STOP-IT sooner – short sharp anti-bug course for intra-abdominal sepsis?

Complicated intra-abdominal infection is a common case on the units admitting general surgical patients. By definition it is infection that extends beyond the viscus of origin into the peritoneal space, and is associated with abscess or peritonitis.
The traditional management approach has been to do the source control and then soak in antibiotics until signs of inflammatory response have settled, usually a good couple of weeks. This is despite guidelines moving towards a 4-10 day course in recent years (eg IDSA or WSES).


In this pragmatic, multicenter (USA and Canada), non-inferiority, randomised trial, published in NEJM, 518 patients with complicated intra-abdominal infection (source control achieved) AND fever or leucocytosis or peritonitis-related gastrointestinal dysfunction (not managing half their ideal nutrition), were given one of two antimicrobial plans:

  1. 4 day course post procedure.
  2. continued therapy until 48 hours of temperature <38, white cells <11, >50% standard nutrition,  or up to 10 days.


The primary outcome was a composite of occurrence of surgical site infection, recurrence of intra-abdominal infection, or death.

There was no difference in primary outcome (~20%), or any element of it (eg death – both around 1%).

There was no difference in emergence of C.difficile or resistant organisms. 

IA infection from NEJM

Potential problems with using this study to inform practice on ICU:

  • The patients were generally mildly unwell (APACHE 2 of 10) – perhaps not as sick as the cases we deal with. The subgroup with APACHE > 10 also had no difference in outcome but this is a post-hoc remark only.
  • The power calculation was based on a higher complication rate, leaving the study underpowered.
  • A considerable number of each group didn’t get the specified treatment (71/260 in control, 47/258 in experimental), mainly because the antibiotic course length was too short or long. Non-adherence can mean bias, in this case favouring the null hypothesis.
  • Funding disappeared after the fist interim analysis.
  • It doesn’t tell us what to do with immunosuppressed patients.


This suggests the 20% that develop complications don’t get them from a lack of antibiotic but probably form failed procedural source control.

All in all this supports, but not confirms, the practice of those in favour of short antimicrobial courses for intra-abdominal infection (post attempted source control). Saving pennies and perhaps looking after the microbiome are sensible considerations after all. 

Refine STOP-IT and set it on the ICU.




As a reminder, the IDSA list the following as risk factors for source control failure in these patients:

  1. Delay in the initial intervention (124 h)
  2. High severity of illness (APACHE II score 15)
  3. Advanced age
  4. Comorbidity and degree of organ dysfunction
  5. Low albumin level
  6. Poor nutritional status
  7. Degree of peritoneal involvement or diffuse peritonitis Inability to achieve adequate debridement or control of drainage
  8. Presence of malignancy