High Joule heat as a risk factor for post-endoscopic submucosal dissection electrocoagulation syndrome: A multicenter prospective study

Abstract

BACKGROUND

Thermal damage may lead to inflammation of the peeled mucosal surface during endoscopic submucosal dissection (ESD).

AIM

To determine the effect of Joule heat on the onset of post-ESD electrocoagulation syndrome (PECS).

METHODS

In this prospective study, PECS was characterized by in-hospital fever (white blood cell count: ≥ 10000 μ/L or body temperature ≥ 37.5 °C) and abdominal pain (visual analog scale score ≥ 30 mm during hospitalization or increased by ≥ 20 mm from baseline at admission). High Joule heat was defined as 15390 J. Between April 2020 and April 2024, 209 patients underwent colorectal ESD; those with intraoperative perforation or penetration were excluded. The remaining 202 patients were divided into the PECS and non-PECS groups.

RESULTS

PECS occurred in 30 (14.9%) patients. Multivariate analysis revealed high Joule heat as an independent factor associated with PECS (odds ratio = 7.96; 95% confidence interval: 2.91-21.8, P < 0.01). The procedure time and presence of lesions in the right colon were not associated with PECS.

CONCLUSION

Accumulated thermal damage on the peeled mucosal surface should be considered during PECS onset. This thermal damage is likely a major component of the mechanism underlying PECS.

Key Words: Colorectal tumors; Colorectal endoscopic submucosal dissection; Joule heat; Post-endoscopic submucosal dissection electrocoagulation syndrome; Thermal damage

Core Tip: This study demonstrated that thermal damage to the peeled mucosal surface during endoscopic submucosal dissection (ESD) is associated with the onset of post-ESD electrocoagulation syndrome (PECS). Analysis of 202 patients with colorectal ESD from April 2020 to April 2024 showed high Joule heat as an independent risk factor for PECS, occurring in 14.9% of the patients. Other factors, such as lesion location in the right colon and procedure time, were not associated with PECS. Accumulated thermal damage on the peeled mucosal surface likely plays an important role in PECS onset.

INTRODUCTION

Colorectal endoscopic submucosal dissection (ESD) is performed as a minimally invasive procedure aimed at achieving the optimal en bloc resection for early-stage colorectal tumors[1-4]. However, ESD is linked to significant complications, such as bleeding (0.7%-3.1%) and perforation (2%-14%)[4-10]. Another notable adverse event associated with ESD is the post-ESD electrocoagulation syndrome (PECS). The reported incidence of PECS ranges from 9% to 40%. However, in many cases, there is a tendency for improvement with conservative therapy[11-15]. In some cases, PECS can present as a delayed perforation, highlighting the importance of identifying patients at risk using clinically relevant PECS predictors[12,16].

Currently, two hypotheses have been proposed to elucidate the mechanisms behind PECS. One study suggested that ESD exposes the intestinal mucosa to intestinal bacteria, resulting in inflammation[13,17]. Another study suggested that the thermal damage during ESD may lead to inflammation of the peeled mucosal surface[13,18]. Neither endoscopic clipping closure nor prophylactic antibiotic, aimed at preventing inflammation from intestinal bacterial infection of the exposed mucosa, was effective in reducing the incidence of PECS[19,20]. However, few studies have investigated the thermodynamic impact during ESD on the occurrence of PECS. A recent study reported that transmural thermodynamic impact on the dissected surface during ESD is likely to be the mechanism underlying the occurrence of PECS[21]; however, this was a retrospective trial. Therefore, in this prospective, multicenter, observational study, we aimed to evaluate the thermodynamic impact on the dissected surface on the incidence of PECS in patients receiving treatment with colorectal ESD.

MATERIALS AND METHODS

Design and patient enrollment

We conducted a prospective, multicenter, observational study involving patients undergoing colorectal ESD as a treatment modality at tertiary Japanese institutions. Patient enrollment was based on the applicability criterion for colorectal ESD outlined in the Japan Gastroenterological Endoscopy Society[22]. In this study, the inclusion criteria were as follows: Carcinomas with shallow T1a invasion (submucosal invasion depth < 1000 μm); lesions that were difficult to remove en bloc using endoscopic mucosal resection, such as large protruding tumors and large depressed tumors; and mucosal lesions accompanied by submucous fibrosis. Patients with T1b invasion (submucosal invasion depth ≥ 1000 μm), multiple lesions per ESD session, and perforation or penetration during colorectal ESD were excluded. Patients underwent outpatient physician’s examinations after 2-3 weeks of colorectal ESD. Physicians evaluated the white blood cell count, C-reactive protein levels, and the intensity of localized abdominal pain using the visual analog scale (VAS) in an outpatient setting.

The study protocol was formulated with Hitachi General Hospital as the main coordinating institution, and was subsequently endorsed by the research ethics committee of joint-research facility. The study was officially registered in the University Hospital Medical Information Network Clinical Trials Registry under the identifier UMIN000041580, and adhered to the guidelines outlined in the 2010 Consolidated Standards of Reporting Trials. Before enrollment, all patients provided written informed consent.

Definition of PECS

The occurrence of post-PECS was defined by patients meeting both of the following conditions: (1) Localized abdominal pain, evaluated with a VAS score ≥ 30 mm during admission, or an increase of ≥ 20 mm from the baseline VAS score at hospital admission; and (2) Fever, indicated by a white blood cell count ≥ 10000/μL or body temperature ≥ 37.5 °C during admission.

ESD procedures

Bowel preparation was conducted by ingesting 2 L of polyethylene glycol before ESD. Subsequently, colorectal ESD procedures were carried out using sedation induced by midazolam (1-5 mg) and/or pentazocine (15-30 mg/kg body weight). Electrocardiographic and respiratory monitoring were performed for each session. ESD was performed by endoscopists comprising experts or trainees. When trainees performed colorectal ESD, they did so under the guidance of experts. Endoscopists who performed > 40 colorectal ESD procedures were defined as experts (four endoscopists), while those who performed < 40 procedures were defined as trainees (13 endoscopists)[23]. A colonoscope (PCF-Q290TI; Olympus, Tokyo, Japan) with a hood (DH-29CR; Fujifilm, Tokyo, Japan) attached to its tip was utilized during colorectal ESD procedures. Marking, mucosal detachment, and mucosal incision were performed using a dual knife (KD-655Q; Olympus). Before performing mucosal detachment, 0.4% sodium hyaluronate diluted in physiological saline was injected directly into the submucosal layer beneath the lesion. We achieved hemostasis using a coagrasper (FD-411QR; Olympus) and an endoscopic clip (HX-610-090; Olympus) for bleeding that occurred during marking, mucosal detachment, and mucosal incision.

Calculation of Joule heat

High-frequency was generated using a high-frequency generator (VIO 300D; ERBE Co. Ltd., Tubingen, Germany) equipped with an electrosurgical coagulation unit from the tip of the dual knife for mucosal detachment. We used a device to convert the time measured in the swift coagulation mode into Joule heat by installing a small micro switch [Miniature basic switch (D3V-016-1C23); OMRON Corporation, Kyoto, Japan] adjacent to the foot switch of the electrosurgical coagulation unit. The small micro switch was activated by pressing the foot switch (Figure 1).

Figure 1  A device used to convert the time measured in the swift coagulation mode into Joule heat by installing a small micro switch adjacent to the foot switch of the electrosurgical coagulation unit.

We calculated the applied power based on a diagram (Effect 4) depicting the relationship between power (W) and resistance (Ω) using swift coagulation mode[21]. The resistance of the lesion targeted for resection was defined as 635 Ω, considering that the internal resistance of the human body typically falls within the range of 535-735 Ω[24]. The electrosurgical coagulation unit can automatically adjust its power output during lesion detachment, but monitoring this automatic adjustment is challenging. By identifying the intersection of the vertical axis [power (W)] and horizontal axis [resistance (Ω)], the power corresponding to 635 Ω was determined to be 80 W[21]. Based on the relationship that power consumption (J) multiplies power (W) by time (s), we calculated the thermally loaded Joule heat on the mucosal surface. This allowed us to compute the total Joule heat applied to the surface. Additionally, high Joule heat was defined as exceeding 15390 J[21].

Post-ESD hospitalization

In all patients, vital signs were monitored on post-operative day (POD) 1 and blood samples were obtained to assess signs of bleeding. Additionally, patients were assessed for pain using VAS, and chest X-ray examination was performed to verify the presence of free air. The patient was allowed to resume eating on POD 1, beginning with dinner. If bleeding, localized abdominal pain, or fever did not occur following the resumption of eating, the patient was discharged on POD 6.

Statistical analysis

Clinicopathological factors and outcomes were analyzed using the Mann-Whitney U test (body mass index, age), the χ² test/Fisher exact test (except body mass index and age) in univariate analysis. The risk assessment of PECS was conducted using logistic regression analysis, incorporating significant factors identified in univariate analysis. The incidence of PECS has been estimated at 10% in previous reports[11-15,25,26]. Considering the higher proportion of trainees, we assumed 7% higher incidence of PECS. With a α error of 0.05 (two-sided) and a 1-β error of 0.8, the required sample size was determined to be 196 cases. We used SPSS software (version 26, IBM Corp., Armonk, NY, United States) for statistical analysis and set the level of statistical significance at a P value of < 0.05.

RESULTS

Participant selection

Patients were selected from April 1, 2020 to April 30, 2024. Of the 209 recruited patients, seven were excluded due to multiple lesions (n = 1) and perforation or penetration during ESD (n = 6). Distinguishing between PECS and penetration or perforation was necessary. Additionally, properly evaluating the relationship between PECS and total Joule heat was crucial. Failure to consider these influences could have skewed the results. Ultimately, 202 patients from five tertiary Japanese institutions were enrolled (Figure 2).

Figure 2 Study flowchart. ESD: Endoscopic submucosal dissection; PECS: Post-endoscopic submucosal dissection electrocoagulation syndrome.

Baseline characteristics

The data of 202 patients (mean age: 69.9 years) were statistically analyzed; of these, 72.8% were male. The patients were categorized based on characteristics such as the sample size (≥ 30 mm, < 30 mm), tumor morphology (protruding tumor, non-granular lateral spreading tumors, granular lateral spreading tumors), tumor location (right colon, left colon, rectum), and Eastern Cooperative Oncology Group performance status (0, 1) (Table 1).

Table 1 Baseline characteristics of eligible patients.

Patient features
Number of patients	202
Male, n (%)	147 (72.8)
Age (years), mean ± SD	69.9 ± 10.1
BMI (kg/m2), mean ± SD	23.7 ± 3.6
Specimen size (mm), mean ± SD	31.7 ± 12.8
Tumor location, n (%)
Right colon	81 (40.1)
Left colon	61 (30.2)
Rectum	60 (29.7)
Tumor morphology, n (%)
0-Is/Ip	88 (43.6)
LST-G	97 (48.0)
LST-NG (0-IIc)	17 (8.4)
ECOG performance status after ESD, n (%)
0	194 (96.0)
≥ 1	8 (4.0)

BMI: Body mass index; Is: Sessile type; Ip: Pedunculated type; LST-G: Lateral spreading tumor granular type; LST-NG: Lateral spreading tumor non-granular type; IIc: Superficial depressed type; ECOG: Eastern Cooperative Oncology Group; ESD: Endoscopic submucosal dissection.

Study outcomes

We conducted an evaluation of the factors associated with PECS occurrence, and high Joule heat showed significantly higher results in univariate analysis. Additionally, high Joule heat was identified as a significantly independent factor for PECS occurrence in multivariate analysis (odds ratio = 7.96; 95% confidence interval: 2.91-21.8; P < 0.01) (Table 2).

Table 2 Clinicopathological factors in the post-endoscopic submucosal dissection electrocoagulation syndrome and non-post-endoscopic submucosal dissection electrocoagulation syndrome groups.

	PECS group, n = 30	Non-PECS group, n = 172	P value		OR (95%CI)
			Univariate analysis	Multivariate analysis
Sex, n (%)			0.52
Male	21 (70.0)	126 (73.3)
Female	9 (30.0)	46 (26.7)
Age, n (%)			0.54
< 75 years	17 (56.7)	112 (65.1)
≥ 75 years	13 (43.3)	60 (34.9)
Specimen size, n (%)			0.24
< 30 mm	11 (36.7)	83 (48.3)
≥ 30 mm	19 (63.3)	89 (48.3)
Tumor location, n (%)			1.00
Right colon	12 (40.0)	69 (40.1)
Left colon	10 (33.3)	51 (29.7)
Rectum	8 (26.7)	52 (30.2)
Tumor morphology, n (%)			0.85
0-Is/Ip	13 (43.3)	75 (43.6)
LST-G	16 (53.4)	81 (47.1)
LST-NG (0-IIc)	1 (3.3)	16 (9.3)
Depth of pathological invasion, n (%)			0.54
Tis (M)	28 (93.4)	152 (88.4)
T1a (SM < 1000 μm)	1 (3.3)	11 (6.4)
T1b (SM ≥ 1000 μm)	1 (3.3)	9 (5.2)
Endoscopist, n (%)			0.30
Experts	12 (40.0)	54 (31.4)
Trainees	18 (60.0)	118 (68.6)
Submucosal fibrosis, n (%)			0.30
Present	6 (20.0)	27 (15.7)
Absent	24 (80.0)	145 (84.3)
ESD procedure time, n (%)			0.18	0.16
< 90 minutes	11 (36.7)	91 (52.9)			1 (reference)
≥ 90 minutes	19 (63.3)	81 (47.1)			1.76 (0.80-3.86)
Total Joule heat, n (%)			< 0.01	< 0.01
Low Joule heat	4 (13.3)	105 (61.0)			1 (reference)
High Joule heat	26 (86.7)	67 (39.0)			7.96 (2.91-21.8)

PECS: Post-endoscopic submucosal dissection electrocoagulation syndrome; ESD: Endoscopic submucosal dissection; T1a (SM): Submucosal invasion depth < 1000 μm; T1b (SM): Submucosal invasion depth ≥ 1000 μm; Tis (M): Carcinoma in situ or intramucosal carcinoma; IIc: Superficial depressed type; LST-NG: Lateral spreading tumor non-granular type; LST-G: Lateral spreading tumor granular type; Ip: Pedunculated type; Is: Sessile type; OR: Odds ratio; CI: Confidence interval.

Treatment outcomes

We compared the long (n = 100) and short (n = 102) procedure time groups, including variables, such as sex, tumor location, ESD procedure performed by trainees, submucosal fibrosis, and PECS. The presence of submucosal fibrosis was significantly higher in the long procedure time group than that in the short procedure time group. However, there was no significant difference in the occurrence of PECS between the two groups (Table 3).

Table 3 Evaluation of the long procedure time group and short procedure time group.

	Long procedure time group (≥ 90 minutes), n = 100	Short procedure time group (< 90 minutes), n = 102	P value
Sex, n (%)			0.75
Male	73 (73.0)	73 (71.6)
Female	27 (27.0)	29 (28.4)
Tumor location, n (%)			0.67
Right colon	42 (42.0)	38 (37.3)
Left colon	26 (26.0)	35 (34.3)
Rectum	32 (32.0)	29 (28.4)
ESD procedure performed by trainees, n (%)	73 (73.0)	63 (61.8)	0.10
Submucosal fibrosis, n (%)	24 (24.0)	9 (8.8)	< 0.01
PECS, n (%)	19 (19.0)	11 (10.8)	0.33

ESD: Endoscopic submucosal dissection; PECS: Post-endoscopic submucosal dissection electrocoagulation syndrome.

Adverse events

Perforation or penetration occurred in one patient (one patient with intraprocedural perforation, 3.3%) in the PECS group and in five patients (two patients with intraprocedural perforation and three patients with penetration; total 2.9%) in the non-PECS group. These patients were managed conservatively without surgical treatment (Table 4).

Table 4 Cases of adverse events in the post-endoscopic submucosal dissection electrocoagulation syndrome and non-post-endoscopic submucosal dissection electrocoagulation syndrome groups.

	PECS group	Non-PECS group
Intraprocedural perforation or penetration/number of patients in each group	1/30	5/172
ESD procedure performed by trainees/number of patients in each group	1/30	5/172
Peak CRP (mg/dL), mean ± SD	4.11 ± not applicable	1.56 ± 1.42
Duration of hospitalization (days), mean ± SD	6 ± not applicable	6.2 ± 2.2

PECS: Post-endoscopic submucosal dissection electrocoagulation syndrome; ESD: Endoscopic submucosal dissection; CRP: C-reactive protein.

DISCUSSION

To our knowledge, this study is the first to demonstrate the significance of total Joule heat magnitude, calculated using a device developed by us, during colorectal ESD as a risk factor for PECS in a multicenter prospective study. Furthermore, a significant increase in the incidence of PECS occurred when a high Joule heat (≥ 15390 J) was reached.

The mechanism of PECS may involve the occurrence of temporary inflammation due to heat transfer from electrocoagulation to the serous membrane and muscle layer at the resection site[18,27,28]. A previously reported risk factor for PECS is the prolonged procedural time[25,29], resulting in an increased electrical load on the mucosal surface[11]. Furthermore, ischemic changes are caused in the mucosal surface due to excessive electrical load during resection, incision, and hemostasis[30]. These reports suggest that the thermodynamic load imposed on the mucosal surface is connected to the occurrence of PECS. Based on these findings, we demonstrated evidence of an association between the total Joule heat imposed on mucosal surface and the occurrence of PECS[21]. However, this was a retrospective study; therefore, it is necessary to investigate the relationship between the Joule heat of mucosal surface and PECS in a multicenter prospective observational study.

Other reported risk factors for PECS include lesion size[11,13,31], lesion location[11,13,25,26,29], submucosal fibrosis[26,31], and injury to the muscle layer[32]. These risk factors may indirectly result in the need for additional Joule heating to detach the lesions. In previous reports, larger tumor size, fibrosis, and fewer experienced endoscopists were associated with longer procedure times (average: 86 minutes)[22]. This indirectly indicates the burden of thermal damage on the dissection surface. However, in our study, despite longer procedure times (average: 98 minutes), tumor size, fibrosis, and less experienced endoscopists were not identified as risk factors for PECS. The lack of an assessment method for thermal damage is the main reason the risk factors for PECS have not been clarified to date. Risk factors, such as procedure time, lesion size, lesion location, submucosal fibrosis, and injury to the muscle layer, suggest a high Joule heat, and the method of accumulating the coagulation heat emitted from the electrosurgical coagulation unit has provided a rational means to evaluate thermal damage. In our study, high Joule heating emerged as the only independent risk factor for PECS, whereas procedure time, lesion size, lesion location, and submucosal fibrosis were not found to be risk factors. Especially, quantifying the amount of heat applied to the mucosal surface is essential to identify the risk factors for PECS. Our results suggest the possibility to predict the occurrence of PECS by quantitatively evaluating thermal damage.

This study has some potential limitations. Patients who underwent colorectal ESD did not routinely undergo computed tomography after the procedure, which potentially led to missed perforations. When evaluating thermal damage, only the mode of swift coagulation was measured, and it was not possible to assess the cutting mode. This was unavoidable because enlarging the device compromised operator performance. Patient movement may increase the heat applied to the dissection surface. Moreover, when the tumor is located between folds, limited visibility can further elevate the heat exposure. Additionally, thermal damage was assessed solely through the mode of swift coagulation, as the cutting mode could not be evaluated without compromising operator performance. Despite these limitations, the primary strength is that it is a prospective study conducted across multiple centers. Our study represents a significant milestone in the evaluation of quantifiable factors indicating PECS.

CONCLUSION

We demonstrated that the accumulation of Joule heat on the peeled surface during ESD contributes to the onset of PECS. Thermodynamic damage to this mucosal surface seems to be the central aspect of the mechanism underlying the onset of PECS.