Infectious Disease Compendium

Intra-abdominal Abscess


IDSA have guidelines. To summarize: drain the pus, kill the bugs. How hard is it?

CT. Enhanced is best.

Epidemiologic Risks

Trauma, cancer or foreign body like a bone. Something needs to cause a perforation of gut. It might be diverticula, a surgical misadventure or a perforated appendix.  The weirdest cause I ever saw was a bread bag square tie that was (I hope) accidently eaten and perforated the colon.


Pyogenic are often mixed: E. coli + Streptococci (esp milleri) + anaerobes.

Appendicitis may be due to invasive infection of Fusobacterium nucleatum and necrophorum (PubMed)(PubMed)(Pubmed).

Candida is often part of duodenal perforations, rarely seen in this era of H2 blockers and PPI's.

There is a curious literature, that may not pan out, suggesting that adenovirus may play a role: military recruits who get the adenovirus vaccine have less appendicitis and those that do are more like to be adenovirus positive on PCR (Link)

Empiric Therapy

I prefer a third generation cephalosporins PLUS metronidazole It is every bit as good as a carbapenam (Pubmed). Alternatives include a carbapenem OR penicillin/beta-lactamase inhibitors OR a quinolone PLUS metronidazole. Clindamycin can be used in place of metronidazole.

If you are going to opt for the non-operative approach, then ertapenem (1 g/d) for 3 days followed by 7 days of oral levofloxacin 500 mg qd and metronidazole 500 mg tid is non-inferior to taking out the appendix (PubMed).

Monotherapy with moxifloxacin works 85% of the time, killing about 80% of the anaerobes (PubMed). As if a 15% failure rate is good.

The traditional clindamycin/aminoglycoside is probably inferior (PubMed) due to both increasing resistance and lousy pharmacokinetics.

Duration? 4–7 days unless it is difficult to achieve adequate source control (PubMed).  I am a huge fan of the STOP IT trial.  For intrabdominal infection, 4 days was fine as long as there was good source control and longer courses of antibiotics only delayed the needed source control (PubMed).  This was true in the subset of patient who were septic (Pubmed).  It's all about the source control.  And here is a little secret that most people do no know: broadening antibiotic coverage when the patient is looking infected is not a substitute for draining pus/source control.

For complicated appendicitis, postop antibiotics may not do much to prevent wound infections (PubMed).  And for complicated appendicitis, 5 days post op antibiotics is no better than three (PubMed).


The whole fecalith as a cause of appendicitis (PubMed) is likely a load of fecalith. It is more more likely an acute Fusobacterium infection (PubMed)

While taking it out is the usual treatment of acute apendicitis (a first choice in all the guidelines), a trial of antibiotics will prevent most patients from needing surgery. Not unreasonalbe, especially if prior surgeries make a trip to the OR problematic. But, "It remains to be determined whether the benefits of potentially avoiding an operation with the antibiotics-first approach are outweighed by the burden to the patient related to future appendicitis episodes, more days of antibiotic therapy, lingering symptoms, and uncertainty that may affect quality of life (PubMed)."

The change of relapsing after medical therapy for appendicitis is about 15% (PubMed).

Drain drain drain. You have to control the infection for the antibiotics to work.

In both kids (PubMed) and adults antibiotics can cure acute appendicitis without surgery (PubMed). BUT.   Primary appendicectomy has a lower rate of post-operative complications (PubMed).  So the best bet is take it out.

And oddly, the larger the volume used for irrigation with appendicitis in the OR, the greater the risk of infection (PubMed).

Where I live, E. coli is 30% resistant to all forms of penicillin, so avoid penicillins if local resistance is > 10% or so.


I tend not to worry about the enterococcus, it is other beasts that will kill your patient, and it will go away of you drain the pus. Usually. The enterococcus does not increase mortality or need for re-operation, but may increase the rate of abscess formation (PubMed).

Here is the dumbest study/conclusion of all time (PubMed). Patients were randomized to 8 or 15 days of antibiotics for intrabdominal sepsis. The conclusion? "Patients treated for 8 days had a higher median number of antibiotic-free days than those treated for 15 days (15 [6–20] vs 12 [6–13] days, respectively; P < 0.0001) ... Conclusion Short-course antibiotic therapy in critically ill ICU patients with PIAI reduces antibiotic exposure." Thank you Dr. Obvious. But there was no mortality difference.

Curious Cases

Relevant links to my Medscape blog

Migratory Plastic

My Secret


The RUQ sucks. Or so I was always told.

Record Delay

Born Under a Bad Sign


Duration: Disease and Treatment?

Last Updated 3/30/18