The American Journal of Surgery (2010) 200, 291297

Review

Laparoscopic versus open mesh repair for recurrent

inguinal hernia: a meta-analysis of outcomes
Georgia Dedemadi, M.D.a,*, George Sgourakis, M.D.b, Arnold Radtke, M.D.c,
Alexandros Dounavis, M.D.a, Ines Gockel, M.D.c, Ioannis Fouzas, M.D.d,
Constantine Karaliotas, M.D.b, Evangelos Anagnostou, M.D.a
a
Surgical Department of A. Fleming General Hospital, 14, 25th Martiou Str, 15127, Athens, Greece; b2nd Surgical Department
and Surgical Oncology Unit of KorgialenioBenakio, Red Cross Hospital, Athens, Greece; cDepartment of General and Abdominal
Surgery, Johannes Gutenberg University Hospital, Mainz, Germany; dOrgan Transplant Unit, Hippokration Hospital, Aristotle
University Medical School, Thessaloniki, Greece

KEYWORDS: Abstract
Meta-analysis; BACKGROUND: The objective of this study was to examine the outcomes of comparisons between
Evidence based; laparoscopic and open mesh repairs in the setting of recurrent inguinal hernia.
Publication bias; METHODS: The electronic databases MEDLINE, Embase, Pubmed, and the Cochrane Library were
Recurrent inguinal used to search for articles from 1990 to 2008. The present meta-analysis pooled the effects of outcomes
hernia; of a total of 1,542 patients enrolled into 5 randomized controlled trials and 7 comparative studies, using
Totally classic and modern meta-analytic methods.
extraperitoneal; RESULTS: Significantly fewer cases of hematoma/seroma formation were observed in the laparo-
Transabdominal scopic group in comparison with the Lichtenstein group (odds ratio, .38; .15.96; P .04). A matter
preperitoneal; of great importance is the higher relative risk of overall recurrence in the transabdominal preperitoneal
OPM; group compared with the totally extraperitoneal group (relative risk, 3.25; 1.327.9; P .01).
Open preperitoneal CONCLUSIONS: Laparoscopic versus open mesh repair for recurrent inguinal hernia was equivalent
mesh; in most of the analyzed outcomes.
Stoppa; 2010 Elsevier Inc. All rights reserved.
Giant prosthetic
reinforcement of the
visceral sac;
Lichtenstein procedure

The repair of recurrent inguinal hernia is a demanding inal wall has been established during the past decades.2 The
procedure accounting for 10% to 15% of inguinal hernia mesh can be placed either anteriorly under the external
repairs.1 The use of mesh for reinforcement of the abdom- aponeurosis (interparietal) or posteriorly (preperitoneal).3,4

Georgia Dedemadi designed the study, analyzed data, and was
responsible for acquisition of data; George Sgourakis designed the
study, analyzed data, and was responsible for acquisition of data;
Arnold Radtke analyzed data and was responsible for acquisition of
data; Alexandros Dounavis designed study and was responsible for
acquisition of data; Ines Gockel drafted the manuscript and was respon-
sible for substantial review; Ioannis Fouzas was responsible for critical
review and drafting the manuscript; Constantine Karaliotas designed
the study and was responsible for critical revision; and Evangelos
Anagnostou designed the study and was responsible for critical revision.
All authors have read and approved the final version of the manuscript
to be published. Georgia Dedemadi and George Sgourakis take respon-
sibility for the integrity of the data and the accuracy of the data analysis.
* Corresponding author. Tel.: 30-210-6033361; fax: 30-210-
7514179.
E-mail address: gdedemadi@yahoo.gr
Manuscript received June 9, 2009; revised manuscript December 1,
2009

0002-9610/$ - see front matter 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjsurg.2009.12.009
292
Figure 1 Progress through the stages of the meta-analysis.
Issues concerning the most appropriate treatment are still
under consideration.57 The Lichtenstein technique com-
monly is used for the repair of primary inguinal hernia and
may be suitable for repairing recurrent inguinal hernia if the
defect is larger than 4 cm.8 However, the technical ease and
low recurrence rate of this technique has lead to its wide-
spread use to repair recurrent inguinal hernias of any
size.9,10 Laparoscopic repairs are based on the same princi-
ples as preperitoneal mesh repair and therefore combine the
advantages of open preperitoneal repair with those of min-
imal access surgery.11,12 There is a major body of criticism
concerning laparoscopic hernia repair.1315 The main argu-
ments address the potential long-term recurrence rate, in-
traoperative complications, and the need for general anes-
thesia.1,16 Conversely, laparoscopic procedures may be
associated with less postoperative pain, simultaneous repair
of bilateral and other concurrent less-frequent types of in-
guinal hernia, short hospital stay, and early return to normal
activities.13,17,18 To date, the effectiveness of the treatment
modalities, concerning short- and long-term results for re-
current inguinal hernia, has not been compared by meta-
analysis of published reports.
Methods
Literature search
All randomized control trials and comparative studies
concerning recurrent hernia repair were identified.1,19 29
The electronic databases MEDLINE, Embase, Pubmed, and
the Cochrane Library were used to search for relevant arti-
cles published in English from 1990 to 2008 using the
following terms and/or combinations in their titles, abstracts, or
The American Journal of Surgery, Vol 200, No 2, August 2010
keyword lists: randomized controlled trials, double-blind, re-
current inguinal hernia, totally extraperitoneal, transabdominal
preperitoneal,OPM, open preperitoneal mesh, Stoppa, giant
prosthetic reinforcement of the visceral sac, and Lichtenstein
procedure. Where it was applicable the earlier-mentioned
terms were used in [MESH] (PubMed and the Cochrane
Library), otherwise the terms were combined with AND/
OR and asterisks. In addition, the abstracts from na-
tional and international conferences were searched using
online search engines corresponding to the particular
conference.
The scheme for this repetitive search is shown in Figure 1.
After the initial screen additional criteria were imposed as
follows: (1) adult population undergoing recurrent hernia
repair, (2) elective surgeries, (3) Nyhus class II to III her-
nias, and (4) analysis on an intention-to-treat basis.
Data extraction
Two authors (G.D., G.S.) independently selected stud-
ies for inclusion and exclusion, and reached consensus
when they did not agree on the initial assignment. The
following variables concerning studies addressing pa-
tients with recurrent hernia repair were recorded: authors,
journal and year of publication, country of origin, trial
duration, participant demographics, and data concerning
complications and follow-up evaluation. Where neces-
sary, the corresponding authors were contacted to obtain
supplementary information.
Interventions and outcome definition. Definitions were
as follows: early recurrence was defined as hernia recur-
rence during the first 2 months of follow-up evaluation;
totally extraperitoneal (TEP) was defined as laparoscopic
hernia repair with extraperitoneal mesh placement; transab-
G. Dedemadi et al. Laparoscopic versus open mesh repair
dominal preperitoneal (TAPP) was defined as laparoscopic
hernia repair through the peritoneal cavity with extraperi-
toneal mesh placement; Lichtenstein procedure was defined
as open tension-free mesh hernia repair; open preperitoneal
mesh (OPM) was defined as open preperitoneal mesh repair,
including the Stoppa technique and giant prosthetic rein-
forcement of the visceral sac.
Data analysis
A formal systematic review (according to the guidelines
of the QUOROM statement)30,31 was performed for all
randomized controlled trials (RCTs) and comparative stud-
ies concerning laparoscopic versus open recurrent hernia
repair. The primary outcomes used for this study were as
follows: (1) preoperative complications, (2) early recur-
rence, and (3) overall recurrence.
Studies with 3 treatment arms27 were treated as being 2
separate studies for outcome measures.
Statistical analysis
The Maxwell test statistic and the generalized McNemar
statistic were calculated to quantify the level of agreement
between reviewers. Pooled estimates of outcomes were cal-
culated using a fixed-effects model but a randomized-effects
model was used according to heterogeneity. For dichoto-
mous data, results for each trial were expressed as an odds
ratio (OR), relative risk (RR), with 95% confidence inter-
vals (CIs). Tests for heterogeneity and overall effect were
provided for each total or subtotal.
The MosesShapiroLittenberg method was applied to
explore sources of heterogeneity by adding covariates to the
model. Where there was a threshold effect (suspicion for
heterogeneity), the summary of study results was performed
by a receiver operating characteristics curve. Bootstrap CIs
were used to estimate the range of uncertainty for a given
test statistic and to generate the lower and upper 95%
confidence limits.
Bias was studied using sensitivity analysis by removing
individual studies from the data set and analyzing the over-
all effect size and the weighted regression tests described by
Egger et al.32 Results were significant if the P value was less
than .5. The RevMan Version 4.2,33,34 the Statsdirect ver-
sion 2.6.5, the Meta Disk version 1.4,35 and Meta-Win
version 2.136 were used for the data analysis.
Study quality assessment. According to the Cochrane
Handbook for Systematic Reviews of Interventions 4.2.6,34
high-quality RCTs were designated as A, low-quality but
randomized controlled studies were designated as B, and
comparative studies were designated as C, taking into
account the way these studies deal with allocation conceal-
ment information.
293
Results
The Maxwell test statistic was not significant (P .852),
indicating that the raters (assigned with study selection) did not
disagree significantly. The generalized McNemar statistic (P
.56) indicated that the agreement was spread evenly.
Seven comparative and 5 RCTs were ascertained fina-
lly.1,19 29 The literature search and study selection process
therefore yielded 12 studies with 1,542 enrolled patients.
All studies provided a minimum follow-up period of at least
18 months. Four studies provided a mean/median follow-up
period of more than 5 years 1 longer than 4 years 3 longer
than 3 years, and 2 longer than 2 years.
The foremost randomized controlled study restrictions
were relevant as follows: (1) validation of sample size; in
only 2 of the studies1,27 was a power analysis performed
(although 1 study25 would have validated its results because
of the large number of patients per treatment arm), and (2)
disclosure of the number of drop-outs: 2 of 5 studies pro-
vided complete data.23,27 Among the comparative studies 5
were retrospective and only 2 were prospective.
Missing data limited the analysis only to these outcomes
about which trials granted comprehensive information. Base-
line characteristics of patients and results for various outcome
measures in the included trials are summarized in Table 1.
Outcomes. By enclosing 5 RCTs and 7 comparative studies
for recurrent inguinal hernia repair 14 comparisons were
conducted with regard to the following: (1) TEP versus
Lichtenstein (Table 2); (2) laparoscopic versus Lichtenstein
(Table 3); (3) laparoscopic versus OPM (Table 4); (4) lapa-
roscopic versus open (Lichtenstein/OPM) (Table 4); and (5)
TAPP versus TEP (Table 4). Comparisons between TAPP
versus Lichtenstein, TEP versus OPM, and TAPP versus
OPM were not conducted because of insufficient data pro-
vided by the investigators.
The TEP group was equivalent to the Lichtenstein group
in comparisons related to the following outcomes: wound
infection, hematoma/seroma, urinary retention, and the RR
of overall recurrence (Fig. 2 and Table 2).
Heterogeneity among studies in terms of hematoma/se-
roma formation (P .007/I2 75.1%) was observed and
this finding was analyzed further by the MosesShapiro
Littenberg method (see paragraph analyzing heterogeneity).
Funnel plots did not detect any obvious publication bias
concerning all outcomes.
The laparoscopic groups were comparable with the Lich-
tenstein group in relation to the following outcomes: wound
infection, urinary retention, testicular pain/discomfort, pain/
neuralgia, and the RR of early and overall recurrence (Fig.
3 and Table 3). Significantly fewer cases with hematoma/
seroma formation were observed in the laparoscopic group.
This outcome was evaluated by the random-effects model
owing to significant heterogeneity, which also was analyzed
further (see paragraph analyzing heterogeneity). Funnel
plots and normal quantile plots did not detect any obvious
publication bias concerning all outcomes.
294 The American Journal of Surgery, Vol 200, No 2, August 2010

Table 1 Baseline characteristics of patients included in the meta-analysis

Bilateral Wound Hematoma/ Testicular pain/ Pain/ Urinary Urinary Early Overall
Study Group N Age, y M/F recurrence infection seroma discomfort neuralgia retention infection recurrence recurrence

Feliu et al25 TEP 86 57.2 10.8 77/1 8 2 13 2 5 5 1 NR 1

OPM 121 57.8 13.2 105/5 11 0 10 1 1 0 0 NR 2
Alani et al26 TEP 45 60 42/3 7 0 5 NR NR NR NR NR 0
OPM 54 64 51/3 9 0 12 NR NR NR NR NR 0
Dedemadi TEP 26 NR 26/0 1 0 3 3 1 1 NR 0 2
et al27* Lichtenstein 32 NR 32/0 2 1 12 1 2 1 NR 2 5
Dedemadi TAPP 24 NR 24/0/ 1 0 4 0 1 1 NR 1 2
et al27* Lichtenstein 32 NR 32/0 2 1 12 1 2 1 NR 2 5
Eklund et al28 TAPP 73 52 10.4 73/0 NR 1 5 0 NR 7 1 NR 12
Lichtenstein 74 55 11.3 74//0 NR 2 17 3 NR 10 1 NR 12
Kouhia et al29 Lichtenstein 47 55.8 46/1 0 4 6 NR NR NR NR NR 3
TEP 49 57.8 47/2 2 1 13 NR NR NR NR NR 0

NR not reported.
*RCTs with 3 treatment arms and subsequently 2 comparisons.

When comparing the laparoscopic and the OPM groups,
results in terms of hematoma/seroma formation were similar
(Table 4). Sensitivity analysis disclosed that the study by
Alani et al26 contributed to the noncombinability of the
studies (P .009, I2 74%). By omitting this study the
OPM treatment arm presented fewer cases of hematoma/
seroma formation (pooled OR, 2.6; 1.49 4.51; P .001).
Funnel plots did not detect any obvious publication bias.
There was no significant difference in terms of urinary
infection between the laparoscopic and open treatment arms
(Table 4). Neither heterogeneity nor publication bias was
present.
The RR of overall recurrence was less in the TEP treat-
ment arm when compared with that of TAPP. Heterogeneity
and publication bias were absent.
Analyzing heterogeneity.
threshold effect was examined by calculating the Spearman
correlation coefficient. No inverse correlation was observed
(.000; P 1.000), putting the presence of heterogeneity in
question. The study by Richards et al24 showed a marginal
sensitivity by using resampling tests and bootstrapping to
generate CIs around the overall cumulative mean effect size
(OR, .4507; bootstrap CI, 1.2274 to .4889).
Laparoscopic versus Lichtenstein (hematoma/seroma). Het-
erogeneity among studies was observed (P .001/I2
64.4%), and the threshold effect was examined by calculat-
ing the Spearman correlation coefficient. No inverse corre-
lation was observed (.754; P .084), putting the presence
of heterogeneity in question. Resampling tests derived from
999 iterations and bootstrapping were used to generate CIs
around the overall cumulative mean effect size (OR, .6546;
bootstrap CI, 1.1005 to .0266).

TEP versus Lichtenstein (hematoma/seroma). Heterogeneity Laparoscopic versus OPM (hematoma/seroma). Significant
among studies was observed (P .007/I2 75.1%) and the heterogeneity among studies was observed (P .009/I2

Table 2 Comparisons of outcomes between TEP and Lichtenstein

Test for
overall Publication bias Study Favors
Outcome/subgroup Studies N Effect estimate (95% CI) Heterogeneity effect (indicator/P value) quality group
Wound infection 3 294 FE, OR .49 (.122.0) P .379 2 .48 Harbold-Egger: 2A None
bias .84
I2 0% P .489 P .786 1C
Hematoma/seroma 4 344 RE, OR .43 (.091.91) P .007 2 1.24 Egger: 2A None
bias 3.3
I2 75.1% P .264 P .427 2 C
Urinary retention 3 248 FE, OR .47 (.092.42) P .733 2 .3 Egger: bias .3 1 A None
I2 0% P .580 P .923 2 C
Overall recurrence 4 344 FE, RR .48 (.181.33) P .531 2 1.97 Egger: bias .16 2 A None
I2 0% P .160 P .943 2 C
The P values of comparisons are given in the column labeled Test for overall effect and the group favored is depicted in the column labeled Favors
group.
FE fixed-effects model; RE random-effects model; A high-quality studies; C comparative studies.
G. Dedemadi et al. Laparoscopic versus open mesh repair 295

Table 3 Comparisons of outcomes between laparoscopic procedures and Lichtenstein

Test for
overall Publication Bias Study
Outcome/subgroup Studies N Effect estimate (95% CI) Heterogeneity effect (indicator/P value) quality Favors group
Wound infection 5 497 FE, OR .48 (.151.43) P .74 2 1.03 Egger: bias 1.34 4 A None
I2 0% P .30 P .594 1C
Testicular pain/ 3 261 FE, OR .94 (.253.53) P .25 2 .06 Harbold-Egger: 3 A None
discomfort bias 3.75
I2 27% P .80 P .4013
Urinary retention 5 451 FE, OR .65 (.281.47) P .91 2 .69 Egger: bias .09 3 A None
I2 0% P .4 P .87 2 C
Hematoma/seroma 6 547 RE, OR .38 (.15.96) P .01 2 4.16 Egger: bias 2.28 4 A Laparoscopic
I2 64% P .04 P .451 2 C
Pain/neuralgia 3 261 FE, OR .63 (.271.47) P .99 2 .74 Harbold-Egger: 3 A None
bias 4.49E03
I2 0% P .39 P .9553
Early recurrence 4 304 FE, RR .73 (.212.51) P .77 2 .23 Egger: bias .54 2 A None
I2 0% P .62 P .874 2 C
Overall recurrence 7 706 FE, RR .72 (.451.15) P .75 2 1.97 Egger: bias .64 5 A None
I2 0% P .16 P .331 2 C
The P values of comparisons are given in the column labeled Test for overall effect and the group favored is depicted in the column labeled Favors
group.
FE fixed-effects model; RE random-effects model; A high-quality studies; C comparative studies.

74%), and this threshold effect was evaluated by calculating
the Spearman correlation coefficient. An inverse correlation
was observed (.400; P .6) confirming the stated heter-
ogeneity. A sensitivity analysis disclosed that the study by
Alani et al26 contributed to the noncombinability of the
studies. By omitting this study, the OPM treatment arm
presented fewer cases of hematoma/seroma formation.
Validating our model of comparisons. Randomization
tests were used to test the significance of our model struct-
ure (pooling the effects of different procedures under the
term laparoscopic (TAPP/TEP) or open repair (Lich-
tenstein/OPM)). Probability between studies was .287 and
within studies was .076. The total model probability was
.088, justifying our comparisons.
Comments
A meta-analysis of RCTs and comparative studies con-
cerning patients treated for recurrent inguinal hernia with

Table 4 Comparisons of outcomes between laparoscopic and open procedures and TAPP versus TEP

Test for overall Publication bias Study Favors

Outcome/subgroup Studies N Effect estimate (95% CI) Heterogeneity effect (indicator/P value) quality group
Laparoscopic versus
OPM
Hematoma/seroma 4 530 RE, OR 1.86 (.685.07) P .009 2 1,46 Egger: bias 1.44 1 A None
I2 74% P .226 P .8833 3C
Laparoscopic versus
open
Urinary infection 3 433 FE, OR 2.67 (.5113.9) P .716 2 .68 Harbold-Egger: 2 A None
bias 2.48
I2 0% P .40 P .608 1C
TAPP versus TEP
Overall recurrence 3 407 FE, RR 3.25 (1.327.9) P .186 2 6.58 Harbold-Egger: 1 A TEP
bias .96
I2 40.4% P .01 P .6327 2C
The P values of comparisons are given in the column labeled Test for overall effect and the group favored is depicted in the column labeled Favors
group.
FE fixed-effects model; RE random-effects model; A high-quality studies; C comparative studies.
296
Figure 2 Meta-analysis of the RR of overall recurrence in stud-
ies addressing TEP versus Lichtenstein.
laparoscopic procedures were comparable, in most of the
analyzed outcomes, with patients treated with open mesh
repair. Significantly fewer cases of hematoma/seroma for-
mation were observed in the laparoscopic group in compar-
ison with the Lichtenstein group. A matter of importance is
the higher RR of overall recurrence in the TAPP group
compared with the TEP group, whereas the RR of overall
recurrence of the laparoscopic versus the Lichtenstein group
was comparable.
Potential heterogeneity of evaluating studies collectively
either in the open or the laparoscopic group was assessed.
Noncompatibility of the studies was present only in 3 cases,
all concerning the outcome of hematoma/seroma formation.
Sensitivity analysis disclosed the 2 studies contributing to
heterogeneity. In the third case the presence of heterogene-
ity was rejected by the MosesShapiroLittenberg method.
Randomization tests were used to test the significance of
this model structure (pooling the effects of different tech-
niques), showing no significant probability, which justified
our comparisons.
When considering the completeness and applicability of
evidence, the included studies provided satisfactory data to
address the issues properly including complications and
recurrence. In addition, their inclusion criteria (adult popu-
lation undergoing recurrent hernia repair, elective surgeries,
Nyhus class IIIII hernias) appeared to be almost identical,
thus ensuring comparable types of analyzed interventions
and outcomes.
Making allowances for the quality of the evidence, the 12
included studies had a total of 1,542 participants, with good
methodologic quality (5 RCT and 2 prospective studies).
Only 3 of 12 studies that were analyzed were sufficiently
powered to document their findings.1,27,28 Significant but
explicable heterogeneity was revealed in only 3 of 14 com-
parisons without the presence of publication bias. Modern
meta-analytic techniques declined heterogeneity of studies
in 2 of the 3 cases.
The American Journal of Surgery, Vol 200, No 2, August 2010
If we confine the search to the English language litera-
ture this could represent a potential publication bias during
the review process; however, when going through the ab-
stracts, no suitable studies were found in the non-English
language literature. The HarboldEgger test was used to
maintain the power of the Egger test in reducing the false-
positive rate, which was a problem in cases of large treat-
ment effects and few events per trial.
In our effort to make an overall judgment of the external
validity of this meta-analysis the outcomes were balanced
with those of a Cochrane database systematic review (41
eligible trials involving 7,161 participants) concerning the
application of laparoscopic techniques versus open tech-
niques for primary inguinal hernia repair.37 There was no
apparent difference in recurrence between laparoscopic and
open mesh methods of hernia repair in this study either.
Moreover, the prominent advantage of the present meta-
analysis is the analysis of heterogeneity and the validation
of the results by randomization tests and bootstrapping.
There also were some limitations in this meta-analysis.
Important clinical issues such as quality of life and return to
normal physical activity were not evaluated because of the
paucity of comprehensive data among studies. It would be
more apposite should this analysis contain comparisons
between each laparoscopic group separately with the open
approaches, nevertheless only 2 studies in each outcome
provided satisfactory data, which could not guarantee the
appropriateness of the outcome in terms of heterogeneity
and publication bias.
The major reasons for considering collectively the pa-
tients of the laparoscopic approaches as a single group are
that the surgical field and handling of the tissues (dissection
of the cord/round ligament and clip placement) are similar
in both procedures and the anticipated postoperative com-
plications (testicular pain/discomfort, pain/neuralgia, hema-
toma/seroma, and urinary retention) are potentially analo-
gous. In support of this is the current evidence provided by
a Cochrane database systematic review38 in which no ob-
vious differences were observed between TAPP and TEP
concerning hematoma, vascular injuries, deep mesh infec-
Figure 3 Meta-analysis of the RR of overall recurrence in stud-
ies addressing laparoscopic repairs versus Lichtenstein.
G. Dedemadi et al. Laparoscopic versus open mesh repair
tions, and recurrence, although this review was based
mainly on nonrandomized studies of primary inguinal her-
nia repair.
In summary, laparoscopic versus open mesh repairs for
recurrent inguinal hernia were equivalent in most of the
analyzed outcomes and particularly for overall recurrence.
A matter of major importance is the higher RR of overall
recurrence in the TAPP compared with the TEP group.
Reports of the long-term follow-up period (up to 10 years)
of the existing studies or new sufficiently powered random-
ized controlled trials are most welcome.
References
1. Neumayer I, Giobbie-Hurder A, Jonasson O, et al. Open mesh versus
laparoscopic mesh repair of inguinal hernia. N Engl J Med 2004;350:
1819 27.
2. Collaboration EH. Mesh compared with non-mesh methods of open
groin hernia repair: systemic review of randomized controlled trials.
Br J Surg 2000;87:854 9.
3. Lichenstein IL, Shulman AG, Amid PK, et al. The tension-free hemio-
plasty. Am J Surg 1989;157:188 93.
4. Stoppa RE. The treatment of complicated groin and incisional hernias.
World J Surg 1989;13:54554.
5. Jonasson O. Introduction: inguinal hernia management-testing man-
agement strategies in two clinical trials. J Am Coll Surg 2003;196:
735 6.
6. Macintyre IM. Best practice in groin hernia repair. Br J Surg 2003;
90:1312.
7. Gianetta E, Cuneo S, Vitale B, et al. Anterior tension-free repair of
recurrent hernia under local anesthesia: a 7-year experience in a teach-
ing hospital. Ann Surg 2000;231:132 6.
8. Lichtenstein IL, Shulman AG, Amid PK. The cause, prevention, and
treatment of recurrent groin hernia. Surg Clin North Am 1993;73:529 44.
9. Wilson MS, Deans GT, Brough WA. Prospective trial comparing
Lichtenstein with laparoscopic tension-free mesh repair of inguinal
hernia. Br J Surg 1995;82:274 7.
10. Kark AE, Kurzer M, Waters KJ. Tension-free mesh hernia repair:
review of 1098 cases using local anaesthesia in a day unit. Ann R Coll
Surg Engl 1995;77:299 304.
11. McKernan JB, Laws HL. Laparoscopic repair of inguinal hernias using
a totally extraperitoneal prosthetic approach. Surg Endosc 1993;7:
26 8.
12. Liem MS, van der Graaf Y, van Steensel CJ, et al. Comparison of
conventional anterior surgery and laparoscopic surgery for inguinal-
hernia repair. N Engl J Med 1997;336:15417.
13. Johansson B, Hallerbck B, Glise H, et al. Laparoscopic mesh versus
open preperitoneal mesh versus conventional technique for inguinal
hernia repair: a randomized multicenter trial (SCUR Hernia Repair
Study). Ann Surg 1999;230:22531.
14. Stoker DL, Spiegelhalter DJ, Singh R, et al. Laparoscopic versus open
inguinal hernia repair: randomized prospective trial. Lancet 1994;21:
12435.
15. Heikkinen TJ, Haukipuro K, Hulkko A. A cost and outcome compar-
ison between laparoscopic and Lichtenstein hernia operations in a
day-case unit: a randomized prospective study. Surg Endosc 1998;12:
1199 203.
16. Memon MA, Cooper NJ, Memon B, et al. Meta-analysis of random-
ized clinical trials comparing open and laparoscopic inguinal hernia
repair. Br J Surg 2003;90:1479 92.
297
17. Sayad P, Abdo Z, Cacchione R, et al. Incidence of incipient contralat-
eral hernia during laparoscopic hernia repair. Surg Endosc 2000;14:
5435.
18. Geis WP, Crafton WB, Novak MJ, et al. Laparoscopic herniorrhaphy:
results and technical aspects in 450 consecutive procedures. Surgery
1993;114:76572.
19. Mok KT, Wang BW, Chang HT, et al. Laparoscopic versus open
preperitoneal prosthetic herniorrhaphy for recurrent inguinal hernia.
Int Surg 1998;83:174 6.
20. Memon MA, Feliu X, Sallent EF, et al. Laparoscopic repair of recur-
rent hernias. Surg Endosc 1999;13:80710.
21. Jarhult J, Hakanson C, Akerud L. Laparoscopic treatment of recurrent
inguinal hernias: experience from 281 operations. Surg Laparosc En-
dosc Percutan Tech 1999;9:115 8.
22. Kumar S, Nixon SJ, MacIntyre IM. Laparoscopic or Lichtenstein
repair for recurrent inguinal hernia: one units experience. J R Coll
Surg Edinb 1999;44:3012.
23. Beets GL, Dirksen CD, Go PM, et al. Open or laparoscopic preperi-
toneal mesh repair for recurrent inguinal hernia? A randomized con-
trolled trial. Surg Endosc 1999;13:3237.
24. Richards SK, Vipond MN, Earnshaw JJ. Review of the management of
recurrent inguinal hernia. Hernia 2004;8:144 8.
25. Feliu X, Torres G, Vinas X, et al. Preperitoneal repair for recurrent
inguinal hernia: laparoscopic and open approach. Hernia 2004;
8:113 6.
26. Alani A, Duffy F, ODwyer PJ. Laparoscopic or open preperitoneal
repair in the management of recurrent groin hernias. Hernia 2006;10:
156 8.
27. Dedemadi G, Sgourakis G, Karaliotas C, et al. Comparison of laparo-
scopic and open tension-free repair of recurrent inguinal hernias: a
prospective randomized study. Surg Endosc 2006;20:1099 104.
28. Eklund A, Rudberg C, Leijonmarck CE, et al. Recurrent inguinal
hernia: randomized multicenter trial comparing laparoscopic and Lich-
tenstein repair. Surg Endosc 2007;21:634 40.
29. Kouhia ST, Huttunen R, Silvasti SO, et al. Lichtenstein hernioplasty
versus totally extraperitoneal laparoscopic hernioplasty in treatment of
recurrent inguinal herniaa prospective randomized trial. Ann Surg
2009;249:384 7.
30. Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports
of meta-analyses of randomized controlled trials: the QUOROM state-
ment. Quality of reporting of meta-analyses. Lancet 1999;354:1896
900.
31. Mahid SS, Hornung CA, Minor KS, et al. Systematic reviews and
meta-analysis for the surgeon scientist. Br J Surg 2006;93:131524.
32. Egger M, Smith GD, Altman DG. Systematic Reviews in Health Care:
Meta-analysis in Context. 2nd ed. London: BMJ Books; 2001.
33. Review Manager (RevMan) [COMPUTER PROGRAM]. Version 4.2
for Windows. Copenhagen: The Nordic Cochrane Centre, The Co-
chrane Collaboration; 2003.
34. Higgins JPT, Green S, eds. Cochrane Handbook For Systematic Re-
views of Interventions 4.2.6 (Updated September 2006). The Cochrane
Library. Chichester, UK: John Wiley & Sons, Ltd.; 2006.
35. Zamora J, Abraira V, Muriel A, et al. Meta-DiSc: a software for
meta-analysis of test accuracy data. BMC Med Res Methodol 2006;
6:31.
36. Rosenberg MS, Adams DC, Gurevitch J. MetaWin. Statistical Soft-
ware for Meta-Analysis, Version 2. Sunderland, MA: Sinauer Asso-
ciates; 2000.
37. McCormack K, Scott NW, Go PM, et al. Hernia Trialists Collabora-
tion. Laparoscopic techniques versus open techniques for inguinal
hernia repair. Cochrane Database Syst Rev 2003;1:CD001785.
38. Wake BL, McCormack K, Fraser C, et al. Transabdominal pre-peri-
toneal (TAPP) vs totally extraperitoneal (TEP) laparoscopic tech-
niques for inguinal hernia repair. Cochrane Database Syst Rev 2005;
1:CD004703.