See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/330856125

Antibiotic Treatment and Appendectomy for Uncomplicated Acute

Appendicitis in Adults and Children: A Systematic Review and Meta-analysis

Article  in  Annals of Surgery · January 2019

DOI: 10.1097/SLA.0000000000003225

CITATIONS READS

45 1,016

9 authors, including:

Mauro Podda Nicola Cillara

Azienda Ospedaliera Universitaria Cagliari Azienda Sanitaria Locale Cagliari
141 PUBLICATIONS   1,093 CITATIONS    35 PUBLICATIONS   193 CITATIONS   

SEE PROFILE SEE PROFILE

Nicola Fearnhead Carlos Augusto Gomes

Cambridge University Hospitals NHS Foundation Trust Federal University of Juiz de Fora
111 PUBLICATIONS   2,113 CITATIONS    104 PUBLICATIONS   2,042 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Acute Appendicitis View project

IMAGINE View project

All content following this page was uploaded by Mauro Podda on 17 January 2020.

The user has requested enhancement of the downloaded file.

 CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

REVIEW PAPER

Antibiotic Treatment and Appendectomy for Uncomplicated

Acute Appendicitis in Adults and Children
A Systematic Review and Meta-analysis
Mauro Podda, MD,  Chiara Gerardi, Pharm D,y Nicola Cillara, MD,z Nicola Fearnhead, MD, FRCS,§
Carlos Augusto Gomes, MD, PhD,ô Arianna Birindelli, MD,jj Andrea Mulliri, MD, 
Richard Justin Davies, M Chir, FRCS,§ and Salomone Di Saverio, MD, FRCS§

Methods: Systematic literature search was performed using MEDLINE, the

Objective: The aim of this meta-analysis was to summarize the current
Cochrane Central Register of Controlled Trials, and EMBASE databases for
available evidence on nonoperative management (NOM) with antibiotics
randomized and nonrandomized studies comparing antibiotic therapy (AT)
for uncomplicated appendicitis, both in adults and children.
and surgical therapy-appendectomy (ST) for uncomplicated appendicitis.
Summary Background Data: Although earlier meta-analyses demonstrated
Downloaded from http://journals.lww.com/annalsofsurgery by BhDMf5ePHKbH4TTImqenVE17rCHCwnx1lzypUBRFIR8A0mbBsAR5WcBfuwptWMaUO4z+K8EYojE= on 02/06/2019

Literature search was completed in August 2018.

that NOM with antibiotics may be an acceptable treatment strategy for
Results: Twenty studies comparing AT and ST qualified for inclusion in the
patients with uncomplicated appendicitis, evidence is limited by conflicting
quantitative synthesis. In total, 3618 patients were allocated to AT (n ¼ 1743)
results.
or ST (n ¼ 1875). Higher complication-free treatment success rate (82.3% vs
67.2%; P < 0.00001) and treatment efficacy based on 1-year follow-up rate
(93.1% vs 72.6%; P < 0.00001) were reported for ST. Index admission
From the Department of General, Emergency and Robotic Surgery, San Francesco antibiotic treatment failure and rate of recurrence at 1-year follow-up were
Hospital, Nuoro, Italy; yCentro di Politiche Regolatorie del Farmaco, IRCCS – reported in 8.5% and 19.2% of patients treated with antibiotics, respectively.
Istituto di Ricerche Farmacologiche ‘‘Mario Negri’’, Milano, Italy; Rates of complicated appendicitis with peritonitis identified at the time of
zDepartment of Surgery, Santissima Trinità Hospital, Cagliari, Italy;
§Cambridge Colorectal Unit, Addenbrooke’s Hospital, Cambridge University surgical operation (AT: 21.7% vs ST: 12.8%; P ¼ 0.07) and surgical
Hospitals NHS Foundation Trust, Cambridge, UK; ôDepartment of Surgery, complications (AT: 12.8% vs ST: 13.6%; P ¼ 0.66) were equivalent.
Hospital Universitario Terezinha de Jesus, Universidade Federal de Juiz de Conclusions: Antibiotic therapy could represent a feasible treatment option
Fora, Juiz de Fora, Brazil; jjTrauma Surgery Unit, Maggiore Hospital Regional for image-proven uncomplicated appendicitis, although complication-free
Emergency Surgery and Trauma Center, AUSL Bologna Local Health District
Bologna, Bologna, Italy; and Department of Digestive Surgery, University treatment success rates are higher with ST. There is also evidence that
Hospital of Caen, Caen cedex, France. NOM for uncomplicated appendicitis does not statistically increase the
Contributions of authors: M.P.: Study conception and design, literature search, perforation rate in adult and pediatric patients receiving antibiotic treatment.
acquisition, interpretation and analysis of data; drafting and critically revising NOM with antibiotics may fail during the primary hospitalization in about 8%
the article for important intellectual content; and final approval of the version
to be published. C.G.: Study conception and design, literature search, acquisi- of cases, and an additional 20% of patients might need a second hospitaliza-
tion, interpretation and analysis of data; drafting and critically revising the tion for recurrent appendicitis.
article for important intellectual content; and final approval of the version to be
published. N.C.: Interpretation and analysis of data; drafting and critically Keywords: acute appendicitis, antibiotic therapy, appendectomy,
revising the article for important intellectual content; and final approval of the conservative treatment, meta-analysis, nonoperative management,
version to be published. N.F.: Interpretation and analysis of data; critically uncomplicated appendicitis
revising the article for important intellectual content; and final approval of the
version to be published. C.A.G.: Interpretation and analysis of data; critically (Ann Surg 2019;xx:xxx–xxx)
revising the article for important intellectual content; and final approval of the
version to be published. A.B.: Interpretation and analysis of data; critically
revising the article for important intellectual content; and final approval of the
version to be published. A.M.: Interpretation and analysis of data; critically
revising the article for important intellectual content; and final approval of the
version to be published. R.J.D.: Interpretation and analysis of data; drafting
A cute appendicitis remains one of the most common causes of an
acute abdomen and is the commonest surgical disease among
adult and pediatric patients presenting to emergency departments,
and critically revising the article for important intellectual content; editing and both in the United States and Europe.1
revising the English for the final version to be published; and final approval of
the version to be published. S.D.S.: Study conception and design, literature
In North America, the incidence is 100 per 100,000 person-
search, acquisition, interpretation and analysis of data; drafting and critically years, with a stable trend throughout the latest decades of the 20th
revising the article for important intellectual content; Editing and revising the century. Although the disease was relatively uncommon outside
English for the final version to be published; and final approval of the version Western countries during the 20th century, a rising incidence is
to be published.
This research did not receive any specific grant from funding agencies in the
reported from the beginning of the 21st century within newly
public, commercial, or not-for-profit sectors. industrialized countries.2
Ethical approval: No ethical approval was required for this article. Since the first successful appendectomy was performed by
The authors report no conflicts of interest. Claudius Amyand in 1735 at St. George’s Hospital in London, surgery
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of
has been the mainstay of treatment for over 2 centuries. Current
this article on the journal’s Web site (www.annalsofsurgery.com). evidence suggests laparoscopic appendectomy as the gold standard
Reprints: Dr. Mauro Podda, MD, Department of Surgery: General, Emergency and for surgical treatment, with lower incidence of wound infections,
Robotic Surgical Unit, San Francesco Hospital, ASSL Nuoro, ATS Sardegna, postintervention morbidity, shorter hospital stay, and higher quality
Via Mannironi, 1, 08100 Nuoro, Italy. E-mail: mauropodda@ymail.com.
Copyright ß 2019 Wolters Kluwer Health, Inc. All rights reserved.
of life scores when compared with open appendectomy.3
ISSN: 0003-4932/16/XXXX-0001 Although conservative management with antibiotics has been
DOI: 10.1097/SLA.0000000000003225 well established for intra-abdominal infections of other sources (in

Annals of Surgery  Volume XX, Number XX, Month 2019 www.annalsofsurgery.com | 1

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Podda et al Annals of Surgery  Volume XX, Number XX, Month 2019

particular, for uncomplicated acute sigmoid diverticulitis), nonoper- uncomplicated appendicitis in adults and children were included in
ative management (NOM) of uncomplicated appendicitis is still the systematic review and meta-analysis. All studies eligible for
limited by conflicting results coming from recent studies with high inclusion had to report a clear definition of the diagnosis ‘‘uncom-
risk of bias.4,5 plicated acute appendicitis.’’ Only studies that reported at least one of
To date, many randomized controlled trials and nonrandom- the primary outcome or secondary outcomes were included.
ized studies have promoted antibiotic therapy as a safe approach to The exclusion criteria were: studies not reporting data on the
appendectomy for adult patients with uncomplicated appendicitis, selected outcomes of interest, or articles in which the outcomes of
suggesting success rates as high as 90% at 30 days and 75% within interest could not be calculated; studies not specifying the patients
1 year of treatment.6– 9 More recently, results of NOM with anti- selection criteria; studies reporting on complicated appendicitis
biotics in children have confirmed that the conservative strategy is a (gangrenous, perforated appendicitis with abscess or generalized
safe and effective alternative to surgery, with 64% to 86% success peritonitis); studies not reporting the specific antibiotic treatment
rates, lower incidence of complications, and no differences in the rate regimens used for the AT; non-human studies; studies that reported
of complicated appendicitis compared to appendectomy.10,11 only percentages instead of absolute numbers or odds ratios; review
However, the recently published Jerusalem Guidelines and articles; editorials; comments; letters and case reports. The 2
separately the European Association of Endoscopic Surgery (EAES) reviewers independently screened all studies retrieved from the
guidelines, note inadequate evidence to recommend routine NOM, search, and full text articles were obtained if inclusion criteria were
and so generally appendectomy remains the treatment of choice fulfilled. Where there was overlap in patient cohorts of 2 studies, the
recommended in Europe and the United States.3,12 most recent and largest study was included in the systematic review
Despite earlier meta-analyses demonstrating that NOM with and meta-analysis.
antibiotics may be an acceptable treatment strategy for patients with
uncomplicated appendicitis, there is still a lack of evidence regarding Risk of Bias Assessment
effectiveness and safety, potential complications, duration of pain, The risk of bias for the studies enrolled in the systematic
costs, lengths of hospital stay and time to return to normal daily life review and meta-analysis was assessed according to the Cochrane
activity following NOM.5,13–16 handbook for systematic reviews of interventions,19 using the
The aim of this systematic review and meta-analysis was to Cochrane risk of bias tool for RCTs, and the risk of bias in
summarize and systematically review the current available evidence nonrandomized studies tool (ROBINS-I) for PCSs and RCSs.20
on the antibiotic approach to uncomplicated appendicitis both in
adults and children. Quality of Evidence Assessment
The grading of recommendations assessment, development
METHODS and evaluation (GRADE) methodology was applied for assessing
This systematic review and meta-analysis was conducted quality of evidence, and reported in the results.21
according to the recommendations of the preferred reporting items
for systematic reviews and meta-analyses (PRISMA) guidelines,17 Outcomes Measures
the meta-analysis of observational studies in epidemiology checklist  Primary outcome measures were evaluated to assess effectiveness
for observational studies (MOOSE),18 and was specified in a regis- and safety of AT and ST for uncomplicated appendicitis. The
tered protocol (PROSPERO: CRD42018104806). All stages of study following outcomes were reviewed:
identification, selection, quality assessment and data abstraction 1. Complication-free treatment success: success of the initial
were carried out independently by 2 reviewers (M.P. and S.D.S.). treatment (AT or ST) with an uncomplicated course (no
Any discrepancies were resolved by consulting a third reviewer postoperative complications, adverse events, or treatment
(C.G.). failure occurring);
2. Treatment efficacy based on 1-year follow-up (intention-to-
Study Identification treat analysis): efficacy for AT was defined as achieving a
MEDLINE (via PubMed), the Cochrane Central Register of definitive improvement without requiring surgery within a
Controlled Trials, and EMBASE were systematically searched for median follow-up of 1 year. Lack of efficacy in the AT group
relevant studies. Reference lists of relevant studies were searched included both persistence of acute appendicitis during the
manually and the ‘‘related articles’’ function in PubMed was used. hospitalization (index admission AT failure: nonresolving
The search strategy combined text words and MeSH terms related to appendicitis with persistent or worsening symptoms during
antibiotic therapy (AT) versus appendectomy/surgical therapy (ST) the primary hospitalization) and recurrent acute appendicitis.
of uncomplicated appendicitis in adults and children: [appendicitis, On the other hand, efficacy for the ST was defined as uncom-
antibiotic, nonoperative treatment, conservative management, non- plicated appendicitis confirmed at the time of the surgical
operative management] and [appendicitis, appendectomy, appendi- operation or histological verification of appendicitis, and
cectomy, laparoscopic appendectomy]. resolution of symptoms after surgical treatment.
Corresponding search strategies were used for the Cochrane 3. Complicated appendicitis with peritonitis identified at the time
Central Register of Controlled Trials and EMBASE. No language of surgical operation: in the AT group the analysis was carried
restrictions were applied. Literature search was completed in out within the cohort of patients who underwent appendec-
August 2018. tomy after the failure of the AT.
The detailed search strategy is freely accessible in the protocol 4. Postintervention complications (intention-to-treat analysis):
(PROSPERO: CRD42018104806). the number and rates of postoperative abscesses, surgical site
infections, incisional hernias, obstructive symptoms, and other
Study Selection general complications, including adverse reaction to antibi-
For sensitivity reasons, not only randomized controlled trials otics, anesthesiology complications, cardiovascular and pul-
(RCTs) but also prospective cohort studies (PCSs) and retrospective monary adverse events were analyzed on an intention-to-
cohort studies (RCSs) comparing AT and ST as primary treatment for treat basis.

2 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Annals of Surgery  Volume XX, Number XX, Month 2019 Nonoperative Management of Appendicitis

5. Surgical complications: the number and rates of surgical Heterogeneity of the results across studies was assessed using
complications only were analyzed both for patients who the Higgins’ I2 and chi-square tests. A P value of chi-square test less
underwent ST as primary treatment and for those who than 0.10 with an I2 value of greater than 50% was considered
underwent surgery as second line approach, after failure of indicative of substantial heterogeneity. Fixed-effects model (Man-
AT treatment during the hospitalization, or for recurrent tel–Haenszel) was implemented if statistically significant heteroge-
appendicitis. neity was absent. Otherwise, a random-effects model was used for
 Secondary outcome measures were evaluated to assess other meta-analysis if statistically significant heterogeneity was found,
potential benefits and drawbacks of AT and ST in terms of: according to the method of DerSimonian and Laird.23
1. Number and rates of patients treated with a laparoscopic Given that substantial differences in methodology and clinical
approach in both groups. settings were found among individual studies, subgroup analyses
2. Total costs: total medical and surgical costs for the primary were planned with the aim of exploring interstudy heterogeneity
hospital stay (primary costs), including materials, medica- (adults vs children and RCTs vs non-RCTs). Funnel plots were
tions, radiology and surgical resources, pathology, laboratory created to evaluate the risk of publication bias.
tests). Costs were also analyzed for all appendicitis-related
care (conservative treatment success or failure vs surgery RESULTS
without complications or surgery with complications). Cur-
A total of 8120 references were identified through database
rency conversion from original value to USD (United States
searching. Two more references were identified by searching lists of
Dollar) was made on August 15, 2018.
retrieved studies (Fig. 1).
3. Length of primary hospital stay: number of days of primary
Fifty-three full text publications were finally assessed for
inpatient admission.
eligibility, of which 20 comparing AT and ST were included for
4. Total length of stay per patient: number of days of hospitali-
quantitative synthesis. Seven of the included studies were random-
zation for readmissions added to the number of days of
ized controlled trials (RCTs),6 –10,24,25 8 were prospective cohort
primary inpatient admission.
studies (PCSs),11,26– 32 4 were retrospective cohort studies
5. Duration of pain (in days) following AT and ST.
(RCSs),33– 36 and 1 was a quasirandomized study (q-RCT).37 Ten
6. Length of sick leave (in days) following AT and ST.
studies were conducted in adults,6–9,24– 26,31,36,37 and 10 in chil-
7. Length of time off work (in days) following AT and ST.
dren.10,11,27 –30,32–35 In total, 3618 patients were allocated to AT (n ¼
8. Quality of life following AT and ST.
1743) or ST (n ¼ 1875). General characteristics of patients as
reported in the studies are shown in Table 1.
Data Extraction
The 2 reviewers independently reviewed each included article. Study Characteristics
A predefined paper-based sheet was used for data extraction. Large heterogeneity was found among the included studies
Data collected for each article comprised the following with regard to diagnostic criteria for uncomplicated appendicitis.
predefined items: 1) Study identifier (first author, year of publica- Marked heterogeneity was also demonstrated in type of antibiotics
tion); 2) Essential study data (study period, study location, study administered, duration of administration, and different
population); 3) Study design (RCT, q-RCT, PCS, RCS); 4) Treat- outcomes evaluated.
ment arms and number of enrolled subjects; 5) Baseline character- A single pediatric patient was enrolled in the otherwise adult
istics of study subjects (mean age, sex, WBC count, body population pilot study by Talan et al.9 For this reason, running a
temperature, CRP concentration, Alvarado/AIR/PAS score on minimal risk of selection bias, the trial was included among
admission); 6) General characteristics of eligible studies (inclusion adult studies.
and exclusion criteria, sample size calculation, preintervention Hansson et al implemented quasi-randomization by date of
imaging techniques, treatment modalities, definition of the investi- birth. Patients, once randomized to a specific treatment, were allowed
gated primary and secondary outcomes); 7) Treatment outcomes, as to cross-over and receive the alternative treatment based on their
described above. preference or medical judgment, resulting in cross-over (47.5% of
patients in the AT group underwent surgery, and 7.8% in the ST group
Statistical Analysis were finally treated conservatively with antibiotics).37
The meta-analysis was conducted by searching for a numeri- Mahida et al29 assessed the feasibility of NOM with antibiotics
cal estimate of the outcome of interest. only in children with an appendicolith identified on preintervention
Variables for pooled analysis were considered if they were imaging. Poillucci et al36 included 14 patients (8.7%) with a radio-
previously evaluated by at least 2 studies. All statistical analyses logical diagnosis of complicated disease (phlegmon or abscess). The
were carried out using Reviewer Manager software (Review Man- randomized pilot trial by Talan et al9 evaluated safety and feasibility
ager—RevMan—version 5.3.5, 2014, The Nordic Cochrane Centre, of a protocol allowing outpatient antibiotic management. This study
Cochrane Collaboration, www.cochrane-handbook.org). Data entries was therefore excluded from the pooled analysis of length of
in the columns of forest plots were double-checked individually by 2 hospitalization (Supp. Digit. Content. Tab. 1, http://links.lww.com/
reviewers to avoid errors. The odds ratio (OR) with 95% confidence SLA/B582).
interval (95% CI) was calculated for dichotomous variables, and the
standardized mean difference (SMD) with 95% CI for continuous Risk of Bias and Quality of Evidence Assessment
variables. When continuous data were presented as medians and Three of the 8 RCTs were judged at a low risk of bias,6,9,10 2 at
range, the method by Hozo et al to estimate respective means and high risk,25,37 and 3 at unclear risk.7,8,24 All but 3 trials24,25,37
standard deviations was applied.22 The point estimate of the OR generated random sequence adequately and reported allocation
value was considered statistically significant at P level of less than concealment, resulting in a low risk of selection bias. None of the
0.05 if the 95% CI did not cross the value 1. The point estimate of the trials reported attempts at blinding patients, personnel, outcome
SMD value was considered statistically significant at P level of less assessors, or data analysts. None of the trials were considered at
than 0.05 if the 95% CI did not cross the value 0. high risk of selective reporting and incomplete outcome data, as

ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 3

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Podda et al Annals of Surgery  Volume XX, Number XX, Month 2019

Identification

Records identified through Additional records identified

database searching through other sources
(n=8.120) (n=2)

Excluded studies based on

titles and abstract review
(n=7.975)
Screening

Potentially appropriate studies

to be included
in the Meta-analysis Studies, excluded due to:
(n=147) - other aims (n=39)
- other comparisons (n=27)
- no comparisons (n=19)
- study groups not clearly defined (n=6)
- duplicates (n=3)
Eligibility

Full-text articles assessed

for eligibility
(n=53)
Full-text article excluded due to:
- retraction since publication (n=1)
- other study design (n=13)
- irrelevant article type (letter, abstract,
case report, review) (n=19)
Studies included in
qualitative synthesis
(n=20)
Included

Studies included in
quantitative synthesis
(n=20)

FIGURE 1. The PRISMA flow diagram for search and selection of articles included in the systematic review and meta-analysis.

primary endpoints were clearly defined and reported in each study included 19 studies. Taking into account any type of postinterven-
(Supp. Digit. Content. Tab. 2, http://links.lww.com/SLA/B582). Four tional complications (including treatment failure), a significantly
of the 10 non-RCTs were considered at serious risk of bias according higher success rate was reported for ST: 82.3% for ST versus 67.2%
to the ROBINS-I tool,11,29,33,34 whereas the remaining 6 were judged for AT (sample size: 3374; OR 0.30; 95% CI 0.20–0.47; P <
at low or moderate risk (Supp. Digit. Content. Tab. 2, http://link- 0.00001; I2 ¼ 77%). Subgroup analyses of the outcome revealed
s.lww.com/SLA/B582). Graphically, potential publication bias was no significant difference between adults (sample size: 2767; AT:
present for the following outcomes of interest: complication-free 68.7% vs ST: 80.9%; OR 0.37; 95% CI 0.22–0.63; P ¼ 0.0003; I2 ¼
treatment success, treatment efficacy based on 1-year follow-up, 84%) and children (sample size: 607; AT: 60.3% vs ST: 88.9%; OR
postintervention complications, surgical complications, complicated 0.21; 95% CI 0.10–0.44; P < 0.0001; I2 ¼ 51%; test for subgroup
appendicitis with peritonitis identified at the time of surgical opera- differences: P ¼ 0.21; I2 ¼ 35.5%) and between RCTs (sample size:
tion, and length of primary hospital stay. Funnel plots have been 1800; AT: 65.7% vs ST: 79.6%; OR 0.30; 95% CI 0.15–0.58; P ¼
provided as supplemental digital content (Supp. Digit. Content. 0.0004; I2 ¼ 82%) and n-RCTs (sample size: 1574; AT: 68.7% vs ST:
Fig. 1, http://links.lww.com/SLA/B582). Overall quality of 86.1%; OR 0.30; 95% CI 0.16–0.56; P < 0.0001; I2 ¼ 77%; test for
evidence, according to the GRADE criteria, was moderate for subgroup differences: P ¼ 0.96; I2 ¼ 0%) (Fig. 2).
complication-free treatment success, treatment efficacy based on
1-year follow-up, postintervention complications, length of sick Treatment Efficacy Based on 1-year Follow-up
leave, and length of time off work. Surgical complications, length Data from 19 studies were included in the pooled analysis
of primary hospital stay, duration of pain, and total costs had a low investigating the treatment efficacy based on 1-year follow-up
quality of evidence, whereas outcome results for intraoperative (Table 3).
finding of complicated appendicitis with peritonitis were judged A significantly higher success rate was reported for ST: 93.1%
as very low (Table 2). versus 72.6% for AT (sample size: 3374; OR 0.12; 95% CI 0.06–
0.24; P < 0.00001; I2 ¼ 81%). Subgroup analyses of the outcome
Complication-free Treatment Success revealed no significant difference between adults (sample size: 2767;
Overall, 20 studies reported the rate of complication-free AT: 73.6% vs ST: 91.9%; OR 0.18; 95% CI 0.08–0.41; P < 0.0001;
treatment success (Table 3). The study by Koike et al34 did not I2 ¼ 88%) and children (sample size: 607; AT: 68.1% vs ST: 98.1%;
report the results in the ST group, and so, the pooled analysis OR 0.08; 95% CI 0.04–0.16; P < 0.00001; I2 ¼ 0%; test for

4 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ß

Annals of Surgery  Volume XX, Number XX, Month 2019

2019 Wolters Kluwer Health, Inc. All rights reserved.

TABLE 1. General Characteristics of Patients as Reported in the Studies Included for the Systematic Review and Meta-analysis
WBC Count Body CRP Alvarado or AIR or
9
( 10 /L): Temperature: Concentration PAS Score: Mean 
Age—Years: Mean  SD— Admis- Mean  SD —Admis- (mg/L): Mean  SD—Admission
Patients N. Sex (M:F) Mean  SD sion sion SD—Admission
Study Study Study Study
Study Type Period Location Population A S A S A S A S A S A S A S

ANNSURG-D-18-02034
Eriksson S 1995 RCT 1992–1994 Sweden Adults 20 20 14:6 13:7 27.8  10.1 35  16.1 13.8  4.4 13.9  4.1 37.2  0.7 37.1  0.7 41  30 40  38 NR NR
Styrud J 2006 RCT 1996–1999 Sweden Adults 128 124 128:0 124:0 NR NR 12.5  3.8 12.4  3.5 37.5  0.7 37.4  0.8 55  44 54  49 NR NR

Hansson J 2009 q-RCT 2006–2007 Sweden Adults 119 250 62:57 138:112 40  2 37  1 12.2  0.4 13.5  0.3 37.2  0.1 37.5  0.1 51  5 56  3 NR NR
Turhan AN 2009 RCT 2005–2006 Turkey Adults 107 183 65:42 125:58 30.9  1.3 26.2  0.7 NR NR NR NR NR NR 6.5  0.1 6.4  0.1
Vons C 2011 RCT 2004–2007 France Adults 119 120 73:47 70:49 31  9 34  12 13.6  3.6 13.1  3.4 NR NR NR NR NR NR
Armstrong J 2014 RCS 2012–2013 Canada Children 12 12 4:8 6:6 12.2  4.2 12  3.2 16.1  4.4 14.0  4.2 NR NR 13.6  26.8 54  23.4 NR NR
Koike Y 2014 RCS 2004–2010 Japan Children 130 114 66:59 NR 7.0  4.0 NR 12.5  4.0 NR NR NR 26.6  0.8 NR 8.2  0.2 NR
Park HC 2014 PCS 2010–2011 Korea Adults 119 159 57:62 86:73 36.7  14.1 38.4  13.8 11.5  3.8 12.1  4.1 37.5  1 37.8  1.1 41.2  48.3 46.3  49.5 6.9  1 7.0  1
Minneci PC 2015 PCS 2012–2013 USA Children 37 65 24:13 45:20 11  1.1 12  1.1 12.9  1.6 12.9  1.3 NR NR NR NR NR NR
Salminen P 2015 RCT 2009–2012 Finland Adults 257 273 155:102 174:99 33  6.1 35  5.4 11.7  3.9 12  4 NR NR 29  15 36  13.5 NR NR
Svensson JF 2015 RCT 2012–2012 Sweden Children 24 26 14:10 12:14 12.2  2.6 11.1  2.4 14.0  4.0 14.5  6.4 37.3  0.7 37.5  0.6 30.5  53.1 27.0  50.2 NR NR
Hartwich J 2016 PCS 2012–2014 USA Children 24 50 14:11 30:19 12.6  0.6 12.1  0.5 15.2  0.9 15.3  0.5 NR NR NR NR NR NR
y
Mahida JB 2016 PCS 2014–2015 USA Children 5 9 1:4 5:4 14  0.1 11  1.7 13.8  1.8 14.3  0.3 NR NR NR NR NR NR
Tanaka Y 2016 PCS 2007–2013 Japan Children 78 86 52:26 61:25 10.1  2.0 10.4  2.3 14.7  3.7 15.3  3.8 NR NR 47  48 61  57 NR NR
Allievi N 2017 PCS 2011–2015 Italy Adults 284 109 135:149 59:50 36.8  16.6 39.6  15.9 13.0  4.0 13.8  4.2 NR NR 52.9  4.2 65.2  5.0 6.2  1.6 6.8  1.8
Gorter RR 2017 PCS 2012–2014 Netherlands Children 25 19 15:10 14:5 14  2.0 14  2.8 12.0  3.7 12.0  4.9 NR NR 35  32 30  54.2 NR NR
Lee SL 2017 PCS 2015–2016 USA Children 51 32 30:21 17:15 10  1.7 11  2.0 13  1.4 14  2.0 37.5  0.3 37.2  0.3 NR NR 7  0.8 7  0.6
Mudri M 2017 RCS 2012–2015 UK Children 26 26 7:19 18:8 12 11 NR NR NR NR NR NR NR NR
Poillucci G 2017 RCS 2014–2016 Italy Adults 162 184 63:99 86:98 33.6  16.5 35.8  17.2 NR NR NR NR NR NR 6.1  1.3 6.9  1.9
Talan DA 2017 RCT 2015–2015 USA Adults 16 14 9:7 9:5 31  18.4 36  11.8 14.2  3.7 15.3  4.3 36.8  0.3 36.9  0.5 25.9  57.4 64.8  71.6 8  1.7 8  1.7
Total 1.743 1.875 988:752 1.092:667 21.8  11.6 23.4  12.4 13.3  1.2 13.6  1.1 37.2  0.2 37.3  0.2 37.3  12.8 48.5  13.6 NA NA

Nonoperative Management of Appendicitis

A indicates antibiotic group; AIR, appendicitis inflammatory response; NA, not applicable; NR, not reported; PAS, pediatric appendicitis score; PCS, prospective cohort study; q-RCT, quasi-randomized controlled trial; RCS,
retrospective cohort study; RCT, randomized controlled trial; S, surgery group; SD, standard deviation.

Per-protocol analysis.
yUncomplicated acute appendicitis with appendicolith.
www.annalsofsurgery.com | 5

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Podda et al Annals of Surgery  Volume XX, Number XX, Month 2019

Publication Overall Quality

Meta-analysis of RCTs comparing conservative management with antibiotics and appendectomy for: complication-free treatment success, treatment efficacy based on 1-yr follow-up, intraoperative finding of complicated
of Evidence subgroup differences: P ¼ 0.13; I2 ¼ 57.0%) and between RCTs
(sample size: 1800; AT: 73.8% vs ST: 94.7%; OR 0.12; 95% CI
0.05–0.26; P < 0.00001; I2 ¼ 67%) and n-RCTs (sample size: 1574;
Moderate
Moderate

Moderate

Moderate
Moderate
Very low AT: 71.6% vs ST: 90.8%; OR 0.12; 95% CI 0.04–0.35; P < 0.0001;
Low
Low
Low

Low
I2 ¼ 85%; Test for subgroup differences: P ¼ 0.95; I2 ¼ 0%) (Fig. 3).
All the studies but that by Park et al26 reported the rate of index
Not serious

Not serious
Not serious
Not serious
admission antibiotic treatment failure. This was 8.5%. Overall, the
Bias

rate of recurrence at 1-year follow-up was 19.2%. Laparoscopic

Serious
Serious
Serious
Serious
Serious

Serious

appendicitis with peritonitis, postintervention complications, surgical complications, length of primary hospital stay, duration of pain, length of sick leave, length of time off work, and total costs.
appendectomy rates were reported by 8 authors in the AT
group8,10,27,32,33,35,36 and by 14 authors in the ST group7–
10,25,27,30–33,35,36
(Supp. Digit. Content. Tab. 3, http://link-
Inconsistency Indirectness Imprecision
Not serious
Not serious

Not serious
Not serious
Not serious
Not serious
Not serious
Not serious
Not serious

s.lww.com/SLA/B582). However, only 2 studies were eligible for

the quantitative analysis.8,36 Results indicated that no significant
Serious

difference in the rate of laparoscopic appendectomy was present

between the group of patients submitted to surgery as first-line
Quality Assessment

treatment and those who underwent surgery after failure of NOM

Not serious
Not serious

Not serious

Not serious
Not serious
Not serious

with antibiotics (sample size: 624; AT: 54.2% vs ST: 77.4%; OR

Serious

Serious
Serious

Serious

0.31; 95% CI 0.06–1.55; P ¼ 0.16; I2 ¼ 87%).

Intraoperative Finding of Complicated Appendicitis

With Peritonitis
Not serious

Three studies did not report data on this outcome,28,31,34 and 3

studies did not report any such event.9,33,35 The meta-analysis was
Serious
Serious
Serious
Serious
Serious
Serious
Serious
Serious

Serious

conducted on 14 studies.6 –8,11,24– 27,29,30,32,36 All studies, except

Turhan et al25 (based on intraoperative observation alone) reported
the incidence of complicated appendicitis by histologic examination.
Not serious
Risk of

Overall, it was 21.7% for AT versus 12.8% for ST (sample size: 1868;
Bias
Serious

Serious
Serious
Serious
Serious
Serious
Serious
Serious
Serious

OR 2.01; 95% CI 0.93–4.34; P ¼ 0.07; I2 ¼ 71%). The odds ratio for

complicated appendicitis was doubled for patients in the AT group
undergoing surgery following failure of NOM. However, the differ-
RCTs, 1 q-RCT
RCTs, 1 q-RCT
RCTs, 1 q-RCT
RCTs, 1 q-RCT
RCTs, 1 q-RCT
RCTs, 1 q-RCT

RCTs, 1 q-RCT

RCTs, 1 q-RCT
RCT, 1 q-RCT

ence was not statistically significant. Subgroup analyses revealed no

Study Design

significant differences between adults (sample size: 1583; AT: 21.8%

vs ST: 12.7%; OR 2.63; 95% CI 0.98–7.09; P < 0.00001; I2 ¼ 81%)
RCTs

and children (sample size: 285; AT: 20.1% vs ST: 13.9%; OR 1.17;
95% CI 0.35–3.88; P ¼ 0.80; I2 ¼ 30%; Test for subgroup differ-
ences: P ¼ 0.31; I2 ¼ 4.7%) and between RCTs (sample size: 1192;
7
7
6
6
6
6
1
2
2
2

AT: 22.9% vs ST: 11.6%; OR 2.96; 95% CI 0.91–9.60; P ¼ 0.0002;

Studies Patients
No. of No. of

1800
1800
1192
1750
1149
1770

1151
608

491
449

I2 ¼ 77%) and n-RCTs (sample size: 676; AT: 19.6% vs ST: 14.9%;
Q-RCT indicates quasi-randomized controlled trial; RCT, randomized controlled trial.

OR 1.30; 95% CI 0.52–3.21; P ¼ 0.58; I2 ¼ 49%; Test for subgroup

differences: P ¼ 0.28; I2 ¼ 15.9%) (Table 3) (Supp. Digit. Content.
TABLE 2. Level of Evidence According to the GRADE Criteria

Fig. 2, http://links.lww.com/SLA/B582).
8
8
7
7
7
7
2
3
2
3

Postintervention Complications
Intraoperative finding of complicated appendicitis with peritonitis

All studies except Koike et al34 reported the rate of post-

intervention complications. Three studies10,28,29 did not observe any
such adverse events, whereas Allievi et al31 reported the postin-
Summary estimate based on the quality assessment items.

tervention complications rate only for patients in the ST group. The

meta-analysis was conducted on the remaining 15 studies. Overall,
the rate of complications of AT was significantly lower compared to
ST (sample size: 2843; AT: 7.1% vs ST: 14.5%; OR 0.41; 95% CI
Treatment efficacy based on 1-yr follow-up

0.22–0.77; P ¼ 0.006; I2 ¼ 68%). Subgroup analyses revealed that

the postintervention complications rate was statistically higher in the
Complication-free treatment success

ST group limited to the adult population (sample size: 2374; AT:

Total costs (successful treatment)

6.6% vs ST: 14.5%; OR 0.39; 95% CI 0.16–0.94; P ¼ 0.04; I2 ¼

Length of primary hospital stay
Postintervention complications

80%), but not in children (sample size: 469; AT: 9.6% vs ST: 12.5%;
OR 0.57; 95% CI 0.28–1.16; P ¼ 0.12; I2 ¼ 0%), although no
Length of time off work

significant differences were found between the two subgroups (test

Surgical complications

for subgroup differences: P ¼ 0.53; I2 ¼ 0%).

Length of sick leave

AT was associated with a significantly lower rate of compli-

Duration of pain

cations in non-RCT studies (sample size: 1750; AT: 5.3% vs ST:

12.1%; OR 0.31; 95% CI 0.15–0.65; P ¼ 0.002; I2 ¼ 28%) but not in
Outcome

RCTs (sample size: 1093; AT: 8.2% vs ST: 15.9%; OR 0.55; 95% CI
0.21–0.45; P ¼ 0.22; I2 ¼ 81%). The test for subgroup difference


indicated that the design of the study did not modify the effect of

6 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Annals of Surgery  Volume XX, Number XX, Month 2019 Nonoperative Management of Appendicitis

TABLE 3. Summary of Primary Outcomes (Complication-free Treatment Success, Treatment Efficacy Based on 1-yr Follow-up,
Intraoperative Finding of Complicated Appendicitis With Peritonitis, Postintervention Complications, Surgical Complications)
Intraoperative Find-
Complication-free Treatment Efficacy Based ing of Complicated Postintervention Surgical
Treatment Success. on 1-yr Follow-up (ITT). Appendicitis With Complications (ITT). Complications.
N (%) N (%)y Peritonitis. N (%)z N (%) N (%)z
Study A S A S A S A S A S
Eriksson S 1995 13 (65) 18 (90) 13 (65) 20 (100) 1 (14.3) 1 (5) – 2 (10) – 2 (10)
Styrud J 2006 93 (72.6) 107 (86.3) 97 (75.8) 120 (96.8) 12 (38.7) 6 (5) 4 (3.1) 17 (13.7) 4 (12.9) 17 (13.7)
Hansson J 2009 57 (47.9) 143 (57.2) 93 (78.2) 223 (89.2) 6 (26) 50 (20) 36 (30.2) 80 (32) 6 (26) 74 (29.6)
Turhan AN 2009 82 (76.6) 175 (95.6) 87 (81.3) 183 (100) – 31 (16.9) 5 (4.7) 8 (4.4) 5 (26.3) 8 (4.4)
Vons C 2011 63 (52.9) 114 (95) 75 (63) 117 (97.5) 12 (27.2) 21 (18) 12 (10.1) 3 (2.5) 12 (27.3) 3 (2.5)
Armstrong J 2014 7 (58.3) 10 (83.3) 8 (66.7) 12 (100) – – 1 (8.3) 2 (16.7) 1 (25) 2 (16.7)
Koike Y 2014 105 (78.3) NR 105 (78.3) NR NR NR NR NR NR NR
Park HC 2014 95 (79.8) 136 (85.5) 96 (80.6) 152 (95.5) 4 (20) 11 (6.9%) 1 (0.8) 16 (10) 1 (5) 16 (10)
Minneci PC 2015 28 (75.7) 56 (86.2) 28 (75.7) 61 (93.8) 1 (11.1) 8 (12.3) – 5 (7.7) – 5 (7.7)
Salminen P 2015 180 (70) 209 (76.5) 186 (72.4) 254 (93) 14 (16.5) 2 (0.7) 6 (2.3) 45 (16.5) 6 (8.6) 45 (16.5)
Svensson JF 2015 18 (75) 26 (100) 18 (75) 26 (100) – 5 (19.2) – – – –
Hartwich J 2016 17 (70.8) 50 (100) 17 (70.8) 50 (100) NR NR – – – –
Mahida JB 2016 2 (40) 9 (100) 2 (40) 9 (100) – 6 (66.7) – – – –
Tanaka Y 2016 55 (70.5) 84 (97.7) 55 (70.5) 86 (100) 1 (7.1) – – 2 (2.3) – 2 (2.3)
Allievi N 2017 232 (81.7) 89 (81.6) 232 (81.7)§ 78 (71.5)§ NR NR NR 20 (18.3) NR 20 (18.3)
Gorter RR 2017 9 (36) 8 (42.1) 21 (84) 19 (100) – 4 (21) 12 (48) 11 (57.9) – 2 (10.5)
Lee SL 2017 17 (33.3) 25 (78.1) 26 (50.9) 31 (96.8) 8 (50) 10 (31.2) 9 (18) 6 (19) 3 (12.5) 6 (19)
Mudri M 2017 17 (65.4) 21 (80.7) 17 (65.4) 25 (96.1) – – – 4 (15.4) – 4 (15.4)
Poillucci G 2017 87 (53.7) 159 (86.4) 87 (53.7) 159 (86.4) 10 (29.4) 44 (23.8) 4 (2.5) 25 (13.6) 4 (7.1) 25 (13.6)
Talan DA 2017 13 (81.2) 12 (85.7) 14 (87.5) 14 (100) – – 1 (6.2) 2 (14.3) – 2 (14.3)
Total 1.190 (67.2) 1.451 (82.3) 1.277 (72.6) 1.639 (93.1) 69 (21.7) 199 (12.8) 91 (7.1) 248 (14.5) 42 (12.8) 233 (13.6)

Success of the initial treatment with uncomplicated course (no postoperative complications, adverse events, or treatment failure occurring).
yIf not otherwise specified, efficacy means clinical recovery without need for surgical treatment at 1-year follow-up in the antibiotic group, and positive diagnosis of acute
appendicitis during operation and resolution of symptoms after surgical treatment in the surgery group.
zIn the antibiotic group, after failure of the primary treatment and subsequent surgery.
§In the antibiotic group the failure rate was intended as the necessity of appendectomy during index admission (acute failure) or a new episode of suspected appendicitis requiring
surgical treatment within 1 year from discharge (delayed failure).
A indicates antibiotic group; CT, computed tomography; ITT, intention-to-treat analysis; NR, not reported; NA, not applicable; S, surgery group; US, ultrasound scan.

interventions (P ¼ 0.37; I2 ¼ 0%) (Table 3) (Supp. Digit. Content. (sample size: 616; AT: 0% vs ST: 0.6; OR 1.33; 95% CI 0.15–11.81;
Fig. 3, http://links.lww.com/SLA/B582). P ¼ 0.49; I2 ¼ 0%) and abscess formation (sample size: 1351; AT:
0.9% vs ST: 1.9; OR 1.08; 95% CI 0.41–2.80; P ¼ 0.88; I2 ¼ 0%)
Surgical Complications were similar between the 2 groups (Supp. Digit. Content. Tab. 4,
Nineteen of 20 studies reported data on surgical complica- http://links.lww.com/SLA/B582), (Supp. Digit. Content. Fig. 5,
tions. Allievi et al31 reported the surgical complications rate only for http://links.lww.com/SLA/B582).
patients in the ST group. Koike et al did not provide results for this
outcome of interest.34 Three studies10,28,29 did not observe any Total Costs
adverse events. The pooled analysis involved 15 studies. Overall, The pooled analysis of primary costs included 7 stud-
the rates of surgical complications after AT and ST were equivalent ies.9,10,26 –28,34,37
(sample size: 1894; AT: 12.8% vs ST: 13.6%; OR 1.16; 95% CI Overall, AT resulted in significantly lower costs when com-
0.59–2.28; P ¼ 0.66; I2 ¼ 57%). Subgroup analyses of the outcome pared to ST (sample size: 1147; $ 2509.14  1621.5 vs $
revealed no significant difference between adults (sample size: 1598; 4898.57  3641.5; SMD 3.65; 95% CI 5.36–1.93; P <
AT: 14.0% vs ST: 14.5%; OR 1.32; 95% CI 0.53–3.28; P ¼ 0.55; I2 0.00001, I2 ¼ 98%). Two studies were included in the meta-analysis
¼ 74%) and children (sample size: 296; AT: 7.1% vs ST: 8.7%; OR of all appendicitis-related care costs.10,27 Results demonstrated
0.86; 95% CI 0.31–2.36; P ¼ 0.77; I2 ¼ 0%; Test for subgroup significantly lower costs in the AT group (sample size: 152; AT: $
differences: P ¼ 0.54; I2 ¼ 0%). In the same way, subgroup analysis 4074.50  204.3 vs ST: $ 5117  124.4; SMD 0.69; 95% CI
of the outcome revealed no significant difference between RCTs 1.02–0.35; P < 0.0001, I2 ¼ 43%) (Supp. Digit. Content. Tab.
(Sample size: 1179; AT: 16.9% vs ST: 15.3%; OR 1.77; 95% CI 3, http://links.lww.com/SLA/B582), (Supp. Digit. Content. Fig. 6,
0.58–5.35; P ¼ 0.32; I2 ¼ 78%) and non-RCT studies (sample size: http://links.lww.com/SLA/B582).
715; AT: 6.8% vs ST: 10.6%; OR 0.64; 95% CI 0.32–1.29; P ¼ 0.21;
I2 ¼ 0%; test for subgroup differences: P ¼ 0.13; I2 ¼ 56.6%) Length of Primary Hospital Stay
(Table 3) (Supp. Digit. Content. Fig. 4, http://links.lww.com/SLA/ The pooled analysis of length of primary hospital stay
B582). included 13 studies.6 –8,10,24– 27,30,31,33,36,37
Overall, the rate of wound infections (sample size: 1492; AT: Overall, AT and ST showed an equivalent length of primary
4.2% vs ST: 6.9%; OR 0.90; 95% CI 0.24–3.41; P ¼ 0.87; I2 ¼ 68%), hospital stay (sample size: 3077; Days 2.9  1.3 vs Days 3.3  1.7;
bowel obstruction (sample size: 1107; AT: 3.2% vs ST: 3.9%; OR SMD 0.55; 95% CI 1.49–0.39; P ¼ 0.25; I2 ¼ 99%). Subgroup
0.90; 95% CI 0.41–2.01; P ¼ 0.80; I2 ¼ 0%), incisional hernia analyses of the outcome revealed no significant difference between

ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 7

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Podda et al Annals of Surgery  Volume XX, Number XX, Month 2019

FIGURE 2. Meta-analysis of complication-free treatment success. Subgroup analyses: adults versus children [A], and RCTs versus n-
RCTS [B].

8 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Annals of Surgery  Volume XX, Number XX, Month 2019 Nonoperative Management of Appendicitis

FIGURE 3. Meta-analysis of treatment efficacy based on 1-year follow-up. Subgroup analyses: adults versus children [A], and RCTs
versus n-RCTS [B].

ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 9

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Podda et al Annals of Surgery  Volume XX, Number XX, Month 2019

adults (sample size: 2737; SMD 0.20; 95% CI 1.34–0.93; P ¼ complicated appendicitis at delayed surgery in the overall population
0.73; I2 ¼ 99%) and children (sample size: 340; SMD 1.40; 95% CI (21.7% vs 12.8%, P ¼ 0.07), especially in adults (21.8% vs 12.7%, P
3.71–0.92; P ¼ 0.24; I2 ¼ 98%; test for subgroup differences: P ¼ ¼ 0.06). Nevertheless, there are some advantages to antibiotic
0.36; I2 ¼ 0%). therapy over appendectomy with lower rate of postintervention
In non-RCT studies (sample size: 1770; SMD 1.52; 95% CI complications (7.1% vs 14.5%, P ¼ 0.006) and reduced healthcare
2.42–0.63; P ¼ 0.0009; I2 ¼ 98%) AT resulted in a significantly costs. All other outcomes that impact the patient experience, includ-
shorter length of stay, whereas in RCTs (sample size: 1307; SMD ing complications following surgery, length of hospitalization, dura-
0.35; 95% CI 1.30–2.00; P ¼ 0.68; I2 ¼ 99%) the difference was tion of pain, length of sick leave, and length of time off work, did not
not statistically significant. Test for subgroup differences (P ¼ 0.05; show any statistically significant difference between the 2 different
I2 ¼ 73.7%) indicated that there was no statistically significant treatment modalities.
subgroup effect, suggesting that the study design did not modify Our results are in line with those reported by Sallinen et al in
the effect of AT in comparison with ST (Supp. Digit. Content. Tab. 4, their recent meta-analysis. These authors found a similar incidence of
http://links.lww.com/SLA/B582), (Supp. Digit. Content. Fig. 7, recurrent appendicitis at 1 year (22.6%) compared with our results
http://links.lww.com/SLA/B582). (19.2%) and concluded that the tradeoff between AT and ST (3%
Total length of stay per patient, duration of pain, length of sick fewer major complications, 7% fewer minor complications, a mean
leave, length of time off work, the total length of stay per patient were of 4 days’ shorter sick leave, and 92% fewer appendectomies in the
documented in 4 studies.31–34 It did not differ significantly between first month) must be balanced against a 23% recurrence rate within
AT and ST (sample size: 744; Days 2.9  1.2 vs Days 3.8  1.3; SMD 1 year of follow-up and slightly longer hospital stay.14
0.32; 95% CI 0.74–0.09; P ¼ 0.12; I2 ¼ 81%). In accordance with the Affordable Care Act (ACA), patients
Duration of pain was evaluated in 3 studies.8,32,37 Results of need to be informed of the most recent clinical evidence regarding
the pooled analysis did not demonstrate any statistically significant management of appendicitis, and engaged in sharing decision-mak-
difference between AT and ST (sample size: 691; Days 3.2  1.7 vs ing processes.38 Despite this established principle, little is known
Days 4.9  3.1; SMD 1.62; 95% CI 3.61–0.38; P ¼ 0.11; I2 ¼ about patients’ attitudes and expectations when a diagnosis of
99%). appendicitis is made. In 2018, an anonymous Internet-based survey
The length of sick leave was reported by 5 authors.6,7,27,32,37 conducted by Hanson et al, asked survey respondents to choose a
Overall, results of the pooled analysis did not differ significantly treatment between laparoscopic appendectomy, open appendectomy,
between AT and ST (sample size: 1336; Days 8.4  6 vs Days 10  6; and NOM with antibiotics in the eventuality they had uncomplicated
SMD 0.57; 95% CI 4.24–3.10; P ¼ 0.76; I2 ¼ 100%). In the same appendicitis. Of the 1738 respondents, 85.8% chose laparoscopic
way, subgroup analyses of the outcome revealed no significant appendectomy, 4.9% open appendectomy, and 9.4% antibiotics
difference between adults (Sample size: 1151; SMD 0.61; 95% alone. This study provides a robust evidence base to support creation
CI 5.10–6.31; P ¼ 0.83; I2 ¼ 100%) and children (sample size: of decision aids that may help patients with uncomplicated appendi-
185; SMD 2.33; 95% CI 5.76–1.09; P ¼ 0.18; I2 ¼ 99%; Test for citis be better involved in decisions about their care.39
subgroup differences: P ¼ 0.39; I2 ¼ 0%). Unfortunately, we were not able to perform any type of meta-
The length of time off work, documented in 3 studies,7,8,31 did analysis for quality of life following NOM with antibiotics and
not show any statistically significant difference between AT and ST appendectomy. Although in children equivalent Pediatric Quality
(sample size: 884; Days 7.9  1.9 vs Days 11.7  2.5; SMD 0.52; of Life scores have been reported, it is interesting to note that results
95% CI 1.24–0.20; P ¼ 0.15; I2 ¼ 96%) (Supp. Digit. Content. of the mental SF-12v2 component score demonstrated higher values
Tab. 4, http://links.lww.com/SLA/B582), (Supp. Digit. Content. Fig. for adult patients who underwent surgery than for those treated with
8, http://links.lww.com/SLA/B582). NOM in the study by Talan et al. This finding adds to the debate on
the current perceptions of the antibiotic-first strategy for uncompli-
Quality of Life cated appendicitis, possibly demonstrating anxiety about future
Three studies provided data regarding quality of life.9,27,32 episodes of abdominal pain in patients who did not receive definitive
However, due to the different scales used to assess the outcome, it surgical treatment, and introducing an element of decisional regret.9
was not possible to perform a pooled analysis. In the study by Talan Few studies have focused on how to distinguish patients who
et al, AT patients had higher physical SF-12v2 Health Survey scores might respond well to NOM with antibiotics from those who require
than ST patients both at 2-weeks (median 54 vs 44) and at 1-month appendectomy. Hansson et al, in their report on 581 patients pub-
follow-up (median 56 vs 47).9 lished in 2014, found that patients with assumed uncomplicated
On the other hand, results of the mental SF-12v2 component appendicitis who fulfilled all criteria with CRP <60 g/L, WBC <12
showed higher values for patients who underwent ST both at 2 weeks  109/L and age <60 years had an 89% chance of recovery with
(median 58 vs 55) and at 1 month follow-up (median 56 vs 55). In antibiotics.40 In another recent study, patients with a longer duration
children, Minneci et al (95.7 vs 91.3) and Lee et al (100 vs 100) of symptoms prior to admission (> 24 h) were more likely to have
reported similar Pediatric Quality of Life scores between AT and ST successful NOM, probably because the lack of progression to
groups at 1-year and at 1-month, respectively.27,32 complicated disease is associated with an indolent clinical evolution.
Other independent predictors of success included lower temperature,
DISCUSSION imaging-confirmed uncomplicated appendicitis with lower modified
This comprehensive systematic review and meta-analysis of Alvarado score (< 4), and smaller diameter of the appendix.41
20 studies including over 3600 patients has demonstrated that The prospective trial by Mahida et al reported that the failure
antibiotic therapy as a primary nonoperative management strategy rate of NOM in children affected by uncomplicated appendicitis with
for uncomplicated appendicitis in both adults and children is associ- appendicolith was high (60%) at a median follow-up of less than 5
ated with a treatment failure rate of 27.7% at 1 year follow-up, a months.29 The presence of an appendicolith has been associated with
lower complication-free treatment success rate compared to appen- high failure rates in the reports published by Tanaka et al (failure rate:
dectomy (67.2% vs 82.3%, P < 0.0006) and a tendency toward a 47%), Svensson et al (failure rate: 60%), and Lee et al, concluding
doubled (although not statistically significant) incidence of that patients with evidence of appendicolith on imaging had an initial

10 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Annals of Surgery  Volume XX, Number XX, Month 2019 Nonoperative Management of Appendicitis

failure rate of NOM more than twice that of patients without an analyzed, AT represented the most cost-saving option ($ 4074 vs
appendicolith.10,30,32 $ 5117).
The theory hypothesizing that perforated appendicitis might Our meta-analysis demonstrated that AT and ST had an
be a different disease entity from uncomplicated appendicitis, rather equivalent length of primary hospital stay. In non-RCT studies,
than being the natural evolution of the disease, has some support in a AT resulted in a significantly shorter length of stay, whereas in
recent meta-analysis demonstrating that delaying appendectomy for RCTs it did not. A possible explanation for these results is that, in the
up to 24 hours after admission does not appear to be a risk factor for context of experimental trials, patients randomized to NOM with
complicated appendicitis, postoperative morbidity, or surgical-site antibiotics have a predetermined length of hospitalization. Indeed, in
infection.42 European trials, patients randomized to antibiotic-first therapy were
In our meta-analysis, although the odds ratio for complicated required to be hospitalized for 3 days, whereas the general trend in
appendicitis was doubled in adult patients undergoing surgery fol- the USA is toward shorter hospitalization.27 Even when analyzing the
lowing failure of NOM with antibiotics, the difference did not reach outcome for patients who required further hospitalizations for treat-
statistical significance. In the subgroup analysis of pediatric patients, ing surgical complications or recurrent episodes of acute appendici-
where all the participants underwent preliminary ultrasound or CT tis, the total length of stay did not differ between the 2 groups.
scan, the odds ratio was 1.17, demonstrating comparable risk of The implementation of treatment and follow-up protocols
complicated appendicitis between AT and ST. With the caveat that based on outpatient antibiotic management, and new evidence
some of the studies included in this met-analysis are derived from indicating safety and feasibility of same-day laparoscopic appendec-
low quality scientific evidence, our pooled results indicate that a trial tomy for uncomplicated appendicitis may result in optimization of
of antibiotic therapy does not lead to a statistically significant the resource used by reducing inpatient admissions and hospital costs
increased risk of developing diffuse peritonitis. for both nonoperative and surgical treatment in the future.9,47
When analyzing patients who underwent surgery after failure In recent years, there has been a worldwide increase in
of AT, similar rates of surgical complications were reported in the AT infections caused by multidrug resistant organisms, as a result of
and ST groups (12.8% vs 13.6%), suggesting that the decision to widespread antibiotic use and excessive antimicrobial prescribing
delay appendectomy can be safely made without excess risk of practice. In particular, antibiotic regimens that involve the use of
developing postoperative complications as a result. In the same ampicillin/sulbactam are no longer recommended because of the
way, the systematic review by Gorter et al concluded that children increasing rate of b-lactamase producing Escherichia coli, resulting
with uncomplicated appendicitis who needed a delayed appendec- in a reduction in antibiotic susceptibility.48 Piperacillin-Tazobactam,
tomy for early failure, recurrent appendicitis, or interval appendec- as well as Penem antibiotics, have been used as first-line antibiotic
tomy following NOM did not experience more surgical therapy for patients with uncomplicated appendicitis to address the
complications than those who underwent immediate appendec- increasing rate of E coli resistant to third-generation cephalosporins,
tomy.43 ampicillin, and amoxicillin.6,10,27,28 The use of Ertapenem or Mer-
The incidence of intraoperative finding of complicated appen- openem to ensure that most of the bacteria associated with appendi-
dicitis, however, may not be considered a reliable outcome of safety citis-related infection are adequately covered could lead to increased
in the evaluation of NOM. Complicated appendicitis might already carbapenem resistance, with a major problem in controlling severe
have been present in a percentage of patients at the time of randomi- infections when they occur, especially in patients with neutropenic
zation, and so, the question arises as to whether this could be related sepsis and complicated intra-abdominal infections.
to a lack of accuracy in the diagnostic process, rather than a real Recently, the results of the study entitled ‘‘Randomized
progression of uncomplicated appendicitis on to perforation. Within clinical trial of antibiotic therapy for uncomplicated appendicitis’’
RCTs that used CT scan to reach a precise diagnosis of uncompli- by Park et al challenged the need for even antibiotic therapy in
cated appendicitis, the rate of perforated appendicitis varied from uncomplicated appendicitis and reported promising results regarding
18% in the surgery group in the study by Vons et al [Vons] to 1% in possible spontaneous resolution in some patients. Analysis of the
the trial by Salminen et al.6 This suggests that even CT scan is not primary outcome measure indicated that treatment failure rates in
able to distinguish with absolute certainty uncomplicated from patients presenting with CT-confirmed uncomplicated appendicitis
perforated forms of appendicitis, especially in the absence of clear were similar among those receiving supportive care with either a no-
findings of extraluminal air, increased wall thickness > 3 mm, and antibiotic regimen or a 4-day course of antibiotics, with no difference
intraluminal fecalith.44 in the rates of perforated appendicitis between the 2 groups was
Based on the study published by Wu et al, NOM without reported.49 Whether recovery from nonperforated appendicitis is the
interval appendectomy is the least costly and most effective treat- result of antibiotic therapy or natural clinical remission, and so
ment strategy for adult patients with uncomplicated appendicitis. whether antibiotics are superior to simple supportive care for uncom-
Health economic modeling suggests that surgery would become the plicated appendicitis remains to be established. If future research
preferred strategy only if combined NOM failure during first hospi- demonstrates that antibiotics do not provide any advantage over
talization and recurrence rates exceeded 56%.45 NOM was also the observation alone in uncomplicated appendicitis, this could have a
most cost-effective strategy in children, even when considering major impact in reducing the use of antimicrobial agents, especially
combined rates of rescue appendectomy for antibiotic-first failure in this era of increasing antimicrobial resistance worldwide.
and interval appendectomy set at 41% (which is significantly higher This systematic review and meta-analysis is the largest pub-
than the 18.3%–35.7% rate reported by recent RCTs). According to lished analysis to date on this topic, with outcomes analyzed for both
the same models, NOM would remain cost-effective up to a 1-year adult and pediatric populations. In addition, the GRADE methodol-
recurrence rate of 32.3%.46 ogy has been used to evaluate the quality of the evidence as part of
Successful antibiotic therapy resulted in an approximate 50% our study. A limitation of our study derives from the difficulty of
reduction in costs in our meta-analysis ($ 2509 vs $ 4898). Surpris- establishing appropriate endpoints in order to compare such different
ingly, even when all appendicitis-related care costs (ie, including types of treatment for uncomplicated appendicitis, with mortality
antibiotic treatment failure and subsequent surgery, and costs due to being such a rare event and for example, the need for further
management of surgical complications in both groups) were interventions always going to be lower in the ST group.

ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 11

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Podda et al Annals of Surgery  Volume XX, Number XX, Month 2019

In our meta-analysis, even taking account of the 14.5% of ST patients keen to avoid appendectomy, and that this approach is
group patients suffering postintervention complications, and the successful in the majority of cases. However, patients must also be
27.4% of AT group patients experiencing a failure of NOM with informed that NOM may fail within 24 to 48 hours in about 8% of
antibiotics, both treatment effectiveness and complication-free treat- cases, and an additional 20% might need a second hospitalization
ment success rates were better in the ST group, with the obvious for recurrent AA. NOM for uncomplicated appendicitis does not
advantage that, once the inflamed appendix has been removed, the statistically increase the perforation rate in adult and pediatric
objective of the complete and long-term avoidance of further appen- patients receiving antibiotic treatment, and the decision to delay
dicitis has been achieved. Nevertheless, NOM with antibiotics might appendectomy does not result in increased risk of postoperative
spare a surgical procedure, and thus its potentially related compli- complications.
cations, in more than 70% of cases. Little evidence on quality of life While this systematic review and meta-analysis presents evi-
or patient experience may also represent relevant limitations. dence to suggest that NOM with antibiotic is safe and, in most cases,
Another important limitation of the present systematic review effective, properly powered and well-constructed studies are still
with meta-analysis is the small number of well-designed RCTs that required to establish the optimal management strategy for treating
have reported on this subject to date, especially regarding pediatric uncomplicated appendicitis.
patients. Overall quality of evidence for each of the outcome
measures was very low to moderate, and statistical heterogeneity REFERENCES
was high for the main outcomes. As diagnostic workup and treatment 1. Gwynn LK. The diagnosis of acute appendicitis: clinical assessment versus
of uncomplicated appendicitis varies widely across different coun- computed tomography evaluation. J Emerg Med. 2001;21:119–123.
tries and hospitals, drawing up definitive and universal recommen- 2. Ferris M, Quan S, Kaplan BS, et al. The global incidence of appendicitis: a
dations on NOM remains elusive. For example, the study by Van systematic review of population-based studies. Ann Surg. 2017;266:237–241.
Rossem et al showed that preoperative imaging was conducted in 3. Di Saverio S, Birindelli A, Kelly MD, et al. WSES Jerusalem guidelines for
diagnosis and treatment of acute appendicitis. World J Emerg Surg.
32.8% of UK patients with suspected acute appendicitis, in contrast 2016;11:34.
to 99.5% of patients in the Netherlands. 4. Podda M, Cillara N, Di Saverio S, et al., ACOI (Italian Society of Hospital
This led to a large difference in the normal appendectomy rate Surgeons) Study Group on Acute Appendicitis. Antibiotics-first strategy for
(20.6% in the UK vs 3.2% in the Netherlands). As NOM for uncomplicated acute appendicitis in adults is associated with increased rates of
uncomplicated appendicitis requires preoperative imaging, adoption peritonitis at surgery. A systematic review with meta-analysis of randomized
controlled trials comparing appendectomy and non-operative management
of this approach will require re-evaluation of management algo- with antibiotics. Surgeon. 2017;15:303–314.
rithms to include imaging in some healthcare systems.50 5. Harnoss JC, Zelienka I, Probst P, et al. Antibiotics versus surgical therapy for
Once considered one of the most significant limitations to uncomplicated appendicitis: systematic review and meta-analysis of con-
the widespread adoption of NOM for uncomplicated appendicitis, trolled trials (PROSPERO 2015: CRD42015016882). Ann Surg.
the lack of investigation on the long-term clinical efficacy of 2017;265:889–900.
antibiotics has been addressed in 2018, when Salminen et al 6. Salminen P, Paajanen H, Rautio T, et al. Antibiotic therapy vs appendectomy
for treatment of uncomplicated acute appendicitis: the APPAC randomized
published the 5-year follow-up results of the APPAC randomized clinical trial. JAMA. 2015;313:2340–2348.
clinical trial. The authors demonstrated that, among patients who 7. Styrud J, Eriksson S, Nilsson I, et al. Appendectomy versus antibiotic
were initially treated with antibiotics for uncomplicated appendi- treatment in acute appendicitis: a prospective multicenter randomized con-
citis, the likelihood of late recurrence within 5 years was 39.1%, trolled trial. World J Surg. 2006;30:1033–1037.
with only 2.3% of patients submitted to surgery for recurrent 8. Vons C, Barry C, Maitre S, et al. Amoxicillin plus clavulanic acid versus
appendicitis diagnosed with complicated forms of the disease, appendicectomy for treatment of acute uncomplicated appendicitis: an open-
label, non-inferiority, randomised controlled trial. Lancet. 2011;377:1573–
and overall complication rate significantly reduced in the antibiotic 1579.
group compared to the appendectomy group (6.5% vs 24.4%, P < 9. Talan DA, Saltzman DJ, Mower WR, et al., Olive View–UCLA Appendicitis
0.001). This long-term follow-up supported the feasibility of NOM Study Group. Antibiotics-first versus surgery for appendicitis: a US pilot
with antibiotics as an alternative to surgery for uncomplicated randomized controlled trial allowing outpatient antibiotic management. Ann
appendicitis.51 Emerg Med. 2017;70:1–11.
As the superiority of one therapy over the other cannot be 10. Svensson JF, Patkova B, Almström M, et al. Nonoperative treatment with
antibiotics versus surgery for acute nonperforated appendicitis in children: a
established due to the fundamental difference of such treatment pilot randomized controlled trial. Ann Surg. 2015;261:67–71.
strategies in the management of uncomplicated appendicitis, further 11. Gorter RR, van der Lee JH, Heijsters FACJ, et al., APAC Study Group.
scientific efforts should be focused on the attempt to provide Outcome of initially nonoperative treatment for acute simple appendicitis in
surgeons with clinical, laboratory, and radiological scores which children. J Pediatr Surg. 2018;53:1849–1854.
allow the early identification of those patients who might respond 12. Gorter RR, Eker HH, Gorter-Stam MA, et al. Diagnosis and management of
well to NOM with antibiotics. acute appendicitis. EAES consensus development conference 2015. Surg
Endosc. 2016;30:4668–4690.
In the meantime, the results of the present meta-analysis may
13. Wilms IM, de Hoog DE, de Visser DC, et al. Appendectomy versus antibiotic
well have the potential to guide and change practice patterns toward treatment for acute appendicitis. Cochrane Database Syst Rev.
more rational and stratified diagnostic and therapeutic pathways, 2011;(11):CD008359.
based on current evidence, and avoiding simplistic ‘‘one size fits 14. Sallinen V, Akl EA, You JJ, et al. Meta-analysis of antibiotics versus
all’’ approaches. appendicectomy for non-perforated acute appendicitis. Br J Surg. 2016;103:
656–667.
CONCLUSIONS 15. Findlay JM, Kafsi JE, Hammer C, et al. Nonoperative management of
appendicitis in adults: a systematic review and meta-analysis of randomized
This study provides evidence required for clinicians to pro- controlled trials. J Am Coll Surg. 2016;223:814–824.
vide evidence-based advice to patients with radiologically con- 16. Sakran JV, Mylonas KS, Gryparis A, et al. Operation versus antibiotics—the
firmed uncomplicated appendicitis regarding both antibiotic ‘‘appendicitis conundrum’’ continues: a meta-analysis. J Trauma Acute Care
Surg. 2017;82:1129–1137.
therapy and surgical appendectomy as primary treatment options.
17. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting
Patients should be advised that, although limited by a lower efficacy systematic reviews and meta-analyses of studies that evaluate healthcare
rate compared to surgery, NOM with antibiotics is a safe option for interventions: explanation and elaboration. BMJ. 2009;339:b2700.

12 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: D.C.; ANNSURG-D-18-02034; Total nos of Pages: 13;
ANNSURG-D-18-02034

Annals of Surgery  Volume XX, Number XX, Month 2019 Nonoperative Management of Appendicitis

18. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies 36. Poillucci G, Mortola L, Podda M, et al. ACTUAA-R Collaborative Working
in epidemiology: a proposal for reporting. Meta-analysis Of Observational Group on Acute Appendicitis. Laparoscopic appendectomy vs antibiotic
Studies in Epidemiology (MOOSE) group. JAMA. 2000;283:2008–2012. therapy for acute appendicitis: a propensity score-matched analysis from a
19. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews multicenter cohort study. Updates Surg. 2017;69:531–540.
of Interventions Version 5.3.5. The Cochrane Collaboration, 2014. Available 37. Hansson J, Körner U, Khorram-Manesh A, et al. Randomized clinical trial of
at www.handbook.cochrane.org. Accessed August 2018. antibiotic therapy versus appendicectomy as primary treatment of acute
20. Sterne JA, Hernán MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk appendicitis in unselected patients. Br J Surg. 2009;6:473–481.
of bias in non-randomised studies of interventions. BMJ. 2016;355:i4919. 38. Telem DA. Shared decision making in uncomplicated appendicitis: it is time to
21. Balshem H, Helfand M, Schünemann HJ, et al. GRADE guidelines: 3. Rating include nonoperative management. JAMA. 2016;315:811–812.
the quality of evidence. J Clin Epidemiol. 2011;64:401–406. 39. Hanson AL, Crosby RD, Basson MD. Patient preferences for surgery or
22. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the antibiotics for the treatment of acute appendicitis. JAMA Surg. 2018;153:471–
median, range, and the size of a sample. BMC Med Res Methodol. 2005;5:13. 478.
23. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 40. Hansson J, Khorram-Manesh A, Alwindawe A, et al. A model to select
1986;7:177–188. patients who may benefit from antibiotic therapy as the first line treatment of
acute appendicitis at high probability. J Gastrointest Surg. 2014;18:961–967.
24. Eriksson S, Granström L. Randomized controlled trial of appendicectomy
versus antibiotic therapy for acute appendicitis. Br J Surg. 1995;82:166–169. 41. Loftus TJ, Brakenridge SC, Croft CA, et al. Successful nonoperative man-
agement of uncomplicated appendicitis: predictors and outcomes. J Surg Res.
25. Turhan AN, Kapan S, Kutukcu E, et al. Comparison of operative and non
2018;222:212–218.
operative management of acute appendicitis. TJTES. 2009;15:459–462.
42. van Dijk ST, van Dijk AH, Dijkgraaf MG, et al. Meta-analysis of in-hospital
26. Park HC, Kim MJ, Lee BH. The outcome of antibiotic therapy for uncompli-
delay before surgery as a risk factor for complications in patients with acute
cated appendicitis with diameters  10 mm. Int J Surg. 2014;12:897–900.
appendicitis. Br J Surg. 2018;105:933–945.
27. Minneci PC, Mahida JB, Lodwick DL, et al. Effectiveness of patient choice in
43. Gorter RR, The SML, Gorter-Stam MAW, et al. Systematic review of
nonoperative vs surgical management of pediatric uncomplicated acute
nonoperative versus operative treatment of uncomplicated appendicitis. J
appendicitis. JAMA Surg. 2016;151:408–415.
Pediatr Surg. 2017;52:1219–1227.
28. Hartwich J, Luks FI, Watson-Smith D, et al. Nonoperative treatment of acute
44. Verma R, Grechushkin V, Carter D, et al. Use and accuracy of computed
appendicitis in children: a feasibility study. J Pediatr Surg. 2016;51:111–116.
tomography scan in diagnosing perforated appendicitis. Am Surg.
29. Mahida JB, Lodwick DL, Nacion KM, et al. High failure rate of nonoperative 2015;81:404–407.
management of acute appendicitis with an appendicolith in children. J Pediatr
45. Wu JX, Dawes AJ, Sacks GD, et al. Cost effectiveness of nonoperative
Surg. 2016;51:908–911.
management versus laparoscopic appendectomy for acute uncomplicated
30. Tanaka Y, Uchida H, Kawashima H, et al. Long-term outcomes of operative appendicitis. Surgery. 2015;158:712–721.
versus nonoperative treatment for uncomplicated appendicitis. J Pediatr Surg.
46. Wu JX, Sacks GD, Dawes AJ, et al. The cost-effectiveness of nonoperative
2015;50:1893–1897.
management versus laparoscopic appendectomy for the treatment of
31. Allievi N, Harbi A, Ceresoli M, et al. Acute appendicitis: still a surgical acute, uncomplicated appendicitis in children. J Pediatr Surg.
disease? Results from a propensity score-based outcome analysis of conser- 2017;52:1135–1140.
vative versus surgical management from a prospective database. World J Surg.
47. Farach SM, Danielson PD, Walford NE, et al. Same-day discharge after
2017;41:2697–2705.
appendectomy results in cost savings and improved efficiency. Am Surg.
32. Lee SL, Spence L, Mock K, et al. Expanding the inclusion criteria for 2014;80:787–791.
nonoperative management of uncomplicated appendicitis: outcomes and cost.
48. Pitout JD, Laupland KB. Extended-spectrum beta-lactamase-producing Enter-
J Pediatr Surg. 2017 [Epub ahead of print].
obacteriaceae: an emerging public-health concern. Lancet Infect Dis.
33. Armstrong J, Merritt N, Jones S, et al. Non-operative management of early, 2008;8:159–166.
acute appendicitis in children: is it safe and effective? J Pediatr Surg.
49. Park HC, Kim MJ, Lee BH. Randomized clinical trial of antibiotic therapy for
2014;49:782–785.
uncomplicated appendicitis. Br J Surg. 2017;104:1785–1790.
34. Koike Y, Uchida K, Matsushita K, et al. Intraluminal appendiceal fluid is a
50. van Rossem CC, Bolmers MD, Schreinemacher MH, et al. Diagnosing acute
predictive factor for recurrent appendicitis after initial successful non-opera-
appendicitis: surgery or imaging? Colorectal Dis. 2016;18:1129–1132.
tive management of uncomplicated appendicitis in pediatric patients. J Pediatr
Surg. 2014;49:1116–1121. 51. Salminen P, Tuominen R, Paajanen H, et al. Five-year follow-up of antibiotic
therapy for uncomplicated acute appendicitis in the APPAC randomized
35. Mudri M, Coriolano K, Bütter A. Cost analysis of nonoperative management
clinical trial. JAMA. 2018;320:1259–1265.
of acute appendicitis in children. J Pediatr Surg. 2017;52:791–794.

ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 13

Copyright © 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
View publication stats