Debate continues regarding the appropriate selection of patients for Flexible Sigmoidoscopy (FS) versus Whole Colon Imaging (WCI), either by Optical Colonoscopy (OC) or CT Colonography (CTC). FS is a quicker, less invasive alternative to Optical Colonoscopy (OC) for direct visualization of the colon during investigation of colorectal symptoms, however it will miss pathology proximal to the splenic flexure. OC offers possible visualization of the entire colon, however is a more technically difficult and time-consuming investigation [9]. CTC has been shown to be equally sensitive in the diagnosis of colorectal cancer when compared with OC and is increasingly popular for WCI, however also requires unpleasant bowel preparation and confers a high rate of incidental findings [5,6]. There is some debate concerning the selection criteria for each of these investigations and several recent studies have suggested that FS may be appropriate for patients presenting with “left sided” symptoms in the absence of anaemia or an abdominal mass on examination [2-4].

For example, Thompson et al undertook a prospective study of 16,433 patients over 16 years referred to colorectal clinic. They found that 5.8% had colorectal cancer: 86.2% within the recto-sigmoid colon (“distal”) and 13.8% elsewhere (“proximal”). 96.3% of all referrals had no anaemia or mass – “symptoms only”. 95% of cancers in these “symptoms only” patients were distal; 5% were proximal. FS for “symptoms only” patients missed only1.3% of cancers, suggesting this may be a reasonable option for investigating similar patients [4]. Subsequently, Atkin et al and Cross et al produced a multicentre retrospective study and cohort study respectively by utilising data from 2 randomised controlled trials. They included 7380 patients over the age of 55 referred with possible colorectal cancer to 21 NHS hospitals between 2004- 2007. 73% had change in bowel habit, 38% PR bleeding, 29% abdominal pain and 23% anaemia. 7.5% of patients had colorectal cancer: 77% distal to the splenic flexure, 22% proximal, 1% multifocal. Proximal cancers were diagnosed in 4.8% of patients with anaemia, with or without an abdominal mass. It was noted that a broad anaemia definition of <13 g/dL for men and <12 g/dL for women identified 80% of proximal cancers. Where there was no anaemia or mass: change in bowel habit to looser or more frequent stool, or PR bleeding with or without change in bowel habit yielded only 0.5% incidence of proximal cancer. These low risk “symptoms only” cases made up 41% of the recruited patients. Again, it was suggested that FS alone may safe in these patients [2,3].

However, Herrod et al undertook a systematic review of 7 studies looking at cancer location with “left sided” symptoms. They found that isolated change in bowel habit, with or without rectal bleeding, was a presenting symptom in 73% of left-sided cancers (95% CI 41–96%, I2 = 99%) but also in 13% of right- sided cancers (95% CI 2–30%, I2 = 96%). In all patients with cancer who presented with isolated change in bowel habit, with or without rectal bleeding, the cancer was right sided in 8% of cases (95% CI 4–12%, I2 = 69%). This suggests that there may be a significantly higher number of patients with proximal colon pathology which would be missed by FS alone [10].

There are other factors to be taken into consideration regarding WCI versus FS for the investigation of colorectal symptoms. For example, although FS for left-sided symptoms is considered a quicker and less technically challenging procedure when compared with OC, it has been suggested that only 65% successfully visualise the colon to the hepatic flexure, over 7% are abandoned due to poor bowel preparation and as many as 31.0-34.8% may require subsequent WCI [4,11]. It should be borne in mind that all patients diagnosed with colorectal cancer will require completion WCI [1]. Furthermore, although FS alone for left-sided symptoms may miss only a small proportion of cancers, it will miss all extra-colonic pathology as well as any benign proximal colonic pathology such as diverticulosis or adenomas.

OC and CTC remain the two core options for WCI, with OC generally considered to be the gold standard, and barium enema having been superseded by CTC [3]. However, CTC has been shown to have equivocal sensitivity when compared with OC - quoted at approximately 96% vs 94% respectively in a large systematic review and meta-analysis by Pickhardt et al involving 11,151 patients [8]. They further noted user heterogeneity to be much lower for CTC when compared with OC, which requires significant skill in endoscopy to complete efficiently and safely [9]. Although CTC is sensitive, it has some drawbacks. For example, the SIGGAR trial found that 30% of patients undergoing CTC for possible colorectal cancer required additional colonic investigations, versus only 8% in the OC group [7]. There is also a high rate of incidental extracolonic pathology detected, reportedly 58.7%.

Approximately 14.2% of these findings require investigation and only 2-3% are thought to be significant, such as malignancies and aneurysms [12,13]. The incidence with which these extracolonic incidentals may explain the patients’ symptoms is reported to be around 2.8%, however one study also suggested no difference in the incidence of extracolonic pathology between symptomatic (diagnostic) and asymptomatic (screening) groups [12,14]. Finally, it should also be remembered that CTC and OC are both able to diagnose benign but potentially significant proximal colonic pathology such as diverticulosis and adenomas in patients with “left sided” symptoms, though the incidence of right sided cancers may be low. Papers concerning the use of FS have recognised these limitations and often recommended further cost-benefit analysis regarding the use of WCI versus FS [2-4].

It is interesting to note that part of the argument given for FS over WCI in these patients, in light of the time and cost benefits, is that only 0.5-1.3% of colorectal cancers would be missed [2,3]. However, as outlined above, up to 8% of cancers in patients presenting with symptoms alone may be right sided [10]. Furthermore, completion WCI is still recommended for all colorectal cancers in light of a synchronous cancer incidence of approximately 2.3-3.5% [1,15,16]. This objective risk of missed cancer is deemed unacceptable, whereas a risk of at least 0.5-1.3% would be deemed acceptable were more liberal FS guidelines to be adopted.

Our results examined the possibility of missed non-”left sided” pathology (both right-sided colonic and other intra-abdominal causes) if guidelines were adopted recommending more liberal use of FS. Specifically, it is interesting to compare the current ACPGBI guidelines for the investigation of possible colorectal cancer with the widely cited recommendations by Cross et al. and similar papers, which suggest FS for investigating presentations of “left sided” symptoms in the absence of anaemia or an abdominal mass [1,2]. Our results revealed that 20% of those with non-”left sided” pathology would have qualified for FS alone, with 4 potentially significant findings missed, by application of the ACPGBI guidelines. However, 55% would have qualified for FS alone, with 12 potentially significant findings missed, by application of the Cross et al recommendations, including 4 right-sided polyps and an ileal neuroendocrine cancer (T3N1M0). It is important to note that by use of CTC, both groups also comprised potentially significant non-neoplastic findings such as right-sided diverticulosis, AAAs, an ovarian cyst, and lesions within the liver and spleen.