Penetrating versus non‐penetrating mesh fixation in laparoscopic groin hernia repair­

Objectives

This is a protocol for a Cochrane Review (intervention). The objectives are as follows:

To assess the benefits and harms of penetrating versus non‐penetrating mesh fixation in adults receiving laparoscopic groin hernia repair.

Background

Description of the condition

It is estimated that surgeons worldwide repair more than 20 million groin hernias each year, making it one of the most common surgical procedures [1]. Groin hernia is mainly a clinical diagnosis based on the patient's history and the clinical examination [2]. A groin hernia is a condition where a bulge appears in the groin area due to protrusion of a hernia sac (peritoneum) through a defect in the abdominal wall [3]. The hernia sac contains preperitoneal fat, intestines, and/or other abdominal organs [2]. Groin hernias can often be manually reduced back into the abdominal cavity [2]; otherwise, they are named irreducible hernias [4]. Individuals with an irreducible hernia who also experience pain have signs of a compromised blood supply, indicating a strangulated hernia that requires acute surgery [4].

Groin hernias comprise femoral and inguinal hernias [5], and inguinal hernias can further be subdivided into lateral/indirect and medial/direct hernias [5, 6]. Around 96% of repaired groin hernias are inguinal hernias and 4% are femoral hernias [7], and groin hernia repairs peak when patients are around 75 to 80 years old [8]. Inguinal hernia repairs are also most frequent in both sexes, constituting 99% of groin hernia repairs in men and 78% in women [9]. However, since 27% of men and 3% of women undergo inguinal hernia repair at some point in their lives [10], most inguinal hernia repairs are performed on men [8, 9]. Regarding femoral hernias, slightly more repairs are performed in women than in men [8, 9], but 22% of groin hernia repairs in women are for a femoral hernia, while this is only the case in 1% of men [9].

Individuals with groin hernias have symptoms ranging from minor cosmetic concerns to severe pain. It is often difficult to determine preoperatively if a groin hernia is femoral or inguinal, but the distinction is clear during a laparoscopic repair [2]. Compared with inguinal hernias, femoral hernias have a higher risk of strangulation, which is associated with higher morbidity and increased mortality [2, 11]. Since femoral hernias in women are fairly common, women with groin hernias are recommended to undergo surgery regardless of their symptoms [2]. In contrast, not all adult men need surgery. Groin hernia repair can induce chronic postoperative pain [2], and men with no or mild symptoms can avoid surgery due to the low risk of acute repair [2, 12]. Ultimately, around 70% of these men will require surgery within 10 years due to the development of pain [12]. However, men with a symptomatic groin hernia are recommended to undergo surgery due to the unknown risk of strangulation if surgery is postponed [2].

Description of the intervention and how it might work

Surgery is the only definitive treatment for groin hernias, and the use of mesh to reinforce the groin is recommended in adults [2, 13]. The implementation of mesh has reduced the risk of recurrence compared with non‐mesh repairs [4] without increasing the risk of chronic pain [4, 14]. The reoperation rate for recurrence after mesh repair is around 5% five years postoperatively [15], and the frequency of chronic pain has traditionally been described as 10% to 12% [2]. However, pain is highly dependent on the definition and assessment method used, and an updated overview is warranted [16]. This Cochrane review will assess different mesh fixation methods during laparoscopic groin hernia repair. The review will therefore mainly include adults, since mesh repairs are not used in young children and are usually avoided in teenagers due to their developing groin area [17, 18].

Laparoscopic groin hernia repairs performed by surgeons with sufficient expertise have benefits over open repairs regarding acute and chronic pain and a faster recovery [2, 4]. Therefore, laparoscopic repairs are recommended for both men and women with primary or bilateral hernias [2, 4]. The most common laparoscopic groin hernia repairs are the transabdominal preperitoneal (TAPP) repair and the total extraperitoneal (TEP) repair [2]. In both repair techniques, surgeons place a mesh in the preperitoneal groin area. During TAPP repair, surgeons access the intraperitoneal space and open the peritoneum to place the mesh, whereas in TEP repairs, the groin is accessed without entering the intraperitoneal space, and the mesh can be placed without opening the peritoneum. TAPP and TEP repairs have comparable intraoperative, short‐, and long‐term outcomes [2, 19, 20].

There are largely three ways to secure the mesh during laparoscopic repairs: not fixating the mesh at all, using a penetrating fixation technique, or using a non‐penetrating fixation technique [2, 13]. In penetrating mesh fixation, the mesh is secured by tissue penetrating devices such as tacks, staples, or sutures, which can be made of absorbable or permanent material [2]. In non‐penetrating mesh fixation, the mesh is secured using topical glue or a self‐adhesive mesh with integrated adhesives or grips [2]. The international groin hernia guideline from 2018 states that the evidence is very uncertain, but mesh fixation is recommended for large medial hernias [2]. For other hernia types, the mesh can likely be placed in the groin without fixation for most groin hernias in both TAPP and TEP repairs without affecting the risk of recurrence or chronic pain [2, 21]. Even so, surgeons often fixate the mesh hoping to avoid recurrences, and mesh fixation is especially common during TAPP repairs [21, 22]. There are some theoretical advantages and disadvantages to the different fixation methods. Penetrating mesh fixation could reduce the risk of recurrence due to a theoretically more solid fixation than non‐penetrating methods. Furthermore, surgeons might choose tacks over glue to reduce the operative time. However, non‐penetrating mesh fixation could result in a lower risk of pain compared with penetrating mesh fixation due to a less traumatic fixation, possibly avoiding nerve damage. Other advantages of non‐penetrating fixation could be a reduced risk of serious vascular intraoperative events and a lower risk of haematoma.

Why it is important to do this review

Fixating the mesh or simply leaving it in place without any fixation likely results in little to no difference in the risk of recurrence [2, 23, 24] and chronic pain during TEP repairs [2, 24], while there is sparse evidence for TAPP repairs [2, 22, 23, 24]; surgeons could therefore likely avoid mesh fixation during many laparoscopic groin hernia repairs. Nevertheless, many surgeons do fixate the mesh during laparoscopic repairs [22], and a well‐performed systematic review will help surgeons choose the best mesh fixation method. To our knowledge, previous systematic reviews have not compared all penetrating versus all non‐penetrating mesh fixation techniques, but instead compared glue versus penetrating mesh fixation, as well as penetrating fixation versus a combination of non‐fixation and non‐penetrating fixation [25, 26, 27, 28, 29, 30, 31]. A Cochrane review is therefore warranted to synthesise the evidence and compare mesh fixation with tacks, staples, and sutures (penetrating) versus fixation with glue and self‐adhesive meshes (non‐penetrating) in TEP and TAPP repairs. This Cochrane review will aid surgeons and patients in their decision of which mesh fixation method to use when choosing to fixate the mesh during laparoscopic groin hernia repair.

Objectives

To assess the benefits and harms of penetrating versus non‐penetrating mesh fixation in adults receiving laparoscopic groin hernia repair.

Methods

This protocol has been developed following the guidelines in the Cochrane Handbook for Systematic Reviews of Interventions [32], and reported according to PRISMA 2020 [33].

Criteria for considering studies for this review

Types of studies

We will include only randomised controlled trials (RCTs), irrespective of publication status, year of dissemination, and language. We will exclude all other study types, for the reasons described below.

• Quasi‐randomised studies: these are non‐randomised studies [34] due to non‐random allocation, such as date of birth or weekday of admission [35].

• Cross‐over RCTs: the interventions are surgical procedures, and patients cannot cross over to another intervention.

• Cluster‐RCTs: results from a cluster‐RCT could reflect different practices and surgeon skills at different hospitals rather than a true effect of the interventions.

• All other study types with original data: lack of randomisation may introduce bias, such as confounding by indication in observational studies [36].

Types of participants

We will include studies that report on men and women ≥ 18 years old who have undergone elective laparoscopic groin hernia repair, either inguinal or femoral hernia repair. If studies include teenagers, then ≥ 90% must be ≥ 18 years old. If studies include acute repairs, then ≥ 95% must be elective repairs. Apart from the criteria for age and acute/elective repairs, studies that contain a subset of relevant participants will only be included if data for the eligible participants can be extracted from the study or provided by the study authors.

Individuals with an inguinal hernia recurrence have a higher risk of developing both further recurrences [37, 15] and chronic pain compared with primary hernias [38, 39]. Nevertheless, studies with recurrent groin hernias will also be included to increase the external validity. Furthermore, we expect that recurrences will be equally distributed between the intervention groups due to the randomisation. We will explore the inclusion of participants with recurrences in subgroup analyses (see Investigation of heterogeneity and subgroup analysis).

Types of interventions

We will include studies that have randomised on penetrating versus non‐penetrating mesh fixation techniques for elective laparoscopic groin hernia repair, either using absorbable or permanent material. We will compare the following intervention groups.

• Penetrating mesh fixation:

  • Tacks (permanent or absorbable), sutures (permanent or absorbable), and staples (only available in a permanent material).

  • A network meta‐analysis has shown comparable recurrence and chronic pain rates between different penetrating mesh fixation materials [40]. Therefore, different types of penetrating fixation and materials will be pooled.

• Non‐penetrating mesh fixation:

  • Topical glue (including fibrin sealants) and self‐adhesive mesh with integrated adhesives or grips.

  • RCTs have shown comparable short‐ and long‐term outcomes for glue fixation and self‐adhesive meshes [41,42,43]. Therefore, the two non‐penetrating fixation methods will be pooled.

We will exclude studies if they only compare:

• penetrating versus penetrating mesh fixation;

• non‐penetrating versus non‐penetrating mesh fixation;

• mesh fixation (penetrating or non‐penetrating) versus no fixation of the mesh.

We will only consider laparoscopic repairs, including both TAPP and TEP repairs performed either as standard laparoscopic repair or robot‐assisted surgery. We will group TAPP and TEP repairs due to comparable short‐ and long‐term outcomes [2, 19, 20]. We will also include variations of TAPP and TEP repairs, such as the enhanced‐view TEP (e‐TEP) repair [44]. We will include permanent and partly absorbable meshes, whereas fully absorbable meshes, including biological meshes, and plugs will be excluded. Any peritoneal closure technique is acceptable.

Outcome measures

Studies must report one of the critical or important outcomes to be included in the review, but they will be included regardless of whether data are reported in a usable way or not. Studies that fulfil the eligibility criteria without reporting any critical or important outcomes will be categorised as awaiting classification. There is no published standard core outcome set after groin hernia surgery in adults [45, 46]; the outcomes were chosen after consumer involvement (see Consumer involvement). Critical outcomes will be long‐term outcomes, while important outcomes will mainly be intraoperative and short‐term outcomes.

Critical outcomes

Chronic pain

• Definition: pain ≥ 6 months postoperatively [47, 48]. Pain will be treated as a dichotomous outcome or a continuous outcome, or both, depending on the data.

• Measurement tool: pain must have been measured by a visual analogue scale (VAS), a numerical rating scale (NRS), or a pain‐related patient‐reported outcome measure (PROM) such as the Activity Assessment Scale [49], the Carolinas Comfort Scale [50], or the Inguinal Pain Questionnaire [51]. We will not include studies in this analysis if participants only verbally stated yes/no, or if the study omitted information about how pain was measured. If possible, we will prioritise PROMs if a study used various measurement tools. However, a hierarchy of outcome measures will ultimately depend on the available data, and we will prioritise measurement tools used by the majority of studies.

• Time point: pain assessed ≥ 6 months postoperatively. Observational questionnaire studies suggest that the prevalence of chronic pain tends to decrease up to three to four years after groin hernia repair [52, 53]. We will therefore opt to use the latest time point if studies have several pain assessments, but this also depends on the follow‐up rate and the comparability of time points between studies, i.e. we will ultimately prioritise the time points reported by most studies. We will choose only one time point for each study but may conduct post hoc analyses by subdividing the results into different follow‐up intervals (see Investigation of heterogeneity and subgroup analysis).

Recurrence

• Definition: recurrence or reoperation for a recurrence, treated as a dichotomous outcome.

• Measurement tool: we will accept any definition of recurrence by the studies.

• Time point: studies must report a minimum follow‐up of 12 months since few recurrences develop within the first postoperative year [15]. We will therefore opt to use the latest time point, but this also depends on the follow‐up rate and the comparability of time points between studies, i.e. we will ultimately prioritise the time points reported by most studies. We will choose only one time point for each study but may conduct post hoc analyses by subdividing the results into different follow‐up intervals (see Investigation of heterogeneity and subgroup analysis).

Important outcomes

Acute pain

• Defined as pain ≤ 30 days postoperatively [54] with the same considerations regarding measurement tools and time point as for chronic pain, treated as a dichotomous or a continuous outcome, or both, depending on the data.

Haematoma

• The definition used by the included studies and occurring ≤ 30 days postoperatively, treated as a dichotomous outcome.

Surgical site infection

• The definition used by the included studies, covering wound or mesh‐related infection, or both, and treated as a dichotomous outcome.

• Given that symptoms of mesh infection may develop years after groin hernia repair [55], there will be no restriction on the time point.

Serious vascular intraoperative events

• The definition used by the included studies, treated as a dichotomous outcome. We will describe details in the review if possible.

Operative time

• The definition used by the included studies, treated as a continuous outcome.

Convalescence

• The definition and time points used by the included studies, treated as a dichotomous or a continuous outcome, or both. This will be a post hoc decision and will be described as a difference between the protocol and the review.

Outcomes assessed by PROMs other than pain and convalescence

• The definition and time points used by the included studies, treated as a dichotomous or a continuous outcome, or both. This will be a post hoc decision and will be described as a difference between the protocol and the review.

Mortality

• We will consider only deaths related to groin hernia repairs, treated as a dichotomous outcome.

Search methods for identification of studies

The authors and a research librarian developed the search strategy for the electronic bibliographic database Embase Ovid. In the review, the search strategy for Embase Ovid will be adapted to the Cochrane Central Register of Controlled Trials (CENTRAL) and MEDLINE Ovid. There will be no restrictions on the searches in either the electronic bibliographic databases or other sources.

Electronic searches

Electronic bibliographic databases

The comprehensive search consists of "groin hernia AND laparoscopic repair AND RCT".

We will use Cochrane's highly sensitive filters for identifying RCTs in MEDLINE Ovid and Embase Ovid [35].

We will search three databases:

• CENTRAL in the Cochrane Library (search date);

• MEDLINE Ovid (1950 to search date);

• Embase Ovid (1974 to search date). The search strategy is available in Supplementary material 1.

Trial registers

We will search the following trial registers:

• ClinicalTrials.gov (clinicaltrials.gov/);

• World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) (www.who.int/clinical‐trials‐registry‐platform).

Search strategies for trial registers are available in Supplementary material 2.

Searching other resources

We will perform backward citation searching by studying the reference lists of the included studies. We will perform forward citation searching for the included studies using Google Scholar (scholar.google.com/). Additionally, we will study the reference lists of narrative and systematic reviews on the same topic, along with relevant guidelines. We will also identify potentially eligible conference abstracts retrieved from the electronic bibliographic databases.

Data collection and analysis

Selection of studies

We will import all records from electronic bibliographic databases to Covidence [56]. After removing duplicates, we will conduct a title and abstract screening as well as a full‐text screening. All records will be screened independently by two of the review authors against the eligibility criteria, and disagreements will be discussed within the author team until consensus is reached. We will contact study authors by email twice if data are described too vaguely in the full text to permit an eligibility decision. We will manage screening of trials identified in the trial registers in Microsoft Excel [57], using the same screening procedure as for records identified in the electronic bibliographic databases.

We will search for the full‐text study when identifying potentially eligible trials and conference abstracts. We will email study authors twice to enquire after unpublished studies. We will exclude trials or abstracts if full‐text studies cannot be retrieved.

To detect changes and comments to the included RCTs post‐publication, we will search for the included studies in the Retraction Watch Database (retractiondatabase.org). We will also search the electronic bibliographic databases for comments, letters, and errata related to the included studies, and include any relevant information in the review.

The unit of interest in the review will be studies and not reports. If there are several reports of the same study, we will consider only one report as the primary report, which will be justified. We will include relevant information from secondary reports.

The review authors understand English, Swedish, Danish, and Norwegian fluently. Other languages will be translated during the screening process using Chat Generative Pre‐trained Transformer (ChatGPT) (chat.openai.com).

Data extraction and management

We will extract data on study setting, participant characteristics, intervention characteristics, outcome characteristics, funding source and role of funder, authors' conflicts of interest, and errata or retraction statements. We will create a pilot data extraction form in Covidence [56], which we will test on five studies and discuss within the author group. Thereafter, two review authors will independently extract data into Covidence. We will email study authors twice for clarification or missing data as needed. Any discrepancies in data will be discussed within the author group until consensus is reached. We will transfer data from Covidence to RevMan [58], and all data in the final review will be cross‐checked manually with each included study.

We will extract the following outcome data for each intervention group.

• Dichotomous data: numbers, either directly from the included studies or calculated using sample size and percentages.

• Continuous data: mean and standard deviation (SD) or median and range/interquartile range. If studies are considered appropriate for inclusion in a meta‐analysis but lack an SD, we will calculate this information using formulas described in Chapter 6 of the Cochrane Handbook [59]. We will roughly evaluate the distribution of mean and SD if the lowest and highest values are known [60]. If only median and range/interquartile range are presented, we will impute mean and SD according to the methods in the Cochrane Handbook [59].

• Only effect estimates available: if studies considered appropriate for inclusion in a meta‐analysis only present effect estimates with confidence intervals (CIs) or P values for an outcome, we will calculate standard errors (SEs) based on the CIs or P values according to the Cochrane Handbook [59]. Thereafter, we will use RevMan to calculate the same effect measure and CIs for all studies reporting this outcome, and these CIs will be transformed into SEs [59].

Risk of bias assessment in included studies

We will perform risk of bias assessment on studies that report data in a usable way for the outcomes presented in the summary of findings table, as follows:

• Chronic pain (dichotomous outcome): ≥ 6 months postoperatively

• Recurrence: ≥ 12 months postoperatively

• Acute pain: ≤ 30 days postoperatively

• Haematoma: ≤ 30 days postoperatively

• Surgical site infection: no restriction on time point

• Serious vascular intraoperative events: intraoperatively

• Operative time: intraoperatively

We will use the RoB 2 tool, which includes the following five domains [61, 62]:

• bias arising from the randomisation process;

• bias due to deviations from intended interventions;

• bias due to missing outcome data;

• bias in measurement of the outcome; and

• bias in selection of the reported result.

Regarding the domain 'bias due to deviations from intended interventions', we will quantify the intention‐to‐treat effect (i.e. "quantifying the effect of assignment to the interventions at baseline, regardless of whether the interventions were received as intended") [61].

The bias assessment will begin with a pilot assessment of five studies independently conducted by two review authors in the RoB 2 Excel tool [63]. After comparing and discussing any disagreements, two review authors will independently assess the risk of bias for all studies. Any disagreements will be discussed within the author group. We will email study authors twice for clarification or missing data as needed.

Signalling questions will be answered for each domain, including a support of the judgement [61]. The RoB 2 tool will propose a risk of bias judgement for each domain as low risk of bias, some concerns, or high risk of bias. The proposed judgements of some concerns and high risk of bias will be verified by the first review author by considering if there are issues that are likely to affect the possibility of drawing trustworthy conclusions. Based on the verification, the proposed judgements may be changed after discussion in the author group alongside a note and a support of the new judgement [61]. Each study will receive an overall risk of bias for each assessed outcome [61].

• Low risk of bias: low risk of bias in all domains.

• Some concerns: one or a few domains with some concerns, and no domain with a high risk of bias.

• High risk of bias: high risk of bias in any domain, or some concerns in many domains.

The risk of bias assessment will be incorporated into the analyses by [64]:

• providing a general interpretation of the risk of bias across studies for each domain and based on the overall judgement;

• presenting overall risk of bias judgements in forest plots, ordered by low to high risk of bias;

• considering risk of bias in sensitivity analyses (see Sensitivity analysis).

We will present conflicts of interest and funding information in the 'Characteristics of included studies' table along with a judgement and rationale of 'notable concern' or 'no notable concern', which also covers insufficient information to assess if there is a concern or not [64]. Studies with notable concerns will be handled in sensitivity analyses (see Sensitivity analysis).

Measures of treatment effect

We will use RevMan to analyse data [58]. We will present dichotomous outcomes as risk ratio (RR) with 95% CI. If necessary, 0.5 will be added to the cells with zero events [59], but studies with no events in both arms will be excluded from the meta‐analyses [60]. We will present continuous outcomes as mean difference (MD) with 95% CI. If different scales are used to measure the same continuous outcome, we will present a standardised mean difference (SMD) with 95% CI. We will interpret the size of the effects in relation to the clinical relevance [65].

Unit of analysis issues

The unit of analysis will be participants and not groins, that is if a participant had bilateral repair and had one complication in only one groin or had the same complication in both groins, this will only be considered as one complication in one participant. If a study only reports outcomes related to groins, we will attempt to obtain data related to participants. If this is not possible, we will treat these groins as participants and examine the effect of this in sensitivity analyses.

In the case of studies with more than two intervention groups, we will include only the interventions relevant to this review.

Dealing with missing data

If summary data for an outcome are missing, we will email the study authors twice. If individual data are missing due to loss of follow‐up for the critical outcomes of chronic pain and recurrence, we will perform sensitivity analyses by implementing data for participants using 'worst‐case' and 'best‐case' analyses [60].

Reporting bias assessment

The author group will visually assess results that have been synthesised in meta‐analyses including minimum five RCTs for publication bias by evaluating the symmetry in funnel plots [66, 67]. If there is asymmetry, the possibility of publication bias will be considered and discussed within the author group along with other explanations outlined in Table 13.3.b in Chapter 13 of the Cochrane Handbook [66]. If publication bias is a possibility, we will conduct sensitivity analyses (see Sensitivity analysis). We will use RevMan [58] to create funnel plots, with the X‐axis representing effect estimates and the Y‐axis representing the SE of the logarithmic effect estimate [66].

Synthesis methods

We plan to combine results in meta‐analyses if a minimum of two RCTs are sufficiently clinically and methodologically homogenous [60]. Furthermore, we will also present outcomes assessed in only one RCT in a forest plot. Regarding the clinical aspects, we will evaluate participant and intervention similarity. Regarding the methodological aspects, we will evaluate the definition and measurement of outcomes. We will include studies regardless of risk of bias, which we will assess in sensitivity analyses. Meta‐analyses will be conducted in RevMan [58] and presented in forest plots with effect estimates as described in Measures of treatment effect. We will apply the random‐effects model using DerSimonian and Laird given that we assume that the included RCTs will estimate similar, but not equal, intervention effects. We will use the Mantel‐Haenszel method for categorical data, the inverse‐variance method for continuous data, and the generic inverse‐variance method employing a random‐effects model if only the effect estimates and SEs are available.

We will assess statistical heterogeneity by visually inspecting the overlap of CIs and by interpreting the I² statistic, which we will interpret as follows [60]:

• 0% to 40%: might not be important;

• 30% to 60%: may represent moderate heterogeneity;

• 50% to 90%: may represent substantial heterogeneity;

• 75% to 100%: considerable heterogeneity.

If there is no plausible explanation for statistical heterogeneity (see Investigation of heterogeneity and subgroup analysis), we will only present forest plots with I² < 75%, since 75% to 100% represents considerable heterogeneity [60].

If meta‐analysis is not possible, we will present a range and distribution of observed effects, as well as a narrative summary of the effect estimates [68].

Investigation of heterogeneity and subgroup analysis

Subgroup analyses may be conducted for one of two reasons [60]: to assess outcomes in different subgroups of studies; or to explore the reason for statistical heterogeneity.

We will conduct the subgroup analyses regardless of the statistical heterogeneity. The aim is to assess results in different subgroups of studies and to simultaneously explore reasons for any observed statistical heterogeneity. We plan the following subgroup analyses.

Chronic pain and recurrence

• Studies that include recurrent groin hernias versus studies that only include primary hernias. Individuals with a recurrence have a higher risk of chronic pain [38, 39] and a higher risk of more recurrences [15, 37]. Theoretically, individuals with a higher risk of chronic pain might benefit from non‐penetrating fixation, and those with a higher risk of more recurrences might benefit from penetrating mesh fixation.

• Studies where both intervention groups received a permanent mesh versus studies where both groups received a partly absorbable mesh. The rationale is to assess any effect of mesh properties and fixation interaction, which has been observed in a database study [69].

• Studies where both intervention groups received a lightweight mesh versus studies where both groups received a heavyweight mesh. A systematic review with network meta‐analysis has demonstrated a lower recurrence rate in participants who underwent laparoscopic inguinal hernia repair using heavyweight mesh compared with lightweight mesh [70]. The rationale for comparing this subgroup is to assess any effect of mesh properties and fixation interaction, which has been observed in a database study [69].

• Studies with different follow‐up times. Both chronic pain and recurrence might be subdivided into different follow‐up intervals. For chronic pain, these will be 6 to 12 months and > 12 months. For recurrence, these will be 12 to 24 months and > 24 months.

Operative time

• Surgeons might choose penetrating tacks or staples because they believe that it is faster than glue or self‐adhesive mesh. The subgroup analyses performed will depend on the data, but they could be tacks and staples versus self‐adhesive mesh as well as tacks and staples versus glue fixation.

We will explore the reason for statistical heterogeneity by testing subgroup differences across subgroups using a random‐effects model [60]. A P value of ≤ 0.05 will indicate an effect modifier that can explain the observed statistical heterogeneity.

Equity‐related assessment

As outlined in the Background, groin hernia is a very common condition [1] that mostly affects men [10]. There is no evidence to suggest that health inequities influence either the development of groin hernia or the effect of the interventions. Therefore, we will not investigate health inequities in this review.

Sensitivity analysis

We plan sensitivity analyses, according to Chapter 10 of the Cochrane Handbook [60]. We will perform sensitivity analyses to assess the robustness of results from meta‐analyses, which we will present in a summary table.

It is likely that other issues will be identified during the review process [60], but the planned sensitivity analyses are as follows.

• Assessing the impact of bias (all outcomes with meta‐analyses)

  • Removing studies with an overall high risk of bias

• Unit of analysis issue (chronic pain and recurrence)

  • Studies reporting outcomes related to groins instead of participants will be excluded.

• Loss to follow‐up (chronic pain and recurrence)

  • Implementing data for participants using 'worst‐case' and 'best‐case' analyses [60]

• Conflicts of interest and funding (chronic pain and recurrence)

  • Removing studies with notable concerns

• Outliers (chronic pain and recurrence)

  • Removing studies only if there is an obvious explanation for the outlying result

• When there is a possible risk of publication bias

  • Compare random‐effects estimates and fixed‐effect estimates. If these are similar, we will assume little or no small‐study effects on the effect estimate. If random‐effects and fixed‐effect estimates are different, with random‐effects estimates closer to the results of the smaller studies, then analyses will be restricted to the larger studies if they are more rigorously conducted [66].

Certainty of the evidence assessment

Summary of findings table

We will create a summary of findings table according to Chapter 14 of the Cochrane Handbook [71], using GRADEpro GDT software [72]. We will present the following outcomes.

• Chronic pain (dichotomous outcome): ≥ 6 months postoperatively

• Recurrence: ≥ 12 months postoperatively

• Acute pain: ≤ 30 days postoperatively

• Haematoma: ≤ 30 days postoperatively

• Surgical site infection: no restriction on time point

• Serious vascular intraoperative events: intraoperatively

• Operative time: intraoperatively

The summary of findings table will summarise the magnitude of relative and absolute effects, the amount of available evidence, and the certainty of the evidence.

GRADE

We will perform a GRADE assessment according to Chapter 14 of the Cochrane Handbook [71] as well as the GRADE Handbook [67], using GRADEpro GDT [72].

We plan assessments for all outcomes synthesised in meta‐analyses. Two review authors will independently assess the certainty of the evidence based on the five GRADE considerations (risk of bias [62], inconsistency, indirectness, imprecision, and publication bias). RCTs start at high certainty of evidence, with downgrading taking place only after consensus, possibly involving all review authors. We will assess the certainty of the evidence as high, moderate, low, or very low.

The interpretation and communication of the effect estimates from meta‐analyses and the certainty of evidence will be guided by Table 15.6.b in Chapter 15 of the Cochrane Handbook [73].

Consumer involvement

Patients of different age, sex, and groin hernia status (repaired or unrepaired primary or recurrent hernia) were consulted for input on the critical and important outcomes. They were asked about which outcomes are important to them. Thereafter, they were presented with predefined outcomes and were asked to grade the importance of these and provide feedback. Chronic pain was by far the most important outcome for all patients, followed by recurrence, and these two were chosen as the critical outcomes. Operative time was not considered important for patients since laparoscopic repairs are performed under general anaesthesia. However, we have chosen operative time as an important outcome since this parameter is of interest to surgeons and administrators and can influence the choice of mesh fixation technique if other outcomes are comparable. The remaining six important outcomes were chosen after considering patient input.

In the review, laypersons with and without a history of groin hernia will assess the Plain language summary for readability.

Supporting Information

Supplementary materials are available with the online version of this article: 10.1002/14651858.CD016122.

Supplementary materials are published alongside the article and contain additional data and information that support or enhance the article. Supplementary materials may not be subject to the same editorial scrutiny as the content of the article and Cochrane has not copyedited, typeset or proofread these materials. The material in these sections has been supplied by the author(s) for publication under a Licence for Publication and the author(s) are solely responsible for the material. Cochrane accordingly gives no representations or warranties of any kind in relation to, and accepts no liability for any reliance on or use of, such material.

Supplementary material 1 Search strategies

Supplementary material 2 Search strategy for trial registers

New

Additional information

Contributions of authors

Stina Öberg, Jason Joe Baker, and Jacob Rosenberg contributed substantially to the conception and design of the protocol. Stina Öberg drafted the work, and Jason Joe Baker and Jacob Rosenberg revised the draft critically. All authors have approved the final version of the protocol.

Declarations of interest

Stina Öberg, Jason Joe Baker, and Jacob Rosenberg report no conflicts of interest relevant to this review.

Stina Öberg and Jason Joe Baker are Managing Editor and Assistant Managing Editor for the Cochrane Colorectal Group, and Jacob Rosenberg is the Co‐ordinating Editor. However, the authors had no influence on the editorial process of this protocol.

Jacob Rosenberg is an author of an eligible randomised controlled trial [74]. He will not be involved in the risk of bias assessment for this study or any GRADE assessments that involve data from it.

Sources of support

Internal sources

• None, Other

  None

External sources

• None, Other

  None

Registration and protocol

Cochrane approved the proposal for this review in February 2024.

Data, code and other materials

Data sharing is not applicable to this article as it is a protocol, so no datasets were generated or analysed.

Notes

Published notes in RevMan are for editor use only. Authors should leave this section blank.