Wednesday, 23 May 2012

Appendicitis - is surgery necessary?

There is one thing (out of a list of many) that makes me disappointed with a surgical trainee; it comes after they describe a new case to me and offer their preferred surgical treatment. I then ask them for the evidence supporting their recommendation. They say: “Well, I saw a guy do one once.” This short statement says so much. Firstly, how we are influenced by what we see, particularly when somebody considered to be senior or authoritative does it. It also shows how readily we recommend treatments without good knowledge of the outcomes of that treatment, or of the alternatives. It is easier just to think: “If this guy did it, then it must be OK”.

This is why appendicectomy is so commonly done. Randomised trials have told us that removing the appendix is not necessary on first presentation, and it is associated with a worse long term outcome. Yet if you present to any of my hospitals with suspected appendicitis, you are unlikely to be leaving hospital without having your appendix removed.

Now a recent review published in the British Medical Journal has summarised the randomised trials that compared immediate appendicectomy to antibiotics and observation for patients presenting with uncomplicated appendicitis. There were four studies involving a total of 900 patients. They concluded that the overall complication rate was significantly lower in the group initially treated non-operatively. It seems that we may have been overestimating the benefits of having an appendix removed straight away, and we might also have been underestimating the harms from the surgery (such as infections and adhesive bowel obstructions). What a surprise.

The fact that some patients later had an appendicectomy does not alter the results of initially treating them non-operatively. The bottom line is that most appendicectomies can be avoided, resulting in and overall reduction in the complication rate.

On questioning surgical colleagues I come up against unscientific, emotive responses like “What if it was your child?” The answer of course, is that I would want them to have antibiotics and observation rather than an immediate appendicectomy.

It often boils down to comments like “But this is what we do here” from my colleagues. That is the attitude that kept venesection (blood letting) going for a thousand years, and another hundred years after it was shown to be ineffective. It is time for tradition to take a back seat to scientific evidence.


  1. So what can be done to educate the waves of surgical trainees every year when the very surgeons that guide their training and instill practices are averse to change?

  2. Thanks bonedocjr,
    You have hit the nail on the head: junior surgeons are being taught by older surgeons in a typical "apprenticeship" model, where they learn what the older surgeon does, then they go off and do it themselves. There is an emphasis on technique, and applying that technique, not on gathering and critically evaluating the scientific evidence around this, particularly when it comes to WHEN to apply those techniques (and of course, when NOT to apply those techniques).
    The answer is to educate the established surgeons, and the trainees, in awareness of the problem and how to be better scientists themselves. For trainees, we need to build a better academic framework around the training program both officially (by building it into the curriculum and more importantly, the final exams) and unofficially, by building the culture of rational practice through evidence based medicine.
    You may look despairingly at the training programs now, but I can tell you that they are much more evidence-based than 20 years ago when I went through, and they will be even better in another 20 years. Science based practices (like medicine) will always evolve to be more rational, but like evolution it takes time.
    You can play your part in this evolution by asking the right questions and sticking to scientific principles.

  3. Absolutely, I am of the strong opinion that surgery should be based on evidence. I was discussing the article with a couple of emergency medicine registrars (my current placement) and they echoed the same views (almost in union): "thats just one paper...its probably wrong...they didnt measure all the outcomes...i remember reading it a while back there were flaws...these patients will get an operation anyway so we should operate at the start than delay the inevitable"
    This is the problem with medicine: in medical school we are taught to be skeptical about research, and we develop a strong skepticism to any research that contradicts years of the point of distrust. The problem isn't just at the trainers, but seeps down into trainees and medical school education. Much work needs to be done if we are to move away from non-evidence based routine practices!

  4. Thanks again bonedocjr,
    It is OK to criticise research, but it needs to be done scientifically. There is no such thing as a perfect study, but criticism of one study makes the next study better, and opens our eyes to potential errors.
    The problem with your colleagues is that they are fitting the literature to fit their world view (prevailing bias) by dismissing the good trials and accepting the poorer studies on face value.

  5. For an alleged skeptic, you seem to have accepted the study without any reservations. I did not. The study is flawed on many levels. See my blog for details. Http://

    1. Thanks Skeptical Scalpel,
      I refer readers to your post on this.
      It is pretty clear that there is non-operative treatment is less harmful than we thought, that it is likely to obviate the need for surgery, and that surgery is associated with complications. I don’t think that this is what patients are told when they come in to my hospitals, and they are not given much time to think about it before having their appendix out. This is the point of my blog: whatever the true difference is between antibiotic therapy and immediate surgery (and it is pretty clear that we can place varying importance on different aspects of this), the benefit of surgery over non-operative treatment is at least overestimated. We need to give the patients the real information and they can make their own decision; where I work, they are basically told that surgery will be unlikely to cause problems later, and they will not get better without it.
      I also think that the idea of adhesions not being a problem in laparoscopic procedures sounds good, but like a lot of things that are biologically plausible, they are not as clear as you would expect. The adhesion rates after the CLASICC trial would be an example.
      Also, surgeons here will remove a normal appendix if found at the time of surgery. Given that the appendix may not be the “junk” body part we thought it was, why do we do this if it was not causing the problem?
      My other point is that if excision is necessary for acute appendicitis, how come we don’t do it for acute diverticulitis?
      Sorry to have raised more questions than I answered, but that is usually the case.

  6. You choose to ignore the relapse rate after antibiotic treatment as if it is not an issue. Symptomatic adhesions are rare after even open appendectomy. Just as acute diverticulitis is a completely different disease than acute appendicitis and therefore is treated differently, so too is the rate of adhesions and complications after colon resection (CLASICC Trial) compared to appendectomy.

  7. Thanks Skeptical Scalpel,
    I was not ignoring the recurrence rate; I am happy to discuss it. The recurrence rate is important but it should be noted that recurrence can also be treated with antibiotics. Also, your “0%” recurrence rate after surgery is simplistic: if a normal appendix is removed, you have not cured the cause of their abdominal pain, and when they re-present with abdominal pain from IBD or whatever, they are off your list of recurrent appendicitis, but they still aren’t cured. Further, when provided with data on the recurrence rate, many patients will still choose non-operative treatment.
    I think we agree on the data, but have different conclusions. That is because we are asking different questions.
    Your opinion (correct me if I am wrong) is that removing the appendix is a quick and safe procedure, and if it is often needed later anyway, why not just remove it initially? This is a reasonable conclusion from the data, and largely answers the question: Which is the most definitive procedure that will stop me from coming back?
    My question is: Is surgery necessary at initial presentation? My conclusion is: No, and that non-operative treatment is a reasonable treatment option. Not only is this a reasonable conclusion, it is also big news. Most surgeons do not know this, and if they do, they certainly don’t pass it on to the patients. I also feel that some surgeons take your opinion (which, as I said, is reasonable) and somehow conclude that all cases of appendicitis MUST be treated with surgery, immediately.
    A surgeon at my institution recently pulled out the Parachute Analogy for this topic, such is the vehemence with which appendicectomy is defended. I have a good response for that, but you will have to wait for a future posting. Stay tuned.

    1. This comment has been removed by the author.

  8. In most centers, including mine, the rate of finding a normal appendix at surgery is <5% thanks to CT scanning.

    Please remember that while the antibiotics are taking effect, the patient is suffering. When I remove an appendix laparoscopically, the patients awaken feeling much better with far less pain from the small incisions than they had preoperatively.

    If the treatment of appendicitis had first been antibiotics for years and then appendectomy had been invented, the procedure would have been hailed as a major advance. I can't imagine why anyone would choose a treatment that was only effective 60% of the time and risk the recurrence of a painful and possibly complicated condition that is so easily, safely and effectively treated with surgery.

    1. ... because the best available evidence tells us that we will have less complications, or at least be no worse off with non-operative treatment.
      Based on the evidence available, including the chance of recurrence, presence of pain etc, I would choose non-operative treatment. You would choose surgery. Patients should be allowed to make that decision after being provided with the relevant information. As a surgeon, you know that patients are happiest when they have been involved in the decision making, even when complications occur.
      The argument is similar for acute disc prolapse: no long term difference between operative and non-operative treatment, except surgery relieves the pain immediately. The procedure is quick, minimally invasive and safe, but there is a similar argument about complications from surgery. Yet many people I know personally have chosen to put up with the pain until it settled, and are happy that they have avoided surgery.

  9. This comment has been removed by the author.