Appendicitis is a common problem in the emergency department, and for many years the standard of care has been to perform surgery to resolve it. When I trained in surgery (before transferring to EM) this was typically done as an open procedure. These days less invasive laparascopic techniques are more common, but it’s still an operation, a hospital stay and an intra-abdominal procedure.
In recent years there have been a number of trials looking at the use of an antibiotic regime as opposed to a surgical approach, and we looked at one of these back in 2016 (1). Interestingly, that 2016 study asked for a US based study (still not sure why) that was larger, and that appears to be what we have this week in the NEJM (2). The 2016 trial was similarly covered on the excellent SGEM. SGEM#180: The First Cut is the Deepest – N.O.T. for Paediatric Appendicitis
There is certainly evidence around that we can treat a proportion of patients with antibiotics, that seems to be well established, but there is also a significant rate of recurrence and so there are still some doubts amongst clinicians as to whether it is a sensible strategy.
The abstract is below, but as we always say, please read the full paper yourself and draw your own conclusions. You should also read the accompanying commentary (3).
What kind of paper is this?
It’s a randomised controlled trial which is exactly what we want to see in trials of interventions. It’s good to see this design in a surgical study.
Tell me about the patients
The study enrolled adult patients aged over 18 in 25 Washington state hospitals. They focused (quite reasonably) on uncomplicated appendicits and so those who were septic, diffusely peritonitic etc. were excluded. The pre-specified a group with an appendolith on imaging as they are thought to have a higher rate of recurrence. It appears that all patients had imaging as part of their work up. This is increasingly common in our practice too, but probably limits the generalisability to health economies with good access to radiology.
What was the intervention?
in the surgery arm of the trial they had an appendicectomy. In the antibiotic arm they had 24 hours of IV antibiotics followed by 10 days of oral antibiotics (as per local guidelines). Confusingly though it looks as though 47% were discharged directly from the ED (though it is unclear how long they were in the ED).
What about the outcomes?
This trial looked at 30-day health status as the primary outcome which is interesting. As an overall measure it’s great, but in a condition like this I wonder if a more specific disease outcome would be of benefit. These are present in the secondary outcomes where they followed the patients up for a year which is important as recurrent appendicitis is one of the main concerns about adopting a conservative management to the condition.
What about the main results?
They recruited 1552 patients from 8168 patients screeened. 1397 made it through to 30 day follow up. 1332 made it through to 90 day follow up. Patients were pretty close at baseline.
There were no differences in the ES-5D score of health status (the primary outcome) 0.92 vs. 0.91 which is consistent with non-inferiority. There was no statistical difference found in the pre-planned subgroup analysis of those with appendicoliths, but this may have been underpowered as the difference was 41% vs. 25%.
Within the antibiotic group 20% of patients had an appendicectomy within 30 days and 29% within 90 days.
There’s a great summary here from @skepticalscalpel
So what does this mean?
This trial is consistent with other studies.
I discussed this paper with Ross Fisher, Virchester’s honourary Prof. of Surgery and his observation was that this is a little simplistic. Assuming that there is natural history of whatever “appendicitis” might be is not borne out in practice and the supposition that complex appendicitis’ are all the same is based purely on supposition, and can push clinicians to interpret imaging as almost binary. We know that we can treat appendicitis by simply removing an organ with only prophylactic antibiotic cover and comparing that to actively treating an infection is maybe not as big a question as suggested here.
It’s also worth noting that the follow up period of 90 days is pretty short, and even within that time a significant number of people had to have an operation in the antibiotic group. I would have preferred to see a longer follow up period (perhaps a year or more). This is especially important when you consider the shape of the cumulative mortality curve above. It really doesn’t look as if it’s flattening off, and so I think we should get more evidence on when the rate of surgical intervention starts to flatten off.
My take is that you can treat with antibiotics, but the failure rate is high and a significant proportion of patients will subsequently require a surgical procedure (and thus presumably there will be more ‘illness’ with a conservative approach). I think it’s a conversation that might help shared decision making as I don’t think there is a clear conclusion here. Personally, I’d go for the surgery but that’s arguably because I trust my colleagues and am familiar with surgery. I can quite happily see others come to a different conclusion in very low risk patients.
- JC: Does every appendicitis need admission? St.Emlyn’s https://www.stemlynsblog.org/jc-does-every-appendicitis-need-admission-st-emlyns/
- A Randomized Trial Comparing Antibiotics with Appendectomy for Appendicitis https://www.nejm.org/doi/full/10.1056/NEJMoa2014320
- Antibiotics for Appendicitis — Proceed with Caution https://www.nejm.org/doi/full/10.1056/NEJMe2029126
- SGEM#180: The First Cut is the Deepest – N.O.T. for Paediatric Appendicitis
1 thought on “JC: Antibiotics or Surgery for appendicitis.”
I think what I always take from these studies is the somewhat reassuring idea that in a remote wilderness setting I might get away with treating appendicitis with antibiotics for a week or two and until someone can reach healthcare. (Obviously if evacuation was available that’s still the sensible option but it’s not always that straightforward!)