Management and associated factors of delayed perforation after gastric endoscopic submucosal dissection

Table 1 Clinicopathological findings of 4943 early gastric cancer patients undergoing endoscopic submucosal dissection

Clinicopathological finding	n (%)
Age (yr)
median (range)	69 (27-96)
< 70	2608 (52.8)
≥ 70	2335 (47.2)
Sex
Male	3930 (80.0)
Female	1013 (20.0)
Chronological periods
1st period: 1999-2005	2285 (46.2)
2nd period: 2006-2012	2658 (53.8)
Clinical indications
Absolute indications	2884 (58.3)
Expanded indications	1737 (35.1)
Locally recurrent EGC	141 (2.9)
Outside indications	181 (3.7)
Stomach status
Normal stomach	4704 (95.2)
Remnant stomach	152 (3.1)
Gastric tube	87 (1.7)
Location
Upper	904 (18.3)
Middle	2100 (42.5)
Lower	1939 (39.2)
Circumference
Greater curvature	807 (16.3)
Lesser curvature	2005 (40.6)
Anterior wall	963 (19.5)
Posterior wall	1168 (23.6)
Size (mm)
median (range)	15 (0.4-120)
≤ 20	3457 (69.9)
> 20	1486 (30.1)
Depth of invasion
M	4075 (82.4)
SM	868 (17.6)
Ulceration
Absent	4073 (82.4)
Present	870 (17.6)
Histological type
Differentiated	4581 (92.7)
Undifferentiated	362 (7.3)
Type of resection
En bloc resection	4859 (98.3)
Piecemeal resection	84 (1.7)
Procedure time (h)
mean ± SD	1.4 ± 1.1
< 2	3811 (77.1)
≥ 2	1132 (22.9)

EGC: Early gastric cancer; M: Mucosa; SM: Submucosa.

Table 2 Set-up for the high-frequency generators for endoscopic submucosal dissection along with the IT knife for early gastric cancer

Procedure	Device	Mode	Output
ICC200
Marking	Needle knife	Forced coag	20W
Precutting	Needle knife	ENDO CUT	Effect 3, 80W
Mucosal incision	IT knife	ENDO CUT	Effect 3, 80W
	Needle knife
Submucosal dissection	IT knife	ENDO CUT	Effect 3, 80W
		Forced coag	50W
	Needle knife	ENDO CUT	Effect 3, 80W
		Forced coag	50W
Endoscopic hemostasis	IT knife	Forced coag	50W
	Needle knife
	Hot biopsy	Soft coag	80W
	Coagrasper
ESG100
Marking	Needle knife	Forced coag 1	20W
Precutting	Needle knife	Pulse cut slow	40W
Mucosal incision	IT knife	Pulse cut slow	40W
	Needle knife
Submucosal dissection	IT knife	Pulse cut slow	40W
		Forced coag 2	50W
	Needle knife	Pulse cut slow	40W
		Forced coag 2	50W
Endoscopic hemostasis	IT knife	Forced coag 2	50W
	Needle knife
	Hot biopsy	Soft coag	80W
	Coagrasper
VIO300D
Marking	Needle knife	Swift coag	Effect 2, 50W
Precutting	Needle knife	ENDO CUT I	Effect 2, CUT duration 2, CUT interval 3
Mucosal incision	IT knife	ENDO CUT I or Q	Effect 2, CUT duration 2, CUT interval 3
		DRY CUT	Effect 4, 50W
	Needle knife	ENDO CUT I	Effect 2, CUT duration 2, CUT interval 3
		DRY CUT	Effect 4, 50W
Submucosal dissection	IT knife	ENDO CUT I or Q	Effect 2, CUT duration 2, CUT interval 3
		DRY CUT	Effect 4, 50W
		Swift coag	Effect 5, 50W
	Needle knife	ENDO CUT I	Effect 2, CUT duration 2, CUT interval 3
		DRY CUT	Effect 4, 50W
		Swift coag	Effect 5, 50W
Endoscopic hemostasis	IT knife	Swift coag	Effect 5, 50W
	Needle knife
	Hot biopsy	Soft coag	Effect 5, 80W
	Coagrasper

Table 3 Clinical management of delayed perforation induced by gastric endoscopic submucosal dissection

Case	Age (yr)	Sex	Stomach status	Time until the occurrence of delayed perforation (h)	Panperitonitis or severe mediastinitis	Management of delayed perforation	Hospital stay (d)
1	68	Male	Gastric tube	11	Absent	Conservative management	45
2	75	Male	Normal stomach	35	Absent	Conservative management	18
3	80	Male	Normal stomach	6	Absent	Conservative management with endoloop-endoclip technique	18
4	64	Female	Gastric tube	7	Absent	Conservative management with endoloop-endoclip technique	25
5	73	Male	Normal stomach	9	Present (Panperitonitis)	Emergency surgery	15
6	62	Female	Normal stomach	27	Present (Panperitonitis)	Emergency surgery	18
7	56	Female	Normal stomach	172	Present (Panperitonitis)	Emergency surgery	15

Table 4 Factors associated with delayed perforation induced by gastric endoscopic submucosal dissection n (%)

Clinicopathological finding	Univariate analysis			Multivariate analysis, OR (95%CI), P value
	ESD cases without perforation (n = 4813)	ESD cases with delayed perforation (n = 7)	P value
Age (yr)			1.00	-
< 70	2538 (99.8)	4 (0.2)
≥ 70	2275 (99.9)	3 (0.1)
Sex			0.16	-
Male	3828 (99.9)	4 (0.1)
Female	985 (99.7)	3 (0.3)
Chronological periods			1.00	-
1st period: 1999-2005	2194 (99.9)	3 (0.1)
2nd period: 2006-2012	2619 (99.8)	4 (0.2)
Clinical indications			0.02	NS
Outside indications	169 (98.8)	2 (1.2)
Other indications1	4644 (99.9)	5 (0.1)
Stomach status			0.006	11.0 (1.7-73.3), 0.013
Normal stomach/Remnant stomach	4732 (99.9)	5 (0.1)
Gastric tube	81 (97.6)	2 (2.4)
Location			0.047	NS
Upper/Middle	2894 (99.8)	7 (0.2)
Lower	1919 (100)	0 (0.0)
Circumference			0.09	-
Greater curvature	774 (99.6)	3 (0.4)
Others2	4039 (99.9)	4 (0.1)
Size (mm)			0.43	-
≤ 20	3395 (99.9)	4 (0.1)
> 20	1418 (99.8)	3 (0.2)
Depth of invasion			0.34	-
M	3988 (99.9)	5 (0.1)
SM	825 (99.8)	2 (0.2)
Ulceration			0.34	-
Absent	3982 (99.9)	5 (0.1)
Present	831 (99.8)	2 (0.2)
Histological type			0.09	-
Differentiated	4466 (99.9)	5 (0.1)
Undifferentiated	347 (99.4)	2 (0.6)
Type of resection			1.00	-
En bloc resection	4743 (99.9)	7 (0.1)
Piecemeal resection	70 (100)	0 (0.0)
Procedure time (h)			0.046	NS
< 2	3758 (99.9)	3 (0.1)
≥ 2	1055 (99.6)	4 (0.4)

¹Other indications, absolute indications, expanded indications and locally recurrent early gastric cancer;

²Others, lesser curvature, anterior wall and posterior wall. ESD: Endoscopic submucosal dissection; M: Mucosa; SM: Submucosa; NS: Not significant.