REVIEWARTICLE

Laparoscopic Versus Open Appendectomy for Complicatedand Uncomplicated Appendicitis in Children

Sheraz R. Markar & Simon Blackburn & Richard Cobb &

Alan Karthikesalingam & Jessica Evans & James Kinross &

Omar Faiz

Received: 27 April 2012 /Accepted: 1 July 2012 /Published online: 19 July 2012# 2012 The Society for Surgery of the Alimentary Tract

AbstractBackground Appendectomy is one of the most common emergency operations performed in the pediatric population. Theaim of this pooled analysis is to compare the outcome from complicated appendicitis (CA) and uncomplicated appendicitis(UA) following laparoscopic appendectomy (LA) and open appendectomy (OA) in children.Methods A systematic literature search was performed. Primary outcome measures were incidence of complications, intra-abdominal abscess, and wound infection. Secondary outcomes were length of operation, length of hospital stay, incidence ofbowel obstruction, and readmission.Results Seventy-three thousand one hundred fifty appendectomies for UA and 34,474 appendectomies for CAwere included.For UA, the only significant difference between the groups was a reduced length of hospital stay following LA. LA in CAwas associated with reduced complications (pooled odds ratio [POR]00.53; P<0.05), wound infections (POR00.42;P<0.05), length of hospital stay (WMD0−0.67; P<0.05), and bowel obstruction episodes (POR00.8; P<0.05), but anincreased incidence of intra-abdominal abscess and length of operation.Conclusion Pooled analysis demonstrates that, in children with uncomplicated acute appendicitis, LA is associated with areduced hospital stay but broad equivalence in postoperative morbidity when compared with the conventional approach.Although overall morbidity is reduced when the laparoscopic approach is utilized, in cases of CA, the risk of intra-abdominalabscess is increased.

Keywords Laparoscopy . Appendectomy . Pediatrics .

Children . Appendicitis

Introduction

Acute appendicitis is a common cause of acute abdominalpain in the Western World,1 with a reported lifetimeincidence of 8 %.2 As a direct result, appendectomy isone of the most frequently performed surgical procedures.The open approach to appendectomy was originally de-scribed by McBurney.3 Since the advent of laparoscopy,appendectomy has increasingly been performed using aminimally invasive approach, following the first reportby Semm in 1983.4

Previous meta-analyses have demonstrated a reducedincidence of surgical site infection and length of hospitalstay following laparoscopic appendectomy (LA) in adults.5,6

Some studies, however, have suggested that LA is associated

S. R. Markar (*) : J. Kinross :O. FaizDepartment of Academic Surgery, St Mary’s Hospital,Imperial College London,Praed Street,London, UKe-mail: sheraz_markar@hotmail.com

S. BlackburnDepartment of Paediatric Surgery, Southampton General Hospital,Southampton, UK

R. Cobb : J. EvansDepartment of General Surgery, Kingston Hospital,London, UK

A. KarthikesalingamDepartment of Outcomes Research, St George’s Hospital,London, UK

O. FaizSt Mark’s Hospital,Northwick Park,London, UK

J Gastrointest Surg (2012) 16:1993–2004DOI 10.1007/s11605-012-1962-y

with higher rates of intra-abdominal abscess formation, longeroperative times, and higher surgical costs when compared toits open counterpart.7,8

LA, however, is currently not universally accepted as thestandard of care for the treatment of acute appendicitis inchildren,9,10 and differences in the patient population meanthat direct extrapolation of adult data to children is invalid.Children have the potential to benefit greatly from thepresumed benefits of a laparoscopic approach, includingreduced postoperative complications, earlier mobilization,and earlier discharge from the hospital.11 The use ofLA for complicated or perforated appendicitis in childrenremains controversial, with some studies suggesting ahigher intra-abdominal abscess rate when compared toopen appendectomy (OA) 12.

The aim of this pooled analysis is to comprehensivelyreview the current literature comparing OA and LA inpediatric patients. In order to guide future managementdecisions, outcome from complicated appendicitis (CA)and uncomplicated appendicitis (UA) will be compared.

Materials and Methods

A systematic literature search of MEDLINE (January 1950–February 2012), Embase (January 1974–February 2012),Web of Science (January 1990–February 2012), and theCochrane Library (2011 Issue 7) databases was performed.The search terms “laparoscopy,” “appendicectomy,” “ap-pendectomy,” “children,” “paediatric,” and “pediatric” andthe medical subject headings “pediatrics,” “laparoscopy,”“appendectomy,” “appendicitis,” “evidence-based medicine,”and “evidence-based practice” were used in combination withBoolean operators AND or OR. The electronic search wassupplemented by a hand-search of published abstracts fromthe European Association for Endoscopic Surgery, theSurgical Research Society, the Society of Academic andResearch Surgery, the Association of Surgeons of GreatBritain and Ireland, the American College of Surgeons, andthe Society of American Gastrointestinal and EndoscopicSurgeons. Reference lists of all relevant studies were reviewedand the search included the Current Controlled Trials Register(http://www.controlled-trials.com).

Abstracts of citations identified by the search werescrutinized by two independent observers to determineeligibility for inclusion in this pooled analysis. To ensurethis pooled analysis reflected current surgical practice,only studies published since 2000 were considered forinclusion. Studies were included if they met each of thefollowing criteria: comparative or randomized controlledtrials, patients divided into LA and OA groups, studiesfocused on a pediatric population, studies with a cleardescription of outcomes for CA and UA groups, and

studies that included results for at least two of the out-comes under investigation in this analysis. Trials withoutdivision of patients into two groups (laparoscopic andopen surgery), studies in adults (an average patient age>15 years), and studies with no clear distinction betweenCA and UA were excluded. CA in this analysis wasdefined as appendicitis with perforation, gangrene, orabscess formation.

Primary outcome measures were postoperative complica-tions, intra-abdominal abscess, and wound infection (withinthe first month of surgery as a direct result of the initialoperation). Secondary outcomes were length of operation(in minutes), length of hospital stay (in days), postoperativebowel obstruction (described within the studies with a rangeof follow-up from 1 month to 3 years), and readmission(described within the studies with a range of follow-up from1 month to 3 years).

Statistical Analyses

Data from eligible trials were entered into a computerizedspreadsheet for analysis. The quality of each trial wasassessed using the Oxford Centre for Evidence-BasedMedicine—Levels of Evidence (March 2009).13 StatsDirect2.5.7 (StatsDirect, Altrincham, UK) was used for statisticalanalysis. Weighted mean differences with 95 % confidenceintervals (CI) were calculated to assess the size of the effectof LA on continuous variables (length of operation andlength of hospital stay). Pooled odds ratio (POR) with95 % CI was calculated for the effect of LA on discretevariables (postoperative complications, intra-abdominalabscess, wound infections, bowel obstruction, and readmis-sion). Pooled outcome measures were determined usingrandom effects models as described by DerSimonian andLaird.14 Heterogeneity among trials was assessed byCochran Q statistic,15 a null hypothesis test in which P<0.05 is taken to indicate the presence of significant hetero-geneity. The Egger test was used to assess the funnel plot forsignificant asymmetry, indicating possible publication orother biases.

Results

The initial search identified 43 publications (Fig. 1). Afterscreening, 19 publications that met the inclusion criteriawere identified,16–34 with 3 randomized controlledstudies.20,24,29 One further publication was excluded as thisfailed to provide the number of patients with CA and UA inthe study.17 Eleven publications compared LA and OA inchildren for UA. Seventy-three thousand one hundred fiftyappendectomies for UA were included in this analysis, ofwhich 27,249 were laparoscopic and 45,901 were open.

1994 J Gastrointest Surg (2012) 16:1993–2004

Sixteen publications were included that compared LA vs.OA in children for CA. Thirty-four thousand four hundredseventy-four appendectomies for CA were included in thisanalysis, of which 10,965 were laparoscopic and 23,509were open. Table 1 describes the basic demographic datafrom each study, along with a description of pediatricpatients age included in this study.

Primary Outcome Measures

Postoperative Complications

Six trials 16,18,20,23,24,27 reported the incidence of postoper-ative complications for LA and OA for UA in children.Twenty-six (2.62 %) and 20 (5.74 %) complications wereseen in the laparoscopic and open groups, respectively.There was no significant difference between the groups(POR01.09; 95 % CI00.51 to 2.3; P00.83) (Fig. 2). Therewas no evidence of statistical heterogeneity (Cochran

Q02.54; P00.77; I200 %); however, there was evidenceof statistical bias (Egger01.77; P<0.05).

Ten publications 16,18,23,24,27–30,32,33 reported the inci-dence of postoperative complications following LA andOA for CA in children. Ninety-three (15.30 %) and 132(29.33 %) complications were seen in the laparoscopic andopen groups, respectively. LA was associated with a signif-icantly reduced incidence of postoperative complications(POR00.53; 95 % CI00.28 to 0.99; P<0.05) (Fig. 3). Therewas evidence of statistical heterogeneity (Cochran Q0

25.31; P<0.05; I2064.4 %). There was no evidence ofstatistical bias (Egger0−1.55; P00.34).

Intra-abdominal Abscess

Nine trials 16,18–21,23–25,27 reported the incidence of intra-abdominal abscess following LA and OA for UA in chil-dren. One hundred fifty-three (0.57 %) and 180 (0.40 %)incidences of postoperative intra-abdominal abscess were

PRISMA Flow Diagram

Records identified throughdatabase searching

(n = 43)

Scr

een

ing

Incl

ud

edE

ligib

ility

Iden

tifi

cati

on

Additional records identifiedthrough other sources

(n = 0)

Records after duplicates removed(n = 43)

Records screened(n = 43)

Records excluded as notfocused on a pediatric

cohort or no comparisonbetween laparoscopic and

open groups(n = 24)

Full-text articles assessedfor eligibility

(n = 19) Full-text articles excluded,as did not provide

numbers in complicatedand uncomplicated groups

(n = 1)Studies included inqualitative synthesis

(n = 18)

Studies included inquantitative synthesis

(pooled-analysis)(n = 18)

Fig. 1 PRISMA flowchartdescribing the literature searchemployed

J Gastrointest Surg (2012) 16:1993–2004 1995

seen in the laparoscopic and open groups, respectively.There was no significant difference between the groups(POR01.01; 95 % CI00.51 to 2; P00.97) (Fig. 4). Therewas evidence of significant statistical heterogeneity (CochranQ020.14; P<0.05; I2070.2 %). However, there was noevidence of statistical bias (Egger0−0.26; P00.81).

Fourteen publications 16,18,19,21,23,24,27–34 reported theincidence of postoperative intra-abdominal abscess follow-ing LA and OA for CA in children. Three hundred ninety-seven (3.69 %) and 603 (2.59 %) incidences of postopera-tive intra-abdominal abscess were seen in the laparoscopic

and open groups, respectively. LA was associated with asignificantly increased incidence of intra-abdominal abscessformation (POR01.32; 95 % CI01.15 to 1.5; P<0.05)(Fig. 5). There was no evidence of statistical heterogeneity(CochranQ011.95; P00.53) or bias (Egger0−0.18; P00.59).

Wound Infection

Nine trials 16,18–21,23–25,27 reported the incidence of woundinfection following LA and OA for UA in children. Twohundred four (0.76 %) and 453 (0.99 %) incidences of

Table 1 Demographic data

Author Age (LA) (years) Age (OA) (years) Study age range (years) M/F ratio (LA) M/F ratio (OA)

Canty 16 – – 1–19 – –

Ikeda 18 9 (2–15) 10 (2–15) 2–15 29:24 31:16

Jen 19 12.4±4.1 11.1±4.3 – 18,559:13,626 40,202:23,418

Lavonius 20 12.5 (7–15) 11.9 (8–15) 7–15 – –

Lee 21 12.8±3.2 10.4±3.7 0.5–17 2,071:1,480 2,597:1,502

Li 22 8.07±0.3 7.85±0.24 3–15 48:21 66:25

McKinlay 23 144 months 105 months – 115:89 74:45

Oka 24 11±4.3 10.6±3.5 2–20 79:62 227:149

Serour 25 11.2±3.3a 11.2±3.3a 2–18 1:1.1 1.9:1

Tsao 26 11±3.7 9.2±5.1 – 355:273 301:176

Vengunta 27– – – – –

Faruquzzaman 28 8.7±1.08 9±1.15 4–12 26:25 28:33

Lintula 29 10±3 11±3 4–15 9:4 9:3

Miyano 30 7.6±1.15 7.7±1.35 2–13 5:7 8:3

Nataraja 31 11 (2–16) 11 (1–16) 1–16 24:15 90:61

Nwokoma 32 10.2±1.29 9.7±1.41 3–15 12:11 10:7

Wang 33 7.4±4.3 6.6±3.2 – 45:35 28:20

Yagmurlu 34 10.2±0.5 8.6±0.5 2–18 2.5:1 2.3:1

aMean age of all patients undergoing appendectomy included in the study

Fig. 2 Forrest plot for theeffect of LA on postoperativecomplications in childrenwith UA

1996 J Gastrointest Surg (2012) 16:1993–2004

postoperative wound infection were seen in the laparoscopicand open groups, respectively. There was no significantdifference between the groups (POR00.6; 95 % CI00.31to 1.17; P00.13) (Fig. 6). There was evidence of statisticalheterogeneity (Cochran Q023.21; P<0.05; I2074.1 %) butno evidence of statistical bias (Egger0−0.24; P00.83).

Thirteen publications 16,18,19,21,23,24,27–30,32–34 reportedthe incidence of postoperative wound infection followingLA and OA for CA in children. Three hundred fifty-nine(3.34 %) and 952 (4.09 %) wound infections were seen inthe laparoscopic and open groups, respectively. LA wasassociated with a significantly reduced incidence of woundinfection (POR00.42; 95 % CI00.27 to 0.67; P<0.05)(Fig. 7). There was evidence of significant statistical hetero-geneity (Cochran Q028.25; P<0.05; I2057.5 %) and bias(Egger0−1.5; P<0.05) (Tables 2 and 3).

Secondary Outcome Measures

Length of Operation (in Minutes)

Six publications 16,18,20,22,23,27 reported the length of oper-ation for LA and OA for UA in children. There was nosignificant difference between the groups (weighted meandifference0−5.03; 95 % CI0−32.47 to 22.42; P00.72).There was evidence of significant statistical heterogeneity(Cochran Q0320.71; P<0.05). There were insufficient datato enable calculation of statistical bias.

Eleven publications 16,18,22,23,27–30,32–34 reported the lengthof operation for LA and OA for CA in children. LA wasassociated with a significantly increased length of operation(weight mean difference02.3; 95%CI01.25 to 3.34;P<0.05).There was significant evidence of statistical heterogeneity

Fig. 3 Forrest plot for theeffect of LA on postoperativecomplications in childrenwith CA

Fig. 4 Forrest plot for theeffect of LA on postoperativeintra-abdominal abscessformation in childrenwith UA

J Gastrointest Surg (2012) 16:1993–2004 1997

(Cochran Q0154.32; P<0.05; I2095.5 %). There was noevidence of statistical bias (Egger06.08; P00.07).

Length of Hospital Stay (in Days)

Eight publications 16,18–23,27 reported the length of hospitalstay following LA and OA for UA in children. LA wasassociated with a significantly reduced length of hospitalstay (weighted mean difference0−1.18; 95 % CI0−1.61

to −0.74; P<0.05). There was evidence of statisticalheterogeneity (Cochran Q0294.01; P<0.05; I2099 %).There was no significant evidence of statistical bias(Egger0−10.4; P00.06).

Twelve publications 16,18,19,21,22,27–30,32–34 reported thelength of hospital stay for LA and OA for CA in children.LA was associated with a significantly reduced lengthof hospital stay (weighted mean difference=−0.67;95 % CI0−0.95 to −0.4; P<0.05). There was evidence of

Fig. 5 Forrest plot for theeffect of LA on postoperativeintra-abdominal abscessformation in childrenwith CA

Fig. 6 Forrest plot for theeffect of LA on postoperativewound infection in childrenwith UA

1998 J Gastrointest Surg (2012) 16:1993–2004

significant statistical heterogeneity (Cochran Q0152.41;P<0.05; I2094.1 %) and bias (Egger0−3.22; P<0.05).

Bowel Obstruction

Six publications 16,18,19,23,24,26 reported the incidence ofpostoperative bowel obstruction following LA and OA forUA in children. Sixty-one (0.25 %) and 100 (0.23 %) ofpatients developed postoperative bowel obstruction followingsurgery in the laparoscopic and open groups, respectively.

There was no significant difference between the groups(POR01.09; 95 % CI00.79 to 1.5; P00.6). There was noevidence of statistical heterogeneity (Cochran Q00.15;P00.93; I200 %). There were insufficient data to allowcalculation of statistical bias.

Nine publications 16,18,19,23,24,26,30,32,34 reported the inci-dence of postoperative bowel obstruction following LAand OA for CA in children. One hundred twenty-two(1.23 %) and 325 (1.48 %) patients were affected bypostoperative bowel obstruction following LA and OA,

Fig. 7 Forrest plot for theeffect of LA on postoperativewound infection in childrenwith CA

Table 2 Primary outcomes for children with UA

Author Patient no.(LA)

Patient no.(OA)

Complications(LA)

Complications(OA)

IA abscess(LA)

IA abscess(OA)

Wound infection(LA)

Wound infection(OA)

Canty 16 653 86 8 1 5 1 3 0

Ikeda 18 31 25 2 0 0 0 1 0

Jen 19 22,939 42,274 – – 72 75 146 313

Lavonius 20 17 15 2 0 0 0 0 0

Lee 21 2,711 2,605 – – 71 98 48 129

Li 22 45 67 – – – – – –

McKinlay 23 158 49 5 1 0 1 3 0

Oka 24 96 262 4 16 2 4 2 9

Serour 25 121 217 – – 0 1 0 0

Tsao 26 442 285 – – – – – –

Vengunta 27 36 16 5 2 3 0 1 2

Total (%) 27,249 45,901 26 (2.62) 20 (5.74) 153 (0.57) 180 (0.4) 204 (0.76) 453 (0.99)

IA abscess intra-abdominal abscess

J Gastrointest Surg (2012) 16:1993–2004 1999

respectively. LAwas associated with a significantly reducedincidence of postoperative bowel obstruction (POR00.8;95 % CI00.65 to 0.99; P<0.05). There was no evidenceof statistical heterogeneity (Cochran Q05.98; P00.65;I200 %) or bias (Egger0−0.19; P00.61).

Readmission

Six publications 19–21,24–26 reported the incidence ofreadmission with a follow-up ranging from 1 month to3 years following LA and OA for UA in children. Threehundred twelve (1.19 %) and 554 (1.21 %) patients werereadmitted following surgery in the laparoscopic andopen groups, respectively. There was no significant dif-ference between the groups (POR00.68; 95 % CI00.32to 1.42; P00.3). There was significant evidence ofstatistical heterogeneity (Cochran Q025.28; P<0.05;I2084.2 %). There was no evidence of statistical bias(Egger0−1.15; P00.53).

Five publications 19,21,24,26,29 reported the incidence ofreadmission with a follow-up period ranging from 1 monthto 3 years following LA and OA for CA in children. Threehundred ninety-eight (3.85 %) and 946 (4.10 %) readmis-sions were recorded for the laparoscopic and open groups,respectively. There was no statistically significant differencebetween the groups (POR00.9; 95 % CI00.74 to 1.09;P00.27). There was no evidence of statistical heteroge-neity (Cochran Q04.62; P00.33; I2013.5 %) or bias(Egger0−0.54; P00.47) (Tables 4 and 5).

Discussion

Analysis of primary outcomes revealed no significant effectof LA upon the incidence of complications of treatment ofUA. Primary outcome analysis for CA, however, revealedthat LA was associated with reduced overall postoperativecomplications and wound infections, but an increased inci-dence of intra-abdominal abscess formation. Postoperativecomplications are described as binary occurrences and notwell characterized in several of the studies included in thisanalysis. There was little distinction made between compli-cations that resulted in a significant impact on the patient,requiring a further intervention and increased length ofhospital stay, and those complications that did not. Thus,although we have demonstrated a clear decrease in postop-erative complications with LA, this result must be inter-preted with an appropriate level of caution due theheterogeneity in the description of these complications.

CA is associated with increased infectious complications,i.e., wound infection and intra-abdominal abscess, whichhave previously been considered a relative contraindicationfor LA.33 The results of this pooled analysis, however, showa significantly reduced incidence of postoperative woundinfection associated with LA in CA (3.34 vs. 4.1 %). Thisfinding could reflect the benefit of minimizing wound sizeor increased use of an endoscopic bag in LA for appendixextraction and, thus, the avoidance of contact between theinfected specimen and the wound.35 This result finds reso-nance in previous studies from large administrative and

Table 3 Primary outcomes in children with CA

Author Patient no.(LA)

Patient no.(OA)

Complications(LA)

Complications(OA)

IA abscess(LA)

IA abscess(OA)

Wound infection(LA)

Wound infection(OA)

Canty 16 302 87 27 7 19 5 0 1

Faruquzzaman 28 51 61 18 42 6 7 2 13

Ikeda 18 22 22 5 6 4 0 2 5

Jen 19 9,246 21,347 – – 266 454 295 761

Lee 21 838 1,393 – – 71 98 48 129

Li 22 15 16 – – – – – –

Lintula 29 13 12 2 2 1 0 0 2

McKinlay 23 46 70 17 23 7 7 5 8

Miyano 30 12 11 3 9 2 2 0 4

Nataraja 31 13 59 – – 0 4 – –

Nwokoma 32 23 17 5 3 1 0 1 2

Oka 24 43 114 9 22 7 10 1 6

Tsao 26 186 192 – – – – – –

Vengunta 27 16 8 4 5 3 1 1 3

Wang 33 80 48 3 13 2 7 1 6

Yagmurlu 34 59 52 – – 8 8 3 12

Total (%) 10,965 23,509 93 (15.30) 132 (29.33) 397 (3.69) 603 (2.59) 359 (3.34) 952 (4.10)

IA abscess intra-abdominal abscess

2000 J Gastrointest Surg (2012) 16:1993–2004

clinical datasets in adults, demonstrating that LA has aprotective effect and reduces the incidence of postoperativewound infection.36,37

A significantly increased incidence of intra-abdominalabscess formation was demonstrated following LA for CA(3.69 vs. 2.59 %). Previous studies have also shown similarfindings of increased intra-abdominal abscess formationfollowing LA in perforated appendicitis in adults 38,39 andchildren.40 During the induction of pneumoperitoneum inLA, the infected contents from a CA may seed across the

abdomen, providing sources for postoperative abscess for-mation. Also, during OA, the appendix is usually dividedexternally and the stump often inverted, thus in theorydecreasing the incidence of intraperitoneal contamination.However, in LA, dissection and division of the appendixoccur within the peritoneal cavity and, thus, represent apotential source of contamination.38 Further studies havesuggested that the incidence of intra-abdominal abscessformation decreases during the learning curve for LA, andthis may represent an inadequacy of laparoscopic peritoneal

Table 4 Secondary outcomes for children with UA

Author Operativetime±SD(LA) (min)

Operativetime±SD(OA) (min)

Length of hospitalstay±SD (LA)(days)

Length of hospitalstay±SD (OA)(days)

Bowelobstruction(LA)

Bowelobstruction(OA)

Readmission(LA)

Readmission(OA)

Canty 16 52 52 2 3 0 0 – –

Ikeda 18 104.25±5.37 60.75±3.94 6.75±1.08 10.5±1 0 0 – –

Jen 19– – 1.9±1.4 2.1±1.3 59 99 251 424

Lavonius 20 40 34 2.5 2.3 – – 0 0

Lee 21– – 1.7±1.5 2.1±1.6 – – 60 125

Li 22 28.73±0.93 43.87±0.68 3.91±0.12 6.37±0.09 – – – –

McKinlay 23 56 44 1.8 2.2 1 0 – –

Oka 24 – – – – 0 0 0 3

Serour 25 – – – – – – 0 1

Tsao 26 – – – – 1 1 1 1

Vengunta 27 55 33 2 2.5 – – – –

Total (%) 61 (0.25) 100 (0.23) 312 (1.19) 554 (1.21)

Table 5 Secondary outcomes for children with CA

Author Operativetime ± SD(LA) (min)

Operativetime ± SD(OA) (min)

Length of hospitalstay ± SD (LA)(days)

Length of hospitalstay ± SD (OA)(days)

Bowelobstruction(LA)

Bowelobstruction(OA)

Readmission(LA)

Readmission(OA)

Canty 16 68 58 7 7 8 1 – –

Faruquzzaman 28 112±36.3 72±30.3 5.7±1.93 7.4±1.86 – – – –

Ikeda 18 111.25±4.02 71±4.12 14±2 16.5±2.16 0 1 – –

Jen 19 – – 5.2±3.2 5.5±3.4 108 307 336 812

Lee 21 – – 5±3 5.7±3.4 – – 60 125

Li 22 55.8±3.2 57.94±0.77 6.33±0.37 8.44±0.24 – – – –

Lintula 29 63±31 37±18 5±1.9 4±1.2 – – 1 0

McKinlay 23 66 50 – – 2 5 – –

Miyano 30 119±5.35 107±5 8.7±1.47 13.3±1.68 1 3 – –

Nataraja 31– – – – – – – –

Nwokoma 32 97±3.51 69.8±3.83 5.1±1.35 4.5±1.68 2 0 – –

Oka 24– – – – 0 1 1 3

Tsao 26– – – – 0 6 0 6

Vengunta 27 68.5 40.5 5.5 7 – – – –

Wang 33 88.5±28.8 71.8±30.6 6.5±2.2 7.8±2.9 – – – –

Yagmurlu 34 60±3 57±3 7.88±0.63 8.75±0.63 1 1 – –

Total (%) 122 (1.23) 325 (1.48) 398 (3.85) 946 (4.10)

J Gastrointest Surg (2012) 16:1993–2004 2001

washout during the procedure.41,42 By limiting the inclusioncriteria to articles published since 2000, there was anattempt to reduce the effect of the learning curve associatedwith LA in children and to give a true representation ofcurrent clinical practice. However, in this analysis, it wasnot possible to account for the experience of the individualsurgeon undertaking the procedure, which represents a con-founding factor that may have influenced the findings pre-sented. The time to diagnosis and time to surgery were notdocumented in any of the studies included in this analysisand, thus, represent another factor that may have influencedthe degree of intraperitoneal contamination. In fact, thedegree of intraperitoneal contamination was not describedin a sufficient number of studies included in this analysis toallow stratification of this as a risk factor. This certainly mayhave affected the results presented.

Secondary outcome analysis revealed that the length ofoperation was significantly increased for LA in CA, butsimilar for LA and OA in UA. Previous studies have dem-onstrated similar findings in adults with an increase inoperative duration seen with LA, especially in cases ofperforated appendicitis.43 This most likely reflects the ergo-nomic challenges associated with laparoscopic surgery inmore challenging cases. Some studies have demonstratedthat, with increased experience, the operative time for a caseof CA is similar for LA and OA,44 this requires focusedstudy in the pediatric population.

Length of hospital stay was significantly reduced for LAin both CA and UA. These results probably reflect thegeneric benefits of minimally invasive surgery, in whichreduced postoperative pain enables earlier mobilizationand subsequent discharge from the hospital.11 Future re-search is required to specifically examine the influence ofa laparoscopic approach on postoperative pain. In the pedi-atric population, it is possible that a swifter return to normalactivities might reduce the psychological effects of hospi-talization, though objective evidence of this is lacking. Thereduced length of hospital stay also provides an additionaleconomic benefit with a likely reduced hospital cost follow-ing LA. Due to limitations in the data provided, however, itwas not possible to analyze hospital cost in this study. Inadults, there is no conclusive evidence to demonstrateincreased or reduced cost associated with LA whencompared to OA.45,46 Future studies must examine thecost-effectiveness of LA in appendicitis in children with areduced hospital stay being offset by the cost of laparoscopicinstrumentation.

The unmeasured benefit of laparoscopy is that it providesboth a diagnostic and therapeutic modality in the setting ofappendicitis, which is of particular benefit in females.Through direct visualization of the abdominal and pelviccontents, a diagnosis of a patient with equivocal signs can bereached and treatment instituted. In prepubescent females,

lower abdominal pain is less attributable to abnormalities inpelvic female organs than in adults, thus the diagnosticadvantages of laparoscopy are reduced in this cohort.

Analysis of postoperative bowel obstruction revealed asignificantly reduced incidence of postoperative bowel ob-struction associated with LA in CA. This is probably due tothe less invasive nature of laparoscopic surgery, reducingthe formation of intraperitoneal adhesions and subsequentbowel obstruction.47 Previous studies have demonstrated aconsiderable incidence of bowel obstruction in adultsfollowing OA.48,49 There was significant heterogeneity inthe length of follow-up employed in each of the studiesincluded in this analysis. Thus, it is inappropriate to drawfirm conclusions based upon this analysis regarding bowelobstruction following LA and OA in children. Long-termfollow-up data are required to gauge the lifetime risk ofbowel obstruction following childhood LA. None of theresults presented in this analysis demonstrate a significanteffect of LA or OA on upon readmission rates.

The age of patients included in the studies under analysiswas highly variable, ranging from 2 to 20 years; however,patient age appeared to be similar between the laparoscopicand open groups (Table 1). Age is an important factor thataffects presentation of appendicitis, children under the ageof 5 often present with atypical symptoms and present latewith perforation. This younger cohort of patients often has amore complicated postoperative course with a prolongedileus and increased risk of intra-abdominal abscess forma-tion, and they may be disproportionately represented inpatients who developed complications in this analysis. Afurther limitation is the disparity of the sample size underinvestigation, and in particular, the effect of the very largecohort studied by Jen et al.19 This may have led to skewingof the results of this pooled analysis and, thus, represents aconfounding variable that must be taken into account in theinterpretation of these results. Patient size and body massindex are also important factors that were not describedsufficiently in the studies and may have some effects onoutcome. There is a learning curve associated with theadoption of laparoscopic surgery, and this may be morepronounced in laparoscopy performed in the confined spaceseen in very small children and neonates. This may repre-sent a further valid reason for the centralization of surgery insmall children and neonates to centers of pediatric surgicalexcellence with highly experienced laparoscopic pediatricand neonatal surgeons.

Conclusion

The available data demonstrate some benefits from LA inthe treatment of acute appendicitis in children, which aremore pronounced in CA. Pooled analysis demonstrates that,

2002 J Gastrointest Surg (2012) 16:1993–2004

in children with uncomplicated acute appendicitis, LA isassociated with a reduced hospital stay but broad equivalencein postoperative morbidity when compared with the conven-tional approach. Although overall morbidity is reduced whenthe laparoscopic approach is utilized, in cases of CA, the riskof intra-abdominal abscess is increased.

Conflicts of Interests None.

References

1. Browse’s introduction to the symptoms and signs of surgicaldisease. 4 ed. Hodder Arnolc 2006.

2. Addiss DG, Shaffer N, Fowler S, Tauxe RV. The epidemiology ofappendicitis and appendectomy in the United States. Am JEpidemiol 1990; 132: 910 –25

3. McBurney C. The incision made in the abdominal wall in the casesof appendicitis, with a description of a new method of operating.Ann Surg 1894; 20: 38 – 43

4. Semm K. Endoscopic appendectomy. Endoscopy 1983; 15: 59-645. Sauerland S, Leferine R, Neugebauer ER. Laparoscopic versus

open surgery for suspected appendicitis. Cochrane Database SystRev 2004; 18: CD001546

6. Bennett J, Boddy A, Rhodes M. Choice of approach forappendicectomy: a meta-analysis of open versus laparoscopicappendicectomy. Surg Laparosc Endosc Percutan Tech 2007;17: 245 – 55.

7. Ortega AE, Hunter JG, Peters JH, Swanstrom LL, Schirmer B. Aprospective, randomized comparison of laparoscopic appendectomywith open appendectomy. Laparoscopic Appendectomy StudyGroup. Am J Surg 1995; 169: 208 – 12

8. Tang E, Ortega AE, Anthone GJ, Beart RW Jr. Intraabdominalabscesses following laparoscopic and open appendectomies. SurgEndosc 1996; 10: 327 – 8

9. Horwitz JR, Custer MD, May BH, Mehall JR, Lally KP. Shouldlaparoscopic appendectomy be avoided for complicated appendicitisin children? J Pediatr Surg 1997; 32 (11): 1601 – 3

10. Faiz O, Blackburn SC, Clark J, Bottle A, Curry JI, Farrands P,Aylin P. Laparoscopic and conventional appendicectomy inchildren: outcomes in English hospitals between 1996 and 2006.Pediatr Surg Int 2008; 24: 1223 – 7

11. Aziz O, Athanasiou T, Tekkis PP, Purkayastha S, Haddow J,Malinovski V, Paraskeva P, Darzi A. Laparoscopic versus openappendectomy in children: a meta-analysis. Ann Surg 2006; 243:17 – 27

12. Pokala N, Sadhasivam S, Kiran RP, Parithivel V. Complicatedappendicitis—is the laparoscopic approach appropriate? Acomparative study with the open approach: outcome in a commu-nity hospital setting. Am Surg 2007; 73: 737 – 41

13. Oxford Centre for Evidence-Based Medicine—Levels of evidence,March 2009

14. DerSimonian R, Laird N. Meta-analysis in clinical trials. ControlClin trials 1986; 7: 177 – 88

15. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 1539 – 58

16. Canty TG, Collins D, Losasso B, Lynch F, Brown C. LaparoscopicAppendectomy for Simple and Perforated appendicitis in Children:The Procedure of Choice? J Pediatr Surg 2000; 35: 1582 – 5

17. Esposito C, Borzi P, Valla JS, Mekki M, Nouri A, Becmeur F, AllalH, Settimi A, Shier F, Sabin MAG, Mastrianni L. Laparoscopicversus Open Appendectomy in Children: A retrospective compar-ative study of 2,332 cases. World J Surg 2007; 31: 750 – 5

18. Ikeda H, Ishimaru Y, Takayasu H, Okamura K, Kisaki Y, Fujino J.Laparoscopic versus Open Appendectomy in Children withUncomplicated and Complicated Appendicitis. J Pediatr Surg2004; 39: 1680 – 5

19. Jen HC, Shew SB, Laparoscopic versus open appendectomy inchildren: outcomes comparison based on a statewide analysis. JSurg Res 2010; 161: 13 – 7

20. Lavonius MI, Liesjarvi S, Ovaska J, Pajulo O, Ristkari S,Alanen M. Laparoscopic versus open appendectomy in chil-dren: a prospective randomised study. Eur J Pediatr Surg2001; 11: 235 – 8

21. Lee SL, Yaghoubian A, Kaji A. Laparoscopic vs Open Appendec-tomy in Children: Outcomes Comparison based on Age, Sex, andPerforation Status. Arch Surg 2011; 146: 1118 – 21

22. Li P, Xu Q, Ji Z, Gao Y, Zhang X, Duan Y, Guo Z, Zheng B, Guo X,Wu X. Comparison of surgical stress between laparoscopic and openappendectomy in children. J Pediatr Surg 2005; 40: 1279 – 83

23. McKinlay R, Neeleman S, Klein R, Stevens K, Greenfeld J, GhoryM, Consentino C. Intraabdominal abscess following open andlaparoscopic appendectomy in the pediatric population. SurgEndosc 2003; 17: 730 – 3

24. Oka T, Kurkchubasche AG, Bussey JGWesselhoeft CW, Tracy TF,Luks FI. Open and laparoscopic appendectomy are equally safeand acceptable in children. Surg Endosc 2004; 18: 242 – 5

25. Serour F,WitzlingM,Gorenstein A. Is laparoscopic appendectomy inchildren associated with an uncommon postoperative complication?Surg Endosc 2005; 19: 912 – 22

26. Tsao KJ, St Peter SD, Valusek PA, Keckler SJ, Sharp S,Holcomb GW 3rd, Snyder CL, Ostile DJ. Adhesive smallbowel obstruction after appendectomy in children: comparisonbetween the laparoscopic and open approach. J Pediatr Surg2007; 42: 939 – 42

27. Vegunta RK, Ali A, Wallace LJ, Switzer DM, Pearl RH. Laparo-scopic appendectomy in children: technically feasible and safe inall stages of acute appendicitis. Am Surg 2004; 70: 198 – 201

28. Faruquzzaman, Mazumder SK. Complicated appendectomy inchildren in relation to laparoscopic vs open procedures. BratislLek Listy 2010; 111: 610 – 5

29. Lintula H, Kokki H, Vanamo K, Antila P, Eskelinen M. Laparos-copy in children with complicated appendicitis. J Pediatr Surg2002; 37: 317 – 20

30. Miyano G, Okazaki T, Kato Y, Marusasa T, Takahashi T, Lane GJ,Yamataka A. Open versus laparoscopic treatment for pan-peritonitissecondary to perforated appendicitis in children: a prospectiveanalysis. J Laparoendosc Adv Surg Tech A 2010; 20: 655 – 7

31. Nataraja RM, Bandi A, Clarke SA, Haddad MJ. Comparison ofintra-abdominal abscess formation following laparoscopic andopen appendicectomy in children. J Laparoendosc Adv Surg TechA 2010; 20: 391 – 4

32. Nwokoma NJ, Swindells MG, Pahl K, Mathur AB, Minocha A,Kurlkarni M, Tsang T. Pediatric advance appendicitis: open versuslaparoscopic approach. Surg Laparosc Endosc Percutan Tech2009; 19: 110 – 3

33. Wang X, Zhang W, Yang X, Shao J, Zhou X, Yuan J. Complicatedappendicitis in children: is laparoscopic appendectomy appropriate?A comparative study with open appendectomy – our experience. JPediatr Surg 2009; 44: 1924 – 7

34. Yagmurlu A, Vernon A, Barnhart DC, Georgeson KE, Harmon CM.Laparoscpic appendectomy for perforated appendicitis: a comparisonwith open appendectomy. Surg Endosc 2006; 20: 1051 – 4

35. Fleming FJ, KimMJ,Messing S, Gunzler D, SalloumR,Monson JR.Balancing the risk of postoperative surgical infections: a multivariateanalysis of factors associated with laparoscopic appendectomy fromthe NSQIP database. Ann Surg 2010; 252: 895 – 900

36. Guller U, Hervey S, Purves H, Muhlbaier LH, Peterson ED,Eubanks S, Pietrobon R. Laparoscopic versus open appendectomy:

J Gastrointest Surg (2012) 16:1993–2004 2003

outcomes comparison based on a large administrative database.Ann Surg 2004; 239: 43 – 52

37. Romy S, Eisenring MC, Bettschart V, Petignat C, Francioli P,Troillet N. Laparoscope using and surgical site infections indigestive surgery. Ann Surg 2008; 247: 627 – 32

38. Krisher SL, Browne A, Dibbins A, Tkacz N, Curci M. Intra-abdominal abscess after laparoscopic appendectomy for perforatedappendicitis. Arch Surg 2001; 136: 438 – 41

39. Paik PS, Towson JA, Anthone GJ, Ortega AE, Simons AJ, BeartRW Jr. Intra-abdominal abscesses following laparoscopic and openappendectomies. J Gastrointest Surg 1997; 1: 188 – 92

40. Horwitz JR, Custer MD, May JR, Mehall JR, Laily KP. Shouldlaparoscopic appendectomy be avoided for complicated appendicitisin children? J Pediatr Surg 1997; 32: 1601 – 3

41. Khalili TM, Hiatt JR, Savar A, Lau C, Marguiles DR. Perforatedappendicitis is not a contraindication to laparoscopy. Am Surg1999; 65: 965 – 7

42. Katkhouda N, Friedlander MH, Grant SW, Achanta KK, Essani R,Paik P, Velmahos G, Campos G, Mason R, Mavor E. Intraabdo-minal abscess rate after laparoscopic appendectomy. Am J Surg2000; 180: 456 – 9

43. Wei B, Qi CL, Chen TF, Zheng ZH, Huang JL, Hu BG, Wei HB.Laparoscopic versus open appendectomy for acute appendicitis: ametaanalysis. Surg Endosc 2011; 25: 1199 – 208

44. Piskun G, Kozik D, Raipal S, Shaftan G, Fogler R. Comparison oflaparoscopic, open, and converted appendectomy for perforatedappendicitis. Surg Endosc 2001; 15: 660 – 2

45. Chu T, Chandhoke RA, Smith PC, Schwaitzberg SD. The impactof surgeon choice on the cost of performing laparoscopicappendectomy. Surg Endosc 2011; 25: 1187 – 91

46. Hsieh CS, Chen YL, Lee MH, Chang HC, Chen ST, Kuo SJ. Alower costly laparoscopic appendectomy: our experience of morethan 2000 cases. Int J Surg 2010; 8: 140 – 3

47. Rosin D, Zmora O, Hoffman A, Khaikin M, Bar Zakai B, Munz Y,Shabtai M, Ayalon A. Low incidence of adhesion-related bowelobstruction after laparoscopic colorectal surgery. J LaparoendoscAdv Surg Tech A. 2007; 17: 604 – 7

48. Andersson RE. Small bowel obstruction after appendicectomy. BrJ Surg 2001; 88: 1387 – 91

49. Leung TT, Dixon E, Gill M, Mador BD, Moulton KM, KaplanGG, MacLean AR. Bowel obstruction following appendectomy:what is the true incidence? Ann Surg 2009; 250: 51 – 3

2004 J Gastrointest Surg (2012) 16:1993–2004