Rectal washout during abdominoperineal resection for rectal cancer has no impact on the oncological outcome

Intraoperative rectal washout is performed to eliminate exfoliated intraluminal cancer cells and thereby decrease the risk of local recurrence. Rectal washout in abdominoperineal resection has not been studied. The aim of this study was to assess the oncological outcome after rectal washout in abdominoperineal resection for rectal cancer and to find evidence as to whether rectal washout should be performed or not.

Abdominoperineal resection is most often reserved for cancers located in the lower third of the rectum ≤5 cm from the anal verge [10]. In this area the mesorectum tapers, making it technically difficult to achieve a clear circumferential resection margin. The risk of intraoperative perforation is also increased, translating to poorer oncological outcomes, including higher rates of LR and impaired overall and disease-free survival [10][11][12][13][14]. Therefore, it is important to explore any possible impact of RW on oncological outcomes in APR. The Swedish national guidelines for rectal cancer care recommend RW in anterior resection but also advocate it at the discretion of the surgeon in APR [10]. RW is usually performed with the lumen occluded distal to the tumour. For low rectal cancers this is technically not always possible, and the alternative is to wash out the rectal lumen without an occlusive clamp before placement of a purse-string suture. The hypothesis behind RW in APR is elimination of intraluminal cancer cells that otherwise would have been left behind in the distal rectal stump and thereby increase the risk of LR if intraoperative perforation or leakage from the purse-string suture occur.
This registry study aims to assess the oncological outcome in terms of LR, distant metastasis (DM), overall recurrence (OAR) and overall and relative survival after RW in APR for rectal cancer and to find evidence for whether RW should be performed in APR or not.

Swedish Colorectal Cancer Registry and study population
The study population comprised all patients registered in the Swedish Colorectal Cancer Registry (SCRCR) who underwent elective surgery with APR for rectal cancer (TNM Stages I-III) between 2007 and 2013. Data on tumour characteristics and demographics and preoperative, perioperative, postoperative and follow-up data were collected. The SCRCR is a national population-based registry and has a coverage ratio of 99.7% [15]. Primary data are reported 30 days after surgery or at diagnosis if no surgery is performed and follow-up data are registered 3 and 5 years postoperatively. The SCRCR has been described in detail in other publications [15,16].

Definitions
Rectal cancer is defined as an adenocarcinoma with its lower edge located within 15 cm of the anal verge as measured with a rigid sigmoidoscope during withdrawal.
Hospital volume is defined according to the number of APRs for rectal cancer performed annually. A volume of 1-10 procedures is referred to as low, 11-25 as medium and ≥26 as high.
A colorectal surgeon is defined as an accredited colorectal surgeon or a surgeon with a colorectal interest.
The definition of an intraoperative perforation is an unintentional perforation of the rectum that occurs during surgery.
R0 is defined as a locally radical procedure with neither macroscopic tumour tissue left behind according to the surgeon nor microscopic tumour tissue at the resection margins according to the pathologist.
Local recurrence is defined as tumour growth located below the level of the promontory related to the primary rectal tumour. Tumour tissue in the ovary, liver, peritoneum, bone, lung, brain or any other organ as well as in any lymph node that is not located in the pelvis is defined as DM. LR and DM are registered regardless of how the diagnosis was made (e.g. clinical, radiological, pathological or endoscopic examination). OAR includes either isolated LR or DM, or both LR and DM.

Statistical analysis
Categorical data are presented as numbers with percentages and continuous data are presented as median with interquartile range.
The chi-square test, Fisher's exact test and the independent sample t-test were used to compare groups when appropriate. Patients were followed for 5 years postoperatively. Survival analyses were performed. Kaplan-Meier with log-rank test and univariable and multivariable Cox regression analysis were used. In multivariable analysis, clinically relevant variables considered as potential confounders for LR, DM, OAR and overall and relative survival (i.e. age, gender, TNM stage, tumour height, neoadjuvant radiotherapy, neoadjuvant chemotherapy, intraoperative perforation, adjuvant chemotherapy) were included in the model. Surgical competence was not included in multivariable analysis because only a few cases were treated by a general surgeon and all occurred in one group. Relative survival was calculated with the R package relsurv and the Andersen multiplicative model [17]. Population life tables from the Human Life-

Study population
A total of 2425 patients with R0 resection, no LR, DM or death within 90 days postoperatively, valid 5-year follow-up and available RW data were grouped depending on whether RW was performed

What does this paper add to the literature?
This is the first study to investigate the oncological outcome after rectal washout during abdominoperineal resection for rectal cancer in order to find evidence as to whether rectal washout should be performed or not.
or not (Figure 1). The groups differed significantly with regard to tumour height, TNM stage, hospital volume, laparoscopic surgery and intraoperative perforation (Table 1).

Recurrence
As shown in Table 2 (Table 3). Furthermore, subgroup analysis of the patients where intraoperative perforation occurred (n = 133), showed no differences in rates of LR, DM and OAR between the RW and no RW groups (Appendix S1 in the Supporting Information).

Survival
Overall and relative survival are presented in Figure 2. The 5year overall survival in the RW group was 0.72 compared with 0.72 in the no RW group (p = 1.00). The 5-year relative survival in the RW and the no RW group was 0.87 and 0.85, respectively (p = 0.67).

Univariable and multivariable analysis
In univariable and multivariable analysis, RW did not have a significant impact on the oncological outcome in terms of LR, DM, OAR and 5-year overall and relative survival ( Table 4). The number of confounders that could be included in the multivariable analysis was limited owning to the small number of LR events.
Multivariable analysis adjusted for laparoscopic surgery and hospital volume was performed but did not affect the results. Thus, these are not presented.

DISCUSS ION
Whether or not RW was performed during APR for rectal cancer had no effect on oncological outcomes. To our knowledge, this is the first study to investigate the impact of RW in APR for rectal cancer.  water and ethanol [20,24,25]. Recent studies recommend a washout volume of >1500-2000 ml [24]. RW with dilute Betadine in APR for the purpose of removing residual stool has been described by Perry et al. [26], and a study of RW before colorectal anastomosis showed a reduction in bacterial counts in the rectal stump when sodium hypochlorite and povidone-iodine were used [27]. Further studies are necessary to investigate other possible benefits of RW before the procedure is completely dismissed in APR. From our data we cannot exclude the possibility that RW may be important in other aspects, such as reducing perineal infections.
To obtain the 5-year oncological outcome this study included patients undergoing APR between 2007 and 2013. Since then, neoadjuvant therapy has improved and is in transition due to the results of the RAPIDO trial where neoadjuvant short-course radiotherapy followed by chemotherapy was given [28]. Also, minimally invasive surgery has continued to increase and is currently the surgical technique of choice in over 50% of rectal cancer resections in Sweden [15]. In Sweden, Denmark, Australia, New Zealand and the United States, APR has been reported to account for 19%-38% of rectal cancer resections [14,15,29,30]. In our previous survey of the cur- The best way to perform RW is still unknown. As mentioned by Kodeda et al. [31], to resolve this question negative and conflicting results also need to be published, and clinical studies investigating details on how to perform RW should continue to be conducted.
Large international multicentre studies will be required to achieve evidence-based guidelines on the role of RW in rectal cancer surgery, including indications, the most effective washout fluid and volume in terms of reducing the risk of LR.
In conclusion, our data do not support routine RW in APR in order to improve oncological outcomes.

ACK N OWLED G EM ENTS
The authors thank the SCRCR for providing data and Helene Jacobsson (Clinical Studies Sweden, Forum South, Skåne University Hospital) for statistical support.

CO N FLI C T O F I NTE R E S T S
The authors declare that they have no conflicts of interest.

AUTH O R CO NTR I B UTI O N
FJ and PB were responsible for the concept and the design of the study. RSN collected and analysed the data. RSN wrote the first draft of the manuscript. All authors contributed to the interpretation of the data and to the critical revision of the article. All authors read and approved the final manuscript.

E TH I C S A PPROVA L
This study was approved by the Swedish Ethical Review Authority (2018/1040). The Declaration of Helsinki guidelines were followed.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available on request from the authors.