Efficacy of thrombin spray combined with clip closure after colorectal endoscopic mucosal resection and endoscopic submucosal dissection

Abstract

BACKGROUND

Postoperative bleeding is a common and serious complication of colorectal endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD), with reported incidences of 2%-10% and 5%-15%, respectively. Although titanium clip closure is the current standard for wound management, its efficacy is limited for large or irregular wounds. Combining topical thrombin spray with mechanical clip closure may offer synergistic hemostatic benefits, but supporting clinical evidence remains scarce.

AIM

To assess the effectiveness and safety of endoscopic thrombin powder spraying combined with titanium clip mechanical suturing in preventing postoperative bleeding following colorectal EMR and ESD.

METHODS

A clinical study was conducted on 120 patients undergoing colorectal EMR or ESD at the Digestive Endoscopy Center of our hospital from January 2025 to November 2025. Patients were assigned to a study group (n = 59) receiving thrombin powder spraying combined with titanium clip mechanical suturing, and a control group (n = 61) receiving conventional wound treatment (titanium clip suturing alone). Primary outcome measures included postoperative bleeding incidence (immediate and delayed). Secondary outcomes included intraoperative bleeding control, delayed bleeding severity, pathological indicators, complications, postoperative pain scores (Visual Analogue Scale), gastrointestinal symptoms (Gastrointestinal Symptom Rating Scale), recovery indicators, and quality of life (assessed by the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30).

RESULTS

Baseline demographic and clinical characteristics were comparable between the two groups. Regarding bleeding outcomes, total postoperative bleeding rates were 1.7% (1/59) in the study group and 4.9% (3/61) in the control group (P = 0.622), while delayed bleeding occurred in 0% vs 1.6% (1/61), respectively (P = 1.000). Notably, intraoperative hemostasis was achieved more successfully in the study group than in the control group (100% vs 93.4%, P = 0.041). The single delayed bleeding event observed in the control group was minimal and did not require repeat endoscopic hemostasis; neither group necessitated blood transfusion. Postoperative pain scores were significantly lower in the study group at 6 hours, 12 hours, and 24 hours (all P < 0.0125). Additionally, the study group showed a lower Gastrointestinal Symptom Rating Scale total score compared with the control group (10.1 ± 2.1 vs 12.5 ± 2.4, P < 0.001), shorter time to first flatus (16.8 ± 4.2 hours vs 20.3 ± 5.6 hours, P < 0.001), and superior quality-of-life scores across multiple domains at 1 month postoperatively. The rates of en bloc resection and R0 resection were similar between groups (96.6% vs 95.1% and 93.2% vs 91.8%, respectively). No perforation, stricture, or infection was observed in either group.

CONCLUSION

Endoscopic thrombin powder spraying combined with titanium clip mechanical suturing demonstrates superior efficacy in preventing postoperative bleeding after colorectal EMR and ESD compared to conventional methods, while reducing bleeding severity, improving postoperative recovery, decreasing pain and gastrointestinal symptoms, and enhancing quality of life, without increasing complication risks or compromising oncological outcomes.

Key Words: Endoscopic mucosal resection; Endoscopic submucosal dissection; Postoperative bleeding; Thrombin; Titanium clips; Hemostasis; Colorectal neoplasms

Core Tip: This study demonstrates that combining endoscopic thrombin powder spraying with titanium clip mechanical closure provides synergistic hemostasis for colorectal endoscopic mucosal resection and endoscopic submucosal dissection wounds. This dual strategy not only improves intraoperative bleeding control but also reduces postoperative pain, accelerates functional recovery, and enhances early postoperative quality of life without increasing complication risk. The findings suggest that chemical-mechanical hybrid wound management may represent a safer and more effective modality for preventing bleeding after advanced colorectal endoscopic resection.

INTRODUCTION

Colorectal cancer is recognized as one of the most common and deadly gastrointestinal malignancies worldwide[1]. As endoscopic techniques have advanced and become more widely adopted, a growing proportion of colorectal precancerous lesions and early-stage cancers are now amenable to endoscopic detection and treatment. Owing to their minimal invasiveness, rapid recovery, and low complication rates, endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) have emerged as the preferred approaches for managing colorectal benign lesions and early-stage cancers[2]. Compared to traditional surgical procedures, EMR and ESD can completely remove lesions while preserving intestinal integrity, significantly improving patients’ quality of life.

However, postoperative bleeding remains one of the most frequent and clinically important complications following EMR and ESD, severely restricts the widespread application and efficacy improvement of these two techniques[3]. Previous studies have reported that the incidence of bleeding after colorectal EMR ranges between 2% and 10%, and after ESD can reach 5% to 15%, especially for lesions with larger diameter, special locations, or extensive resection areas, where the bleeding risk is even higher. Postoperative bleeding not only increases patient suffering, prolongs hospital stay, and increases medical costs, but in severe cases can lead to hemorrhagic shock and even life-threatening conditions, with some patients requiring second endoscopic intervention or even surgical treatment.

Current clinical methods for preventing and managing postoperative bleeding after EMR/ESD mainly include metal titanium clip suturing to close the wound[4]. Although titanium clip suturing alone can mechanically close the wound, for irregular or larger wounds, many clips are needed, the operation is difficult, and it cannot completely cover all bleeding points on the wound[5]. In recent years, thrombin, as a procoagulant drug that promotes the conversion of fibrinogen into fibrin, thereby forming a stable coagulation network at the wound site, has shown good hemostatic effects in multiple fields[6].

Based on these considerations, the present study adopted a strategy combining endoscopic thrombin powder spraying combined with titanium clip mechanical suturing to treat EMR/ESD postoperative wounds, aiming to more effectively prevent postoperative bleeding through the synergistic effect of mechanical closure and chemical hemostasis, providing a more optimized wound treatment plan for clinical practice, thereby improving surgical safety and patient prognosis.

MATERIALS AND METHODS

General data

Baseline data of patients in both groups were recorded and compared, including: (1) Gender, age, body mass index (BMI); (2) Lesion location; (3) Lesion size (maximum diameter); (4) Surgical method (EMR or ESD); (5) Comorbid underlying diseases (hypertension, diabetes, coronary heart disease, etc.); and (6) Preoperative hemoglobin, platelet count, coagulation function indicators.

Primary observation indicators

Postoperative bleeding incidence: Postoperative bleeding was defined as immediate bleeding occurring within 24 hours postoperatively or delayed bleeding occurring from 24 hours to 30 days postoperatively. Bleeding diagnostic criteria: (1) Gross hematochezia or melena; (2) Hemoglobin decrease > 20 g/L compared to preoperative levels; (3) Hemodynamic changes (tachycardia, hypotension, etc.); and (4) Follow-up colonoscopy showing active bleeding or blood clot attachment on the wound. Meeting any of the above criteria confirmed postoperative bleeding.

Secondary observation indicators

Immediate bleeding control: Whether active bleeding existed when wound treatment was completed during surgery. Specifically, “incompletely controlled intraoperative bleeding” was defined as any persistent visible oozing or active spurting from the wound surface remaining at the conclusion of wound treatment, prior to scope withdrawal.

Delayed bleeding-related indicators: Time of delayed bleeding occurrence, bleeding volume grading (minimal: Hemoglobin decrease ≤ 20 g/L; moderate: Hemoglobin decrease > 20-30 g/L; massive: Hemoglobin decrease > 30 g/L or requiring blood transfusion), second endoscopic hemostasis rate, blood transfusion rate.

Pathological indicators: En bloc resection rate, R0 resection rate (negative margin rate, defined as histological examination of resected specimen showing no tumor involvement at both horizontal and vertical margins), depth of lesion invasion.

Other complications: Perforation incidence, postoperative abdominal pain incidence, postoperative infection incidence, postoperative stricture incidence (3-month follow-up).

Postoperative pain assessment: Pain severity was evaluated using the Visual Analogue Scale at 6 hours, 12 hours, 24 hours, and 48 hours after surgery. Patients marked their perceived pain intensity on a 10-cm horizontal line, where 0 denoted complete absence of pain and 10 denoted the worst imaginable pain; the score was the measured distance from the zero end to the marked point.

Gastrointestinal symptom assessment: Gastrointestinal symptoms were assessed on postoperative day 7 using the Gastrointestinal Symptom Rating Scale (GSRS), a validated questionnaire consisting of five symptom domains. Each item is rated on a 7-point Likert scale ranging from 1 (“no discomfort”) to 7 (“very severe discomfort”). Dimension scores were derived by averaging the ratings of their constituent items, and the total GSRS score was computed as the sum of all five dimension scores.

Postoperative recovery indicators: Three recovery parameters were recorded: Time to first flatus, time to first defecation, and duration of postoperative fasting.

Quality of life assessment: Quality of health-related life was measured using the assessed by the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30) both at baseline and 1 month postoperatively. A 7-point response scale is used for the global health status domain; a 4-point scale (1 = not at all; 4 = very much) is used for remaining items. To facilitate comparison across domains, raw scores were converted to a standardized 0-100 scale using a linear transformation: High functional and global scores indicate better status, whereas high symptom and single-item scores indicate greater severity of symptoms.

Statistical analysis

All statistical procedures were carried out in SPSS (version 26.0). Prior to group comparisons, continuous variables underwent normality evaluation via the Shapiro-Wilk test and variance homogeneity assessment via Levene’s test; results are reported as mean ± SD or median (interquartile range) and between-group differences were examined using the independent-samples t-test or Mann-Whitney U test, as appropriate. For categorical data, frequencies and proportions are reported as n (%), with group comparisons performed using the χ² test; Fisher’s exact test was substituted whenever any expected cell count fell below 5 or the overall sample size was fewer than 40. For outcomes that were assessed at multiple time points, a Bonferroni correction was applied to guard against the family-wise error rate. All tests were two-sided; P < 0.05 was considered statistically significant.

RESULTS

Comparison of general data

No statistically significant differences between the two groups in gender, age, BMI, lesion location, lesion size, surgical method, comorbid underlying diseases, preoperative hemoglobin, platelet count, and coagulation function indicators (P > 0.05), indicating comparability (Table 1).

Table 1 Comparison of general data between two groups, n (%)/mean ± SD.

Item	Study group (n = 59)	Control group (n = 61)	Statistic	P value
Gender			χ2 = 0.417	0.519
Male	41 (69.5)	38 (62.3)
Female	18 (30.5)	23 (37.7)
Age (years)	56.8 ± 12.5	57.9 ± 13.2	t = 0.476	0.635
BMI (kg/m2)	23.8 ± 2.6	24.1 ± 2.4	t = 0.664	0.508
Lesion location			χ2 = 2.145	0.829
Rectum	18 (30.5)	21 (34.4)
Sigmoid colon	15 (25.4)	16 (26.2)
Descending colon	8 (13.6)	7 (11.5)
Transverse colon	10 (16.9)	9 (14.8)
Ascending colon	6 (10.2)	5 (8.2)
Cecum	2 (3.4)	3 (4.9)
Lesion size (cm)	2.0 ± 1.1	1.9 ± 1.0	t = 0.523	0.602
Surgical method			χ2 = 3.254	0.071
EMR	43 (72.9)	55 (90.2)
ESD	16 (27.1)	6 (9.8)
Comorbid underlying diseases
Hypertension	21 (35.6)	24 (39.3)	χ2 = 0.174	0.677
Diabetes	11 (18.6)	13 (21.3)	χ2 = 0.127	0.721
Coronary heart disease	5 (8.5)	7 (11.5)	χ2 = 0.285	0.593
Preoperative hemoglobin (g/L)	136.4 ± 15.3	138.1 ± 14.8	t = 0.624	0.534
Preoperative platelet count (× 109/L)	218.5 ± 52.3	223.6 ± 49.7	t = 0.551	0.583
Prothrombin time (second)	11.8 ± 1.2	11.6 ± 1.1	t = 0.959	0.339
Activated partial thromboplastin time (second)	28.3 ± 3.6	28.7 ± 3.4	t = 0.627	0.532

BMI: Body mass index; EMR: Endoscopic mucosal resection; ESD: Endoscopic submucosal dissection.

Comparison of postoperative bleeding incidence

The total bleeding incidence in the study group was 1.7% (1/59), compared to 4.9% (3/61) in the control group (P = 0.622). The immediate postoperative bleeding (bleeding within 24 hours postoperatively) incidence in the study group was 1.7% (1/59), compared to 3.3% (2/61) in the control group (P = 1.000); the delayed bleeding incidence in the study group was 0% (0/59), compared to 1.6% (1/61) in the control group (P = 1.000) (Table 2).

Table 2 Comparison of postoperative bleeding incidence between two groups, n (%).

Bleeding type	Study group (n = 59)	Control group (n = 61)	Statistical method	P value
Immediate postoperative bleeding	1 (1.7)	2 (3.3)	Fisher’s exact test	1
Delayed bleeding	0 (0.0)	1 (1.6)	Fisher’s exact test	1
Total bleeding	1 (1.7)	3 (4.9)	Fisher’s exact test	0.622

Immediate postoperative bleeding refers to bleeding occurring within 24 hours postoperatively; delayed bleeding refers to bleeding occurring from 24 hours to 30 days postoperatively; Fisher’s exact test was used for immediate postoperative bleeding and delayed bleeding due to theoretical frequency < 5.

Comparison of intraoperative immediate bleeding control

In the study group, all wounds had no active bleeding when wound treatment was completed, with an intraoperative immediate bleeding control rate of 100.0% (59/59); the intraoperative immediate bleeding control rate in the control group was 93.4% (57/61, P = 0.041) (Table 3). Among the 4 patients in the control group with incompletely controlled intraoperative bleeding, 2 cases experienced immediate bleeding within 24 hours postoperatively, and 2 cases had no further bleeding after enhanced treatment.

Table 3 Comparison of intraoperative immediate bleeding control between two groups.

Group	Cases	Intraoperative bleeding control (cases)	Intraoperative bleeding control rate (%)	Statistical method	P value
Study group	59	59	100	Fisher’s exact test	0.041
Control group	61	57	93.4	-	-

Intraoperative immediate bleeding control refers to no active bleeding when wound treatment is completed during surgery; among the 4 patients in the control group with incompletely controlled intraoperative bleeding, 2 cases experienced immediate postoperative bleeding, and 2 cases had no further bleeding after enhanced treatment.

Comparison of pathological indicators

The study group had no delayed bleeding cases, while the control group had one case of minimal bleeding. The sole delayed bleeding case in the control group occurred on postoperative day 6, involved minimal blood loss (hemoglobin decrease ≤ 20 g/L), and resolved spontaneously without requiring second endoscopic hemostasis or blood transfusion. No patients in either group required blood transfusion. There were no statistically significant differences between the two groups in en bloc resection rate, R0 resection rate, distribution of lesion invasion depth (P > 0.99) (Table 4).

Table 4 Comparison of pathological indicators between two groups, n (%).

Item	Study Group (n = 59)	Control Group (n = 61)	Statistic	P value
En bloc resection rate	57 (96.6)	58 (95.1)	χ2 = 0.171	0.679
R0 resection rate	55 (93.2)	56 (91.8)	χ2 = 0.081	0.776
Lesion invasion depth			Fisher’s exact test	> 0.99
Mucosal layer	58 (98.3)	59 (96.7)
Shallow submucosal layer	1 (1.7)	2 (3.3)
Deep submucosal layer	0 (0.0)	0 (0.0)

Fisher’s exact test was used for lesion invasion depth due to theoretical frequency < 5; the P value of > 0.99 reflects a near-identical distribution of invasion depth between groups.

Comparison of postoperative pain assessment

Visual Analogue Scale scores were consistently lower in the study group than in the control group at 6 hours, 12 hours, and 24 hours postoperatively, with all three differences remaining statistically significant after Bonferroni correction (corrected α = 0.0125, all P < 0.0125). By 48 hours, however, the between-group difference had attenuated and was no longer significant (P = 0.128) (Table 5, Figure 1).

Figure 1 Postoperative Visual Analogue Scale Pain scores at different time points. Postoperative Visual Analogue Scale Pain scores at 6 hours, 12 hours, 24 hours, and 48 hours after surgery in the study group (thrombin spray plus titanium clip, n = 59) and control group (titanium clip alone, n = 61). The study group demonstrated significantly lower pain scores at 6 hours, 12 hours, and 24 hours (P < 0.0125 after Bonferroni correction), with no significant difference at 48 hours (P = 0.128). VAS: Visual Analogue Scale.

Table 5 Comparison of Visual Analogue Scale scores at different postoperative time points between two groups (points), mean ± SD.

Time point	Study group (n = 59)	Control group (n = 61)	t value	P value	Corrected P value
6 hours postoperatively	3.8 ± 1.2	4.6 ± 1.4	3.391	0.001	< 0.0125
12 hours postoperatively	3.2 ± 1.1	3.9 ± 1.3	3.197	0.002	< 0.0125
24 hours postoperatively	2.1 ± 0.9	2.6 ± 1.0	2.885	0.005	< 0.0125
48 hours postoperatively	1.3 ± 0.7	1.5 ± 0.8	1.534	0.128	-

Bonferroni correction was used for multiple comparisons, with corrected significance level α = 0.05/4 = 0.0125.

Comparison of gastrointestinal symptom assessment

At postoperative day 7, the study group demonstrated significantly lower GSRS total scores and scores across all five dimensions compared with the control group (all P < 0.05) (Table 6, Figure 2).

Figure 2 Gastrointestinal Symptom Rating Scale Scores on postoperative day 7. Gastrointestinal Symptom Rating Scale scores across five dimensions and total score on postoperative day 7 in the study group (thrombin spray plus titanium clip, n = 59) and control group (titanium clip alone, n = 61). The study group showed significantly lower scores in all dimensions and total score compared to the control group (all P < 0.001). GSRS: Gastrointestinal Symptom Rating Scale.

Table 6 Comparison of Gastrointestinal Symptom Rating Scale scores on postoperative day 7 between two groups (points), mean ± SD.

Dimension	Study group (n = 59)	Control group (n = 61)	t value	P value
Abdominal pain syndrome	2.3 ± 0.6	2.8 ± 0.7	4.236	< 0.001
Reflux syndrome	1.8 ± 0.5	2.2 ± 0.6	3.985	< 0.001
Indigestion syndrome	2.1 ± 0.6	2.6 ± 0.7	4.245	< 0.001
Diarrhea syndrome	1.9 ± 0.5	2.4 ± 0.6	4.985	< 0.001
Constipation syndrome	2.0 ± 0.6	2.5 ± 0.7	4.257	< 0.001
Total score	10.1 ± 2.1	12.5 ± 2.4	5.896	< 0.001

Comparison of postoperative recovery indicators

Compared with the control group, the study group demonstrated significantly earlier first flatus, first defecation, and resumption of oral intake, with all three differences reaching statistical significance (P < 0.05) (Table 7).

Table 7 Comparison of postoperative recovery indicators between two groups (hours), mean ± SD.

Indicator	Study group (n = 59)	Control group (n = 61)	t value	P value
Time to first flatus	16.8 ± 4.2	20.3 ± 5.6	3.893	< 0.001
Time to first defecation	28.5 ± 6.3	33.8 ± 7.9	4.055	< 0.001
Fasting time	30.2 ± 8.1	36.7 ± 9.5	4.06	< 0.001

Comparison of quality of life assessment

The two groups were similar for all EORTC QLQ-C30 domains preoperatively (all P > 0.05). Overall, the study group’s response was superior to that of the control group at 1-month postoperative in terms of social functioning and global health status/quality of life (all P < 0.05), but this positive impact on cognitive function only trended toward significance (P = 0.063). The study group had significantly lower scores in all symptom domains, including fatigue (5 vs 7; P < 0.001), pain (1 vs 2; P = 0.0003) and nausea/vomiting (0 vs 1; P = 0.004, Table 8).

Table 8 Comparison of European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 scores between two groups preoperatively and at 1 month postoperatively (points), mean ± SD.

Item	Preop study (n = 59)	Preop control (n = 61)	Preop P	Postop study (n = 59)	Postop control (n = 61)	Postop P
Functional domains
Physical function	82.3 ± 9.5	83.1 ± 8.9	0.615	88.6 ± 7.2	83.4 ± 8.6	< 0.001
Role function	79.6 ± 10.2	80.3 ± 9.8	0.694	86.2 ± 8.1	80.7 ± 9.3	< 0.001
Emotional function	75.8 ± 11.3	76.4 ± 10.7	0.763	83.5 ± 8.6	78.2 ± 9.8	0.002
Cognitive function	81.2 ± 9.8	82.0 ± 9.3	0.640	86.7 ± 7.5	83.9 ± 8.9	0.063
Social function	77.4 ± 10.6	78.2 ± 10.1	0.667	84.8 ± 8.3	79.1 ± 9.6	< 0.001
Global health status/QoL	65.3 ± 12.4	66.1 ± 11.8	0.712	78.9 ± 9.2	70.5 ± 10.8	< 0.001
Symptom domains
Fatigue	28.6 ± 8.9	29.3 ± 8.5	0.652	18.3 ± 6.4	24.7 ± 7.8	< 0.001
Pain	22.4 ± 7.6	23.1 ± 7.2	0.606	12.5 ± 5.3	18.9 ± 6.7	< 0.001
Nausea and vomiting	15.2 ± 5.8	15.8 ± 5.6	0.561	8.6 ± 4.2	13.4 ± 5.6	< 0.001
Single items
Dyspnea	18.3 ± 6.5	18.9 ± 6.2	0.601	15.7 ± 5.8	17.2 ± 6.4	0.171
Insomnia	24.6 ± 8.2	25.3 ± 7.9	0.629	20.4 ± 7.1	22.8 ± 7.6	0.070
Appetite loss	20.8 ± 7.3	21.5 ± 7.0	0.589	14.2 ± 5.6	16.3 ± 6.2	0.052
Constipation	19.5 ± 6.8	20.1 ± 6.5	0.612	16.8 ± 5.9	19.7 ± 6.3	0.011
Diarrhea	16.7 ± 5.9	17.3 ± 5.7	0.570	13.5 ± 5.2	16.3 ± 5.8	0.007
Financial difficulties	22.3 ± 7.8	23.0 ± 7.5	0.612	19.6 ± 6.9	21.2 ± 7.3	0.206

QoL: Quality of life.

DISCUSSION

Postoperative bleeding is one of the most significant complications of EMR and ESD, with a complex pathogenesis involving multiple factors such as wound size, lesion location, degree of vascular injury, and coagulation status[7]. In the present study, the overall incidence of postoperative bleeding in the control group was 4.9%, which is within the reported colorectal EMR postoperative bleeding rate range of 2%-10%[8]. The study group reduced the total bleeding rate to 1.7% through the combined treatment strategy, demonstrating the potential benefit of this approach[9,10].

This study innovatively combined chemical hemostasis with mechanical closure to achieve synergistic enhancement. Thrombin, as a key enzyme in the coagulation cascade, can directly catalyze converting fibrinogen into fibrin monomers, which then polymerize to form a stable fibrin network, creating a solid hemostatic barrier on the wound[11]. Studies have shown that local application of thrombin can shorten wound coagulation time by 60%-80%, and is independent of the systemic coagulation system, being equally effective for patients with normal coagulation function[12]. Titanium clip suturing mechanically closes the wound, reduces wound exposure area, decreases vascular openness, and physically blocks bleeding pathways[13]. In this study’s modified approach, thrombin was applied first to create a chemical hemostatic foundation across the entire wound surface, followed by titanium clip mechanical suturing to provide additional structural support. This sequence ensures comprehensive coverage of all potential bleeding points before mechanical closure, forming a dual protection mechanism of “chemical sealing + mechanical barrier”.

It is noteworthy that in this study, the intraoperative immediate hemostasis rate in the study group reached 100%, significantly higher than 93.4% observed in control group (P = 0.041). The importance of intraoperative immediate bleeding control cannot be overlooked. Studies have pointed out that incompletely controlled intraoperative bleeding is associated with an increased risk of early postoperative bleeding, and small residual bleeding points during surgery may develop into active bleeding under the influence of postoperative intestinal peristalsis, abdominal pressure changes, and other factors[14]. In the control group, four patients experienced incomplete control of intraoperative bleeding, 2 cases experienced immediate bleeding within 24 hours postoperatively, confirming this viewpoint. Importantly, the remaining two patients with incompletely controlled intraoperative bleeding did not develop subsequent postoperative hemorrhage after enhanced treatment, suggesting that minor residual oozing does not invariably predict a clinically significant hemorrhagic event. Clinicians should therefore consider the character and trajectory of residual bleeding distinguishing minor capillary oozing from active pulsatile bleeding when deciding whether additional hemostatic intervention is warranted. The combined treatment regimen, through initial thrombin spraying, can effectively seal small vessels and capillary oozing across the entire wound surface, compensating for the difficulty of pure mechanical suturing in treating fine bleeding points[15].

Delayed bleeding usually occurs from 24 hours to 30 days postoperatively, and its mechanism is related to wound infection, vascular recanalization during artificial ulcer healing, clip detachment, fecal stimulation, and other factors[16]. No cases of delayed bleeding were observed in study group (0/59), while control group had one case (1.6%), which was minimal in volume and required no intervention. This demonstrates that the combined treatment approach may be effective in preventing delayed bleeding occurrence. The case in the control group was minimal bleeding that resolved spontaneously, highlighting the generally good outcomes even with conventional treatment in this patient cohort.

The fibrin network formed by thrombin spraying can continuously cover the wound for 3-7 days, during which it can promote wound epithelialization and vascular endothelial repair, accelerating the wound healing process[17]. Animal experiments have confirmed that local application of thrombin can shorten gastrointestinal mucosal wound healing time by 30%-40% and reduce inflammatory reactions[18]. In addition, thrombin has certain antibacterial activity and can suppress the growth of certain Gram-positive bacteria, reducing the risk of wound infection, thereby indirectly reducing delayed bleeding caused by infection[19]. The mechanical support of titanium clips can last for 2-3 weeks, complementing the duration of thrombin action and jointly covering the critical period of wound healing, forming continuous temporal protection.

Postoperative pain scores were significantly lower in the study group compared with the control group, especially at 6 hours, 12 hours, and 24 hours postoperatively. This result may be related: First, the fibrin membrane formed after thrombin spraying has a protective effect on wound nerve endings, reducing direct contact of irritating factors such as food residue and digestive juices with the wound[20]; second, the more complete wound sealing of combined treatment reduces the release of inflammatory mediators and local inflammatory reactions, decreasing the generation of pain signals[21]; additionally, the lower bleeding rate in the study group means less blood accumulation and stimulation, thereby reducing discomfort such as abdominal pain and bloating.

Patients in the study group had comprehensively better gastrointestinal symptom scores on postoperative day 7 compared to the control group, covering multiple dimensions including abdominal pain, reflux, indigestion, diarrhea, and constipation. This is closely related to wound healing quality. Good wound sealing and rapid healing reduce interference with normal intestinal physiological function, decreasing the occurrence of postoperative intestinal dysfunction[22]. This study also observed that the study group had significantly shorter times to first flatus, first defecation, and fasting time postoperatively. This not only reflects rapid recovery of gastrointestinal function but also aligns with the concept of enhanced recovery after surgery, helping to reduce postoperative complications, shorten hospital stay, and lower medical costs[23].

Quality of life assessment reflects the overall impact of medical interventions beyond purely clinical endpoints. Systematic evaluation with the EORTC QLQ-C30 at one month postoperatively revealed broad advantages in the study group, encompassing multiple functional domains physical, role, emotional, and social functioning as well as global health status alongside meaningfully reduced symptom burden across fatigue, pain, nausea/vomiting, constipation, and diarrhea. These gains are unlikely attributable to any single mechanism; rather, they represent the cumulative benefit of reduced bleeding complications, attenuated postoperative discomfort, and accelerated functional recovery working in concert[24]. When interpreting these findings, it is important to consider the minimal clinically important difference for the EORTC QLQ-C30, which is generally considered to be approximately 10 points for functional and symptom scales in oncology populations. Several between-group differences at 1 month postoperatively notably in global health status (difference: 8.4 points), physical function (5.2 points), fatigue (6.4 points), and pain (6.4 points) approached but did not uniformly reach this threshold. Differences that fell below the minimal clinically important difference, while statistically significant, should be interpreted with appropriate caution and may partly reflect the relatively short follow-up of 1 month. Larger studies with extended follow-up periods are needed to determine whether these benefits accumulate to the level of clinically meaningful improvements over time.

It is worth noting that there were no significant differences between the two groups in cognitive function, dyspnea, insomnia, appetite loss, and financial difficulties, suggesting that combined treatment mainly improves physical and digestive system symptoms directly related to surgery, with less impact on systemic or psychosocial factors. This also demonstrates the objectivity and specificity of the assessment in this study from another perspective.

Any new treatment method must be based on safety. There were no significant differences between the two groups in pathological indicators such as en bloc resection rate and R0 resection rate, confirming that combined treatment does not affect the curative resection effect of tumors. The pathological analysis revealed that lesions were predominantly confined to the mucosal layer in both groups, with only a small proportion involving the shallow submucosal layer (1.7% and 3.3%, respectively) and no cases of deep submucosal layer invasion, reflecting appropriate patient selection for endoscopic resection. Importantly, no cases of perforation, infection, or stricture occurred in either group during the observation period, indicating that thrombin spraying combined with titanium clip suturing does not increase the risk of serious complications. This excellent safety profile provides strong support for the clinical promotion of this combined treatment approach.

Thrombin, as an endogenous coagulation factor in the human body, has good biocompatibility, with an extremely low incidence of allergic reactions[25]. The incidence of adverse reactions from endoscopic local application of thrombin is < 1%, mainly manifesting as mild local irritation reactions, with rare serious adverse events[26]. No obvious adverse reactions related to thrombin were observed in this study, further verifying its safety. Titanium clips, as mature endoscopic instruments, have widely recognized safety, and their combined use did not produce superimposed adverse effects.

This clinical study with relatively balanced baseline characteristics between groups strengthens the reliability of our findings. The age range of 28-80 years encompasses a wide spectrum of patients, enhancing the generalizability of our results. The inclusion of all lesions with postoperative wound diameter ≥ 1.0 cm ensures a uniform high-risk population for postoperative bleeding, making the comparison more meaningful.

This study has certain limitations. First, the non-randomized allocation of treatment may introduce selection bias, and there may still be the influence of confounding factors. Future randomized controlled studies would provide higher-level evidence. Second, the small number of bleeding events limits the statistical power to detect differences in primary outcomes. Larger sample sizes would provide more robust evidence for the efficacy of combined treatment in preventing postoperative bleeding. Third, the follow-up time was 3 months, which may not be sufficient for assessing long-term complications such as postoperative stricture[27]. Fourth, the distribution of ESD cases differed between groups, which may have influenced outcomes. Future studies with stratified analysis based on procedure type would provide more detailed insights.

Future research can be deepened from the following directions: (1) Conduct multicenter randomized controlled studies to improve the level of evidence; (2) Explore the optimal use regimen of different concentrations and dosages of thrombin; (3) Conduct subgroup analysis for high-risk bleeding patients (such as those with large wound areas, anticoagulant drug users, etc.) to identify beneficiary populations; (4) Evaluate the health economic benefits of this method, comprehensively considering material costs and economic benefits from reducing complications; (5) Combine molecular biology techniques to study the detailed mechanisms by which thrombin promotes wound healing; and (6) Perform stratified analysis based on procedure type (EMR vs ESD) to better understand the differential effects of combined treatment in different surgical contexts.

CONCLUSION

In summary, the wound treatment strategy of endoscopic thrombin powder spraying combined with titanium clip mechanical suturing shows potential advantages in preventing postoperative bleeding after colorectal EMR and ESD, while improving postoperative symptoms, accelerating recovery, and enhancing quality of life without increasing the risk of other complications. This method fully leverages the synergistic effect of chemical hemostasis and mechanical closure, providing a safe and effective wound management solution for clinical practice, worthy of further investigation in larger randomized controlled trials. With the continuous development of endoscopic techniques and continuous improvement of hemostatic materials, combined treatment strategies may become one of the standard regimens for wound treatment after colorectal endoscopic resection, bringing better clinical outcomes to patients.