You are on page 1of 12

Hernia (2019) 23:461–472

https://doi.org/10.1007/s10029-019-01989-7

REVIEW

Open versus laparoscopic mesh repair of primary unilateral


uncomplicated inguinal hernia: a systematic review
with meta‑analysis and trial sequential analysis
N. L. Bullen1   · L. H. Massey1 · S. A. Antoniou2,3 · N. J. Smart1 · R. H. Fortelny4,5

Received: 17 May 2019 / Accepted: 20 May 2019 / Published online: 3 June 2019
© Springer-Verlag France SAS, part of Springer Nature 2019

Abstract
Background  One standard repair technique for groin hernias does not exist. The objective of this study is to perform an update
meta-analysis and trial sequential analysis to investigate if there is a difference in terms of recurrence between laparoscopic
and open primary unilateral uncomplicated inguinal hernia repair.
Methods  The reporting methodology conforms to PRISMA (Preferred Reporting Items for Systematic reviews and Meta-
Analyses) guidelines. Randomised controlled trials only were included. The intervention was laparoscopic mesh repair
(transabdominal preperitoneal (TAPP) and totally extraperitoneal (TEP)). The control group was Lichtenstein repair. The
primary outcome was recurrence rate and secondary outcomes were acute and chronic post-operative pain, morbidity and
quality of life.
Results  This study included 12 randomised controlled trials with 3966 patients randomised to Lichtenstein repair (n = 1926)
or laparoscopic repair (n = 2040). There were no significant differences in recurrence rates between the laparoscopic and
open groups (odds ratio (OR) 1.14, 95% CI 0.51–2.55, p = 0.76). Laparoscopic repair was associated with reduced rate of
acute pain compared to open repair (mean difference 1.19, CI − 1.86, − 0.51, p ≤ 0.0006) and reduced odds of chronic pain
compared to open (OR 0.41, CI 0.30–0.56, p ≤ 0.00001). The included trials were, however, of variable methodological
quality. Trial sequential analysis reported that further studies are unlikely to demonstrate a statistically significant difference
between the two techniques.
Conclusion  This meta-analysis and trial sequential analysis report no difference in recurrence rates between laparoscopic
and open primary unilateral inguinal hernia repairs. Rates of acute and chronic pain are significantly less in the laparoscopic
group.

Keywords  Inguinal hernia · Lichtenstein · TAPP · TEP

Introduction

The article is part of the Topical Collection “Forum on primary


Lifetime occurrence of groin hernia is 27–43% in men and
monolateral uncomplicated inguinal hernia”. 3–6% in women [1]. One standard repair technique for all
groin hernias does not exist. Mesh repair is recommended
Electronic supplementary material  The online version of this as first choice either by an open procedure or laparoscopic
article (https​://doi.org/10.1007/s1002​9-019-01989​-7) contains
supplementary material, which is available to authorized users.

3
* N. L. Bullen European University Cyprus, Nicosia, Cyprus
n.bullen@nhs.net 4
Department of General, Visceral and Oncological Surgery,
1 Wilhelminenspital, 1160 Vienna, Austria
Department of Colorectal Surgery, Royal Devon and Exeter
5
NHS Foundation Trust, Royal Devon and Exeter Hospital, Medical Faculty, Sigmund Freud University, Freudplatz 3,
Barrack Road, Exeter EX2 5DW, UK 1020 Vienna, Austria
2
Surgical Department, St Loukas Hospital, Thessaloniki,
Greece

13
Vol.:(0123456789)

462 Hernia (2019) 23:461–472

technique providing the surgeon is sufficiently experienced follow-up duration was less than 12 months and if hernias
in the specific procedure [2]. The Lichtenstein technique is were recurrent or bilateral. As some studies included small
considered the reference standard for open inguinal hernia amounts of data on these patients and might introduce heter-
repair [3]. Recurrence rates are reported to be less than 1% ogeneity into analyses, it was decided to exclude those where
[3, 4]. Morbidity is mainly related to chronic pain which the proportion of recurrent or bilateral hernias was more
can occur in up to 63% of patients and affects quality of life than 10% of the total. The intervention was any technique
in 5–10% [5]. The increase in laparoscopic inguinal hernia of laparoscopic mesh repair, including both transabdominal
repairs has largely aimed at reducing rates of chronic pain preperitoneal (TAPP) and totally extraperitoneal (TEP). The
without compromising recurrence rates. control group was open mesh repair with the Lichtenstein
Evidence from previous meta-analyses is contradictory. technique.
Laparoscopic repair with both transabdominal preperitoneal The electronic databases of Embase, Medline through
(TAPP) and totally extraperitoneal (TEP) has significantly PubMed, the Cochrane Central Register of Controlled Tri-
reduced rates of chronic pain [6–10] and has improved als (CENTRAL, provider Wiley Online Library) and Open-
patient satisfaction [9]. However, TEP has been shown to Grey were searched from their inception until April 2019.
be associated with an increased risk of recurrence com- A combination of the following MeSH terms (Medical Sub-
pared to open surgery [6], specifically when follow-up data ject Headings) were used: “Inguinal”, “Groin”, “Hernia”,
exceeded 3 years [7]. TAPP is linked to an increased risk “Herniorrhaphy”, “Mesh”, “Prosthetic material”, “Lichten-
of perioperative complications compared to open surgery stein”, “Laparoscopic”, “Endoscopic”, “Transabdominal
[6]. More recent studies have shown equivalent recurrence Preperitoneal” (TAPP), and “Totally Extraperitoneal” (TEP)
rates [8, 10]. (Online Appendix 1). Randomised trials were searched for
Some of the previous analyses had methodological short- using the Cochrane highly sensitive strategies for identify-
comings such as small numbers of randomized controlled ing randomised trials [13]. No language restrictions were
trials [8] and inclusion of trials of short follow-up duration applied. Eligibility assessment was performed independently
[9] and since the 2014 meta-analysis, three further rand- in an unblinded standardised manner by two reviewers. Disa-
omized controlled trials have become available. greements were resolved by consensus.
The objective of this study is to perform an update meta-
analysis to investigate if there is a difference in terms of Data collection
recurrence and post-operative pain between laparoscopic and
open primary inguinal hernia repair, furthermore, to per- Two independent reviewers were involved in the study selec-
form a trial sequential analysis and investigate whether the tion (NB, LM). Reviewers were blinded to studies selected
required information size has been reached and evidence is for inclusion by the other reviewer. Discrepancies were
conclusive, or alternatively compute the number of patients resolved by a third party (NJS). Bibliographic references
required to be enrolled in further trials. of published studies and reviews were also interrogated.
Data were extracted using a standardised data collection
form. One reviewer extracted the data (NB) and the second
Methods reviewer checked the extracted data (LM).
The primary outcome measure was hernia recurrence.
The protocol for this systematic review was established Data related to the primary outcome measure were collected
prior to initiation of the study and was registered under the from the abstract, main text, tables or graphs. The method
number CRD42019125335 in the International Prospective for diagnosis recurrence was documented (clinical or radio-
Register of Systematic Reviews (PROSPERO) database logical) and the duration of follow-up.
[11]. The reporting methodology conforms to the PRISMA The secondary outcome measures were short- and long-
(Preferred Reporting Items for Systematic reviews and Meta- term post-operative pain, re-operation rate, surgical site
Analyses) guidelines (Online Appendix 4) [12]. infection, surgical site occurrence (haematoma/seroma) and
quality of life. The time intervals for short-term assessment
Eligibility criteria, study selection and search of post-operative pain were defined as up to 30 days. Longer-
strategy term follow-up for chronic pain was greater than 6 months.
The ICD-11 (International Classification of Diseases) defi-
Randomised controlled trials only were included compar- nition of chronic pain is pain lasting longer than 3 months
ing laparoscopic versus open mesh repair for primary ingui- [14]. However, as the usage of synthetic materials for hernia
nal hernia. No other study design was considered. Studies repair may lengthen the inflammatory response, chronic pain
enrolling patients aged over 16 years with primary unilateral is defined as pain lasting more than 6 months [15]. Most
inguinal hernias were included. Studies were excluded if studies were expected to assess post-operative pain using the

13
Hernia (2019) 23:461–472 463

Fig. 1  PRISMA flow chart of


search history and study selec-

Idenficaon
tion Records identified through Additional records identified
database searching through other sources
(n = 146) (n = 3)

Records after duplicates removed


(n = 83)

Screening
Records screened Records excluded
(n = 83) (n = 45)

Full-text articles assessed Full-text articles excluded,


for eligibility with reasons
Eligibility

(n = 38) (n = 18)
14 = included too many
recurrent/bilateral hernia
3 = F/u too short
Studies included in 1 = not relevant
qualitative synthesis
(n = 20)
Included

Unique studies included in


quantitative synthesis
(meta-analysis)
(n = 12)

mean visual analogue scale (VAS) from 0 to 10. VAS scores Pooled odds ratios (ORs) with 95% CIs were calculated to
ranging from 0 to 100 were converted to an 11-scale climax. measure the effect of each type of procedure on dichoto-
Descriptive data on soft tissue collection/seroma/haematoma mous variables. Publication bias was assessed visually
(such as method of diagnosis) were documented. evaluating the symmetry of funnel plots if at least ten tri-
als were included in the meta-analysis. Statistical analysis
was performed using RevMan 5.3 (Review Manger 5.3,
Synthesis of results The Nordic Cochrane Centre, Copenhagen, Denmark).
Sensitivity analyses calculating the risk difference were
The presence of heterogeneity was evaluated by assess- performed if one or more studies reported zero events in
ing the consistency of study population, intervention, both arms. Subgroup analyses of studies comparing open
perioperative care characteristics and method of outcome mesh repair versus TAPP or TEP were performed.
assessment, by inspecting the forest plots, and by com- Trial sequential analysis was performed to assess the
puting the I2 values. A random effects model according possibility of type I error and to compute the information
to DerSimonian and Laird was applied to synthesise data size. The Lan and DeMets method was used to construct
irrespective of statistical heterogeneity as we considered monitoring boundaries and set adjusted thresholds for sta-
pooling preperitoneal with transperitoneal techniques tistical significance [17]. The information size was calcu-
and various duration of follow-up to introduce concep- lated at α = 0.05, β = 0.2, relative risk reduction of 50%
tual heterogeneity. Difference in means (MD) with 95% and a cumulative incidence of chronic post-operative pain
confidence intervals (CIs) was calculated to assess the of 5.2% based on trials at low risk of bias, defined as out-
size of the effect of the use of laparoscopic or open mesh come assessment by a blinded investigator. The Z-value
repair on continuous variables. The standard error and curve was constructed based on consecutive Z-values
standard deviation were obtained from CIs by using the calculated upon two-sided significant testing. Trial
formula suggested by the Cochrane Collaboration [16]. sequential monitoring boundaries were constructed using

13

464 Hernia (2019) 23:461–472

Fig. 2  a Risk of bias graph. b


Risk of bias summary

conventional testing and applying the O’Brien-Fleming were selected for full-text review. There was satisfactory
α-spending function [18]. Futility testing was performed agreement between the reviewers (NB & LM) with eight
and the respective futility boundaries were constructed to articles causing conflict, which were resolved by a third
assess whether the two interventions differ more than the reviewer (NJS) and one of these was subsequently included.
anticipated intervention effect. Trial sequential analysis A total of 20 articles reporting on randomised controlled tri-
was performed using The Trial Sequential Analysis soft- als (RCTs) fulfilled the eligibility criteria and were included
ware 0.9.5.10 Beta (Copenhagen Trial Unit, Copenhagen, in the qualitative analysis. Five trials reported short- and
Denmark). long-term follow-up across several papers (between two
and four separate papers); overall, 12 unique studies were
Methodological assessment included in the quantitative analysis. The literature search
processes are summarised in Fig. 1. Search strategies can
Risk of bias of the included studies was assessed using be found in Online Appendix 1. The risk of bias assess-
Cochrane Collaboration’s tool [19]. This tool considers ran- ment of articles that were judged to be of acceptable or high
dom sequence generation, allocation concealment, blinding quality is illustrated in Fig. 2a, b. None of the trials were
of participants, personnel and outcome assessors; incom- patient-blinded.
plete outcome data, selective outcome reporting and other
potential threats to validity.
Study characteristics

The twelve included RCTs were published between 1997 and


Results 2019 and all were written in the English language. A total
of 3966 patients were included, randomised to Lichtenstein
Study selection and quality assessment repair (n = 1926) or laparoscopic repair (n = 2040). (Lapa-
roscopic repair included TAPP n = 387 and TEP n = 1653.)
The searches of the electronic databases retrieved 146 Study characteristics are summarised in Table 1. Half of the
records. After the exclusion of duplicates, 83 titles and trials included only male patients and of the others report-
abstracts remained for screening. From these, 38 articles ing sex of patients: 93% were male. Duration of follow-up

13
Table 1  Characteristics of included studies
Authors Year Country Num- Surgeon Male/ Proportion Number Number Type of Follow-up Patients Primary Definition of
(refer- ber of experience female recurrent/ of patients of patients mesh (months) lost to Outcome primary out-
ences) centres ratio bilateral Lichten- laparo- follow-up come
hernia stein scopic (%)
technique
Hernia (2019) 23:461–472

Sevi̇nç 2019 Turkey 1 > 100 TEP 273:29 0 160 160 TEP Polypro- Mean 5.6 Chronic Pain
[20] pylene 40.95 ± 17.9 groin for > 3 months
pain
Koju [21] 2017 Nepal 1 NR NR 7.8% 51 51 TAPP NR 12 NR Post- Pain at 24 h
operative
pain
Westin/ 2016/2013 Sweden 2 NR All male 0 191 193 TEP Polypro- 12 2.3 Chronic Any pain at
Dahl- pylene groin 1 year
strand heavy- pain
[22, 23] weight
Wang [24] 2013 China 1 > 20 TEP/ 210:42 NR 84 TAPP 84 Polypro- Mean 16 3.6 Primary NR
TAP TEP 84 pylene (3–32) outcome
light- not
weight stated
Eker/Lan- 2012/2010 Nether- 6 > 30 TEP 649:11 8.1% 317 323 TEP Polypro- Median 60 24.7 Post- Pain up to
geveld lands pylene (27–69) operative 2 weeks post
[25, 26] pain surgery
Eklund 2010/2009/2006 Sweden 11 > 25 TEP All male 0 705 665 TEP Polypro- Median 61 7 Recur- Expansile cough
[27–29] pylene (53–109) rence impulse
Pokorny 2008 Austria 12 > 30 TEP 185:13 0 69 TAPP 93, Polypro- 36 7 Recur- NR
[30] and TEP 36 pylene rence
TAPP
Butters/ 2007/2004 Germany 1 > 100 All male 0 93 94 TAPP NR Median 52 16 Recur- NR
Köninger TAPP (46–60) rence
[31, 32]
Lau [33] 2006 China 1 > 200 TEP All male 0 100 100 TEP Polypro- 12 13 Recur- Reducible ingui-
pylene rence, nal swelling.
light- post- Pain: NR
weight operative
pain,
chronic
pain
Heikkinen 2014/1998/1997 Finland 2 NR 120:3 3.3% 61 TAPP 40, Polypro- Mean 72 1.6 Cost and Cost analysis
[34–37] TEP 22 pylene short- detailed
light- term out-
weight comes

13
465

466 Hernia (2019) 23:461–472

ranged from 12 months to 5 years. All trials reported only

Overall hospi-

bulge in groin
the use of polypropylene mesh for both laparoscopic and

Operative Pain 12/24 h.


Definition of
primary out-

Recurrence:
open repairs.

tal cost
Hernia recurrence was measured as a primary outcome
come

measure in four trials [27, 30, 31, 33]: two reported primar-
ily chronic pain [20, 22], two post-operative pain [21, 25],

term out-

outcome
Outcome

one operative time [38] and one cost [34]. Most of the trials

comes,
Primary

Primary
short-

stated
time,

cost
only included patients with primary unilateral inguinal her-

not
nia except for three trials where the proportion of bilateral
or recurrent hernias was less than 10% of the total patients
follow-up
Patients

[21, 25, 34].


lost to

12.1
Two studies were multi-arm trials with Pokorny et al. [30]
(%)

NR

reporting a three-arm trial (Lichtenstein, TAPP and Shoul-


dice repair) and Butters et al. [31] reporting a five-arm trial
Median 18
Mean 13.5
Follow-up

(Lichtenstein, Bassini, Shouldice, TAPP and TEP). Both


(months)

(8–28)

these trials clearly reported data for the separate groups,


and hence for the purpose of this meta-analysis, the patients
receiving Shouldice and Bassini repairs were excluded.
Polypro-

Polypro-

weight
pylene

pylene
Type of

Additional details on the individual trials methodology is


light-
mesh

reported in Table 2 (Online Appendix 2).


of patients

technique

25 TAPP
Number

70 TEP
laparo-

Data synthesis
scopic

Recurrence
of patients
Lichten-
Number

stein

25

70

Recurrence rates were reported in all 12 trials over a range


from 1 to 5 years. Recurrence was specifically defined by
Proportion
recurrent/

two trials as a reducible swelling in the groin and one as


bilateral
hernia

the presence of an expansile cough impulse. The rest of the


trials did not define recurrence. Most trials assessed hernia
0

recurrence by clinical examination with ultrasonography if


All male

All male

inconclusive. A total of 3623 patients were analysed and


experience female
Male/

ratio

there were no significant differences in effects (OR 1.14, CI


0.51–2.55, p = 0.76) (Fig. 3). There was some evidence of
between-study heterogeneity (I2 = 58%). The funnel plot did
Surgeon

not suggest the presence of publication bias (Fig. 4).


NR

NR

Sensitivity analysis for recurrence rates was performed


and found similar results (risk difference 0.0, CI − 0.01,
centres

0.02, p = 0.74). This has allowed the inclusion of studies


ber of
Num-

that reported a zero recurrence rate hence providing a more


accurate analysis (Fig. 5).
Country

To account for zero events in the subgroup analysis com-


Turkey

Turkey

paring TAPP versus TEP, we performed sensitivity analysis


calculating the risk difference. The test for subgroup dif-
ferences suggested no statistically significant difference
in effects of TAPP versus TEP with regard to recurrence
(χ2 = 0.91, p = 0.34, I2 = 0%) (Fig. 6).
Table 1  (continued)

2004

2003
Year

NR Not recorded

Post‑operative pain
Authors

Anadol

Gokalp

Acute post-operative pain was reported in ten trials with


ences)
(refer-

[38]

[39]

mean VAS scores reported at 24 h in five trials. However,

13
Hernia (2019) 23:461–472 467

Fig. 3  Forest plot of odds ratio for recurrence between laparoscopic and standard mesh for open inguinal hernia. M-H Mantel–Haenszel, CI con-
fidence interval

pain was 3445. Laparoscopic repair was associated with


reduced odds of chronic pain compared to open repair (OR
0.41, 95% CI 0.30–0.56, p ≤ 0.00001) (Fig. 8). There was
moderate between-study heterogeneity (I2 = 33%). Visual
inspection of the funnel plot did not suggest the presence of
publication bias (Fig. 9).
VAS scores for chronic pain were reported by two trials
but specifically mean VAS score was only reported by one
trial.

Other results

Results for wound complications and re-operation rates are


displayed in Online Appendix 3. Quality of life was not
Fig. 4  Funnel plot of treatment effect estimates against the standard
error of the treatment effect for recurrence reported using the same validated questionnaires so could
not be analysed.

only four reported the standard deviations and so could be


Trial sequential analysis
included in the analysis. The number of patients followed up
for acute pain in the analysis was 672. Laparoscopic repair
The Z-curve crosses the futility boundaries; further studies
was associated with reduced rate of acute pain compared to
are thus highly unlikely to demonstrate a statistically signifi-
open repair (MD 1.19, 95% CI − 1.86, − 0.51, p ≤ 0.0006)
cant effect between the two techniques (Fig. 10).
(Fig. 7). There was evidence of substantial between-study
heterogeneity (I2 = 96%).

Chronic pain Discussion

Chronic pain was reported in 12 trials. The definition of This meta-analysis of randomised controlled trials provides
chronic pain was variable and most studies did not provide up-to-date evidence on the comparative effect of laparo-
a definition. Rates of chronic pain were analysed between 1 scopic and open mesh repair of primary unilateral ingui-
and 5 years. The number of patients followed up for chronic nal hernia. It included 12 studies and 3966 patients and

13

468 Hernia (2019) 23:461–472

Fig. 5  Forest plot of the sensitivity analysis of recurrence rates between the laparoscopic and open repairs. M-H Mantel–Haenszel, CI confidence
interval

Fig. 6  Forest plot of the sensitivity analysis of recurrence rates between TAPP and TEP and open repairs. M-H Mantel–Haenszel, CI confidence
interval

demonstrated no difference with respect to recurrence rates. on primary unilateral inguinal hernia [10], which suggests a
However, rates of chronic pain were significantly less within degree of external validity.
the laparoscopic groups. In contrast to previous meta-analyses published on this
These findings are important when the surgical commu- topic, this study focused on primary unilateral inguinal her-
nity reflects on current guidelines [2] and their adoption into nia. Studies with greater than 10% rate of bilateral or recur-
practice. It is noteworthy that in the UK, only approximately rent hernias were excluded from analysis. This is important
20% of primary unilateral inguinal hernias are repaired as recurrence rates for primary recurrent hernia have been
laparoscopically with wide variations between hospitals quoted at 15% [2]. In addition, this analysis has grouped TEP
[40]. The findings of this review regarding recurrence and and TAPP for synthesis. This is because evidence shows
chronic pain are concordant with those from registry studies they have similar complication risks, post-operative acute

13
Hernia (2019) 23:461–472 469

Fig. 7  Forest plot of the mean difference for visual analogue scale (VAS) for acute pain between laparoscopic and open mesh repair for inguinal
hernia. CI confidence interval

Fig. 8  Forest plot of odds ratio for chronic pain between laparoscopic and standard mesh for open inguinal hernia. M-H Mantel–Haenszel, CI
confidence interval

and chronic pain incidence and recurrence rates [2]. Sub-


group analysis was in line with this finding, as it does not
suggest significant difference between the two techniques
with regard to recurrence. Other strengths of this review
include prior registration with the PROSPERO database
prior to commencement, hence minimising reporting bias.
There were no deviations from the protocol and a formal
assessment of publication bias has been performed by way
of funnel plots.
The results of this systematic review and meta-analy-
sis are limited by the methodological inadequacies of the
included randomised studies relating to their design, conduct
and statistical analysis that are covered in the “other bias”
Fig. 9  Funnel plot of treatment effect estimates against the standard
section of the Cochrane risk of bias assessment (Fig. 2a, b).
error of the treatment effect for chronic pain

13

470 Hernia (2019) 23:461–472

Fig. 10  Trial sequential analysis

This necessitates some caution when interpreting the find- invasive (TAPP or TEP) inguinal hernia repair has been met
ings of the review. [44]. One study used expertise randomisation (i.e. differ-
Randomised trials exist on a continuum from those that ent surgeons performed the laparoscopic and open surgery).
are explanatory to those that are pragmatic [41]. In surgery Although this study design has its positive features, failure
and other complex interventions, pragmatic studies are to be to account for surgeon effects in the analysis is an inherent
favoured as they investigate effectiveness (i.e. what happens source of bias and is evidenced by the fact that one surgeon
to the average patient in the hands of the average surgeon) accounted for approximately one-third of the recurrences
and reflect quotidian practice. Studies that are confined to a [27].
low number of centres lack generalisability and cannot reli- The conduct of RCTs is of fundamental importance
ably investigate effectiveness due to inherent systemic biases when interpreting their applicability to one’s own practice.
within the health care unit (usually hospital) and the limited The risk of bias assessment highlights a number of areas
number of surgeons involved. Only three studies could truly of concern, particularly with regard to blinding of sur-
be considered multicentre [25, 27, 30]. The minimum sur- geons, participants and assessors. Furthermore, the lack of
geon skill level mandated for participation in each of the standardised definitions for the outcomes and a widespread
included trials also varies considerably. Laparoscopic ingui- failure to use appropriately validated assessment tools for
nal hernia repair is a challenging operation for the surgeon recurrence, complications (e.g. Clavien Dindo [45]), pain
and there is evidence that it is associated with a steep learn- and quality of life are of great concern. Bhangu et al. [46]
ing curve during which recurrence rates increase [42]. The have presented data showing that outcome reporting from
European Hernia Society states that 100 laparoscopic cases RCTs concerning inguinal hernia repair is inconsistent and
are required to achieve outcomes comparable with open poorly defined, which has limited meta-analyses. A core
mesh repair [2]. In addition, complication rates decrease outcome dataset would standardise reporting and conse-
by 50% after 50 cases. In this review, only three studies quently improve quality of RCTs on inguinal hernias, but
required surgeons of experience in laparoscopic hernia to as yet one does not exist.
be more than 100 [20, 31, 33]. The other studies either did The role of clinical trials units (CTU) in the success
not report surgeon experience or included surgeons still on of RCTs remains hotly debated, but their role in develop-
the learning curve. Furthermore, surgeon volume and opera- ing robust designs with appropriate statistical analysis is
tive quality, although linked, are not synonymous and there widely acknowledged [47]. In this review, the lack of CTU
is no evidence of objective operative quality assessment in input is highlighted by the failure to provide a sample size
either arm (c.f. the ROMIO study [43]) and no evidence calculation in half of the studies [24, 30, 31, 34, 38, 39].
provided that the appropriate quality standard for minimally Of those that did provide a sample size calculation, only

13
Hernia (2019) 23:461–472 471

four accounted for loss to follow-up [20, 22, 25, 27]. Clear for management of chronic postoperative inguinal pain. Hernia
statements of the use of intention to treat analysis and of 19(1):33–43
6. O’reilly EA, Burke JP, O’connell PR (2012) A meta-analysis
intended study design (superiority or non-inferiority etc.) of surgical morbidity and recurrence after laparoscopic and
were similarly sparse. Only one study [25] could truly be open repair of primary unilateral inguinal hernia. Ann Surg
considered high quality, and while a specific CTU was not 255(5):846–853
used in this study, the REPAIR group from Rotterdam has 7. Bobo Z, Nan W, Qin Q, Tao W, Jianguo L, Xianli H (2014) Meta-
analysis of randomized controlled trials comparing Lichtenstein
extensive expertise in RCTs. and totally extraperitoneal laparoscopic hernioplasty in treatment
The trial sequential analysis utilising all 12 RCTs has of inguinal hernias. J Surg Res 192(2):409–420
suggested that further studies are highly unlikely to dem- 8. Scheuermann U, Niebisch S, Lyros O, Jansen-Winkeln B, Gockel
onstrate a statistically significant effect between the two I (2017) Transabdominal Preperitoneal (TAPP) versus Lichten-
stein operation for primary inguinal hernia repair: a systematic
techniques in terms of recurrence for primary simple uni- review and meta-analysis of randomized controlled trials. BMC
lateral inguinal hernia. However, the methodological qual- Surg 17(1):55
ity of the RCTs is such that we cannot state this with any 9. Patterson TJ, Beck J, Currie PJ, Spence RA, Spence G (2019)
certainty regarding other outcomes, particularly those that Meta-analysis of patient-reported outcomes after laparoscopic ver-
sus open inguinal hernia repair. Br J Surg. https:​ //doi.org/10.1002/
matter to patients such as chronic pain and quality of life. bjs.11139​
In conclusion, this meta-analysis and trial sequential 10. Köckerling F, Stechemesser B, Hukauf M, Kuthe A, Schug-Pass C
analysis reports no difference in recurrence rates between (2016) TEP versus Lichtenstein: which technique is better for the
laparoscopic and open primary unilateral inguinal hernia repair of primary unilateral inguinal hernias in men? Surg Endosc
30(8):3304–3313
repair. Laparoscopic groin hernia repair is associated with 11. University of York, Centre for Reviews and Dissemination. PROS-
less acute pain and reduced odds for chronic pain. PERO; International prospective register of systematic reviews.
http://www.crd.york.ac.uk/PROSP​ERO. Accessed 1 Apr 2019
12. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioan-
Funding  There were no sources of funding associated with this review. nidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D (2009)
The PRISMA statement for reporting systematic reviews and
meta-analyses of studies that evaluate health care interventions:
Compliance with ethical standards  explanation and elaboration. PLoS Med 6(7):e1000100
13. Robinson KA, Dickersin K (2002) Development of a highly sensi-
Conflict of interest  NLB, LHM, SAA, NJS and RHF declare no con- tive search strategy for the retrieval of reports of controlled trials
flict of interest. using PubMed. Int J Epidemiol 31(1):150–153
14. Treede RD, Rief W, Barke A, Aziz Q, Bennett MI, Benoliel
Ethical approval  Approval from the institutional review board was not R, Cohen M, Evers S, Finnerup NB, First MB, Giamberardino
required for this study. MA (2015) A classification of chronic pain for ICD-11. Pain
156(6):1003
Human and animal rights  This article is a systematic review and meta- 15. Alfieri S, Amid PK, Campanelli G, Izard G, Kehlet H, Wijsmuller
analysis and it contains data collected through literature review. It does AR, Di Miceli D, Doglietto GB (2011) International guidelines
not include research directly involving human or animal participation. for prevention and management of post-operative chronic pain
following inguinal hernia surgery. Hernia 15:239. https​://doi.
Informed consent  For this literature review, formal consent is not org/10.1007/s1002​9-011-0798-9
required. 16. Higgins JPT, Deeks JJ (2011) Obtaining standard deviations from
standard errors, confidence intervals, t values and p values for dif-
ferences in means. Cochrane handbook for systematic reviews of
interventions. Version 5.1.0. https​://handb​ook-5-1.cochr​ane.org/.
References Accessed 1 Apr 2019
17. DeMets D, Lan KK (1994) Interim analysis: the alpha spending
function approach. Stat Med 12:1341–1352
1. Kingsnorth A, LeBlanc K (2003) Hernias: inguinal and inci- 18. O’Brien PC, Fleming TR (1979) A multiple testing procedure for
sional. Lancet 362(9395):1561–1571 clinical trials. Biometrics 35:549–556
2. The HerniaSurge Group (2018) International guidelines for 19. Higgins JP, Altman DG (2009) Assessing risk of bias in
groin hernia management. Hernia 22(1):1–65 included studies. In: Higgins JP, Green S (eds) Cochrane hand-
3. Simons MP, Aufenacker TJ, Berrevoet F, Bingener J, Bisgaard book for systematic reviews of interventions. Wiley, West Sus-
T, Bittner R, Bonjer HJ, Bury K, Campanelli G, Chen DC, sex, pp 187–235
Chowbey PK (2017) World guidelines for groin hernia man- 20. Sevi̇nç B, Damburaci N, Güner M, Karahan Ö (2019) Comparison
agement. http://news.europ​eanhe​rnias​ociet​y.eu/sites​/www.europ​ of early and long term outcomes of open Lichtenstein repair and
eanhe​r nias​ociet​y.eu/files​/media​s/PDF/Herni​aSurg​eGuid​eline​ totally extraperitoneal herniorrhaphy for primary inguinal hernias.
sPART​1TREA​TMENT​.pdf. Accessed 1 Apr 2019 Turk J Med Sci 49(1):38–41
4. Nordin P, Bartelmess P, Jansson C, Svensson C, Edlund G (2002) 21. Koju R, Koju RB, Malla B, Dongol Y, Thapa LB (2017) Transab-
Randomized trial of Lichtenstein versus Shouldice hernia repair dominal pre-peritoneal mesh repair versus Lichtenstein’ s
in general surgical practice. Br J Surg 89(1):45–49 hernioplasty. J Nepal Health Res Counc 15(2):135–140
5. Lange JFM, Kaufmann R, Wijsmuller AR, Pierie JPEN, Ploeg RJ, 22. Westin L, Wollert S, Ljungdahl M, Sandblom G, Gunnarsson
Chen DC, Amid PK (2015) An international consensus algorithm U, Dahlstrand U (2016) Less pain 1 year after total extra-peri-
toneal repair compared with Lichtenstein using local anesthesia:

13

472 Hernia (2019) 23:461–472

data from a randomized controlled clinical trial. Ann Surg Lichtenstein hernioplasties. Surg Endosc Other Interv Tech
263(2):240–243 18(3):518–522
23. Dahlstrand U, Sandblom G, Ljungdahl M, Wollert S, Gun- 35. Heikkinen TJ, Haukipuro K, Hulkko A (1998) A cost and outcome
narsson U (2013) TEP under general anesthesia is superior to comparison between laparoscopic and Lichtenstein hernia opera-
Lichtenstein under local anesthesia in terms of pain 6 weeks after tions in a day-case unit. Surg Endosc 12(10):1199–1203
surgery: results from a randomized clinical trial. Surg Endosc 36. Heikkinen TJ, Haukipuro K, Koivukangas P, Hulkko A (1998) A
27(10):3632–3638 prospective randomized outcome and cost comparison of totally
24. Wang WJ, Chen JZ, Fang Q, Li JF, Jin PF, Li ZT (2013) Com- extraperitoneal endoscopic hernioplasty versus Lichtenstein her-
parison of the effects of laparoscopic hernia repair and Lichten- nia operation among employed patients. Surg Laparosc Endosc
stein tension-free hernia repair. J Laparoendosc Adv Surg Tech 8(5):338–344
23(4):301–305 37. Heikkinen T, Haukipuro K, Leppälä J, Hulkko A (1997) Total
25. Eker HH, Langeveld HR, Klitsie PJ, van’t Riet M, Stassen LP, costs of laparoscopic and Lichtenstein inguinal hernia repairs: a
Weidema WF, Steyerberg EW, Lange JF, Bonjer HJ, Jeekel J randomized prospective study. Surg Laparosc Endosc 7(1):1–5
(2012) Randomized clinical trial of total extraperitoneal inguinal 38. Anadol ZA, Ersoy E, Taneri F, Tekin E (2004) Outcome and cost
hernioplasty vs Lichtenstein repair: a long-term follow-up study. comparison of laparoscopic transabdominal preperitoneal hernia
Arch Surg 147(3):256–260 repair versus open Lichtenstein technique. J Laparoendosc Adv
26. Langeveld HR, van’t Riet M, Weidema WF, Stassen LP, Steyer- Surg Tech 14(3):159–163
berg EW, Lange J, Bonjer HJ, Jeekel J (2010) Total extraperitoneal 39. Gokalp A, Inal M, Maralcan G, Baskonus I (2003) A prospec-
inguinal hernia repair compared with Lichtenstein (the LEVEL- tive randomized study of Lichtenstein open tension-free versus
Trial): a randomized controlled trial. Ann Surg 251(5):819–824 laparoscopic totally extraperitoneal techniques for inguinal hernia
27. Eklund A, Montgomery A, Bergkvist L, Rudberg C (2010) repair. Acta Chir Belg 103(5):502–506
Chronic pain 5 years after randomized comparison of laparoscopic 40. http://www.rcseng​ .ac.uk/commis​ ionin​ g/groin-​ hernia​ -commis​ sion​
and Lichtenstein inguinal hernia repair. Br J Surg 97(4):600–608 ing-guide​. Accessed 1 May 2019
28. Eklund AS, Montgomery AK, Rasmussen IC, Sandbue RP, Berg- 41. https​://www.preci​s-2.org/. Accessed 1 May 2019
kvist LÅ, Rudberg CR (2009) Low recurrence rate after laparo- 42. Neumayer L, Giobbie-Hurder A, Jonasson O, Fitzgibbons R Jr,
scopic (TEP) and open (Lichtenstein) inguinal hernia repair: a Dunlop D, Gibbs J, Reda D, Henderson W (2004) Open mesh
randomized, multicenter trial with 5-year follow-up. Ann Surg versus laparoscopic mesh repair of inguinal hernia. N Engl J Med
249(1):33–38 350(18):1819–1827
29. Eklund A, Rudberg C, Smedberg S, Enander LK, Leijonmarck 43. Blencowe NS, Skilton A, Gaunt D, Brierley R, Hollowood A,
CE, Österberg J, Montgomery A (2006) Short-term results of a Dwerryhouse S, Higgs S, Robb W, Boddy A, Hanna G, Barham
randomized clinical trial comparing Lichtenstein open repair with CP (2019) Protocol for developing quality assurance measures to
totally extraperitoneal laparoscopic inguinal hernia repair. Br J use in surgical trials: an example from the ROMIO study. BMJ
Surg: Inc Eur J Surg Swiss Surg 93(9):1060–1068 Open 9(3):e026209
30. Pokorny H, Klingler A, Schmid T, Fortelny R, Hollinsky C, Kawji 44. Daes J, Felix E (2017) Critical view of the myopectineal orifice.
R, Steiner E, Pernthaler H, Függer R, Scheyer M (2008) Recur- Ann Surg 266(1):e1–e2
rence and complications after laparoscopic versus open inguinal 45. Clavien PA, Barkun J, De Oliveira ML, Vauthey JN, Dindo D,
hernia repair: results of a prospective randomized multicenter Schulick RD, De Santibañes E, Pekolj J, Slankamenac K, Bassi
trial. Hernia 12(4):385–389 C, Graf R (2009) The Clavien-Dindo classification of surgical
31. Butters M, Redecke J, Köninger J (2007) Long-term results of a complications: five-year experience. Ann Surg 250(2):187–196
randomized clinical trial of Shouldice, Lichtenstein and transab- 46. Bhangu A, Singh P, Pinkney T, Blazeby JM (2015) A detailed
dominal preperitoneal hernia repairs. Br J Surg: Inc Eur J Surg analysis of outcome reporting from randomised controlled trials
Swiss Surg 94(5):562–565 and meta-analyses of inguinal hernia repair. Hernia 19(1):65–75
32. Köninger J, Redecke J, Butters M (2004) Chronic pain after hernia 47. Gohel MS, Chetter I. Are clinical trials units essential for a suc-
repair: a randomized trial comparing Shouldice, Lichtenstein and cessful trial?. Bmj. 2015 May 27;350:h2823
TAPP. Langenbeck’s Arch Surg 389(5):361–365
33. Lau H, Patil NG, Yuen WK (2006) Day-case endoscopic totally Publisher’s Note Springer Nature remains neutral with regard to
extraperitoneal inguinal hernioplasty versus open Lichtenstein jurisdictional claims in published maps and institutional affiliations.
hernioplasty for unilateral primary inguinal hernia in males. Surg
Endosc Other Interv Tech 20(1):76–81
34. Heikkinen T, Bringman S, Ohtonen P, Kunelius P, Haukipuro
K, Hulkko A (2004) Five-year outcome of laparoscopic and

13

You might also like