- Email: [email protected]
Prophylactic endoscopic coagulation to prevent delayed post-EMR bleeding in the colorectum: a prospective randomized controlled trial (with videos)
ORIGINAL ARTICLE
Prophylactic endoscopic coagulation to prevent delayed post-EMR bleeding in the colorectum: a prospective randomized controlled trial (with videos) Hyun Seok Lee, MD, PhD,1,2 Seong Woo Jeon, MD, PhD,1,2 Yong Hwan Kwon, MD, PhD,1,2 Su Youn Nam, MD, PhD,2 Seonghwan Shin, MD,2 Ryanghi Kim, MD,2 Sohyun Ahn, MD2 Daegu, South Korea
GRAPHICAL ABSTRACT
Background and Aims: Post-EMR bleeding (PEB) is the most common adverse event after EMR. However, there are no established endoscopic methods for the prevention of PEB. This study aimed to investigate whether prophylactic endoscopic coagulation (PEC) using coagulation probes reduces the incidence of overall delayed PEB. Methods: We performed a randomized controlled study of patients undergoing EMR for large (10 mm) sessile lesions and laterally spreading tumors. Patients were randomized 1:1 to the EMR with coagulation group (n Z 285) or EMR (control) group (n Z 285). Immediate bleeding during colon EMR or clean-based ulcer after EMR was excluded. Clinically significant PEB was defined as bleeding requiring endoscopic hemostasis, hospitalization, or a decrease in the hemoglobin level >2 g/dL. Results: A total of 569 patients were analyzed. The incidence of overall PEB was significantly lower in the EMR with coagulation group than in the control group (12.6% [36/285] vs 18.7% [53/284], P Z .048). However, this was largely because of a reduction in minor bleeding. There was no difference in clinically significant PEB (1.8% [5/285] vs 3.2% [9/284], P Z .276). Rectal location was a risk factor associated with overall PEB (odds ratio, 1.256; 95% confidence interval, 1.12-1.41; P < .001). Conclusions: Although this study found reduced PEB with prophylactic cautery of visible vessels, this was largely because of a reduction in minor bleeding with no benefit observed for clinically significant bleeding. Overall, PEB was more frequent with rectal lesions. (Clinical trial registration number: KCT0000779.) (Gastrointest Endosc 2019;-:1-10.)
(footnotes appear on last page of article)
Polypectomy of colorectal polyps reduces the incidence and mortality of colorectal cancer.1 However, polypectomy is associated with adverse events, including post-EMR bleeding (PEB).2 PEB can occur during (immediate) or after (delayed) the procedure.3 Based on various reports,
www.giejournal.org
delayed PEB could occur in approximately .3% to 15.1% of cases.4-10 Among the risk factors for delayed PEB, size and location of polyps are the most widely known and well-defined risk factors.11,12 However, many risk factors remain equivocal and inconsistent. Delayed PEB can be
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 1
Prophylactic endoscopic coagulation to prevent delayed PPB
Lee et al
serious because it can result in hospitalization, repeated colonoscopy, and blood transfusion.13 Clinically minor PEB might cause an unplanned visit to an outpatient clinic or a repeated check of the hemoglobin level but would be ultimately identified as clinically nonsignificant compared with severe PEB. Closure of EMR sites with clips is effective in reducing PEB, but this technique is time-consuming and costly.14 A retrospective study showed that prophylactic endoscopic coagulation (PEC) might prevent delayed PEB after endoscopic resection of superficial neoplastic lesions in the stomach.15 PEC is easy to perform and widely applicable. It can be applied to postpolypectomy defects of any size, unlike endoscopic clip closure.16 However, there are some concerns about perforation caused by PEC in the colorectum that are not concerning in the stomach. A previous prospective study did not show the efficacy of PEC to reduce the incidence of delayed PEB on colorectal lesions larger than 20 mm.16 Our hypothesis was that if a lesion 10 mm was included and the number of subjects increased accordingly, the result could show a significant reduction of delayed PEB. Thus, we performed a prospective randomized controlled trial to determine whether PEC using coagulation probes reduces the incidence of delayed PEB after resection of large (10 mm) sessile or laterally spreading colorectal lesions and to evaluate the incidence of perforation after PEC.
METHODS Patients This randomized controlled trial was conducted at Kyungpook National University Hospital between June 2013 and July 2017. Patients undergoing colonoscopic EMR for large (10 mm) sessile lesions and laterally spreading tumors were considered eligible for enrollment. For patients with more than 1 polyp, all patients with postEMR ulcers were enrolled in the study. Written informed consent was obtained from each patient. Randomization of all patients was performed using a table of random numbers generated by a computer program. The protocol was registered with the Clinical Research Information Service (a representative clinical trials registry platform in Korea) before enrolling the first participant (registration number KCT0000779). This study was conducted after obtaining ethical approval from the Kyungpook National University Hospital internal review board (KNUMC_12-1037).
Colonoscopic EMR Patients were assigned randomly to 1 of 2 treatment groups in a 1:1 fashion. Patients were blinded to their intervention apart from their sedation. Exclusion criteria were polyp diameter <10 mm, a pedunculated polyp, immediate 2 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
bleeding within 1 minute after EMR, an ulcer with a clean base without erythema or visible vessels, bleeding from other GI sites except for colorectum, incompletely resected lesion, combined with other procedures such as epinephrine injection or endoscopic clipping, injury of the proper muscular layer (suspected or confirmed), and patient with coagulopathy, hematologic disease, endstage renal disease, thrombocytopenia, or a history of prior PEB. Patients on antiplatelet agents were asked to stop taking them 5 to 7 days before the procedure and resume up to 48 hours after the procedure. Patients on warfarin were advised to stop taking it 5 days before the procedure and resume on the day after the procedure.17 Before colonoscopic examination, the bowel was cleansed with 2 L polyethylene glycol plus ascorbic acid solution. Intravenous sedation was performed using midazolam, and pethidine was used to reduce pain. Colonoscopic examinations were performed with a conventional video colonoscope (CF-H260AL; Olympus Optical Co, Tokyo, Japan). All colonoscopic EMR procedures were performed by experienced endoscopists with more than 5 years of experience of performing therapeutic endoscopy or endoscopic fellows under direct supervision. These endoscopists had performed PEC routinely in real clinical practice for lesions believed to be at high risk of PEB. Colonic insufflation was performed with air. The size, location, number, and macroscopic form of the polyps were documented. The gross morphologies of the polyps were classified using the updated Paris classification as follows: sessile (Is) lesions were defined as raised lesions without evidence of a stalk or pedicle, and subpedunculated (Isp) lesions were defined as those with a clearly defined neck but with no distinct pedicle.18 Laterally spreading tumors were defined as large flat polyps with diameters >10 mm.19 Laterally spreading tumors were classified into 4 subtypes based on their granular or nongranular, homogeneous or nonhomogeneous appearance as follows: granular homogenous type, granular nodular mixed type, nongranular flat elevated type, and nongranular pseudodepressed type. The size of the lesion was estimated using open (7-mm) biopsy forceps held against the polyp.20 For EMR as a polypectomy technique, a solution of normal saline solution without epinephrine was injected into the submucosa beneath the lesions. Subsequently, snare resection was performed using a cut current (VIO 300D, Endocut Q mode, effect 3; Erbe, Tubingen, Germany). The excised tissue was retrieved for histologic analysis. All study patients had scheduled follow-ups at our hospital within 30 days after colonoscopic EMR procedures. Any postprocedural bleeding and timing in relation to the index procedure were recorded. Patients also underwent a telephone interview at 30 days by the designated nurse who was blinded to patients’ randomization. If an episode of PEB was notified during the phone interview at 30 days, patients were told to revisit our hospital www.giejournal.org
Lee et al
Prophylactic endoscopic coagulation to prevent delayed PPB
Figure 1. Prophylactic endoscopic coagulation procedure. A, A 10-mm polyp in the sigmoid colon after submucosal saline solution injection. B, Resection ulcer demonstrating blood vessels after EMR. C, Prophylactic endoscopic coagulation is applied to vessels in the post-EMR ulcer using the coagulation probe. D, Final resection ulcer after prophylactic endoscopic coagulation.
promptly to have their hemoglobin level checked. When a repeated colonoscopy was performed in patients with clinically significant PEB, the origin of bleeding was identified and the bleeding site controlled by applying endoclipping or endoscopic coagulation. In patients with multiple polyp resections, lesions with adherent clots, visible vessels, or pigmented protuberances were considered responsible for the bleeding. If there were no polyps with signs of bleeding, the largest polyp was considered responsible for the bleeding.
3; Erbe). If there were multiple polyps in a patient, multiple EMR procedures were performed for all polyps. All polyps in the same patient were treated with PEC after EMR in the EMR with coagulation group and with EMR only in the control group. However, if any of the post-EMR ulcers showed immediate bleeding within 1 minute, the lesions were treated with coagulation and the patients excluded from study. With the exception of the intervention, the periprocedural treatment and care in both groups were the same.
Intervention
Delayed PEB
If the lesion did not bleed within the 1-minute observation after the completion of a technically successful colonoscopic EMR for colorectal lesions, consenting patients were randomized 1:1 to the EMR with coagulation group or EMR (control) group. The control group received no additional treatment after EMR. The EMR with coagulation group was treated with PEC. The nonbleeding post-EMR ulcers were treated with a coagulation probe (MTW Endoskopie Manufaktur, Wesel, Germany [from June 2013 to June 2014] or Finemedix, Daegu, South Korea [from July 2014 to July 2017]) until the visible vessels or erythema disappeared (Fig. 1, Videos 1-3, available online at www. giejournal.org). All PECs were applied using coagulation current (VIO 300D, soft coagulation mode at 80 W, effect
Delayed PEB was defined as bleeding that occurred up to 30 days after colonoscopic EMR. Overall, PEB was classified as clinically significant or minor bleeding. Clinically significant PEB was defined as massive hematochezia that required endoscopic hemostasis, presentation to the emergency department, hospitalization, or a decrease in the hemoglobin level >2 g/dL. If hematochezia occurred repeatedly for at least 2 days, despite a small amount, and if the amount of bleeding did not decrease, was one of the decision-making criteria for colonoscopic hemostasis. The hemoglobin levels were not checked repeatedly in all patients. We measured the hemoglobin level when the patient showed repeated hematochezia despite a small amount, an unplanned visit, or the
www.giejournal.org
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 3
Prophylactic endoscopic coagulation to prevent delayed PPB
Enrollment
Lee et al
Colorectal EMR (n = 3252) Excluded (n = 2682) polyp diameter <10 mm (n = 1262) pedunculated polyp (n = 328) immediate bleeding within 1 minute after EMR (n = 272) ulcer with a clean base without erythema or visible vessels (n = 413) incompletely resected lesion (n = 39) combined with other procedures such as epinephrine injection or endoscopic clipping (n = 305) injury of the proper muscular layer (suspected or confirmed) (n =47) patient with coagulopathy, hematological disease, end-stage renal disease, thrombocytopenai, or a history of prior PPB (n = 16)
Randomized (n = 570)
Allocation
EMR with coagulation group (n = 285)
EMR (control) group (n = 285)
Excluded (n = 1) Upper Gl bleeding within 30 days (n = 1)
Excluded (n = 0)
Analysis Analyzed (n = 285)
Analyzed (n = 284) Figure 2. Flow diagram of the study. PPB, post-EMR bleeding.
features of clinically significant PEB including massive hematochezia, before endoscopic hemostasis, presentation to the emergency department, and hospitalization. Minor PEB was defined as self-limited hematochezia and a decrease in the hemoglobin level <2 g/dL that did not require endoscopic hemostasis. The primary outcome measure was the incidence of delayed overall PEB, including clinically significant and minor bleeding, in patients with and without PEC. Overall, PEB was stratified into clinically significant and minor. Other evaluated outcomes were baseline patient demographics and polyp-related factors, the rate of PEB according to the endoscopist experience and the types of coagulation probes used, and factors associated with overall and clinically significant PEB.
cation sequence was prepared using a computer-generated random number table. Patients were blinded to the intervention they received. The incidence of delayed PEB was calculated per patient treated, regardless of the number of EMR procedures for a single patient. Values were expressed as means and percentages, and continuous variables were presented as mean standard deviation. A paired t test was performed to analyze continuous variables and the c2 test to evaluate categorical variables. Multivariate analysis was conducted by logistic regression. All statistical analyses were performed using SPSS version 14.0 (SPSS Inc, Chicago, Ill). The confidence interval (CI) was set at 95%, and P < .05 were considered to be statistically significant.
RESULTS Statistical analysis We determined our sample size by assuming that a preventive method would decrease the PEB rate by about 50%, based on various reports of delayed PEB (.3%15.1%).4-10 The sample size for this study was calculated based on a predicted reduction of the delayed PEB rate from 12.5% to 5.5% using PEC. Given an alpha value of .05, a power of 80%, and a 10% drop-out rate, the sample size required 285 patients per group. Patients were assigned randomly to 1 of 2 groups in a 1:1 fashion. The allo4 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
Study flow In total, 570 consecutive patients undergoing colonoscopic EMR procedures were considered eligible for this study. Patients were randomly assigned to the EMR with coagulation group (n Z 285) or the EMR (control) group (n Z 285) after EMR. In the control group, 1 patient was excluded because of upper GI bleeding within 30 days after EMR. Finally, 285 patients were included in the EMR with coagulation group and 284 patients in the EMR www.giejournal.org
Lee et al
Prophylactic endoscopic coagulation to prevent delayed PPB
TABLE 1. Baseline characteristics of patient demographics and polyp-related factors EMR with coagulation (n [ 285)
EMR (control) (n [ 284)
61.2 10.9
62.7 10.4
.085
168 (58.9)
189 (66.5)
.061
Oral administration of antiplatelets
28 (9.8)
36 (12.7)
.282
Oral administration of anticoagulants
5 (1.8)
1 (0.4)
.102
HTN
72 (25.3)
91 (32.0)
.074
DM
39 (13.7)
45 (15.8)
.468
P value
Patient demographics Age, y Men
Comorbidity
Chronic renal disease
6 (2.1)
2 (.7)
.156
Chronic liver disease
2 (.7)
1 (.4)
.565
16 (5.6)
9 (3.2)
.155
356
370
Experienced endoscopists
238 (66.9)
235 (63.5)
Endoscopic fellows
118 (33.1)
135 (36.5)
87 (24.4)
86 (23.2)
Heart disease Polyp-related factors Total no. of polyps removed No. of polyps removed by
.345
Coagulation probe MTW Endoskopie Manufaktur Finemedix
.706
269 (75.6)
284 (76.8)
No. of polyps
1.25 .64
1.31 .81
.348
Size of polyps, mm
15.3 6.3
14.5 5.7
.084
Distribution of polyps
.711
Cecum or ascending colon
155 (43.5)
169 (45.7)
Transverse colon
52 (14.6)
64 (17.3)
Descending colon
29 (8.2)
26 (7.0)
Sigmoid colon
73 (20.5)
66 (17.8)
Rectum
47 (13.2)
45 (12.2)
Semipedunculated (Isp)
82 (23.0)
98 (26.5)
Sessile (Is)
104 (29.2)
101 (27.3)
LST-G-H
49 (13.8)
51 (13.8)
LST-G-NM
18 (5.1)
16 (4.3)
LST-NG-F
94 (26.4)
100 (27.0)
9 (2.5)
4 (1.1)
Adenocarcinoma
46 (12.9)
42 (11.4)
Adenoma (tubular or tubulovillous)
Macroscopic form of polyps
LST-NG-PD
.629
Histopathologic findings of polyps
.967
238 (66.9)
256 (69.2)
Sessile serrated adenoma
29 (8.1)
29 (7.8)
Hyperplastic polyp
34 (9.6)
34 (9.2)
Others En bloc resection
9 (2.5)
9 (2.4)
282 (79.2)
304 (82.2)
.244
Values are n (%) or mean standard deviation. HTN, Hypertension; DM, diabetes mellitus; LST-G-H, laterally spreading tumor, granular homogenous type; LST-G-NM, laterally spreading tumor, granular nodular mixed type; LST-NG-F, laterally spreading tumor, nongranular flat elevated type; LST-NG-PD, laterally spreading tumor, nongranular pseudodepressed type.
www.giejournal.org
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 5
Prophylactic endoscopic coagulation to prevent delayed PPB
Lee et al
TABLE 2. Outcomes of PEB in patients with and without prophylactic endoscopic coagulation EMR with coagulation (n [ 285)
EMR (control) (n [ 284)
36 (12.6)
53 (18.7)
.048
5 (1.8)
9 (3.2)
.276
31 (10.9)
44 (15.5)
.104
30/185 (16.2)
40/183 (21.9)
.168
5/185 (2.7)
7/183 (3.8)
.544
25/185 (13.5)
33/183 (18.0)
.234
6/100 (6.0)
13/101 (12.9)
.096
0/100 (0)
2/101 (2.0)
.157
6/100 (6.0)
11/101 (10.9)
.213
4/72 (5.6)
10/68 (14.7)
.071
0/72 (0)
2/68 (2.9)
.143
3/72 (4.2)
8/68 (11.8)
.095
32/213 (15.0)
43/216 (19.9)
.183
5/213 (2.3)
7/216 (3.2)
.575
28/213 (13.1)
36/216 (16.7)
.306
Overall PEB Clinically significant Minor
P value
Endoscopists Experienced endoscopists Overall PEB Clinically significant Minor Endoscopic fellows Overall bleeding Clinically significant Minor Coagulation probe MTW Endoskopie Manufaktur Overall bleeding Clinically significant Minor Finemedix Overall bleeding Clinically significant Minor Clinically significant PEB Time of bleeding presentation
.302
<24 h
2 (.7)
24-48 h
1 (.4)
0 (0)
3-7 days
1 (.4)
1 (.4)
8-14 days
0 (0)
2 (.7)
15-30 days
1 (.4)
0 (0)
Rectal lesions Size of polyps, mm, mean (SD)
6 (2.1)
2 (.7)
4 (1.4)
.409
21.4 (6.43)
20.33 (8.16)
.806 .997
Antiplatelet/anticoagulant agents
2 (.7)
2 (.7)
Hospitalization
5 (1.8)
8 (2.8)
.396
Need for intervention
3 (1.1)
7 (2.5)
.200
3 (1.1)
7 (2.5)
.200
0
0
N/A
Colonoscopy Angiography/embolization
1 (.4)
1 (.4)
.998
Perforation
Transfusion
0
0
N/A
Death
0
0
N/A
Values are n or n/N (%) unless otherwise defined. PEB, Post-EMR bleeding; SD, standard deviation; N/A, not applicable.
(control) group. The flow diagram of the study is shown in Figure 2.
Baseline characteristics There were no significant differences between the 2 groups of patients in terms of demographics and
6 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
polyp-related factors, including the total number, mean number, mean size, distribution, macroscopic form, histopathologic findings, and en bloc resection rates (Table 1). There were 356 polyps in the EMR with coagulation group and 370 in the control group. The most common location of polyps was the cecum or the
www.giejournal.org
Lee et al
Prophylactic endoscopic coagulation to prevent delayed PPB
TABLE 3. Factors associated with overall post-EMR bleeding Univariate analysis
Multivariate analysis
Variables
Odds ratio
95% Confidence interval
P value
Age
.990
.97-1.01
.328
Sex
.833
.52-1.34
.451
Antiplatelet/anticoagulant use
1.446
.78-2.68
.242
Prophylactic endoscopic coagulation
.630
.40-.99
.049
Odds ratio
95% Confidence interval
P value
.590
.37-.95
.029
Experienced endoscopist
2.250
1.31-3.86
.003
2.116
1.22-3.67
.008
Type of coagulation probe (Finemedix)
1.907
1.04-3.49
.037
1.720
.93-3.19
.086
1.256
1.12-1.41
<.001
Polyp size >20 mm
1.526
.93-2.50
.094
Proximal colonic location
1.154
.92-1.45
.222
Rectal location
1.254
1.12-1.41
<.001
Macroscopic form of polyp
1.006
.83-1.23
.952
Piecemeal resection
1.128
.67-1.90
.650
Histologic type of polyp
.932
.83-1.05
.236
ascending colon in both groups (43.5% [155/356] in the EMR with coagulation group and 45.7% [169/370] in the control group). The most common macroscopic form of polyps was sessile (0-Is) polyp, found in 29.2% (104/ 356) of the EMR with coagulation group and 27.3% (101/370) of the control group. Histopathologically, most lesions were polyps with adenoma (66.9% [238/ 356] in the EMR with coagulation group and 69.2% [256/370] in the control group).
Outcomes of PEB The outcomes of PEB in patients with and without PEC are detailed in Table 2. The incidence of overall PEB was significantly lower in the EMR with coagulation group than in the control group (12.6% [36/285] vs 18.7% [53/ 284], P Z .048), whereas the rates of clinically significant PEB (1.8% [5/285] vs 3.2% [9/284], P Z .276) and minor PEB (10.9% [31/285] vs 15.5% [44/284], P Z .104) were comparable between the 2 groups. There were no differences in the rates of overall, clinically significant, and minor PEB according to endoscopist experience and the types of coagulation probes between the 2 groups. In patients with clinically significant PEB, there were no significant differences in the time of bleeding presentation, mean polyp size, rate of the rectal lesion, antithrombotic agents, hospitalization, and need for intervention between the 2 groups. Clinically significant PEB occurred in 14 patients (2.5%) and occurred within 24 hours in 8 patients (57.1% [8/14]). There was no perforation or death after polypectomy in both groups. An 81-year-old woman in the control group developed hematochezia 2 weeks after EMR on a 22-mm rectal polyp. Her hemoglobin was decreased to 11 g/dL from the initial
www.giejournal.org
15 g/dL. She refused hospitalization with endoscopic hemostasis. Fortunately, her hematochezia stopped spontaneously the next day.
Risk factors for PEB We performed univariate and multivariate analyses for factors associated with overall PEB. Univariate analyses showed that experienced endoscopist (P Z .003), type of coagulation probe (P Z .037), and rectal location (P < .001) were statistically significant. Multivariate analyses demonstrated that experienced endoscopist (odds ratio [OR], 2.116; 95% CI, 1.22-3.67; P Z .008) and rectal location (OR, 1.256; 95% CI, 1.12-1.41; P < .001) were statistically significant. Moreover, PEC was revealed as a preventive factor (OR, .590; 95% CI, .37-.95; P Z .029) (Table 3). We also performed univariate and multivariate analyses for factors associated with clinically significant PEB. Univariate analyses indicated that polyp size >20 mm (P Z .005) and rectal location (P Z .001) were statistically significant (Table 4). Moreover, an increase in polyp size by 1 mm was a risk factor associated with clinically significant PEB in univariate analyses (OR, 1.088; 95% CI, 1.03-1.16; P Z .005) (data not shown). Multivariate analyses indicated that polyp size >20 mm (OR, 1.08; 95% CI, 1.01-1.15; P Z .021) and rectal location (OR, 1.37; 95% CI, 1.101.71; P Z .005) were statistically significant (Table 4).
DISCUSSION In this prospective randomized controlled trial 569 patients undergoing colorectal EMR procedures were analyzed to examine the effectiveness of PEC in the prevention of delayed PEB. The rate of overall PEB was
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 7
Prophylactic endoscopic coagulation to prevent delayed PPB
Lee et al
TABLE 4. Factors associated with clinically significant post-EMR bleeding Univariate analysis
Multivariate analysis
Variables
Odds ratio
95% Confidence interval
P value
Age
1.011
.96-1.06
.679
Sex
.27
.06-1.24
.092
Antiplatelet/anticoagulant use
2.77
.85-9.08
.092
Prophylactic endoscopic coagulation
.546
.18-1.65
.283
Experienced endoscopist
3.354
.74-15.14
.116
Type of coagulation probe (Finemedix)
1.986
.44-8.98
.373
Polyp size >20 mm
1.09
1.03-1.16
.005
Proximal colonic location
.801
.47-1.37
.418
Rectal location
1.429
1.15-1.78
.001
Macroscopic form of polyp
1.214
.87-1.69
.248
Piecemeal resection
2.523
.89-7.19
.083
Histologic type of polyp
1.11
.89-1.40
.347
significantly lower in the EMR with coagulation group than in the control group (12.6% vs 18.7%, P Z .048). Rectal location of polyps was a risk factor associated with overall PEB, and PEC was shown as a preventive factor. Experienced endoscopists were related to performing EMR on a relatively higher number of polyps (473 vs 253) and larger polyps (16.0 6.4 mm vs 12.8 4.4 mm, P < .001) when compared with endoscopic fellows in this study. However, no differences were observed between the EMR with coagulation and control groups in the rates of clinically significant and minor PEB. Clinically significant PEB was associated with polyp size >20 mm and rectal location. Colorectal EMR is an effective and relatively safe procedure. However, delayed PEB remains a significant adverse event. Many risk factors for delayed PEB have been investigated but found to be inconsistent among studies.21 Moreover, no previous investigations have used coagulation probes as methods for prevention of delayed PEB. Therefore, this prospective randomized controlled trial assessed the effectiveness of PEC using coagulation probes after EMR. The incidence of overall delayed PEB (15.6%) in the current study was higher than that in previous studies.4,11,22-25 Previous studies included small polyps and colorectal polypectomies performed with various methods such as cold biopsy sampling and cold snare polypectomy. Moreover, most of these previous analyses had a retrospective design and included only major bleeding cases. A randomized trial on larger polys (>20 mm) using prophylactic coagulation forceps did not show the incidence of minor PEB and found that clinically significant PEB occurred in 6.6% of patients (5.2% in patients receiving PEC and 8.0% in control subjects).16 Although the present study was very similar 8 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
Odds ratio
95% Confidence interval
P value
1.08
1.01-1.15
.021
1.37
1.10-1.71
.005
to this study, we used coagulation probes for prophylaxis and included not only clinically significant PEB but also minor PEB. Minor PEB might result in an unplanned visit to an outpatient clinic and a repeated check of the hemoglobin level, although it would finally reveal not to be clinically significant. It would be a burden because most patients may have returned to their workplace or routine life. Therefore, we evaluated the overall bleeding including minor bleeding. In a recently published article, the incidences of overall, significant, and minor bleeding were .44%, .01%, and .24%, respectively,2 but the use of prophylactic endoscopic therapy after EMR was not specifically documented. In the present study, the incidence of clinically significant PEB (2.5%) was similar to that in previous studies.3,22,23,25 Several previous researchers have reported that polyp size >10 mm or increases in polyp size by 1 mm were risk factors for delayed PEB.3,4,11,12,24,25 Therefore, we focused on patients with polyps >10 mm. In the present study, polyp size >20 mm was a risk factor for clinically significant PEB as well as increases in polyp size by 1 mm. Clinically significant PEB occurred within 24 hours in 8 patients. This result is consistent with a randomized trial on larger polys (>20 mm) using prophylactic coagulation forceps, which showed that clinically significant PEB occurred within 24 hours in most patients (60.9%).16 In that trial of 347 patients, en bloc resection rate was about 14.5% and clinically significant PEB occurred in 6.6% of patients. Patients with intraprocedural bleeding were included, and those with multiple EMRs in 1 session were excluded. However, in our study of 569 patients, en bloc resection rate was 80.7%. Patients with intraprocedural bleeding within 1 minute after EMR were excluded, and those with multiple EMRs in 1 session were included. www.giejournal.org
Lee et al
These differences could influence the incidence of overall PEB and be related to a relatively lower incidence of clinically significant PEB despite a larger number of our study patients compared with the previous trial.16 A few endoscopic options have been investigated for the prevention of delayed PEB for sessile polyps. The use of argon plasma coagulation9 and coagulation forceps16 did not reduce the incidence of clinically significant delayed PEB. Likewise, in our study the use of coagulation probes also did not reduce the incidence of clinically significant delayed PEB. However, unlike argon plasma coagulation, this method reduced the incidence of the overall delayed PEB. Although PEC was not effective in preventing clinically significant and minor delayed PEB, it was safe and easy to perform and did not result in periprocedural adverse events or hospitalization. In practice, we perform the PEC routinely after most colorectal EMR procedures in our hospital, regardless of this clinical trial, and we considered that the incidence of clinically significant PEB after PEC would be lower in real clinical practice than that in our study.26,27 Interestingly, rectal lesions were a risk factor for overall and clinically significant delayed PEB in our study. Previous research demonstrated an increased risk of delayed PEB in right-side polypectomy.6,16,21 A few hypotheses on this association have been proposed, such as a thinner cecal wall and ileal fluids containing digestive enzymes and bile acids.11 However, there was evidence of publication bias on an assessment of polyp location, and the association became nonsignificant after correction of publication bias using trim-and-fill techniques.21 One reason for this inconsistency of the location could be the absolute small number of the incidences of PEB in our study. This study has some limitations. First, the sample size could be insufficient to show a more significant reduction of clinically significant and minor PEB. Second, the study duration was relatively long, because patients with even slight oozing within 1 minute after EMR at the time of randomization were excluded. Third, the P value for the rate of overall PEB was just under .05, which might not show the definite or conclusive differences between 2 groups. There is no convincing clinical benefit demonstrated from using PEC. Fourth, it might be biased in the analysis of patients with minor bleeding. Given that undergoing repeated colonoscopic examinations in these patients may be unnecessary and ethically inappropriate, the reasonable method in this study was that the largest polyp was considered responsible for the bleeding in patients with minor bleeding.3 Fifth, we did not analyze the cost-effectiveness of PEC in this study. The cost of a coagulation probe is approximately $50, which is not inexpensive. Although this study found reduced PEB with prophylactic cautery of visible vessels, this was largely because of a reduction in minor bleeding with no benefit observed for www.giejournal.org
Prophylactic endoscopic coagulation to prevent delayed PPB
clinically significant bleeding. Overall, PEB was more frequent with rectal lesions. Therefore, careful consideration of these factors will improve the outcomes of colorectal EMR with respect to PEB. REFERENCES 1. Zauber AG, Winawer SJ, O'Brien MJ, et al. Colonoscopic polypectomy and long-term prevention of colorectal-cancer deaths. N Engl J Med 2012;366:687-96. 2. Derbyshire E, Hungin P, Nickerson C, et al. Post-polypectomy bleeding in the English National Health Service Bowel Cancer Screening Programme. Endoscopy 2017;49:899-908. 3. Park SK, Seo JY, Lee MG, et al. Prospective analysis of delayed colorectal post-polypectomy bleeding. Surg Endosc 2018;32:3282-9. 4. Rutter MD, Nickerson C, Rees CJ, et al. Risk factors for adverse events related to polypectomy in the English Bowel Cancer Screening Programme. Endoscopy 2014;46:90-7. 5. Rosen L, Bub DS, Reed JF 3rd, et al. Hemorrhage following colonoscopic polypectomy. Dis Colon Rectum 1993;36:1126-31. 6. Burgess NG, Williams SJ, Hourigan LF, et al. A management algorithm based on delayed bleeding after wide-field endoscopic mucosal resection of large colonic lesions. Clin Gastroenterol Hepatol 2014;12:1525-33. 7. Buchner AM, Guarner-Argente C, Ginsberg GG. Outcomes of EMR of defiant colorectal lesions directed to an endoscopy referral center. Gastrointest Endosc 2012;76:255-63. 8. Di Giorgio P, De Luca L, Calcagno G, et al. Detachable snare versus epinephrine injection in the prevention of postpolypectomy bleeding: a randomized and controlled study. Endoscopy 2004;36:860-3. 9. Lee CK, Lee SH, Park JY, et al. Prophylactic argon plasma coagulation ablation does not decrease delayed postpolypectomy bleeding. Gastrointest Endosc 2009;70:353-61. 10. Li LY, Liu QS, Li L, et al. A meta-analysis and systematic review of prophylactic endoscopic treatments for postpolypectomy bleeding. Int J Colorectal Dis 2011;26:709-19. 11. Buddingh KT, Herngreen T, Haringsma J, et al. Location in the right hemi-colon is an independent risk factor for delayed postpolypectomy hemorrhage: a multi-center case-control study. Am J Gastroenterol 2011;106:1119-24. 12. Sawhney MS, Salfiti N, Nelson DB, et al. Risk factors for severe delayed postpolypectomy bleeding. Endoscopy 2008;40:115-9. 13. Heldwein W, Dollhopf M, Rosch T, et al. The Munich Polypectomy Study (MUPS): prospective analysis of complications and risk factors in 4000 colonic snare polypectomies. Endoscopy 2005;37:1116-22. 14. Pohl H, Grimm IS, Moyer MT, et al. Clip closure prevents bleeding after endoscopic resection of large colon polyps in a randomized trial. Gastroenterology. Epub 2019 Mar 15. 15. Takizawa K, Oda I, Gotoda T, et al. Routine coagulation of visible vessels may prevent delayed bleeding after endoscopic submucosal dissectiondan analysis of risk factors. Endoscopy 2008;40: 179-83. 16. Bahin FF, Naidoo M, Williams SJ, et al. Prophylactic endoscopic coagulation to prevent bleeding after wide-field endoscopic mucosal resection of large sessile colon polyps. Clin Gastroenterol Hepatol 2015;13: 724-30. 17. Veitch AM, Vanbiervliet G, Gershlick AH, et al. Endoscopy in patients on antiplatelet or anticoagulant therapy, including direct oral anticoagulants: British Society of Gastroenterology (BSG) and European Society of Gastrointestinal Endoscopy (ESGE) guidelines. Endoscopy 2016;48: 385-402. 18. Participants in the Paris Workshop. Update on the Paris classification of superficial neoplastic lesions in the digestive tract. Endoscopy 2005;37: 570-8. 19. Kudo S, Lambert R, Allen JI, et al. Nonpolypoid neoplastic lesions of the colorectal mucosa. Gastrointest Endosc 2008;68:S3-47.
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 9
Prophylactic endoscopic coagulation to prevent delayed PPB
Lee et al
20. Hwang MJ, Kim KO, Kim AL, et al. Histologic discrepancy between endoscopic forceps biopsy and endoscopic mucosal resection specimens of colorectal polyp in actual clinical practice. Intest Res 2018;16:475-83. 21. Jaruvongvanich V, Prasitlumkum N, Assavapongpaiboon B, et al. Risk factors for delayed colonic post-polypectomy bleeding: a systematic review and meta-analysis. Int J Colorectal Dis 2017;32: 1399-406. 22. Watabe H, Yamaji Y, Okamoto M, et al. Risk assessment for delayed hemorrhagic complication of colonic polypectomy: polyp-related factors and patient-related factors. Gastrointest Endosc 2006;64: 73-8. 23. Hui AJ, Wong RM, Ching JY, et al. Risk of colonoscopic polypectomy bleeding with anticoagulants and antiplatelet agents: analysis of 1657 cases. Gastrointest Endosc 2004;59:44-8. 24. Choung BS, Kim SH, Ahn DS, et al. Incidence and risk factors of delayed postpolypectomy bleeding: a retrospective cohort study. J Clin Gastroenterol 2014;48:784-9. 25. Kim JH, Lee HJ, Ahn JW, et al. Risk factors for delayed postpolypectomy hemorrhage: a case-control study. J Gastroenterol Hepatol 2013;28:645-9. 26. Saturni S, Bellini F, Braido F, et al. Randomized controlled trials and real life studies. Approaches and methodologies: a clinical point of view. Pulm Pharmacol Ther 2014;27:129-38.
10 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
27. Kim HS, Lee S, Kim JH. Real-world evidence versus randomized controlled trial: clinical research based on electronic medical records. J Korean Med Sci 2018;33:e213. Abbreviations: CI, confidence interval; OR, odds ratio; PEB, post-EMR bleeding; PEC, prophylactic endoscopic coagulation. DISCLOSURE: The following author disclosed financial relationships relevant to this publication: S. W. Jeon: Major shareholder and CEO of Finemedix Co, Ltd. All other 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.2019.05.039 Received December 17, 2018. Accepted May 27, 2019. Current affiliations: Department of Internal Medicine, School of Medicine, Kyungpook National University (1) and Department of Internal Medicine (2), Kyungpook National University Hospital, Daegu, South Korea. Reprint requests: Seong Woo Jeon, MD, PhD, Department of Internal Medicine, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, 807 Hoguk-ro, Buk-gu, Daegu, 41404, Korea.
www.giejournal.org
Prophylactic endoscopic coagulation to prevent delayed post-EMR bleeding in the colorectum: a prospective randomized controlled trial (with videos) Hyun Seok Lee, MD, PhD,1,2 Seong Woo Jeon, MD, PhD,1,2 Yong Hwan Kwon, MD, PhD,1,2 Su Youn Nam, MD, PhD,2 Seonghwan Shin, MD,2 Ryanghi Kim, MD,2 Sohyun Ahn, MD2 Daegu, South Korea
GRAPHICAL ABSTRACT
Background and Aims: Post-EMR bleeding (PEB) is the most common adverse event after EMR. However, there are no established endoscopic methods for the prevention of PEB. This study aimed to investigate whether prophylactic endoscopic coagulation (PEC) using coagulation probes reduces the incidence of overall delayed PEB. Methods: We performed a randomized controlled study of patients undergoing EMR for large (10 mm) sessile lesions and laterally spreading tumors. Patients were randomized 1:1 to the EMR with coagulation group (n Z 285) or EMR (control) group (n Z 285). Immediate bleeding during colon EMR or clean-based ulcer after EMR was excluded. Clinically significant PEB was defined as bleeding requiring endoscopic hemostasis, hospitalization, or a decrease in the hemoglobin level >2 g/dL. Results: A total of 569 patients were analyzed. The incidence of overall PEB was significantly lower in the EMR with coagulation group than in the control group (12.6% [36/285] vs 18.7% [53/284], P Z .048). However, this was largely because of a reduction in minor bleeding. There was no difference in clinically significant PEB (1.8% [5/285] vs 3.2% [9/284], P Z .276). Rectal location was a risk factor associated with overall PEB (odds ratio, 1.256; 95% confidence interval, 1.12-1.41; P < .001). Conclusions: Although this study found reduced PEB with prophylactic cautery of visible vessels, this was largely because of a reduction in minor bleeding with no benefit observed for clinically significant bleeding. Overall, PEB was more frequent with rectal lesions. (Clinical trial registration number: KCT0000779.) (Gastrointest Endosc 2019;-:1-10.)
(footnotes appear on last page of article)
Polypectomy of colorectal polyps reduces the incidence and mortality of colorectal cancer.1 However, polypectomy is associated with adverse events, including post-EMR bleeding (PEB).2 PEB can occur during (immediate) or after (delayed) the procedure.3 Based on various reports,
www.giejournal.org
delayed PEB could occur in approximately .3% to 15.1% of cases.4-10 Among the risk factors for delayed PEB, size and location of polyps are the most widely known and well-defined risk factors.11,12 However, many risk factors remain equivocal and inconsistent. Delayed PEB can be
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 1
Prophylactic endoscopic coagulation to prevent delayed PPB
Lee et al
serious because it can result in hospitalization, repeated colonoscopy, and blood transfusion.13 Clinically minor PEB might cause an unplanned visit to an outpatient clinic or a repeated check of the hemoglobin level but would be ultimately identified as clinically nonsignificant compared with severe PEB. Closure of EMR sites with clips is effective in reducing PEB, but this technique is time-consuming and costly.14 A retrospective study showed that prophylactic endoscopic coagulation (PEC) might prevent delayed PEB after endoscopic resection of superficial neoplastic lesions in the stomach.15 PEC is easy to perform and widely applicable. It can be applied to postpolypectomy defects of any size, unlike endoscopic clip closure.16 However, there are some concerns about perforation caused by PEC in the colorectum that are not concerning in the stomach. A previous prospective study did not show the efficacy of PEC to reduce the incidence of delayed PEB on colorectal lesions larger than 20 mm.16 Our hypothesis was that if a lesion 10 mm was included and the number of subjects increased accordingly, the result could show a significant reduction of delayed PEB. Thus, we performed a prospective randomized controlled trial to determine whether PEC using coagulation probes reduces the incidence of delayed PEB after resection of large (10 mm) sessile or laterally spreading colorectal lesions and to evaluate the incidence of perforation after PEC.
METHODS Patients This randomized controlled trial was conducted at Kyungpook National University Hospital between June 2013 and July 2017. Patients undergoing colonoscopic EMR for large (10 mm) sessile lesions and laterally spreading tumors were considered eligible for enrollment. For patients with more than 1 polyp, all patients with postEMR ulcers were enrolled in the study. Written informed consent was obtained from each patient. Randomization of all patients was performed using a table of random numbers generated by a computer program. The protocol was registered with the Clinical Research Information Service (a representative clinical trials registry platform in Korea) before enrolling the first participant (registration number KCT0000779). This study was conducted after obtaining ethical approval from the Kyungpook National University Hospital internal review board (KNUMC_12-1037).
Colonoscopic EMR Patients were assigned randomly to 1 of 2 treatment groups in a 1:1 fashion. Patients were blinded to their intervention apart from their sedation. Exclusion criteria were polyp diameter <10 mm, a pedunculated polyp, immediate 2 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
bleeding within 1 minute after EMR, an ulcer with a clean base without erythema or visible vessels, bleeding from other GI sites except for colorectum, incompletely resected lesion, combined with other procedures such as epinephrine injection or endoscopic clipping, injury of the proper muscular layer (suspected or confirmed), and patient with coagulopathy, hematologic disease, endstage renal disease, thrombocytopenia, or a history of prior PEB. Patients on antiplatelet agents were asked to stop taking them 5 to 7 days before the procedure and resume up to 48 hours after the procedure. Patients on warfarin were advised to stop taking it 5 days before the procedure and resume on the day after the procedure.17 Before colonoscopic examination, the bowel was cleansed with 2 L polyethylene glycol plus ascorbic acid solution. Intravenous sedation was performed using midazolam, and pethidine was used to reduce pain. Colonoscopic examinations were performed with a conventional video colonoscope (CF-H260AL; Olympus Optical Co, Tokyo, Japan). All colonoscopic EMR procedures were performed by experienced endoscopists with more than 5 years of experience of performing therapeutic endoscopy or endoscopic fellows under direct supervision. These endoscopists had performed PEC routinely in real clinical practice for lesions believed to be at high risk of PEB. Colonic insufflation was performed with air. The size, location, number, and macroscopic form of the polyps were documented. The gross morphologies of the polyps were classified using the updated Paris classification as follows: sessile (Is) lesions were defined as raised lesions without evidence of a stalk or pedicle, and subpedunculated (Isp) lesions were defined as those with a clearly defined neck but with no distinct pedicle.18 Laterally spreading tumors were defined as large flat polyps with diameters >10 mm.19 Laterally spreading tumors were classified into 4 subtypes based on their granular or nongranular, homogeneous or nonhomogeneous appearance as follows: granular homogenous type, granular nodular mixed type, nongranular flat elevated type, and nongranular pseudodepressed type. The size of the lesion was estimated using open (7-mm) biopsy forceps held against the polyp.20 For EMR as a polypectomy technique, a solution of normal saline solution without epinephrine was injected into the submucosa beneath the lesions. Subsequently, snare resection was performed using a cut current (VIO 300D, Endocut Q mode, effect 3; Erbe, Tubingen, Germany). The excised tissue was retrieved for histologic analysis. All study patients had scheduled follow-ups at our hospital within 30 days after colonoscopic EMR procedures. Any postprocedural bleeding and timing in relation to the index procedure were recorded. Patients also underwent a telephone interview at 30 days by the designated nurse who was blinded to patients’ randomization. If an episode of PEB was notified during the phone interview at 30 days, patients were told to revisit our hospital www.giejournal.org
Lee et al
Prophylactic endoscopic coagulation to prevent delayed PPB
Figure 1. Prophylactic endoscopic coagulation procedure. A, A 10-mm polyp in the sigmoid colon after submucosal saline solution injection. B, Resection ulcer demonstrating blood vessels after EMR. C, Prophylactic endoscopic coagulation is applied to vessels in the post-EMR ulcer using the coagulation probe. D, Final resection ulcer after prophylactic endoscopic coagulation.
promptly to have their hemoglobin level checked. When a repeated colonoscopy was performed in patients with clinically significant PEB, the origin of bleeding was identified and the bleeding site controlled by applying endoclipping or endoscopic coagulation. In patients with multiple polyp resections, lesions with adherent clots, visible vessels, or pigmented protuberances were considered responsible for the bleeding. If there were no polyps with signs of bleeding, the largest polyp was considered responsible for the bleeding.
3; Erbe). If there were multiple polyps in a patient, multiple EMR procedures were performed for all polyps. All polyps in the same patient were treated with PEC after EMR in the EMR with coagulation group and with EMR only in the control group. However, if any of the post-EMR ulcers showed immediate bleeding within 1 minute, the lesions were treated with coagulation and the patients excluded from study. With the exception of the intervention, the periprocedural treatment and care in both groups were the same.
Intervention
Delayed PEB
If the lesion did not bleed within the 1-minute observation after the completion of a technically successful colonoscopic EMR for colorectal lesions, consenting patients were randomized 1:1 to the EMR with coagulation group or EMR (control) group. The control group received no additional treatment after EMR. The EMR with coagulation group was treated with PEC. The nonbleeding post-EMR ulcers were treated with a coagulation probe (MTW Endoskopie Manufaktur, Wesel, Germany [from June 2013 to June 2014] or Finemedix, Daegu, South Korea [from July 2014 to July 2017]) until the visible vessels or erythema disappeared (Fig. 1, Videos 1-3, available online at www. giejournal.org). All PECs were applied using coagulation current (VIO 300D, soft coagulation mode at 80 W, effect
Delayed PEB was defined as bleeding that occurred up to 30 days after colonoscopic EMR. Overall, PEB was classified as clinically significant or minor bleeding. Clinically significant PEB was defined as massive hematochezia that required endoscopic hemostasis, presentation to the emergency department, hospitalization, or a decrease in the hemoglobin level >2 g/dL. If hematochezia occurred repeatedly for at least 2 days, despite a small amount, and if the amount of bleeding did not decrease, was one of the decision-making criteria for colonoscopic hemostasis. The hemoglobin levels were not checked repeatedly in all patients. We measured the hemoglobin level when the patient showed repeated hematochezia despite a small amount, an unplanned visit, or the
www.giejournal.org
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 3
Prophylactic endoscopic coagulation to prevent delayed PPB
Enrollment
Lee et al
Colorectal EMR (n = 3252) Excluded (n = 2682) polyp diameter <10 mm (n = 1262) pedunculated polyp (n = 328) immediate bleeding within 1 minute after EMR (n = 272) ulcer with a clean base without erythema or visible vessels (n = 413) incompletely resected lesion (n = 39) combined with other procedures such as epinephrine injection or endoscopic clipping (n = 305) injury of the proper muscular layer (suspected or confirmed) (n =47) patient with coagulopathy, hematological disease, end-stage renal disease, thrombocytopenai, or a history of prior PPB (n = 16)
Randomized (n = 570)
Allocation
EMR with coagulation group (n = 285)
EMR (control) group (n = 285)
Excluded (n = 1) Upper Gl bleeding within 30 days (n = 1)
Excluded (n = 0)
Analysis Analyzed (n = 285)
Analyzed (n = 284) Figure 2. Flow diagram of the study. PPB, post-EMR bleeding.
features of clinically significant PEB including massive hematochezia, before endoscopic hemostasis, presentation to the emergency department, and hospitalization. Minor PEB was defined as self-limited hematochezia and a decrease in the hemoglobin level <2 g/dL that did not require endoscopic hemostasis. The primary outcome measure was the incidence of delayed overall PEB, including clinically significant and minor bleeding, in patients with and without PEC. Overall, PEB was stratified into clinically significant and minor. Other evaluated outcomes were baseline patient demographics and polyp-related factors, the rate of PEB according to the endoscopist experience and the types of coagulation probes used, and factors associated with overall and clinically significant PEB.
cation sequence was prepared using a computer-generated random number table. Patients were blinded to the intervention they received. The incidence of delayed PEB was calculated per patient treated, regardless of the number of EMR procedures for a single patient. Values were expressed as means and percentages, and continuous variables were presented as mean standard deviation. A paired t test was performed to analyze continuous variables and the c2 test to evaluate categorical variables. Multivariate analysis was conducted by logistic regression. All statistical analyses were performed using SPSS version 14.0 (SPSS Inc, Chicago, Ill). The confidence interval (CI) was set at 95%, and P < .05 were considered to be statistically significant.
RESULTS Statistical analysis We determined our sample size by assuming that a preventive method would decrease the PEB rate by about 50%, based on various reports of delayed PEB (.3%15.1%).4-10 The sample size for this study was calculated based on a predicted reduction of the delayed PEB rate from 12.5% to 5.5% using PEC. Given an alpha value of .05, a power of 80%, and a 10% drop-out rate, the sample size required 285 patients per group. Patients were assigned randomly to 1 of 2 groups in a 1:1 fashion. The allo4 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
Study flow In total, 570 consecutive patients undergoing colonoscopic EMR procedures were considered eligible for this study. Patients were randomly assigned to the EMR with coagulation group (n Z 285) or the EMR (control) group (n Z 285) after EMR. In the control group, 1 patient was excluded because of upper GI bleeding within 30 days after EMR. Finally, 285 patients were included in the EMR with coagulation group and 284 patients in the EMR www.giejournal.org
Lee et al
Prophylactic endoscopic coagulation to prevent delayed PPB
TABLE 1. Baseline characteristics of patient demographics and polyp-related factors EMR with coagulation (n [ 285)
EMR (control) (n [ 284)
61.2 10.9
62.7 10.4
.085
168 (58.9)
189 (66.5)
.061
Oral administration of antiplatelets
28 (9.8)
36 (12.7)
.282
Oral administration of anticoagulants
5 (1.8)
1 (0.4)
.102
HTN
72 (25.3)
91 (32.0)
.074
DM
39 (13.7)
45 (15.8)
.468
P value
Patient demographics Age, y Men
Comorbidity
Chronic renal disease
6 (2.1)
2 (.7)
.156
Chronic liver disease
2 (.7)
1 (.4)
.565
16 (5.6)
9 (3.2)
.155
356
370
Experienced endoscopists
238 (66.9)
235 (63.5)
Endoscopic fellows
118 (33.1)
135 (36.5)
87 (24.4)
86 (23.2)
Heart disease Polyp-related factors Total no. of polyps removed No. of polyps removed by
.345
Coagulation probe MTW Endoskopie Manufaktur Finemedix
.706
269 (75.6)
284 (76.8)
No. of polyps
1.25 .64
1.31 .81
.348
Size of polyps, mm
15.3 6.3
14.5 5.7
.084
Distribution of polyps
.711
Cecum or ascending colon
155 (43.5)
169 (45.7)
Transverse colon
52 (14.6)
64 (17.3)
Descending colon
29 (8.2)
26 (7.0)
Sigmoid colon
73 (20.5)
66 (17.8)
Rectum
47 (13.2)
45 (12.2)
Semipedunculated (Isp)
82 (23.0)
98 (26.5)
Sessile (Is)
104 (29.2)
101 (27.3)
LST-G-H
49 (13.8)
51 (13.8)
LST-G-NM
18 (5.1)
16 (4.3)
LST-NG-F
94 (26.4)
100 (27.0)
9 (2.5)
4 (1.1)
Adenocarcinoma
46 (12.9)
42 (11.4)
Adenoma (tubular or tubulovillous)
Macroscopic form of polyps
LST-NG-PD
.629
Histopathologic findings of polyps
.967
238 (66.9)
256 (69.2)
Sessile serrated adenoma
29 (8.1)
29 (7.8)
Hyperplastic polyp
34 (9.6)
34 (9.2)
Others En bloc resection
9 (2.5)
9 (2.4)
282 (79.2)
304 (82.2)
.244
Values are n (%) or mean standard deviation. HTN, Hypertension; DM, diabetes mellitus; LST-G-H, laterally spreading tumor, granular homogenous type; LST-G-NM, laterally spreading tumor, granular nodular mixed type; LST-NG-F, laterally spreading tumor, nongranular flat elevated type; LST-NG-PD, laterally spreading tumor, nongranular pseudodepressed type.
www.giejournal.org
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 5
Prophylactic endoscopic coagulation to prevent delayed PPB
Lee et al
TABLE 2. Outcomes of PEB in patients with and without prophylactic endoscopic coagulation EMR with coagulation (n [ 285)
EMR (control) (n [ 284)
36 (12.6)
53 (18.7)
.048
5 (1.8)
9 (3.2)
.276
31 (10.9)
44 (15.5)
.104
30/185 (16.2)
40/183 (21.9)
.168
5/185 (2.7)
7/183 (3.8)
.544
25/185 (13.5)
33/183 (18.0)
.234
6/100 (6.0)
13/101 (12.9)
.096
0/100 (0)
2/101 (2.0)
.157
6/100 (6.0)
11/101 (10.9)
.213
4/72 (5.6)
10/68 (14.7)
.071
0/72 (0)
2/68 (2.9)
.143
3/72 (4.2)
8/68 (11.8)
.095
32/213 (15.0)
43/216 (19.9)
.183
5/213 (2.3)
7/216 (3.2)
.575
28/213 (13.1)
36/216 (16.7)
.306
Overall PEB Clinically significant Minor
P value
Endoscopists Experienced endoscopists Overall PEB Clinically significant Minor Endoscopic fellows Overall bleeding Clinically significant Minor Coagulation probe MTW Endoskopie Manufaktur Overall bleeding Clinically significant Minor Finemedix Overall bleeding Clinically significant Minor Clinically significant PEB Time of bleeding presentation
.302
<24 h
2 (.7)
24-48 h
1 (.4)
0 (0)
3-7 days
1 (.4)
1 (.4)
8-14 days
0 (0)
2 (.7)
15-30 days
1 (.4)
0 (0)
Rectal lesions Size of polyps, mm, mean (SD)
6 (2.1)
2 (.7)
4 (1.4)
.409
21.4 (6.43)
20.33 (8.16)
.806 .997
Antiplatelet/anticoagulant agents
2 (.7)
2 (.7)
Hospitalization
5 (1.8)
8 (2.8)
.396
Need for intervention
3 (1.1)
7 (2.5)
.200
3 (1.1)
7 (2.5)
.200
0
0
N/A
Colonoscopy Angiography/embolization
1 (.4)
1 (.4)
.998
Perforation
Transfusion
0
0
N/A
Death
0
0
N/A
Values are n or n/N (%) unless otherwise defined. PEB, Post-EMR bleeding; SD, standard deviation; N/A, not applicable.
(control) group. The flow diagram of the study is shown in Figure 2.
Baseline characteristics There were no significant differences between the 2 groups of patients in terms of demographics and
6 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
polyp-related factors, including the total number, mean number, mean size, distribution, macroscopic form, histopathologic findings, and en bloc resection rates (Table 1). There were 356 polyps in the EMR with coagulation group and 370 in the control group. The most common location of polyps was the cecum or the
www.giejournal.org
Lee et al
Prophylactic endoscopic coagulation to prevent delayed PPB
TABLE 3. Factors associated with overall post-EMR bleeding Univariate analysis
Multivariate analysis
Variables
Odds ratio
95% Confidence interval
P value
Age
.990
.97-1.01
.328
Sex
.833
.52-1.34
.451
Antiplatelet/anticoagulant use
1.446
.78-2.68
.242
Prophylactic endoscopic coagulation
.630
.40-.99
.049
Odds ratio
95% Confidence interval
P value
.590
.37-.95
.029
Experienced endoscopist
2.250
1.31-3.86
.003
2.116
1.22-3.67
.008
Type of coagulation probe (Finemedix)
1.907
1.04-3.49
.037
1.720
.93-3.19
.086
1.256
1.12-1.41
<.001
Polyp size >20 mm
1.526
.93-2.50
.094
Proximal colonic location
1.154
.92-1.45
.222
Rectal location
1.254
1.12-1.41
<.001
Macroscopic form of polyp
1.006
.83-1.23
.952
Piecemeal resection
1.128
.67-1.90
.650
Histologic type of polyp
.932
.83-1.05
.236
ascending colon in both groups (43.5% [155/356] in the EMR with coagulation group and 45.7% [169/370] in the control group). The most common macroscopic form of polyps was sessile (0-Is) polyp, found in 29.2% (104/ 356) of the EMR with coagulation group and 27.3% (101/370) of the control group. Histopathologically, most lesions were polyps with adenoma (66.9% [238/ 356] in the EMR with coagulation group and 69.2% [256/370] in the control group).
Outcomes of PEB The outcomes of PEB in patients with and without PEC are detailed in Table 2. The incidence of overall PEB was significantly lower in the EMR with coagulation group than in the control group (12.6% [36/285] vs 18.7% [53/ 284], P Z .048), whereas the rates of clinically significant PEB (1.8% [5/285] vs 3.2% [9/284], P Z .276) and minor PEB (10.9% [31/285] vs 15.5% [44/284], P Z .104) were comparable between the 2 groups. There were no differences in the rates of overall, clinically significant, and minor PEB according to endoscopist experience and the types of coagulation probes between the 2 groups. In patients with clinically significant PEB, there were no significant differences in the time of bleeding presentation, mean polyp size, rate of the rectal lesion, antithrombotic agents, hospitalization, and need for intervention between the 2 groups. Clinically significant PEB occurred in 14 patients (2.5%) and occurred within 24 hours in 8 patients (57.1% [8/14]). There was no perforation or death after polypectomy in both groups. An 81-year-old woman in the control group developed hematochezia 2 weeks after EMR on a 22-mm rectal polyp. Her hemoglobin was decreased to 11 g/dL from the initial
www.giejournal.org
15 g/dL. She refused hospitalization with endoscopic hemostasis. Fortunately, her hematochezia stopped spontaneously the next day.
Risk factors for PEB We performed univariate and multivariate analyses for factors associated with overall PEB. Univariate analyses showed that experienced endoscopist (P Z .003), type of coagulation probe (P Z .037), and rectal location (P < .001) were statistically significant. Multivariate analyses demonstrated that experienced endoscopist (odds ratio [OR], 2.116; 95% CI, 1.22-3.67; P Z .008) and rectal location (OR, 1.256; 95% CI, 1.12-1.41; P < .001) were statistically significant. Moreover, PEC was revealed as a preventive factor (OR, .590; 95% CI, .37-.95; P Z .029) (Table 3). We also performed univariate and multivariate analyses for factors associated with clinically significant PEB. Univariate analyses indicated that polyp size >20 mm (P Z .005) and rectal location (P Z .001) were statistically significant (Table 4). Moreover, an increase in polyp size by 1 mm was a risk factor associated with clinically significant PEB in univariate analyses (OR, 1.088; 95% CI, 1.03-1.16; P Z .005) (data not shown). Multivariate analyses indicated that polyp size >20 mm (OR, 1.08; 95% CI, 1.01-1.15; P Z .021) and rectal location (OR, 1.37; 95% CI, 1.101.71; P Z .005) were statistically significant (Table 4).
DISCUSSION In this prospective randomized controlled trial 569 patients undergoing colorectal EMR procedures were analyzed to examine the effectiveness of PEC in the prevention of delayed PEB. The rate of overall PEB was
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 7
Prophylactic endoscopic coagulation to prevent delayed PPB
Lee et al
TABLE 4. Factors associated with clinically significant post-EMR bleeding Univariate analysis
Multivariate analysis
Variables
Odds ratio
95% Confidence interval
P value
Age
1.011
.96-1.06
.679
Sex
.27
.06-1.24
.092
Antiplatelet/anticoagulant use
2.77
.85-9.08
.092
Prophylactic endoscopic coagulation
.546
.18-1.65
.283
Experienced endoscopist
3.354
.74-15.14
.116
Type of coagulation probe (Finemedix)
1.986
.44-8.98
.373
Polyp size >20 mm
1.09
1.03-1.16
.005
Proximal colonic location
.801
.47-1.37
.418
Rectal location
1.429
1.15-1.78
.001
Macroscopic form of polyp
1.214
.87-1.69
.248
Piecemeal resection
2.523
.89-7.19
.083
Histologic type of polyp
1.11
.89-1.40
.347
significantly lower in the EMR with coagulation group than in the control group (12.6% vs 18.7%, P Z .048). Rectal location of polyps was a risk factor associated with overall PEB, and PEC was shown as a preventive factor. Experienced endoscopists were related to performing EMR on a relatively higher number of polyps (473 vs 253) and larger polyps (16.0 6.4 mm vs 12.8 4.4 mm, P < .001) when compared with endoscopic fellows in this study. However, no differences were observed between the EMR with coagulation and control groups in the rates of clinically significant and minor PEB. Clinically significant PEB was associated with polyp size >20 mm and rectal location. Colorectal EMR is an effective and relatively safe procedure. However, delayed PEB remains a significant adverse event. Many risk factors for delayed PEB have been investigated but found to be inconsistent among studies.21 Moreover, no previous investigations have used coagulation probes as methods for prevention of delayed PEB. Therefore, this prospective randomized controlled trial assessed the effectiveness of PEC using coagulation probes after EMR. The incidence of overall delayed PEB (15.6%) in the current study was higher than that in previous studies.4,11,22-25 Previous studies included small polyps and colorectal polypectomies performed with various methods such as cold biopsy sampling and cold snare polypectomy. Moreover, most of these previous analyses had a retrospective design and included only major bleeding cases. A randomized trial on larger polys (>20 mm) using prophylactic coagulation forceps did not show the incidence of minor PEB and found that clinically significant PEB occurred in 6.6% of patients (5.2% in patients receiving PEC and 8.0% in control subjects).16 Although the present study was very similar 8 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
Odds ratio
95% Confidence interval
P value
1.08
1.01-1.15
.021
1.37
1.10-1.71
.005
to this study, we used coagulation probes for prophylaxis and included not only clinically significant PEB but also minor PEB. Minor PEB might result in an unplanned visit to an outpatient clinic and a repeated check of the hemoglobin level, although it would finally reveal not to be clinically significant. It would be a burden because most patients may have returned to their workplace or routine life. Therefore, we evaluated the overall bleeding including minor bleeding. In a recently published article, the incidences of overall, significant, and minor bleeding were .44%, .01%, and .24%, respectively,2 but the use of prophylactic endoscopic therapy after EMR was not specifically documented. In the present study, the incidence of clinically significant PEB (2.5%) was similar to that in previous studies.3,22,23,25 Several previous researchers have reported that polyp size >10 mm or increases in polyp size by 1 mm were risk factors for delayed PEB.3,4,11,12,24,25 Therefore, we focused on patients with polyps >10 mm. In the present study, polyp size >20 mm was a risk factor for clinically significant PEB as well as increases in polyp size by 1 mm. Clinically significant PEB occurred within 24 hours in 8 patients. This result is consistent with a randomized trial on larger polys (>20 mm) using prophylactic coagulation forceps, which showed that clinically significant PEB occurred within 24 hours in most patients (60.9%).16 In that trial of 347 patients, en bloc resection rate was about 14.5% and clinically significant PEB occurred in 6.6% of patients. Patients with intraprocedural bleeding were included, and those with multiple EMRs in 1 session were excluded. However, in our study of 569 patients, en bloc resection rate was 80.7%. Patients with intraprocedural bleeding within 1 minute after EMR were excluded, and those with multiple EMRs in 1 session were included. www.giejournal.org
Lee et al
These differences could influence the incidence of overall PEB and be related to a relatively lower incidence of clinically significant PEB despite a larger number of our study patients compared with the previous trial.16 A few endoscopic options have been investigated for the prevention of delayed PEB for sessile polyps. The use of argon plasma coagulation9 and coagulation forceps16 did not reduce the incidence of clinically significant delayed PEB. Likewise, in our study the use of coagulation probes also did not reduce the incidence of clinically significant delayed PEB. However, unlike argon plasma coagulation, this method reduced the incidence of the overall delayed PEB. Although PEC was not effective in preventing clinically significant and minor delayed PEB, it was safe and easy to perform and did not result in periprocedural adverse events or hospitalization. In practice, we perform the PEC routinely after most colorectal EMR procedures in our hospital, regardless of this clinical trial, and we considered that the incidence of clinically significant PEB after PEC would be lower in real clinical practice than that in our study.26,27 Interestingly, rectal lesions were a risk factor for overall and clinically significant delayed PEB in our study. Previous research demonstrated an increased risk of delayed PEB in right-side polypectomy.6,16,21 A few hypotheses on this association have been proposed, such as a thinner cecal wall and ileal fluids containing digestive enzymes and bile acids.11 However, there was evidence of publication bias on an assessment of polyp location, and the association became nonsignificant after correction of publication bias using trim-and-fill techniques.21 One reason for this inconsistency of the location could be the absolute small number of the incidences of PEB in our study. This study has some limitations. First, the sample size could be insufficient to show a more significant reduction of clinically significant and minor PEB. Second, the study duration was relatively long, because patients with even slight oozing within 1 minute after EMR at the time of randomization were excluded. Third, the P value for the rate of overall PEB was just under .05, which might not show the definite or conclusive differences between 2 groups. There is no convincing clinical benefit demonstrated from using PEC. Fourth, it might be biased in the analysis of patients with minor bleeding. Given that undergoing repeated colonoscopic examinations in these patients may be unnecessary and ethically inappropriate, the reasonable method in this study was that the largest polyp was considered responsible for the bleeding in patients with minor bleeding.3 Fifth, we did not analyze the cost-effectiveness of PEC in this study. The cost of a coagulation probe is approximately $50, which is not inexpensive. Although this study found reduced PEB with prophylactic cautery of visible vessels, this was largely because of a reduction in minor bleeding with no benefit observed for www.giejournal.org
Prophylactic endoscopic coagulation to prevent delayed PPB
clinically significant bleeding. Overall, PEB was more frequent with rectal lesions. Therefore, careful consideration of these factors will improve the outcomes of colorectal EMR with respect to PEB. REFERENCES 1. Zauber AG, Winawer SJ, O'Brien MJ, et al. Colonoscopic polypectomy and long-term prevention of colorectal-cancer deaths. N Engl J Med 2012;366:687-96. 2. Derbyshire E, Hungin P, Nickerson C, et al. Post-polypectomy bleeding in the English National Health Service Bowel Cancer Screening Programme. Endoscopy 2017;49:899-908. 3. Park SK, Seo JY, Lee MG, et al. Prospective analysis of delayed colorectal post-polypectomy bleeding. Surg Endosc 2018;32:3282-9. 4. Rutter MD, Nickerson C, Rees CJ, et al. Risk factors for adverse events related to polypectomy in the English Bowel Cancer Screening Programme. Endoscopy 2014;46:90-7. 5. Rosen L, Bub DS, Reed JF 3rd, et al. Hemorrhage following colonoscopic polypectomy. Dis Colon Rectum 1993;36:1126-31. 6. Burgess NG, Williams SJ, Hourigan LF, et al. A management algorithm based on delayed bleeding after wide-field endoscopic mucosal resection of large colonic lesions. Clin Gastroenterol Hepatol 2014;12:1525-33. 7. Buchner AM, Guarner-Argente C, Ginsberg GG. Outcomes of EMR of defiant colorectal lesions directed to an endoscopy referral center. Gastrointest Endosc 2012;76:255-63. 8. Di Giorgio P, De Luca L, Calcagno G, et al. Detachable snare versus epinephrine injection in the prevention of postpolypectomy bleeding: a randomized and controlled study. Endoscopy 2004;36:860-3. 9. Lee CK, Lee SH, Park JY, et al. Prophylactic argon plasma coagulation ablation does not decrease delayed postpolypectomy bleeding. Gastrointest Endosc 2009;70:353-61. 10. Li LY, Liu QS, Li L, et al. A meta-analysis and systematic review of prophylactic endoscopic treatments for postpolypectomy bleeding. Int J Colorectal Dis 2011;26:709-19. 11. Buddingh KT, Herngreen T, Haringsma J, et al. Location in the right hemi-colon is an independent risk factor for delayed postpolypectomy hemorrhage: a multi-center case-control study. Am J Gastroenterol 2011;106:1119-24. 12. Sawhney MS, Salfiti N, Nelson DB, et al. Risk factors for severe delayed postpolypectomy bleeding. Endoscopy 2008;40:115-9. 13. Heldwein W, Dollhopf M, Rosch T, et al. The Munich Polypectomy Study (MUPS): prospective analysis of complications and risk factors in 4000 colonic snare polypectomies. Endoscopy 2005;37:1116-22. 14. Pohl H, Grimm IS, Moyer MT, et al. Clip closure prevents bleeding after endoscopic resection of large colon polyps in a randomized trial. Gastroenterology. Epub 2019 Mar 15. 15. Takizawa K, Oda I, Gotoda T, et al. Routine coagulation of visible vessels may prevent delayed bleeding after endoscopic submucosal dissectiondan analysis of risk factors. Endoscopy 2008;40: 179-83. 16. Bahin FF, Naidoo M, Williams SJ, et al. Prophylactic endoscopic coagulation to prevent bleeding after wide-field endoscopic mucosal resection of large sessile colon polyps. Clin Gastroenterol Hepatol 2015;13: 724-30. 17. Veitch AM, Vanbiervliet G, Gershlick AH, et al. Endoscopy in patients on antiplatelet or anticoagulant therapy, including direct oral anticoagulants: British Society of Gastroenterology (BSG) and European Society of Gastrointestinal Endoscopy (ESGE) guidelines. Endoscopy 2016;48: 385-402. 18. Participants in the Paris Workshop. Update on the Paris classification of superficial neoplastic lesions in the digestive tract. Endoscopy 2005;37: 570-8. 19. Kudo S, Lambert R, Allen JI, et al. Nonpolypoid neoplastic lesions of the colorectal mucosa. Gastrointest Endosc 2008;68:S3-47.
Volume
-,
No.
-
: 2019 GASTROINTESTINAL ENDOSCOPY 9
Prophylactic endoscopic coagulation to prevent delayed PPB
Lee et al
20. Hwang MJ, Kim KO, Kim AL, et al. Histologic discrepancy between endoscopic forceps biopsy and endoscopic mucosal resection specimens of colorectal polyp in actual clinical practice. Intest Res 2018;16:475-83. 21. Jaruvongvanich V, Prasitlumkum N, Assavapongpaiboon B, et al. Risk factors for delayed colonic post-polypectomy bleeding: a systematic review and meta-analysis. Int J Colorectal Dis 2017;32: 1399-406. 22. Watabe H, Yamaji Y, Okamoto M, et al. Risk assessment for delayed hemorrhagic complication of colonic polypectomy: polyp-related factors and patient-related factors. Gastrointest Endosc 2006;64: 73-8. 23. Hui AJ, Wong RM, Ching JY, et al. Risk of colonoscopic polypectomy bleeding with anticoagulants and antiplatelet agents: analysis of 1657 cases. Gastrointest Endosc 2004;59:44-8. 24. Choung BS, Kim SH, Ahn DS, et al. Incidence and risk factors of delayed postpolypectomy bleeding: a retrospective cohort study. J Clin Gastroenterol 2014;48:784-9. 25. Kim JH, Lee HJ, Ahn JW, et al. Risk factors for delayed postpolypectomy hemorrhage: a case-control study. J Gastroenterol Hepatol 2013;28:645-9. 26. Saturni S, Bellini F, Braido F, et al. Randomized controlled trials and real life studies. Approaches and methodologies: a clinical point of view. Pulm Pharmacol Ther 2014;27:129-38.
10 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2019
27. Kim HS, Lee S, Kim JH. Real-world evidence versus randomized controlled trial: clinical research based on electronic medical records. J Korean Med Sci 2018;33:e213. Abbreviations: CI, confidence interval; OR, odds ratio; PEB, post-EMR bleeding; PEC, prophylactic endoscopic coagulation. DISCLOSURE: The following author disclosed financial relationships relevant to this publication: S. W. Jeon: Major shareholder and CEO of Finemedix Co, Ltd. All other 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.2019.05.039 Received December 17, 2018. Accepted May 27, 2019. Current affiliations: Department of Internal Medicine, School of Medicine, Kyungpook National University (1) and Department of Internal Medicine (2), Kyungpook National University Hospital, Daegu, South Korea. Reprint requests: Seong Woo Jeon, MD, PhD, Department of Internal Medicine, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, 807 Hoguk-ro, Buk-gu, Daegu, 41404, Korea.
www.giejournal.org