following the resolution of diffuse peritonitis in which a loculated area of infection and suppuration is “walled off” and persists; or after a perforation of a viscus, which is effectively localized by peritoneal defenses. Visceral abscesses are caused by hematogenous or lymphatic dissemination of bacteria to a parenchymatous viscus; retroperitoneal abscesses may result from a perforation of a hollow viscus into the retroperitoneum as well as by hematogenous or lymphatic spread (3, 4). Another distinction is between postoperative abscesses - for the development of which we surgeons feel responsible - and spontaneous abscesses - unassociated with previous operation. A further clinically significant separation is between simple abscesses and complex. (e.g. multiple, mutiloculated, associated with tissue necrosis, enteric communication or tumor) which require a more aggressive therapy and carry poorer prognosis (5). The anatomical classification - based on the specific anatomical location of abscess - which typically develops in one of the few constant potential spaces - has diminished in significance since the availability of modern imaging and percutaneous drainage techniques.

Table I. Classification of abdominal abscesses.

Table I

Classification of abdominal abscesses.

Note that abscesses signify an intermediate natural outcome of contamination/infection. On the one end of spectrum infection persists, spreads and kills, on the other end, the process is entirely cleared by host's defenses-assisted by our therapy. Abscesses lies in between; where the peritoneal defenses are only partially effective - being disturbed by an overwhelming number of bacteria, micro - environmental hypoxemia or acidosis, and adjuvants of infections such as necrotic debris, hemoglobin, fibrin and barium sulfate. Abdominal abscesses do not kill the patient immediately but if left untreated they become gradually lethal, unless spontaneous drainage ensues.


Generally speaking the bacteriology of abdominal abscesses is polymicrobial. Abscesses which develop in the aftermath of secondary peritonitis (e.g. appendiceal or diverticular abscess) possess the mixed aerobic-anaerobic flora of secondary peritonitis (1). It appears that while endotoxin generating facultative anaerobes such as E. coli, are responsible for the phase of acute peritonitis, the obligate anaerobes, such as Bacteroides fragilis, are responsible for late abscess formation. These bacteria act in synergy: both are necessary to produce an abscess and the obligate anaerobe can increase the lethality of an otherwise non-lethal inoculum of the facultative microorganism. The vast majority of visceral abscesses (e.g. hepatic and splenic) are polymicrobial - aerobic, anaerobic, Gram negative and positive (6). This is also true for retroperitoneal abscesses (7). Primary abscesses such as the psoas one are often mono-bacterial, with the Staphylococci predominatly (4). Postoperative abscesses are often characterized by the flora typical of tertiary peritonitis-representing superinfection with yeasts and other commensulas (1). The low virulence of these organisms, which represent a marker of tertiary peritonitis, and not its cause, reflects the global immunodepression of the affected patients.

Clinical features

Clinical presentation of abdominal abscesses is as heterogeneous and multifaceted as the abscesses themselves. The spectrum is vast - systemic repercussions of the infection vary between a frank septic shock to nothing at all - being suppressed by immunoparesis and antibiotics. Locally the abscess may be felt - through the abdominal wall, the rectum or vagina; in most instances however it remains physically occult. In our modern times, when any fever is an alleged indication for antibio-tics, most abscesses are initially “partially treated” or “masked” - presenting as a systemic inflammatory response syndrome (SIRS) with or without multi-organ dysfunction. Paralytic ileus is another not uncommon presentation of abdominal abscesses; in the postoperative situation it is an ileus, which “fails to resolve”.


Life has become simple! Modern abdominal imaging has revolutionized the diagnosis of abdominal abscesses. Yes, you still need to suspect the abscess and carefully examine your patient but the definitive diagnosis (and usually the treatment) depends of imaging techniques. Computed tomography (CT), ultrasound (US) and various radioisotope-scanning techniques are available (8). Which is the best?

Radioisotope scanning with whichever radiolabeled material does not provide any anatomical data beyond localization of an inflammatory site; they are not accurate enough to permit percutaneous (PC) drainage. The usefulness of these methods is limited therefore to the continuous survival of nuclear medicine units and an excuse to publish papers. Practically, these tests have no role at all. Both US and CT provide good anatomical definition including the abscess's site, size and structure'; both can guide PC drainage. US is portable, cheaper, and more accurate at detecting abscesses in the right upper abdomen and pelvis. It is, however, extremely operator dependent. We - surgeons - are better trained to read CT scans than US; hence we prefer the CT which allows us to visualize the entire abdomen, independently assessing the anatomy of the abscess, and plan its optimal management. CT-intravenously and luminary contrast-enhanced, is also helpful in classifying the abscess either as simple or complex (table I).

It appears that performing multiple tests-adding a CT to an US-is not productive (9). Likewise, CT or US scanning during the first postoperative week is futile (10) because neither technique can distinguish between a sterile fluid collection (e.g. residual lavage fluid), an infected fluid collection or a frank abscess. The best and only way to document the infective nature of the visualized fluid is a diagnostic PC aspiration-subjecting the aspirate for a Gram stain and culture. CT features, which however may hint that one deals with a proper abscess, are a contrast-enhancing, well defined rim, and the presence of air bubbles.


Abdominal abscesses are managed with drainage; when the source is present it should be dealt with, antibiotic treatment is of marginal importance.

Antibiotics: the truth is that no real evidence exists to prove that anti-antimicrobial agents - which anyway poorly penetrate into an established abscess (11) - are necessary at all in addition to the complete evacuation of the pus. Think about the good old days-not many years' ago-when pelvic abscesses were observed until reaching “maturity” and then drained through the rectum or the vagina; no antibiotics were used and the recovery was immediate and complete. The prevalent standard of care, however, although lacking evidence, maintains that when an abscess is strongly suspected or diagnosed-antibiotic therapy should be initiated. The latter should initially be empirically targeted against the usual expected polymicrobial spectrum of bacteria; when the causative bacteria are identified, the coverage can be changed or reduced as indicated. How long to administer antibiotics? Again there are no scientific data to formulate logical guidelines. Commonsense dictates that prolonged administration after the effective drainage is not necessary. Theoretically antibiotics may combat bacteremia during drainage and eradicate locally spilled microorganisms; but after the pus has been evacuated - leading to a clinical response - antibiotics should be discontinued. The presence of a drain is not an indication to continue with administration (12).

Conservative treatment

Traditionally, multiple hepatic abscesses, as a consequence to portal pyemia, which are not amenable to drainage, are treated with antibiotics - with a variable response rate. A few recent level V communications claimed that non-operative treatment, with prolonged administration of antibiotics, is also effective in children who develop abdominal abscesses following an appendectomy for acute appendicitis (13). The problem with such studies is that the alleged “abscesses”, which were imagined on US or CT, were never proven as such. Instead, it probably represented sterile collections - the majority requiring no therapy at all.


Philosophy & timing: presently the prevailing paradigm, when an abscess is suspected on a CT or an US, is to hit the patient with antibiotics and rush to drainage. In the hysterical hurry “to treat”, clinical lessons leaned over centuries are often ignored. Only a generation ago, a patient who spiked temperature after an appendectomy was patiently but carefully observed without antibiotics (which even did not exist); usually the temperature - signifying residual local inflammatory response syndrome (LIRS) (2) - subsided spontaneously. In the minority of patients “septic” fever persisted, reflecting maturing local suppuration. The later was eventually drained through the rectum when assessed as “mature”. Today, conversely, antibiotics are immediately given to mask the clinical picture, and imaging techniques are instantly ordered to diagnose “red herrings”, which in turn promote unnecessary invasive procedures. Remember: in a stable patient fever is a symptom of effective host defenses - not an indication to be aggressive - invasive.


When an abscess is suspected a few dilemmas arise and should be dealt with stepwise:

Is it an abscess or a sterile collection?

The aforementioned CT features may be helpful but the clinical scenario is as important-especially when postoperative abscesses are concerned. Abscesses are rarely mature for drainage before a week has passed since the operation; two weeks after the operation the cause of “sepsis” is rarely within the abdomen. When in doubt image-guided diagnostic aspiration is indicated.

Percutaneous (PC) versus open surgical drainage?

During the 1980's multiple level V and IV case series suggested that the results of PC drainage are at least as good as that achieved by an operation. In a level III study, in well stratified patients, surgical or PC drainage documented a mortality of 14% and 12%, respectively (14). In another level III study of 91 patients the mortality rate was 29%. Only 1.7% of the patients with an APACHE II score of less than 15 died compared with 78% when APACHE II was equal to or more than 15. Only 8% of patients with APACHE II equal to or more than 15 undergoing PC drainage survived, compared with 30% of patients who underwent surgical drainage procedures. The authors claimed that, paradoxically, despite the attractiveness of a PC technique for abscess drainage in the most ill patients, a better, although not statistically improved, chance for survival was noted with surgical treatment. They recommended that surgical treatment not be avoided because the patient is considered to be too ill (15). Be it as may, although there is no clear evidence to attribute lesser mortality or morbidity to PC drainage versus surgical drainage, the former represent a minimally-invasive procedure which can spare the patient the unpleasantness of another open abdominal operation.

Malangoni et al. suggested that complex abscesses (e.g. multiple, mutiloculated, associated with tissue necrosis, enteric communication or tumor) poorly respond to PC drainage while most simple abscesses do (5). Others however claim that the above factors represent no contra-indication to PC drainage which may be successfully attempted nevertheless (16, 17). The concept that a “temporizing” PC abscess drainage in gravely ill patients with complex abscesses offers significant therapeutic benefit -allowing a definitive semi-elective laparotomy in better stabilized patients, has been repeatedly made but never proven (18).

It appears that PC drainage and surgical drainage techniques should not be considered competitive but rather complementary. Grade C recommendation.

If an abscess is accessible by PC techniques, it is reasonable to consider a non-operative approach to the problem (19). The surgeon together with the radiologist should consider each abscess individually taking into the consideration the “pro & cons” depict in table II.

Table II. Intra-abdominal abscesses: percutaneous (PC) versus open surgical drainage. Considerations in selecting the approach.

Table II

Intra-abdominal abscesses: percutaneous (PC) versus open surgical drainage. Considerations in selecting the approach.

Aspiration only versus catheter drainage?

A single PC needle aspiration may successfully eradicate an abscess-especially when small and containing low-viscosity fluid. However there is level II+ evidence that PC catheter drainage is more effective (grade B recommendation) (20).

Size of PC catheters-drains?

Level V reports claimed advantage for large-bore trocar catheters for PC drainage (21, 22). A level III study concluded however that size 7 French PC sump drains were as effective as size 14 French (23).

Management of PC drains

There is not much science here; those are small tubes and should be regularly flushed with saline to remain patent. The drain site should be cleaned and observed: there is a single case report of necrotizing fasciitis of the abdominal wall around a PC drain site (24). PC drains are removed when clinical sepsis has resolved and the daily output (minus the saline injected) is below 25 ml. On the average, after PC drainage of a simple abdominal abscess, the drain is removed after seven days.


A clinical improvement should be seen within 24 to 72 hours following PC drainage. A level V study showed that persistent fever and leukocytosis at the fourth day after PC drainage correlated with a failure (25). Non-responders should be re-imaged with a CT-combined with water-soluble contrast injected through the drain. Depending on the finding a decision should be taken by the surgeon, in consultation with the radiologist, as to the next appropriate course of action - are - PC drain or an operation. Persistence of high output drainage in a patient who is clinically well can be better investigated with a tube sinogram to delineate the size of the residual abscess cavity. Abscess cavities, which do not collapse commonly recur. (8).

Failure of PC drainage: when to “switch over” to surgical drainage?

Patients who deteriorate after the first attempt at PC drainage should be operated upon promptly. No evidence is required to prove the common sense notion that further procrastination may be disastrous.

In stable non-responders to the initial PC drainage a second attempt may be appropriate, according to the considerations mentioned in table II. Inability to successfully affect the second PC drainage, or its clinical failure, mandates an open procedure.

Surgical management of intra-abdominal abscesses

About a third of intra-abdominal abscesses are not suitable to PC drainage and require an open operation (26). A few practical dilemmas exist:

Exploratory laparotomy vs. “direct” surgical approach

A “blind” exploratory laparotomy to search for abscesses-“somewhere”, so common less than 20 years ago is currently very rarely necessary. A “direct” approach is obviously more “benign”, sparing the previously uninvolved peritoneal spaces and avoiding bowel injury and wound complications. It is almost always possible in spontaneous abscesses, which are so well defined on CT. But those are also the kind of abscesses, which usually respond to PC drainage. Although postoperative abscesses are today anatomically well localized on CT - those which fail PC drainage are usually “complex”, thus often not amenable to a “direct” approach (e.g. interloop abscess) or requiring in addition the control of an intestinal source (3). Criteria to consider in selecting the approach are summarized in table III.

Table III. Exploratory laparotomy vs. “direct” open drainage of abdominal abscesses.

Table III

Exploratory laparotomy vs. “direct” open drainage of abdominal abscesses.

Direct approach: extra-peritoneal versus trans-peritoneal?

A level III study showed no significant differences in overall mortality and morbidly between the two approaches; the transperitoneal route was, however, associated with a higher incidence of injury to the bowel (21). It logical to suggest that the extra-peritoneal approach should be utilized whenever anatomically possible (3). Subphrenic and subhepatic abscesses can be approached extra-peritoneally through a subcostal incision or-if posterior-thorough the bed of the 12thrib. Old timers are still familiar with these techniques which are currently rarely utilized-being replaced by PC drainage. Peri-colic, appendicular and all sorts of retroperitoneal abscesses are best approached through a loin incision. Also late-appearing pancreatic abscesses can be drained extra-peritoneally - from the flank-occasionally needing bilateral approach. Pelvic abscesses are best drained through the rectum or vagina.


Classically, at the end of the open procedure a drain has been placed within the abscess cavity, brought to the skin away from the main incision. The type, size and number of drains used to depend more on local traditions and preferences than on science. Similarly, the postoperative management of drains involved cumbersome rituals with the drains sequentially shortened, based on serial contrast sinogram - to ascertain the gradual collapse of the cavities and drain - tracts. House surgeons and nurses forever changed dressings and irrigated the drains - again according to the locally prevailing ritual. Also undocumented, our experience is that this scenario should belong to history. With adequate surgical drainage, when the source of infection has been controlled, when the abscess cavity is “filled” with omentum or adjacent structures, and prophylactic perioperative antibiotics are administered - no drains are necessary. Trust the peritoneal cavity to deal with the residual bacteria better in the absence of a foreign-body-a drain. We do not remember when was the last time we had to “shorten” a drain or to obtain a drainsent inogram.


There is a paucity of high level evidence concerning the management of intra-abdominal abscesses. Therefore, use your logic-supported by grade C recommendations. Tailor your approach to the anatomy of the abscess, the physiology of the patient, and the local facilities available to you. Do not procrastinate, do not forget to deal with the source, do not over rely on antibiotics, and get rid off the pus.

Sepsis-the host generated systemic inflammatory response to the abscess may persist, and progress to organ failure, even after the abscess has been adequately managed (28). Try not to be too late.


Wittmann D H, Schein M, Condon R E. Current management of secondary peritonitis. Ann Surg. (1996);224:10–18. [PMC free article: PMC1235241] [PubMed: 8678610]
Schein M, Wittmann D H, Wise L, Condon R E. Abdominal contamination, infection and sepsis: a continuum. Br J Surg. (1997);84:269–2723. [PubMed: 9052453]
Rotstein O D, Meakins J L. Diagnostic and therapeutic challenges of intra-abdominal infections. World J Surg. (1990);14:159–166. [PubMed: 2183478]
Santaella R O, Fishman E K, Lipsett P A. Primary vs. secondary iliopsoas abscess. Presentation, microbiology, and treatment. Arch Surg. (1995);130:1309–1311. [PubMed: 7492279]
Malangoni M, Shumate C R, Thomas H A, Richardson J D. Factors infleuncing the treatment of intra-abdominal abscesses. Am J Surg. (1990);159:167–171. [PubMed: 2294795]
Brook I, Frazier E H. Microbiology of liver and spleen abscesses. J Med Microbiol. (1998);47:1075–1080. [PubMed: 9856643]
Brook I, Frazier E H. Aerobic and anaerobic microbiology of retroperitoneal abscesses. Clin Infect Dis. (1998);26:938–941. [PubMed: 9564479]
Gleason T G, Pruett T L, Sawyer R G. Intra-abdominal abscesses: emphasis on mageguided diagnosis and therapy. J Intensive Care Med. (1998);13:320–328.
Dobrin P B, Gully P H, Greenlee H B. et al. Radiologic diagnosis of an intra-abdominal abscess. Do multiple tests help? Arch Surg. (1986);121:41–46. [PubMed: 3510605]
Norwood S H, Civetta J M. Abdominal CT scanning in critically ill surgical patients. Ann Surg. (1985);202:166–175. [PMC free article: PMC1250870] [PubMed: 4015222]
Galandiuk S, Lamos J, Montgomery W, Young S, Polk H C. Antibiotic penetration of experimental intra-abdominal abscesses. Am Surg. (1995);61:521–525. [PubMed: 7762902]
Schein M, Wittmann D H, Lorenz W. Duration of postoperative antibiotic administration following emergency abdominal surgery. Eur J Surg. (1996);162 (supp 576):1–75.
Okoye B O, Rampersad B, Marantos A. et al. Abscess after appendicectomy in children: the role of conservative management. Br J Surg. (1998);85:1111–1113. [PubMed: 9718008]
Hemming A, Davis N L, Robins E. surgical versus percutaneous drainage of intra-abdominal abscesses. Am J Surg. (1991);161:593. [PubMed: 2031543]
Levison M A, Zeigler D. Correlation of APACHE II Score, drainage technique and outcome in postoperative intra-abdominal abscess. Surg Gynecol Obstet. (1991);172:89–94. [PubMed: 1989123]
Fulcher A S, Turner M A. Percutaneous drainage of enteric-related abscesses. Gastroenterologist. (1996);4:276–285. [PubMed: 8957100]
Goletti O, Lippolis P V, Chiarugi M. et al. Percutaneous ultrasound-guided drainage of intra-abdominal abscesses. Br J Surg. (1993);80:336–339. [PubMed: 8472145]
van Sonnenberg E, Wing V W, Casola G. et al. Temporizing effect of percutaneous drainage of complicated abscesses in critically ill patients. Am J Roentgenol. (1984);142:821–826. [PubMed: 6199965]
Levison M A. Percutaneous versus open operative drainage of intra-abdominal abscesses. Infect Dis Clin North Am. (1992);6:525–544. [PubMed: 1431037]
Rajak C L, Gupta S, Jain S. et al. Percutaneous treatment of liver abscesses: needle aspiration versus catheter drainage. Am J Roentgenol. (1998);170:1035–1039. [PubMed: 9530055]
Voros D, Gouliamos A, Kotoulas G. et al. Percutaneous drainage of intra-abdominal abscesses using large lumen tubes under computed tomographic control. Eur J Surg. (1996);162:895–898. [PubMed: 8956959]
Singh K, Lamparelli M, Foley R J. Use of a laparoscopic trocar for percutaneous introduction of drains into intra-abdominal abscesses. Br J Surg. (1996);83:670–671. [PubMed: 8689215]
Rothlin M A, Schob O, Klotz H. et al. Percutaneous drainage of abdominal abscesses: are large-bore catheters necessary? Eur J Surg. (1998);164:419–424. [PubMed: 9696442]
Holley D T, McGrath P C, Sloan D A. Necrotizing fasciitis as a complication of percutaneous catheter drainage of an intra-abdominal abscess. Am Surg. (1994);60:197–199. [PubMed: 8116981]
Brolin R E, Flancbaum L, Ercoli F R. et al. Limitations of percutaneous catheter drainage of abdominal abscesses. Surg Gynecol Obstet. (1991);173:203–210. [PubMed: 1925881]
Shuler F W, Newman C N, Angood P B, Tucker J G, Lucas G W. Nonoperative management for intra-abdominal abscesses. Am Surg. (1996);62:218–222. [PubMed: 8607582]
Stone H H, Mullins R J, Dunlop W E, Strom P R. Extraperitoneal versus transperitoneal drainage of the intra-abdominal abscess. Surg Gynecol Obstet. (1984);159:549–552. [PubMed: 6390758]
Norton L W. Does drainage of intraabdominal pus reverse multiple organ failure? Am J Surg. (1985);149:347–350. [PubMed: 3976989]
Copyright © 2001, W. Zuckschwerdt Verlag GmbH.
Bookshelf ID: NBK6937


  • PubReader
  • Print View
  • Cite this Page

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...