Standardised Pfannenstiel versus supraumbilical muscle‐sparing transverse extraction sites in colon cancer surgery: An IDEAL stage 4 report

Abstract

Aim

To compare long‐term incisional hernia (IH) risk between Pfannenstiel and supraumbilical muscle‐sparing transverse extraction sites following minimally invasive colon cancer surgery with standardised fascial closure. Reported according to the IDEAL framework (Stage 4).

Method

This single‐centre study included patients (Stages I–III colon adenocarcinoma, 2014–2017) undergoing laparoscopic resection via Pfannenstiel or supraumbilical incisions. Fascial closure used continuous slowly absorbable poly‐4‐hydroxybutyrate sutures (ratio ≥ 4:1). The primary endpoint was CT‐verified IH at the extraction site at 5 years. Inverse Probability of Treatment Weighting (IPTW) adjusted for baseline covariates.

Results

The final cohort comprised 245 patients (Pfannenstiel, n = 141; supraumbilical, n = 104). The median follow‐up was 5.3 and 5.6 years, respectively. In the unadjusted analysis, the 5‐year IH risk was 15.4% (95% CI: 9.5–21.3) for Pfannenstiel and 18.8% (95% CI: 11.2–26.4) for supraumbilical incisions (p = 0.49). Following IPTW adjustment, the estimated risks were 14.6% (95% CI: 8.9–20.3) and 13.6% (95% CI: 7.7–19.4), respectively (adjusted risk difference − 1.0%; 95% CI: −9.2–7.2, p = 0.81). Only two (0.8%) of the 245 patients underwent surgical hernia repair.

Conclusion

The IH risk seems not associated with the extractions site incision, Pfannenstiel and supraumbilical muscle‐sparing transverse extraction sites, when standardised closure techniques are applied. These findings challenge the assumption that Pfannenstiel sites are inherently superior for hernia prevention. The choice of extraction site based on anastomosis technique may not be justified; instead, it should be guided by oncological requirements and procedural ergonomics rather than perceived differences in hernia risk.

Trial Registration

Danish Patient Safety Authority (#31–1522‐51).

What does this paper add to the literature?

A muscle‐sparing supraumbilical transverse incision is comparably robust to the Pfannenstiel approach regarding hernia risk provided that gold‐standard fascial closure is employed.

INTRODUCTION

Following abdominal surgery, incisional hernia (IH) represents a frequent and significant long‐term sequela. Prevalence rates fluctuate substantially in the literature, estimated at 2% to nearly 30% depending on the population and follow‐up modality [1, 2, 3]. Established patient‐related predictors of hernia formation include higher body mass index (BMI), tobacco use, diabetes mellitus and advanced age [3, 4, 5, 6]. Surgical factors also play an important role; these include the specific operative approach, incision characteristics and the presence of surgical site infection [2, 7, 8]. Computed tomography (CT) offers the highest diagnostic sensitivity, detection rates of up to 35%, and even in minimally invasive surgery (MIS) cohorts, radiologically confirmed IH rates persist at approximately 25% [9, 10]. Beyond the clinical risks of incarceration and strangulation, the development of IH significantly impairs patient quality of life, leading to chronic discomfort, cosmetic dissatisfaction and functional limitations [11].

Enhanced recovery pathways for colon cancer originated in Denmark, first reported by Kehlet et al. in 1995 [12], and have since been adopted nationwide. Although transverse incisions were initially adopted to optimise epidural‐managed pain control, the advent of minimally invasive surgery (MIS) as the gold standard has rendered epidurals largely obsolete. Nevertheless, transverse incisions, such as Pfannenstiel or supraumbilical approaches, remain preferable to midline incisions among Danish surgeons.

At Copenhagen University Hospital—North Zealand (NOH), all colorectal cancer resections were standardised in 2008 as complete mesocolic excision (CME) [13]. For MIS, resections with ileocolic anastomoses involve an extracorporeal hand‐sewn anastomosis performed after bowel extraction through a supraumbilical transverse incision. For MIS sigmoid resections, a Pfannenstiel incision is used for bowel extraction. Fascial closure has also been standardised since 2008, employing the small‐bite technique [14] from 2013.

Interventions under study

The interventions evaluated are two established, distinct off‐midline specimen extraction sites—the Pfannenstiel incision and the supraumbilical muscle‐sparing transverse incision—within a framework of standardised fascial closure. Both techniques were performed using a continuous, slowly absorbable poly‐4‐hydroxybutyrate suture with a targeted suture‐to‐wound length ratio of ≥ 4:1 for fascial closure of the bowel extraction site after elective laparoscopic resection for stage I–III colon adenocarcinoma. This standardised fascial closure technique has been considered the gold standard for a decade in accordance with the European Hernia Society recommendations [15].

What is novel about the intervention in this paper

The choice of extraction site is often linked to the intended type of anastomosis (intracorporeal vs extracorporeal). Intracorporeal ileocolic anastomosis has become increasingly popular, and some of the proposed advantages, such as a lower risk of IH, reduced ileus and shorter length of stay, have been associated with the extraction incision [16, 17, 18]. However, there is a paucity of data comparing these outcomes between off‐midline extraction sites in MIS, particularly between supraumbilical muscle‐sparing transverse incisions and Pfannenstiel incisions. Although some data suggest variation in risk, previous comparisons have often been confounded by heterogeneous closure methods.

This study provides a direct long‐term comparison of two specific transverse extraction sites while controlling for the confounding effect of heterogeneous closure techniques, specifically addressing the comparative robustness of the supraumbilical muscle‐sparing site against the Pfannenstiel site. Furthermore, the systematic use of 5‐year CT surveillance to detect IH provides a more accurate assessment of the long‐term risk of IH than clinical examination alone.

IDEAL stage of assessment

This study represents an IDEAL Stage 4 (Surveillance) report. The interventions are now well‐established clinical practices at Danish centres. This report evaluates the long‐term performance, late complications (hernia formation) and durability of these techniques over a 5‐year follow‐up period in a large, consecutive patient cohort. The findings serve to monitor the safety and efficacy of the standardised protocol and to identify any lessons from prolonged experience that might question the assumption of a lower risk of IH after intracorporeal ileocolic anastomosis.

The objective of this study was to evaluate whether specimen extraction via a supraumbilical muscle‐sparing transverse incision carries a distinct risk of IH compared with the Pfannenstiel approach, within a cohort adhering to internationally standardised closure protocols.

METHODS

Design

Adhering to the STROBE reporting guidelines [19] and IDEAL Stage 4 requirements [20], this single‐centre cohort study employed a retrospective analysis of prospectively acquired cancer surveillance CT scans to evaluate the long‐term outcome (IH) of two established incision techniques. Data collection was authorised by the Regional Data Protection Agency (HIH‐2013‐031 I‐02468) and the Danish Patient Safety Authority (31–1522‐51). The study utilised REDCap (Research Electronic Data Capture) [21] for data management, hosted on servers within the Capital Region of Denmark. In accordance with Danish regulations, approval from the Ethics Committee was not required.

The study population was derived from the Copenhagen cOmplete Mesocolic Excision Study (COMES) registry, which encompasses all patients treated for stages I–III colon adenocarcinoma with elective curative resection in the Capital Region of Denmark from 2008 to 2017 [22, 23, 24]. Inclusion was restricted to patients treated at NOH, as this centre uniquely implemented a standardised oncological surveillance protocol comprising annual thoracic and abdominal CT scans for 5 years. Variables regarding preoperative status, intraoperative details (specifically incision characteristics and closure materials) and postoperative outcomes were extracted prospectively from electronic health records (EHR).

Population

Patients were eligible for inclusion if they underwent elective, curative‐intent laparoscopic resection for stage I–III colon adenocarcinoma at NOH from 1 January 2014, to 31 December 2017. We excluded individuals with a preoperative stoma, those whose extraction incision utilised a prior surgical scar, and cases complicated by early surgical site events (e.g. fascial dehiscence, infection) or re‐entry through the incision within 60 days. Furthermore, procedures converted to open surgery or utilising a midline extraction site were omitted from the analysis.

Exposure and outcomes

Standardised complete mesocolic excision was performed in all cases. The primary exposure variable was the specific extraction site utilised during MIS. The cohort was categorised into two groups: the Pfannenstiel incision group (standard for left‐sided resections) and the supraumbilical muscle‐sparing transverse incision group (standard for right‐sided resections). The supraumbilical incision generally involved minimal to no division of the rectus abdominis muscle. Conversely, the Pfannenstiel technique employed a low transverse fascial incision superior to the pubic symphysis, sparing the rectus muscles, with vertical peritoneal access. Fascial closure was uniform across the cohort, utilising a poly‐4‐hydroxybutyrate suture (Monomax 0, B. Braun, Melsungen, Germany) aiming at a distance of 5 mm from the fascial edges and with a maximum interval distance of 5 mm. All operations were conducted or supervised by a group of five specialist colorectal surgeons.

The primary endpoint was the presence of IH at the specimen extraction site; herniation at trocar sites was not classified as an event. A systematic reassessment of all available annual follow‐up CT scans, performed during the hospital's standardised 5‐year cancer follow‐up programme, was conducted between January 2022 and July 2023. If abdominal CT scanning was performed between the annual intervals for any indication, these were also assessed as part of this study. CT imaging was re‐evaluated for IH by two independent reviewers via the regional picture archiving and communication system. Assessment for fascial defects on CT is considered a simple procedure not requiring a specialist radiologist. As detailed previously [25], Reviewer 1 (MCR) underwent specific training by Reviewer 2 (NNB), who already possessed validated experience from earlier research [6], for radiological hernia detection prior to the study. To validate the data on IH, Reviewer 2 independently re‐evaluated a subset of scans. Reviewer 1 assessed scans without blinding to the incision site, while Reviewer 2 remained blinded to the stratification group. Discrepancies were resolved through consensus discussion.

Consistent with our prior methodology [25], IH was defined as a distinct fascial defect located within the region of the extraction scar. Fascial defects adjacent to port sites were excluded, as their dimensions and precise locations were not prospectively logged or marked for follow‐up. Hernias identified clinically (through physical examination or subsequent repair or other laparotomy) were also recorded during routine annual visits or if symptoms were reported.

Events considered as competing risks included reoperation for IH, subsequent abdominal surgery involving fascial incision > 60 days post‐index procedure and mortality prior to IH occurrence. Both were informative/dependent censoring. Before conducting the analyses, we were aware that no patients included emigrated, which would have been the only other relevant competing risk.

Measures

Surgical and complication data were sourced from the COMES database, the rigorous audit process of which has been described previously [26, 27]. For the current study, the dataset was augmented by audited incision types and CT‐detected IH findings. For every identified IH, the anatomical location and date of initial detection were logged. To ensure data integrity, specific data points were double‐checked during collection and subsequent to the establishment of the database, supplemented by validation checks performed on a random sample of patients [25].

Statistical analyses

The statistical strategy was predefined and aligned with the standard survival analysis reporting protocols of the COMES database [23, 24, 25, 28, 29, 30]. Continuous variables were reported as medians with interquartile ranges (IQR), while categorical variables were expressed as frequencies and percentages. Group comparisons employed the Kruskal–Wallis test for continuous data and Fisher's exact test for categorical data. We conducted time‐to‐event analyses using a competing risk framework to estimate the 5‐year cumulative incidence of IH, accounting for death and laparotomy as competing events. Unadjusted data are displayed via cumulative incidence and survival curves.

To mitigate selection bias and approximate the results of a randomised trial, Inverse Probability of Treatment Weighting (IPTW) was applied to estimate marginal or population‐averaged exposure effects [31]. This method utilises propensity scores to weight individual patient data according to the inverse probability of the assigned exposure (incision type). IPTW yields unbiased estimates of the average treatment effect in time‐to‐event analyses provided that observed baseline covariates are balanced between groups [32]. Stabilised weights were employed to balance the following clinically significant covariates: age, sex, BMI, ASA classification (aggregating ASA III and IV), diabetes, smoking status and adjuvant chemotherapy.

Covariate balance—including two‐way interactions and quadratic terms for continuous variables—was evaluated by calculating the absolute mean differences between the Pfannenstiel and supraumbilical transverse groups post‐IPTW, using the ‘cobalt’ R package. We considered standardised absolute differences in means and raw mean differences of < 0.1, alongside variance ratios between 0.5 and 2, as indicative of negligible imbalance. Post‐weighting covariate distributions were also inspected graphically. All analyses were performed in R version 4.5.0 [33]. Patients with missing data for balanced covariates were excluded; otherwise, the full dataset was utilised.

RESULTS

In the database, 451 patients were identified who underwent elective resection with curative intent. Following the application of specific exclusion criteria, 206 patients were excluded, leaving a final cohort of 245 patients (Figure 1). Stratification by extraction site revealed that 141 patients (57.6%) received a Pfannenstiel incision, while 104 (42.4%) underwent a supraumbilical muscle‐sparing transverse incision. Table 1 details the baseline demographics, surgical procedures and chemotherapy administration. The median duration of follow‐up did not differ significantly between groups, standing at 5.3 years (IQR 4.3–6.0) for the Pfannenstiel cohort and 5.6 years (IQR 5.0–6.4) for the supraumbilical cohort (p = 0.134). No hernia was clinically diagnosed before it was detectable on a CT scan assessed in this study.

FIGURE 1.

Flow chart of patient inclusion and exclusion.

TABLE 1.

Baseline characteristics, procedures performed and chemotherapy administered for the 245 patients included, stratified by incision and proportions (%).

Variables	Pfannenstiel incision (n = 141)	Supraumbilical muscle‐sparing transverse incision (n = 104)	p
Median age—years (IQR)	67.7 [60.0–74.0]	75.7 [70.0–80.8]	< 0.001
Male/female sex—no. (%)	84 (59.6)/57 (40.4)	46 (44.2)/58 (55.8)	0.020
Median BMI—kg/m2 (i.q.r)	25.0 [22.8–27.8]	25.8 [22.5–29.3]	0.398
Active smoker—no. (%)	20 (14.2)	9 (8.7)	0.231
Diabetes—no. (%)	10 (7.1)	15 (14.4)	0.086
ASA physical status—no. (%)			< 0.001
1	56 (39.7)	19 (18.3)
2	67 (47.5)	46 (44.2)
3 and 4	18 (12.8)	39 (37.5)
WHO performance score—no. (%)			<0.001
0	120 (84.1)	68 (65.4)
1	18 (12.8)	23 (22.1)
2	3 (2.1)	12 (11.5)
3	0	1 (1.0)
Neoadjuvant chemotherapy	5 (3.5)	8 (7.7)	0.163
Postoperative UICC stage (%)			0.121
I	53 (37.3)	24 (23.3)
II	43 (30.3)	42 (40.8)
III	45 (31.7)	36 (35.0)
NA a	1 (0.7)	1 (1.0)
Procedure, n (%)			< 0.001
Right hemicolectomy	0	65 (62.5)
Extended right hemicolectomy	0	24 (23.1)
Right‐sided subtotal colectomy	2 (1.4)	1 (1.0)
Left hemicolectomy	17 (12.1)	1 (1.0)
Sigmoid resection	118 (83.7)	0
Other segmental colon resection	1 (0.7)	12 (11.5)
Total colectomy	3 (2.1)	1 (1.0)
Adjuvant chemotherapy	56 (39.7)	27 (26.0)	0.035

Note: Continuous data are presented as the median and interquartile range (IQR). Categorical data are presented as frequencies. Kruskal–Wallis tests were used for continuous data, and Fisher's exact tests for categorical data.

^a No residual tumour tissue in the specimen after neoadjuvant chemotherapy.

The covariate balance before and after IPTW adjustment is shown in Figure 2. In the unadjusted analysis, the absolute risk of IH was 15.4% (95% CI: 9.5–21.3) for the Pfannenstiel group and 18.8% (95% CI: 11.2–26.4) for the supraumbilical group. This corresponded to an unadjusted absolute risk difference of 3.4% (95% CI: −6.2–13.0, p = 0.49) in favour of the Pfannenstiel group. Unadjusted 5‐year cumulative incidence curves for IH, reoperation, and death before recurrence are presented in Figure 3. Following IPTW adjustment, the estimated risk of IH was 14.6% (95% CI: 8.9–20.3) and 13.6% (95% CI: 7.7–19.4), respectively. The adjusted absolute risk difference was −1.0% (95% CI: −9.2–7.2, p = 0.81). No significant differences were observed between the groups when accounting for death and subsequent laparotomy as competing risks (Table 2).

FIGURE 2.

Baseline covariate balance before and after adjustment using inverse probability of treatment weighting (IPTW).

FIGURE 3.

Unadjusted competing risk analyses of the 5‐year cumulative incidence of incisional hernia, and the competing risks of subsequent laparotomy and death before IH occurrence, stratified by incision type.

TABLE 2.

Competing risk analyses, with and without inverse probability of treatment weighting (IPTW), of the 5‐year cumulative incidence of incisional hernia in 245 patients undergoing colon cancer resection, stratified by incision type.

Outcomes	Pfannenstiel incision	Supraumbilical muscle‐sparing transverse incision	Absolute risk difference	p
Cumulative incidences—without IPTW
Incisional hernia	15.4% (9.5–21.3)	18.8% (11.2–26.4)	3.4% (−6.2–13.0)	0.49
Subsequent laparotomy	14.6% (8.8–20.4)	10.7% (4.5–16.9)	−3.9% (−12.4–4.6)	0.37
Death before recurrence	8.2% (3.9–12.4)	14.0% (7.5–12.4)	5.8% (−1.4–13.0)	0.11
Cumulative incidences—with IPTW
Incisional hernia	14.6% (8.9–20.3)	13.6% (7.7–19.4)	−1.0% (−9.2–7.2)	0.81
Subsequent laparotomy	11.5% (6.7–16.2)	7.8% (3.2–12.4)	−3.6% (−10.3–3.0)	0.29
Death before recurrence	11.4% (5.2–17.7)	9.9% (4.8–15.0)	−1.5% (−9.4–16.3)	0.69

Note: Data are % (95% CI).

Surgical repair of IH was performed in two patients (1.9%) in the supraumbilical group, whereas no patients in the Pfannenstiel group required repair (p = 0.179). The majority of radiologically detected hernias remained asymptomatic throughout the surveillance period.

DISCUSSION

This study demonstrated comparable rates of incisional herniation between the two off‐midline extraction sites, with adjusted 5‐year incidences of 14.6% for Pfannenstiel and 13.6% for supraumbilical transverse incisions. These results are particularly salient given the contemporary surgical trend towards intracorporeal ileocolic anastomosis. Proponents of the intracorporeal technique frequently cite the ability to utilise a Pfannenstiel extraction site as a primary advantage for hernia prevention. However, our findings suggest that an extracorporeal approach via a supraumbilical site is equally robust, provided a standardised, muscle‐sparing technique is utilised.

Consequently, the choice between these anastomotic techniques should be dictated by intraoperative and postoperative patient‐related recovery metrics rather than perceived differences in IH risk. These metrics were thoroughly assessed in a well‐designed triple‐blinded randomised controlled trial (RCT) by Dohrn et al. [34], who compared intracorporeal with hand‐sewn single‐layer end‐to‐end extracorporeal ileocolic anastomosis in resections for right‐sided colon cancer in a standardised enhanced postoperative recovery setting. They found no statistically significant differences in patient‐reported recovery, including postoperative pain, nausea, time to ambulation, time to first passing of gas or bowel movements, or length of hospital stay. All intraoperative and postoperative variables showed no difference, with the exception of a clinically insignificant longer duration of anastomotic creation for the intracorporeal approach [35]. Many studies, as included in a meta‐analysis [17], indicate better short‐term outcomes after intracorporeal anastomosis, but these studies' generalisability is limited by bias related to design and no use of enhanced recovery programmes.

A cohort study [16] also assessed the association between intracorporeal and extracorporeal anastomosis without showing differences in short‐term outcomes, such as wound complications, anastomotic leakage or time to passing gas. While they reported a longer hospital stay in the extracorporeal anastomosis group, the supporting statistics remain questionable; the methods of data analysis were not described, and means and standard deviations were reported for length‐of‐stay data that were evidently highly right‐skewed. Furthermore, while that study found no significant 1‐year IH risk difference, the results are difficult to interpret as the extraction incisions were not standardised, and the anatomical locations, types (vertical vs horizontal) and fascial closure techniques remained undescribed.

Given the questionable patient‐related benefits, the potential implications for resource utilisation must be carefully evaluated, particularly within public, tax‐funded healthcare systems such as those in the United Kingdom and Denmark. In the Capital Region of Denmark, an intracorporeal anastomosis—whether laparoscopic or robot‐assisted—currently costs approximately £580 (€670) more than a hand‐sewn, continuous, single‐layer extracorporeal ileocolic or colocolic anastomosis. This significant cost differential must be weighed carefully when considering the implementation of intracorporeal anastomosis programmes.

With the global 5‐year survival rate for colon cancer continuing to improve, increasing numbers of patients are reaching a follow‐up stage where late sequelae like IH become clinically relevant. Quality of life (QoL) is an essential metric in contemporary cancer care, and IH is known to adversely affect domains such as global health, social functioning and physical role. Furthermore, the presence of an IH is strongly linked to debilitating symptoms, including chronic discomfort, dyspnoea and sleep disturbances, highlighting the need for effective preventive surgical strategies.

In accordance with the European Hernia Society's recommendations, the use of slowly absorbable sutures is paramount for reducing IH risk. The poly‐4‐hydroxybutyrate suture utilised in this study is unique in maintaining over 50% of its tensile strength beyond 12 weeks. This study is among the first to examine specific off‐midline risk profiles in a cohort where the 4:1 suture‐to‐wound length ratio was uniformly applied as a standard of care. As this study utilises real‐world data, rather than data from a prospectively controlled trial, the bite sizes and intervals were not measured. This is a strength, as it increases the generalisability of the study, although it may also be a limitation that these measures were not available.

While incarceration remains the most severe complication of IH, its prevalence is generally low (1.7% to 3.7%). Our data showed that surgical repair was sought by only two patients (0.8%), a significant departure from historic Danish cohorts where repair rates often reached 72%. This shift is likely dual‐factorial: minimally invasive techniques produce smaller, frequently asymptomatic defects and there is a broader clinical transition towards non‐operative management. When incarceration risks are minimal, the potential for chronic pain after IH repair (reported at 16% to 20%) must be carefully balanced against the benefits of intervention.

When planning the study, we concluded that a retrospective survey to assess symptom type and severity would be of limited value. Such data would likely be compromised by recall bias and missing information from non‐responders or deceased patients. Furthermore, symptomatic patients with incisional hernias (IH) might not be referred for repair if their general practitioner perceives the surgical risks as outweighing the symptoms. Notably, both symptomatic and asymptomatic patients remain at risk of strangulation. Consequently, relying on the incidence of surgical repair carries a risk of underestimating the magnitude of the problem; therefore, using CT findings and clinically diagnosed IH is a more robust primary outcome.

Strengths and limitations

A primary strength of this work is the use of systematic CT verification, which avoids the significant underestimation of IH rates common in studies relying on diagnostic or procedural codes. However, radiologically detected hernias can sometimes result in an overestimation of the clinical burden. The single‐centre nature of the study ensures that all patients received standardised oncological and fascial closure techniques, though this inherently limits the total sample size and inevitably leads to lower generalisability, with results requiring validation in larger multicentre settings.

The retrospective analysis of prospectively gathered EHR data and the systematic reassessment of CT images remain limitations of this study. As the imaging was performed for oncological surveillance rather than hernia detection, abdominal straining was not routinely performed, potentially limiting the visualisation of subtle fascial defects. Additionally, while the medical reviewers were specifically trained, the lack of consensus on a universal CT definition for IH introduces a potential for inter‐observer variation.

CONCLUSION

This study indicates that specimen extraction via a supraumbilical muscle‐sparing transverse incision does not carry a greater risk of incisional hernia than the Pfannenstiel approach following minimally invasive surgery for colon cancer. As an IDEAL Stage 4 report, these results provide reassurance that both off‐midline approaches yield acceptable long‐term hernia risks when gold‐standard closure techniques are followed. These findings challenge the assumption that Pfannenstiel extraction sites are inherently superior for hernia prevention and, by extension, that intracorporeal anastomosis should be preferred on this basis. Further research should focus on hernia prevention in general and on the quality‐of‐life impact of radiologically identified, yet clinically occult, hernias in the era of minimally invasive surgery.

AUTHOR CONTRIBUTIONS

Maja Christine Rasmussen: Conceptualization; data curation; formal analysis; methodology; visualization; writing – original draft; writing – review and editing. Niklas Nygaard Baastrup: Data curation; methodology; writing – review and editing. Anna Sofie Friis Olsen: Data curation; writing – review and editing. Anders Kierkegaard Gundestrup: Data curation; validation; writing – review and editing. Charlotte Tiffanie Bendtz Kanstrup: Data curation; writing – review and editing. Jakob Kleif: Conceptualization; methodology; formal analysis; writing – review and editing. Claus Anders Bertelsen: Conceptualization; data curation; formal analysis; methodology; project administration; funding acquisition; supervision; validation; writing – review and editing.

CONFLICT OF INTEREST STATEMENT

The authors have no conflicts of interest.

DISCLOSURE

Previous representations: Poster presentation of preliminary results at the 17th Scientific and Annual Conference of the European Society of Coloproctology, Dublin, Ireland, 21–23 September 2022.

ETHICS STATEMENT

Data collection was authorised by the Regional Data Protection Agency (HIH‐2013‐031 I‐02468) and the Danish Patient Safety Authority (31‐1522‐51). In accordance with Danish regulations, approval from the Ethics Committee was not required.

Supporting information

Data S1. STROBE Statement—Checklist of items that should be included in reports of cohort studies.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.