Sage Journals: Discover world-class research

Abstract

Objectives

To report complications after colonoscopy and surgery in patients with neoplasia detected through a population based colorectal cancer (CRC) screening programme in the capital region of Sweden.

Methods

Patients who after a positive FOBT screening result underwent colonoscopy from 1 January 2008 to 30 June 2012 were included. Mortality and complications within 30 days after colonoscopy or subsequent surgery were identified through national registers, and complications were assessed through review of medical charts. Complications were graded using the Clavien-Dindo classification.

Results

After 2984 colonoscopies, the complication rate was 1%. The risk of post-polypectomy bleeding was 14/1000. The risk of perforation was 1/1000 after a diagnostic colonoscopy and 2.5/1000 after a colonoscopy with polypectomy. One patient developed a post-polypectomy syndrome. There was one death which was not related to the colonoscopy. After surgery for 37 adenomas and 155 CRCs, the total complication rates were 27% and 50%, respectively. The rate of anastomotic leakage was 13% and 12% after surgery for adenomas and CRC, respectively. There were no deaths after surgery. The overall complication rate after colonoscopy and surgery for adenomas and cancer was 4%.

Conclusions

Overall complication rates were acceptable and mortality low; however, the rate of anastomotic leakage after surgery for both adenomas and CRC was higher than expected.

Keywords

Colorectal cancer screening complications

Introduction

With an incidence of 447,000 new cases in Europe in 2012, colorectal cancer (CRC) is the second most common form of cancer in Europe, and the second most common cause of death from cancer (215,000 cases in 2012).¹ Since 2003, the European Council has recommended implementation of population-based primary screening programmes using faecal occult blood testing (FOBT) for men and women aged 50–74. Faecal immunochemical testing (FIT) is currently the recommended screening test.² Screening with endoscopy is a possible alternative to FOBT, and four large randomized controlled trials (RCT) assessing sigmoidoscopy as the primary screening method have demonstrated a reduction in distal CRC incidence and mortality of 41% and 61%, respectively, and RCTs evaluating colonoscopy are currently ongoing.^3,4 Regardless of whether FOBT or endoscopy is used for screening, colonoscopy plays a key role in the diagnostic evaluation of positive findings by other methods, and these findings may subsequently require intervention, either endoscopically or surgically. There is a vast literature on complications after colonoscopy,^5–9 and some studies on complications after screening colonoscopies, including 30-day morbidity and mortality.^9,10 However, the literature on the collected burden of complications from population-based screening programmes for CRC is limited.

Colonoscopy by itself, but also subsequent interventions following a positive colonoscopy, put the screening participants at risk of complications. This risk must be within reasonable limits, in particular following interventions designed to remove pre-malignant lesions, as the true risk of malignant transformation is unknown in each individual case.

In this study, positive FOBT screening results were followed by colonoscopy. We aimed to evaluate 30-day morbidity and mortality caused by colonoscopy, including both diagnostic and therapeutic procedures, and by subsequent surgery in case of cancer, suspected cancer, or in adenomas not suitable for endoscopic removal.

Methods

The programme and registers

In 2008, a population-based screening programme for CRC was launched in the capital region of Sweden, Stockholm, and by 2009 the island of Gotland was included. The region has a total of 2.3 million inhabitants. The programme is based on biennial screening with guaiac-based FOBT (Hemoccult®), and the target population is all residents aged 60–69. Two birth cohorts were successively included in the programme every year, and by 2014 all eligible birth cohorts had been included. The programme is centrally administered and does not involve primary care. Individuals invited to screening are mailed a test-kit, and a reminder after eight weeks if no tests are returned. Cases with positive test results (≥1 of three cards) are referred for a colonoscopy, to be performed within two weeks at designated centres. All data regarding the screening process, including detailed information on the colonoscopy procedure, findings, and histopathological diagnosis, are prospectively collected in the Stockholm-Gotland screening register. The results of the first five years of the programme have been previously reported.¹¹ From the screening register, data on all patients with a positive FOBT followed by a colonoscopy performed between 1 January 2008 and 30 June 2012 were retrieved.

The National Board of Health and Welfare, a Swedish government agency under the Ministry of Health and Social Affairs, administers a number of registries, including the Swedish National Inpatient Register (IPR) and the Cause of Death Register. Over 99% of all hospital discharges are recorded in the IPR, and previous validation has shown that 85–95 % of all diagnoses in the IPR are correct.¹² We defined complications within 30 days as events requiring over-night hospital care, and extracted data from the IPR regarding diagnoses and proceedings according to the International Classification of Diseases (ICD-10), length of stay, and hospital name. This was done for all in-patient care admissions from 2008 to 2012, on the patients in the screening register who underwent screening colonoscopy between 1 January 2008 and 30 June 2012, thus generating a minimum of six months of follow-up. In addition, for the same cohort of patients, data regarding date of death were extracted from the Cause of Death Register from 2008 to 2013.

The Swedish Colorectal Cancer Registry (SCRCR), a national quality registry established in 1995, prospectively and continuously collects data on all patients with adenocarcinomas of the colon and rectum. The SCRCR holds detailed information on preoperative staging, surgical procedures, and pathological findings, and on oncological treatment, complications, and recurrence within five years after surgery. The validity of registered data is high and SCRCR covers ≥98% of all CRCs in Sweden.^13,14

Complications and mortality

Data on all patients in the screening programme who, after their first positive FOBT, underwent a colonoscopy were matched with the IPR and the Cause of Death Register to find all in-care medical events and possible deaths within 30 days after screening colonoscopy. All adverse events deviating from a normal postoperative course were registered after scrutinizing the hospital charts. In patients with multiple complications, the most severe of these was registered according to the Clavien–Dindo classification of surgical complications.¹⁵

In order to detect complications and mortality within 30 days after surgery for adenomas, data on all patients in the screening register with one positive FOBT, and a subsequent colonoscopy, were matched to the IPR to find patients with relevant codes of surgical procedures according to the Swedish version of the Nordic Medico-Statistical Committee Classification of Surgical Procedures¹⁶ within six months after the screening colonoscopy. The following procedure codes were included: JFA00 – JFA98 (Local operations on intestine), JFB00 – JFB 97 (Partial excision of intestine), JFC00 – JFC 51 (Anastomosis of intestine), JFF00 – JFF97 (Exteriorization of intestine and creation of intestinal stomas), JFH 00 – JFH 96 (Total colectomy), JFW96 – JFW 98 (Other operations on intestine), JGA 00 – JGA98 (Proctotomy and local operations on rectum), JGB 00 – JGB97 (Excision of rectum), JGW00 – JGW98 (Other operations on rectum), and JHA00 – JHA30 (Incision, biopsy and excision of lesion of anus and perianal tissue). The limit of six months was selected to capture surgical procedures which may not have been performed promptly due to the lack of malignant diagnosis. Patients meeting these criteria, but who were also found in SCRCR, were not included in the adenoma group. Hospital charts of the remaining patients were reviewed for mortality and complications within 30 days after adenoma surgery. Complications were registered according to the Clavien–Dindo classification of surgical complications. All adverse events deviating from a normal postoperative course were registered, and in patients with multiple complications the most severe one was registered.

Data on all individuals in the screening register with one positive FOBT within the screening programme with a subsequent colonoscopy between 1 January 2008 and 30 June 2012 were matched to the SCRCR to identify patients with a histopathologically verified screening detected CRC. Among these patients, hospital charts of all those undergoing resectional surgery were reviewed for assessment of complications and mortality within 30 days after surgery for CRC.

The study was approved by the Regional Ethics Committee in Stockholm.

Statistical analysis

Fisher’s exact test was used to determine significant differences in complications according to tumour stage in patients undergoing surgery for CRC, and to determine differences in complications between prevalent and incident screening rounds. P-values < 0.05 were considered statistically significant. Stata 12.0 (StataCorp, 4905 Lakeway Drive, Collage Station, TX, USA) was used for the statistical analysis.

Results

Of 3522 screened individuals with a positive FOBT result, 2984 underwent colonoscopy, of whom 37 and 155 patients subsequently underwent surgery for screening detected adenomas and CRC, respectively. After matching data with the IPR, the Cause of Death register, the SCRCR, and with relevant procedure codes, the medical charts were further reviewed for assessment of complications and mortality within 30 days after the respective procedure. An overview of the complications after colonoscopy and surgery for screening detected adenomas and CRC is presented in Figure 1.

Figure 1.

Complications after colonoscopy and surgery for screening-detected adenomas and CRC. *Rate of post-polypectomy bleeding and perforation, respectively.

Complications and mortality after screening colonoscopy

The matching procedure identified 105 patients attending in-patient care within 30 days after screening-colonoscopy, 16 of whom were considered to have non-colonoscopy related causes (e.g. visits to a plastic surgeon, hand surgeon, psychiatrist, ophthalmologist, etc.) and were excluded from further analysis. A further 40 patients had no colonoscopy related complications, but were identified as they were subject to surgery for screening detected cancer, surgery for non-screening related causes, and re-colonoscopy because of incomplete bowel laxation at the first colonoscopy, or other causes. Data were missing for four patients.

Complications of any kind were found in 44 of 2984 patients (1 %), and one patient died within 30 days after the screening-colonoscopy (Table 1). All episodes of bleeding and perforation occurred after polypectomy, which was performed in 1204 patients undergoing screening colonoscopy. The risk of bleeding after polypectomy was 14/1000 colonoscopies. The overall risk of perforation was 1/1000 colonoscopies, and 2.5/1000 after a colonoscopy with polypectomy. Of the three patients with perforation, one patient was operated on with a right sided colectomy and a stoma, one was treated with a laparoscopic lavage, and the third was initially treated with a drainage, later operated on with lavage, and subsequently had multiple complications. One patient was re-admitted to hospital with fever and abdominal pain after polypectomy, but a CT-scan did not reveal any signs of perforation. This patient was defined as having a post-polypectomy syndrome.

Table 1.

Complications (n = 44) and mortality (n = 1), within 30 days after 2984 screening colonoscopies.

Bleeding, n = 18
Intervention; clips/adrenalin/other	11
Re-admission, no intervention	6
Bleeding gastric ulcer	1
Thromboembolic, n = 6
Stroke	4
Pulmonary embolism	1
Unstable angina pectoris (stent LAD-stenosis)	1
Infectious, n = 3
Fever, abdominal pain but CT neg, antibiotics (after polypectomy)	1
Infection in prosthetic knee	1
CT verified diverticulitis	1
Re-operations, n = 6
Perforation after polypectomy, laparoscopy and lavage	1
Perforation after polypectomy, right sided hemi colectomy	1
Perforation after polypectomy, multiple complications, stoma	1
Fever, abdominal pain, laparoscopy; carcinosis	1
Colonic obstruction, colostomy	1
Inguinal hernia with perforation in diverticulum	1
Re-admissions, miscellaneous, n = 11
Fall, anaemia, gastroscopy neg.	1
Headache, nausea, vomiting CT neg.	2
Joint pain (+ anaemia)	2
Dyspnoea, CT neg.	1
Cardio-/thoracic symptoms, investigation neg.	2
Abdominal pain, CT neg, no intervention	1
Social reasons	1
Migraine	1
Mortality	1

Additional morbidity following screening colonoscopy is presented in Table 1, including four patients with stroke. Two of these patients were on anticoagulants, and one had temporarily omitted anticoagulation therapy prior to colonoscopy. One patient died within 30 days after colonoscopy of non-colonoscopy related causes. This patient was re-admitted because of jaundice and a GI-bleeding caused by an esophagitis and further investigations diagnosed carcinomatosis with multiple hepatic metastases from a neuroendocrine cancer.

Complications and mortality after surgery for adenomas

Fifty-six patients had surgical procedures registered within six months of the screening colonoscopy, but 14 patients only had elective colonoscopies. Five patients underwent surgical procedures unrelated to adenomas. Of these, three were operated on because of tumours in the distal ileum, one for diverticulosis, and one for a perianal abscess. The remaining 37 patients underwent surgery for adenomas motivated by findings at the screening colonoscopy.

The overall complication rate in the adenoma-surgery group was 10/37 (27%) and there was no mortality (Table 2). In patients operated on with an anastomosis (23 patients with 24 anastomoses), anastomotic leakage was detected in three patients (13% – all re-operated), two with a diverting loop-ileostomy and one with laparoscopic lavage and drainage. One patient undergoing transanal micro surgery underwent re-operation because of a post-operative haemorrhage, and one patient underwent re-operation because of a wound dehiscence.

Table 2.

Complications (n = 10), according to the Clavien–Dindo classification, within 30 days after surgery, in 37 patients with screening-detected adenomas.

Grade I, n = 2
Vomiting	1
Transient incontinence	1
Grade II, n = 2
Wound infection, antibiotics	1
Urinary tract infection, antibiotics	1
Grade III, n = 5
Wound dehiscence; re-op	1
Anastomotic leakage; re-op loop-ileostomy	2
Anastomotic leakage, laparoscopic lavage and drain	1
Abscess in rectal stump; trans rectal drainage	1
Grade IV, n = 1
Bleeding, ICU, hemodynamically instable, re-op	1

Note: No deaths occurred.

Complications and mortality after surgery for CRC

The screening register and SCRCR identified 158 patients undergoing surgery for screening detected CRC, and hospital charts were reviewed, except in three patients with missing data. All complications according to Clavien–Dindo are presented in Table 3. There were 77/155 complications (50%), and no 30-day mortality. The number of complications ≥ Grade II was 58/155 (37%). Anastomotic leakage was detected in 16 of 138 patients with an anastomosis (12%), eight of whom underwent re-operation. The remaining eight patients were treated conservatively with drainage (transrectally/endoscopically/radiologically) and antibiotics (n = 4) or antibiotics only (n = 4). There was no statistically significant correlation between the total number of complications and the TNM stage (p = 0.8). Comparing the prevalent and incident screening rounds, a numerical difference was seen in the overall complication rate, with 54% (61/113) and 38% (16/42) complication rated in the prevalent and incident screening rounds, respectively. However, this difference was not statistically significant (p = 0.4).

Table 3.

Complications (n = 77), according to the Clavien–Dindo classification, within 30 days after surgery, in 155 patients with screening-detected colorectal cancer.

Grade I, n = 19
Intestinal failure and electrolyte imbalance	5
Various wound problems	7
Rectal bleeding or anaemia, no transfusion	2
Remaining urinary catheter at discharge	3
Remaining drainage after bile-leakage at discharge (synchronous op of liver metastases)	1
Vitreous detachment	1
Grade II, n = 34
Infectious complications, n = 27
Elevated CRP, antibiotics, no certain leakage	9
Anastomotic leakage/abscess, antibiotics	4
Pneumonia, antibiotics	3
Wound infection, antibiotics	8
UTI, antibiotics	3
Blood transfusion	1
Atrial fibrillation	2
Deep venous thrombosis	1
Re-admission: intestinal failure/stoma problems/ abdominal pain	3
Grade III, n = 19
Anastomotic leakage; re-op colostomy	4
Anastomotic leakage; re-op laparoscopic drainage	1
Anastomotic leakage; endosponge	1
Probable anastomotic leakage: trans rectal/endoscopic/ radiological drainage, antibiotics	3
Wound dehiscence; re-op	4
Bile-leakage; re-admission, ERCP, stent	1
Drainage insufficiency; re-op	1
Inguinal hernia with incarceration; re-op	1
Bowel obstruction; re-op ileostomy	1
TIA	1
Bleeding after suprapubic catheter	1
Grade IV, n = 5
Anastomotic leakage; re-op, sepsis, ICU	2
Anastomotic leakage; sepsis, ICU, cardiac arrest, ECMO, loop-ileostomy	1
Wound dehiscence, re-op, multiple complications	2

Note: No deaths occurred.

Discussion

The present study assessed the collected complication rates following colonoscopy work-up and subsequent surgery in a population based CRC screening programme with FOBT. The complication rates were 1% after colonoscopy, 27% following surgery for pre-malignant lesions, and 50% following surgery for CRC. Although significant, these rates are considered to be within reasonable limits compared with the literature. However, the rate of anastomotic leakage was relatively high in both the adenoma- and CRC-surgery group.

The detected rates of post-polypectomy bleeding, and the risk of perforation in colonoscopies, with and without polypectomy, were found to be in concordance with the quality indicator guidelines for endoscopic procedures according to the American Society for Gastrointestinal Endoscopy.¹⁷ There was one death within 30 days after colonoscopy, but review of the medical charts found no obvious correlation to the index endoscopy itself. The incidence of stroke following colonoscopy was 1.3/1000 in this screening series. Ko et al.¹⁰ studied almost 19,000 patients (age ≥ 40) and reported a 30-day post colonoscopy incidence of 0.33/1000 in stroke or transient ischaemic attack, and an expected adjusted rate of stroke 0.48 per 1000 over 30 days. The present study is smaller and includes older patients, which makes direct comparison difficult, but data highlight that a risk of severe morbidity appears to be associated with the screening colonoscopy itself. Apart from post-polypectomy bleeding and perforation, it is, however, difficult to establish a definitive relationship between the colonoscopy and any subsequent morbidity although it may appear plausible.

The total complication rate after resection surgery for adenomas was less than that after surgery for CRC, but the rate of anastomotic leakage was 13%, which was higher than expected. According to Swedish national quality data (2009–2013), anastomotic leakage rates after CRC were 8% and 4%, and overall postoperative complication rates (Clavien–Dindo ≥ grade 2) were around 40% and 30% after surgery for rectal and colon cancer, respectively.^13,14 In benign surgery, the rate could be expected to be in the same range, or lower. In the present study, anastomotic leakage rates were equivalent in the adenoma and cancer surgery groups. However, the number of patients with anastomotic leakage after adenoma surgery was small, and a larger study group would be needed before drawing firm conclusions on prevalence or underlying causes of anastomotic leakage in this group of patients. A high adenoma detection rate (ADR) is considered a quality measure of colonoscopy, and has been shown to reduce the rate of interval cancers.¹⁸ However, a high ADR will generate complications imposed by the endoscopic and surgical efforts taken to remove these adenomas, which have an unknown natural history if left without intervention. Removal of adenomas without obvious risk of malignancy could increase the morbidity imposed by screening, and hence the risk of overtreatment within screening for CRC should be considered.

The overall complication rate was 50% after CRC surgery, and 37% if Grade I was excluded. Rates and assessment of complications vary in the literature, but in rectal cancer surgery total complication rates of 50% have been reported.^19,20 Previous studies have demonstrated an anastomotic leakage frequency of 7–12% for the rectum and 3% in the colon.^20,21 The frequency of anastomotic leakage for both colon and rectum in the present study was 12%, without correlation between serious complications and tumour stage. The validity of data in the present study (i.e. operation procedure and diagnosis codes) relies on the reporting accuracy to the IPR, which is high due to economic reimbursement systems, and one of the strengths of the study is that all complications were also assessed through in-detail review of medical charts, generating robust data. The definition of anastomotic leakage was very broad, including all events with any suspicion of this condition.

The benefit of screening for colorectal cancer has long been proven, however, screening investigations (i.e. colonoscopy) as well as endoscopical and surgical treatment of screening detected disease will generate morbidity, and even mortality. Standards for screening colonoscopy are well established, but the chain of events after a positive screening test does not end with colonoscopy. Therefore, monitoring and assessment of morbidity and mortality after surgery for screening detected adenomas and cancer is of equal importance, to assure high-quality medical service through all stages of screening investigations and treatment. In our study, the total complication rate after colonoscopy and surgery for adenomas and cancer was 4%. The complication rates after colonoscopy and surgery were largely in concordance with findings from previous studies. Accordingly, these results confirm that the total complication rate after investigations and treatment of screening detected disease is acceptable, and mortality is low.

Conclusion

This study addresses the total morbidity and mortality after colonoscopy and surgery for adenomas and cancer imposed by CRC screening. Overall complications rates were acceptable, and mortality low. However, the rate of anastomotic leakage after surgery for both adenomas and CRC was higher than expected. The risk of complications associated with investigations and treatment of screening detected disease should be considered, monitored, and continuously evaluated.

Footnotes

Acknowledgements

Contributing with extraction of register data: Robert Johansson, MSc, Regional Cancer Centre North, Department of Radiation Science, Oncology, Umeå University, Sweden. Sini Kilpeläinen, Statistician, Regional Cancer Centre Stockholm Gotland, Sweden. Tongplaew Singnomklao, Data manager, Swedish Colorectal Cancer Registry, Regional Cancer Centre Stockholm Gotland, Sweden.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financial support was provided through the Regional Agreement on Medical Training and Clinical Research (ALF) between the Stockholm Community Council and Karolinska Institutet, and also by the Swedish Cancer Society and the Stockholm Cancer Society.

References

Ferlay

Steliarova-Foucher

Lortet-Tieulent

. Cancer incidence and mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer 2013; 49: 1374–1403.

Council of the European Union. Council Recommendation of 2 December 2003 on cancer screening. 2003.

Brenner

Stock

Hoffmeister

. Effect of screening sigmoidoscopy and screening colonoscopy on colorectal cancer incidence and mortality: systematic review and meta-analysis of randomised controlled trials and observational studies. BMJ 2014; 348: g2467–g2467.

Kaminski

Bretthauer

Zauber

. The NordICC Study: rationale and design of a randomized trial on colonoscopy screening for colorectal cancer. Endoscopy 2012; 44: 695–702.

de Jonge

Nicolaas

van Baalen

. The incidence of 30-day adverse events after colonoscopy among outpatients in the Netherlands. Am J Gastroenterol 2012; 107: 878–884.

Day

Kwon

Inadomi

. Adverse events in older patients undergoing colonoscopy: a systematic review and meta-analysis. Gastrointest Endosc 2011; 74: 885–896.

Fisher

Maple

Ben-Menachem

. Complications of colonoscopy. Gastrointest Endosc 2011; 74: 745–752.

Rabeneck

Paszat

Hilsden

. Bleeding and perforation after outpatient colonoscopy and their risk factors in usual clinical practice. Gastroenterology 2008; 135: 1899–1906.

Stock

Ihle

Sieg

. Adverse events requiring hospitalization within 30 days after outpatient screening and nonscreening colonoscopies. Gastrointest Endosc 2013; 77: 419–429.

10.

Riffle

Michaels

. Serious complications within 30 days of screening and surveillance colonoscopy are uncommon. Clin Gastroenterol Hepatol 2010; 8: 1100–1100.

11.

Blom

Kilpelainen

Hultcrantz

. Five-year experience of organized colorectal cancer screening in a Swedish population – increased compliance with age, female gender, and subsequent screening round. J Med Screen 2014; 21: 144–150.

12.

Ludvigsson

Andersson

Ekbom

. External review and validation of the Swedish national inpatient register. BMC Public Health 2011; 11: 450–450.

13.

Swedish Colorectal Cancer Registry, Rectal Cancer 2013, http://www.cancercentrum.se/Global/Diagnoser/kolorektal/rapporter/rekti2013_20150129.pdf (2013).

14.

Swedish Colorectal Cancer Registry, Colon Cancer 2013, http://www.cancercentrum.se/Global/Diagnoser/kolorektal/rapporter/colon2013.pdf (2013).

15.

Dindo

Demartines

Clavien

. Classification of surgical complications – A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240: 205–213.

16.

NOMESCO. Classification of Surgical Procedures (NCSP), version 1.16. Nordic Centre for Classifications in Health Care, WHO Collaborating Centre for the Family of International Classifications in the Nordic Countries 2011.

17.

Rex

Schoenfeld

Cohen

. Quality indicators for colonoscopy. Gastrointest Endosc 2015; 81: 31–53.

18.

Kaminski

Regula

Kraszewska

. Quality indicators for colonoscopy and the risk of interval cancer. N Engl J Med 2010; 362: 1795–1803.

19.

Ballian

Yamane

Leverson

. Body mass index does not affect postoperative morbidity and oncologic outcomes of total mesorectal excision for rectal adenocarcinoma. Ann Surg Oncol 2010; 17: 1606–1613.

20.

Guillou PJ, Quirke P, Thorpe H, et al. Short-term endpoints of conventional versus laparoscopic-assisted surgery in patients with colorectal cancer (MRC CLASICC trial): multicentre, randomised controlled trial. Lancet 2005; 365: 1718–1726.

21.

Peeters

Tollenaar

Marijnen

CAM

. Risk factors for anastomotic failure after total mesorectal excision of rectal cancer. Br J Surg 2005; 92: 211–216.

Complications after colonoscopy and surgery in a population-based colorectal cancer screening programme

Abstract

Objectives

Methods

Results

Conclusions

Keywords

Introduction

Methods

The programme and registers

Complications and mortality

Statistical analysis

Results

Complications and mortality after screening colonoscopy

Complications and mortality after surgery for adenomas

Complications and mortality after surgery for CRC

Discussion

Conclusion

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

References