Does surgeon volume matter in the outcome of endoscopic inguinal hernia repair?

Introduction For open and endoscopic inguinal hernia surgery, it has been demonstrated that low-volume surgeons with fewer than 25 and 30 procedures, respectively, per year are associated with significantly more recurrences than high-volume surgeons with 25 and 30 or more procedures, respectively, per year. This paper now explores the relationship between the caseload and the outcome based on the data from the Herniamed Registry. Patients and methods The prospective data of patients in the Herniamed Registry were analyzed using the inclusion criteria minimum age of 16 years, male patient, primary unilateral inguinal hernia, TEP or TAPP techniques and availability of data on 1-year follow-up. In total, 16,290 patients were enrolled between September 1, 2009, and February 1, 2014. Of the participating surgeons, 466 (87.6 %) had carried out fewer than 25 endoscopic/laparoscopic operations (low-volume surgeons) and 66 (12.4 %) surgeons 25 or more operations (high-volume surgeons) per year. Results Univariable (1.03 vs. 0.73 %; p = 0.047) and multivariable analysis [OR 1.494 (1.065–2.115); p = 0.023] revealed that low-volume surgeons had a significantly higher recurrence rate compared with the high-volume surgeons, although that difference was small. Multivariable analysis also showed that pain on exertion was negatively affected by a lower caseload <25 [OR 1.191 (1.062–1.337); p = 0.003]. While here, too, the difference was small, the fact that in that group there was a greater proportion of patients with small hernia defect sizes may have also played a role since the risk in that group was higher. In this analysis, no evidence was found that pain at rest [OR 1.052 (0.903–1.226); p = 0.516] or chronic pain requiring treatment [OR 1.108 (0.903–1.361); p = 0.326] were influenced by the surgeon volume. Summary As confirmed by previously published studies, the data in the Herniamed Registry also demonstrated that the endoscopic/laparoscopic inguinal hernia surgery caseload impacted the outcome. However, given the overall high-quality level the differences between a “low-volume” surgeon and a “high-volume” surgeon were small. That was due to the use of a standardized technique, structured training as well as continuous supervision of trainees and surgeons with low annual caseload.

unilateral inguinal hernia, TEP or TAPP techniques and availability of data on 1-year follow-up. In total, 16,290 patients were enrolled between September 1, 2009, and February 1, 2014. Of the participating surgeons, 466 (87.6 %) had carried out fewer than 25 endoscopic/laparoscopic operations (low-volume surgeons) and 66 (12.4 %) surgeons 25 or more operations (high-volume surgeons) per year. Results Univariable (1.03 vs. 0.73 %; p = 0.047) and multivariable analysis [OR 1.494 (1.065-2.115); p = 0.023] revealed that low-volume surgeons had a significantly higher recurrence rate compared with the highvolume surgeons, although that difference was small. Multivariable analysis also showed that pain on exertion was negatively affected by a lower caseload \25 [OR 1.191 (1.062-1.337); p = 0.003]. While here, too, the difference was small, the fact that in that group there was a greater proportion of patients with small hernia defect sizes may have also played a role since the risk in that group was higher. In this analysis, no evidence was found that pain at rest [OR 1.052 (0.903-1.226); p = 0.516] or chronic pain requiring treatment [OR 1.108 (0.903-1.361); p = 0.326] were influenced by the surgeon volume. Summary As confirmed by previously published studies, the data in the Herniamed Registry also demonstrated that the endoscopic/laparoscopic inguinal hernia surgery caseload impacted the outcome. However, given the overall high-quality level the differences between a ''low-volume'' surgeon and a ''high-volume'' surgeon were small. That was due to the use of a standardized technique, structured training as well as continuous supervision of trainees and surgeons with low annual caseload.
Keywords Inguinal hernia Á TEP Á TAPP Á Surgeon volume Á Outcome In the Guidelines of the European Hernia Society (EHS), the open Lichtenstein and Plug techniques as well as the endoscopic techniques (TEP, TAPP) are recommended as the best evidence-based options for the repair of a primary unilateral inguinal hernia, providing the surgeon is sufficiently experienced in the specific procedure [1,2]. The Consensus Development Conference of the European Association of Endoscopic Surgery (EAES) and the Guidelines of the International Endohernia Society (IEHS) formulated as a statement that endoscopic groin hernia repair was considered to be more complex than open groin hernia repair [3][4][5]. Therefore, the learning curve for performing endoscopic inguinal hernia repair is longer than for open Lichtenstein repair and ranges between 50 and 100 procedures, with the first 30-50 being the most critical [1]. The Danish Hernia Database demonstrated on the basis of 14,532 endoscopic/ laparoscopic inguinal hernia operations that, in institutions with fewer than 50 endoscopic/laparoscopic inguinal hernia repairs per year, the recurrence rate at 9.97 versus 6.06 % was significantly higher compared with in institutions with more than 50 endoscopic/laparoscopic inguinal hernia operations per year (p \ 0.0001) [6].
In the Swedish Hernia Registry, there was a significantly higher rate of recurrences for surgeons who carried out one-to-five repairs a year compared with surgeons who performed more repairs [7].
Data on open inguinal hernia surgery in the Statewide Planning and Research Cooperative System Database on 151,322 patients with primary inguinal hernia repairs revealed that low-volume surgeons with fewer than 25 procedures per year had significantly more recurrences than high-volume surgeons with 25 or more procedures per year (hazard ratio 1.23; 95 % confidence interval 1.11-1.36; p \ 0.001) [ [9]. Based on data from the Herniamed Registry [10], this paper now explores whether in a hernia registry too, with several surgeons participating in endoscopic/laparoscopic inguinal hernia surgery, a difference was also identified between those surgeons with fewer than 25 procedures per year compared with surgeons with 25 and more procedures.

Materials and methods
The Herniamed quality assurance study is a multicenter, internet-based hernia registry [10] into which 460 participating hospitals and surgeons engaged in private practice (Herniamed Study Group) in Germany, Austria, and Switzerland (Status: March 19, 2015) had entered data prospectively on their patients who had undergone hernia surgery. All postoperative complications occurring up to 30 days after surgery are recorded. On one-year follow-up, postoperative complications are once again reviewed when the general practitioner and patients complete a questionnaire. On one-year follow-up, the general practitioner and patients are also asked about any recurrences, pain at rest, pain on exertion, and chronic pain requiring treatment.
In the present analysis, prospective data on male primary unilateral inguinal hernias, operated on in either the total extraperitoneal patch plasty (TEP) or transabdominal patch plasty (TAPP) technique, were analyzed to identify whether surgery had been performed by a surgeon with fewer than 25 or with 25 or more endoscopic/laparoscopic inguinal hernia operations per year. The registry does not, of course, provide any information on the actual experience of individual surgeons.
Inclusion criteria were minimum age of 16 years, male patient, primary unilateral inguinal hernia, TEP or TAPP techniques, and availability of data on one-year follow-up ( Fig. 1). In total, 16,290 patients were enrolled between September 1, 2009, and February 1, 2014. Of the participating surgeons, 466 (87.6 %) surgeons had carried out fewer than 25 endoscopic/laparoscopic operations (lowvolume surgeons) and 66 (12.4 %) surgeons with 25 or more operations (high-volume surgeons) per year ( Table 1). The low-volume surgeons' group had carried out 9482 (58.2 %), and the high-volume surgeons' group 6808 (41.8 %) of the total number of endoscopic/laparoscopic procedures ( Table 2). The surgeons with fewer than 25 procedures had performed on average 9.47 ± 5.99 operations, and the surgeons with 25 or more procedures 44.12 ± 21.41 operations.
The demographic and surgery-related parameters included age (years), BMI (kg/m 2 ), ASA score (I-IV), proportion of medial, lateral, femoral, and scrotal EHS classification as well as the hernia defect size based on EHS classification (Grade I = \1.5 cm, Grade II = 1.5-3 cm, Grade III = [3 cm) [11]. Where an operation entailed several hernia classifications, the latter were summarized as having a ''combined'' status.
The risk factors included COPD, diabetes, cortisone, immunosuppression, nicotine abuse, coagulopathy or antithrombotic therapy based on antiplatelet or anticoagulant medication. Risk factors were dichotomized, i.e., ''yes'' if at least one risk factor was positive and ''no'' otherwise. The dependent variables were intra-and postoperative complication rates, reoperation rates due to postoperative complications, recurrence rates, and rates of pain at rest, pain on exertion, and chronic pain requiring treatment. All analyses were performed with the software SAS 9.2 (SAS Institute Inc., Cary, NY, USA) and intentionally calculated to a full level of 5 %, i.e., they were not corrected in respect of multiple tests, and each p value B0.05 represents a significant result. To discern differences between the groups in unadjusted analyses, Fisher's exact test was used for categorical outcome variables, and the robust t test (Satterthwaite) for continuous variables.
To rule out any confounding of data caused by different patient characteristics, the results of unadjusted analyses were verified via multivariable analyses in which, in addition to the surgeon volume, other influence parameters were simultaneously reviewed.
Since the main focus of this analysis is on comparison of surgeon's caseloads per year (\25/C25), most of the descriptive statistical analyses in this paper are shown separately for the two groups. All categorical patient data are therefore presented in contingency tables as absolute and relative frequencies for these categories. For continuous data, the mean values and standard deviations are given.
The binary regression model for dichotomous target variables was used to identify the influence of the various factors in multivariable analysis. In addition to the surgeon's caseload per year (\25/C25), other potential influence parameters included: ASA score I, II, III, IV, defect size EHS classification I (\1.5 cm), II (1.5-3 cm), III ([3 cm), age, BMI, risk factors, and EHS classification (lateral, medial, scrotal, femoral). As a result, the odds ratios (OR) and corresponding 95 % confidence intervals based on the Wald test are given for estimates. For influence variables with more than two categories, one of these values was used in each case as a reference category. For the continuous variable age (years), the 10-year odds ratio is given and for BMI (kg/m 2 ) a 5-point odds ratio. The results are sorted on the basis of influence and presented in tabular form.

Comparison of patient collective
With regard to age, patients operated on by surgeons with C25 procedures per year had a significantly higher age and were on average one year older (56.1 ± 15.3 vs. 57.1 ± 15.4 years, p \ 0.001) ( Table 3). As regards the BMI, no difference was identified between the patient collectives of surgeons with \25 and C25 endoscopic/laparoscopic procedures per year (Table 3).
For the unadjusted tests aimed at identifying a relationship between the caseloads per surgeon and year (\25/ C25) and the categorical influence variables, significant differences were noted for almost all influence variables. Low-volume surgeons operated more often on patients with a low ASA score (e.g., ASA I: 35.9 vs. 28.4 %) as well as with smaller defect sizes (EHS I = \1.5 cm: 15.4 vs. 10.6 %) ( Table 4). On the other hand, high-volume surgeons had patients with higher ASA scores (e.g., ASA III/ IV: 16.0 vs. 10.9 %), larger defect sizes (e.g., EHS In terms of the risk factors, global analysis, i.e., occurrence of at least one risk factor, also revealed a significant difference (Table 4). In total, 26.0 % of patients operated on by low-volume surgeons had at least one risk factor, while the proportion of those with at least one risk factor operated on by high-volume surgeons was only 22.3 % (p = 0.001). That effect was mainly attributable to the difference in the nicotine abuse rate (11.8 vs. 7.5 %; p \ 0.001). The proportion of patients with antithrombotic therapy based on antiplatelet and anticoagulant treatment was significantly higher in the patient collectively operated on by the high-volume surgeons (Table 4).

Unadjusted analysis of outcomes by volume
Unadjusted analysis of the relationship between the caseload per surgeon and year did not show any significant difference in the overall intraoperative complication rate between \25 and C25 (p = 0.526, Table 5). However, surgeons with\25 endoscopic/laparoscopic procedures per year caused significantly more organ injuries, especially vascular injuries (p = 0.010, Table 5). As regards the overall postoperative complication rates, low-volume surgeons had, at 2.23 %, a significantly lower rate (p \ 0.001) compared with the high-volume surgeons at 4.95 % ( Table 5). That difference was mainly due to the significantly lower seroma rate in favor of the low-volume surgeons (0.91 vs. 4.20 %; p \ 0.001). That may be due to the high proportion of inguinal hernias with EHS III ([3 cm) defect size and scrotal classification which was investigated in the subsequent multivariable analysis. No significant difference was found in the rate of postoperative complications, leading to reoperation, which was 0.94 % for the low-volume surgeons and 0.72 % for the highvolume surgeons (p = 0.133).
Significant advantages were identified in the recurrence (0.73 vs. 1.03 %; p = 0.047) and in the pain on exertion (7.71 vs. 9.35 %; p \ 0.001) rates in favor of the patient collective operated on by the high-volume surgeons on one-year follow-up (Table 5).

Postoperative complications
The results obtained with the model used to investigate the postoperative complication rate are presented in Table 7 (model matching: p \ 0.001). The risk of postoperative complications was negatively impacted by high-volume surgeons, scrotal hernias, higher age, and larger defects.

Pain on exertion
Analysis of the results obtained on investigating pain on exertion on one-year follow-up is summarized in Table 10 (model matching: p \ 0.001). Pain on exertion was

Chronic pain requiring treatment
The results obtained on investigating chronic pain requiring treatment are presented in

Discussion
The learning curve associated with endoscopic/laparoscopic inguinal hernia surgery requiring 50-100 procedures is longer than that involving the open Lichtenstein operation [1]. Under the supervision of experienced laparoscopic surgeons, young trainees can master the learning curve with good results [12]. Apart from the learning curve, other aspects increasingly discussed in surgery are the impact of the caseload of the treating institution and of the individual surgeon. In the hernia surgery setting, this topic has been addressed so far in three studies on, in each case, open incisional hernia surgery [13], open inguinal hernia surgery [8], and endoscopic inguinal hernia surgery in TEP technique [9]. All three studies identified a significant relationship between the individual surgeon's caseload per year and patient outcome.
In the present paper, the results obtained for perioperative complications and 1-year follow-up of endoscopic/laparoscopic inguinal hernia surgery based on data from the Herniamed Registry were analyzed to ascertain whether the number of operations per surgeon and year (\25 vs. C25) impacted the outcome. Differences were identified first of all on comparing the patient collectives undergoing surgery. The high-volume surgeons (C25 operations per year) operated on significantly more patients with higher ASA score, larger defect size, and scrotal hernia. Likewise, patients operated on by the high-volume surgeons had received significantly more often effective treatment with platelet aggregation inhibitors and coumarin derivatives.
Overall, patients operated on by high-volume surgeons had thus a significantly higher risk profile with, accordingly, significantly more postoperative complications observed in the patients operated on by high-volume surgeons. That this, nonetheless, did not result in more postoperative complications requiring reoperation, but rather in a higher rate of seromas amenable to conservative treatment, attesting to the skill of experienced surgeons in Univariable analysis of the findings on 1-year follow-up revealed that patients operated on by the low-volume surgeons had a significantly higher recurrence rate and pain on exertion rate but here, too, the differences at 1.03 versus 0.73 and 9.35 versus 7.71 %, respectively, were small. Univariable analysis of data for pain at rest and chronic pain requiring treatment did not reveal any differences.
Multivariable analysis revealed that scrotal hernia and large defect size had a significant influence on onset of a postoperative complication. The risk of occurrence of a postoperative complication was less in association with medial or lateral EHS classification, higher BMI value and, interestingly, for surgeons with a caseload of fewer than 25 operations per year. The only explanation that can be given for the latter finding is that surgeons with fewer than 25 procedures per year generally had operated on patients with a lower-risk profile.
Multivariable analysis of the influence variables impacting recurrence showed that higher BMI, medial EHS classification, and a caseload of fewer than 25 procedures per year were associated with a higher risk.
Pain at rest was revealed by multivariable analysis to be negatively affected by a smaller defect size, higher BMI value, and femoral EHS classification. Older patients were found to have a lower risk of onset of pain at rest.
Likewise, multivariable analysis showed that onset of pain on exertion was negatively influenced by smaller defect size, higher BMI value, and additionally by a caseload of fewer than 25 surgical procedures per year. Higher age was also found to be associated with a lower risk of pain on exertion.
Equally, chronic pain requiring treatment was negatively impacted by a smaller hernia defect and higher BMI, with here, too, a lower risk in older patients. The caseload per year did not affect that outcome criterion.
As such, the registry data presented in this paper for endoscopic/laparoscopic inguinal hernia surgery confirm that the annual caseload of the individual surgeons exerted a certain amount of influence on the outcome but the differences were not as pronounced as in the publication by the Mayo Clinic [9]. This is no doubt due to the fact that in the German system even trained surgeons who have less experience of a surgical technique work under the supervision of an experienced surgeon, thus assuring that in such settings, too, good results can be achieved [12]. Based on the experience of the surgeon, also of the trained surgeon, the Chairman of a Department of Surgery decides whether the surgeon can perform the operation alone or under the guidance of a more experienced colleague. The registry does not, of course, provide any information on the actual experience of individual surgeons. It must also be borne in mind that unlike the National Danish and Swedish Registries the data in the Herniamed Registry are collected only from hospitals with a special interest in hernia surgery. Furthermore, the high-volume surgeons were responsible for the more difficult cases, i.e., more advanced hernias. The difference would have probably been much greater if the study had been randomized.
In summary, it can be stated that with regard to the quality parameters recurrence rate and pain on exertion, a ''low-volume surgeon'' achieves slightly worse results than a ''high-volume'' surgeon, but overall can assure a highquality level in endoscopic/laparoendoscopic inguinal hernia surgery. The preconditions for a good outcome, also in routine clinical settings and, in particular, for trainee surgeons or surgeons with lower annual caseloads, are the use of a standardized technique, a structured training program, and close supervision of trainees and of surgeons with lower caseloads.