t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

Available online at www.sciencedirect.com

ScienceDirect The Surgeon, Journal of the Royal Colleges of Surgeons of Edinburgh and Ireland www.thesurgeon.net

Review

Totally extraperitoneal laparoscopic hernioplasty versus open extraperitoneal approach for inguinal hernia repair: A meta-analysis of outcomes of our current knowledge Xiang Zhu a, Hongyong Cao a,*, Yong Ma a, Aihua Yuan a, Xiangyang Wu a, Yi Miao b, Song Guo a a

Department of General Surgery, Nanjing First Hospital, Nanjing Hospital Affiliated to Nanjing Medical University, 210000 Nanjing, Jiangsu, China b Department of General Surgery, The People’s Hospital of Jiangsu Province, Affiliated to Nanjing Medical University, 210000 Nanjing, Jiangsu, China

article info

abstract

Article history:

Background: The aim of this article is to explore the clinical effects between open extrap-

Received 18 September 2013

eritoneal approaches and totally extraperitoneal laparoscopic hernioplasty (TEP) in the

Received in revised form

repair of inguinal hernias.

10 November 2013

Methods: The electronic databases Pubmed, Medline, Embase, Web of science and the

Accepted 14 November 2013

Cochrane Library were used to search for articles from January 1992 to March 2013. The

Available online 8 December 2013

present meta-analysis pooled the effects of outcomes of a total of 1157 patients with 1377 hernias enrolled into 10 randomized controlled trials and 2 comparative studies. The data

Keywords:

was analyzed using the statistic software Stata12.0 and IBM SPSS Statistics 19.

Inguinal hernia

Results: Significant advantages of totally extraperitoneal laparoscopic hernioplasty (TEP)

TEP

compared to the open extraperitoneal approach include a lower incidence of total post-

Open extraperitoneal approach

operative complications (Odds Ratio, 0.544; 95% confidence interval, 0.369e0.803), a

Hernioplasty

reduction in urinary problems (0.206[0.064,0.665]), an earlier return to normal activities or

Randomized controlled trials

work (SMD ¼ 1.798[3.322,0.275]), and a shorter length of hospital stay (1.995

Meta-analysis

[2.358,1.632]). No difference was found in operative time, the incidence of hernia recurrence, chronic pain, intraoperative complications, seromas or hematomas, wound infection and testicular problems between the two techniques. One significant advantage for the open extraperitoneal inguinal hernia repair was a lower incidence of peritoneal tears (46.504 [15.399,140.437]). Conclusions: Totally extraperitoneal laparoscopic hernioplasty (TEP) and open extraperitoneal mesh repair are equivalent in most of the analyzed outcomes. TEP is associated with shorter hospital stay, quicker return to normal activities or work, lower incidence of

* Corresponding author. Tel.: þ86 25 18951670281. E-mail address: [email protected] (H. Cao). 1479-666X/$ e see front matter ª 2014 Royal College of Surgeons of Edinburgh (Scottish charity number SC005317) and Royal College of Surgeons in Ireland. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.surge.2013.11.018

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

95

total postoperative complications and urinary problems, while the open extraperitoneal method has less incidence of peritoneal tears. ª 2014 Royal College of Surgeons of Edinburgh (Scottish charity number SC005317) and Royal College of Surgeons in Ireland. Published by Elsevier Ltd. All rights reserved.

Introduction Inguinal hernia repair is one of the most common surgical operations in general surgery. Since Lichtenstein1 described his tension-free hernioplasty in 1989, tension-free hernia repair was quickly accepted by most surgeons as the effective and safe method of hernia repair for its lower recurrence rate, less postoperative pain and easy to learn, various surgical methods were described and invented from that time on. In tension-free hernioplasty, a prosthesic mesh can be placed subaponeurotically or extraperitoneally, either through an open approach or laparoscopically.2 With improved understanding of the groin anatomical structure, especially the Fruchauds myopectineal orifice, reinforcing the extraperitoneal space and completely covering the orifice seem to be the most effective and reasonable method for hernia repair currently.3 Many open techniques that combined the benefit of tensionfree with the advantages of the extraperitoneal approach have been in use for decades. For instances, Stoppa4 developed his technique through a lower midline incision putting a giant prosthesis in the extraperitoneal space with good results. The Kugel and Modified Kugel methods both place the polypropylene mesh in the extraperitoneal space the posterior and anterior approaches respectively.5 In addition, the transinguinal preperitoneal technique (TIPP) and the Prolene hernia system (PHS) both are commonly used technique that proved to be successful.6 Laparoscopic repairs combine the advantages of minimal access surgery with the open extraperitoneal approach. The transabdominal preperitoneal repair (TAPP) and the totally extraperitoneal repair(TEP) are the two most frequently-used methods. More surgeons prefer the latter for its not entering into the peritoneum.7 Many researches have shown that laparoscopic hernia repair may offer less postoperative pain and early return to normal activities compared with open method. However, its potential intraoperative complications, need for general anesthesia and long learning curve have restricted its use to some extent.8 To date, clinical comparisons between TEP and open extraperitoneal herniorrhaphy are not very abundant. There is no meta-analysis directly comparing the outcomes of laparoscopic extraperitoneal herniorrhaphy and open extraperitoneal mesh repair. In the present article, different types of open extraperitoneal repairs with prosthetic meshes are combined as they all achieve similar clinical goals.6

procedures for the repair of groin hernias were identified intensively in the electronic databases Pubmed, Embase, Web of Science, Medline, and the Cochrane Library from January 1992 to March 2013. The search strategies used the following major medical terms: “inguinal hernia”, “extraperitoneal”, “laparoscopic”, “OPM”, “Stoppa”, “Kugel”, “PHS”, and “repair/ hernioplasty”. The function of “related articles” in the database was used to broaden the search results and all abstracts, comparative studies and citations scanned were reviewed comprehensively. Two comparative studies and 10 randomized controlled trials (RCTs) were ascertained finally.

Selection criteria To be included in this analysis, studies had to: compare TEP and the open extraperitoneal procedures for the repair of groin hernias; RCT or well-designed comparative study; be published as full-length articles; report on at least one of the following outcomes: ⑴operative time, ⑵postoperative complications, ⑶hospital stay, ⑷chronic pain, ⑸delay in return to normal activities or work, ⑹recurrences, ⑺intraoperative complications, ⑻conversion, ⑼wound infection, ⑽hematomas, ⑾urinary problems, ⑿seroma, ⒀testicular/scrotal problems, or ⒁peritoneal tears (Table 1). Articles are excluded if they did not compare the two procedures; only reported one method or surgical experience; not a RCT or prospective comparative study; and difficult to extract the appropriate data from its results.

Data extraction Two independent researchers for eligibility in meta-analysis extracted the following information from each article separately: first author, publication year, study design, country of origin, matching criteria, and outcomes. Any disagreements were resolved by consensus.

Study quality assessment According to the Cochrane Handbook for Systematic Reviews of Interventions 5.0.2,9 the risk of bias of each included trial was assessed by two reviewers independently, which was judged using the following methodologic criterias: sequence generation, allocation concealment, blinding of participants, personnel, and outcome assessors, incomplete outcome data, free of selective outcome reporting, and other bias10 (Table 2). If any information was not available, authors with access to the raw data was contacted by e-mail.

Materials and methods Statistical analysis Literature search All randomized controlled trials and prospective case control studies that compared TEP and the open extraperitoneal

The statistical analysis was carried out with statistic software STATA12.0 and IBM SPSS Statistics 19. For dichotomous data, results for each trial were expressed as relative risks(RRs) or

96

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

Table 1 e Basic information of the trials ⑴operative time ⑵total postoperative complications ⑶hospital stay ⑷postoperative chronic pain ⑸delay in return to work/normal activities ⑹recurrences ⑺intraoperative complications ⑻conversion in TEP ⑼wound infection ⑽hematomas ⑾urinary problems ⑿seroma ⒀testicular/scrotal problems ⒁peritoneal tears. Study

Type of inguinal hernia

1 Champault GG et al., 1997 (RCT) 2 Gainant A et al., 2000 (observational) 3 Bender O et al., 2009(RCT) 4 Feliu X et al., 2004 (observational) 5 Sinha R et al., 2006(RCT) 6. Gu¨nal O et al., 2007 (RCT) 7 Suter M et al, 2002 (RCT) 8 Ozmen M et al., 2010 (RCT) 9 Ozmen MM et al., 2004 (RCT) 10 Vatansev C et al., 2002(RCT) 11 Alani A et al., 2006 (RCT) 12 Hamza Y et al., 2010 (RCT)

Sample size (G1/G2)

Group 1 (G1)

Group 2 (G2)

Outcome (extracted for meta-analysis)

Follow-up

Primary unilateral and recurrent Bilateral

49/51

Stoppa

TEP

⑵⑶⑷⑸⑹⑺

3y

74/43(148/86)

Stoppa

TEP

⑻⑼⑽⑾⒁

NG

Primary unilateral Recurrent unilateral

20/20 121/86

Kugel OPM

TEP TEP

⑴⑵⑶⑸⑹⑺ ⑻

2.3 mo 5y

Primary and recurrent Primary unilateral Bilateral Bilateral Unilateral Primary unilateral Recurrent Primary unilateral

121/120 39/40 20/19(40/38) 32/32(64/64) 40/40 21/20 54/45 25/25

APP Nyhus Stoppa Stoppa Nyhus Nyhus OPM OPM

TEP TEP TEP TEP TEP TEP TEP TEP

⑴⑵⑶⑸⑹ ⑴⑵⑶⑷⑹ ⑻⑼⑽⑾⑿⒀⒁ ⑴⑶⑷⑸⑹⑺ ⑻⑼⑽⑿⒁ ⑵⑹⑺ ⑽⑾ ⑴⑵⑶⑸⑹ ⑴⑵⑶⑺⑼ ⑽⑾⑿⒀⒁ ⑴⑵⑶⑷⑼ ⑽⑾⑿⒀⒁ ⑴ ⑵⑷⑹ ⑼⑽⑿⒀ ⑴⑵⑶⑷⑸⑹ ⑻⑼⑽⒀⒁

1y 7y 1y 18 mo 5y NG 5y 6 mo

⑴operative time ⑵total postoperative complications ⑶hospital stay ⑷postoperative chronic pain ⑸delay in return to work/normal activities ⑹recurrences ⑺intraoperative complications ⑻conversion in TEP ⑼wound infection ⑽hematomas ⑾urinary problems ⑿seroma ⒀testicular/ scrotal problems ⒁peritoneal tears.

odds ratios(ORs) with 95% confidence intervals(CIs), continuous variables, the mean difference(MD) or standardized mean difference(SMD) with 95% CI were calculated using a fixedeffects model or a randomized-effects model according to heterogeneity.11 Data for continuous outcomes was excluded if standard deviations of the mean were not given and that could not be calculated or estimated.12 Statistical heterogeneity between trials was assessed using the chi-squared statistic and the extent of inconsistency was assessed using the I2 statistic.13 Subgroup analysis was performed for operative time of

bilateral and unilateral groups. Sensitivity analysis was conducted by removing individual studies that have a high risk of bias.14 Publication bias was tested using the Egger test. Results were significant if the P < 0.05.

Results The flowchart of literature selecting is presented in Fig. 1. Eventually, the meta-analysis included two well-designed

Table 2 e Risk of bias in the randomized controlled trials. Study

Randomization

Allocation concealment

Blinding

Incomplete outcome data

Free of selective reporting

Other sources of bias

Champault GG et al., 1997 Bender O et al.,2009 Sinha R et al., 2006 Gu¨nal O et al., 2007 Suter M et al., 2002

Random number tables Computer Alternately assigned Computer Mentioned

Unclear Unclear Central allocation Unclear Sealed envelopes

Yes Yes Yes Yes Yes

Unclear Unclear Unclear Unclear Unclear

Unclear Unclear Unclear Unclear Unclear

Ozmen M et al,2010 Ozmen MM et al., 2004

Random number tables Random number tables

Unclear Unclear

Yes Yes

Unclear Unclear

Unclear Unclear

Vatansev C et al., 2002 Alani A et al., 2006 Hamza Y et al., 2010

Mentioned Computer Mentioned

Sealed envelopes Unclear Random number allocation

Unclear Unclear Unclear Unclear Doubleblind Unclear Doubleblind Unclear Unclear Double blinded

Yes Yes Yes

Unclear Unclear Unclear

Unclear Unclear Unclear

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

97

Fig. 1 e Flowchart for selection of trials.

comparative studies and 10 randomized controlled trials containing a total of 1157 patients (1377 hernias) older than 18 years of age.3,4,15e24 The 2 comparative studies20,21 had a large sample size, enough information and comparable baseline characteristics. Though one of them21 was written in French, the abstract and useful needed contents were in English and could be extracted. The publication dates ranged from 1997 to 2010. The characteristics of all included trials are summarized in Table 1.

Hernia recurrence Data concerning hernia recurrence rates was included in 93,4,15,17e22 of the analyzed studies. There were a total of eleven recurrences (2.2%) in TEP group and nine (1.5%) in the open group respectively (P ¼ 0.379). The fixed-effects model was appropriate for there was no statistical heterogeneity (Isquared ¼ 0.0%, p heterogeneity ¼ 0.829, p > 0.05). Overall, there was no significant difference in hernia recurrence rates between the open extraperitoneal and TEP group (OR ¼ 1.387; 95% confidence interval [CI], 0.596e3.228, p ¼ 0.448; Fig. 2). Publication bias was tested with Egger’test, and no publication bias was detected in the included studies (p ¼ 0.768).

Chronic pain Five3,17,19,20,24 of the 12 studies reported chronic pain (>3 months). The incidence of total postoperative chronic pain was 1.6% in TEP group and 2.5% in the open extraperitoneal mesh repair group (P ¼ 0.406). No significant heterogeneity was found among them [I-squared ¼ 17.6%, p heterogeneity ¼ 0.297, p > 0.05], so the fixed-effects model was used to perform meta-analysis. The results showed that no statistical difference between the two groups was found for chronic pain [OR ¼ 0.697; 95% confidence interval [CI], 0.235e2.063, p ¼ 0.514; Fig. 3].

Intraoperative complication Six3,15,16,18,21,24 studies reported the incidence of intraoperative complications (such as vascular injury, spermatic cord or inguinal nerves injury). In TEP group, there were twelve instances (3.0%) of intraoperative complications, and there were ten patients (2.2%) in the open group (P ¼ 0.445). The randomized-effects model was used because of the heterogeneity (I-squared ¼ 54.6%, I2 > 50%). The results showed that there was no significant difference in the incidence of

98

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

Fig. 2 e Meta-analysis for the outcome of hernia recurrence.

intraoperative complications between the open extraperitoneal group and TEP group (OR ¼ 1.595; 95% confidence interval [CI], 0.327e7.784, p ¼ 0.564; Fig. 4).

effects model was used. Compared with the open extraperitoneal hernia repair, TEP showed more peritoneal tears during surgery (26.2%, P ¼ 0.000).

Peritoneal tears

Total postoperative complications

The incidence of inadvertent peritoneal tears during operation was examined in six of 12 analyzed studies.3,16e18,20,24 The meta-analysis showed that there was significantly statistical difference in inadvertent peritoneal tears (OR ¼ 46.504; 95% confidence interval [CI], 15.399e140.437, p ¼ 0.000). The heterogeneity of the studies was not significant (Isquared ¼ 43.8%, p heterogeneity ¼ 0.113, p > 0.05), and the fixed-

There was no significant heterogeneity among the eleven3,4,15e22,24 of 12 trials (I-squared ¼ 0.0%, p heterogeneity ¼ 0.613, p > 0.05), therefore, the fixed-effects model was appropriate. TEP group showed less total postoperative complications compared with the open extraperitoneal group (OR ¼ 0.544; 95% confidence interval [CI], 0.369e0.803, p ¼ 0.002; Fig. 5A). The incidence of total postoperative complications was 7.2%

Fig. 3 e Meta-analysis for the outcome of chronic pain.

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

99

Fig. 4 e Meta-analysis for the outcome of intraoperative complication.

(44/615) in TEP group and 11.9% (86/721) in the open extraperitoneal group (P ¼ 0.003). The sensitivity of the results was also tested (Fig. 5B). Moreover, we excluded two trials which were not RCTs,20,21 and received similar results (OR ¼ 0.522; 95% confidence interval [CI], 0.326e0.837, p ¼ 0.007; Fig. 5C). Publication bias was tested with Egger’s test and Begg’s test, and no publication bias was detected in the included studies (p ¼ 0.983 in Egger’s test; Begg’s funnel plot shown as Fig. 5D).

Seromas and hematomas Five3,16,19,20,24 studies reported the incidence of seromas and eight3,15e20,24 reported the hematomas respectively. The fixedeffects model was both used for there was no statistical heterogeneity (I-squared ¼ 0.0%, p heterogeneity ¼ 0.392; Isquared ¼ 0.0%, p heterogeneity ¼ 0.815). There was no significant difference in the incidence of seromas and hematomas

Fig. 5 e A) Meta-analysis for the outcome of total postoperative complications. (B). Sensitivity analysis for the outcome of total postoperative complications. (C). Meta-analysis for the outcome of total postoperative complications; two trials that were not RCTs were excluded from analysis (Feliu X and Gainant A). (D). The Begg’s test publication bias plot for total postoperative complications.

100

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

between the two groups (OR ¼ 0.685, 95% confidence interval [CI], 0.260e1.808, p ¼ 0.445; OR ¼ 0.862, 95% confidence interval [CI], 0.473e1.573, p ¼ 0.629).

Wound infection Seven trials3,16e20,24 reported postoperative wound infection, there were a total of five cases (1.2%) of wound infection in TEP group and twelve (2.5%) in the open group (P ¼ 0.129). The fixed-effects model was chosen because heterogeneity of the studies was not significant (p heterogeneity ¼ 0.930, Isquared ¼ 0.0%). The results showed no significant difference in wound infection (OR ¼ 0.488, 95% confidence interval [CI], 0.188e1.269, p ¼ 0.141).

Scrotal/testicular problems Five trials16,17,19,20,24 could be included for analyzing for scrotal problems. Overall, there was no significant heterogeneity between the two groups (I-squared ¼ 0.0%, p heterogeneity ¼ 0.836) and no significant difference (OR ¼ 0.699, 95% confidence interval [CI], 0.274e1.783, p ¼ 0.453).

Urinary problems The incidence of postoperative urinary problems (ie, urinary retention or infection) was also reported in five15,16,18,20,24 of 12 analyzed studies. The incidence of urinary problems was significantly less after laparoscopic surgery: 2 cases (0.7%) vs. 16 cases (5.1%) for the open methods (P ¼ 0.004). The fixedeffects model was used for there was no statistical heterogeneity (I-squared ¼ 0.0%, p heterogeneity ¼ 0.796). Compared with the open extraperitoneal group, TEP group showed less incidence of urinary problems (OR ¼ 0.206; 95% confidence interval [CI], 0.064e0.665, p ¼ 0.008; Fig. 6). We recalculated the results with relative risk and got the same conclusion. Publication bias was tested with Egger’test, and no publication bias was detected in the included studies (p ¼ 0.538).

Operative time Ten trials reported data for the operative time, but in two studies,3,21 standard deviations of the mean which were not given and could not be calculated or estimated.12 The remaining eight trials4,15e17,20,22e24 was pooled for metaanalysis. Only one20 of the eight trials performed a subgroup analysis for bilateral and unilateral hernia. The randomizedeffects model was used because of the heterogeneity (p heterogeneity ¼ 0.000, I-squared ¼ 96.5%), and the overall effect showed that the operative time of the open extraperitoneal group was shorter than that of TEP group (SMD ¼ 1.045, 95% confidence interval [CI], 0.043e2.047, p ¼ 0.041; Fig. 7A). while a sensitivity analysis and meta-regression disclosed that the study by Gu¨nal O et al.15 contributed to the statistical heterogeneity of studies. We arrived at no significant difference in operative time (SMD ¼ 0.265, 95% confidence interval [CI], 0.350e0.879, p ¼ 0.399; Fig. 7B) following appropriate omissions. The mean operative time (Gu¨nal O et al. excluded) was 55.87  26.64 min in TEP group and 54.36  19.97 min in the open extraperitoneal mesh repair group (P > 0.05). A subgroup analysis based on one trial20 showed there was no significant difference in operative time of bilateral hernia too (SMD ¼ 0.180, 95% confidence interval [CI], 1.093e0.733, p ¼ 0.699). The open extraperitoneal repairs did not appear to have a significant impact on the operative time compared with TEP.

Length of hospital stay (days) Five of the twelve trials gave the total duration of hospital stay.16,18,20,21,24 Regardless of the type of hernia, length of hospital stay was in all cases significantly shorter (P < 0.01) after TEP (1.84  1.45 days) when compared with the open extraperitoneal operation (4.28  2.20 days). The randomizedeffects model was used because of the heterogeneity (Isquared ¼ 66.1%, I2>50%). The results of meta-analysis

Fig. 6 e Meta-analysis for the outcome of urinary problems.

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

101

Fig. 7 e (A) Meta-analysis for the outcome of operative time. (B). Meta-analysis for the outcome of operative time; one trial was excluded after Sensitivity analysis (Gu ¨ nal O).

showed that the total length of hospital stay in TEP group was significantly shorter than that of the open extraperitoneal group (SMD ¼ 1.995, 95% confidence interval [CI], 2.358 w 1.632, p ¼ 0.000; Fig. 8). The same conclusion was obtained by recalculating the results with risk difference (RD) and relative risk (RR) (The heterogeneity in the included trials would be discussed in the comments section).

return to normal activities/work (SMD ¼ 1.798, 95% confidence interval [CI], 3.322 w 0.275, p ¼ 0.021; Fig. 9A). To test the sensitivity of the results, we omitted the trial with poor quality,21 and got the similar results (SMD ¼ 0.934, 95% confidence interval [CI], 1.701 w 0.168, p ¼ 0.017; Fig. 9B). So return to work was significantly earlier (P < 0.05) after TEP (13.07  8.11 vs. 28.11  12.41 days), without difference concerning the sort of work.

Delay in return to normal activities/work (days) Conversions of TEP Six trials3,4,17,18,21,22 investigated the time needed to return to normal activities/work and 54,17,18,21,22 of them whose data could be pooled for meta-analysis. The results showed that TEP appeared to have a significant influence on the time to

There were 5 trials3,17,18,20,21 that reported the incidence of conversions in TEP group. Of a total of 368 patients, there were 15 conversions from TEP to TAPP or open hernia repair

102

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

Fig. 8 e Meta-analysis for the outcome of length of hospital stay.

procedures. The total conversion rate of TEP group in our present study is 4.08%. Table 3 summarize the results of our meta-analysis and describe the detailed differences between the two techniques.25

Discussion There is still an ongoing debate on the first line treatment for inguinal hernia. So far, there is no consensus about the best method for surgical repair.26 Currently, the reinforcement of the Fruchauds myopectineal orifice in the extraperitoneal space seems to be the most logical and effective method of herniorraphy. The meshes are all placed in the extraperitoneal space and completely cover the Fruchauds orifice either through an open approach or a laparoscopic way. The open extraperitoneal techniques utilize the posterior or anterior access to place the prosthesis in the preferred extraperitoneal space without extensive dissection of the inguinal canal. Many open extraperitoneal mesh repair techniques such as the Stoppa technique, Nyhus procedure, the Kugel procedure, TIPP and PHS procedure are all commonly used.27 Laparoscopic inguinal hernia repairs follow the same principles as open extraperitoneal hernia repair and combine the advantages of tension-free mesh repair and minimally invasive surgery.28 TEP procedure uses a giant prosthesic mesh to offer a complete overlap of the Fruchauds myopectineal orifice while avoiding the need for a large transabdominal approach. In spite of these advantages many avoid this technique due to its steep learning curve and the potential risk of major intraoperative complications.3,29 Until now, there has been no meta-analysis directly comparing the outcomes of the present popular two kinds of operation method. In our analysis different kinds of the open extraperitoneal techniques were combined into one group. In our opinion, all the techniques that place mesh in the extraperitoneal space share the same clinical significance. We

believe that there are no clinically relevant differences between them when performed properly. Therefore, further subgroup analysis would only hinder the statistical evaluation.6 Hernia recurrence rate that is one of the most important parameters for evaluating different types of hernia repair techniques. Many authors have reported various factors leading to the recurrence after laparoscopic or open hernioplasty. Such as inadequate dissection, insufficient mesh size, characteristics of hernias, improper fixation, missed hernias, and especially the surgeon’s inexperience.30,31 It is said that hernia recurrence occurred mostly within 2 years for those patients treated with extraperitoneal mesh repair. Recurrence is most commonly caused by technical defects, insufficient prosthesis size and inadequate cover of myopectineal orifice.3 Laparoscopic totally extraperitoneal hernioplasty and open extraperitoneal approach both emphasized the importance of the extraperitoneal reinforcement of the myopectineal orifice. In our present meta-analysis, there was no significant difference in terms of recurrences between open extraperitoneal approach and TEP procedure. The exact incidence of recurrence is unknown because long-term effective and systematic follow-up is difficult to carry out. Just as Lowham AS et al.30 mentioned, we concluded that both of these two extraperitoneal groin hernia repair techniques play the same important role in reducing the hernia recurrence rate compared with the other traditional surgical methods. Generally speaking, early postoperative pain is caused by vascular injury, hematoma compression and the incision itself, while the long-term chronic pain may be due to the intraoperative nerve injury or scar tissue disturbing a nerve, which is relatively frequent in the open hernia repair methods. In the procedure of laparoscopic repair, ilioinguinal or genitofemoral nerves could be circumambulated effectively, avoiding the dissection of the “dangerous triangle of pain”.25 In terms of postoperative chronic pain, however, the results showed that no statistical difference between the two groups.

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

103

Fig. 9 e (A) Meta-analysis for the outcome of time to return to normal activities or work. (B) Meta-analysis for the outcome of time to return to normal activities or work; one trial was excluded after Sensitivity analysis (Gainant A).

Table 3 e Comparison of TEP vs open extraperitoneal hernia repair: summary of meta-analysis. Advantage for TEP

No difference

Advantage for open extraperitoneal hernia repair

Total postoperative Hernia recurrence Less peritoneal tears complications Urinary problems Chronic pain Length of hospital stay Intraoperative complication Return to normal Hematomas activities/work Seromas Wound infection Testicular problems Operative time

The TEP technique has the advantage of lower incidence of postoperative chronic pain syndromes associated with the open extraperitoneal approaches in some sense. So a more gentle surgical operation, a smaller incision, a lesser fixation of the prosthesis, and a lower incidence of inguinal nerve injury would reduce the occurrence of postoperative chronic pain. Complications from hernioplasty can be categorized into intraoperative and postoperative complications.32 Most of the included trials reported the complications of the two procedures. It’s said that the risk of more complications limited the extension of laparoscopic approach, in our meta-analysis, the results showed that there was no significant difference in the incidence of intraoperative complications (injuries to the bowel, large vessels, urinary bladder, or spermatic cords) between them though the incidence of inadvertent peritoneal tears in TEP group far more than that of open approach. Peritoneal tears (PTs) are not uncommon during the course of

104

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

inguinal hernia repairs. In some open repairs, as we all know, the peritoneum was dissected and the distal part of the sac was left undissected, thus creating an intentional peritoneal opening.34 Here we focus on the inadvertent “peritoneal tearing” that is peritoneal injuries during the course of hernia repair surgery in the present article. Although mentioned rarely, PTs are not uncommon during the course of hernia repairs and even considered as an intraoperative complication,35 especially in totally extraperitoneal laparoscopic hernioplasty. PTs may be created at the beginning of the operation due to overinflation of the dissecting balloon, whereas others happen during the dissection of the peritoneum or hernial sac from the surrounding adhesions. The former are usually large tears and necessitate conversion to TAPP or open procedures, while in experienced hands, most of the latter are small, and do not interfere with the operative course.34 Compared with the open repair methods, laparoscopic hernia repair shows a higher incidence of peritoneal tears. Previous studies have reported the incidence of peritoneal injuries, ranging from 0.4 to 67% during the course of TEP,34 the incidence of peritoneal tears was 26.2% in our meta-analysis. Total postoperative complication rates vary from 0 to 17.5% in TEP group and 0e27.5% in open preperitoneal group, and more postoperative complications were found in the open group compared with the TEP group. Urinary problems (ie, urinary retention or infection) were the most common of the postoperative patient-ralated problems. Jensen et al.33 reported that the incidence of postoperative urinary retention was higher when herniorrhaphy was carried out under general anesthesia (3.0%) rather than the local anesthesia (0.4%). However, in our analysis, the incidence of urinary problems when using TEP method was lower than that when using the open extraperitoneal approach though most of the TEP group adopted the general anesthesia. Supporters of laparoscopic groin hernia repair often mention that the quicker return to normal activities or work and the shorter length of hospital stay associated with this technique. The results of our meta-analysis also show significantly shorter hospital stay and much earlier return to work or normal activities in patients undergoing TEP method compared with the open extraperitoneal hernioplasty. However, it is important to note that the duration of hospitalization and postoperative return to normal activities or work affected not only by the patient’s condition but also by many other factors such as the local health-care financing system and the current trend in medical practice.17 Most of the published randomized controlled trials reported longer operative time for TEP method than for open tension-free hernia repair. Surprisingly, in this meta-analysis no difference in operative time was detected between the open extraperitoneal approaches and TEP when we excluded a trial15 after a sensitivity analysis. So the open extraperitoneal hernioplasty didn’t appear to have a significantly shorter operative time compared with TEP repair. Similarly, in our this meta-analysis, no differences in hematomas, seromas, wound infection, or testicular problems were found between the open extraperitoneal and TEP groups. This meta-analysis pooled all currently usable data from published 10 RCTs and 2 clinical comparative studies which we retrieved comparing the open extraperitoneal hernia repair and TEP method, thus providing a relatively reliable

evaluation results. Nevertheless, we are aware that there were also limitations in our study. Though an intensive literature search was conducted, we inevitably have missed important trials that are not in English, which would result in nonpublication bias. The methodological quality of selected studies is another limit of our results. Not all of the included trials were RCTs, moreover, not all of the RCTs provided detailed informations concerning randomization, blinding, allocation concealment and incomplete outcome data, which could cause a high risk of selection and detection bias. The trials analyzed here contained some differences in study structural design and sample size. About the definitions of symptoms to some extent, there were differences between the objective and subjective judgment. One thing to point out was that means and standard deviations were not reported for continuous variables in a very few articles, which was not good for secondary utilization of literature in the future. There was other heterogeneity among the included trials in our meta-analysis. The most obvious cause of heterogeneity would be the different kinds of the open extraperitoneal techniques were included in one group. Secondly, no further classification of hernia types was done in both two groups, such as unilateral or bilateral hernias, primary or recurrent, which would only reduce the sample size making the detection of difference impossible. Thus, future RCTs are still required. Moreover, these studies were conducted in different countries and the operations were all performed by different surgeons. Also, the short duration of follow-up makes it difficult to comment on the long-term complications. In summary, the evidence presented here suggests that the open extraperitoneal inguinal hernia repair is equivalent in most of the analyzed outcomes and has less incidence of peritoneal tears compared with TEP. There are significant benefits of TEP in reducing the length of hospital stay, time to return to normal activities or work, the incidence of total postoperative complications, and reducing the incidence of urinary problems. Further high-quality, long-term follow-up randomized controlled trials are needed to strengthen these findings.

references

1. Lichtenstein IL, Shulman AG, Amid PK, Montllor MM. The tension-free hernioplasty. Am J Surg 1989 Feb;157(2):188e93. 2. Schmedt CG, Da¨ubler P, Leibl BJ, Kraft K, Bittner R. Laparoscopic Hernia Repair Study Team. Simultaneous bilateral laparoscopic inguinal hernia repair: an analysis of 1336 consecutive cases at a single center. Surg Endosc 2002 Feb;16(2):240e4. 3. Sinha R, Sharma N, Dhobal D, Joshi M. Laparoscopic total extraperitoneal repair versus anterior preperitoneal repair for inguinal hernia. Hernia 2006 Apr;10(2):187e91. 4. Suter M, Martinet O, Spertini F. Reduced acute phase response after laparoscopic total extraperitoneal bilateral hernia repair compared to open repair with the Stoppa procedure. Surg Endosc 2002 Aug;16(8):1214e9. 5. Li J, Zhang Y, Hu H, Tang W. Preperitoneal groin hernia repair with Kugel patch through an anterior approach. ANZ J Surg 2008 Oct;78(10):899e902.

t h e s u r g e o n 1 2 ( 2 0 1 4 ) 9 4 e1 0 5

6. Li J, Ji Z, Cheng T. Comparison of open preperitoneal and Lichtenstein repair for inguinal hernia repair: a meta-analysis of randomized controlled trials. Am J Surg 2012 Nov;204(5):769e78. 7. Krishna A, Misra MC, Bansal VK, Kumar S, Rajeshwari S, Chabra A. Laparoscopic inguinal hernia repair: transabdominal preperitoneal (TAPP) versus totally extraperitoneal (TEP) approach: a prospective randomized controlled trial. Surg Endosc 2012 Mar;26(3):639e49. 8. Gong K, Zhang N, Lu Y, Zhu B, Zhang Z, Du D, et al. Comparison of the open tension-free mesh-plug, transabdominal preperitoneal (TAPP), and totally extraperitoneal (TEP) laparoscopic techniques for primary unilateral inguinal hernia repair: a prospective randomized controlled trial. Surg Endosc 2011 Jan;25(1):234e9. 9. Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions, version 5.0.1 Chichester. United Kingdom: The Cochrane Collaboration; 2008. 10. Higgins JP, Altman DG. Assessing risk of bias in included studies. In: Julian P, Higgins SG, editors. Cochrane handbook for systematic reviews of interventions. Chichester, UK: John Wiley and Sons; 2008. pp. 187e241. 11. Chinn S. A simple method for converting an odds ratio to effect size for use in meta-analysis. Stat Med 2000 Nov 30;19(22):3127e31. 12. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol 2005 Apr 20;5:13. 13. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002 Jun 15;21(11):1539e58. 14. Egger M, Smith GD, Altman DG. Systematic reviews in health care: meta-analysis in context. 2nd ed. London: BMJ; 2001. 15. Gu¨nal O, Ozer S, Gu¨rleyik E, Bahc¸ebas‚i T. Does the approach to the groin make a difference in hernia repair? Hernia 2007 Oct;11(5):429e34. 16. Ozmen M, Zulfikaroglu B, Ozalp N, Moran M, Soydinc P, Ziraman I. Femoral vessel blood flow dynamics following totally extraperitoneal vs Stoppa procedure in bilateral inguinal hernias. Am J Surg 2010 Jun;199(6):741e5. 17. Hamza Y, Gabr E, Hammadi H, Khalil R. Four-arm randomized trial comparing laparoscopic and open hernia repairs. Int J Surg 2010;8(1):25e8. 18. Champault GG, Rizk N, Catheline JM, Turner R, Boutelier P. Inguinal hernia repair: totally preperitoneal laparoscopic approach versus Stoppa operation: randomized trial of 100 cases. Surg Laparosc Endosc 1997 Dec;7(6):445e50. 19. Alani A, Duffy F, O’Dwyer PJ. Laparoscopic or open preperitoneal repair in the management of recurrent groin hernias. Hernia 2006 Apr;10(2):156e8. 20. Feliu X, Torres G, Vin˜as X, Martı´nez-Ro´denas F, Ferna´ndezSallent E, Pie J. Preperitoneal repair for recurrent inguinal hernia: laparoscopic and open approach. Hernia 2004 May;8(2):113e6.

105

21. Gainant A, Geballa R, Bouvier S, Cubertafond P, Mathonnet M. Prosthetic treatment of bilateral inguinal hernias via laparoscopic approach or Stoppa procedure. Ann Chir 2000 Jul;125(6):560e5.  lam F, Ozdenkaya Y, Sarı YS. 22. Bender O, Balcı FL, Yu¨ney E, Sag Systemic inflammatory response after Kugel versus laparoscopic groin hernia repair: a prospective randomized trial. Surg Endosc 2009 Dec;23(12):2657e61. 23. Vatansev C, Belviranli M, Aksoy F, Tuncer S, Sahin M, Karahan O. The effects of different hernia repair methods on postoperative pain medication and CRP levels. Surg Laparosc Endosc Percutan Tech 2002 Aug;12(4):243e6. 24. Ozmen MM, Ozalp N, Zulfikaroglu B, Soydinc P, Ziraman I, Hengirmen S. The evaluation of the peak flow velocity and cross-sectional area of the femoral artery and vein following totally extraperitoneal vs preperitoneal open repair of inguinal hernias. Hernia 2004 Dec;8(4):332e5. 25. Schmedt CG, Sauerland S, Bittner R. Comparison of endoscopic procedures vs Lichtenstein and other open mesh techniques for inguinal hernia repair: a meta-analysis of randomized controlled trials. Surg Endosc 2005 Feb;19(2):188e99. 26. Read RC. Milestones in the history of hernia surgery: prosthetic repair. Hernia 2004 Feb;8(1):8e14. 27. Mayagoitia JC, Prieto-Dı´az Cha´vez E, Sua´rez D, Cisneros HA, Tene CE. Predictive factors comparison of complications and recurrences in three tension-free herniorraphy techniques. Hernia 2006 Apr;10(2):147e51. 28. McKernan JB, Laws HL. Laparoscopic repair of inguinal hernias using a totally extraperitoneal prosthetic approach. Surg Endosc 1993 Jan-Feb;7(1):26e8. 29. Putnis S, Berney CR. Totally extraperitoneal repair of inguinal hernia: techniques and pitfalls of a challenging procedure. Langenbecks Arch Surg 2012 Dec;397(8):1343e51. 30. Lowham AS, Filipi CJ, Fitzgibbons Jr RJ, Stoppa R, Wantz GE, Felix EL, et al. Mechanisms of hernia recurrence after preperitoneal mesh repair. Traditional and laparoscopic. Ann Surg 1997 Apr;225(4):422e31. 31. Leibl BJ, Schmedt CG, Kraft K, Ulrich M, Bittner R. Recurrence after endoscopic transperitoneal hernia repair (TAPP): causes, reparative techniques, and results of the reoperation. J Am Coll Surg 2000 Jun;190(6):651e5. 32. Davis CJ, Arregui ME. Laparoscopic repair for groin hernias. Surg Clin North Am 2003 Oct;83(5):1141e61. 33. Jensen P, Mikkelsen T, Kehlet H. Postherniorrhaphy urinary retentiondeffect of local, regional, and general anesthesia: a review. Reg Anesth Pain Med 2002 Nov-Dec;27(6):612e7. 34. Shpitz B, Lansberg L, Bugayev N, Tiomkin V, Klein E. Should peritoneal tears be routinely closed during laparoscopic total extraperitoneal repair of inguinal hernias? A reappraisal. Surg Endosc 2004 Dec;18(12):1771e3. 35. Lau H, Patil NG, Yuen WK, Lee F. Management of peritoneal tear during endoscopic extraperitoneal inguinal hernioplasty. Surg Endosc 2002 Oct;16(10):1474e7.