Wednesday, April 30, 2014

The End of Appendectomies?

We’ve seen dogma challenged regarding diverticulitis and the necessity of antibiotics.  This isn’t the first post regarding a change in initial strategy for appendicitis, however, it’s certainly reasonable to revisit again as the evidence accumulates.

This study is simply a prospective, observational case series of 159 patients with acute, uncomplicated appendicitis.  In 2010, this institution in Italy made surgical appendectomy the exception, rather than the rule.  Patients without serious illness or complicated appendicitis were admitted for short term observation and started on amoxicillin-clavulanate.  Patients who failed to improve or worsened went to the OR.  Others were discharged and re-examined at 5-7 days as an outpatient, and, again, those without significant improvement went to the OR.  Over a 2 year follow-up period patients were assessed by phone.

Within 7 days, there were 19 (12%) treatment failures; 17 of 19 were acute appendicitis, 2 were tubo-ovarian abscess with secondary appendiceal inflammation.  Over the 2 year follow-up, 22 (13.8%) patients had recurrent appendicitis – 14 of which were managed with antibiotics without complication.  8 went to the OR, 6 of which were confirmed as acute appendicitis.

I don’t think we’d have the same issue with misdiagnosed TOAs in our population – 73% of their diagnoses were by ultrasound, and only 17% underwent CT.  12% short-term treatment failure is also nothing to scoff at – and this number is consistent with other studies.  Routine surgery, however, is much costlier, resource-intensive, and carries with it a similar or greater risk of major complications.  It seems to me this is absolutely a viable strategy.

Is it time surgery added “Consider a trial of antibiotic therapy prior to surgery for acute, uncomplicated appendicitis” to their Choosing Wisely list?

“The NOTA Study (Non Operative Treatment for Acute Appendicitis)”


  1. If you look at this paper more carefully, you will find the following:

    These figures are from Table 3 of the paper
    US done 116 (73%)
    US positive 88 (76%)
    CT scan done 27 (17%)
    CT scan positive 21 (78%)
    Clinical diagnosis only of acute appendicitis 16 (10%)

    The authors do not explain why patients with negative ultrasounds and CT scans were included in the cohort of nonoperatively treated patients with appendicitis. If they were going to disregard the results of the imaging studies, they shouldn't have done them in the first place.

    Including the 16 patients with clinical diagnoses only and the 28 with negative ultrasounds and 6 with negative CT scans, a total of 50 (31%) of the patients may not have actually had appendicitis.

    They say that 19 patients failed nonoperative management. If you divide those 19 patients by the 109 true cases of acute appendicitis, you get a failure rate of 17%.

    The study is not quite as convincing as it first seems.

    There is a randomized prospective trial of observation vs. surgery ongoing in Finland. I hope it clarifies the situation.

    1. Great points. I agree their imaging strategy leaves a little ambiguity regarding gold standard for diagnosis – and the failure rate could very well be as high as 17%. The true answer is probably somewhere in between – a "negative" ultrasound does not rule out appendicitis, and these authors were more concerned about excluding folks with complicated (abscess, perforation, etc.) appendicitis than they were with imaging confirmation given an otherwise convincing clinical picture. But, you're absolutely right to criticize their gold standard – and to point out the effect that has on their numbers.

      Looking at the forest for the trees, however – their comprehensive literature review in the introduction details a reasonable number of observational and randomized trials with similar numbers. I'm sure we won't have consensus re: optimal management, trial of antibiotics vs. surgery, but I think we've reached a point where we probably ought to offer a trial antibiotics as part of informed consent as an "alternative" to surgery, and be able to reasonably describe the complication rates of this alternative.

  2. the only concern I have with making practice changes from these studies is the absence of really long term follow up. For appendicitis, to be able to give patients a properly informed choice they would need to know at the very least the recurrence rates at at least 10 years and probably out to 20, even 30 years. In fact the long term follow up of relevance is not know as we don't exactly know how quickly the recurrence rate drops off over time eg if it was say 20% in the first 10 years, 5% in the next 10 and then 1% in the next 10 then yeah 20 years follow up data might be good enough. Sure we could possibly imply these rates based on age related incidence of appendicitis but this might not correlate to individual recurrence rates eg a person prone to getting appendicitis once might be very likely to recur v's someone who's never had appendicitis at the same age.
    Such long term follow up data is obviously hard to study and perhaps we will never see it but surely you need that to make an informed decision. eg if the recurrence rate over 20-30 years was 50% I'd probably want my appendix taken out now assuming I was healthy and at low risk from surgery.
    Arguably then an antibiotic v surgery strategy is perhaps most reasonable within the bounds of an RCT which hopefully seeks to answer these long term follow up questions.

    1. Agree; 2 year follow-up is a nice start, but long-term follow-up would be more helpful in terms of true recurrence. On the flip side, however, that same long-term interval follow-up should be applied to surgical appendectomy, along with operative complications (hernia, SBO, recurrence) for a true comparison.

  3. I work have a co-worker whose son, 5 years old, had been suffering from side flank pain for a week or so. She took him to the emergency room and the emergency room staff checked him out, said he may have a stomach flu (it was right during the peak of flu season) and sent him home with some Tylenol. Three days later his appendix burst, she rushed him to the hospital (a different one, thank goodness) where he spent the next 10 days, they kept him to make sure the antibiotics flushed out all the germs that had gotten free because of the appendix. It's unfortunate that he had this issue when the flu was so rampant but the ER should have done a much better job than that, even an ultrasound may have picked up an inflamed or abnormal appendix in a child whose family has a history of appendix troubles.

    1. Can't comment on the specifics of any individual case, other to say appendicitis can be a difficult diagnosis to make, confounded by many other symptoms and epidemiologic factors, and counterbalanced by the need to be mindful of imaging studies and resource utilization. Hope your ultimate outcomes were OK.


Comments on posts > 10 days old will be moderated; blame spammers.