Abstract
Background: A myriad of localization options are available to endoscopists for colorectal cancer (CRC); however, little is known about the use of such techniques and their relation to repeat endoscopy before CRC surgery. We examined the localization practices of gastroenterologists and compared their perceptions toward repeat endoscopy to those of general surgeons.
Methods: We distributed a survey to practising gastroenterologists through a provincial repository. Univariate analysis was performed using the χ2 test.
Results: Gastroenterologists (n = 69) reported using anatomical landmarks (91.3%), tattooing (82.6%) and image capture (73.9%) for tumour localization. The majority said they would tattoo lesions that could not be removed by colonoscopy (91.3%), high-risk polyps (95.7%) and large lesions (84.1%). They were equally likely to tattoo lesions planned for laparoscopic (91.3%) or open (88.4%) resection. Rectal lesions were less likely to be tattooed (20.3%) than left-sided (89.9%) or right-sided (85.5%) lesions. Only 1.4% agreed that repeat endoscopy is the standard of care, whereas 38.9% (n = 68) of general surgeons agreed (p < 0.001). General surgeons were more likely to agree that an incomplete initial colonoscopy was an indication for repeat endoscopy (p = 0.040). Further, 56% of general surgeons indicated that the findings of repeat endoscopy often lead to changes in the operative plan.
Conclusion: Discrepancies exist between gastroenterologists and general surgeons with regards to perceptions toward repeat endoscopy and its indications. This is especially significant given that repeat endoscopy often leads to change in surgical management. Further research is needed to formulate practice recommendations that guide the use of repeat endoscopy, tattoo localization and quality reporting.
Colorectal cancer (CRC) is the third most commonly diagnosed cancer in Canada. In 2014, an estimated 24 300 individuals will have CRC diagnosed and an estimated 9300 will die of the disease.1 Colonoscopy is considered the gold standard for detection of CRC, with a specificity of 90.0% and a sensitivity of 95.0%.2 Over the past decade, the rise of minimally invasive surgical (MIS) techniques has made endoscopic lesion localization critical to surgical planning. However, little is known about the localization practices of gastroenterologists and general surgeons.
Although colonoscopy demonstrates excellent detection rates for both malignancies and adenomas, its ability to provide precise localization information is less clear. In the literature, estimates of the error rate in tumour localization vary from as low as 4% to as high as 21%.3–10 These errors can have a dramatic impact on surgical management, especially in laparoscopic cases where the surgeon lacks the tactile ability to palpate the colon for the lesion. This may result in conversion from a laparoscopic to open approach, intraoperative colonoscopies and removal of incorrect segments of colon. To help surgical planning and prevent such complications, surgeons often perform repeat colonoscopy before CRC surgery to verify lesion location.
In a recent study by Al Abbasi and colleagues,11 the repeat endoscopy rate at a large tertiary academic centre before CRC surgery was estimated to be up to 40.5%. Factors associated with preoperative repeat endoscopy by the operating surgeon were left-sided colonic neoplasms, planned laparoscopic resection and failure to tattoo the lesion on the initial colonoscopy.11 However, to our knowledge, no studies have assessed the perceptions of gastroenterologists toward repeat endoscopy and its indications.
When a lesion is detected by colonoscopy, endoscopists have a myriad of localization techniques at their disposal. They include use of anatomic landmarks, distance from the anal verge, image capture if the lesion is near an identifiable landmark, hemoclips and tattooing. No formal protocols exist to guide endoscopists in choosing the appropriate localization technique, and little is known about current use of these practices as they relate to repeat endoscopy before CRC surgery. Localization based on anatomic landmarks is often reported in the vast majority of polypectomies;12 however, this technique has been associated with error rates as high as 21%.3–10 Less clear is the practice of colonoscopic image capture, with studies demonstrating varied utilization rates.12,13 Moreover, there is a paucity of literature assessing the use of hemoclips in colonoscopy. Tattoo localization is considered the most accurate localization technique, and lack of tattoo localization is the most cited reason for repeat endoscopy by the operating surgeon.11,14 Nevertheless, the rate of tattooing remains quite low — between 0% and 23%.15,16 Furthermore, while tattoo localization is often recommended for colonic lesions suspicious for malignancy,17 the use of tattooing in different clinical scenarios remains unclear.
Given the limited evidence available on the localization practices of endoscopists, including the use of tattoo localization, the primary objective of our study was to identify the colonoscopic localization practices of gastroenterologists and to clarify the role of tattoo localization in this setting. The secondary objective was to assess the attitudes and perceptions of gastroenterologists toward repeat endoscopy and to compare their perceptions to those of general surgeons identified by a recent survey.
Methods
Instrument design
We developed a preliminary questionnaire based on a Medline literature review to ascertain the current localization practices and attitudes of gastroenterologists toward repeat endoscopy before CRC surgery. The preliminary survey was reviewed by a focus group consisting of 2 academic gastroenterologists, a community gastroenterologist, a gastroenterology resident and a practising general surgeon.
The final questionnaire comprised 16 questions that addressed demographic items, general localization practices, tattoo localization practices and indications for repeat endoscopy as well as attitudes and perceptions toward repeat endoscopy. The survey evaluated the frequency of 5 localization techniques used over the previous 12 months using a 5-point Likert scale (with possible responses being never, rarely, sometimes, frequently and always). Tattoo localization practices under 10 clinical scenarios were assessed using the same 5-point Likert scale. The 5 most common indications for repeat endoscopy were ranked from most to least frequent.11
Data collection and analysis
We disseminated the questionnaire in an online, electronic format using QuestionPro. Participants were identified and recruited via the membership directory of the Ontario Association of Gastroenterology (OAG) and among attendees of the 17th Annual OAG Conference. We performed statistical analyses using SPSS Statistics software version 21.0 (IBM Corp.).
We compared attitudes and perceptions of gastroenterologists toward preoperative repeat endoscopy and its perceived indications with those of practising general surgeons.18 Univariate group comparisons for categorical data were achieved using the χ2 test. We considered results to be significant for all comparisons at p < 0.05. The University Health Network Research Ethics Board approved our study protocol.
Results
Of the 184 active members of the OAG, 69 practising gastroenterologists completed the survey, along with 3 trainees. Trainees were excluded from the statistical analysis, resulting in a response rate of 38%. The characteristics of participants excluding trainees are presented in Table 1. The majority of respondents were men (89.9%), had more than 20 years of practice experience (43.5%) and worked in an urban setting (92.8%). Most respondents were employed in a teaching hospital (60.9%); only 8.7% worked in private clinics. All respondents performed colonoscopies. Colonoscopy volume was widely distributed; whereas the majority of respondents performed between 50 and 100 colonoscopies per month (60.9%), 27.5% performed less than 50 colonoscopies per month.
Localization practices
On average, 56.5% ± 11.5% of respondents routinely shared colonoscopic images and/or videos with the consulting surgeon when referring patients for CRC surgery. The frequency of various localization techniques when a lesion was detected by colonoscopy is highlighted in Table 2. Most respondents used anatomical landmarks (91.3% ± 10.9%), followed by tattooing (82.6% ± 11.3%), image capture (73.9% ± 11.5%) and distance measured from the anal verge (71.0% ± 11.5%). No participants reported the use of hemoclips for localization.
Tattoo localization practices are highlighted in Table 3. The majority of respondents indicated they would tattoo lesions that could not be removed by colonoscopy (92.8% ± 10.9%), polyps with high-risk features (95.7% ± 10.7%) and large lesions suspicious for malignancy (84.1% ± 11.3%). In addition, 76.8% ± 11.5% indicated they would tattoo polyps with features that may require follow-up colonoscopy. Only 2.9% ± 5.0% indicated they would tattoo polyps smaller than 1 cm. The vast majority of respondents would routinely tattoo a lesion regardless of surgical approach (91.3% ± 10.9% for laparoscopic surgery; 88.4% ± 11.1% for open surgical resection). With respect to tumour location, respondents said they would be likely to tattoo left-sided (89.9% ± 11.0%) and right-sided malignancies (85.5% ± 11.2%). Only 20.3% ± 8.2% said they would be likely to tattoo a rectal lesion.
Repeat colonoscopy prior to CRC surgery
Table 4 summarizes the perceived indications for preoperative repeat endoscopy specified by gastroenterologists as well as general surgeons. Gastroenterologists (80.6%) were more likely than general surgeons (56.6%) to identify preoperative planning or tattoo localization by the operating surgeon as the primary indication for repeat endoscopy (p < 0.001). General surgeons were more likely than gastroenterologists to agree that an incomplete initial colonoscopy was an indication for repeat endoscopy (14.9% v. 6.9%; p = 0.040).
Most (63 [91.3%]) gastroenterologists agreed that tattoo localization of a malignancy on the initial colonoscopy is the standard of care. Only 1 (1.4%) gastroenterologist agreed that repeat colonoscopy by the general surgeon before CRC surgery is the standard of care, whereas 68 (38.9%) general surgeons agreed (p < 0.001). The majority of gastroenterologists (48 [69.6%]) and general sugeons (105 [60.0%]) disagreed with the statement, “Repeat colonoscopy prior to surgery has minimal impact on total cost of care” (p = 0.15). A total of 104 (59.4%) general surgeons agreed that “Repeat colonoscopy prior to surgery has minimal impact on time to definitive surgery,” whereas only 28 (40.6%) gastroenterologists agreed (p = 0.003). Further, 84 (56%) of the general surgeons surveyed indicated that the findings of repeat endoscopy often lead to changes in the operative plan. With regard to conducting repeat endoscopy before CRC surgery in the previous 12 months, 46 (26.9%) never performed repeat endoscopy, 73 (42.7%) in less than 50% of CRC cases, 24 (14.0%) in 51%–75% of cases and 28 (16.4%) performed repeat endoscopy in more than 75% of cases.
Discussion
Our study demonstrates that for lesions detected by colonoscopy, gastroenterologists most frequently use anatomical landmarks (91.3%) for tumour localization. This is followed by tattooing (82.6%), image capture (73.9%) and colonoscopy tip measured from the anal verge (71.0%). Use of hemoclips was not reported by any of the respondents in this study. The frequent use of anatomical landmarks corroborates recent findings by Beaulieu and colleagues,12 who demonstrated a 99.1% anatomical location-based reporting rate.12 Despite these results, use of anatomical landmarks compared with intraoperative location has been associated with localization error rates as high as 21%, thus leading many to recommend tattoo localization instead.7,9
Given that lack of tattoo localization was cited as one of the most common indications for repeat endoscopy by general surgeons,18 the results of our study are an important contribution to further understanding this practice by gastroenterologists. For lesions detected at colonoscopy in the previous 12 months, 74% of respondents said they frequently localized the lesion by tattooing. This result is similar to those of Conaghan and colleagues,17 who demonstrated a tattoo localization rate of 65.1% in 85 patients who underwent laparoscopic resection for colorectal tumours. However, in their study, 31% of patients were tattooed at a repeat endoscopy, a procedure that is both costly and has associated risks.17 Respondents in our study were very likely to tattoo lesions that could not be removed by colonoscopy (92.8%), polyps with high-risk features (95.7%) and polyps with features that may require follow-up colonoscopy (76.8%). These findings are in contrast to those of a recent study by Zafar and colleagues,16 who examined tattoo localization in 165 patients with polyps and reported a tattoo rate of only 23%. Our study demonstrates a tattoo rate of only 2.9% for polyps smaller than 1 cm and much higher rates for polyps with high-risk features. To our knowledge, this is the first study to explicitly assess tattooing practices based on polyp features.
There is a paucity of literature evaluating tattoo rates with respect to tumour location. Most of the respondents in our study said they would tattoo right-sided (85.5%) and left-sided malignancies (89.9%), whereas only 20.3% said they would tattoo rectal lesions. Keller and colleagues15 demonstrated an even lower rectal tattoo rate of 4.1% in 49 patients with rectal polyps that were later diagnosed as neoplastic. This discrepancy in tattooing of rectal lesions may best be explained by the clinical scenarios addressed: whereas Keller and colleagues assessed only tattooing of polyps, our study assessed the likelihood of tattooing rectal lesions suspicious for malignancy. Interestingly, our study demonstrates that gastroenterologists are likely to tattoo malignancies in the right colon despite evidence that right-sided lesions close to known anatomical landmarks, such as the ileocecal junction or appendiceal orifice, may not require tattoo localization.19 The finding that gastroenterologists are likely to tattoo left-sided lesions and lesions that will be referred for laparoscopic surgery (91.3%) builds on recent work by Al Abbasi and colleagues,11 who showed a significant increase in the rate of repeat endoscopy before CRC surgery for left-sided lesions and planned laparoscopic surgery.
The preoperative repeat endoscopy rate at a large tertiary care centre has been estimated to be 40.5%, with the most common reasons for repeat endoscopy including tattooing of the lesion or preoperative planning in more than 80% of cases.11 Given the relatively high cost of colonoscopy in Canada20 and the adverse potential risk of perforation with colonoscopy,21 recent efforts have been made to elucidate general surgeons’ perspectives on repeating colonoscopy.18
Our study builds on previous work by our group by assessing and comparing the attitudes and perceptions of gastroenterologists with those of general surgeons. Our study confirmed that the majority of both gastroenterologists and general surgeons identify preoperative planning or tattoo localization by the operating surgeon as the primary indication for repeat endoscopy. Furthermore, we found that general surgeons were more likely than gastroenterologists to believe that preoperative repeat endoscopy is the standard of care (p < 0.001). Only 1.4% of gastroenterologists agreed that repeat endoscopy should be the standard of care. This difference is likely explained by our finding that the majority of general surgeons frequently alter their surgical plans based on the results of the repeat colonoscopy. The featured discrepancies between the opinions of gastroenterologists and general surgeons emphasize the misconceptions that exist with regards to this costly practice. Our finding that 56% of general surgeons agree that repeat colonoscopy often leads to change in surgical management underlines the importance of accurate localization and reporting of this information among endoscopists. This finding is supported by a recent study by Saleh and colleagues,22 who demonstrated that repeat endoscopy was protective against localization error, with a sensitivity greater than 80% in detecting errors. These conflicting perceptions are likely secondary to the lack of standardized protocols that guide endoscopists with respect to the indications for repeat endoscopy and the indications for tattooing. Additionally, the lack of standardized documentation and reporting of localization information compounds this problem.
Limitations
The findings of the present study should be interpreted within the context of the following limitations. First, the results may have been influenced by recall bias. To minimize this possibility, we used Likert scales and ranges rather than soliciting discrete values from respondents. Second, although respondents included gastroenterologists from across the province of Ontario, the results may not be generalizable to other gastroenterology groups. Finally, the results are based on self-reporting by clinicians, and thus we were unable to calculate precise rates of tumour localization practices.
Conclusion
The present study confirms that anatomic landmarks are the most commonly used localization technique by gastroenterologists for lesions detected by colonoscopy. Moreover, gastroenterologists frequently use tattoo localization for high-risk polyps, large lesions suspicious for malignancy, and left- and right-sided malignancies. However, discrepancies exist between gastroenterologists and general surgeons with regards to perceptions toward repeat endoscopy and its indications. Such discrepancies highlight the importance of developing standardized guidelines for tattooing, repeat endoscopy and reporting of localization information among endoscopists. This is especially important given that general surgeons report frequent changes in the operative plan as a result of new information learned at repeat endoscopy. Further research is needed to formulate practice recommendations that guide the use of repeat endoscopy and tattoo localization with the aim of improving quality of patient care while minimizing cost.
Footnotes
This work was presented at Digestive Disease Week (American Society for Gastrointestinal Endoscopy), May 16–19, 2015.
Competing interests: None declared.
Contributors: T.D. Jackson, A. Okrainec, P.G. Rossos and F.A. Quereshy designed the study. A. Azin and M.C. Jimenez acquired the data, which A. Azin, M.C. Jimenez and M.C. Cleghorn analyzed. A. Azin, M.C. Jimenez and M.C. Cleghorn wrote the article, which T.D. Jackson, A. Okrainec, P.G. Rossos and F.A. Quereshy reviewed. All authors approved the final version of the manuscript for publication.
- Accepted February 23, 2015.