Article Text

Download PDFPDF
Non-operative management of appendicitis in children
  1. Nigel J Hall1,2,
  2. Simon Eaton3
  1. 1 Department of Paediatric Surgery and Urology, Southampton Children’s Hospital, Southampton, UK
  2. 2 University Surgery Unit, Faculty of Medicine, University of Southampton, Southampton, UK
  3. 3 Developmental Biology and Cancer Programme, UCL Great Ormond Street Institute of Child Health, London, UK
  1. Correspondence to Nigel J Hall, Faculty of Medicine, University of Southampton, Southampton General Hospital, Southampton, SO16 6YD, UK; n.j.hall{at}soton.ac.uk

Abstract

While appendicectomy has been considered the mainstay of treatment for children with acute appendicitis for many decades, there has been a great deal of recent interest in non-operative treatment (NOT) with antibiotics alone. Initial results suggest that many children with appendicitis can indeed be safely treated with NOT and can be spared the surgeon’s knife. Many as yet unanswered questions remain, however, before NOT can be considered a realistic and reliable alternative to surgery. This review summaries current knowledge and understanding of the role of NOT in children with appendicitis and outlines and discusses the outstanding knowledge gaps.

  • appendicitis
  • paediatric surgery
  • evidence-based medicine

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Acute appendicitis affects between 7% and 8% of the worldwide population at some point in their lifetime and as such is one of the most common general surgical emergencies.1 Appendicitis is of particular relevance to children since the incidence peaks in the second decade of life.1 2 Overall, acute appendicitis is diagnosed in 1%–8% of children presenting to the emergency department with acute abdominal pain.3–5 Diagnosis is either clinical or supported by diagnostic imaging (usually ultrasound or CT scan).

Over 130 years ago, Fitz reported appendicectomy for appendicitis,6 and this operation has become the mainstay of treatment of acute appendicitis. Of note, this was several decades prior to the discovery and widespread availability of modern antibiotics. However, in recent years, the dogma that surgery is required has been challenged and there is growing literature to suggest that antibiotics without surgery may be an effective treatment for acute appendicitis in adults7–9 and more recently in children.10–12 This non-operative treatment (NOT) of acute appendicitis in children remains controversial and unproven at present due to a lack of randomised controlled trials (RCTs) and other large prospective evaluations.13

In this article, we focus primarily on the treatment of children with uncomplicated acute appendicitis (UAA). UAA may be defined as acute appendicitis without perforation, development of peri-appendicular abscess or an appendix mass. This form of appendicitis is receiving greatest attention currently in the context of NOT. Of note, the distinction between UAA and more advanced forms of appendicitis may be difficult to make with certainty on either clinical or radiological grounds, and the definitive diagnosis of acute uncomplicated appendicitis may only be apparent on macroscopic and/or histological appearance of the appendix at appendicectomy. We review the role of NOT of acute appendicitis in children based on the existing literature, highlight current gaps in the evidence base and identify key areas for future knowledge generation as we strive to determine the role of NOT when compared with surgery for this common illness.

Historical context

Although appendicectomy is the mainstay of treatment for acute appendicitis in the developed world, the concept of using antibiotics without surgery to treat appendicitis is not a new one. In 1956, Coldrey challenged the dogma that all patients with acute appendicitis required appendicectomy and reported successful NOT in 137 cases with low morbidity and low mortality.14 A report in 1977 of over 400 cases of acute appendicitis claimed high success and low recurrence (7%) with a combination of antibiotics and traditional Chinese medicines.15 In environments without access to surgical care (eg, submariners, merchant navy), appendicitis has been treated non-operatively with antibiotics for many years, with well-documented evidence of high success rates.16 17 Despite this, throughout recent decades, the majority of adults and children with acute appendicitis have been treated surgically, with an appendicectomy.

An anomalous exception to this has been the child presenting with an advanced form of appendicitis known as an appendix mass. An appendix mass is an inflammatory mass centred on the appendix and usually formed by loops of intestine and omentum, which become adherent to the appendix and inflamed. There may be a small abscess within the mass. Up to 9% of children may present with an appendix mass.18 Cases of appendix mass are often treated with antibiotics alone since the risk of damage to these adjacent structures during dissection of the inflammatory mass is high. The majority of cases of appendix mass respond well to intravenous and subsequently oral antibiotics with resolution of symptoms.18 Some children with successful treatment of appendix mass are electively operated later by interval appendectomy, although this is by no means universal, and only 12% of children who have had successful NOT of appendix mass have a subsequent recurrence of acute appendicitis within 1 year.19 Since most surgeons have experience of treating children with appendicitis non-operatively, expertise with NOT of acute appendicitis does currently exist, and it is not, therefore, a completely new concept.

In the mid-1990s, clinicians began to evaluate the role of NOT for UAA as primary therapy in developed healthcare settings.20 Since then, a number of studies have investigated the relative performance of NOT as primary therapy initially in adults with UAA.7 9 21 22 In adults, it is recognised that a high proportion of cases can be successfully treated without an operation.23 However, the evidence is not yet strong enough to support a widespread shift in treatment approach to this disease, and for the time being, appendicectomy remains commonplace.23 Further large-scale evaluations of NOT of uncomplicated appendicitis in adults, particularly in comparison with appendicectomy, are warranted and ongoing.24 Some of the recent trials also include economic analyses that suggested that costs were significantly lower in patients treated with NOT.25

Non-operative Treatment for UAA in children

Investigation of NOT for the management of UAA in children has lagged behind that in adults to some extent but is currently gaining momentum rapidly. A number of studies have to date reported outcomes of children with UAA treated with NOT pathways either in isolation or in comparison with children who have undergone surgery.10 12 26–33 To date, there has been just one prospective RCT (a pilot RCT) reported of this comparative evaluation.28 Existing studies have been of children who were initially selected based on clinical and/or laboratory test parameters (eg, short duration of history, no faecolith on imaging, moderately raised white cell count).30 34 More recently, NOT has been evaluated in a wider spectrum of children who are less selected.35 Perhaps unsurprisingly, NOT failure rates appear higher in less highly selected patient populations. Of note, however, these studies are not RCTs so it is impossible to be sure of the comparative outcomes between NOT and appendicectomy in comparable populations.

Several systematic reviews and meta-analyses have been undertaken summarising the collective evidence to date.36–39 Importantly, these reports have demonstrated that NOT pathways, which include close monitoring to ensure that children treated without surgery are improving satisfactorily, are safe. Furthermore, NOT appears effective as initial treatment in the vast majority of cases. When considering all reports of NOT (comparative and non-comparative), Georgiou et al reported success rates of NOT (defined as discharge home from hospital without appendicectomy during initial hospital admission) in excess of 97%.37 In the most recent systematic review39 reporting NOT for UAA in children, success rate of NOT was 89% or higher in 14 of the 15 included studies but notably only 58% in one report.40 What is less clear, however, due primarily to the absence of well-designed RCTs, is the relative efficacy of NOT when compared with appendicectomy. A number of groups worldwide are currently in the process of performing the systematic evaluations that are necessary to inform treatment choices in the form of RCTs,41–43 while other groups are using alternate prospective methodology.10

In addition to clinical outcomes, there is much interest in the relative costs of NOT and appendicectomy for children using both real patient data44 and financial modelling.45 It is evident that cost-effectiveness will be an important consideration when analysing and interpreting the comparative effectiveness research.

Non-operative treatment for perforated appendicitis in children

A varying proportion of children, but up to 50% in some series46 present with perforated appendicitis. In similarity to uncomplicated appendicitis, the treatment of perforated appendicitis has traditionally been appendicectomy, but it is now recognised that cases of perforated acute appendicitis can also be treated safely and effectively with antibiotics alone or, in cases of intra-abdominal abscess, with antibiotics and percutaneous image-guided abscess drainage.

Two RCTs have evaluated NOT and appendicectomy for perforated appendicitis in children,47 48 and a recent review evaluated patient-level data from these two RCTs.49 Current data support emergency appendicectomy in children with perforated appendicitis in whom no intra-abdominal abscess is present, but in the presence of an abscess at presentation, comparable outcomes can be achieved with either NOT or surgery.49

Improving the evidence base: what we need to know

If there is a future for NOT in children, then there are a number of knowledge gaps to be filled before appropriate evidence-based decisions can be taken by surgeons, commissioners, parents and children. These include:

  • Are there any disadvantages to removing the appendix?

  • If an operation is not performed, what is the likelihood of the appendix perforating or other serious complications occurring, either during the first admission or subsequently?

  • If the child has successful resolution of symptoms following NOT, what is the likelihood of appendicitis recurring, either in the short or long term?

  • Is it possible to identify those children who will respond successfully to NOT or those that will not respond and potentially perforate if not operated?

  • What will be the psychological impact of not having the appendix removed—will some parents and children regard the remaining appendix as a ‘ticking time bomb’ that might affect their future activities (eg, taking holidays or later working in medical-resource-poor settings)?

  • Is NOT for appendicitis cost-effective? Although it might be assumed that not operating makes treatment cheaper, children are usually discharged from hospital quickly following appendectomy, whereas children undergoing NOT may stay in hospital longer, so that this assumption may not be true.

  • What if the diagnosis is not appendicitis?

Surprisingly, for a common disease, we do not know the answer to two more fundamental questions that underpin several of those questions above. These are: what is the biological function of the appendix, and what is the pathogenesis of acute appendicitis?

Darwin and many others have regarded the appendix as useless, but recent evidence suggests that the appendix has evolved at least 32 different times during mammalian evolution50 and that the function of the appendix during human evolution is to act as a reservoir of beneficial microbiota to rapidly recolonise the intestine following diarrhoeal infections.50 The pathogenesis of acute appendicitis involves an interaction between the microbiota, the immune response and blockage of the appendix, by a faecolith, for example, but we are still far from a complete understanding of the pathogenesis.51 Might there be an advantage to leaving the appendix in place? Complications following appendectomy are rare, and although there are diseases when the appendix is useful as a surgical conduit, these are even more rare and cannot justify non-removal of the appendix. By removal of the appendix, however, we are removing an immune organ and potentially a reservoir of beneficial microbiota. This may predispose to subsequent development of other pathologies, such as recurrent Clostridium difficile infection,52 Crohn’s disease53 and colorectal cancer,54 although this is a controversial area and it is difficult to dissect appendicitis from appendectomy and to control for diagnostic uncertainty.

As the pathogenesis of acute appendicitis is not completely elucidated, the progression of disease from uncomplicated appendicitis to complicated appendicitis with perforation and peritonitis is uncertain. While it has traditionally been assumed that uncomplicated appendicitis will progress to complicated appendicitis without appropriate treatment, there is increasing evidence that this may not be the case. It is possible that there are instead two (or perhaps more) distinct pathological entities and that some cases of appendicitis are ‘destined’ to become complicated, while others are not.55 This is supported by a number of studies that demonstrate no detrimental effect related to in-hospital delays after a diagnosis of appendicitis while waiting for appendicectomy.56 57 It may be that uncomplicated and complicated appendicitis are effectively two separate entities.

In an ideal world, we would be able to select children who could be predicted to respond to NOT and treat them without surgery while reserving surgery for the remainder. At present, we are unable to make this prediction. Although a number of scoring systems exist to assist with diagnosis of acute appendicitis (eg, Alvarado score and the Paediatric Appendicitis Score58), none can currently inform us which children are likely to respond to NOT, which children are not going to respond to NOT and which children may be at risk of perforation. Only when we have large datasets from the prospective studies described above might we be able to develop these systems.

At the same time as generating these large datasets, it is imperative that qualitative work regarding the impact of NOT on children and their families and full cost-effectiveness and cost-utility evaluations are undertaken so we can understand fully the wider implications of NOT beyond the immediate clinical outcomes.

Finally, we must consider diagnostic certainty if we are to pursue NOT in a large number of children. Is there a chance that a child presumed to have appendicitis is treated ‘successfully’ with NOT but actually has a more sinister diagnosis? The most important underlying diagnosis is likely that of a carcinoid tumour. Previous studies of children who present initially with an appendix mass and treated initially with NOT found an incidence of carcinoid tumour of <1%,59 60 which is within the range of overall incidence in the general population of developing a carcinoid tumour at any site.61 With this and other potential diagnoses (eg, inflammatory bowel disease) in mind, it would seem prudent to restrict NOT to the first episode of UAA, proceeding with further investigation and possibly appendicectomy, for recurrence or residual symptoms.

Challenges of designing RCTs

RCTs in surgery are always challenging, particularly when the treatments being compared are markedly different. NOT and surgery represent conceptually very different treatment strategies for both surgeons and parents. This, combined with the fact that RCTs need to recruit children in an emergency setting, means that recruitment is likely to be challenging. While these challenges may be surmountable, particularly with the use of qualitative research techniques of proven efficacy,43 62 there remain methodological challenges, primarily related to selecting an appropriate primary outcome.

RCTs comparing NOT and appendicectomy are likely to be designed to demonstrate either superiority of one treatment over the other or non-inferiority of NOT compared with appendicectomy. A statistical comparison must be made between the incidence of a common primary outcome in each treatment group. This is precisely where the problem arises since the outcomes of primary interest in each group are not identical. The main outcome of interest in individuals allocated to appendicectomy is likely complications of surgery. Many also believe that negative appendicectomy (ie, removal of an appendix in a child who did not actually have appendicitis) is equally important since this would represent surgery that had been performed unnecessarily.41 The negative appendicectomy rate in in the UK is approximately 10% for children treated in specialist paediatric surgical centres and 24% for children treated by adult general surgeons63 but in some centres in which all children have a radiological diagnosis (ultrasound with CT for equivocal findings) may be as low as 1%–2%. Inclusion of negative appendicectomy in a composite primary outcome for an RCT will likely have significant impact on the ‘success rate’ of a surgical treatment arm and thence overall sample size. Similarly, the negative appendectomy rate in centres involved in an RCT will have a strong influence on sample size. In children treated with NOT, failure of NOT to adequately treat appendicitis is likely the most important outcome, but another important factor to consider is need for later appendicectomy for recurrence. If a high proportion of children treated with NOT develop recurrent appendicitis, there is little benefit to an initial course of NOT. Bearing these factors in mind, what then should be the primary outcome for future RCTs? How should ‘failure of treatment’ be defined—what might seem like success to a surgeon might be deemed a failure to a child or their parent.

Current examples of primary outcomes in RCTs completed or in progress are all composite primary outcomes, including various combinations of those individual outcomes detailed above.28 41 42 It remains unknown, however, what surgeons, parents and other stakeholders believe should be measured and reported in RCTs of this nature. We are currently undertaking work to develop a core outcome set for the treatment of children with UAA which will hopefully enlighten this debate and inform our future work.43 64

Conclusion

Current interest in NOT for acute appendicitis in children is at a high level, and efforts are in place to improve the evidence base and importantly to define the precise role of NOT for children with appendicitis. Until this work has been completed and the evidence base improved, we would caution against any change in current practice. In particular, we would not encourage a diagnosis of appendicitis to be made without thorough assessment by a surgeon, nor for NOT to be instigated outside the closely monitored environment of a research study, ideally an RCT.65 We also caution against ‘therapeutic creep’ towards treatment of abdominal pain (not appendicitis) with courses of antibiotics.

If NOT proves to be effective, then the treatment of acute appendicitis may be revolutionised. More likely, we suspect that the data will suggest that some children can be successfully treated without surgery. The challenge will be to define this population precisely so that NOT can be offered as a realistic alternative to them while reserving surgery for those who really need it.

Acknowledgments

SE is supported by Great Ormond Street Hospital NIHR Biomedical Research Centre and by Great Ormond Street Children’s Charity.

References

Footnotes

  • Contributors SE and NJH contributed equally to the writing of this manuscript, including drafting of the initial manuscript and its subsequent revision.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; externally peer reviewed.