Low incidence of recurrence and chronic pain after groin hernia repair in adolescents: a systematic review and meta-analysis

Purpose The best operative management of groin hernia in adolescents is uncertain. The aim of this systematic review was to assess recurrence and chronic pain after mesh versus non-mesh repair for groin hernia in adolescents. Methods A systematic search was done in PubMed, EMBASE, and Cochrane CENTRAL in May 2022 for studies reporting postoperative chronic pain (≥6 months) or recurrence after groin hernia repair in adolescents aged 10–17 years. We included randomized controlled trials and observational studies on primary unilateral or bilateral groin hernia repair. Risk of bias was assessed with the Cochrane risk-of-bias tool and Newcastle-Ottawa Scale. Meta-analysis of the incidence of recurrence was conducted. This review is reported according to PRISMA guideline. Results A total of 21 studies including 3,816 adolescents with groin hernias were included comprising two randomized controlled trials, six prospective, and 13 retrospective cohort studies. For non-mesh repairs, the weighted mean incidence proportion of recurrence was 1.6% (95% CI 0.6–2.5) after 2,167 open repairs and 1.9% (95% CI 1.1–2.8) after 1,033 laparoscopic repairs. For mesh repairs, it was 0.6% (95% CI 0.0–1.4) after 406 open repairs while there were no recurrences after 347 laparoscopic repairs (95% CI 0.0–0.6). Across all surgical techniques, the rate of chronic pain after 1,153 repairs ranged from 0 to 11%. Follow-up time varied and was reported in various ways. Conclusion The incidences of recurrence after groin hernia repair in adolescents were low for both open and laparoscopic mesh and non-mesh repairs. Rates of postoperative chronic pain were low. Trial registration PROSPERO: CRD42022130554. Supplementary Information The online version contains supplementary material available at 10.1007/s00423-023-02947-9.


Introduction
Groin hernia repair is a common surgical intervention in both pediatric and adult populations [1].The best surgical management of groin hernias in adolescents is an area of uncertainty as these patients fall between the pediatric and adult populations.Among adolescents, there is a great physical variation since some patients will be fully grown while others will not, raising the question when an adolescent should be surgically treated as either a child or an adult.Mesh is routinely used in hernia repair in adults, but the use of mesh might be problematic in children due to concerns about placing a foreign body in growing tissue and concerns of living 70+ years with a foreign body.
At present, the surgical approach for adolescents is determined by the personal preferences of the surgeon in collaboration with the patient and parents.Operative methods include open or laparoscopic repair and use of mesh or sutures only.However, the preferred methods can depend on whether the surgeon is mostly handling pediatric or adult patients [2].In an international survey among pediatric surgeons, 83% preferred an open approach while 17% preferred a laparoscopic approach [3].Mesh is preferred in adults as it lowers recurrence rates [4,5], but a common postoperative complication is chronic pain occurring in rates as high as 8-15%, depending on definition and method of assessment [6][7][8][9][10].Young adult males are at greater risk of chronic pain than older males following inguinal hernia repair, although the role of the mesh is uncertain [11].There are indications from a recent retrospective study that mesh in

Population
The participants of the included studies were males or females aged 10-17 years.

Intervention
The intervention of interest was primary unilateral or bilateral groin hernia repair.Both open and laparoscopic repairs were included.

Outcomes
The primary outcomes were hernia recurrence and postoperative chronic pain.Recurrence had to be determined by physical examination or diagnostic imaging, hence, studies with only self-reported recurrence were excluded.If a study did not explicitly outline the method for assessing recurrence, but the study design and study characteristics indicated that recurrence was reliably assessed, the study was included.Reoperation for recurrence was used as a proxy for recurrence.There was no minimum follow-up time on recurrence for inclusion in this review.Postoperative chronic pain was defined as pain ≥6 months after surgery, hence, studies reporting pain had to have a minimum follow-up of six months [15].At least one of the primary outcomes (recurrence or chronic pain) had to be reported for a study to be included.The secondary outcome was postoperative complications.The definition of age was of specific importance as this review aimed to include adolescents.According to the World Health Organization, the definition of adolescents is 10-19-year-olds [16].The hernia literature generally defines adults from 18 years and above.Therefore, we defined adolescents as 10-17-year-olds.

Study design
Study types included in this review were randomized controlled trials (RCT) and observational studies with ≥5 participants.Studies reported in English, Danish, Swedish, or Norwegian were included.

Information sources
The systematic search was conducted in PubMed, EMBASE, and Cochrane CENTRAL.Authors were contacted if reporting of data was unclear.Specifically, authors were contacted if data for the age group of 10-17-year-olds were not presented.The search strategy was developed in cooperation with a professional research librarian.The search was last conducted on 9 May 2022 in all three databases.

Search strategy
The search was developed for PubMed and afterwards adapted to EMBASE and Cochrane CENTRAL.The systematic search was divided into three blocks: hernia AND age AND (pain OR recurrence).In EMBASE, the limitation "exclude medline journals" was used with confidence that all MEDLINE-registered articles were found in PubMed.The search strategy can be accessed in the PROSPERO protocol (registration number: CRD42022130554).The full search strategy in PubMed was: (

Data collection
Data extraction was done by one author and entered into predefined tables in an Excel sheet (Microsoft Excel for Mac, version 15.32), and all data were validated at least twice by the same author.Doubts regarding data extraction were discussed within the author group.Extracted variables included study design, year, outcomes, number and age of participants, hernia characteristics, type of repair, pain, recurrence, follow-up, intra-and postoperative complications, analgesia, and surgeon experience.

Study risk of bias assessment
The risk of bias in RCTs was assessed with a revised Cochrane risk-of-bias tool for randomized trials (RoB 2) [19].The risk of bias in observational studies was assessed with the Newcastle-Ottawa Scale [20].The higher the number of stars a study gets assigned with the Newcastle-Ottawa Scale, the lower the risk of bias.In general, a study can get a maximum of nine stars, indicating low risk of bias.For observational studies, the risk of bias was assessed for the two primary outcomes (recurrence and chronic pain) separately (Table 1 and Table S1).If a study only included either an exposed (mesh) or non-exposed (non-mesh) cohort, the study could be awarded a maximum of eight stars.The risk of bias assessment was done independently by two authors.Discrepancies in the assessment were discussed within the author group until a consensus was reached.

Data analysis
For meta-analysis, data on recurrence were analyzed with OpenMeta[Analyst] [21].Forest plots and weighted mean incidence proportions of recurrence were produced.When meta-analysis was not feasible, data were presented descriptively.A binary random-effects model (DerSimonian and Laird [22]) with 95% confidence intervals was applied.When the lower limit of the confidence interval was negative in meta-analysis, it was reported as zero.A correction factor of 0.5 was used for outcomes with zero events.Sensitivity analysis was conducted on follow-up time, sample size omitting studies with less than 100 participants, and by omitting repairs that included femoral hernia.For quantifying heterogeneity across studies, the I 2 statistic was applied.For chronic pain, meta-analysis was not feasible due to methodological heterogeneity between studies and results were therefore presented descriptively.If a study only reported the number of adolescents with hernia repairs, but did not report the specific number of repairs (e.g. if patients had bilateral repairs), the number of adolescents with hernia repairs would be used as a surrogate measure for the number of hernia repairs knowing that this would likely underestimate the total number of repairs.

Certainty assessment
The Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) framework [23] and the GRADEpro software [24] were used to assess the certainty of evidence for the primary outcome of recurrence.To guide the assessment, the GRADE Handbook [23] and a guide on the application of GRADE for prognostic studies were used [25].

Results
Twenty-one studies were included in this review .See Fig. 1 for flow diagram depicting the screening process.We contacted the corresponding authors of 334 studies to obtain data relevant for this review when data were not reported in the individual studies.The authors of two studies [35,45] supplied unpublished data that enabled inclusion in this review.

Recurrence
All 21 included studies  reported recurrence rates.Recurrence was assessed in various ways: clinical assessment [26,29,35,37,[43][44][45], clinical assessment and additional methods (phone interview, chart review, or imaging) [31-33, 39, 40, 42], questionnaires [30,36], questionnaires and chart review [41,46], chart review only [28], ultrasound [27], and registry data [38].In one study [34], where the method of assessment of recurrence was not reported, the study design was an RCT, and the authors reported that patients were followed for three years giving sufficient confidence in the assessment method to include the study in this review.However, the corresponding author did not respond to our enquiry.In two studies [30,36] with self-reported recurrence, all patients with recurrence were either reoperated or the recurrence was later verified by a doctor.Followup time varied and was reported in various ways (Table 1).
Three studies [34,44,45] used synthetic mesh, two studies [31,37] used biologic mesh, and one study [38] did not report the type of mesh used.
For non-mesh repairs, two studies [36,37] reported rates of chronic pain for open repairs of 7% (15/210) and 0% (0/50), respectively, and one study [32] reported a rate of 0% (0/255) for laparoscopic repairs.For mesh repairs, one study [34] reported a rate of chronic pain for open repairs of 2.9% (5/170), and two studies [37,45] both reported a rate of 0% (0/50 and 0/8, respectively).One study [34] on laparoscopic mesh repair reported a rate of chronic pain of 1.5% (5/336).Two studies [33,34] compared open with laparoscopic repairs.One study [34] did not find a significant difference in chronic pain when comparing the two groups, and the other study [33] did not report data on chronic pain separately for the open and laparoscopic repair groups for the adolescent subgroup.Follow-up time varied and was reported in various ways (Fig. 1).Tools for the assessment of pain varied.One study [32] used the Visual Analog Scale (VAS) and stated that there was no occurrence of chronic pain, but they did not define the term chronic pain or report when pain was measured.
One study [37] measured postoperative pain with the VAS two hours postoperatively, but also stated that no patients had chronic pain without defining the term or tool used for later pain assessment.One study [34] used the Numeric Rating Scale reporting chronic pain in 2% of adolescents, also not stating a definition of chronic pain.One study [36] reporting chronic pain in 7% of adolescents used a telephone questionnaire with yes or no questions including a question on residual pain from the operation.In another questionnaire study [33] using yes or no questions that were either e-mailed or phoned, 11% of adolescents answered yes to the question: do you have any pain associated with your hernia repair?And in one study [45] reporting no occurrence of chronic pain, an interview was conducted at follow-up, but the study did not state the specific method or questions used for the evaluation of chronic pain.Due to heterogeneity of the included studies, meta-analysis on chronic pain was not performed.

Postoperative complications
Eight studies [26, 28-30, 32-34, 37] on a total of 1,363 adolescent patients with 1,432 hernia repairs reported postoperative complications other than pain or recurrence for adolescents.Across all studies, the definition and reporting of postoperative complications varied.

Discussion
This systematic review demonstrates that most data in the literature on recurrence after groin hernia repair in adolescents are on open non-mesh repair.Overall, incidences of recurrence were low across all surgical approaches including open and laparoscopic mesh and non-mesh repairs.The rate of chronic pain was also low across all surgical techniques, though results should be interpreted with caution due to data scarcity and study heterogeneity.Data on postoperative complications other than recurrence and chronic pain were sparsely reported but seem to be low.
Most of the data in the literature on recurrence after groin hernia repair concerns inguinal hernia.In children under the age of 12 years, the recurrence rate was 1.4% (137/9,993) after non-mesh repair (96% were open repairs) [48].In adults, a Cochrane review found a recurrence rate of 1.8% (52/2,834) in the mesh repair group, compared with 4% (110/2,741) in the non-mesh repair group [4].This systematic review found lower recurrence rates after both open and laparoscopic mesh and non-mesh groin hernia repair among adolescents compared with the results for adults in the Cochrane review [4].There may be many reasons for this, including differences in follow-up between studies and likely better postoperative tissue healing in children compared with adults.
As with recurrence, most of the literature on postoperative chronic pain after groin hernia repair also concerns inguinal hernia.Overall, the rate of chronic pain is lower in children and adolescents compared with the rate of up to 8-15% reported in adults [6][7][8][9][10].The prevalence of chronic pain was 5.1% after a follow-up time of 6-48 months in children aged six months to 12 years at the time of open non-mesh repair [49].The prevalence of chronic pain was 13.5% after an observation period of 14-18 years in patients who had undergone open nonmesh repair before the age of 5 years [50].Despite data scarcity, this systematic review indicates that rates of postoperative chronic pain in adolescents are similar compared to younger children and substantially lower than in adult hernia repair regardless of operative technique.
This systematic review has several strengths.The review was reported according to PRISMA guideline [13], and a protocol was uploaded on PROSPERO [14] prior to completion of screening and prior to initiation of data extraction, securing transparency.The search was developed with a professional research librarian and performed in several databases supplemented by a snowball search.The search strategy was broad to confidently search all available literature for a population whose age-range can be difficult to isolate in literature searches.All authors with available contact information were contacted if the reporting of data was unclear or to retrieve unpublished data if relevant, which was often due to studies not reporting data on the specific age group targeted by this review.The assessment of bias was conducted by two independent researchers and doubts were discussed within the author group.To ensure that postoperative pain was reported correctly as chronic pain, a minimum follow-up time of 6 months after the repair was set as a criterion for inclusion in this review.Some limitations of this systematic review must also be mentioned.We were unable to retrieve 17 of the 777 articles for full-text screening, hence, some articles fulfilling our inclusion criteria might have been missed.There was a possibility of language bias as only studies in English, Danish, Swedish, and Norwegian were included.The Newcastle-Ottawa Scale was used for assessing the risk of bias in observational studies, though the latest version of the Cochrane Handbook recommends using the Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool [51].Underestimation of recurrence rates is probable as we accepted studies using reoperation for recurrence as a proxy for recurrence even though it has been shown to underestimate recurrence rates by up to 40% [52].Furthermore, recurrence was evaluated at various time points and in some studies using uncertain assessment methods.Also, several studies measured recurrence using a combination of initial clinical assessment and later questionnaires with self-reported recurrence.The reporting of pain was done with varying pain assessment tools and at various time points.Lastly, follow-up times were reported in various ways across all studies and in several studies not specifically for the adolescent population inhibiting comparison of studies on follow-up time.

Table 1
Study characteristics presented twice in the table to present data for the mesh and non-mesh groups separately, b : as reported for the whole adolescent cohort including both mesh and non-mesh repairs, c a: study is