Wednesday, March 15, 2017

Nonoperative treatment of appendicitis in children: Is it safe?

After writing my 21st post about appendicitis back in November, I swore I would not write about it again for the foreseeable future.

Well, the future is now because investigators from the United Kingdom and Canada just published a meta-analysis including 10 papers and 413 children about the efficacy and safety of nonoperative treatment for appendicitis in children.

They concluded that nonoperative management is effective in 96% of children with acute uncomplicated appendicitis during their initial hospitalizations with just 17 (4%) children requiring appendectomy before discharge. An additional 68 (16.4%) developed recurrent appendicitis later, and 19 of these patients were treated with the second course of antibiotics. The other 49 underwent appendectomy with histologic evidence of recurrent appendicitis.

Another 11 patients underwent appendectomy in the follow-up period for various reasons. In all, 77 (18.6%) patients initially treated with antibiotics eventually underwent appendectomy.

Although the initial hospital length of stay for appendectomy was shorter than that of patients treated with antibiotics, complication rates were similar.

These findings were met with headlines like "Antibiotics, not surgery, could treat appendicitis in children, study suggests" from The Guardian and "Is Surgery Always Needed for Kids' Appendicitis?" from US News.

What are the problems with this paper?

After searching the literature, the authors identified the 10 studies they deemed suitable—7 prospective and 3 retrospective. Four involved only nonoperative treatment. It is unclear how one could perform a meta-analysis comparing the nonoperative treatment of appendicitis with antibiotics to appendectomy if 40% of the studies did no such comparison.

Of the 10 papers, only one was a randomized controlled trial, and it was a pilot study in which patients, parents, and surgeons were not blinded regarding treatment allocation. The lack of blinding obviously taints the results. The proper way to do this study is to blind the patients, the parents, and the surgeons by obtaining consent for enrollment before knowing which treatment group each patient was randomly assigned to.

The authors of the meta-analysis also pointed out that the randomized trial "was not powered to provide definitive evidence of the efficacy of nonoperative treatment vs. appendectomy."

In the five other studies, surgery and antibiotics were discussed with the parents, and they were allowed to choose which treatment their child would undergo, hardly an objective research method.

Other than saying it varied, the meta-analysis was vague regarding how the diagnosis of acute uncomplicated appendicitis was made in each of the reviewed studies. An unknown number of children may have been diagnosed by clinical criteria only. Therefore, some may not have had appendicitis at all. If they never had appendicitis, they were not likely to get a recurrence no matter what the treatment was.

The combined average length of follow-up for patients in the 10 studies is not stated but ranged from as little as 2 months to 51 months, and seven of the studies had follow-up lengths of less than 2 years.

The authors had an epiphany about this stating, "Although we have not formally analyzed it, we noted a tendency for long-term efficacy to be lower in studies with longer duration of follow-up." Short duration of follow-up is an issue with all studies of nonoperative treatment of appendicitis.

No studies used the same intravenous antibiotic regimen and the duration of IV antibiotics was any one of the following: 2 doses, 24 hours, 48 hours, 72-120 hours, until abdominal tenderness resolved, and until C-reactive protein was less than 5 mg/dL. The oral antibiotic protocols were almost as varied.

I can agree with one conclusion of the meta-analysis. The authors called for larger and proper randomized controlled trials and said, "Until such studies are completed, we would recommend that nonoperative treatment of children with acute uncomplicated appendicitis be reserved for those participating in carefully designed research studies."

This should ease the conscience of any surgeon not involved in a randomized controlled trial who believes appendectomy is still the treatment of choice for children with appendicitis.

19 comments:

William Reichert said...

OK.Time to get serious. So which method of care, surgery first or antibiotics first cost the least?

Skeptical Scalpel said...

As far as I know, cost has not been completely addressed. No doubt antibiotic treatment costs less at first. But if 25% of adults initially treated with antibiotics need surgery for recurrent appendicitis within a year or two, there may be no cost difference on average.

artiger said...

Let's not forget risks of C diff and increasing antibiotic resistance either.

Welcome back Scalpel.

Anonymous said...

I am glad to see that they all concurred with the principle of using a prime number of days as the duration of antibiotics.

Korhomme said...

Cost of treatment is not a (direct) problem so far in the UK.

What, SS, would you now do and recommend?

Skeptical Scalpel said...

Artiger, yes and C diff is not likely to be reported as a complication.

Anon, way to recognize the prime numbers :-).

Kor, I am retired, but if I was still practicing, I would tell patients the following:

Antibiotics have been used to treat uncomplicated appendicitis.
18-27% of patients treated with antibiotics will experience a recurrence within 2 years. Beyond that time, no one knows the recurrence rate.
The treatment is experimental.
The type and doses of antibiotic(s) have not been agreed upon.
I recommend a laparoscopic appendectomy. The operation is safe and cures the disease.

artiger said...

As far as nonoperative treatment of appendicitis, we might could call it safe, but I would question effective.

Oldfoolrn said...

Nonoperative treatment of appendicitis could be complicated by rupture. Not so safe.

Skeptical Scalpel said...

Artiger, I agree.

Old, You may be right but so far that has not been a big problem. Some experts feel that perforated appendicitis is a different entity than uncomplicated appendicitis which usually does not progress to perforation.

Oldfoolrn said...

Interesting. Thanks for enlightening me. I'm about 20 years behind current thinking!

Anonymous said...

Do you think there would be any other potential long-term complications other than recurring appendicitis caused by the nonoperative treatment as opposed to the permanent 'cure' of the appendectomy? If so, what then makes the nonoperative treatment beneficial compared to the appendectomy?

Skeptical Scalpel said...

Anon, I am not sure that there would be any other complications besides the recurrence of appendicitis. If the recurrence was treated with antibiotics again, then the patience would still have the risk of a third or fourth attack.

artiger said...

Anon at 2:40pm, occasionally appendicitis is the result of a tumor, such as a carcinoid or even adenocarcinoma. All the antibiotics in the world are not going to kill a tumor. So yeah, there are long-term complications, because sometimes you don't discover a tumor was involved until you read the pathology report.

Skeptical Scalpel said...

Artiger, yes tumors are possible, but they are found in <1% of specimens removed at appendectomy. I don't believe a strong case can be made that appendectomy should be done to prevent advance cancer of the appendix.

However, it will be interesting to see what happens when the first lawsuit for delay in diagnosis of cancer of the appendix after nonoperative treatment of appendicitis is filed.

artiger said...

Scalpel, I predict an award of at least 8 digits, perhaps 9.

Anonymous said...

Difficult to see here why lack of blinding to assignment would be likely to affect results (except inasmuch as, in some studies, every antibiotic user with recurrent abdominal pain was hustled in for appendectomy, which obviously was not done for pain in the appendectomy group). I presume you do not want everyone to be given sham surgery.

There is no blood-appendix barrier and therefore no a priori reason that appendicitis should always require surgical treatment any more than colitis or pneumonia. The cultural belief that it does is a historical contingency based on the fact that appendectomy was a simple enough surgery that it could be invented before antibiotics.

Second, there is a belief that the appendix is useless. It is not. It helps repopulate the gut with necessary flora after illness, so if in future you are going to travel to or live in places where one gets diarrhea a lot, you will benefit from having an appendix. You will also be at lower risk of C.diff. following antibiotic treatment, which is not meaningless. You probably can't envision being sued for someone's C.diff. a decade down the line because our culture favors "Doing Everything" and will almost never punish you for it. That doesn't mean that those harms aren't equally significant to patients.

Skeptical Scalpel said...

Anon, you are correct in that sham surgery would not be appropriate in the setting. The study needs to be blinded because if parents or doctors conducting the trial can decide to which group the child is assigned, kids who look sicker would be more likely to be enrolled in the surgery arm of the study.

As I wrote back in 2014, a study from Ohio State [http://skepticalscalpel.blogspot.com/2014/05/antibiotics-instead-of-surgery-for.html] had this very problem. I wrote “The nonrandomized nature of the ... study created imbalances in the cohorts as 6 (13%) of the 47 patients who underwent surgery had complicated appendicitis (2 with gangrenous and 4 with perforated appendicitis), compared to no instances of complicated appendicitis in the nonoperative group.”

The subject of whether the appendix is useful or not is debatable. For example, several studies have shown that appendectomy is protective against the development of ulcerative colitis.

Geboes K: Appendiceal function and dysfunction: what are the implications for inflammatory bowel disease? Nature Clin Pract (Gastroenterology and Hepatology) 2:339, 2005. 16265384
Radford-Smith GL, Edwards JE, Purdie DM, et al: Protective role of appendicectomy on onset and severity of ulcerative colitis and Crohn's disease. Gut 51:808, 2002. 12427781
Andersson RE, Olaison G, Tysk C, et al: Appendectomy and protection against ulcerative colitis. N Engl J Med 344:808, 2001. 11248156
Frisch M, Pedersen BV, Andersson RE: Appendicitis, mesenteric lymphadenitis, and subsequent risk of ulcerative colitis: cohort studies in Sweden and Denmark. BMJ 338:716, 2009. 19273506
Matsushita M, Takakuwa H, Matsubayashi Y, et al: Appendix is a priming site in the development of ulcerative colitis. World J Gastroenterol 31:4869, 2005. 16097061

I don’t understand your comment about antibiotics. You say “you will also be at lower risk of C. diff following antibiotic treatment.” Since the number one cause of C. diff colitis is antibiotic treatment, I fail to see how exposing more people to courses of antibiotics lasting 7-10 days will result in fewer cases of C. diff colitis.

Luqman said...

I believe he meant the risk of C. diff will be reduced in the future due to preservation of the appendix.

Skeptical Scalpel said...

Luqman, I got that. My response was that exposing large numbers of people to a 7-10 day course of antibiotics to treat a disease that is cured by surgery might also cause C. diff colitis. In fact, it might lead to even more cases than those found in people without appendices.

Post a Comment

Note: Only a member of this blog may post a comment.