- Email: [email protected]
Risk factors for early and late adenoma recurrence after advanced colorectal endoscopic resection at an expert Western center
ORIGINAL ARTICLE
Risk factors for early and late adenoma recurrence after advanced colorectal endoscopic resection at an expert Western center Andrew Emmanuel, MBChB, MS,1,2 Christo Lapa, MD,1 Anil Ghosh, MBBS, PhD,1 Shraddha Gulati, MBBS,2 Margaret Burt, RN,1,2 Bu’Hussain Hayee, PhD,2 Amyn Haji, MA, MSc, MD1,2 London, United Kingdom
Background and Aims: Few large Western series examine risk factors for recurrence after endoscopic resection (ER) of large (20 mm) colorectal laterally spreading tumors. Recurrence beyond initial surveillance is seldom reported, and differences between residual/recurrent adenoma and late recurrence are not scrutinized. We report the incidence of recurrence at successive surveillance intervals, identify risk factors for recurrent/residual adenoma and late recurrence, and describe the outcomes of ER of recurrent adenomas. Methods: Recurrence was calculated for successive surveillance periods after colorectal ER. Multiple logistic regression was used to identify independent risk factors for recurrent/residual adenoma and late recurrence (12 months). Results: Six hundred twenty colorectal ERs were performed, and 456 eligible patients (98%) had completed 3- to 6-month surveillance. Residual/recurrent adenoma (3-6 months) was detected in 8.3%, at 12 months in 6.1%, between 24 and 36 months in 6.4%, and after 36 months in 13.5%. Independent risk factors for residual/recurrent adenoma were piecemeal resection (odds ratio [OR], 13.0; P Z .01), adjunctive argon plasma coagulation (OR, 2.4; P Z .01), and lesion occupying 75% of the luminal circumference (OR, 5.6; P < .001) and for late recurrence were lesion size >60 mm (OR, 6.3; P < .001) and piecemeal resection (OR, 4.4; P Z .04). Of 66 patients with recurrence, 5 required surgery, 8 left the treatment pathway, 20 are still receiving ER or surveillance, and 33 had ER with normal subsequent surveillance. Conclusions: Recurrence occurs at successive periods of surveillance after ER even beyond 3 years. Aside from piecemeal resection, risk factors for residual/recurrent adenoma and late recurrence are different. Recurrence can be challenging to treat, but surgery is rarely required. (Gastrointest Endosc 2019;-:1-10.)
Recurrence after endoscopic resection (ER) of large colorectal laterally spreading tumors (LSTs) is not infrequent.1,2 It is difficult to predict, and many patients who do not develop recurrence are subjected to frequent and Abbreviations: APC, argon plasma coagulation; ER, endoscopic resection; ESD, endoscopic submucosal dissection; LST, laterally spreading tumor; pEMR, piecemeal endoscopic mucosal resection. DISCLOSURE: All authors disclosed no financial relationships relevant to this publication. Copyright ª 2019 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 https://doi.org/10.1016/j.gie.2019.01.031 Received October 2, 2018. Accepted January 28, 2019. Current affiliations: Department of Colorectal Surgery (1), King’s Institute of Therapeutic Endoscopy (2), King’s College Hospital NHS Foundation Trust, London, UK. Reprint requests: Amyn Haji, King’s Institute of Therapeutic Endoscopy, King’s College Hospital NHS Foundation Trust, London SE59RS, UK.
www.giejournal.org
ultimately unnecessary surveillance procedures. Despite the now widespread use of ER, few large Western series examine the risk factors for recurrence. In those that do, diverse possible risk factors are reported with very few factors consistently demonstrated between studies.3 Furthermore, early recurrent or residual adenoma at the first surveillance is usually investigated, but factors associated with longer-term recurrence are almost never reported. Few Western data report the incidence of recurrence beyond 12 to 18 months of follow-up.2,4-7 Most large Western series to date have exclusively focused on EMR, but it is important to consider the evolving nature of advanced colorectal ER. In many expert Western centers, increasing experience has resulted in better lesion assessment and selection, higher quality ER, and an increased use of ESD as a routine component of the treatment strategy.2,8,9 Endoscopic treatment of recurrent lesions, although frequently feasible, can be extremely challenging, and Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 1
Risk factors for early and late adenoma recurrence
Emmanuel et al
there is little consensus for the resection strategy between studies.2,10-13 Given that radical surgery may be the only alternative option for these patients, it is important to define the outcomes for patients treated for recurrent adenomas after initial expert resection. In this study we report the short- and long-term incidence of recurrence after ER of large colorectal lesions and identify risk factors for recurrent or residual adenoma at the first surveillance and for late recurrence after 12 months of follow-up or longer. We also describe the outcomes of ER of recurrent adenomas after initial expert resection.
METHODS This was a retrospective study of prospectively collected data on ERs of large (20 mm) colorectal superficial neoplastic lesions between January 2010 and March 2018 at a tertiary referral center. Patients were referred to our service if the endoscopist performing the index procedure believed the lesion was beyond the capabilities of the referring institution as a result of large size, difficult location, challenging morphology, or previous failed attempts at resection. All patients with successful initial ER who had undergone at least the first surveillance colonoscopy were included. Those with noncurative ER of invasive cancer were excluded, even if they had declined radical surgery.
Procedures Our approach to assessment and resection has been previously described.8,9 Procedures were generally performed with the patient under conscious sedation and analgesia with intravenous midazolam and fentanyl administered by the endoscopist. Patients were discharged home on the day of the procedure. All assessments and procedures were performed by 2 experienced interventional endoscopists or senior endoscopy fellows under their direct supervision. All lesions were assessed before resection with chromoendsocopy using indigo carmine and magnification colonoscopy (model CF-H260AZL colonoscope [Olympus, Tokyo, Japan] and model EC-600Z colonoscope [Fujifilm, Düsseldorf, Germany]) and classified according to Kudo pit pattern and vascular pattern (Showa classification).14,15 Selected lesions were also evaluated with variable high-frequency miniprobe US (Fujifilm) to evaluate for submucosal invasion. All lesions were considered for ER except those with invasive cancer >T1 (Kudo type Vn pit pattern, sparse vascular pattern, at least muscularis propria invasion on high-frequency miniprobe colonoscopic US) for which surgery was advised. Techniques used for resection included ESD, hybrid ESD, and EMR or piecemeal EMR (pEMR). The decision for the resection technique was based on the patient and 2 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
lesion characteristics specific to each, bearing in mind our general considerations when undertaking ER.16 In brief, we generally use ESD where there is a suspicion of invasive cancer (Kudo type Vi pit, irregular or sparse narrow-band imaging) or with lesions with a significant risk of harboring invasive cancer such as LST granular mixed-nodular type, LST nongranular, or a pseudodepressed (IIc) component. We also often choose ESD or hybrid ESD to aid resection where there is profound submucosal fibrosis. Magnification did not specifically influence the selection of ESD or EMR but, in our opinion, makes the detection of type Vi pit or irregular vascular pattern easier to detect. If these features are discovered, ESD is the preferred resection technique. EMR and pEMR were performed using a sequential injection and resection technique. The submucosal injection consisted of succinylated gelatine (Gelofusine; B. Braun, Crissier, Switzerland) with adrenaline diluted to 1:100 000 and dyed with methylene blue or indigo carmine. A variety of snare types as appropriate to the lesion were used. ESD was performed using an Olympus PCF Q260JL with ST distal hood (Fujifilm) and the FlushKnife (Fujifilm). In brief, a standard technique was typically used consisting of a mucosal incision on the anal side of the lesion, trimming and flap creation, followed by submucosal dissection under the lesion. The mucosal incision was then completed around the lesion. Hybrid ESD was used to aid resection of some previously heavy manipulated lesions and consisted of variable ESD dissection to aid resection or snare capture of areas of submucosal fibrosis. The Erbe VIO 300D electrosurgery device (Erbe Elekromedezin, Tübingen, Germany) was used with appropriate settings depending on the resection technique. After resection, lesions were retrieved and submitted for histopathologic analysis. The resection bed was routinely examined again using magnification chromoendoscopy for any residual polyp. For lesions with profound submucosal fibrosis, argon plasma coagulation (APC) was sometimes used as an adjunctive measure after resection of visible adenoma (not to ablate visible residual tissue), although this was dependent on the endoscopist’s preference at the time of resection, and there was no specific protocol for the use of adjunctive APC. The presence of fibrosis was judged by the endoscopist and recorded but was not characterized in a standardized manner according to reported grading systems.17
Surveillance Patients underwent surveillance endoscopy at 3 to 6 months and 12 months and, if no recurrence was detected, colonoscopy at 36 months. After the 12-month surveillance, patients could be discharged to their referring hospital for local surveillance because our unit receives referrals form a wide geographic area. Surveillance up to 12 months was performed by members of the advanced ER team. The protocol for surveillance involved routine www.giejournal.org
Emmanuel et al
examination of the scar using magnification chromoendoscopy and magnification narrow-band imaging or blue light imaging (BLI). If there was no evidence of recurrent adenoma on magnification assessment, the scar was not routinely biopsy sampled.
Data collection Data were collected on patient demographics, lesion characteristics, details of the resection procedure, histopathology findings, findings at surveillance, and details of any recurrence and subsequent treatment. Reports of surveillance procedures performed at patients’ local institution beyond 12 months were also obtained. Recurrent/residual adenoma was defined as histologically confirmed recurrence at the site of previous resection at 3 to 6 months surveillance. Late recurrence was defined as adenoma at the site of previous resection at 12 months’ surveillance in patients who had no evidence of recurrent/ residual adenoma at the first (3-6 months) surveillance. Recurrence at each period of surveillance was defined as detection of recurrent adenoma at the scar from previous resection in patients who had no evidence of recurrence at any previous surveillance. For analysis of outcomes of recurrent adenomas, any treatment for recurrence after an initial apparently successful ER in our unit was included, regardless of any attempted prior resection or substantial manipulation (6 biopsy samples) at other institutions before referral.
Statistical analysis Results are reported using means and standard deviation for continuous variables and frequencies for categorical variables. Potential risk factors for recurrent/residual adenoma and for late recurrence were identified by univariate analysis using the t test to compare means and the c2 test to compare categorical variables. Multiple logistic regression using stepwise variable selection was performed for factors associated with recurrent/residual adenoma or late recurrence with P < .1 on univariate analysis to identify independent associations. Ethical approval for this study was granted by the National Research Ethics Committee.
Risk factors for early and late adenoma recurrence
mean age of the analyzed patients was 71 years. The mean lesion size was 53.7 mm (median, 50 mm; range, 20-160). Three hundred thirty-eight were resected using EMR or pEMR and 118 using ESD or hybrid ESD. Twelve lesions contained invasive cancer with curative ER (invasion <1000 mm from muscularis mucosa and absence of poor differentiation, tumor budding, or lymphovascular invasion). Fifty percent of lesions had been subjected to substantial manipulation before referral, and 34% of lesions resected using ESD or hybrid ESD had been subjected to prior attempted resection or substantial manipulation compared with 18% of those resected by EMR/pEMR (odds ratio [OR], 2.4; 95% confidence interval [CI], 1.5-3.6; P < .001).
Incidence of recurrence according to length of follow-up Residual or recurrent adenoma at 3 to 6 months was detected in 8.3% (95% CI, 6.0%-11.4%) (Fig. 2), at 12 months in 6.1% (95% CI, 3.5%-9.7%), between 24 and 36 months in 6.4% (95% CI, 2.6%-12.8%), and after 36 months in 5 of 37 patients (13.5%; 95% CI, 4.5%-28.8%).
Recurrent/residual adenoma at 3 to 6 months Univariate analysis (Table 1) showed residual/recurrent adenoma was significantly more frequent in those referred after prior attempts at ER (P < .001), lesions occupying 75% of the luminal circumference (P < .001), larger lesions (P Z .02), piecemeal resections (P < .001), EMR/pEMR or hybrid ESD compared with ESD (P Z .04), where adjunctive APC was used (P < .001), and in cases where the mucosal defect was closed with endoscopic clips (P Z .02). The multiple logistic regression model (Table 2) showed that independent predictors of residual/recurrent adenoma were piecemeal resection (OR, 13.0; 95% CI, 1.7-98.4; P Z .01), adjunctive use of APC (OR, 2.4; 95% CI, 1.1-5.0; P Z .02), and lesion occupying 75% of the luminal circumference (OR, 5.6; 95% CI, 2.4-12.9; P < .001). Large size was not an independent predictor of residual/recurrent adenoma.
Late (≥12 months) recurrence RESULTS ER was performed for 620 LSTs 20 mm over the study period. The flow diagram in Figure 1 shows the patients’ surveillance and reasons for exclusion. Forty-one ERs were performed for recurrent adenomas 20 mm. ER was initially unsuccessful in 5 (.9%). Invasive cancer with noncurative ER was found in 30 patients. Twenty-one patients were medically unfit or refused further surveillance, and 42 are awaiting surveillance that is not yet due. Six patients died of unrelated causes before the first surveillance. Of the remaining 464 eligible patients, 456 (98%) had completed the first surveillance. The www.giejournal.org
Two hundred sixty-two patients without residual/recurrent adenoma at first surveillance have undergone 12 months or longer surveillance. Late recurrence was more frequent in large lesions (P < .001), in those referred after prior attempts at ER (P < .009), lesions occupying 75% of the luminal circumference (P < .001), piecemeal resections (P < .003), in cases where the mucosal defect was closed with endoscopic clips (P Z .02), and in patients with form of postprocedure adverse event (P Z .02), although there was no association specifically with significant postprocedure bleeding (P Z .48) or any form of colonic wall muscle injury (P Z .13) (Table 1). The multiple logistic regression model (Table 2) showed that Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 3
Risk factors for early and late adenoma recurrence
Emmanuel et al
Endoscopic resection ≥20mm N=620
Eligible for 1st surveillance N=464 Lost to follow-up N=8 1st (3-6 month) surveillance N=456 (98%)
Eligible for 2nd surveillance N=303
Ineligible for 1st surveillance - Procedure due to recurrence, n=41 - Invasive cancer, non curative ER, n=30 - Unsuccessful ER, n=5 - Non-neoplastic histology, n=9 - Died before surveillance, n=6 - Surveillance not due, n=42 - Surgery for other pathology, n=2 - Not fit for surveillance, n=7 - Refused surveillance, n=14
Ineligible for 2nd surveillance -
Died before surveillance, n=9 Surveillance not due, n=93 Surgery for recurrence, n=1 Surgery for other pathology, n=6 - Not fit for surveillance, n=18 - Refused surveillance or asked for local hospital surveillance, n=26
Lost to follow-up N=14 2nd (12 month) surveillance N=289 (95%)
3rd (18-35 month) surveillance N=173
>36 months surveillance N=58
Figure 1. Flow diagram showing follow-up of patients after endoscopic resection (ER).
independent predictors of late recurrence were lesion size >60 mm (OR, 6.3; 95% CI, 2.7-14.9; P < .001) and piecemeal resection (OR, 4.4; 95% CI, 1.3-15.3; P Z .02).
Outcomes of treatment of recurrent adenomas Sixty-six patients developed recurrent adenomas (Fig. 3). Overall, 5 required surgery, and the rest were deemed suitable for further endoscopic treatment, although not all accepted or were medically fit for treatment. During the course of treatment and surveillance, 8 patients left the treatment pathway because of lack of medical fitness, death, or inability to contact. Twenty patients are still receiving endoscopic treatment or await surveillance after endoscopic treatment, whereas in 33 patients endoscopic cure 4 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
appears to have been successful with normal subsequent surveillance. Only 1 patient, who originally had a 40-mm LST (tubulovillous adenoma) resected by pEMR, developed invasive cancer in a recurrent lesion, although the patient’s treatment for recurrence had been significantly delayed because of treatment for other medical problems. Reassessing the lesion showed evidence of type V Kudo pit pattern, and surgery was recommended. The interval between the first detection of recurrence and surgery was 35 months. After right hemicolectomy the cancer was staged as T2 N0. Of the recurrent adenomas that have been treated, 97% were tubulovillous or tubular adenomas, 3% were serrated adenomas, only 1 contained high-grade dysplasia, and the remainder were low-grade dysplasia. Patients treated www.giejournal.org
Emmanuel et al
3-6 month surveillance N=456
Risk factors for early and late adenoma recurrence
Recurrence N=38 (8.3%)
Late Residual/recurrent P recurrence P adenoma (3-6 mo) value (≥12 mo) value
No recurrence N=418
12 month surveillance N=262
Age
Recurrence N=16 (6.1%)
No recurrence N=245 24-36 months surveillance N=109
TABLE 1. Exploratory univariate analysis of factors potentially associated with recurrent/residual adenoma and late recurrence
Recurrence N=7 (6.4%)
No recurrence N=102
<70 y
15 (39.5%)
70-79 y
14 (36.8%)
13 (44.8%)
80 y
9 (23.7%)
7 (24.1%)
9 (31.0%)
.20
Size 20-30 mm
3 (1.1%)
31-40 mm
6 (15.8%)
41-60 mm
12 (31.6%)
5 (17.2%)
>60 mm
17 (44.7%)
20 (69.1%)
.02
1 (3.4%)
<.001
3 (10.3%)
Location Right-sided colon
12 (31.6%)
Left-sided colon
8 (21.1%)
5 (17.2%)
Rectum/ rectosigmoid
18 (47.4%)
11 (37.9%)
.47
13 (44.8%)
.36
Morphology Is/Isp
>36 months surveillance N=37
.94
Recurrence N=5 (13.5%)
2 (.1%)
.06
1 (3.4%)
LST G H
21 (55.3%)
15 (51.7%)
LST G MN
15 (39.5%)
12 (41.4%)
0 (0%)
1 (3.4%)
LST NG
.07
Kudo pit pattern*
No recurrence N=32 Figure 2. Flow diagram showing the incidence of first-detected recurrence at successive periods of surveillance.
I or II
0 (0%)
III or IV
.47
0 (0%)
36 (94.7%)
27 (93.1%)
V
0 (0%)
1 (3.4%)
Not recorded
2 (.1%)
1 (3.4%)
.81
Vascular pattern* Faint/normal
with ER for recurrent adenoma have required a median of 2 ERs (range, 2-7). Table 3 shows a comparison of lesions at initial treatment and treatment for recurrent lesions (all instances of ER for recurrence included, including multiple treatments of repeat recurrence). For lesions 20 mm, ESD or hybrid ESD was used significantly more frequently to resect recurrent lesions (44% vs 26%, P Z .006). Adjunctive use of APC or soft coagulation was used significantly more frequently after resection of recurrent adenomas (P < .001). There were no significant differences in overall adverse events, intraprocedural bleeding, significant postprocedural bleeding, or colonic muscle injuries of any grade.
DISCUSSION
0 (0%)
.67
0 (0)%
Network/dense
36 (94.7%)
28 (96.6%)
Irregular/sparse
0 (0%)
0 (0%)
Not recorded
2 (.1%)
1 (3.4%)
.78
Prior manipulationy Attempt at ER
16 (42.1%)
.003
11 (37.9%)
Heavy
11 (28.9%)
11 (37.9%)
Light
11 (28.9%)
7 (24.1%)
.009
Difficult access Yes
13 (34.2%)
No
25 (65.8%)
.20
8 (27.6%)
.56
21 (72.4%)
Lesion 75% luminal circumference Yes
12 (31.6%)
No
26 (68.4%)
<.001
6 (20.7%)
<.001
23 (79.3%)
Technique
Little is known about the incidence of recurrence after ER of large LSTs with follow-up longer than 12 to 18 months, which is rarely, if ever, reported. It is possible that the paucity of published data has led to misconceptions www.giejournal.org
EMR/hybrid ESD ESD
37 (97.4%) 1 (2.6%)
.04
27 (93.1%)
.22
2 (6.9%) (continued on the next page)
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 5
Risk factors for early and late adenoma recurrence
Emmanuel et al
TABLE 1. Continued Late Residual/recurrent P recurrence P adenoma (3-6 mo) value (≥12 mo) value Piecemeal resection Yes
37 (97.4%)
No
1 (2.6%)
<.001 26 (89.7%)
.003
3 (10.3%)
Adjunctive APC Yes
16 (42.1%)
<.001 10 (34.5%)
No
22 (57.9%)
19 (65.5%)
.003
Intraprocedural bleeding Yes
5 (13.2%)
No
33 (86.8%)
.75
1 (3.4%)
.15
28 (96.6%)
Muscle injury Yes
2 (5.3%)
No
36 (94.7%)
.29
2 (6.9%)
.13
27 (93.1%)
Resection defect closed Yes
2 (5.3%)
No
36 (94.7%)
.005
2 (6.9%)
.02
27 (93.1%)
Postprocedure adverse event Yes
2 (5.3%)
No
36 (94.7%)
.90
5 (17.2%)
.02
24 (82.8%)
Postprocedure bleeding Yes
0 (0%)
No
38 (100%)
.46
0 (0%)
.48
29 (100%)
Trainee performed Yes
2 (5.3%)
No
36 (94.7%)
.18
1 (3.4%)
.26
28 (96.6%)
High-grade dysplasia Yes
7 (18.4%)
No
31 (81.6%)
.83
7 (24.1%)
.91
22 (75.9%)
Is/Isp, Sessile/sub-pedunculated; LST G H, laterally spreading tumor granular homogeneous; LST G MN, laterally spreading tumor granular mixed-nodular type; LST NG, laterally spreading tumor non-granular; ER, endoscopic resection; ESD, endoscopic submucosal dissection; APC, argon plasma coagulation. *Highest grade pit or vascular pattern (Showa classification) visualized on lesion. yHeavy manipulation Z 6 biopsy samples, tattoo under base of lesion; Light manipulation Z <6 biopsy samples.
regarding recurrence as conventional understanding, according to most published series to date, is that almost all instances of recurrence will be detected within the first 6 months after ER.3 However, we found that recurrence is consistently detected at each successive period of surveillance after initial resection in patients who had no evidence of recurrence on previous surveillance examinations using magnification to examine the scar. Recurrence occurred for the first time between 24 and 35 months after ER in 6.4% of patients. Perhaps even more 6 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
interesting is that 5 of 37 patients (13.5%) developed recurrence more than 3 years after ER. The latter sample size is small and is perhaps an overestimation of the true incidence; however, it is clear that a significant incidence of recurrence is to be expected even several years after ER. Few consistent risk factors for recurrence after ER of large LSTs have been identified despite the current widespread use of colorectal ER. A systematic review and metaanalysis of recurrence after EMR identified only piecemeal resection as an independent predictor of recurrence. However, very few studies involving significant numbers of ER of large colorectal lesions were included, highlighting the current paucity of evidence.3 Also, risk factors tend to be reported either for overall recurrence or early recurrent or residual adenoma without analyzing differences between early and late recurrence or assessing the longerterm incidence of recurrence.2,5,7,18,19 Consistent with some other studies, we found that piecemeal resection and the use of APC were independently associated with recurrent or residual adenoma at first surveillance. However, none of the other risk factors identified in the seminal article from the Australian Colonic EMR (ACE) study (size >40 mm and intraprocedural bleeding) was identified as independent risk factors in our cohort. Lesion occupying 75% of the luminal circumference was the only other independent predictor of residual or recurrent adenoma. We speculate that 1 possible reason for these differences is the use of preand postresection examination using magnification to ensure complete resection. There is evidence that recurrence occurs less frequently with the use of magnification before resection and is also useful after resection.20,21 Another reason may be the use of ESD by achieving en bloc resection or hybrid ESD by achieving improved clearance of heavily scarred lesions from prior manipulation. The use of APC was independently associated with recurrent or residual adenoma at first surveillance. The ACE study reported similar findings but believe this is related to using ablation techniques on residual adenoma, which is unreliable. This group has subsequently reported on the use of cold snare avulsion to remove visible residual adenoma followed by snare-tip soft coagulation, which appears to achieve much better results.22 We agree with this principle and generally remove all macroscopically evident adenoma before using APC. However, APC was not used as part of a protocol or standard technique but rather at the discretion of the endoscopist. It is therefore most commonly used in difficult cases where profound fibrosis complicates the resection or in multiple piecemeal resections. In such cases there is a much higher likelihood of unidentified residual adenoma remaining, which explains the recurrence. Nevertheless, recognizing that when such methods are used there is a higher likelihood of residual or recurrent adenoma may help dictate surveillance strategies. www.giejournal.org
Emmanuel et al
Risk factors for early and late adenoma recurrence
TABLE 2. Independent risk factors for recurrent/residual adenoma and late recurrence showing factors included in the multiple logistic regression model using stepwise variable selection Residual/recurrent adenoma (3-6 mo) OR (95% CI)
Late recurrence (≥12 mo)
P value
OR (95% CI)
P value
Piecemeal resection
13.0 (1.7-98.4)
.01
4.4 (1.3-15.3)
.02
Lesion 75% luminal circumference
5.6 (2.4-12.9)
<.001
d
d
Use of APC
2.4 (1.1-5.1)
.02
d
d
d
d
6.3 (2.7-14.9)
<.001
Size >60 mm
OR, Odds ratio; CI, confidence interval; APC, argon plasma coagulation.
On the other hand, only piecemeal resection and a very large lesion size (>60 mm) were independent predictors of late recurrence 12 months after initial resection. Most large series have not differentiated between risk factors for residual/recurrent adenoma at first surveillance and later recurrence.2,4,5,7,19,20 Insight into potential different risk factors for residual/recurrent adenoma and late recurrence may help tailor patient specific surveillance strategies. Differences in lesion assessment and resection strategies between Eastern and Western experts are well recognized, which previously made results from Japanese expert centers less relevant to Western endoscopists.23-25 The strongest evidence of risk factors for recurrence in Western practice has come from the results of the Australian Colonic EMR study.2 However, ER in the West has progressed steadily with some expert centers incorporating ESD more frequently and perhaps better quality lesion assessment and ER with experience.9 For example, we have merged several aspects of Western and Eastern practice, routinely using magnification endoscopy for lesion assessment and increasingly using ESD.8 Furthermore, patient and lesion characteristics may vary considerably between regions, especially in terms of injudicious attempts at resection or heavy manipulation of lesions before they are referred to an expert center. Our findings will help to define the risk factors for recurrence and outcomes of treatment of recurrent lesions using advanced ER in a modern expert Western center involving a lesion-specific approach to resection that incorporates techniques such as ESD for both initial resection and treatment of large recurrent lesions. To our knowledge, this is the first such Western series. Recent studies and some Western experts have indicated that ESD is only indicated for very few lesions.25,26 However, this is often based only on the fact that relatively few invasive cancers ultimately prove to be minimal submucosal invasive cancers (<1000 mm) and therefore cured by ER. In our opinion, this is not the only consideration. First, although it may be true that relatively few lesions with invasive cancer meet common criteria for curative ER, it is not always possible to identify these lesions before resection because many lesions have the potential to harbor covert invasive cancers without obvious surface www.giejournal.org
characteristics such as type V pit.27 In our opinion it is ideal to resect all lesions with a significant risk of harboring covert invasive cancer en bloc. Second, curative resection is not the only consideration. One of the primary benefits of ESD is allowing accurate pathologic diagnosis and staging if invasive cancer is found. This is particularly relevant for rectal tumors where understaging can have devastating consequences, as can overstaging, resulting in radical resection with the high risks of associated morbidity with rectal surgery. Piecemeal EMR can result in missed diagnoses and also makes accurate staging difficult if not impossible. Third, en bloc resection has the added benefit of reducing recurrence. As our experience has accumulated, we have more recently moved toward treating most rectal lesions by ESD for this additional advantage. As further experience is gained, it may be feasible to similarly expand the indications for ESD elsewhere in the colon to help reduce recurrence further. Finally, ESD and hybrid ESD have proved useful in achieving resection in lesions complicated by profound fibrosis after previous manipulation or attempted resection. For these reasons, we believe the indications for ESD far exceed only cases where curative resection of invasive cancer can be achieved. Given that recurrent or residual adenoma is not infrequent after ER, it is important to define the actual outcome for patients with recurrence. Most series report only on the treatment of early recurrent or residual adenoma, but detailed outcomes are not always clear. We found it difficult to define when successful endoscopic treatment of recurrence has been achieved, because there is a high incidence of multiple recurrences after ER for recurrent or residual adenoma. Significantly, patients with recurrence have required a range of 2 to 7 ERs. Only 1 patient required 7 ERs, and endoscopic cure appears to have been achieved finally after en bloc resection by ESD of the last recurrent adenoma. The patient has now been free from further recurrence at 2 successive surveillance endoscopies. Because this was a very low rectal lesion, the only alternative would be abdominoperineal resection and permanent stoma; therefore, the patient has been satisfied with this result. Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 7
Risk factors for early and late adenoma recurrence
Emmanuel et al
- Awaiting ER, N=6 - Refused or unfit for ER, N=2
Recurrent adenoma N=66
Surgery N=4
- Awaiting surveillance, N=6 - Unfit for surveillance, N=2 - Lost to follow-up, N=2
Endoscopic resection N=54
No recurrence N=23
- Awaiting ER, N=1 - Unfit for ER, N=1
Recurrence N=21 (48%)
Surgery N=1
Awaiting surveillance N=1
Endoscopic resection N=18
No recurrence N=5
Awaiting ER N=1
Recurrence N=12 (71%)
Surgery N=0
Awaiting surveillance N=3
Endoscopic resection N=11
No recurrence N=4
Recurrence N=4 (50%)
- Awaiting ER, N=1 - Died, N=1
Endoscopic resection N=2
No recurrence N=0
Awaiting ER N=1
Recurrence N=2 (100%)
Endoscopic resection N=1 (No recurrence)
Figure 3. Flow diagram showing the outcomes of all recurrent adenomas. ER, Endoscopic resection.
Recurrence is challenging to treat, reflected in the more frequent use of ESD to achieve clearance for large (20 mm) recurrences. Nevertheless, there were no differences in overall adverse events or procedure-specific adverse events between ER of large recurrences and initial ER. This is in keeping with series from Japan reporting the use of ESD for treatment of recurrent or residual adenoma and confirms it is safe and useful in 8 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
an expert Western center given the necessary expertise and experience.10,11 This study is limited by its retrospective design, although clinical data for ERs is collected prospectively, and all consecutive patients were included. These results are from a U.K. tertiary referral center and may not be generalizable, although these large complex lesions should, in our opinion, be treated in high-volume expert www.giejournal.org
Emmanuel et al
Risk factors for early and late adenoma recurrence
TABLE 3. Comparison of characteristics, treatment strategy, and treatment outcomes between adenomas at initial resection and resection of recurrent adenomas ≥20 mm Lesion type
Mean size, mm
Initial resection
Recurrent adenoma
P value
52.3
28.6
<.001
NA (all 20 mm)
35 (38%)
NA
months (piecemeal resection, large lesions size >60 mm) are different. Recurrence is challenging to treat and often requires multiple ERs to achieve control, but surgery is very rarely required. ESD or hybrid ESD is often used to resect large recurrent lesions and can safely be used in expert centers with the necessary experience and expertise.
Size classification Diminutive Significant
58 (62%)
Technique, all lesions ESD/hybrid
140 (26%)
26 (28%)
EMR
394 (74%)
67 (72%)
ESD/hybrid
140 (26%)
23 (44%)
EMR
.73
Technique, large lesions (20 mm) .006
394 (74%)
29 (56%)
Adjunctive APC or coagulation*
83 (15%)
61 (66%)
<.001
Intraprocedural bleeding
82 (15%)
17 (18%)
.48
Colonic muscle injury, any degree
16 (3%)
1 (1%)
.29
Postprocedure bleeding
8 (2%)
1 (1%)
.75
Overall adverse events
2 (2%)
30 (6%)
.16
ESD, Endoscopic submucosal dissection; APC, argon plasma coagulation; NA, not applicable.
centers. In particular, not all units will have the expertise and experience to use ESD techniques in a significant proportion of cases. However, we believe that ESD is now an important component of any tertiary referral advanced ER practice and will be increasingly adopted by advanced interventional endoscopists in the West. We have chosen to analyze ER by various techniques together. However, this reflects the “real-world” approach adopted by a modern Western expert center, providing meaningful data on expected outcomes. The use of ESD may have influenced the proportion of recurrences in this study as a result of en bloc resection for some larger lesions. However, although Western experts are likely to increasingly incorporate ESD in their standard practice, it is unlikely that they would primarily treat all lesions by ESD. Therefore, it would not provide clinically useful data to subanalyze groups according to resection technique.25 In conclusion, recurrence consistently occurs at successive periods of surveillance after initial successful ER even beyond 3 years. Consistent risk factors for recurrence have been difficult to identify and may have changed as advanced practice in ER has changed with increasing experience in expert Western centers. Independent risk factors for residual or recurrent adenoma at first surveillance (piecemeal resection, use of APC, occupying 75% of luminal circumference) and for late recurrence 12 www.giejournal.org
REFERENCES 1. Hassan C, Repici A, Sharma P, et al. Efficacy and safety of endoscopic resection of large colorectal polyps: a systematic review and metaanalysis. Gut 2016;65:806-20. 2. Moss A, Williams SJ, Hourigan LF, et al. Long-term adenoma recurrence following wide-field endoscopic mucosal resection (WF-EMR) for advanced colonic mucosal neoplasia is infrequent: results and risk factors in 1000 cases from the Australian Colonic EMR (ACE) study. Gut 2015;64:57-65. 3. Belderbos TDG, Leenders M, Moons LMG, et al. Local recurrence after endoscopic mucosal resection of nonpedunculated colorectal lesions: systematic review and meta-analysis. Endoscopy 2014;46: 388-402. 4. Margagnoni G, Angeletti S, D’Ambra G, et al. Outcome and risk of recurrence for endoscopic resection of colonic superficial neoplastic lesions over 2 cm in diameter. Dig Liver Dis 2016;48:399-403. 5. Seidel J, Färber E, Baumbach R, et al. Complication and local recurrence rate after endoscopic resection of large high-risk colorectal adenomas of 3 cm in size. Int J Colorectal Dis 2016;31:603-11. 6. Longcroft-Wheaton G, Duku M, Mead R, et al. Risk stratification system for evaluation of complex polyps can predict outcomes of endoscopic mucosal resection. Dis Colon Rectum 2013;56: 960-6. 7. Knabe M, Pohl J, Gerges C, et al. Standardized long-term follow-up after endoscopic resection of large, nonpedunculated colorectal lesions: a prospective two-center study. Am J Gastroenterol 2014;109:183-9. 8. Emmanuel A, Gulati S, Burt M, et al. Combining eastern and western practices for safe and effective endoscopic resection of large complex colorectal lesions. Eur J Gastroenterol Hepatol 2018;30:506-13. 9. Emmanuel A, Gulati S, Burt M, et al. Using endoscopic submucosal dissection as a routine component of the standard treatment strategy for large and complex colorectal lesions in a Western tertiary referral unit. Dis Colon Rectum 2018;61:743-50. 10. Sakamoto T, Saito Y, Matsuda T, et al. Treatment strategy for recurrent or residual colorectal tumors after endoscopic resection. Surg Endosc 2011;25:255-60. 11. Kuroki Y, Hoteya S, Mitani T, et al. Endoscopic submucosal dissection for residual/locally recurrent lesions after endoscopic therapy for colorectal tumors. J Gastroenterol Hepatol 2010;25:1747-53. 12. Zhou P, Yao L, Qin X, et al. Endoscopic submucosal dissection for locally recurrent colorectal lesions after previous endoscopic mucosal resection. Dis Colon Rectum 2009;52:305-10. 13. Kim HG, Thosani N, Banerjee S, et al. Underwater endoscopic mucosal resection for recurrences after previous piecemeal resection of colorectal polyps (with video). Gastrointest Endosc 2014;80:1094-102. 14. Kudo S, Tamura S, Nakajima T, et al. Diagnosis of colorectal tumorous lesions by magnifying endoscopy. Gastrointest Endosc 1996;44:8-14. 15. Wada Y, Kudo S, Kashida H, et al. Diagnosis of colorectal lesions with the magnifying narrow-band imaging system. Gastrointest Endosc 2009;70:522-31. 16. Emmanuel A, Gulati S, Burt M, et al. Colorectal endoscopic submucosal dissection: patient selection and special considerations. Clin Exp Gastroenterol 2017;10:121-31.
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 9
Risk factors for early and late adenoma recurrence
Emmanuel et al
17. Matsumoto A, Tanaka S, Oba S, et al. Outcome of endoscopic submucosal dissection for colorectal tumors accompanied by fibrosis. Scand J Gastroenterol 2010;45:1329-37. 18. Tate DJ, Desomer L, Klein A, et al. Adenoma recurrence after piecemeal colonic EMR is predictable: the Sydney EMR recurrence tool. Gastrointest Endosc 2017;85:647-56. 19. Briedigkeit A, Sultanie O, Sido B, et al. Endoscopic mucosal resection of colorectal adenomas > 20 mm: risk factors for recurrence. World J Gastrointest Endosc 2016;8:276-81. 20. Oka S, Tanaka S, Saito Y, et al. Local recurrence after endoscopic resection for large colorectal neoplasia: a multicenter prospective study in Japan. Am J Gastroenterol 2015;110:697-707. 21. Cipolletta L, Bianco MA, Garofano ML, et al. Can magnification endoscopy detect residual adenoma after piecemeal resection of large sessile colorectal lesions to guide subsequent treatment? A prospective single-center study. Dis Colon Rectum 2009;52:1774-9. 22. Tate DJ, Bahin FF, Desomer L, et al. Cold-forceps avulsion with adjuvant snare-tip soft coagulation (CAST) is an effective and safe strategy
10 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
23.
24.
25.
26. 27.
for the management of non-lifting large laterally spreading colonic lesions. Endoscopy 2018;50:52-62. Bourke M. Current status of colonic endoscopic mucosal resection in the west and the interface with endoscopic submucosal dissection. Dig Endosc 2009;21(Suppl 1):S22-7. Lee EY, Bourke MJ. EMR should be the first-line treatment for large laterally spreading colorectal lesions. Gastrointest Endosc 2016;84: 326-8. Fuccio L, Repici A, Hassan C, et al. Why attempt en bloc resection of non-pedunculated colorectal adenomas? A systematic review of the prevalence of superficial submucosal invasive cancer after endoscopic submucosal dissection. Gut 2018;67:1464-74. Burgess NG, Bourke MJ. Endoscopic resection of colorectal lesions: the narrowing divide between East and West. Dig Endosc 2016;28:296-305. Burgess NG, Hourigan LF, Zanati SA, et al. Risk stratification for covert invasive cancer among patients referred for colonic endoscopic mucosal resection: a large multicenter cohort. Gastroenterology 2017;153:732-42.
www.giejournal.org
Risk factors for early and late adenoma recurrence after advanced colorectal endoscopic resection at an expert Western center Andrew Emmanuel, MBChB, MS,1,2 Christo Lapa, MD,1 Anil Ghosh, MBBS, PhD,1 Shraddha Gulati, MBBS,2 Margaret Burt, RN,1,2 Bu’Hussain Hayee, PhD,2 Amyn Haji, MA, MSc, MD1,2 London, United Kingdom
Background and Aims: Few large Western series examine risk factors for recurrence after endoscopic resection (ER) of large (20 mm) colorectal laterally spreading tumors. Recurrence beyond initial surveillance is seldom reported, and differences between residual/recurrent adenoma and late recurrence are not scrutinized. We report the incidence of recurrence at successive surveillance intervals, identify risk factors for recurrent/residual adenoma and late recurrence, and describe the outcomes of ER of recurrent adenomas. Methods: Recurrence was calculated for successive surveillance periods after colorectal ER. Multiple logistic regression was used to identify independent risk factors for recurrent/residual adenoma and late recurrence (12 months). Results: Six hundred twenty colorectal ERs were performed, and 456 eligible patients (98%) had completed 3- to 6-month surveillance. Residual/recurrent adenoma (3-6 months) was detected in 8.3%, at 12 months in 6.1%, between 24 and 36 months in 6.4%, and after 36 months in 13.5%. Independent risk factors for residual/recurrent adenoma were piecemeal resection (odds ratio [OR], 13.0; P Z .01), adjunctive argon plasma coagulation (OR, 2.4; P Z .01), and lesion occupying 75% of the luminal circumference (OR, 5.6; P < .001) and for late recurrence were lesion size >60 mm (OR, 6.3; P < .001) and piecemeal resection (OR, 4.4; P Z .04). Of 66 patients with recurrence, 5 required surgery, 8 left the treatment pathway, 20 are still receiving ER or surveillance, and 33 had ER with normal subsequent surveillance. Conclusions: Recurrence occurs at successive periods of surveillance after ER even beyond 3 years. Aside from piecemeal resection, risk factors for residual/recurrent adenoma and late recurrence are different. Recurrence can be challenging to treat, but surgery is rarely required. (Gastrointest Endosc 2019;-:1-10.)
Recurrence after endoscopic resection (ER) of large colorectal laterally spreading tumors (LSTs) is not infrequent.1,2 It is difficult to predict, and many patients who do not develop recurrence are subjected to frequent and Abbreviations: APC, argon plasma coagulation; ER, endoscopic resection; ESD, endoscopic submucosal dissection; LST, laterally spreading tumor; pEMR, piecemeal endoscopic mucosal resection. DISCLOSURE: All authors disclosed no financial relationships relevant to this publication. Copyright ª 2019 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 https://doi.org/10.1016/j.gie.2019.01.031 Received October 2, 2018. Accepted January 28, 2019. Current affiliations: Department of Colorectal Surgery (1), King’s Institute of Therapeutic Endoscopy (2), King’s College Hospital NHS Foundation Trust, London, UK. Reprint requests: Amyn Haji, King’s Institute of Therapeutic Endoscopy, King’s College Hospital NHS Foundation Trust, London SE59RS, UK.
www.giejournal.org
ultimately unnecessary surveillance procedures. Despite the now widespread use of ER, few large Western series examine the risk factors for recurrence. In those that do, diverse possible risk factors are reported with very few factors consistently demonstrated between studies.3 Furthermore, early recurrent or residual adenoma at the first surveillance is usually investigated, but factors associated with longer-term recurrence are almost never reported. Few Western data report the incidence of recurrence beyond 12 to 18 months of follow-up.2,4-7 Most large Western series to date have exclusively focused on EMR, but it is important to consider the evolving nature of advanced colorectal ER. In many expert Western centers, increasing experience has resulted in better lesion assessment and selection, higher quality ER, and an increased use of ESD as a routine component of the treatment strategy.2,8,9 Endoscopic treatment of recurrent lesions, although frequently feasible, can be extremely challenging, and Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 1
Risk factors for early and late adenoma recurrence
Emmanuel et al
there is little consensus for the resection strategy between studies.2,10-13 Given that radical surgery may be the only alternative option for these patients, it is important to define the outcomes for patients treated for recurrent adenomas after initial expert resection. In this study we report the short- and long-term incidence of recurrence after ER of large colorectal lesions and identify risk factors for recurrent or residual adenoma at the first surveillance and for late recurrence after 12 months of follow-up or longer. We also describe the outcomes of ER of recurrent adenomas after initial expert resection.
METHODS This was a retrospective study of prospectively collected data on ERs of large (20 mm) colorectal superficial neoplastic lesions between January 2010 and March 2018 at a tertiary referral center. Patients were referred to our service if the endoscopist performing the index procedure believed the lesion was beyond the capabilities of the referring institution as a result of large size, difficult location, challenging morphology, or previous failed attempts at resection. All patients with successful initial ER who had undergone at least the first surveillance colonoscopy were included. Those with noncurative ER of invasive cancer were excluded, even if they had declined radical surgery.
Procedures Our approach to assessment and resection has been previously described.8,9 Procedures were generally performed with the patient under conscious sedation and analgesia with intravenous midazolam and fentanyl administered by the endoscopist. Patients were discharged home on the day of the procedure. All assessments and procedures were performed by 2 experienced interventional endoscopists or senior endoscopy fellows under their direct supervision. All lesions were assessed before resection with chromoendsocopy using indigo carmine and magnification colonoscopy (model CF-H260AZL colonoscope [Olympus, Tokyo, Japan] and model EC-600Z colonoscope [Fujifilm, Düsseldorf, Germany]) and classified according to Kudo pit pattern and vascular pattern (Showa classification).14,15 Selected lesions were also evaluated with variable high-frequency miniprobe US (Fujifilm) to evaluate for submucosal invasion. All lesions were considered for ER except those with invasive cancer >T1 (Kudo type Vn pit pattern, sparse vascular pattern, at least muscularis propria invasion on high-frequency miniprobe colonoscopic US) for which surgery was advised. Techniques used for resection included ESD, hybrid ESD, and EMR or piecemeal EMR (pEMR). The decision for the resection technique was based on the patient and 2 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
lesion characteristics specific to each, bearing in mind our general considerations when undertaking ER.16 In brief, we generally use ESD where there is a suspicion of invasive cancer (Kudo type Vi pit, irregular or sparse narrow-band imaging) or with lesions with a significant risk of harboring invasive cancer such as LST granular mixed-nodular type, LST nongranular, or a pseudodepressed (IIc) component. We also often choose ESD or hybrid ESD to aid resection where there is profound submucosal fibrosis. Magnification did not specifically influence the selection of ESD or EMR but, in our opinion, makes the detection of type Vi pit or irregular vascular pattern easier to detect. If these features are discovered, ESD is the preferred resection technique. EMR and pEMR were performed using a sequential injection and resection technique. The submucosal injection consisted of succinylated gelatine (Gelofusine; B. Braun, Crissier, Switzerland) with adrenaline diluted to 1:100 000 and dyed with methylene blue or indigo carmine. A variety of snare types as appropriate to the lesion were used. ESD was performed using an Olympus PCF Q260JL with ST distal hood (Fujifilm) and the FlushKnife (Fujifilm). In brief, a standard technique was typically used consisting of a mucosal incision on the anal side of the lesion, trimming and flap creation, followed by submucosal dissection under the lesion. The mucosal incision was then completed around the lesion. Hybrid ESD was used to aid resection of some previously heavy manipulated lesions and consisted of variable ESD dissection to aid resection or snare capture of areas of submucosal fibrosis. The Erbe VIO 300D electrosurgery device (Erbe Elekromedezin, Tübingen, Germany) was used with appropriate settings depending on the resection technique. After resection, lesions were retrieved and submitted for histopathologic analysis. The resection bed was routinely examined again using magnification chromoendoscopy for any residual polyp. For lesions with profound submucosal fibrosis, argon plasma coagulation (APC) was sometimes used as an adjunctive measure after resection of visible adenoma (not to ablate visible residual tissue), although this was dependent on the endoscopist’s preference at the time of resection, and there was no specific protocol for the use of adjunctive APC. The presence of fibrosis was judged by the endoscopist and recorded but was not characterized in a standardized manner according to reported grading systems.17
Surveillance Patients underwent surveillance endoscopy at 3 to 6 months and 12 months and, if no recurrence was detected, colonoscopy at 36 months. After the 12-month surveillance, patients could be discharged to their referring hospital for local surveillance because our unit receives referrals form a wide geographic area. Surveillance up to 12 months was performed by members of the advanced ER team. The protocol for surveillance involved routine www.giejournal.org
Emmanuel et al
examination of the scar using magnification chromoendoscopy and magnification narrow-band imaging or blue light imaging (BLI). If there was no evidence of recurrent adenoma on magnification assessment, the scar was not routinely biopsy sampled.
Data collection Data were collected on patient demographics, lesion characteristics, details of the resection procedure, histopathology findings, findings at surveillance, and details of any recurrence and subsequent treatment. Reports of surveillance procedures performed at patients’ local institution beyond 12 months were also obtained. Recurrent/residual adenoma was defined as histologically confirmed recurrence at the site of previous resection at 3 to 6 months surveillance. Late recurrence was defined as adenoma at the site of previous resection at 12 months’ surveillance in patients who had no evidence of recurrent/ residual adenoma at the first (3-6 months) surveillance. Recurrence at each period of surveillance was defined as detection of recurrent adenoma at the scar from previous resection in patients who had no evidence of recurrence at any previous surveillance. For analysis of outcomes of recurrent adenomas, any treatment for recurrence after an initial apparently successful ER in our unit was included, regardless of any attempted prior resection or substantial manipulation (6 biopsy samples) at other institutions before referral.
Statistical analysis Results are reported using means and standard deviation for continuous variables and frequencies for categorical variables. Potential risk factors for recurrent/residual adenoma and for late recurrence were identified by univariate analysis using the t test to compare means and the c2 test to compare categorical variables. Multiple logistic regression using stepwise variable selection was performed for factors associated with recurrent/residual adenoma or late recurrence with P < .1 on univariate analysis to identify independent associations. Ethical approval for this study was granted by the National Research Ethics Committee.
Risk factors for early and late adenoma recurrence
mean age of the analyzed patients was 71 years. The mean lesion size was 53.7 mm (median, 50 mm; range, 20-160). Three hundred thirty-eight were resected using EMR or pEMR and 118 using ESD or hybrid ESD. Twelve lesions contained invasive cancer with curative ER (invasion <1000 mm from muscularis mucosa and absence of poor differentiation, tumor budding, or lymphovascular invasion). Fifty percent of lesions had been subjected to substantial manipulation before referral, and 34% of lesions resected using ESD or hybrid ESD had been subjected to prior attempted resection or substantial manipulation compared with 18% of those resected by EMR/pEMR (odds ratio [OR], 2.4; 95% confidence interval [CI], 1.5-3.6; P < .001).
Incidence of recurrence according to length of follow-up Residual or recurrent adenoma at 3 to 6 months was detected in 8.3% (95% CI, 6.0%-11.4%) (Fig. 2), at 12 months in 6.1% (95% CI, 3.5%-9.7%), between 24 and 36 months in 6.4% (95% CI, 2.6%-12.8%), and after 36 months in 5 of 37 patients (13.5%; 95% CI, 4.5%-28.8%).
Recurrent/residual adenoma at 3 to 6 months Univariate analysis (Table 1) showed residual/recurrent adenoma was significantly more frequent in those referred after prior attempts at ER (P < .001), lesions occupying 75% of the luminal circumference (P < .001), larger lesions (P Z .02), piecemeal resections (P < .001), EMR/pEMR or hybrid ESD compared with ESD (P Z .04), where adjunctive APC was used (P < .001), and in cases where the mucosal defect was closed with endoscopic clips (P Z .02). The multiple logistic regression model (Table 2) showed that independent predictors of residual/recurrent adenoma were piecemeal resection (OR, 13.0; 95% CI, 1.7-98.4; P Z .01), adjunctive use of APC (OR, 2.4; 95% CI, 1.1-5.0; P Z .02), and lesion occupying 75% of the luminal circumference (OR, 5.6; 95% CI, 2.4-12.9; P < .001). Large size was not an independent predictor of residual/recurrent adenoma.
Late (≥12 months) recurrence RESULTS ER was performed for 620 LSTs 20 mm over the study period. The flow diagram in Figure 1 shows the patients’ surveillance and reasons for exclusion. Forty-one ERs were performed for recurrent adenomas 20 mm. ER was initially unsuccessful in 5 (.9%). Invasive cancer with noncurative ER was found in 30 patients. Twenty-one patients were medically unfit or refused further surveillance, and 42 are awaiting surveillance that is not yet due. Six patients died of unrelated causes before the first surveillance. Of the remaining 464 eligible patients, 456 (98%) had completed the first surveillance. The www.giejournal.org
Two hundred sixty-two patients without residual/recurrent adenoma at first surveillance have undergone 12 months or longer surveillance. Late recurrence was more frequent in large lesions (P < .001), in those referred after prior attempts at ER (P < .009), lesions occupying 75% of the luminal circumference (P < .001), piecemeal resections (P < .003), in cases where the mucosal defect was closed with endoscopic clips (P Z .02), and in patients with form of postprocedure adverse event (P Z .02), although there was no association specifically with significant postprocedure bleeding (P Z .48) or any form of colonic wall muscle injury (P Z .13) (Table 1). The multiple logistic regression model (Table 2) showed that Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 3
Risk factors for early and late adenoma recurrence
Emmanuel et al
Endoscopic resection ≥20mm N=620
Eligible for 1st surveillance N=464 Lost to follow-up N=8 1st (3-6 month) surveillance N=456 (98%)
Eligible for 2nd surveillance N=303
Ineligible for 1st surveillance - Procedure due to recurrence, n=41 - Invasive cancer, non curative ER, n=30 - Unsuccessful ER, n=5 - Non-neoplastic histology, n=9 - Died before surveillance, n=6 - Surveillance not due, n=42 - Surgery for other pathology, n=2 - Not fit for surveillance, n=7 - Refused surveillance, n=14
Ineligible for 2nd surveillance -
Died before surveillance, n=9 Surveillance not due, n=93 Surgery for recurrence, n=1 Surgery for other pathology, n=6 - Not fit for surveillance, n=18 - Refused surveillance or asked for local hospital surveillance, n=26
Lost to follow-up N=14 2nd (12 month) surveillance N=289 (95%)
3rd (18-35 month) surveillance N=173
>36 months surveillance N=58
Figure 1. Flow diagram showing follow-up of patients after endoscopic resection (ER).
independent predictors of late recurrence were lesion size >60 mm (OR, 6.3; 95% CI, 2.7-14.9; P < .001) and piecemeal resection (OR, 4.4; 95% CI, 1.3-15.3; P Z .02).
Outcomes of treatment of recurrent adenomas Sixty-six patients developed recurrent adenomas (Fig. 3). Overall, 5 required surgery, and the rest were deemed suitable for further endoscopic treatment, although not all accepted or were medically fit for treatment. During the course of treatment and surveillance, 8 patients left the treatment pathway because of lack of medical fitness, death, or inability to contact. Twenty patients are still receiving endoscopic treatment or await surveillance after endoscopic treatment, whereas in 33 patients endoscopic cure 4 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
appears to have been successful with normal subsequent surveillance. Only 1 patient, who originally had a 40-mm LST (tubulovillous adenoma) resected by pEMR, developed invasive cancer in a recurrent lesion, although the patient’s treatment for recurrence had been significantly delayed because of treatment for other medical problems. Reassessing the lesion showed evidence of type V Kudo pit pattern, and surgery was recommended. The interval between the first detection of recurrence and surgery was 35 months. After right hemicolectomy the cancer was staged as T2 N0. Of the recurrent adenomas that have been treated, 97% were tubulovillous or tubular adenomas, 3% were serrated adenomas, only 1 contained high-grade dysplasia, and the remainder were low-grade dysplasia. Patients treated www.giejournal.org
Emmanuel et al
3-6 month surveillance N=456
Risk factors for early and late adenoma recurrence
Recurrence N=38 (8.3%)
Late Residual/recurrent P recurrence P adenoma (3-6 mo) value (≥12 mo) value
No recurrence N=418
12 month surveillance N=262
Age
Recurrence N=16 (6.1%)
No recurrence N=245 24-36 months surveillance N=109
TABLE 1. Exploratory univariate analysis of factors potentially associated with recurrent/residual adenoma and late recurrence
Recurrence N=7 (6.4%)
No recurrence N=102
<70 y
15 (39.5%)
70-79 y
14 (36.8%)
13 (44.8%)
80 y
9 (23.7%)
7 (24.1%)
9 (31.0%)
.20
Size 20-30 mm
3 (1.1%)
31-40 mm
6 (15.8%)
41-60 mm
12 (31.6%)
5 (17.2%)
>60 mm
17 (44.7%)
20 (69.1%)
.02
1 (3.4%)
<.001
3 (10.3%)
Location Right-sided colon
12 (31.6%)
Left-sided colon
8 (21.1%)
5 (17.2%)
Rectum/ rectosigmoid
18 (47.4%)
11 (37.9%)
.47
13 (44.8%)
.36
Morphology Is/Isp
>36 months surveillance N=37
.94
Recurrence N=5 (13.5%)
2 (.1%)
.06
1 (3.4%)
LST G H
21 (55.3%)
15 (51.7%)
LST G MN
15 (39.5%)
12 (41.4%)
0 (0%)
1 (3.4%)
LST NG
.07
Kudo pit pattern*
No recurrence N=32 Figure 2. Flow diagram showing the incidence of first-detected recurrence at successive periods of surveillance.
I or II
0 (0%)
III or IV
.47
0 (0%)
36 (94.7%)
27 (93.1%)
V
0 (0%)
1 (3.4%)
Not recorded
2 (.1%)
1 (3.4%)
.81
Vascular pattern* Faint/normal
with ER for recurrent adenoma have required a median of 2 ERs (range, 2-7). Table 3 shows a comparison of lesions at initial treatment and treatment for recurrent lesions (all instances of ER for recurrence included, including multiple treatments of repeat recurrence). For lesions 20 mm, ESD or hybrid ESD was used significantly more frequently to resect recurrent lesions (44% vs 26%, P Z .006). Adjunctive use of APC or soft coagulation was used significantly more frequently after resection of recurrent adenomas (P < .001). There were no significant differences in overall adverse events, intraprocedural bleeding, significant postprocedural bleeding, or colonic muscle injuries of any grade.
DISCUSSION
0 (0%)
.67
0 (0)%
Network/dense
36 (94.7%)
28 (96.6%)
Irregular/sparse
0 (0%)
0 (0%)
Not recorded
2 (.1%)
1 (3.4%)
.78
Prior manipulationy Attempt at ER
16 (42.1%)
.003
11 (37.9%)
Heavy
11 (28.9%)
11 (37.9%)
Light
11 (28.9%)
7 (24.1%)
.009
Difficult access Yes
13 (34.2%)
No
25 (65.8%)
.20
8 (27.6%)
.56
21 (72.4%)
Lesion 75% luminal circumference Yes
12 (31.6%)
No
26 (68.4%)
<.001
6 (20.7%)
<.001
23 (79.3%)
Technique
Little is known about the incidence of recurrence after ER of large LSTs with follow-up longer than 12 to 18 months, which is rarely, if ever, reported. It is possible that the paucity of published data has led to misconceptions www.giejournal.org
EMR/hybrid ESD ESD
37 (97.4%) 1 (2.6%)
.04
27 (93.1%)
.22
2 (6.9%) (continued on the next page)
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 5
Risk factors for early and late adenoma recurrence
Emmanuel et al
TABLE 1. Continued Late Residual/recurrent P recurrence P adenoma (3-6 mo) value (≥12 mo) value Piecemeal resection Yes
37 (97.4%)
No
1 (2.6%)
<.001 26 (89.7%)
.003
3 (10.3%)
Adjunctive APC Yes
16 (42.1%)
<.001 10 (34.5%)
No
22 (57.9%)
19 (65.5%)
.003
Intraprocedural bleeding Yes
5 (13.2%)
No
33 (86.8%)
.75
1 (3.4%)
.15
28 (96.6%)
Muscle injury Yes
2 (5.3%)
No
36 (94.7%)
.29
2 (6.9%)
.13
27 (93.1%)
Resection defect closed Yes
2 (5.3%)
No
36 (94.7%)
.005
2 (6.9%)
.02
27 (93.1%)
Postprocedure adverse event Yes
2 (5.3%)
No
36 (94.7%)
.90
5 (17.2%)
.02
24 (82.8%)
Postprocedure bleeding Yes
0 (0%)
No
38 (100%)
.46
0 (0%)
.48
29 (100%)
Trainee performed Yes
2 (5.3%)
No
36 (94.7%)
.18
1 (3.4%)
.26
28 (96.6%)
High-grade dysplasia Yes
7 (18.4%)
No
31 (81.6%)
.83
7 (24.1%)
.91
22 (75.9%)
Is/Isp, Sessile/sub-pedunculated; LST G H, laterally spreading tumor granular homogeneous; LST G MN, laterally spreading tumor granular mixed-nodular type; LST NG, laterally spreading tumor non-granular; ER, endoscopic resection; ESD, endoscopic submucosal dissection; APC, argon plasma coagulation. *Highest grade pit or vascular pattern (Showa classification) visualized on lesion. yHeavy manipulation Z 6 biopsy samples, tattoo under base of lesion; Light manipulation Z <6 biopsy samples.
regarding recurrence as conventional understanding, according to most published series to date, is that almost all instances of recurrence will be detected within the first 6 months after ER.3 However, we found that recurrence is consistently detected at each successive period of surveillance after initial resection in patients who had no evidence of recurrence on previous surveillance examinations using magnification to examine the scar. Recurrence occurred for the first time between 24 and 35 months after ER in 6.4% of patients. Perhaps even more 6 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
interesting is that 5 of 37 patients (13.5%) developed recurrence more than 3 years after ER. The latter sample size is small and is perhaps an overestimation of the true incidence; however, it is clear that a significant incidence of recurrence is to be expected even several years after ER. Few consistent risk factors for recurrence after ER of large LSTs have been identified despite the current widespread use of colorectal ER. A systematic review and metaanalysis of recurrence after EMR identified only piecemeal resection as an independent predictor of recurrence. However, very few studies involving significant numbers of ER of large colorectal lesions were included, highlighting the current paucity of evidence.3 Also, risk factors tend to be reported either for overall recurrence or early recurrent or residual adenoma without analyzing differences between early and late recurrence or assessing the longerterm incidence of recurrence.2,5,7,18,19 Consistent with some other studies, we found that piecemeal resection and the use of APC were independently associated with recurrent or residual adenoma at first surveillance. However, none of the other risk factors identified in the seminal article from the Australian Colonic EMR (ACE) study (size >40 mm and intraprocedural bleeding) was identified as independent risk factors in our cohort. Lesion occupying 75% of the luminal circumference was the only other independent predictor of residual or recurrent adenoma. We speculate that 1 possible reason for these differences is the use of preand postresection examination using magnification to ensure complete resection. There is evidence that recurrence occurs less frequently with the use of magnification before resection and is also useful after resection.20,21 Another reason may be the use of ESD by achieving en bloc resection or hybrid ESD by achieving improved clearance of heavily scarred lesions from prior manipulation. The use of APC was independently associated with recurrent or residual adenoma at first surveillance. The ACE study reported similar findings but believe this is related to using ablation techniques on residual adenoma, which is unreliable. This group has subsequently reported on the use of cold snare avulsion to remove visible residual adenoma followed by snare-tip soft coagulation, which appears to achieve much better results.22 We agree with this principle and generally remove all macroscopically evident adenoma before using APC. However, APC was not used as part of a protocol or standard technique but rather at the discretion of the endoscopist. It is therefore most commonly used in difficult cases where profound fibrosis complicates the resection or in multiple piecemeal resections. In such cases there is a much higher likelihood of unidentified residual adenoma remaining, which explains the recurrence. Nevertheless, recognizing that when such methods are used there is a higher likelihood of residual or recurrent adenoma may help dictate surveillance strategies. www.giejournal.org
Emmanuel et al
Risk factors for early and late adenoma recurrence
TABLE 2. Independent risk factors for recurrent/residual adenoma and late recurrence showing factors included in the multiple logistic regression model using stepwise variable selection Residual/recurrent adenoma (3-6 mo) OR (95% CI)
Late recurrence (≥12 mo)
P value
OR (95% CI)
P value
Piecemeal resection
13.0 (1.7-98.4)
.01
4.4 (1.3-15.3)
.02
Lesion 75% luminal circumference
5.6 (2.4-12.9)
<.001
d
d
Use of APC
2.4 (1.1-5.1)
.02
d
d
d
d
6.3 (2.7-14.9)
<.001
Size >60 mm
OR, Odds ratio; CI, confidence interval; APC, argon plasma coagulation.
On the other hand, only piecemeal resection and a very large lesion size (>60 mm) were independent predictors of late recurrence 12 months after initial resection. Most large series have not differentiated between risk factors for residual/recurrent adenoma at first surveillance and later recurrence.2,4,5,7,19,20 Insight into potential different risk factors for residual/recurrent adenoma and late recurrence may help tailor patient specific surveillance strategies. Differences in lesion assessment and resection strategies between Eastern and Western experts are well recognized, which previously made results from Japanese expert centers less relevant to Western endoscopists.23-25 The strongest evidence of risk factors for recurrence in Western practice has come from the results of the Australian Colonic EMR study.2 However, ER in the West has progressed steadily with some expert centers incorporating ESD more frequently and perhaps better quality lesion assessment and ER with experience.9 For example, we have merged several aspects of Western and Eastern practice, routinely using magnification endoscopy for lesion assessment and increasingly using ESD.8 Furthermore, patient and lesion characteristics may vary considerably between regions, especially in terms of injudicious attempts at resection or heavy manipulation of lesions before they are referred to an expert center. Our findings will help to define the risk factors for recurrence and outcomes of treatment of recurrent lesions using advanced ER in a modern expert Western center involving a lesion-specific approach to resection that incorporates techniques such as ESD for both initial resection and treatment of large recurrent lesions. To our knowledge, this is the first such Western series. Recent studies and some Western experts have indicated that ESD is only indicated for very few lesions.25,26 However, this is often based only on the fact that relatively few invasive cancers ultimately prove to be minimal submucosal invasive cancers (<1000 mm) and therefore cured by ER. In our opinion, this is not the only consideration. First, although it may be true that relatively few lesions with invasive cancer meet common criteria for curative ER, it is not always possible to identify these lesions before resection because many lesions have the potential to harbor covert invasive cancers without obvious surface www.giejournal.org
characteristics such as type V pit.27 In our opinion it is ideal to resect all lesions with a significant risk of harboring covert invasive cancer en bloc. Second, curative resection is not the only consideration. One of the primary benefits of ESD is allowing accurate pathologic diagnosis and staging if invasive cancer is found. This is particularly relevant for rectal tumors where understaging can have devastating consequences, as can overstaging, resulting in radical resection with the high risks of associated morbidity with rectal surgery. Piecemeal EMR can result in missed diagnoses and also makes accurate staging difficult if not impossible. Third, en bloc resection has the added benefit of reducing recurrence. As our experience has accumulated, we have more recently moved toward treating most rectal lesions by ESD for this additional advantage. As further experience is gained, it may be feasible to similarly expand the indications for ESD elsewhere in the colon to help reduce recurrence further. Finally, ESD and hybrid ESD have proved useful in achieving resection in lesions complicated by profound fibrosis after previous manipulation or attempted resection. For these reasons, we believe the indications for ESD far exceed only cases where curative resection of invasive cancer can be achieved. Given that recurrent or residual adenoma is not infrequent after ER, it is important to define the actual outcome for patients with recurrence. Most series report only on the treatment of early recurrent or residual adenoma, but detailed outcomes are not always clear. We found it difficult to define when successful endoscopic treatment of recurrence has been achieved, because there is a high incidence of multiple recurrences after ER for recurrent or residual adenoma. Significantly, patients with recurrence have required a range of 2 to 7 ERs. Only 1 patient required 7 ERs, and endoscopic cure appears to have been achieved finally after en bloc resection by ESD of the last recurrent adenoma. The patient has now been free from further recurrence at 2 successive surveillance endoscopies. Because this was a very low rectal lesion, the only alternative would be abdominoperineal resection and permanent stoma; therefore, the patient has been satisfied with this result. Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 7
Risk factors for early and late adenoma recurrence
Emmanuel et al
- Awaiting ER, N=6 - Refused or unfit for ER, N=2
Recurrent adenoma N=66
Surgery N=4
- Awaiting surveillance, N=6 - Unfit for surveillance, N=2 - Lost to follow-up, N=2
Endoscopic resection N=54
No recurrence N=23
- Awaiting ER, N=1 - Unfit for ER, N=1
Recurrence N=21 (48%)
Surgery N=1
Awaiting surveillance N=1
Endoscopic resection N=18
No recurrence N=5
Awaiting ER N=1
Recurrence N=12 (71%)
Surgery N=0
Awaiting surveillance N=3
Endoscopic resection N=11
No recurrence N=4
Recurrence N=4 (50%)
- Awaiting ER, N=1 - Died, N=1
Endoscopic resection N=2
No recurrence N=0
Awaiting ER N=1
Recurrence N=2 (100%)
Endoscopic resection N=1 (No recurrence)
Figure 3. Flow diagram showing the outcomes of all recurrent adenomas. ER, Endoscopic resection.
Recurrence is challenging to treat, reflected in the more frequent use of ESD to achieve clearance for large (20 mm) recurrences. Nevertheless, there were no differences in overall adverse events or procedure-specific adverse events between ER of large recurrences and initial ER. This is in keeping with series from Japan reporting the use of ESD for treatment of recurrent or residual adenoma and confirms it is safe and useful in 8 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
an expert Western center given the necessary expertise and experience.10,11 This study is limited by its retrospective design, although clinical data for ERs is collected prospectively, and all consecutive patients were included. These results are from a U.K. tertiary referral center and may not be generalizable, although these large complex lesions should, in our opinion, be treated in high-volume expert www.giejournal.org
Emmanuel et al
Risk factors for early and late adenoma recurrence
TABLE 3. Comparison of characteristics, treatment strategy, and treatment outcomes between adenomas at initial resection and resection of recurrent adenomas ≥20 mm Lesion type
Mean size, mm
Initial resection
Recurrent adenoma
P value
52.3
28.6
<.001
NA (all 20 mm)
35 (38%)
NA
months (piecemeal resection, large lesions size >60 mm) are different. Recurrence is challenging to treat and often requires multiple ERs to achieve control, but surgery is very rarely required. ESD or hybrid ESD is often used to resect large recurrent lesions and can safely be used in expert centers with the necessary experience and expertise.
Size classification Diminutive Significant
58 (62%)
Technique, all lesions ESD/hybrid
140 (26%)
26 (28%)
EMR
394 (74%)
67 (72%)
ESD/hybrid
140 (26%)
23 (44%)
EMR
.73
Technique, large lesions (20 mm) .006
394 (74%)
29 (56%)
Adjunctive APC or coagulation*
83 (15%)
61 (66%)
<.001
Intraprocedural bleeding
82 (15%)
17 (18%)
.48
Colonic muscle injury, any degree
16 (3%)
1 (1%)
.29
Postprocedure bleeding
8 (2%)
1 (1%)
.75
Overall adverse events
2 (2%)
30 (6%)
.16
ESD, Endoscopic submucosal dissection; APC, argon plasma coagulation; NA, not applicable.
centers. In particular, not all units will have the expertise and experience to use ESD techniques in a significant proportion of cases. However, we believe that ESD is now an important component of any tertiary referral advanced ER practice and will be increasingly adopted by advanced interventional endoscopists in the West. We have chosen to analyze ER by various techniques together. However, this reflects the “real-world” approach adopted by a modern Western expert center, providing meaningful data on expected outcomes. The use of ESD may have influenced the proportion of recurrences in this study as a result of en bloc resection for some larger lesions. However, although Western experts are likely to increasingly incorporate ESD in their standard practice, it is unlikely that they would primarily treat all lesions by ESD. Therefore, it would not provide clinically useful data to subanalyze groups according to resection technique.25 In conclusion, recurrence consistently occurs at successive periods of surveillance after initial successful ER even beyond 3 years. Consistent risk factors for recurrence have been difficult to identify and may have changed as advanced practice in ER has changed with increasing experience in expert Western centers. Independent risk factors for residual or recurrent adenoma at first surveillance (piecemeal resection, use of APC, occupying 75% of luminal circumference) and for late recurrence 12 www.giejournal.org
REFERENCES 1. Hassan C, Repici A, Sharma P, et al. Efficacy and safety of endoscopic resection of large colorectal polyps: a systematic review and metaanalysis. Gut 2016;65:806-20. 2. Moss A, Williams SJ, Hourigan LF, et al. Long-term adenoma recurrence following wide-field endoscopic mucosal resection (WF-EMR) for advanced colonic mucosal neoplasia is infrequent: results and risk factors in 1000 cases from the Australian Colonic EMR (ACE) study. Gut 2015;64:57-65. 3. Belderbos TDG, Leenders M, Moons LMG, et al. Local recurrence after endoscopic mucosal resection of nonpedunculated colorectal lesions: systematic review and meta-analysis. Endoscopy 2014;46: 388-402. 4. Margagnoni G, Angeletti S, D’Ambra G, et al. Outcome and risk of recurrence for endoscopic resection of colonic superficial neoplastic lesions over 2 cm in diameter. Dig Liver Dis 2016;48:399-403. 5. Seidel J, Färber E, Baumbach R, et al. Complication and local recurrence rate after endoscopic resection of large high-risk colorectal adenomas of 3 cm in size. Int J Colorectal Dis 2016;31:603-11. 6. Longcroft-Wheaton G, Duku M, Mead R, et al. Risk stratification system for evaluation of complex polyps can predict outcomes of endoscopic mucosal resection. Dis Colon Rectum 2013;56: 960-6. 7. Knabe M, Pohl J, Gerges C, et al. Standardized long-term follow-up after endoscopic resection of large, nonpedunculated colorectal lesions: a prospective two-center study. Am J Gastroenterol 2014;109:183-9. 8. Emmanuel A, Gulati S, Burt M, et al. Combining eastern and western practices for safe and effective endoscopic resection of large complex colorectal lesions. Eur J Gastroenterol Hepatol 2018;30:506-13. 9. Emmanuel A, Gulati S, Burt M, et al. Using endoscopic submucosal dissection as a routine component of the standard treatment strategy for large and complex colorectal lesions in a Western tertiary referral unit. Dis Colon Rectum 2018;61:743-50. 10. Sakamoto T, Saito Y, Matsuda T, et al. Treatment strategy for recurrent or residual colorectal tumors after endoscopic resection. Surg Endosc 2011;25:255-60. 11. Kuroki Y, Hoteya S, Mitani T, et al. Endoscopic submucosal dissection for residual/locally recurrent lesions after endoscopic therapy for colorectal tumors. J Gastroenterol Hepatol 2010;25:1747-53. 12. Zhou P, Yao L, Qin X, et al. Endoscopic submucosal dissection for locally recurrent colorectal lesions after previous endoscopic mucosal resection. Dis Colon Rectum 2009;52:305-10. 13. Kim HG, Thosani N, Banerjee S, et al. Underwater endoscopic mucosal resection for recurrences after previous piecemeal resection of colorectal polyps (with video). Gastrointest Endosc 2014;80:1094-102. 14. Kudo S, Tamura S, Nakajima T, et al. Diagnosis of colorectal tumorous lesions by magnifying endoscopy. Gastrointest Endosc 1996;44:8-14. 15. Wada Y, Kudo S, Kashida H, et al. Diagnosis of colorectal lesions with the magnifying narrow-band imaging system. Gastrointest Endosc 2009;70:522-31. 16. Emmanuel A, Gulati S, Burt M, et al. Colorectal endoscopic submucosal dissection: patient selection and special considerations. Clin Exp Gastroenterol 2017;10:121-31.
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 9
Risk factors for early and late adenoma recurrence
Emmanuel et al
17. Matsumoto A, Tanaka S, Oba S, et al. Outcome of endoscopic submucosal dissection for colorectal tumors accompanied by fibrosis. Scand J Gastroenterol 2010;45:1329-37. 18. Tate DJ, Desomer L, Klein A, et al. Adenoma recurrence after piecemeal colonic EMR is predictable: the Sydney EMR recurrence tool. Gastrointest Endosc 2017;85:647-56. 19. Briedigkeit A, Sultanie O, Sido B, et al. Endoscopic mucosal resection of colorectal adenomas > 20 mm: risk factors for recurrence. World J Gastrointest Endosc 2016;8:276-81. 20. Oka S, Tanaka S, Saito Y, et al. Local recurrence after endoscopic resection for large colorectal neoplasia: a multicenter prospective study in Japan. Am J Gastroenterol 2015;110:697-707. 21. Cipolletta L, Bianco MA, Garofano ML, et al. Can magnification endoscopy detect residual adenoma after piecemeal resection of large sessile colorectal lesions to guide subsequent treatment? A prospective single-center study. Dis Colon Rectum 2009;52:1774-9. 22. Tate DJ, Bahin FF, Desomer L, et al. Cold-forceps avulsion with adjuvant snare-tip soft coagulation (CAST) is an effective and safe strategy
10 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
23.
24.
25.
26. 27.
for the management of non-lifting large laterally spreading colonic lesions. Endoscopy 2018;50:52-62. Bourke M. Current status of colonic endoscopic mucosal resection in the west and the interface with endoscopic submucosal dissection. Dig Endosc 2009;21(Suppl 1):S22-7. Lee EY, Bourke MJ. EMR should be the first-line treatment for large laterally spreading colorectal lesions. Gastrointest Endosc 2016;84: 326-8. Fuccio L, Repici A, Hassan C, et al. Why attempt en bloc resection of non-pedunculated colorectal adenomas? A systematic review of the prevalence of superficial submucosal invasive cancer after endoscopic submucosal dissection. Gut 2018;67:1464-74. Burgess NG, Bourke MJ. Endoscopic resection of colorectal lesions: the narrowing divide between East and West. Dig Endosc 2016;28:296-305. Burgess NG, Hourigan LF, Zanati SA, et al. Risk stratification for covert invasive cancer among patients referred for colonic endoscopic mucosal resection: a large multicenter cohort. Gastroenterology 2017;153:732-42.
www.giejournal.org