doi:10.1111/codi.12988
Original article
Straight to flexible sigmoidoscopy: rationalization of 2-week wait referrals in suspected colorectal cancer D. G. Couch*, J. H. Murphy†, K. M. Boyle* and D. M. Hemingway* *Department of Colorectal Surgery, Leicester Royal Infirmary, Leicester, UK and †Department of Colorectal Surgery, Derby Royal Hospital, Derby, UK Received 18 November 2014; accepted 12 March 2015; Accepted Article online 6 May 2015
Abstract Aim The 2-week wait pathway was designed to decrease the time from presentation to primary care of patients with ‘red flag’ symptoms of suspected cancer for review by a specialist for the diagnosis or exclusion of cancer. In our tertiary referral centre we have found that 968 colonoscopies per year are required to satisfy the demand for the 2-week wait, leading to limited colonoscopy availability for other services. We sought to determine the yield of colorectal cancer found at colonoscopy referred via the 2week wait and referenced to the original red flag symptoms. This was in order to select the most efficacious alternative primary investigation based upon presenting symptoms. Method Electronic records were retrospectively analysed. All patients who went through the 2-week wait for suspicion of colorectal cancer in 2013 and were found to have colorectal cancer on colonoscopy were included. Patients not undergoing colonoscopy as the first investigation were excluded. The splenic flexure was deemed to be within the range of a flexible sigmoidoscope.
Introduction The 2-week wait pathway introduced nationally in the UK in December 2000 is designed to decrease the time from presentation to primary care with ‘red flag’ symptoms of suspected cancer, to allow review by a specialist for the diagnosis or exclusion of cancer, with the underlying goal of improving cancer survival [1]. In the context of colorectal cancer, nationally issued guidance advises colonoscopy to be the investigation of choice. Barium enema or flexible sigmoidoscopy should be Correspondence to: Mr Daniel Couch, Leicester Royal Infirmary, Infirmary Square, Leicester LE1 5WW, UK. E-mail: [email protected]
980
Results In all, 2950 referrals were made. 968 colonoscopies were performed as the primary investigation of which 35 were found to have colorectal cancer. No patients referred with rectal bleeding and another symptom had a tumour more proximal to the range of flexible sigmoidoscopy. 80% of tumours proximal to the splenic flexure were suitable for CT diagnosis alone. Conclusion Our data support the use of flexible sigmoidoscopy alone as an initial investigation for patients presenting with rectal bleeding with or without additional colorectal symptoms. Patients with anaemia (without bleeding) or change in bowel habit (without bleeding) may be investigated with CT colonography alone; colonoscopy may then be used selectively prior to surgery. Keywords flexible sigmoidoscopy, 2-week wait, endoscopy, colorectal cancer What does this paper add to the literature? The paper assesses the outcome of primary investigation of patients referred within the 2-week wait pathway aimed to exclude colorectal cancer when endoscopic services are limited.
reserved for patients unable to tolerate full colonoscopy. CT colonography has been advised as an alternative investigation to colonoscopy where a centre has sufficient expertise [2]. The 2-week wait system places a heavy burden on surgical outpatient and endoscopy services [3]. Innovations aiming to tackle this burden have been introduced with pathways such as ‘straight to test’ from primary care bypassing outpatients, but the problem of high volumes of colonoscopy requests may not have been resolved [4]. In our tertiary referral centre almost 1000 colonoscopies were carried out in 2013 to satisfy the demand for the 2-week wait at the expense of other service delivery. This demand for colonoscopy is outstripping
Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 17, 980–983
D. G. Couch et al.
Straight to flexible sigmoidoscopy
our centre’s ability to investigate all patients without delaying investigations for patients in other pathways such as bowel cancer follow-up. In the present study we set out to determine the yield of colorectal cancer found during colonoscopy in patients with red flag symptoms with the aim of selecting the most effective primary investigation.
Method Data were collected retrospectively from the Leicester Colorectal Special Interest Group electronic database. Patients diagnosed with colorectal cancer during 2013 were identified, and those not diagnosed with cancer who followed the lower gastrointestinal 2-week wait pathway were excluded. Patients who underwent colonoscopy as the first investigation were included. Patients undergoing any other test before colonoscopy were excluded. Patients undergoing flexible sigmoidoscopy were also excluded. Remaining patients’ records were then examined to identify the symptoms of initial referral to the 2-week wait scheme. Lesions identified at or distal to the splenic flexure were assumed to have been within reach of the flexible sigmoidoscope.
Results During 2013, 2950 2-week wait referrals were made from primary care. Of these, 968 colonoscopies were performed as a primary investigation. Thirty-five (3.6%) patients were found to have a colorectal cancer during colonoscopy (Table 1). Table 2 gives the tumour site in patients referred with anaemia alone. Six of the 15 cancers were in the caecum, five in the ascending colon, one at the hepatic flexure, two at the transverse colon and one at the splenic flexure. Table 3 details tumour site in patients referred with change in bowel habit
alone who underwent colonoscopy. Two of the 10 cancers were in the caecum, two in the descending colon, three in the sigmoid colon, one at the rectosigmoid junction and two in the rectum.
Discussion Since its introduction the 2-week wait pathway has been demonstrated to have a low rate of detection of colorectal cancer. This finding is supported by the data in the present study and by reports from other centres [5]. Furthermore the pathway has not been shown to improve survival from colorectal cancer as was originally hoped [6], but despite the lack of evidence for any value of the strategy the volume of 2-week referrals has continued to rise to a point where it is inhibiting the ability of secondary and tertiary care to provide services to other patients. In line with published guidance, flexible sigmoidoscopy has been used in our centre as the investigation of choice for rectal bleeding as the only symptom through the 2-week wait pathway. This guidance has also submitted patients with any change in bowel habit or anaemia to colonoscopy unless contraindicated. There is a grey area for those patients presenting with bleeding and symptoms such as change in bowel habit or anaemia suggesting the possibility of more proximal pathology. Until recently in our centre these have had colonoscopy as the first investigation in order that lesions above the splenic flexure would not be missed, but the demand for colonoscopy has prevented us from delivering this investigation for patients outside of the pathway such as those requiring assessment of colitis, bowel cancer surveillance or the investigation of CTidentified abnormalities. Compared with colonoscopy in our centre, CT colonography has been under-used in patients referred through the 2-week wait.
Table 1 Site and number of colorectal tumours found at colonoscopy as per presenting symptom. Presenting symptom as per 2-week wait referral
Tumour site proximal to splenic flexure
Tumour site distal to splenic flexure
Total
Anaemia Change in bowel habit Change in bowel habit and anaemia Visible rectal bleeding and anaemia Visible rectal bleeding and change in bowel habit Visible rectal bleeding and change in bowel habit and anaemia
14 2 1 (ascending colon) 0 0
1 8 0 1 6
15 10 1 1 6
2
2
0
35
Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 17, 980–983
981
D. G. Couch et al.
Straight to flexible sigmoidoscopy
Table 2 Tumour sites reported during colonoscopy of 2-week wait patients referred with anaemia alone. Tumour site
n
Caecum Ascending colon Hepatic flexure Transverse colon Splenic flexure
6 5 1 2 1
Table 3 Tumour sites at colonoscopy of 2-week wait patients referred with change in bowel habit alone. Tumour site
n
Caecum Descending colon Sigmoid colon Rectosigmoid junction Rectum
2 2 3 1 2
Resource limitations have not allowed us to expand the provision of our endoscopic service further and we have therefore been obliged to limit the use of colonoscopy within the 2-week wait pathway. In its place, through planned expansion of gastrointestinal radiology expertise and capacity, we now intend to use CT colonography and flexible sigmoidoscopy as far as is possible for 2-week wait patients without compromising diagnostic accuracy. By working back from cancer diagnosis to the symptom complex at the time of the original referral we have determined that within the 2013 patient cohort patients who presented with rectal bleeding and either a change in bowel habit or anaemia in addition were all found to have colorectal cancer within the range of a fibre-optic sigmoidoscope. This suggests that, within the patient population in the present study, patients with these symptoms had ‘distal’ rather than ‘proximal’ symptoms. As a result of this data analysis, we have changed the 2-week wait referral protocol to limit initial investigation to flexible sigmoidoscopy in any patient presenting with a symptom complex including visible rectal bleeding. In doing so the data indicate that the risk of missing a more proximal cancer is negligible. In practice this ‘straight to test’ protocol brings the patient directly to endoscopy on receipt of the referral letter. This preinvestigation consultation ensures that the referral is justified and appropriate. Should it be felt by the reviewing clinician that further tests be performed should the ini-
982
tial flexible sigmoidoscopy be normal, for example where there is a history of profound anaemia in the presence of negligible anal bleeding, a further CT colonogram can be arranged. This change in protocol has been ratified by the Leicester Colorectal Specialist Interest Group (a multidisciplinary group of clinicians interested in colorectal disease) and the University Hospitals of Leicester Cancer Board. Furthermore, the present study has also shown that 20% of cancer patients originally referred with change in bowel habit had a proximal lesion, the remainder being at or distal to the splenic flexure. In 80% of patients presenting with anaemia, the tumour was at or proximal to the hepatic flexure. With this evidence, our centre now provides CT colonography alone for patients presenting with anaemia or change in bowel habit because of its increased availability, its ability to survey other intra-abdominal structures and because it allows substitution of bowel preparation for water-soluble contrast medium and faecal tagging. A possible objection to our changes in handling 2week wait referrals could be that, in the shift towards flexible sigmoidoscopy for patients with any rectal bleeding regardless of change in bowel habit or anaemia, proximal cancers will be missed. This study did not identify any patient who presented with these symptoms in whom a cancer proximal to the splenic flexure was found. In addition to our evidence there are other studies that support the use of flexible sigmoidoscopy in this context. One large recent investigation which followed 1690 patients over a median of 35 months found that, when predominantly left-sided symptoms were investigated with flexible sigmoidoscopy, only two (0.24%) were subsequently found to have a more proximal colonic lesion [7]. In a larger series it was found that when patients were referred through the 2-week wait with bleeding, change in bowel habit or abdominal pain (without iron deficiency anaemia),
Original article
Straight to flexible sigmoidoscopy: rationalization of 2-week wait referrals in suspected colorectal cancer D. G. Couch*, J. H. Murphy†, K. M. Boyle* and D. M. Hemingway* *Department of Colorectal Surgery, Leicester Royal Infirmary, Leicester, UK and †Department of Colorectal Surgery, Derby Royal Hospital, Derby, UK Received 18 November 2014; accepted 12 March 2015; Accepted Article online 6 May 2015
Abstract Aim The 2-week wait pathway was designed to decrease the time from presentation to primary care of patients with ‘red flag’ symptoms of suspected cancer for review by a specialist for the diagnosis or exclusion of cancer. In our tertiary referral centre we have found that 968 colonoscopies per year are required to satisfy the demand for the 2-week wait, leading to limited colonoscopy availability for other services. We sought to determine the yield of colorectal cancer found at colonoscopy referred via the 2week wait and referenced to the original red flag symptoms. This was in order to select the most efficacious alternative primary investigation based upon presenting symptoms. Method Electronic records were retrospectively analysed. All patients who went through the 2-week wait for suspicion of colorectal cancer in 2013 and were found to have colorectal cancer on colonoscopy were included. Patients not undergoing colonoscopy as the first investigation were excluded. The splenic flexure was deemed to be within the range of a flexible sigmoidoscope.
Introduction The 2-week wait pathway introduced nationally in the UK in December 2000 is designed to decrease the time from presentation to primary care with ‘red flag’ symptoms of suspected cancer, to allow review by a specialist for the diagnosis or exclusion of cancer, with the underlying goal of improving cancer survival [1]. In the context of colorectal cancer, nationally issued guidance advises colonoscopy to be the investigation of choice. Barium enema or flexible sigmoidoscopy should be Correspondence to: Mr Daniel Couch, Leicester Royal Infirmary, Infirmary Square, Leicester LE1 5WW, UK. E-mail: [email protected]
980
Results In all, 2950 referrals were made. 968 colonoscopies were performed as the primary investigation of which 35 were found to have colorectal cancer. No patients referred with rectal bleeding and another symptom had a tumour more proximal to the range of flexible sigmoidoscopy. 80% of tumours proximal to the splenic flexure were suitable for CT diagnosis alone. Conclusion Our data support the use of flexible sigmoidoscopy alone as an initial investigation for patients presenting with rectal bleeding with or without additional colorectal symptoms. Patients with anaemia (without bleeding) or change in bowel habit (without bleeding) may be investigated with CT colonography alone; colonoscopy may then be used selectively prior to surgery. Keywords flexible sigmoidoscopy, 2-week wait, endoscopy, colorectal cancer What does this paper add to the literature? The paper assesses the outcome of primary investigation of patients referred within the 2-week wait pathway aimed to exclude colorectal cancer when endoscopic services are limited.
reserved for patients unable to tolerate full colonoscopy. CT colonography has been advised as an alternative investigation to colonoscopy where a centre has sufficient expertise [2]. The 2-week wait system places a heavy burden on surgical outpatient and endoscopy services [3]. Innovations aiming to tackle this burden have been introduced with pathways such as ‘straight to test’ from primary care bypassing outpatients, but the problem of high volumes of colonoscopy requests may not have been resolved [4]. In our tertiary referral centre almost 1000 colonoscopies were carried out in 2013 to satisfy the demand for the 2-week wait at the expense of other service delivery. This demand for colonoscopy is outstripping
Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 17, 980–983
D. G. Couch et al.
Straight to flexible sigmoidoscopy
our centre’s ability to investigate all patients without delaying investigations for patients in other pathways such as bowel cancer follow-up. In the present study we set out to determine the yield of colorectal cancer found during colonoscopy in patients with red flag symptoms with the aim of selecting the most effective primary investigation.
Method Data were collected retrospectively from the Leicester Colorectal Special Interest Group electronic database. Patients diagnosed with colorectal cancer during 2013 were identified, and those not diagnosed with cancer who followed the lower gastrointestinal 2-week wait pathway were excluded. Patients who underwent colonoscopy as the first investigation were included. Patients undergoing any other test before colonoscopy were excluded. Patients undergoing flexible sigmoidoscopy were also excluded. Remaining patients’ records were then examined to identify the symptoms of initial referral to the 2-week wait scheme. Lesions identified at or distal to the splenic flexure were assumed to have been within reach of the flexible sigmoidoscope.
Results During 2013, 2950 2-week wait referrals were made from primary care. Of these, 968 colonoscopies were performed as a primary investigation. Thirty-five (3.6%) patients were found to have a colorectal cancer during colonoscopy (Table 1). Table 2 gives the tumour site in patients referred with anaemia alone. Six of the 15 cancers were in the caecum, five in the ascending colon, one at the hepatic flexure, two at the transverse colon and one at the splenic flexure. Table 3 details tumour site in patients referred with change in bowel habit
alone who underwent colonoscopy. Two of the 10 cancers were in the caecum, two in the descending colon, three in the sigmoid colon, one at the rectosigmoid junction and two in the rectum.
Discussion Since its introduction the 2-week wait pathway has been demonstrated to have a low rate of detection of colorectal cancer. This finding is supported by the data in the present study and by reports from other centres [5]. Furthermore the pathway has not been shown to improve survival from colorectal cancer as was originally hoped [6], but despite the lack of evidence for any value of the strategy the volume of 2-week referrals has continued to rise to a point where it is inhibiting the ability of secondary and tertiary care to provide services to other patients. In line with published guidance, flexible sigmoidoscopy has been used in our centre as the investigation of choice for rectal bleeding as the only symptom through the 2-week wait pathway. This guidance has also submitted patients with any change in bowel habit or anaemia to colonoscopy unless contraindicated. There is a grey area for those patients presenting with bleeding and symptoms such as change in bowel habit or anaemia suggesting the possibility of more proximal pathology. Until recently in our centre these have had colonoscopy as the first investigation in order that lesions above the splenic flexure would not be missed, but the demand for colonoscopy has prevented us from delivering this investigation for patients outside of the pathway such as those requiring assessment of colitis, bowel cancer surveillance or the investigation of CTidentified abnormalities. Compared with colonoscopy in our centre, CT colonography has been under-used in patients referred through the 2-week wait.
Table 1 Site and number of colorectal tumours found at colonoscopy as per presenting symptom. Presenting symptom as per 2-week wait referral
Tumour site proximal to splenic flexure
Tumour site distal to splenic flexure
Total
Anaemia Change in bowel habit Change in bowel habit and anaemia Visible rectal bleeding and anaemia Visible rectal bleeding and change in bowel habit Visible rectal bleeding and change in bowel habit and anaemia
14 2 1 (ascending colon) 0 0
1 8 0 1 6
15 10 1 1 6
2
2
0
35
Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 17, 980–983
981
D. G. Couch et al.
Straight to flexible sigmoidoscopy
Table 2 Tumour sites reported during colonoscopy of 2-week wait patients referred with anaemia alone. Tumour site
n
Caecum Ascending colon Hepatic flexure Transverse colon Splenic flexure
6 5 1 2 1
Table 3 Tumour sites at colonoscopy of 2-week wait patients referred with change in bowel habit alone. Tumour site
n
Caecum Descending colon Sigmoid colon Rectosigmoid junction Rectum
2 2 3 1 2
Resource limitations have not allowed us to expand the provision of our endoscopic service further and we have therefore been obliged to limit the use of colonoscopy within the 2-week wait pathway. In its place, through planned expansion of gastrointestinal radiology expertise and capacity, we now intend to use CT colonography and flexible sigmoidoscopy as far as is possible for 2-week wait patients without compromising diagnostic accuracy. By working back from cancer diagnosis to the symptom complex at the time of the original referral we have determined that within the 2013 patient cohort patients who presented with rectal bleeding and either a change in bowel habit or anaemia in addition were all found to have colorectal cancer within the range of a fibre-optic sigmoidoscope. This suggests that, within the patient population in the present study, patients with these symptoms had ‘distal’ rather than ‘proximal’ symptoms. As a result of this data analysis, we have changed the 2-week wait referral protocol to limit initial investigation to flexible sigmoidoscopy in any patient presenting with a symptom complex including visible rectal bleeding. In doing so the data indicate that the risk of missing a more proximal cancer is negligible. In practice this ‘straight to test’ protocol brings the patient directly to endoscopy on receipt of the referral letter. This preinvestigation consultation ensures that the referral is justified and appropriate. Should it be felt by the reviewing clinician that further tests be performed should the ini-
982
tial flexible sigmoidoscopy be normal, for example where there is a history of profound anaemia in the presence of negligible anal bleeding, a further CT colonogram can be arranged. This change in protocol has been ratified by the Leicester Colorectal Specialist Interest Group (a multidisciplinary group of clinicians interested in colorectal disease) and the University Hospitals of Leicester Cancer Board. Furthermore, the present study has also shown that 20% of cancer patients originally referred with change in bowel habit had a proximal lesion, the remainder being at or distal to the splenic flexure. In 80% of patients presenting with anaemia, the tumour was at or proximal to the hepatic flexure. With this evidence, our centre now provides CT colonography alone for patients presenting with anaemia or change in bowel habit because of its increased availability, its ability to survey other intra-abdominal structures and because it allows substitution of bowel preparation for water-soluble contrast medium and faecal tagging. A possible objection to our changes in handling 2week wait referrals could be that, in the shift towards flexible sigmoidoscopy for patients with any rectal bleeding regardless of change in bowel habit or anaemia, proximal cancers will be missed. This study did not identify any patient who presented with these symptoms in whom a cancer proximal to the splenic flexure was found. In addition to our evidence there are other studies that support the use of flexible sigmoidoscopy in this context. One large recent investigation which followed 1690 patients over a median of 35 months found that, when predominantly left-sided symptoms were investigated with flexible sigmoidoscopy, only two (0.24%) were subsequently found to have a more proximal colonic lesion [7]. In a larger series it was found that when patients were referred through the 2-week wait with bleeding, change in bowel habit or abdominal pain (without iron deficiency anaemia),
