Article Text
PDF
Abstract
Objective To compare the sensitivity of autofluorescence endoscopy (AFE) and white light video endoscopy (WLE) for the detection of colorectal adenomas in high-risk patients belonging to Lynch syndrome (LS) or familial colorectal cancer (CRC) families.
Methods This was a prospective single-centre study carried out in a tertiary referral centre. The subjects were 75 asymptomatic patients originating from LS or familial CRC families. Patients were examined with either WLE followed by AFE or AFE followed by WLE. Back-to-back colonoscopy was performed by two blinded endoscopists. All lesions were removed during the second endoscopic procedure. Lesions missed during the second procedure were identified and removed on third pass. The sensitivity calculations for colorectal adenomas were based on histology results. The main outcome measures were the difference in sensitivity between WLE and AFE for the detection of adenomas in patients with LS or familial CRC.
Results At least one adenoma was detected in 41 (55%) patients. WLE identified adenomas in 28/41 patients and AFE in 37/41 patients, corresponding to a 32% increase. In total 95 adenomas were detected, 65 by WLE and 87 by AFE, resulting in a significantly higher sensitivity of AFE compared with WLE (92% vs 68%; p=0.001). The additionally detected adenomas with AFE were significantly smaller than the adenomas detected by WLE (mean 3.0 mm vs 4.9 mm, p<0.01).
Conclusions AFE improves the detection of colorectal adenomas in patients with LS or familial CRC. The results of this study suggest that AFE may be preferable for surveillance of these high-risk patients.
- Autofluorescence endoscopy
- cancer syndromes
- colonoscopy
- familial colorectal cancer
- Lynch syndrome
- surveillance colonoscopy
- white light video endoscopy
Statistics from Altmetric.com
- Autofluorescence endoscopy
- cancer syndromes
- colonoscopy
- familial colorectal cancer
- Lynch syndrome
- surveillance colonoscopy
- white light video endoscopy
Introduction
The detection and removal of colorectal adenomas has been proven effective in reducing mortality and incidence of colorectal cancer (CRC).1 Therefore, endoscopic surveillance is highly recommended in high-risk populations such as Lynch syndrome (LS) or familial CRC.2 Standard screening and surveillance colonoscopy is performed using flexible white light video endoscopy (WLE). However, with this technique, 2–26% of adenomatous polyps are missed.3 Furthermore, flat and depressed adenomas are often invisible to WLE.4 5 Such lesions reportedly make up 36% of neoplasia in a standard population,6 and they are presumably more common in high-risk subjects. Identification of these lesions is especially important in these high-risk populations. In LS mutation carriers, the mortality due to CRC remains >10-fold increased, even when undergoing a (bi-) annual WLE colonoscopy screening.2 This may to a considerable extent be due to rapid progression of smaller, missed lesions.
Autofluorescence endoscopy (AFE) is a technique specifically designed to probe large areas of mucosa to detect neoplasia. The technique has been approved by the US Food and Drug Administration for use in bronchoscopy and may also be of special use in the colon. The technique is based on the phenomenon that when tissue is exposed to light of a short wavelength (typically blue light) some endogenous biological substances (fluorophores) are excited, leading to subsequent emission of fluorescence light of a longer wavelength. This phenomenon is called tissue autofluorescence.7 The technique does not require administration of fluorescent dyes. Several pathological processes, in particular inflammation and neoplasia, change the concentration and distribution of the various endogenous fluorophores in the tissue and consequently alter the tissue's endogenous fluorescence. Other phenomena such as the increase in mucosal thickness and variation in distribution of haemoglobin in early colonic neoplastic lesions leads to attenuation of the emitted tissue autofluorescence.
In AFE the autofluorescent image is induced by using a white light source with a green filter. The tissue is thereby exposed to the remaining visible blue and red light. The reflected blue light is blocked by a filter in the CCD (charge-coupled device) camera. The emitted green autofluorescence from the tissue and the reflected red light are used to obtain a false-colour image. The system consisting of a light source, processor and dual CCD camera produces an image in which the normal colonic mucosa is depicted in cyan blue and adenomatous lesions in brick red.
Available data indeed suggest that AFE can detect small or early stage lesions and flat or depressed adenomas with a high sensitivity compared with WLE.8 9 However, prospective comparative studies with current standard video colonoscopy and AFE for surveillance of colorectal adenomas and CRC are lacking. The aim of this study therefore was to compare the sensitivity of AFE and WLE for the detection of colorectal adenomas in patients with LS or familial CRC.
Methods
Study population
Patients aged ≥18 year scheduled for surveillance colonoscopy were eligible for this study if they originated from LS or familial CRC families. Patients were categorised as patients with LS if they fulfilled the Amsterdam II criteria or if they carried a proven mutation in one of the mismatch repair genes (MLH1, MSH2 or MSH6).10 Patients were categorised as having familial CRC if they had one first-degree relative with CRC diagnosed at a young age (<50 years) or two first-degree relatives regardless of age.11
Patients with inflammatory bowel disease, familial adenomatous polyposis, Peutz–Jeghers syndrome or juvenile polyposis were excluded from the study, as were patients with a coagulopathy or on anticoagulant treatment that could not be discontinued.
The study protocol was approved by the institutional review board of the Erasmus MC University Medical Center, and informed consent for participation in the study was obtained from all subjects before inclusion.
Imaging systems
Standard WLE was performed using a standard flexible video endoscope (CF160, Olympus Optical, Tokyo, Japan) connected to a xenon light source. AFE was performed using the Onco-Life system (Xillix Technologies Corporation, Richmond, British Columbia, Canada). The Onco-Life system consists of a metal-halide light source, processor and dual CCD camera that is attached to the eyepiece of a standard fibreoptic endoscope (CF40, Olympus Optical, Tokyo, Japan). The Onco-Life system operates in two modes, providing both autofluorescence and conventional white light imaging. Modes can be switched instantly by pressing a lever on the camera head.
In fluorescence mode, blue light (400–450 nm) triggers autofluorescence from the tissue within the endoscopic field of view. The emitted green fluorescence passes through a dichroic mirror and a 490–560 nm bandpass filter to an intensified CCD, whereas the red reflected light is projected on a second CCD. The processor combines the two images into a single dual-colour digital image that can be displayed on a standard RGB (red green blue) monitor. The resolution and contrast of these images, although less than those of white light images, are sufficient to allow visualisation of the usual features evident by standard endoscopy.
In the white light mode, the Onco-Life system light source provides a standard broad-spectrum light (400–700 nm) for illumination of the endoscopic field. The reflected light is captured by a colour CCD camera. The white light images are processed and displayed in the conventional manner.
Endoscopic procedure
All patients were prepared for colonoscopy by ingesting 4 litres of a standard oral polyethylene glycol–electrolyte lavage solution (Coloforte, Ipsen Farmeutica, The Netherlands). All patients received 40 mg of butylscopolamine intravenously to improve visualisation of the colon. Colonoscopy was performed under conscious sedation with midalozam in combination with fentanyl citrate, and cardiopulmonary monitoring was used.
All patients underwent two colonoscopic examinations in one session (figure 1). An independent observer (DR or PP) recorded every macroscopically visible lesion suspicious of adenoma during withdrawal. Every procedure was videotaped and every lesion was immediately stored as a JPEG still image. Each lesion was graded by the endoscopist with respect to size (in millimetres, by comparison with the known diameter of an open biopsy forceps), morphology (flat or depressed, sessile, polypoid) and location. The latter was done both by measurement of the distance in centimetres from the anal verge with a fully stretched endoscope, and by identification of the located segment of the colon (ie, caecum/ascending colon/transverse colon/descending colon/sigmoid/rectum). The first colonoscopy was performed with either a standard video colonoscope (WLE) or with AFE by an experienced endoscopist (figure 1). All lesions were left in situ. Immediately after the first endoscopic procedure, the first endoscopist left the endoscopy suite and a second experienced endoscopist, who was unaware of the results of the first endoscopic procedure, performed the second endoscopic procedure with the alternative endoscopic method. All endoscopists involved had similar adenoma detection rates over the past years as identified from our endoscopy database, and endoscopists alternated with respect to the type of endoscopy performed in this study (WLE or AFE). All identified lesions were removed during the second endoscopic procedure and sent for histology. After finishing the second procedure, the independent observer unblinded the results of both WLE and AFE.
Flow chart of the study design. AFE, autofluorescence endoscopy; WLE, white light endoscopy.
Lesions found during WLE and AFE were matched based on the location in the colon and the comparison of the photographs taken during both procedures of any lesions. AFE was switched back to WLE only for the subsequent polypectomy. In the case of a switch back for stools, there was immediately a switch back to AFE to examine the remaining colonic mucosa to ensure that the colonic mucosa was not assessed by WLE.
In cases where lesions had been detected during the first endoscopic procedure but missed by the second procedure, the colon was re-examined on a third pass with either WLE or AFE to remove lesions left in situ (figure 1).
Data collection
Demographic data, medical history, family history, known mismatch repair gene mutations, drug use, degree of cleansing of the colon (1 excellent, 2 good, 3 moderate, 4 poor), endoscopic withdrawal time of each colonoscopy and all adverse events were recorded in the case record form. The withdrawal time was determined from a central review of videos by the first author (DR) and included the time for withdrawing the endoscope from the caecum to removal from the rectum, without the additional time needed to evaluate suspicious lesions and/or to perform polypectomy.
All removed lesions were stained with H&E and assessed by an experienced gastrointestinal pathologist, who was blinded for the macroscopic aspect of the lesion. Lesions were categorised by the pathologist as hyperplasia, adenoma (tubular, tubulovillous, villous or traditional serrated) or carcinoma.12 Neoplasia was defined as the presence of adenoma or carcinoma in the specimen.
Outcome parameters
The primary outcome parameter was the difference in sensitivity between WLE and AFE for the detection of adenomas during colonoscopy in patients with LS or familial CRC.
The secondary outcome parameters were: (1) the histological difference in the lesions detected only by WLE or only by AFE; (2) withdrawal time of the procedure for WLE and AFE; and (3) the size of the lesions detected with WLE and AFE.
Sample size calculation
Based on data on file of patients with LS or familial CRC who had undergone surveillance by WLE in 2004 in our department, at least one lesion was expected in 60% of patients. Autofluorescence was considered to be of benefit when 20% more adenomas would be detected in comparison with WLE.13 Based on an α of 0.05 and a power of 0.80 the required sample size was 100 patients (McNemar test of equality of paired proportions, nQuery Advisor).
The study protocol required an interim analysis after 50 enrolments. If a highly significant difference in the primary measure outcome was found (p<0.01; to limit the risk of a false-positive finding), the study would be terminated. Based on the interim analysis, the study was terminated after 51 patients. However, in the original study protocol AFE was routinely performed after WLE. In order to control for unintended biases, we extended the study by performing back-to-back colonoscopy in the reverse order—that is, AFE followed by WLE. Based on the results of the interim analysis (adenoma detected in 39% of patients with WLE compared with 59% with AFE and thus a 50% increase in adenoma-positive patients during AFE in comparison with WLE) and using an α of 0.05 and a power of 0.80, the required sample size for the study extension study was 22 patients (Mc Nemar test of equality of paired proportions, nQuery Advisor).
Statistical analysis
Data were analysed with the SPSS version 12.0 (SPSS, Chicago, Illinois, USA). Data are given as the mean and SD. Results of WLE and AFE were compared with each other and with histology results. These analyses ignored correlations between findings within patients.
The sensitivity of WLE and AFE for adenomas was calculated using histology as the standard. The McNemar test was used to calculate differences between WLE and AFE regarding the number of adenoma-positive patients, the sensitivity for adenomas and the detection rate of non-adenomatous lesions. The Mann–Whitney U test was used to calculate differences in the size of lesions detected by WLE and AFE only. The Fisher exact test was used to calculate differences in the endoscopic prediction and morphology of detected adenomas. The paired t test was used to compare the endoscopic withdrawal time of AFE and WLE. Multivariable proportional odds logistic regression analysis was performed to identify possible correlations between the withdrawal time and the number of adenomas detected with each technique, and for the difference in number of adenomas detected against the difference in the withdrawal times. An adenoma detection ratio representing the number of times AFE detected more adenomas than WLE per fixed amount of procedure time was calculated. This was done by dividing the sum of the withdrawal times per modality by the number of detected adenomas per modality (the procedure time). Secondly, the AFE procedure time was divided by the WLE procedure time (second calculation). The ratio was then calculated by dividing 1 by the result of the second calculation. For example, a ratio of 1 corresponded to no difference in the adenoma detection rate for AFE per fixed amount of procedure time compared with WLE. A ratio <1 corresponded to a lower adenoma detection of AFE per fixed amount of procedure time, while a ratio >1 corresponded to a higher adenoma detection rate. A two-sided p value of <0.05 was considered statistically significant.
Results
In total, 76 patients gave informed consent for inclusion in the study. One of them could not be included because back-to-back colonoscopy could not be performed due to very poor bowel preparation. The remaining 75 were included in the trial. Fifty-one patients were evaluated by WLE followed by AFE and 24 patients were evaluated by AFE followed by WLE (figure 1). Forty-one (55%) patients fulfilled the criteria of familial CRC and 34 (45%) patients were classified as having LS. Twenty-nine of the latter 34 (85%) patients were proven mutation carriers, whereas the other five fulfilled the Amsterdam II criteria, but no gene mutation had so far been identified. Thirty-four (45%) of the 75 patients were male and the mean age of the study population was 48 (±12) years. Twenty-three (31%) of the 75 patients entered a colonoscopic surveillance programme. The remaining 52 (69%) were already included in a colonoscopic surveillance programme, and the mean time since the last surveillance colonoscopy was 2.7 (±2.9) years. Bowel preparation was excellent in 31 (41%) patients, good in 27 (36%) patients and moderate in 17 (23%) patients. The caecum was intubated in all WLE and AFE procedures. No complications occurred during or after WLE and AFE procedures.
A third inspection was required in 26 (35%) patients for the removal of 44 lesions. In 22 of them, this third inspection was necessary for removal of 36 lesions that had been identified during WLE, but were missed during subsequent AFE (table 1). Histology showed that 3 (8%) of these 36 lesions were adenomas; the remaining 33 (92%) were hyperplastic lesions. In the remaining four patients requiring a third inspection, this was done to remove eight lesions that had been identified during AFE but were missed during subsequent WLE. Histology showed that all eight (100%) lesions were adenomas.
Detected lesions by WLE and AFE
In this study, a total of 173 lesions were detected: three CRCs, 95 adenomas and 75 hyperplastic lesions. AFE detected a total of 123 lesions, while WLE detected 129 lesions (table 1).
The mean withdrawal time of AFE (11.3±3.9 min) was significantly longer (p<0.001, paired t test) than that of WLE (9.8±3.0 min).
Multivariable proportional odds logistic regression analysis, however, revealed no significant relationship between withdrawal time and number of adenomatous lesions detected with both techniques (p=0.29). The adenoma detection ratio of AFE was 1.15; thus, per fixed amount of procedure time, AFE detected 1.15 more adenomas compared with WLE.
Adenoma-positive patients
Of the 75 patients included, 41 (55%) were found to have one or more adenomas (figures 2–4). WLE detected one or more adenomas in 28 patients, while AFE detected one or more adenomas in 37 patients. This 32% increase (95% CI −0.01% to −0.23%) in the number of adenoma-positive patients identified by AFE was significant (p=0.03). Twenty-four (59%) of the 41 patients were found to be adenoma positive by both WLE and AFE. Furthermore, WLE detected adenomas in four patients in whom no neoplastic lesions were detected by AFE. In contrast, AFE detected adenomas in 13 other patients in whom WLE detected no abnormalities. Of these 13 patients, seven were classified as familial CRC and six as LS.
Example images of two adenomas detected by both white light endoscopy (A) and autofluorescence endoscopy (B).
Multiple images of adenomas detected by autofluorescence endoscopy. Note the green-blue image of the colon mucosa, contrasting with the dark red-coloured adenomas.
Flat adenomas in the white light endoscopy (A) and autofluorescence (B) modality of the Onco-Life system.
Neoplasia detection rate
In total, 98 neoplastic lesions were detected in our study population: 91 tubular adenomas with low-grade dysplasia, two tubulovillous adenomas with low-grade dysplasia, two serrated adenomas and three adenocarcinomas. Fifty-seven of the 95 (60%) adenomas were detected by both WLE and AFE, including two tubulovillous adenomas with low-grade dysplasia and one serrated adenoma. Thirty-eight of the 95 (40%) adenomas were only seen with one of the two light techniques. WLE enabled detection of eight adenomas that were missed by AFE, resulting in an overall adenoma detection rate by WLE of 68% (65/95) (table 2). Of the latter eight adenomas, all had a size ranging from 3 to 5 mm. Seven of the eight adenomas had a sessile morphology, while one had a pedunculated morphology. All the eight adenomas were located in the proximal colon, two in the caecum, four in the ascending colon and two in the transverse colon. AFE, on the other hand, detected 30 additional adenomas that were not seen with WLE, resulting in an adenoma detection rate of 92% (87/95). AFE thus enabled the detection of 34% (95% CI −0.34% to −0.11%) more adenomas than WLE.
Detection of adenomas and hyperplastic polyps with WLE and AFE
Each lesion was immediately graded by the endoscopist as either non-adenomatous or suspicious for adenoma. Forty-nine (75%) of the 65 detected adenomas by WLE were correctly classified by the endoscopist, while 80 (92%) of the 87 adenomas detected by AFE were correctly classified (p=0.003, Fisher exact test).
The sensitivity of AFE (92%, 95% CI 0.78% to 0.94%) for the detection of adenomas was significantly higher (p = 0.001, McNemar test) than that of WLE (68%, 95% CI 0.55% to 0.74%). There was no significant difference in adenoma detection rate between the two endoscopists (p=0.46, Student t test). The majority (16/30) of the adenomas detected by AFE only were located proximal to the splenic flexure. The mean size of the adenomas detected by AFE only was significantly smaller than the size of those detected by both WLE and AFE (3.0±1.1 mm vs 4.9+2.1 mm; p<0.001, Mann–Whitney U test). Eleven (10%) adenomas were flat; 10 of them were detected by AFE compared with five by WLE (p=0.063, Fisher exact test).
Non-adenomatous lesions
A total of 75 hyperplastic lesions were identified (figure 5). WLE (sensitivity 81%, 95% CI 0.41% to 0.73%) detected significantly more hyperplastic lesions (p=0.001, McNemar test) than AFE (sensitivity 44%, 95% CI 0.21% to 0.44%). Of the 14 hyperplastic lesions detected only by AFE, eight were initially classified as adenomatous polyps during the endoscopic procedure, but histological evaluation showed these lesions to be hyperplastic.
Images of two hyperplastic lesions detected by autofluorescence endoscopy. Note the bright-green colour of the hyperplastic lesion which corresponds to the surrounding colonic mucosa.
LS versus familial CRC
Of the 41 patients with one or more adenomas, 14 (34%) were classified as having LS while the remaining 27 (66%) were classified as having familial CRC (table 3).
Adenoma detection rate and size per patient group
Discussion
This study has compared the yield of AFE with standard WLE for the detection of adenomatous lesions in a high-risk population. Both techniques yielded an equal number of lesions. However, AFE identified more adenomas including flat and serrated adenomas, whereas WLE identified more hyperplastic lesions. Using AFE, the overall adenoma detection rate increased significantly by 34%, corresponding to a higher sensitivity of AFE compared with WLE (92% vs 68%, p=0.001). Also, AFE identified a significant 32% more patients with adenomas. This yield of AFE compared with WLE is slightly higher than in a previous Canadian study, which reported an increased adenoma detection rate of 18% with AFE.13 The Canadian study population consisted of patients with an indication for surveillance after previous adenoma or colon cancer resection, while our study population consisted of high-risk patients with LS and familial cancer. Furthermore, the Canadian study had a somewhat different design, as the white light mode and autofluorescent modality were compared during the same procedure by the same endoscopist. This by definition meant that the endoscopist while performing AFE was aware of the WLE results. This contrasted with the currently presented study in which video WLE and AFE were performed by independent endoscopists unaware of the results of the other procedure. These differences may explain the level of difference in adenoma detection rate. More importantly, both studies consistently support the hypothesis that AFE significantly increases the detection of colorectal adenomas.
The higher detection rate of AFE could also be influenced by the experience of the endoscopist. However, both endoscopists in our study were highly experienced, with a similar level of endoscopic competence and with similar adenoma detection rates in the past during surveillance colonoscopies.
A strength of our study was that AFE and WLE were performed in a crossover manner. In the original study design AFE was routinely performed after WLE (n=51). However, as this could have introduced bias, the study was extended with 24 additional patients. These patients underwent back-to-back colonoscopy in a reverse order—that is, AFE followed by WLE. The current study design enabled a true comparison of the diagnostic yield of both WLE and AFE. All endoscopies were done by two independent endoscopists who were unaware of the results of the prior endoscopy instead of a design with one endoscopist and segmental unblinding.
A potential shortcoming of our study is the difference in withdrawal time between WLE and AFE. Previous studies have reported that a longer withdrawal time is associated with improved adenoma detection.14–16 In our study, the mean endoscopic withdrawal time for both WLE and AFE was significantly longer than the recommended minimum of 6 min.17 18 There was no time restraint on either procedure, and the endoscopists were aware that the patients participating in the study were high-risk cases. Withdrawal with AFE averaged 2 min more than with WLE (11.3 vs 9.8 min). This can in part be explained by the sensitivity of AFE to stool particles demanding additional cleansing. Stool particles can have the same red appearance as adenomatous tissue. For this reason, any ‘red flag’ that is spotted also needs to be checked with white light, requiring switching between both light modalities of the endoscope. In the case of a switch back for stool, we immediately switched back to AFE, ensuring that the colonic mucosa was not assessed by WLE. In the case of an adenoma, the switch back to WLE was extended for the duration of the polypectomy, after which AFE was continued for the remainder of the colon. One might consider the switching back as a major limitation of AFE, but this should be balanced against the additional number of detected adenomas with AFE. Most importantly, the difference in adenoma detection rates was not explained by the difference in withdrawal time between the two techniques (p=0.29) (figure 6).
Correlations between the endoscopic withdrawal time and detection rate of colorectal adenomas for white light endoscopy (WLE) and autofluorescence endoscopy (AFE). The dots represent the number of detected adenomas (y-axis) and the corresponding endoscopic withdrawal time (x-axis) for WLE and AFE.
In our study, the adenomas detected by AFE only were significantly smaller than those detected by WLE, and mostly had a tubular histology with low-grade dysplasia. Adenomas in patients with LS seem to progress more rapidly (2–3 years) to invasive CRC compared with those in the general population (8–10 years), so smaller lesions are also considered of particular importance in our patient category.19 Even with biennial colonoscopic surveillance the risk of CRC in patients with LS patients remains 10-fold increased compared with the general population.2 Rapid progression of small adenomas to invasive cancer is one of the likely explanations of the interval cancers that occur in LS despite surveillance colonoscopy.20 21 This means that we have to put effort into finding new endoscopic techniques to increase adenoma detection rates. Furthermore, these new endoscopic techniques should be evaluated in randomised studies with long-term follow-up to determine whether they will improve survival.
The number of hyperplastic lesions detected by AFE was significant lower than the number detected by WLE. Hyperplastic lesions do not exhibit neoplastic changes that might influence the tissue autofluorescence. During AFE, hyperplastic lesions do not appear as ‘flagged’ lesions but in fact only appear as an irregularity of the mucosal surface, sometimes with minimal red or white coloration that can be distinguished from a background with a normal autofluorescence. One could consider the lower detection rate of hyperplastic lesions as an additional benefit of AFE, as hyperplastic lesions generally do not carry a malignant potential and their removal is associated with costs and a small but relevant complication risk. On the other hand, it has been acknowledged that colorectal carcinoma can also develop via the ‘serrated’ pathway from hyperplastic polyps to sessile serrated polyps to carcinoma.22 In this respect, it is important to note that AFE enabled identification of all the traditional serrated adenomas.
Our study demonstrated that AFE is a suitable and reliable endoscopic technique for the detection of colorectal adenomas in patients with LS as well as those with familial CRC. Other endoscopic techniques to improve the detection of adenomas include pancolonic chromoendoscopy and narrow band imaging (NBI). Pancolonic chromoendoscopy has been shown to improve the detection of small adenomas with low-grade dysplasia in LS.23 24 However, pancolonic chromoendoscopy is time consuming because of the required dye spraying and therefore less practical in routine practice. The value of NBI is not yet clear, since previous studies show different results.25 26 More studies are needed to evaluate the yield of NBI in high-risk patients.
In conclusion, the use of AFE improves the detection of colorectal adenomas in patients with LS or familial CRC in comparison with WLE. Although more time consuming, the additional value of AFE is mainly in the detection of small lesions, which may well be relevant to this high-risk population. Therefore, AFE may be considered for colonoscopic surveillance of these high-risk patients.
Significance of this study
What is already known about this subject?
With conventional white light endoscopy, 2–26% of adenomatous polyps are missed.
Flat and depressed adenomas are often invisible to white light endoscopy.
Identification of these lesions is especially important in these high-risk populations.
Autofluorescence endoscopy is a technique specifically designed to probe large areas of mucosa to detect adenomas.
What are the new findings?
Screening with autofluorescence endoscopy identified considerably more patients with adenomas among subjects with Lynch syndrome or familial colorectal cancer.
Autofluorescence endoscopy resulted in a significantly higher adenoma detection rate than white light endoscopy.
The number of hyperplastic lesions detected by autofluorescence endoscopy was significantly lower than the number detected by white light endoscopy.
How might it impact on clinical practice in the foreseeable future?
The results of this study suggest that autofluorescence is preferable for screening and surveillance of these high-risk patients.
References
Footnotes
Competing interests None.
Ethics approval This study was conducted with the approval of the institutional review board of the Erasmus MC University Medical Center.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Digest