Impact of prior abdominal surgery on post-polypectomy complications following endoscopic mucosal resection and endoscopic submucosal dissection

Abstract

BACKGROUND

Endoscopic resection, encompassing both endoscopic mucosal resection and endoscopic submucosal dissection (ESD), is the mainstay intervention for colorectal polyps; however, post-procedural complications persist. Although patient and polyp factors are known modulators, the effect of prior abdominal surgery (PAS), which potentially leads to adhesions and an elevated risk, is poorly defined. However, a key unanswered question is whether PAS is an independent risk factor. Resolving this issue is essential for optimizing pre-operative risk assessment and personalized patient management.

AIM

To evaluate the association between PAS and complication risk after endoscopic resection procedures to inform clinical decision-making.

METHODS

This single-center retrospective cohort study included 80 patients who underwent endoscopic resection for colorectal polyps, stratified into PAS (n = 31) and non-PAS (n = 49) groups. Demographics, polyp characteristics, procedural variables, and post-procedural recovery indices were extracted. The primary endpoint was 30-day clinically significant complications (post-polypectomy bleeding, perforation, infection, fever ≥ 38 °C, or moderate-to-severe abdominal pain). Independent predictors were identified using multivariate logistic regression and discriminative performance was quantified using receiver operating characteristic curve analysis.

RESULTS

The overall complication rate was higher in the PAS group than in the non-PAS group (P < 0.05). Patients with PAS exhibited larger polyps, higher polyp multiplicity, a greater proportion of ESD, longer procedure times, more frequent hemoclip use, prolonged hospital stay, delayed resumption of oral intake and ambulation, and elevated post-procedural pain scores (P < 0.05). Multivariate analysis identified PAS, arterial hypertension, hyperlipidemia, anticoagulant/antiplatelet therapy, polyp diameter, polyp multiplicity, ESD as the resection modality, prolonged procedure time, and hemoclip deployment as independent risk factors for complications. Receiver operating characteristic - derived area under the curve values for these variables ranged from 0.71 to 0.84, indicating moderate-to-good predictive accuracy.

CONCLUSION

PAS independently predicts endoscopic resection complications, increasing technical difficulty, procedural time, and the risk of adverse events. Identifying pre-procedure PAS status should inform risk stratification and peri-operative management.

Key Words: Endoscopic polypectomy; Endoscopic mucosal resection; Endoscopic submucosal dissection; Post-operative complications; Prior abdominal surgery

Core Tip: Prior abdominal surgery (PAS) is an independent risk factor for polypectomy complications after endoscopic mucosal resection and endoscopic submucosal dissection. This retrospective cohort study of 80 patients demonstrated that PAS increased technical difficulty, prolonged procedure time, and elevates rates of bleeding, perforation, infection, and post-procedural pain. Larger and multiple polyps, the endoscopic submucosal dissection modality, and hemoclip use further contribute to the risk of complications. Preoperative recognition of PAS status can guide individualized risk stratification, optimize procedural planning, and inform perioperative management strategies, ultimately improving patient safety and clinical outcomes in colorectal polypectomies.

INTRODUCTION

Adenomatous polyps, which are benign neoplasms of the gastrointestinal mucosa, are well-established premalignant lesions, whose timely removal reduces the incidence and mortality[1]. Endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) have therefore become the standard-of-care for sessile and flat lesions ≥ 10 mm. Compared to open or laparoscopic surgery, EMR/ESD is minimally invasive, preserves bowel continuity, and accelerates convalescence. As the cornerstones of minimally invasive techniques, EMR and ESD offer significant clinical value by enabling organ preservation and reducing surgical morbidity. Nevertheless, although well established and generally safe, these procedures are associated with risks. Post-procedural bleeding, perforation, and intraluminal sepsis still occur in 5%-15% of cases, prolonging hospital stay and compromising quality of life[2].

Prior abdominal surgery, particularly laparotomy, has been linked to longer procedural times, impaired luminal visualization, and higher adverse event rates after endoscopic resection[3,4]. Putative mechanisms include adhesive disease, obliteration of normal tissue planes, altered mesenteric perfusion, and increased mural friability[5]. Although polyp size, location, Paris morphology, resection modality, and antithrombotic therapy are validated risk modifiers, the independent contribution of previous abdominal surgery remains contentious and no quantitative risk score incorporates this variable[6]. Clarifying this association would refine the preprocedural risk stratification and inform tailored perioperative management.

Therefore, we conducted a retrospective cohort study to quantify the impact of prior abdominal surgery on complication rates after EMR and ESD, and to derive evidence-based recommendations for clinical practice.

MATERIALS AND METHODS

Study population

This study retrospectively reviewed consecutive patients who underwent endoscopic resection of colorectal polyps at our center between January 2021 and December 2024. This study and the electronic medical records were validated.

Inclusion criteria: (1) Age ≥ 18 years; (2) Histologically verified colorectal adenomatous polyp[7]; (3) Treatment by EMR or ESD; (4) Written informed consent obtained before the procedure; and (5) Complete documentation.

Exclusion criteria: (1) Severe cardiac, hepatic, or renal failure; (2) Inherited coagulopathy or immunodeficiency; (3) Pregnancy or lactation; (4) Previous curative resection for colorectal cancer; and (5) Chemotherapy or radiotherapy within six months before index endoscopy.

Eighty patients fulfilled these criteria and were allocated to either the prior abdominal surgery (PAS) group (n = 31) or the non-PAS group (n = 49) according to surgical history.

Observation indicators - baseline and procedural characteristics

Demographics and comorbidities: Age, sex, body-mass index, smoking (≥ 1 cigarette every day for ≥ 1 year), alcohol consumption (≥ 3 episodes every week for ≥ 1 year), arterial hypertension, type 2 diabetes mellitus, hyperlipidemia, and pre-procedural medications including anticoagulants (warfarin, direct oral anticoagulants) and antiplatelet agents (acetylsalicylic acid, clopidogrel).

Polyp characteristics: Location (left colon: Descending, sigmoid, rectum; right colon: Cecum, ascending, transverse), maximum diameter (mm), morphology according to the Paris classification[8] (pedunculated, semi-pedunculated, sessile), and multiplicity (single vs multiple).

Procedural variables: They included resection modality (EMR or ESD), procedure duration (min, from first submucosal injection to final hemostasis), use of hemoclips for defect closure, and immediate intraprocedural bleeding or perforation.

Recovery indices: Time to first oral fluid intake (hours), time to first ambulation (hours), and total length of hospital stay (days).

Post-procedural adverse events

Bleeding: Hematochezia within 24 hours with a ≥ 2 g/dL drop in hemoglobin or requirement for endoscopic re-intervention or transfusion.

Perforation: Endoscopically visible pneumoperitoneum or free intraperitoneal air on plain abdominal radiography or computed tomography.

Infection: Core temperature > 38.5 °C for > 24 hours plus leukocytosis (> 12 × 10⁹/L) with or without peritoneal signs.

Pain: Moderate-to-severe abdominal pain defined as a Visual Analog Scale score ≥ 4.

Statistical analysis

All analyses were performed using SPSS version 26.0. Continuous variables are expressed as mean ± SD and compared with the independent-samples t test. Categorical variables are presented as n (%) and compared with the χ² test or Fisher’s exact test as appropriate. Independent predictors of post-procedural complications were identified using multivariate logistic regression analysis. Statistical significance was defined as a two-tailed P-value < 0.05.

RESULTS

Baseline characteristics

No significant between-group differences were observed in sex, age, body-mass index, smoking status, alcohol consumption, or diabetes mellitus (P > 0.05). In contrast, the prevalence of arterial hypertension, hyperlipidemia, anticoagulant/antiplatelet therapy, and PAS differed significantly between the two cohorts (P < 0.05; Table 1).

Table 1 Baseline characteristics of the study cohort, mean ± SD/n (%).

Variable	PAS (n = 31)	NPAS (n = 49)	χ2/t	P value
Sex			0.024	0.877
Male	17 (54.84)	26 (53.06)
Female	14 (45.16)	23 (46.94)
Ages (olds)	56.73 ± 8.12	57.80 ± 8.09	0.576	0.567
BMI (kg/m2)	23.45 ± 2.10	22.70 ± 1.89	1.656	0.102
Smoking history	17 (54.84)	19 (38.78)	1.614	0.204
Alcohol consumption	9 (29.03)	13 (26.53)	0.060	0.807
Comorbidities
Arterial hypertension	18 (58.07)	16 (32.65)	5.017	0.025
Type 2 diabetes mellitus	10 (32.26)	8 (16.33)	2.764	0.096
Hyperlipidemia	12 (38.71)	7 (14.29)	6.255	0.012
Anticoagulant/antiplatelet therapy	15 (48.39)	9 (18.37)	8.148	0.004

BMI: Body-mass index; PAS: Prior abdominal surgery; NPAS: Non-prior abdominal surgery.

Polyp characteristics and procedural variables

There were no significant intergroup differences in polyp location or Paris morphology (P > 0.05). In contrast, the maximum polyp diameter, multiplicity, resection modality (EMR vs ESD), procedure duration, and rate of hemoclip deployment differed significantly between the PAS and no-prior surgery groups (P < 0.05, Table 2).

Table 2 Comparison of polyposis lesions and surgical-related indicators between two patient groups, mean ± SD/n (%).

Variable	PAS (n = 31)	NPAS (n = 49)	χ2/t	P value
Polyp location			0.192	0.661
Left colon	18 (58.07)	26 (53.06)
Right colon	13 (41.93)	23 (46.94)
Polyp size (mm)	16.45 ± 4.98	13.20 ± 4.67	2.956	0.004
Number of polyps	1.87 ± 0.72	1.52 ± 0.39	2.818	0.006
Polyp morphology			0.298	0.585
Pedunculated	12 (38.71)	22 (44.90)
Sessile	19 (61.29)	27 (55.10)
Resection modality			6.140	0.013
EMR	15 (48.39)	37 (75.51)
ESD	16 (51.61)	12 (24.49)
Procedure duration (minutes)	45.29 ± 11.13	33.70 ± 10.57	4.681	0.000
Prophylactic clip deployment	25 (80.65)	22 (44.90)	10.010	0.002

EMR: Endoscopic mucosal resection; ESD: Endoscopic submucosal dissection; PAS: Prior abdominal surgery; NPAS: Non-prior abdominal surgery.

Post-procedural recovery indices

Patients with PAS exhibited significantly longer hospital stays, delayed resumption of oral intake, later ambulation, and higher post-procedural Visual Analog Scale pain scores than those without a surgical history (P < 0.05, Table 3).

Table 3 Comparison of postoperative recovery indicators between two patient groups, mean ± SD.

Variable	PAS (n = 31)	NPAS (n = 49)	t	P value
Length of stay (days)	8.19 ± 2.16	5.76 ± 1.78	5.472	0.000
Time to resumption of oral intake (days)	4.06 ± 1.35	2.69 ± 0.93	5.376	0.000
Time to mobilization (days)	5.38 ± 1.72	3.89 ± 1.35	4.319	0.000
Post-operative VAS score	5.24 ± 1.28	3.27 ± 1.06	7.467	0.000

VAS: Visual Analogue Scale; PAS: Prior abdominal surgery; NPAS: Non-prior abdominal surgery.

Post-operative complications

The incidence of post-polypectomy complications was significantly higher in the PAS group than in the no prior surgery group (P < 0.05, Table 4). Specifically, the surgical-history group demonstrated markedly higher frequencies of hemorrhage (19.35% vs 10.20%), perforation (12.90% vs 4.08%), and post-procedural pain (35.48% vs 16.33%), collectively indicating that PAS was an independent risk factor for endoscopic complications.

Table 4 Comparison of postoperative complication rates between the two patient groups, n (%).

Groups	n	Hemorrhage	Perforation	Pyrexia	Infection	Abdominal pain	Complications
PAS	31	6 (19.35)	4 (12.90)	3 (9.68)	4 (12.90)	8 (35.48)	25 (80.65)
NPAS	49	5 (10.20)	2 (4.08)	4 (8.16)	3 (6.12)	8 (16.33)	22 (44.90)
χ2							10.010
P value							0.002

PAS: Prior abdominal surgery; NPAS: Non-prior abdominal surgery.

Multivariable logistic regression of post-polypectomy complications

Using the occurrence of any complication as the binary dependent variable (1 = yes, 0 = no), we entered the following covariates into a multivariate logistic model: Arterial hypertension, hyperlipidemia, anticoagulant or antiplatelet therapy, PAS, polyp diameter, polyp multiplicity, ESD vs EMR resection modality, procedure duration, and the proportion of hemoclip deployment. All nine variables emerged as independent predictors of adverse events (P < 0.05; Table 5).

Table 5 Multivariate logistic regression analysis of postoperative complications following endoscopic polypectomy.

Variable	β	SE	Wald χ2	P value	OR	95%CI
Hypertension	1.049	0.475	4.888	0.027	2.856	1.126-7.240
Hyperlipidemia	1.332	0.550	5.865	0.015	3.790	1.289-11.139
Anticoagulant/antiplatelet therapy	1.427	0.515	7.678	0.006	4.167	1.518-11.434
Prior abdominal surgery	1.437	0.517	7.733	0.005	4.208	1.528-11.585
Polyp size	0.145	0.053	7.316	0.006	1.156	1.041-1.283
Polyp number	1.030	0.425	5.867	0.015	2.801	1.217-6.447
Resection modality	1.191	0.489	5.918	0.015	3.289	1.260-8.583
Procedure duration	0.104	0.028	13.896	0.000	1.110	1.051-1.172
Prophylactic clip deployment	1.632	0.538	9.210	0.002	5.114	1.782-14.671

OR: Odds ratio; CI: Confidence interval.

Predictive performance of individual risk factors

Receiver operating characteristic curves were constructed for each variable retained in the multivariate model. All nine factors demonstrated moderate discriminative capacity for postpolypectomy complications, with area under the curve values ranging from 0.63 to 0.81 (Figures 1 and 2; Table 6).

Figure 1 Receiver operating characteristic curves for the impact of various factors on the occurrence of postoperative complications in patients undergoing endoscopic polypectomy. ROC: Receiver operating characteristic.

Figure 2  Contribution of different variables to the overall quality of the postoperative complication prediction model.

Table 6 Assessment of the value of various factors in predicting postoperative complications in patients undergoing endoscopic polypectomy.

Variable	AUC	95%CI	Sensitivity (%)	Specificity (%)	Maximum Youden index
Hypertension	0.627	0.500-0.754	73.53	90.00	0.635
Hyperlipidemia	0.622	0.492-0.752	70.59	80.00	0.506
Anticoagulant/antiplatelet therapy	0.650	0.522-0.778	67.65	95.36	0.630
Prior abdominal surgery	0.663	0.541-0.784	77.42	55.10	0.325
Polyp size	0.670	0.547-0.792	41.94	85.71	0.277
Polyp number	0.633	0.506-0.759	70.91	48.98	0.200
Resection modality	0.636	0.508-0.763	51.61	75.51	0.271
Procedure duration	0.777	0.671-0.883	67.74	83.67	0.514
Prophylactic clip deployment	0.679	0.560-0.798	80.65	55.10	0.358

AUC: Area under the curve; CI: Confidence interval.

DISCUSSION

Postoperative complications, ranging from bleeding and perforation to infection and abdominal pain, are major clinical challenges in endoscopic resection. Although polyp characteristics and procedural factors are well-established risk factors, the specific effect of patient history remains poorly defined. Endoscopic resection is the first-line curative modality for colorectal adenomatous polyps; however, procedure-related morbidity remains high. This study specifically addressed the knowledge gap regarding the influence of PAS influences the complication profiles. Using a stratified cohort design, we demonstrated that PAS independently doubled the risk of adverse events after EMR or ESD. This finding underscores the need to integrate surgical history into perioperative risk algorithms.

Previous laparotomy precipitates extensive intraperitoneal adhesions that distort anatomical planes, tether and angulate bowel loops, and reduce mural compliance[9,10]. These alterations compress the luminal workspace, degrade endoscopic visualization, and impair the operator’s ability to delineate the lesion margins, thereby introducing procedural uncertainty[11,12]. Elevated mechanical tension within the adhesive bands predisposes patients to tissue tearing during submucosal dissection, amplifying the risk of perforation and intraprocedural hemorrhage[13].

Consistent with this pathomechanism, patients with PAS required a mean procedure time of 45.29 ± 11.13 minutes - significantly longer than the 33.70 ± 10.57 minutes recorded in surgically naive controls - and underwent ESD in 51.61% of cases vs 24.49% in the control group. The preferential use of ESD presumably reflects deeper submucosal invasion or fibrosis within the adhesive segments, requiring en bloc excision for curative resection. Protracted operative duration, in turn, increases cumulative mucosal trauma, anesthetic exposure and the probability of adverse events[14-16]. In addition to mechanical impediments, ischemia-reperfusion injury within adhesive fields disrupts microvascular perfusion, whereas post-operative fibrosis and aberrant angiogenesis impair subsequent wound repair[17,18]. These biological sequelae manifest as attenuated epithelial proliferation and disorganized granulation tissue formation, providing a plausible mechanistic basis for the observed increase in delayed bleeding and intraluminal sepsis[19].

The four-fold increase in hemoclip deployment observed in patients with a previous laparotomy reflects heightened operator awareness of bleeding propensity. Although clips provide immediate hemostasis, their retention in the ulcer bed may perpetuate local inflammation and impair granulation tissue formation, paradoxically predisposing patients to late-onset hemorrhage[20,21]. This finding highlights the need for adjunctive pharmacological or mechanical hemostatic strategies that do not impede mucosal healing[22]. Multivariable modelling corroborated established risk modifiers - arterial hypertension, hyperlipidemia, antithrombotic therapy, large polyp diameter, multiplicity, ESD modality, and prolonged procedure time - and quantified their individual discriminative power (area under the curve of 0.63-0.81)[23-25]. Hypertension and hyperlipidemia promote arteriosclerotic fragility, whereas antithrombotic agents impair primary hemostasis. Large or multiple polyps necessitate wider resection fields that transect submucosal arteries, which inherently carry a higher vascular injury risk; and every additional 10 minutes of operative time incrementally increases exposure-related trauma and operator fatigue, compounding error probability.

The limitations of our study include its single-center retrospective design, which may have introduced a selection bias. We did not stratify the analyses by type, number, or time elapsed since previous abdominal operations, nor did we quantify the adhesion burden using preprocedural imaging or intraoperative scoring systems. Future prospective studies integrating radiographic adhesion scoring are needed to validate these findings and refine the risk stratification.

CONCLUSION

PAS is an independent modifiable predictor of morbidity after colorectal EMR/ESD. Surgeons should flag these patients for intensive pre-operative work-up, consider prophylactic hemostatic adjuncts, schedule procedures under optimal anticoagulation windows, and institute enhanced post-procedural surveillance to mitigate complications and safeguard patient safety.