International Journal of Surgery 105 (2022) 106837

Contents lists available at ScienceDirect

International Journal of Surgery

journal homepage: www.elsevier.com/locate/ijsu

Prospective Cohort Study

Significant factors influencing chronic postoperative inguinal pain: A

conditional time-dependent observational cohort study
L.M. van den Dop a, *, F.P.J. den Hartog a, D. Sneiders a, G. Kleinrensink a, J.F. Lange a,
J.F. Gillion b, Hernia-Club Members
a
Department of Surgery, Erasmus University Medical Center, Rotterdam, the Netherlands
b
Department of Surgery, Ramsay Santé-Antony Private Hospital, Antony, France

A R T I C L E I N F O A B S T R A C T

Keywords: Background: Inguinal hernia (IH) repair is a common surgical procedure. Focus has shifted from recurrences to
Inguinal hernia chronic postoperative inguinal pain (CPIP). To assess the natural course of CPIP and identify patient factors
Herniorrhaphy influencing the onset of CPIP, an observational registry-based study was performed.
CPIP
Materials and methods: Data prospectively collected from the Club-Hernie national database was retrieved from
Surgical technique
Quality of life
2011 until 2021. Patients who underwent elective surgery for inguinal hernia were divided in an irrelevant pain
Registry group and relevant pain group. Relevant pain at one year and two years were compared with patients with
irrelevant pain at all-time points (preoperatively, one month, one year and two years). Quality of life questions
were compared between relevant pain at one year and two years.
Results: 4.016 patients were included in the analysis. Mean age was 65.1 years, 90.3% of patients was male.
Factors correlated with CPIP onset were age, gender, ASA, recurrent surgery, surgical technique, nerve handling
and fixation type. Relevant pain at one month was a greater risk for CPIP than preoperative pain (12.3% vs
3.6%). In the majority of patients (83.2%) CPIP was ameliorated at two years. Hernia related complaints differed
significantly between CPIP at one year and two years.
Conclusion: Postoperative pain after one month was a greater risk factor for CPIP development than preoperative
pain. CPIP at one year seems to have a different pain etiology than CPIP at two years. Patient and surgical factors
influence the onset of CPIP at one year, however the natural course of these complaints shows great decline at
two years, largely without reinterventions.

1. Introduction shrinkage by fibrosis, followed by folding or balling up of the mesh) and

non-absorbable sutures for mesh fixation [4]. Although the nerve
Inguinal hernia (IH) repair is one of the most ubiquitous procedures handling considerations apply less for the laparoendoscopic approach,
in daily surgical practice. In The Netherlands alone, approximately CPIP is also reported in this technique [5].
20.000 inguinal hernia repairs are performed annually [1]. With the use Intra-operative nerve transection for the prevention of CPIP has been
of prosthetic meshes, focus has shifted from the incidence of recurrence thoroughly investigated by many studies, and also in a meta-analysis by
to the development of chronic postoperative inguinal pain (CPIP). Cirocchi and colleagues [6–8]. Although decrease in pain in the early
Chronic pain has been identified by the International Association for the postoperative period is observed, this difference is not observed
Study of PAIN (IASP) as pain lasting longer than three months [2]. Due anymore one year postoperatively. Conversely, more groin paresthesia
to differences in definitions of CPIP, the incidences range from 1 to 60% is found when comparing neurectomy with nerve preservation after six
[3]. It is estimated that CPIP affects the daily life of 5–10% of patients months.
[1]. The type of fixation used has also been suggested to interfere with the
In open IH repair, reasons for developing CPIP have been attributable onset of CPIP [9]. Many types of fixation can be used to hold the mesh in
to nerve injury, nerve entrapment in the mesh, meshoma formation (i.e. place (e.g. resorbable and non-resorbable sutures/staples, glue,

* Corresponding author. Erasmus University Medical Center, Department of Surgery, Room Ee-173, Dr. Molewaterplein 40, 3015 GD Rotterdam, the Netherlands,
PO BOX 2040, 3000, CA Rotterdam, the Netherlands.
E-mail address: [email protected] (L.M. van den Dop).

https://doi.org/10.1016/j.ijsu.2022.106837
Received 21 February 2022; Received in revised form 28 July 2022; Accepted 11 August 2022
Available online 18 August 2022
1743-9191/© 2022 The Author(s). Published by Elsevier Ltd on behalf of IJS Publishing Group Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
L.M. van den Dop et al. International Journal of Surgery 105 (2022) 106837

self-adhesive). While glue in the form of fibrin or cyanoacrylate has been thoroughly. There is some information that CPIP dissipates over time,
proven to reduce short-term postoperative pain, doubt still exists but no information is given about the patient and surgical factors that
whether this will be significant for the onset of inguinal pain in the long may come into play [14]. Surgeons can inform patients about the
run [9,10]. chances of developing CPIP, which is about ten percent depending on
While many studies focus on the factors that influence the onset of open or endoscopic techniques, but information about the long-term
CPIP [4,11–13], the natural course of CPIP has not been investigated course of CPIP and chances of CPIP persisting for over one year with

Fig. 1. Flowchart representing the patient inclusion and time-dependent conditional probability of relevant pain one month postoperatively, one year post­
operatively and two years postoperatively.

2
L.M. van den Dop et al. International Journal of Surgery 105 (2022) 106837

respect to their patient and surgical characteristics is lacking. This would handling (i.e. not seen, preservation or resection) of the iliohypogastric
be valuable information for patients experiencing CPIP. nerve (IHN) and the ilioinguinal nerve (IIN).
In order to provide an outlook on the course of CPIP and quantitate To measure preoperative and postoperative pain the Verbal Rating
quality of life deterioration, the aim of this study is to assess the con­ Scale (VRS) was used. This scale is divided in four scores (i.e. no pain,
ditional probability of continuous pain and discomfort, depending on mild pain, moderate pain, severe pain). For present comparison, these
the earlier status of pain and discomfort, and identifying patient and/or groups are clustered in irrelevant pain (no pain, mild pain) and relevant
surgical factors contributing to pain and/or discomfort with a complete pain (moderate and severe pain).
follow-up of two years. Patients who reported relevant pain at one year and two years were
compared to patients that reported no pain at one month, one year and
2. Methods two years with respect to baseline, hernia and surgical characteristics.
Quality of Life was assessed with the use of a validated telephone
This prospectively collected observational study was registered at questionnaire in the form of a PROM concept. Patients were asked a self-
researchregistry.com (UIN: researchregistry7671) and conducted ac­ assessment of their complaints in a simple manner and understandable
cording to the STROBE (Strengthening the Reporting of Observational for everyone with capacity of the French language. QoL was compared
studies in Epidemiology) and STROCSS (Strengthening the Reporting of to the patient group that reported relevant pain at one year and two year
Cohort Studies in Surgery) statements [15]. to distinguish possible differences in experienced afflictions.
Data utilized in this study was prospectively collected by the mem­
bers of the Hernia-Club Registry [16] and retrospectively reviewed. This
2.4. Statistical analysis
registry functions as a prospectively compiled online database of sur­
gical procedures for all ventral abdominal wall hernias. The Hernia-Club
Baseline characteristics are presented as means with standard devi­
(club-hernie-mesh.com) is a society of French surgeons with an interest
ation, medians with inter-quartile range or as numbers of patients and
in parietal surgery, and who have been gathering the prospective ano­
percentages, as appropriate. Patients who had no pain at any measured
nymised data for all ventral abdominal wall hernia patients since 2011.
time point (one month, one year and two years postoperatively) were
compared to patients with relevant pain at one year and to patients with
2.1. Data collection
relevant pain at and two years postoperatively, using multivariable bi­
nary logistic regression, including the variables age, gender, ASA clas­
Surgeons from all over France enlisted patients consecutively and
sification, smoking status, activity level, diabetes, primary/recurrent
unselectively in close-ended boxes. Patients gave their formal consent
hernia, EHS hernia classification, and mesh fixation, presenting odds
for auditing of the original medical records in the case of discrepancies
ratios (OR) with 95% confidence intervals (95% CI). Surgical approach
between the database and patient reported outcomes during the follow-
and nerve handling were excluded from multivariable analysis, as these
up visit, independent of the attending surgeon.
variables correlate heavily to the mesh fixation that was used. This
Data collection for follow-up was performed by a clinical research
correlation was discovered by Cross-Tabulation. Univariable binary lo­
assistant (CRA) by telephone calls. This included validated Quality of
gistic regression was used for surgical approach, iliohypogastric nerve
Life (QoL) questionnaires in the form of Patient Reported Outcome
handling and ilioinguinal nerve handling. For iliohypogastric nerve
Measurements (PROMs).
handling, only patients undergoing the Lichtenstein procedure were
Patients were informed that their data was registered in a pseudo­
considered. For ilioinguinal nerve handling only patients undergoing the
nymized manner, and only the CRA and operating surgeon were able to
Lichtenstein and TIPP procedures were considered. Model fitting was
relate the pseudonymized number with the patient. Data was stored in a
assessed with the Log-Likelihood statistic. QoL outcomes were compared
protected databank in Switzerland. This registry is abiding to the re­
using Pearson’s Chi-Square test. Correction for multiple testing was not
quirements of the French ‘Commission Nationale de l’Informatique et
performed.
des Libertés’ (CNIL; registration number 1993959v0).
Additional information regarding registration in the Hernia-Club
3. Results
database can be found in the articles by Romain et al. and De Smet
et al. [13,17].
In total, 4.016 repairs were extracted from the Club-Hernie database
that completed the full two-year follow-up between 2011 and 2021, and
2.2. Patients and methods
were included in the analysis.
All patients undergoing elective inguinal hernia repair from 2011 to
2021 were identified from the Hernia-Club registry. Only patients with 3.1. Patient, hernia and surgical characteristics
minimum two years of follow-up were included in the study (Fig. 1).
Baseline characteristics of the included patients is shown in Table 1.
2.3. Outcomes Mean age of patients was 65.1 years and patients were predominantly
male (90.3%). Generally, patients had good physical health with only
Data derived from the registry included patient demographics such 7.7% of patients having a BMI higher than 30, only 9.9% of patients had
as age, gender, body mass index (BMI), American association of Anes­ an ASA classification exceeding two, 52.8% was active (i.e. doing ex­
thesiologists (ASA) classification, smoking, comorbidities such as dia­ ercise), and 4.1% of patients had diabetes.
betes, preoperative pain and physical activity. Primary hernias were the main type of hernia (92.8%), with lateral
Hernia characteristics identified were whether the hernia was pri­ and medial inguinal types representing the majority of hernia types.
mary or recurrent and classified according to the EHS classification of The surgical techniques were clustered into two groups: open and
inguinal hernias [18]. laparoendoscopic techniques.
Surgical characteristics included surgical techniques: Lichtenstein, Most procedures encompassed an open technique (55.2%). Meshes
transabdominal preperitoneal (TAPP) repair, transinguinal preper­ or other prosthetics were used in 96.7% repairs, and most prosthetics if
itoneal (TIPP) repair, totally extraperitoneal (TEP) repair, use of pros­ fixated, were fixated by resorbable staples (46.0%). Almost all surgeries
thetics, fixation of these prosthetics (e.g. resorbable and non-resorbable were performed under general anaesthesia, and many surgeons did not
sutures, resorbable and non-resorbable staples, glue, self-adhesive), type undertake nerve identification, 54.0% of the IHNs and 38.8% of the IINs
of anaesthesia (i.e. general, spinal or local) and intra-operative nerve the course was not recognized during open surgery.

3
L.M. van den Dop et al. International Journal of Surgery 105 (2022) 106837

Table 1 3.2. Conditional probability of CPIP

Baseline characteristics of patients with 2-year follow-up.
N = 4016 (%) Fig. 1 shows the development and evolution of CPIP in the study
population. Preoperatively, 2.929 (72.9%) patients had irrelevant pain
Patient characteristics
and 1.087 (27.1%) patients had relevant pain.
Age 65.1 (14.3) At one month postoperatively, the large majority of patients: 96.8%
Male sex 3628 (90.3)
BMI 25.1 (3.6)
(n = 3.886) had irrelevant pain and 3.2% (n = 130) had relevant pain.
Patients with BMI > 30 311 (7.7) Ninety-nine patients (3.4%) who did not have pain preoperatively, had
ASA classification pain at one month. One-thousand-and-fifty-six patients (97.1%) who
1 2036 (50.7) had pain preoperatively, did not have pain at one month.
2 1574 (39.2)
At one year postoperatively, 3.850 (95.9%) patients had irrelevant
3 397 (9.9)
Smoking pain and 167 (4.1%) patients had relevant pain. One-hundred-and-fifty-
Never smoked 2134 (53.1) two (4.0%) patients who did not have pain at one month, had pain at one
Ex-smoker > 12 months 1029 (25.6) year. One-hundred-and-twelve (86.8%) patients who had pain at one
Occasional smoker 177 (4.4) month, did not perceive their pain as relevant at one year.
Daily smoker 653 (16.3)
Activities
At two years postoperatively, 3.769 patients (93.8%) had irrelevant
None 1894 (47.2) pain and 247 (6.2%) patients had relevant pain. Two-hundred-and-
Sporadic (<1 time a week) 593 (14.8) nineteen (5.8%) patients who did not have pain at one year, had pain
Moderate (1 time a week 646 (16.1) at two years. One-hundred-and-thirty-nine (83.2%) patients who had
Intense (>1 times a week) 860 (21.4)
pain at one year, did not have pain at two years.
Work
Unemployed 1472 (36.7) Of the patients that experienced relevant pain preoperatively, 30
Administrative work 883 (22.0) (2.8%) continued to have relevant pain at one month. Sixteen (12.3%)
Moderate physical work 643 (16.0) patients experiencing pain at one month, continued developing CPIP at
Physical work 911 (22.7) one year, while only 39 (3.6%) of the patients experiencing preoperative
Diabetes 163 (4.1)
Preoperative pain
pain but who were pain free at one month, developed CPIP at one year.
None 777
Mild 2099 3.3. No pain at any point versus relevant pain at one year postoperatively
Moderate 53
Severe 1087
Patients who reported no pain at one month, one year and two years
Hernia characteristics
Primary hernia 3725 (92.8) postoperatively were compared to patients who reported relevant pain
Lateral (total) 2855 at one year postoperatively. These results are presented in Table 2.
L1 602 Patients of ages 60 and below were more likely to have relevant pain
L2 1499 (OR 1.763; 95% CI 1.150–2.703; p = 0.009) than patients over 60 years
L3 754
Medial (total) 1518
old. Males were at a significantly lower odds of having relevant pain (OR
M1 261 0.466; 95% CI 0.258–0.840; p = 0.011). This difference is also present in
M2 747 ASA 2 patients, when compared to ASA 1 patients (OR 1.976; 95% CI
M3 510 1.284–3.040; p = 0.002). Additionally, fewer ASA 1 patients were pre­
Femoral (total) 260
sent in the relevant pain group (p = 0.002). Recurrent IHs were at
F1 124
F2 108 significantly higher risk of relevant pain group than patients operated
F3 28 for a primary IH (OR 2.125; 95% CI 1.015–4.450; p = 0.046).
Surgical characteristics Lichtenstein repair was performed more often in the relevant pain
Open 2219 (55.2) group (p = 0.004). TIPP repair, when compared to Lichtenstein repair,
Lichtenstein 932 (23.2)
TIPP 1287 (32.0)
resulted in lower odds of relevant pain (OR 0.485; 95% CI 0.312–0.752;
Laparoendoscopic 1786 (44.0) p = 0.001). Not fixating the mesh was more prevalent in the no pain
TAPP 992 (24.7) group (p = 0.016). The opposite is true for fixation of the mesh with non-
TEP 776 (19.3) resorbable staples (OR 2.445; 95% CI 1.120–4.940; p = 0.013) and for
Received prosthetics 3883 (96.7)
self-adhesive mesh (OR 5.550; 95% CI 1.773–17.376; p = 0.003), both
Fixation
No fixation 2700 of which resulted in pain more often. Any type of nerve handling of the
Sutures (resorbable/not resorbable) 134/204 IHN was not statistically significant correlated with severe pain at one
Staples (resorbable/not resorbable) 578/183 year (p = 0.246, p = 0.194, p = 0.953). When the IIN was preserved
Glue 27 instead of not identified, pain occurred more often (OR 1.713; 95% CI
Self-adhesive 132
Type of anaesthesia
1.062–2.763; p = 0.027).
General 3786
Spinal or TAP Block 122 3.4. No pain at any point versus relevant pain at two years
Local anaesthesia 75 postoperatively
Iliohypogastric nerve:
Not applicable/not seen 2889
Preserved 998 Patients who reported no pain at one month, one year and two years
Resected 104 postoperatively were compared to patients who reported relevant pain
Ilioinguinal nerve: at two years postoperatively. These results are presented in Table 3.
Not applicable/not seen 2584 Younger patients (≤60 years old) were observed in the group with
Preserved 1272
Resected 129
relevant pain at two-years (OR 1.450; 95% CI 1.052–1.999; p = 0.023).
Males were less likely than females to have relevant pain (OR 0.490;
95% CI 0.310–0.773; p = 0.002). More patients in the relevant pain
group were occasional smokers (OR 2.566; 95% CI 1.490–4.417; p <
0.001). More patients in relevant pain group were operated for a F1
femoral hernia (p = 0.036).

4
L.M. van den Dop et al. International Journal of Surgery 105 (2022) 106837

Table 2
Patients with no pain at 1 month, 1 year and 2 years versus patients with relevant pain at 1 year. Missing values were omitted from analyses.1: only Lichtenstein repairs
considered in analysis.2: only Lichtenstein and TIPP repairs considered in analysis. *: too few (<10) patients in group in order to interpret logistic regression. **: Chi-
squared test. OR: odds ratio. UV: univariable logistic regression, MV: multivariable logistic regression.
No pain N = 2796 (%) Relevant pain N = 167 (%) OR (UV, 95% CI) OR (MV, 95% CI) P-value

Patient characteristics
Age
<60 years 842 (30.1) 56 (33.5) 1.763 (1.150–2.703) 0.009
>60 years 1952 (69.8) 111 (66.5) (ref)
Male sex 2552 (91.3) 142 (85.0) 0.466 (0.258–0.840) 0.011
BMI at year 1 24.7 (5.3) 24.3 (4.54) 1.013 (0.966–1.062) 0.598
ASA classification
1 1423 (50.9) 64 (38.3) (ref) 0.002**
2 1088 (38.9) 76 (45.5) 1.976 (1.284–3.040) 0.002
3 280 (10.0) 27 (16.2) 1.870 (0.799–4.375) 0.149
Smoking
Never smoked 1497 (53.8) 90 (53.9) (ref) 0.974**
Ex-smoker > 12 months 735 (26.4) 40 (24.2) 0.844 (0.511–1.395) 0.508
Occasional smoker 111 (4.0) 7 (4.2) * 0.577
Daily smoker 439 (15.8) 29 (17.6) 0.987 (0.583–1.672) 0.961
Activities
None 1312 (47.2) 97 (58.8) (ref) 0.004**
Sporadic (<1 time a week) 415 (14.9) 21 (12.7) 0.715 (0.405–1.261) 0.246
Moderate (1 time a week 469 (16.9) 22 (13.3) 0.736 (0.417–1.299) 0.291
Intense (>1 a week) 586 (21.1) 25 (15.2) 0.739 (0.435–1.257) 0.265
Diabetes
No 2246 (80.3) 127 (76.0) (ref) 0.965**
Yes 126 (4.5) 7 (4.2) * 0.385
Hernia characteristics
Primary hernia 2617 (93.6) 153 (91.6) (ref) 0.314**
First recurrence 137 (4.9) 11 (6.6) 2.125 (1.015–4.450) 0.046
Second recurrence 16 (0.6) 3 (1.8) *
Lateral (total) 1965 (70.2) 124 (74.2)
L1 411 (14.7) 32 (19.2) (ref) 0.116**
L2 1034 (37.0) 64 (38.3) 0.850 (0.488–1.479) 0.565
L3 520 (18.6) 28 (16.8) 0.534 (0.258–1.102) 0.090
Medial (total) 1044 (37.3) 68 (40.7)
M1 180 (6.4) 15 (9.0) (ref) 0.198**
M2 521 (18.6) 28 (16.8) 0.726 (0.339–1.553) 0.409
M3 343 (12.3) 25 (15.0) 0.708 (0.309–1.624) 0.415
Femoral (total) 163 (5.8) 20 (12.0)
F1 72 (2.6) 8 (4.8) * *
F2 78 (2.8) 7 (4.2) * *
F3 13 (0.5) 5 (3.0) * *
Surgical characteristics
Open
Lichtenstein 612 (22.0) 53 (31.7) (ref) 0.004**
TIPP 834 (30.0) 35 (21.0) 0.485 (0.312–0.752) 0.001
Laparoendoscopic
TAPP 767 (27.6) 44 (26.3) 0.662 (0.438–1.002) 0.051
TEP 563 (20.2) 35 (21.0) 0.718 (0.461–1.117) 0.142
Received prosthetics 2719 (92.7) 164 (98.2)
Fixation
No fixation 1877 (67.1) 97 (58.1) (ref) 0.016**
Sutures
Resorbable 98 (3.5) 2 (1.2) * 0.997
Not resorbable 127 (4.5) 8 (4.8) * 0.271
Staples
Resorbable 439 (15.7) 28 (16.8) 0.897 (0.515–1.563) 0.702
Not resorbable 129 (4.6) 17 (10.2) 2.445 (1.210–4.940) 0.013
Glue 19 (0.7) 2 (1.2) * 0.627
Self-adhesive 68 (2.4) 10 (6.0) 5.550 (1.773–17.376) 0.003
Iliohypogastric nerve1:
Not identified 781 (54.0) 27 (30.7) (ref) 0.246**
Preserved 593 (41.0) 57 (64.8) 1.642 (0.777–3.466) 0.194
Resected 68 (4.7) 4 (4.5) * 0.953
Ilioinguinal nerve2:
Not identified 561 (38.8) 25 (28.4) (ref) 0.052**
Preserved 799 (55.2) 61 (69.3) 1.713 (1.062–2.763) 0.027
Resected 80 (5.5) 2 (2.3) * 0.438
Reoperations 12 (0.4) 3 (1.8) *

TEP repair showed a lower odds of having relevant pain when prosthetic meshes resulted in more relevant pain when compared to
compared to Lichtenstein repair (OR 0.616; 95% CI 0.398–0.953; p = no fixation (OR 2.948; 95% CI 1.233–7.049; p = 0.015).
0.029). In the no pain group, the prosthetic mesh was not fixated more A proportionally higher number of resections of the IHN was per­
often than in the relevant pain group (p < 0.001). Self-adhesive formed in the relevant pain group (OR 2.800; 95% CI 1.150–6.816; p =

5
L.M. van den Dop et al. International Journal of Surgery 105 (2022) 106837

Table 3
Patients with no pain at 1 month, 1 year and 2 years versus patients with relevant pain at 2 years. Missing values were omitted from analyses.1: only Lichtenstein repairs
considered in analysis.2: only Lichtenstein and TIPP repairs considered in analysis. *: too few (<10) patients in group in order to interpret logistic regression. **: Chi-
squared test. OR: odds ratio. UV: univariable logistic regression. MV: multivariable logistic regression.
No pain N = 2796 (%) Relevant pain N = 247 (%) OR (UV, 95% CI) OR (MV, 95% CI) P-value

Patient characteristics
Age
≤60 years 842 (30.1) 103 (41.7) 1.450 (1.052–1.999) 0.023
>60 years 1952 (69.8) 144 (58.3) (ref)
Male sex 2552 (91.3) 204 (82.6) 0.490 (0.310–0.773) 0.002
ASA classification
1 1423 (50.9) 135 (54.7) (ref) 0.257**
2 1088 (38.9) 91 (36.8) 1.004 (0.717–1.407) 0.980
3 280 (10.0) 20 (8.1) 0.626 (0.255–1.536) 0.306
Smoking
Never smoked 1497 (53.8) 130 (52.8) (ref) 0.771**
Ex-smoker > 12 months 735 (26.4) 51 (20.7) 1.085 (0.733–1.606) 0.682
Occasional smoker 111 (4.0) 21 (8.5) 2.566 (1.490–4.417) < 0.001
Daily smoker 439 (15.8) 44 (17.9) 1.221 (0.811–1.838) 0.338
Activities
None 1312 (47.2) 121 (49.2) (ref) 0.542**
Sporadic (<1 time a week) 415 (14.9) 34 (13.8) 0.783 (0.496–1.237) 0.295
Moderate (1 time a week 469 (16.9) 40 (16.3) 0.786 (0.509–1.212) 0.276
Intense (>1 a week) 586 (21.1) 51 (20.7) 0.989 (0.673–1.452) 0.953
Diabetes
No 2246 (80.3) 214 (86.9) (ref) 0.050**
Yes 126 (4.5) 5 (2.0) * 0.065
Hernia characteristics
Primary hernia 2617 (93.6) 224 (90.7) (ref) 0.139
First recurrence 137 (4.8) 17 (6.9) 1.635 (0.875–3.053) 0.159
Second recurrence 16 (0.57) 2 (0.81) * 0.573
Lateral (total) 1965 (70.2) 180 (72.8)
L1 411 (14.7) 48 (26.3) (ref) 0.293
L2 1034 (37.0) 94 (38.1) 0.965 (0.629–1.482) 0.871
L3 520 (18.6) 38 (15.4) 0.725 (0.423–1.243) 0.242
Medial (total) 1044 (37.3) 83 (33.6)
M1 180 (6.4) 18 (7.3) (ref) 0.850
M2 521 (18.6) 37 (15.0) 1.018 (0.509–2.039) 0.959
M3 343 (12.3) 28 (11.3) 0.895 (0.416–1.925) 0.777
Femoral (total) 163 (5.8) 26 (10.5)
F1 72 (2.6) 12 (4.9) (ref) 0.036**
F2 78 (2.8) 9 (3.6) * 0.421
F3 13 (0.5) 5 (2.0) * 0.084
Surgical characteristics
Open
Lichtenstein 612 (22.0) 60 (24.4) (ref) 0.325**
TIPP 834 (30.0) 69 (28.0) 0.844 (0.588–1.211) 0.357
Laparoendoscopic
TAPP 767 (27.6) 79 (32.1) 1.051 (0.739–1.494) 0.784
TEP 563 (20.2) 34 (13.8) 0.616 (0.398–0.953) 0.029
Received prosthetics 2719 (97.2) 238 (97.2)
Fixation
No fixation 1877 (67.1) 140 (56.7) (ref) < 0.001**
Sutures
Resorbable 98 (3.5) 5 (2.0) * 0.221
Non-resorbable 127 (4.5) 18 (7.3) 0.983 (0.471–2.051) 0.963
Staples
Resorbable 439 (15.7) 39 (19.8) 1.357 (0.908–2.027) 0.136
Non-resorbable 129 (4.6) 11 (4.5) 1.008 (0.460–2.210) 0.984
Glue 19 (0.7) 1 (0.4) * 0.998
Self-adhesive 68 (2.4) 19 (7.7) 2.948 (1.233–7.049) 0.015
Iliohypogastric nerve1:
Not identified 781 (54.0) 56 (43.4) (ref) 0.199**
Preserved 593 (41.0) 59 (45.7) 1.404 (0.682–2.889) 0.357
Resected 68 (4.7) 14 (10.9) 2.800 (1.150–6.816) 0.023
Ilioinguinal nerve2:
Not identified 561 (38.8) 36 (27.9) (ref) 0.015**
Preserved 799 (55.3) 80 (62.0) 1.560 (1.038–2.346) 0.033
Resected 80 (5.5) 13 (10.1) 2.532 (1.288–4.979) 0.007
Reoperations 6 (0.2) 2 (0.8)

0.023). Non-identification of the IIN occurred more often in the relevant 95% CI 1.288–4.979; p = 0.007).
pain group (p = 0.015). The IIN being preserved resulted in a higher
odds of relevant pain, compared to not identifying the IIN (OR 1.560;
95% CI 1.038–2.346; p = 0.033). More relevant pain resulted from IIN 3.5. Quality of life assessment
resections when compared to non-identification of the IIN (OR 2.532;
Table 4 shows outcomes of QoL questionnaires of patients who had

6
L.M. van den Dop et al. International Journal of Surgery 105 (2022) 106837

Table 4 two years postoperatively found their current symptoms significantly

Quality of life compared between severe pain at 1 year and 2 years post­ less bothering than their preoperative hernia-related symptoms,
operatively. Differences in answers between the groups were calculated with the compared to patients with relevant pain at one year postoperatively (p
Chi2 test. Missing answers were excluded in statistical analyses. = 0.011).
Severe pain at Severe pain at p- Satisfaction with the surgery (Question 8) was significantly higher in
year 1 n = 167 year 2 N = 219 value patients with relevant pain at two years postoperatively (p = 0.001).
(%) (%)

Quality of Life 4. Discussion

Questionnaire
Q1: Does your abdominal wall 0.435
seem firm?
This observational, registry-based study demonstrated that overall,
Firm 154 (92.2) 196 (89.5) chances on developing CPIP are low, even if patients experienced pre­
Not firm 13 (7.8) 22 (10) operative relevant pain. Patient and surgical characteristics such as
Missing 0 (0) 1 (0.5) gender, age, ASA, type of surgery, recurrent surgery, type of fixation and
Q2: Do you feel a swelling or a 0.137
nerve handling are significantly associated with influencing CPIP onset.
lump in your groin?
Yes 11 (6.6) 29 (13.2) Remarkably, patients experiencing pain at one month, had relatively
No 156 (93.4) 189 (86.3) higher risks of developing CPIP than patients that experienced preop­
Missing 0 (0) 1 (0.5) erative pain (12.3% versus 3.6%). Patients who perceived relevant pain
Q3: Where are your symptoms 0.125
at one year, had high chances of alleviating their complaints at two
located?
Side of surgery 133 (79.6) 196 (89.5)
years, without need of reintervention. There was a small chance of
Median 9 (5.4) 6 (2.7) developing new CPIP complaints at two years.
Contralateral 4 (2.4) 8 (3.7) Previous studies showed significant differences in baseline charac­
Bilateral 5 (3.0) 1 (0.5) teristics for patients developing CPIP [11]. In accordance with litera­
Other locations 6 (3.6) 6 (2.7)
ture, the present study also showed significant differences in gender and
Missing 10 (6) 2 (1)
Q4: When do you feel these 0.001 ASA classification for patients developing CPIP at one year [19]. Occa­
symptoms? sional smokers seem to be at an higher risk of being in the relevant pain
During coughing, lifting 20 (12.0) 8 (3.7) at two year group. Femoral hernias seemed to lead to more CPIP com­
After exertion or at the end 18 (10.8) 48 (21.9) plaints after one and two years, which is congruent with literature [20].
of the day
During other activities 121 (72.4) 163 (74.4)
This may be due to the difference in pathophysiology of femoral hernias
Missing 8 (4.8) 0 (0) compared to inguinal hernias.
Q5: How often do you feel 0.001 There seems to be a difference in patient groups when comparing
these symptoms? patients experiencing relevant pain at one year and two years. For
Rarely 56 (33.5) 127 (58.0)
instance, the ASA classification is no longer significantly different be­
A few times per week 76 (45.5) 62 (28.3)
A few times per day 17 (10.2) 14 (6.4) tween the groups at two years, whereas frequency of smoking is. These
Often per day 8 (4.8) 4 (1.8) results suggest that during the longer follow up period other factors
All the time 4 (2.4) 12 (5.5) (affections) than the local inguinal postoperative anatomical conditions
Missing 6 (3.6) 0 (0) started to play a role. Laparoendoscopic techniques do differ in the sense
Q6: Do these symptoms: 0.003
Not hinder your activities? 142 (85.1) 192 (87.7)
that TEP is seemingly protective for CPIP at two years. This could be due
Cause temporary 4 (2.4) 6 (2.7) to the fact that by avoiding the neural planes less nerve handling con­
interruption of your siderations apply, and could therefore limits the chances on inguino­
activities? dynia. Not fixating the prosthetic mesh, possibly interfering with any
Prevent you from doing 2 (1.2) 18 (8.2)
nerves, seems to be favorable at one year and two years.
specific activities?
Missing 19 (11.4) 3 (1.4) Many studies have been conducted on CPIP and surgical techniques.
Q7: The nuisance/pain 0.011 In accordance with literature, more CPIP complaints are observed when
discomfort is: patients received an open repair for their IH (with the highest grade in
Less of a nuisance than my 123 (73.7) 185 (84.5) the Lichtenstein group), when compared to laparoendoscopic tech­
hernia was
More of a nuisance than my 29 (17.4) 20 (9.1)
niques [21]. While the evidence is apparent, many surgeons still practice
hernia was the Lichtenstein procedure in high numbers, perhaps due to the fact that
Missing 15 (9) 14 (6.4) the anterior approach can be performed without general anaesthesia and
Q8: How would you assess 0.001 has long been one of the first procedures that trainees are taught. Pa­
your surgery?
tients with IH opting for surgery should be informed about the CPIP
Bad or medium 71 (42.5) 45 (20.6)
Good or excellent 93 (55.7) 172 (78.5) percentages after Lichtenstein’s procedure. Surgical residents should be
Missing 3 (1.8) 2 (1.6) made aware of the fact that the open Lichtenstein technique is simple
but not easy, considering the large amount of patients experiencing
CPIP.
relevant pain at one year postoperatively, compared to patients who had The difference between relevant pain and irrelevant pain is based on
relevant pain at two years postoperatively. four different “levels” of pain. In studying pain, arbitrary cut-off points
Firmness of the abdominal wall (Question 1) was similar in either for pain have to be chosen. It is known that patients experience pain in a
group (p = 0.435). Feeling a lump or swelling in the groin region and the multitude of different ways and that pain is not merely physical, but
location of any present symptoms (Question 2 and Question 3) were also psychological and social factors play a large role in how pain is expe­
answered similarly (p = 0.137 and p = 0.125, respectively). rienced. Patients might tolerate a high amount of pain if they are still
Significant differences were present in the moment symptoms were able to perform certain daily tasks and it is up to their interpretation how
experienced (Question 4), the frequency of symptoms (Question 5) and severely they are affected. Judgement of pain in this study is determined
the amount of hindrance in activities experienced from symptoms by the patients themselves, and might therefor be subjective with regard
(Question 6, p = 0.001, p = 0.001 and p = 0.003, respectively). When difference in relevant and irrelevant pain.
asked how their current symptoms compare to their preoperative Fixation of prosthetics meshes in IH repair can be done in many
hernia-related symptoms (Question 7), patients with relevant pain at ways. Fixation types included in this study were non-resorbable sutures

7
L.M. van den Dop et al. International Journal of Surgery 105 (2022) 106837

and staples, resorbarble sutures and staples, glue and self-adhesive 6. Conclusion
prosthetics not requiring fixation. Significantly more relevant pain was
observed at one year and two years when fixation was used, although There is a low risk of developing CPIP after IH surgery. CPIP was
due to small numbers of patients, analyses on resorbable and non- relatively more apparent in patients with younger age, higher ASA
resorbable sutures are incomplete. This effect was most apparent classification, female gender, open surgery, and if fixation of prosthetic
when non-resorbable staples or self-adhesive prosthetics were used. This meshes was used. The present study found that postoperative pain at one
could be due to the fact that staples that were non-resorbable entrapped month is a relatively higher risk for later onset of CPIP, when compared
a nerve. Results from this study might indicate that the use of non- to preoperative pain. Patients could be informed if they experience CPIP
resorbable staples should be avoided in IH repair, as other types of fix­ at one year that they are likely to ameliorate their complaints at two-
ations are available. year follow-up. Patients presenting CPIP after two years seem to have
Cirocchi et al. [6] wrote an extensive review with respect to pro­ a different pain etiology than patients presenting CPIP at one year. The
phylactic nerve transection to prevent CPIP. As results varied between present study could aid surgeons when informing patients about the
studies, the authors made it clear that no conclusions could be drawn onset and course of CPIP after IH surgery.
from their meta-analysis. It is appreciable however, that pain on the
short postoperative period (i.e. six months postoperatively) declined Provenance and peer review
after neurectomy. However, at one year, the advantage of the neu­
rectomy was no longer present. This could be attributed to neuroma Not commissioned, externally peer-reviewed.
formation of the IIN or IHN, beginning at postoperative periods
exceeding three months [22]. Difference should be made however, in Ethical approval
the interpretation of the ‘not identified’ group. In TIPP repair, the
recommendation of identification of all three nerves is not standard All patients agreed to their data being stored in a pseudonymized
recommendation as it is with Lichtenstein’s repair [23]. Only the IIN manner in a protected databank in Switzerland. This registry is abiding
should be identified and preserved in TIPP repair, while the IHN is not at to the requirements of the French ‘Commission Nationale de l’Infor­
risk due to absence of extensive pre-muscular dissection, and no fixation matique et des Libertés’ (CNIL; registration number 1993959v0).
is performed. This could be the main reason for the paradoxical CPIP
outcome in unidentified nerves. Next to this, it was not clear whether the Sources of funding
surgeons merely identified the nerves or actively searched for the nerve.
In ‘digging up’ the nerve, unintentional lesions to the nerve could have None.
taken place. Resection of both nerves seemed to lead to more CPIP
complaints at two years, which could be due to neuroma formation and Author contribution
the fact that resection of the nerve could require dissection locally
around it. L.M. van den Dop, MD: Study design, data analysis, writing
Differences in PROM answers between pain at one year and to year F.P.J. den Hartog, MD: Data analysis, writing
were present in time-dependent questions, while localization-specific D. Sneiders, MD: Data analysis, writing
answers were equal. As prior results suggest other factors to play a G. Kleinrensink, PhD: Study design, writing
role in the long-term pain perception, the same could be true for the J.F Lange, MD, PhD: Study design, writing
PROM answers. To underline this, patients experiencing pain at two J.F. Gillon, MD4: Study design, writing
years judge their CPIP as less of a nuisance compared to the pain they Hernia-Club Members: Data collection
experienced from their hernia, compared to patients experiencing pain
at one year (9.1% versus 17.4%, respectively). Furthermore, patients Conflicts of interest
experiencing pain at two years predominantly (78.5%) assess their
surgery as good or excellent while patients experiencing pain at one year None.
asses their surgery as good or excellent in a mere 55.7%.
Research registration Unique Identifying number (UIN)
5. Limitations
Name of the registry: Researchregistry.com.
The present study is based on retrospective data from a prospectively Unique Identifying number or registration ID: Researchregistry7671.
maintained cohort. As patient inclusion into this database is not ran­ https://www.researchregistry.com/register-now#home/registrat
domized, although the experienced surgeons that include patients iondetails/621399ce0d04e5001e4c5ba4/
attempt to minimize selection, there will undoubtedly be some form of
selection bias present in the selected sample of patients. Guarantor
Surgeons that participate in the Hernia Club database are experi­
enced hernia surgeons. Inguinal hernia surgery is not only performed by L.M. van den Dop, MD.
hernia surgeons, but also by most other types of surgeons around the
world. This inhibits the ability to generalize the outcomes of the Hernia Acknowledgements
Club database to the broader surgeon community.
Identifying the characteristics that make one group different from We thank members of the Club-Hernie members for input of their
another in a non-randomized setting, using their baseline and surgical data, and therefore the making of this study.
characteristics, can be done by statistical analysis, which can bring forth
correlations between these characteristics and the outcomes that are Appendix A. Supplementary data
assessed. However, this does not directly mean that there is a direct
causal connection between these characteristics and the patients’ out­ Supplementary data to this article can be found online at https://doi.
comes. Results from this study should therefore be taken with caution. org/10.1016/j.ijsu.2022.106837.

8
L.M. van den Dop et al. International Journal of Surgery 105 (2022) 106837

References [13] B. Romain, T. Fabacher, P. Ortega-Deballon, L. Montana, J.P. Cossa, J.F. Gillion, et
al., Longitudinal cohort study on preoperative pain as a risk factor for chronic
postoperative inguinal pain after groin hernia repair at 2-year follow-up, Hernia 26
[1] J.F. Lange, V.M. Meyer, D.A. Voropai, E. Keus, A.R. Wijsmuller, R.J. Ploeg, et al.,
(1) (2022) 189–200.
The role of surgical expertise with regard to chronic postoperative inguinal pain
[14] R.N. van Veen, A.R. Wijsmuller, W.W. Vrijland, W.C. Hop, J.F. Lange, J. Jeekel,
(CPIP) after Lichtenstein correction of inguinal hernia: a systematic review, Hernia
Randomized clinical trial of mesh versus non-mesh primary inguinal hernia repair:
20 (3) (2016) 349–356.
long-term chronic pain at 10 years, Surgery 142 (5) (2007) 695–698.
[2] J. Scholz, N.B. Finnerup, N. Attal, Q. Aziz, R. Baron, M.I. Bennett, et al., The IASP
[15] G. Mathew, R. Agha, S. Group, STROCSS 2021: strengthening the reporting of
classification of chronic pain for ICD-11: chronic neuropathic pain, Pain 160 (1)
cohort, cross-sectional and case-control studies in surgery, Ann Med Surg (Lond).
(2019) 53–59.
72 (2021), 103026.
[3] J.F. Lange, R. Kaufmann, A.R. Wijsmuller, J.P. Pierie, R.J. Ploeg, D.C. Chen, et al.,
[16] I. Kyle-Leinhase, F. Kockerling, L.N. Jorgensen, A. Montgomery, J.F. Gillion, J.A.
An international consensus algorithm for management of chronic postoperative
P. Rodriguez, et al., Comparison of hernia registries: the CORE project, Hernia 22
inguinal pain, Hernia 19 (1) (2015) 33–43.
(4) (2018) 561–575.
[4] C. Nikkolo, U. Lepner, Chronic pain after open inguinal hernia repair, Postgrad.
[17] G.H.J. de Smet, D. Sneiders, Y. Yurtkap, A.G. Menon, J. Jeekel, G.J. Kleinrensink,
Med. 128 (1) (2016) 69–75.
et al., Functional outcomes in symptomatic versus asymptomatic patients
[5] W. Reinpold, Risk factors of chronic pain after inguinal hernia repair: a systematic
undergoing incisional hernia repair: replacing one problem with another? A
review, Innov Surg Sci 2 (2) (2017) 61–68.
prospective cohort study in 1312 patients, Int. J. Surg. 82 (2020) 76–84.
[6] R. Cirocchi, M. Sutera, P. Fedeli, G. Anania, P. Covarelli, F. Suadoni, et al.,
[18] M. Miserez, J.H. Alexandre, G. Campanelli, F. Corcione, D. Cuccurullo, M.
Ilioinguinal nerve neurectomy is better than preservation in Lichtenstein hernia
H. Pascual, et al., The European hernia society groin hernia classification: simple
repair: a systematic literature review and meta-analysis, World J. Surg. 45 (6)
and easy to remember, Hernia 11 (2) (2007) 113–116.
(2021) 1750–1760.
[19] M.F. Bjurstrom, M.R. Irwin, D.C. Chen, M.T. Smith, A. Montgomery, Sex
[7] N. Crea, G. Pata, Effects of prophylactic ilioinguinal nerve excision in mesh groin
differences, sleep disturbance and risk of persistent pain associated with groin
hernia repair: short- and long-term follow-up of a randomized clinical trial, Am.
hernia surgery: a nationwide register-based cohort study, J. Pain 22 (11) (2021)
Surg. 76 (11) (2010) 1275–1281.
1360–1370.
[8] F. Karakayali, E. Oksuz, E. Turk, M. Pekmez, Z. Karabulut, T. Yilmaz, et al.,
[20] U. Dahlstrand, G. Sandblom, P. Nordin, S. Wollert, U. Gunnarsson, Chronic pain
Effectiveness of multiple neurectomies to prevent chronic groin pain after tension-
after femoral hernia repair: a cross-sectional study, Ann. Surg. 254 (6) (2011)
free hernia repair, Int. Surg. 95 (1) (2010) 40–48.
1017–1021.
[9] B. de Goede, P.J. Klitsie, B.J. van Kempen, L. Timmermans, J. Jeekel, G. Kazemier,
[21] G. HerniaSurge, International guidelines for groin hernia management, Hernia 22
et al., Meta-analysis of glue versus sutured mesh fixation for Lichtenstein inguinal
(1) (2018) 1–165.
hernia repair, Br. J. Surg. 100 (6) (2013) 735–742.
[22] A. Yavuz, H. Kulacoglu, E. Olcucuoglu, S. Hucumenoglu, C. Ensari, Z. Ergul, et al.,
[10] A. Kaul, S. Hutfless, H. Le, S.A. Hamed, K. Tymitz, H. Nguyen, et al., Staple versus
The faith of ilioinguinal nerve after preserving, cutting, or ligating it: an
fibrin glue fixation in laparoscopic total extraperitoneal repair of inguinal hernia: a
experimental study of mesh placement on inguinal floor, J. Surg. Res. 171 (2)
systematic review and meta-analysis, Surg. Endosc. 26 (5) (2012) 1269–1278.
(2011) 563–570.
[11] U. Franneby, G. Sandblom, P. Nordin, O. Nyren, U. Gunnarsson, Risk factors for
[23] J.F. Gillion, J.M. Chollet, Chronic pain and quality of life (QoL) after transinguinal
long-term pain after hernia surgery, Ann. Surg. 244 (2) (2006) 212–219.
preperitoneal (TIPP) inguinal hernia repair using a totally extraperitoneal,
[12] B. Romain, J.F. Gillion, P. Ortega-Deballon, N. Meyer, H. Club, Patient’s
parietalized, Polysoft (R) memory ring patch : a series of 622 hernia repairs in 525
satisfaction at 2 years after groin hernia repair: any difference according to the
patients, Hernia 17 (6) (2013) 683–692.
technique? Hernia 22 (5) (2018) 801–812.