Posterior mesh inguinal hernia repairs: a propensity score matched analysis of laparoscopic and robotic versus open approaches

Purpose International guidelines suggest the use of lapro-endoscopic technique for primary unilateral inguinal hernia (IHR) because of lower postoperative pain and reduction in chronic pain. It is unclear if the primary benefit is due to the minimally invasive approach, the posterior mesh position or both. Further research evaluating posterior mesh placement using open preperitoneal techniques is recommended. A potential benefit of open preperitoneal repair is the avoidance of general anesthesia, as these repairs can be performed under local anesthesia. This study compares clinical and patient-reported outcomes after unilateral laparo-endoscopic, robotic, and open posterior mesh IHRs. Methods We performed a propensity score matched analysis of patients undergoing IHR between 2012 and 2021 in the Abdominal Core Health Quality Collaborative registry. 10,409 patients underwent a unilateral IHR via a posterior approach. Hernia repairs were performed via minimally invasive surgery (MIS) which includes laparoscopic and robotic transabdominal preperitoneal (TAPP), laparoscopic totally extraperitoneal (TEP), or open transrectus preperitoneal/open preperitoneal (TREPP/OPP) approaches. Propensity score matching (PSM) utilizing nearest neighbor matching accounted for differences in baseline characteristics and possible confounding variables between groups. We matched 816 patients in the MIS cohort with 816 patients in the TREPP/OPP group. Outcomes included patient reported quality of life, hernia recurrence, and postoperative opioid use. Results Improvement was seen after TREPP/OPP as compared to MIS IHR in EuraHS at 30 days (Median(IQR) 7.0 (2.0–16.64) vs 10 (2.0–24.0); OR 0.69 [0.55–0.85]; p = 0.001) and 6 months (1.0 (0.0–4.0) vs 2.0 (0.0–4.0); OR 0.63 [0.46–85]; p = 0.002), patient-reported opioid use at 30-day follow-up (18% vs 45% OR 0.26 [0.19–0.35]; p < 0.001), and rates of surgical site occurrences (0.8% vs 4.9% OR 0.16 [0.06–0.35]; p < 0.001). There were no differences in EuraHS scores and recurrences at 1 year. Conclusions This study demonstrates a potential benefit of open posterior mesh placement over MIS repair in short-term quality of life and seroma formation with equivalent rates of hernia recurrence. Further study is needed to better understand these differences and determine the reproducibility of these findings outside of high-volume specialty centers. Supplementary Information The online version contains supplementary material available at 10.1007/s10029-022-02680-0.


Introduction
Inguinal hernia repair (IHR) is performed in over 20 million individuals annually worldwide [1]. Considerable debate remains over the optimal surgical approach for unilateral inguinal hernia repair. International guidelines suggest that in skilled hands, posterior laparo-endoscopic mesh repair may lead to reduced early postoperative pain and reduced incidence of chronic groin pain [2][3][4], while maintaining equivalent rates of recurrence [3,5]. Posterior mesh repair can be accomplished using a minimally invasive (laparoscopic and/or robotic) or an open approach [6][7][8]. Minimally invasive (MIS) IHR is further divided into transabdominal preperitoneal (TAPP) and totally extraperitoneal (TEP) approaches. Several open posterior approaches exist that also allow preperitoneal mesh placement, including the Transrectus Preperitoneal repair (TREPP), Transinguinal Preperitoneal repair (TRIPP), Open Preperitoneal repair (OPP), and the Kugel repair. However, there is currently insufficient comparative data to make any conclusive recommendations. As such, further evaluation of open techniques for posterior mesh placement with comparative analysis to currently accepted minimally invasive repair is warranted.
The disadvantage of open posterior repairs is the potential for violation of both the anterior and posterior planes. However, the Transrectus Preperitoneal (TREPP), Open Preperitoneal (OPP), and the Kugel repair all involve incisions above the traditional open anterior inguinal hernia repair, thus sparing the anterior plane while still allowing for posterior mesh placement. One key differentiator between open posterior and MIS IHR is choice of anesthetic. To perform an MIS IHR, general anesthesia with muscle relaxation is typically required to allow for pneumoperitoneum and placement of the mesh. By contrast, open posterior mesh repairs can often be performed utilizing local anesthesia with intravenous sedation, which improves short-term outcomes [4,9]. Studies evaluating TREPP/OPP have thus far been limited, but some single-institution observational studies have shown promising outcomes [10][11][12][13]. Comparisons of TREPP to MIS TEP and open Lichtenstein approaches demonstrate no significant differences in recurrence rates, postoperative complications, or postoperative pain [5,14], and a recent meta-analysis on TREPP demonstrated low risk of recurrence, chronic pain, hematoma, and wound infection [15]. While evaluation of the quality of hernia repair has traditionally involved comparison of recurrence rates, patients themselves also measure the success of their surgery by their quality of life in recovery. Patient reported outcomes such as post-operative pain, ability to return to work and exercise, and aesthetic outcomes play a significant role in determining the quality of repair [16]. The purpose of this study is to compare postoperative clinical and patientreported outcomes in individuals who underwent unilateral MIS IHR compared to open posterior mesh IHR. We hypothesized that open preperitoneal repairs have similar outcomes to MIS repairs and may be a good option for patients who may benefit from avoidance of general anesthesia.

Data collection
The Abdominal Core Health Quality Collaborative (ACHQC) is a national registry that collects short-and long-term hernia-specific data, including patient reported outcomes related to hernia repairs with the goal of improving surgical quality and value [17].

Outcomes
Data collected include patient demographics and comorbidity, surgical details, clinical outcomes, and patient reported outcomes (PRO) before, during, and after unilateral IHR procedures as described previously [50]. The primary outcome is patient reported quality of life using the EuraHS scores at 30-day, 6-month, and 1-year after surgery. The EuraHS is a validated quality of life measurement tool for inguinal hernia. The tool assesses pain (range 0-30), restriction of activity (range 0-40), and cosmetic discomfort (range 0-20) due to the hernia or from surgery [51] with total scores ranging from 0 to 90, A lower score signifies an improved QOL. Secondary outcomes include composite hernia recurrence, perioperative complications, surgical site occurrence or infection, and opioid use. Clinical or radiographic recurrence is recorded by the clinician at any point after surgery. Patient surveys are filled out in the office or sent to ACHQC patients at 30 days, 6 months and then once per year after surgery. Patient reported outcome recurrence are recorded at or after the 1-year mark. Composite recurrence is defined by the Hernia Recurrence Inventory which includes physical exam or radiographic imaging at any point post-operatively or a patient reported bulge at the site of the hernia at a 1 year or greater after hernia repair.

Statistical methods
Patient-level, hernia, and operative characteristics were compared between individuals who received TREPP and MIS IHR. Pearson's chi-squared and Wilcoxon rank sum tests were used to conduct bivariate tests comparing categorical and continuous covariates, respectively. Time-to-recurrence was examined using Kaplan-Meier recurrence free estimation and log-rank test to compare recurrence curves between operative approaches. An advantage of evaluating recurrence as time-to-event is the ability to use all information available to compute 1-year recurrence probability, including endpoint and censoring information, and to account for varying length of follow up. Because there is much more statistical information, survival analyses tend to have greater statistical power to detect effects over methods for binary outcomes [52,53]. Although the time it might take after an IHR for the hernia to recur is not possible to predict definitively in the clinical setting, our Kaplan-Meier analysis is statistically precise and its estimate of recurrence-free probability is an unbiased representation of the true time-to-event data as reflected by the relatively stable width of the curve's confidence intervals up to 3 years. This method to characterize long-term recurrence has been established in statistical literature and in hernia recurrence reporting [54,55]. Additional pairwise analysis was performed to detect differences in TREPP and each MIS IHR technique. To address selection bias and systematic differences in baseline covariates, we created a propensity-score matched cohort. A logistic regression model was used to estimate the propensity score for operative approach conditional on covariates identified a priori. Covariates included in the propensity score model were age, gender, race, BMI, insurance status, ASA class, comorbidities, indication for surgery (enlarging hernia, painful bulge, recurrent hernia), prior pelvic operation, prior mesh, hernia size, scrotal component, history of substance use, history of opioid use, behavioral health history, and EuraHS quality of life score measured at baseline. A 1:1 nearest-neighbor matching algorithm with a caliper of 0.2 was used to match TREPP with MIS IHRs [56,57]. Balance was assessed by examining the standardized mean differences (SMD) of baseline covariates where a SMD < 0.1 was considered good balance. This resulted in 816 TREPP and 816 MIS IHR patients for comparison. Odds and hazard ratios and their 95% confidence intervals were estimated using logistic, proportional odds, or Cox proportional hazards models for binary, patient-reported, and timeto-event outcomes, respectively. To assess the difference in EuraHS quality of life scores between surgical approaches for populations with the same baseline score, we adjust for baseline scores in a proportional odds regression model.
To address missingness in baseline data, we performed 30 iterations of multiple imputation using a predictive mean matching algorithm to preserve baseline information. Results were pooled using the across approach [58]. Planned Fig. 1 Representation of the myopectineal orifice as seen by a surgeon over time through a 4 cm left lower abdominal incision. A TREPP is performed through a 4-5 cm lower abdominal incision. Using a headlight, the surgeon can visualize the entire myopectineal orifice in its entirety, by having assistants retract the abdominal wall in different directions propensity score matched subgroup analyses were performed to compare TREPP to each type of MIS approach. Lastly, to address possible bias due to informative missingness in the outcomes, we performed a tipping point analysis as a sensitivity analysis.   (Table 3). Additionally, domainspecific subanalysis performed post-hoc revealed lower pain domain scores after TREPP repair at 30 days, but not at 180 days, in pairwise comparisons between each matched TREPP cohort to corresponding laparoscopic TAPP, TEP, and robotic TAPP cohorts (Supplemental Tables 1, 2, 3).

Recurrence rates
At 1 year, 2/296 composite recurrences were reported in the TREPP group and 4/174 composite recurrences reported in the MIS group. The Kaplan-Meier time-to-event log-rank test did not reveal a statistically significant difference in hernia recurrence risk between the TREPP and MIS repair cohorts (Fig. 3). Subgroup analysis using Kaplan-Meier recurrence free estimation and log-rank test as previously described showed minimal differences in the composite recurrence risk between the TREPP, TAPP, TEP and rTAPP groups (Supplemental Fig. 1).  There were no statistically significant differences in 30-day SSOs or surgical site infections requiring procedural intervention, rates of postoperative bleeding, peripheral nerve injury, postoperative respiratory failure, pulmonary embolism, ileus, bowel obstruction, DVT, or UTI between the TREPP and MIS repair cohorts (Table 3). No statistical differences were seen between TREPP and MIS IHR in urinary retention; however, reduced rates of urinary retention were appreciable in TREPP versus TAPP (TREPP 0.5% (3/561); TAPP 2.0% (11/558) P = 0.11) and TREPP versus rTAPP (TREPP 0.5% (3/649); rTAPP 1.9% (12/632) P = 0.17).

Discussion
In this analysis comparing TREPP and MIS approaches to posterior unilateral inguinal hernia repair, TREPP was associated with decreased postoperative pain domain score, lower opioid use and better overall quality of life compared to MIS approaches at 1 month and 6 months after surgery. No statistically significant differences between the repairs were noted in QOL or recurrence at 1 year after surgery. These results suggest that an open posterior mesh IHR utilizing local anesthesia and sedation may have short-term benefits over the more common MIS repairs.
A key difference between the TREPP and MIS repairs is the ability to perform TREPP with local anesthesia and sedation while avoiding general anesthetic and muscle relaxation. Reduced postoperative pain, early mobilization after surgery, and reduced cost associated with local anesthetic use in IHR has been demonstrated in prior studies [4,10,[59][60][61][62], and our study confirms this is true for open posterior mesh repairs as well. Additionally, general anesthetic is independently associated with higher healthcare cost compared to local anesthetic [60].
The improvement in the PRO pain domain and differential opioid consumption seen in this study is likely multifactorial. Peritoneal entry, incision, and closure used for TAPP and robotic TAPP repairs may account for the increased pain, particularly as PRO in the pain domain was less pronounced between TREPP and TEP repairs. Additionally, suture fixation was performed in 96% of the open repairs while tacks were used in 42% of MIS approaches. There are mixed data on the relationship between mesh fixation and postoperative pain, thus obfuscating any clear conclusions on the impact of this difference on our observed outcomes. We also found lower rates of postoperative seroma in the TREPP group. While most seromas are self-limited with minimal clinical significance, the impact on quality of life is unclear and may be a contributing factor to lower QOL scores in the MIS group. Several meta-analyses report that seroma formation rates are higher following MIS compared to open repairs [63][64][65][66]. Studies have shown reduction in seroma formation after MIS repair when transversalis fascia is inverted to close the dead space [37,38]. This technical detail is not captured in the ACHQC and we cannot determine its direct impact on reported outcomes. However, the high single-surgeon volume of TREPP in which this maneuver is routinely performed suggests some correlation with transversalis fascia fixation and decreasing seroma rate. It is also important to point out that, though statistically significant differences in PRO were found, the clinical significance of this finding is not defined. Decreased postoperative opioid use after TREPP can at least partly be attributed to administration of local anesthesia. However, selection bias is a significant factor here due to practice-specific prescribing patterns and focus on opioid reduction. Better patient education on anticipatory management of postprocedural pain and decreased surgeon prescribing can significantly impact opioid consumption, but was not standard across patients studied [67,68]. Still, the differences in opioid use between TREPP and MIS approaches are significant and warrant further investigation. Differential prescribing practices, patient education, surgeon experience, and patient expectations all impact opioid use. Opioid reduction initiatives within the ACHQC for the last two years have had variable success, which may lead to some of the differences seen in our study. A recent analysis of ACHQC data indicates that opioid prescribing is an independent risk factor for opioid consumption, thus implicating individual surgeon practice and management of patient expectations as a potential confounder [69]. Combined with the discrepancy in the number of participating surgeons contributing data to TREPP vs MIS, this introduces additional bias. Nonetheless, the observation of lower opioid use in TREPP suggests that significantly fewer opioids can likely be prescribed in all posterior inguinal hernia repairs and therefore reduce surgeons' impact on the opioid epidemic in the United States.
Reported incidence of urinary retention after IHR ranges from 1 to 22% [70,71]. The use of general anesthesia is an independent risk factor for retention [72,73]. Patients undergoing TREPP were much more likely to avoid general anesthesia which may explain why patients undergoing TREPP had reduced rates of urinary retention over MIS repairs. Though these differences did not reach statistical significance in our analysis, the corresponding p-values and confidence intervals approached the alpha threshold of 0.05, with three-to fourfold increase consistently evident in the rate of retention following a laparoscopic or robotic repair. Given the large number of patients undergoing IHR annually, this is arguably an important observation resulting in reduced quality of life after surgery and prolonged hospital stay for those patients who develop urinary retention [60,[70][71][72][73]. Additional data collection and analysis will be required to further investigate this trend.
No statistically significant differences were seen in recurrence rates after IHR between approaches. This is not surprising given that the end result of the repairs is identical, utilizing similar size mesh for complete posterior coverage of the myopectineal orifice. However, our relatively short duration of follow-up limits any true comparison in terms of hernia recurrence. Lack of long-term follow-up is a known limitation of the ACHQC. Greater than 1-year follow-up in this study is seen in less than 25%. To account for this, we used a Kaplan-Meier disease-free model to predict recurrence based on the hernia inventory scores obtained throughout the study period. Efforts are ongoing to improve long-term follow-up in the ACHQC, for instance, by implementing an interface with Center for Medicare Services (CMS) data.
There are important limitations to this study that deserve discussion. First, a minimal clinically important difference (MCID) has not been established for the EuraHS, thus any clinical significance of our findings has yet to be established. The TREPP repairs were performed by a significantly smaller subset of surgeons compared to MIS repair, which likely introduces significant selection bias into our results. This bias is introduced by the large number of TREPP cases being performed by a small subset of high-volume surgeons. Admittedly, heterogeneity in surgeon training, learning curve, and case volume potentially influence these outcomes. In our analysis, one surgeon (first author) performed the majority of the TREPP repairs but trained an MIS surgeon (second author) on the TREPP approach during the study period. In training, we identified two learning curves to this operation and feel that they are hurdles that can be overcome with a methodical training approach. The first learning curve is to become competent at the operation, which we estimate at 25 cases. The second learning curve is to become an expert teacher of the operation, and we estimate that would require 100-200 cases with deliberate practice [74]. Since both MIS and TREPP repairs are offered in our practice, we feel that in patients who have access to a skilled surgeon, TREPP is a good alternative to MIS and can allow patients to avoid general anesthesia.
Finally, to address missing data, tipping point sensitivity analysis was performed. This suggested that a substantial degree of differential missingness, where unobserved outcome rates/means would need to be more than 75% higher/ lower than observed rates/means, would have to be present in the data to alter the conclusions of this analysis. As such, it is unlikely that missing outcome data are dramatically different from observed data.

Conclusion
This study demonstrates a potential benefit of open posterior mesh placement over MIS repair in short-term quality of life and seroma formation with equivalent rates of hernia recurrence. Further study is needed to better understand these differences and determine the reproducibility of these findings outside of high-volume specialty centers.
Author contributions All authors had full access to all the data in the study. Alexandra Urquiza, in addition to co-authoring the manuscript, is the medical illustrator for Figs. 1 and 3. We want to thank Katherine Melvin, PA-C, Wanda Ebitibitua, Elizabeth Constantine, and Haley Sykes for contributing to data collection for this study. The ACHQC has contracted with Weill Cornell Medicine to provide biostatistical support for ACHQC projects. The work provided for this manuscript was performed under the umbrella of the Weill Cornell Medicine and ACHQC collaboration plan.

Declarations
Conflict of interest Michael Reinhorn, MD, MBA, has received consulting fees from Heron Therapeutics. Nora Fullington, MD, has no conflicts or disclosures. Divyansh Agarwal, MD, PhD, has no conflicts or disclosures. Molly A. Olson, MS, has no conflicts or disclosures. Lauren Ott PA-C has no conflicts or disclosures. Anna Canavan PA-C has no conflicts or disclosures. Bailey Pate has no conflicts or disclosures. Melissa Hubertus has no conflicts or disclosures. Alexandra Urquiza has no conflicts or disclosures. Benjamin Poulose, MD, MPH, has received research support from BD International and Advanced Medical Support; Consulting-Ethicon; he receives salary from the Abdominal Core Health Quality Collaborative (ACHQC) as the Director of Quality and Outcomes. Jeremy Warren, MD, has received fees from Intuitive Surgical as clinical proctor and speaker.
Ethical approval This article does not contain studies directly with human or animal participants performed by any of the authors.
Human and animal rights This manuscript does not directly involve human participants, as it is a review of data already collected in a herniadatabase.

Informed consent For this type of study formal consent is not required.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.