Management of colonic diverticular bleeding: A narrative review with literature search of current evidence and therapeutic strategies

Abstract

BACKGROUND

Colonic diverticular bleeding (CDB) is the most common cause of acute lower gastrointestinal bleeding in Western populations, accounting for 30%-65% of cases. Its management has evolved substantially over the past two decades, shifting from emergent surgery toward a structured stepwise approach that combines risk stratification, endoscopic hemostasis, transcatheter arterial embolization (TAE), and surgery reserved for refractory cases.

AIM

To synthesize current evidence on the epidemiology, diagnostic evaluation, risk stratification, endoscopic and radiologic management, surgical intervention, antithrombotic management, and recurrence prevention of CDB, and to provide a practical evidence-based framework for clinicians.

METHODS

We conducted a narrative review with systematic literature search of the literature using PubMed/MEDLINE, EMBASE, and the Cochrane Library from database inception through January 2026 in accordance with the preferred reporting items for systematic reviews and meta-analyses 2009 Statement guidelines. Search terms combined controlled vocabulary and free text for “colonic diverticular bleeding”, “lower gastrointestinal bleeding”, “endoscopic band ligation”, “transcatheter arterial embolization”, “Oakland score”, and related concepts. Eligible studies included randomized controlled trials, systematic reviews and meta-analyses, multicenter cohort studies, and major society guidelines. Two reviewers independently screened titles, abstracts, and full texts. Data on study design, interventions, outcomes (technical and clinical success, rebleeding, mortality, complications), and effect estimates were extracted. A narrative synthesis was performed given clinical and methodological heterogeneity.

RESULTS

Twenty-six studies met inclusion criteria. Non-steroidal anti-inflammatory drugs [odds ratio (OR) = 6.90, 95% confidence interval (CI): 3.86-12.35] and aspirin (OR = 2.84, 95%CI: 2.19-3.67) are the strongest pharmacologic risk factors. The Oakland score (≤ 8 points) predicts safe outpatient discharge with an area under the receiver operating characteristic curve of 0.87. Computed tomographic angiography (CTA) achieves 90% sensitivity and 92% specificity for active bleeding and is recommended first-line in hemodynamically unstable patients. Endoscopic band ligation (EBL) demonstrated lower early and late rebleeding rates compared with through-the-scope clipping (8% vs 19%, P = 0.012; and 9% vs 29%, P = 0.02, respectively) in meta-analytic data; however, these findings derive predominantly from Japanese retrospective cohorts, and prospective Western validation is lacking. Over-the-scope clips show promise (OR = 0.16 vs EBL for early rebleeding). The largest TAE meta-analysis (n = 3530 patients across 58 trials) reported 90.4% technical success, 73.5% clinical success, and 19.9% 30-day rebleeding. A restrictive transfusion strategy (hemoglobin threshold 7 g/dL; 8 g/dL in cardiovascular disease) is supported across major guidelines. Surgery is now required in fewer than 5% of contemporary cases.

CONCLUSION

Contemporary management of CDB rests on early risk stratification with the Oakland score, judicious use of CTA in unstable patients, colonoscopy with bowel preparation at the next non-urgent opportunity for stable patients, and preferential use of EBL over clipping when hemostasis is required, acknowledging the predominantly Japanese evidence base. Long-term recurrence approaches 40% at 4 years, mandating individualized antithrombotic review and selective elective colectomy in appropriate candidates.

Key Words: Colonic diverticular bleeding; Lower gastrointestinal bleeding; Endoscopic band ligation; Transcatheter arterial embolization; Oakland score; Computed tomographic angiography; Over-the-scope clip; Antithrombotic therapy; Risk stratification; Colonic diverticulosis

Core Tip: Colonic diverticular bleeding (CDB) is the leading cause of acute lower gastrointestinal bleeding in Western populations, yet management remains inconsistently applied. This review synthesizes current evidence across the full CDB care pathway from risk stratification using the Oakland score to endoscopic, radiologic, and surgical decision-making. Endoscopic band ligation outperforms clipping for hemostasis, transcatheter arterial embolization achieves high technical success, and early antithrombotic resumption reduces mortality without excess rebleeding. A practical, evidence-based framework is provided to guide clinicians toward more standardized, outcome-driven care.

INTRODUCTION

Colonic diverticular hemorrhage represents a significant clinical challenge in gastroenterology and is the leading cause of overt lower gastrointestinal bleeding (LGIB) in the United States, occurring in up to 15% of patients with diverticulosis with an incidence of approximately 0.5 per 1000 person-years. The annual incidence of hospitalization for acute LGIB in the United States is approximately 36 per 100000 population[1,2]. As the global population ages and the prevalence of diverticulosis increases with estimates suggesting that fewer than 20% of individuals have diverticulosis at age 40, rising to approximately 60% by age 60 years the burden of diverticular bleeding is expected to rise further[3-5].

Historically, the management of colonic diverticular bleeding (CDB) relied heavily on emergent colectomy, particularly in patients with hemodynamic instability or recurrent hemorrhage[6]. However, advances in endoscopic techniques, interventional radiology, and risk stratification tools have fundamentally transformed the therapeutic landscape. The development of endoscopic band ligation (EBL), through-the-scope (TTS) clipping, and super-selective transcatheter embolization has enabled a more targeted, minimally invasive approach to hemostasis[5]. Despite these advances, challenges remain in recurrence prevention and individualized treatment selection[3,5].

The purpose of this narrative review with systematic literature search is to provide a comprehensive synthesis of the current evidence on the management of CDB, organized by the sequential phases of clinical care: Initial assessment and resuscitation, diagnostic evaluation, endoscopic therapy, interventional radiology, surgical management, antithrombotic medication management, and long-term recurrence prevention. This review draws upon landmark multicenter studies including the CODE-BLUE J-study, randomized controlled trials (RCTs) on colonoscopy timing, meta-analyses of endoscopic and radiologic interventions, and the most recent society guidelines from the American College of Gastroenterology (ACG), the British Society of Gastroenterology (BSG), the European Society of Gastrointestinal Endoscopy (ESGE), the American College of Radiology, and the joint ACG-Society of Abdominal Radiology (SAR) consensus recommendations[5,7-14].

MATERIALS AND METHODS

Search strategy

A systematic literature search was conducted in PubMed/MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials from database inception through January 2026 in accordance with the preferred reporting items for systematic reviews and meta-analyses 2009 Statement guidelines. The search combined Medical Subject Headings and free-text terms for: (“Colonic diverticular” OR “diverticulosis”) AND (“bleed” OR “hemorrhag” OR “haemorrhag”) AND (“endoscopic band ligation” OR “endoscopic clipping” OR “over-the-scope clip” OR “transcatheter arterial embolization” OR “colectomy” OR “Oakland score” OR “SHA2PE” OR “NOBLADS” OR “computed tomography angiography” OR “colonoscopy timing” OR “antithrombotic” OR “anticoagulation” OR “non-steroidal anti-inflammatory”). Additional records were identified through citation tracking of included articles and review of position statements and clinical practice guidelines from major gastroenterology societies.

Eligibility criteria

Inclusion criteria comprised: (1) RCTs evaluating any intervention for CDB or acute LGIB; (2) Systematic reviews and meta-analyses; (3) Large prospective or retrospective multicenter cohort studies (≥ 500 patients) of diagnostic or therapeutic strategies in CDB; (4) Landmark single-center studies of seminal techniques [e.g., EBL, over-the-scope clip (OTSC)]; (5) Major society clinical practice guidelines; and (6) Population-based incidence and risk factor studies. Exclusion criteria comprised: (1) Case reports and case series of fewer than 50 patients; (2) Studies in pediatric populations exclusively; (3) Animal or in vitro studies; (4) Non-English publications without an English abstract; (5) Editorials, commentaries, and conference abstracts without peer-reviewed full text; and (6) Studies addressing diverticular bleeding in the small bowel or stomach exclusively.

Study selection and data extraction

Two reviewers independently screened titles and abstracts, with full-text review of potentially eligible articles. Disagreements were resolved by discussion and consensus, with a third reviewer consulted as needed. Data extracted from each included study comprised: Study design and setting, sample size, patient characteristics, interventions, comparators, primary and secondary outcomes (technical and clinical success, early and late rebleeding rates, complications, mortality, hospital length of stay), and effect estimates with 95% confidence intervals (CIs) where available. Given substantial clinical and methodological heterogeneity across studies, a narrative synthesis approach was used rather than meta-analytic pooling. The flow of study selection is summarized in Figure 1.

Figure 1 Preferred reporting items for systematic reviews and meta-analyses 2020 flow diagram illustrating the systematic literature search and study selection process. Records were identified from PubMed/MEDLINE (n = 842), EMBASE (n = 612), and the Cochrane Library (n = 69), with 27 additional records identified through citation searching and society guideline review. After removal of duplicates and sequential screening, 26 studies met predefined inclusion criteria and were included in the qualitative synthesis. pts: Patients; RCT: Randomized controlled trial.

Quality assessment

Given the clinical and methodological heterogeneity of the included literature spanning RCTs, large multicenter retrospective cohorts, meta-analyses of observational studies, and society guidelines a formal risk-of-bias assessment table and GRADE evidence summary were not applied uniformly across all included studies. Study quality is instead characterized narratively throughout the “results and discussion”, with explicit notation of study design limitations, evidence level, and certainty of evidence where relevant. This approach is consistent with the methodology of a narrative review with systematic literature search, and is disclosed accordingly in the article title and methods.

RESULTS

Epidemiology and pathophysiology

Prevalence and incidence: Diverticulosis is exceedingly common in Western populations, with prevalence increasing markedly with age. A comprehensive JAMA review by Brown et al[4] reported that the prevalence of diverticulosis increases from fewer than 20% at age 40 years to approximately 60% at age 60 years, underscoring the age-dependent burden of this condition. Approximately 3%-15% of patients with diverticulosis will develop diverticular hemorrhage during their lifetime[1,5]. In a long-term Japanese cohort study of patients with colonoscopy-confirmed asymptomatic diverticulosis, Niikura et al[15] reported that the cumulative risks of new-onset diverticular hemorrhage were 0.21% at 1 year, 2.2% at 5 years, and 9.5% at 10 years, with an overall incidence rate of 0.46 per 1000 patient-years. Variables predictive of new-onset diverticular bleeding included age older than 70 years and the presence of bilateral (right and left-sided) diverticulosis[15].

Geographic and population considerations: An important geographic distinction exists in the distribution of diverticular disease. In Western populations, left-sided diverticulosis predominates, with over 90% of cases involving the sigmoid colon[3]. In East Asian populations, right-sided (cecal and ascending) diverticulosis is considerably more prevalent[5,15]. This anatomical distinction has critical implications for interpreting the evidence base and applying it across different healthcare settings. The preponderance of high-quality endoscopic outcome data particularly for EBL derives from Japanese multicenter cohorts in which right-sided diverticulosis predominates. Right-sided diverticula are typically larger, have a wider neck, and may be more amenable to certain endoscopic techniques such as band ligation compared with the narrow-necked sigmoid diverticula common in Western patients. Differences in bowel preparation logistics, colonoscopy access angles, and bleeding behavior between right-sided and left-sided diverticula may further influence diagnostic yield and endoscopic hemostasis outcomes. Accordingly, the findings of Japanese studies on EBL, OTSC, and colonoscopy timing should be interpreted with caution when applied to Western populations, where prospective multicenter trials are still lacking.

Pathogenesis: Bleeding occurs when the vasa recta arteries that penetrate the colonic wall at the site of diverticulum formation rupture into the intestinal lumen[3,16]. This vascular fragility is compounded by mechanical stress on the thinned colonic wall and patient-related factors such as advanced age, hypertension, atherosclerosis, and the use of antithrombotic agents[2,3,5]. Aspirin and non-aspirin non-steroidal anti-inflammatory drugs (NSAIDs) have both been shown to increase the risk of diverticular bleeding, believed to be due to mucosal ulceration within the dome or neck of a diverticulum, subsequently contributing to arterial rupture into the colonic lumen[5]. Unlike diverticulitis, dietary factors have not been clearly implicated in increasing the risk of diverticular hemorrhage[5].

Risk factors: Multiple studies have identified pharmacologic and clinical risk factors for the onset of CDB. In a meta-analysis of case-control and cohort studies by Longo et al[17], NSAIDs and acetylsalicylic acid use were associated with an increased risk of diverticular bleeding [odds ratio (OR) = 6.90, 95%CI: 3.86-12.35 for NSAIDs; OR = 2.84, 95%CI: 2.19-3.67 for aspirin]. In an age- and sex-matched case-control study by Taki et al[2], bilateral colonic diverticulosis, non-selective NSAIDs, low-dose aspirin, and anticoagulants were significant risk factors for the onset of CDB on multivariate analysis, whereas selective cyclooxygenase-2 inhibitors were not associated with increased risk. The incidence of bleeding in direct oral anticoagulant (DOAC) and warfarin users did not differ significantly between the two groups[2].

Data on P2Y12 inhibitors are more limited; however, studies have shown an increased risk of diverticular bleeding and all-cause LGIB with non-aspirin antiplatelets such as clopidogrel[5]. Vitamin K antagonists are known to increase the risk of all-cause gastrointestinal bleeding approximately three-fold compared with placebo[5]. Among DOACs, rivaroxaban has a higher risk of LGIB compared with apixaban [hazard ratio (HR) = 1.53, 95%CI: 1.00-2.33][5]. Combinations of antiplatelet and anticoagulant drugs are associated with an increased risk of LGIB with an overall incidence of 70 per 1000 person-years[5] (Table 1).

Table 1 Risk factors for colonic diverticular bleeding.

Category	Risk factors	Effect size/notes
Patient/anatomic	Age > 70 years	HR: 1.5-2.0; independent predictor of new-onset CDB
	Bilateral diverticulosis (right + left)	Key predictor of onset; higher cumulative incidence
	Hypertension/atherosclerosis	Contributes to vascular fragility of vasa recta
NSAIDs	Non-selective NSAIDs	OR = 6.90 (95%CI: 3.86-12.35); Longo et al[17] meta-analysis
	Aspirin (any dose)	OR = 2.84 (95%CI: 2.19-3.67)
	COX-2 selective inhibitors	Not associated with increased CDB risk
Antiplatelets	P2Y12 inhibitors (clopidogrel)	Increased LGIB risk; limited CDB-specific data
Anticoagulants	Vitamin K antagonists (warfarin)	Approximately 3-fold increased GI bleeding vs placebo
	Rivaroxaban vs apixaban	HR = 1.53 (95%CI: 1.00-2.33) higher LGIB risk with rivaroxaban
	DOACs vs antiplatelets	DOACs not shown to increase new-onset CDB
Combination	Dual antiplatelet + anticoagulant	Incidence approximately 70/1000 person-years
	Triple antithrombotic therapy	Highest risk combination; avoid when possible

GI: Gastrointestinal; NSAIDs: Non-steroidal anti-inflammatory drugs; HR: Hazard ratio; CDB: Colonic diverticular bleeding; COX-2: Cyclooxygenase-2; DOAC: Direct oral anticoagulant; OR: Odds ratio; CI: Confidence interval; LGIB: Lower gastrointestinal bleeding.

Clinical presentation and initial assessment

Clinical presentation: Diverticular bleeding typically presents with painless hematochezia, frequently large-volume in nature, because it is an arterial bleed occurring from the neck or dome of a diverticulum[3,5]. For most patients, bleeding is relatively minor and self-limited, with spontaneous cessation occurring in over 80% of cases[6]. However, in older individuals and those on antithrombotic therapy, bleeding can be brisk and hemodynamically significant[3]. Importantly, hematochezia in the context of hemodynamic instability may represent a brisk upper gastrointestinal bleeding event; therefore, upper endoscopy should be considered in such patients, as approximately 10%-15% of patients presenting with hematochezia will have an upper gastrointestinal source[3,18].

Risk stratification: Initial risk assessment is increasingly important to identify patients at low or high risk of requiring hospital-based intervention. The BSG recommends initial hemodynamic stratification using the shock index (SI), calculated by dividing heart rate by systolic blood pressure[12]. Patients deemed unstable (SI > 1) should proceed to computed tomographic angiography (CTA), while stable patients can be further stratified using dedicated LGIB risk scores[12].

The Oakland score, derived and validated from a United Kingdom audit of patients admitted with LGIB, incorporates age, sex, history of LGIB, digital rectal examination findings, heart rate, systolic blood pressure, and hemoglobin level[5]. A score of ≤ 8 predicts a 95% probability of safe discharge[5] (Table 2). The Oakland score was externally validated using data from a network of 140 hospitals across the United States; the area under the receiver-operating characteristic curve (AUROC) for safe discharge was 0.87[5]. In a systematic review and meta-analysis by Almaghrabi et al[19] comparing four LGIB risk scores across 9 studies and 12 independent cohorts, the Oakland score was the most discriminative for predicting safe discharge (AUROC = 0.86), major bleeding (AUROC = 0.93), and need for transfusion (AUROC = 0.99).

Table 2 Oakland score variables and discharge threshold.

Variable	Category	Points	Interpretation
Age	< 40 years	0
	40-69 years	1
	≥ 70 years	2
Sex	Female	0
	Male	1
Prior LGIB	No	0
	Yes	1
DRE	No blood	0
	Blood	1
Heart rate	≤ 70 bpm	0
	71-89 bpm	1
	90-109 bpm	2
	≥ 110 bpm	3
Systolic BP	≥ 160 mmHg	0
	130-159 mmHg	1
	100-129 mmHg	2
	50-99 mmHg	3
Hemoglobin	≥ 16.0 g/dL	0
	13.0-15.9 g/dL	1
	11.0-12.9 g/dL	2
	9.0-10.9 g/dL	3
	6.0-8.9 g/dL	4
	< 6.0 g/dL	5
Score ≤ 8	≥ 95% probability of safe discharge; ESGE and BSG recommend outpatient evaluation

A score ≤ 8 predicts a ≥ 95% probability of safe discharge (area under the receiver-operating characteristic curve = 0.87). LGIB: Lower gastrointestinal bleeding; DRE: Digital rectal examination; BP: Blood pressure; ESGE: European Society of Gastrointestinal Endoscopy; BSG: British Society of Gastroenterology.

Comparison with other risk scores and practical considerations: The SHA2PE score, derived by Hreinsson et al[20] from a retrospective cohort of 581 patients, incorporates systolic blood pressure ≥ 100 mmHg, hemoglobin > 12 g/dL, non-use of antiplatelets and anticoagulants, and pulse < 100 bpm as low-risk criteria. However, an external validation study demonstrated that the SHA2PE score misclassified 14 patients who required hospital-based intervention as low-risk, raising concerns about its safety as a sole discharge criterion[5]. The NOBLADS score was developed specifically for predicting severe CDB in a Japanese cohort and incorporates NSAIDs, non-antithrombotic use, blood pressure, albumin level, and history of diverticular bleeding; however, its external validation outside Japanese centers remains limited.

Several important limitations of the Oakland score should be recognized when applying it outside its original validation cohort. First, it was derived primarily from a United Kingdom population with predominantly left-sided diverticular disease; its performance in East Asian cohorts with right-sided predominance has not been systematically validated. Second, the score does not account for the specific indication for antithrombotic therapy or the degree of supratherapeutic anticoagulation, both of which substantially affect the clinical trajectory. Third, the score was validated for prediction of “safe discharge” as a composite endpoint and may not reliably predict which patients will require endoscopic or radiologic intervention. Accordingly, the Oakland score is best used as a complementary tool alongside clinical judgment rather than as a standalone discharge decision instrument. In patients with a score ≤ 8 who nonetheless have ongoing hemodynamic instability, rapid drop in hemoglobin, high-dose antithrombotic use, or significant comorbidity burden, inpatient observation and colonoscopy should be strongly considered regardless of the score.

Resuscitation and transfusion: Patients who are hemodynamically unstable should receive intravenous fluid resuscitation with crystalloids with the goal of normalization of blood pressure and heart rate before any diagnostic or therapeutic intervention[5].

It is important to emphasize that transfusion recommendations in LGIB are largely extrapolated from upper gastrointestinal bleeding randomized trials; LGIB specific randomized evidence remains extremely limited. In the landmark RCT by Villanueva et al[21], of 921 patients with acute upper gastrointestinal bleeding, a restrictive transfusion strategy (hemoglobin threshold of 7 g/dL) compared with a liberal strategy was associated with significantly improved survival at 6 weeks[21,22]. This mortality benefit has since been highlighted as a paradigm shift in transfusion practice for gastrointestinal bleeding[23]. In a meta-analysis of 5 RCTs comprising 1965 patients, a restrictive transfusion strategy was associated with a lower risk of all-cause mortality [relative risk (RR) = 0.65, 95%CI: 0.44-0.97] and rebleeding (RR = 0.58, 95%CI: 0.40-0.84)[5].

Data are far more limited in LGIB specifically. In a post hoc analysis of the United Kingdom audit of LGIB, no significant difference was seen between liberal and restrictive red blood cells transfusion strategies for odds of rebleeding (OR = 0.89, 95%CI: 0.6-1.22) or in-hospital mortality (OR = 0.54, 95%CI: 0.3-1.1)[5]. Given that this analysis was post hoc, underpowered, and derived from observational registry data, it should be interpreted cautiously. Dedicated LGIB-specific transfusion threshold RCTs are needed before the evidence level for this recommendation can be upgraded beyond low-to-moderate certainty. Current guidelines recommend a restrictive transfusion strategy for hemodynamically stable patients with LGIB, with a threshold of 7 g/dL[5]. A higher threshold of 8 g/dL is considered reasonable for patients with underlying cardiovascular disease (Table 3)[21,24-27].

Table 3 Transfusion strategy: Evidence base and guideline recommendations.

Domain	Study/threshold	Key finding/recommendation
Hb threshold	7 g/dL (restrictive)	Standard recommendation for hemodynamically stable LGIB
	8 g/dL	Reasonable in significant cardiovascular disease
	Before Hb < 7 if hypotensive	Dilutional drop post-resuscitation underestimates deficit; transfuse early
UGIB RCT evidence	Villanueva et al[21] (n = 921)	Restrictive: Improved 6-week survival, lower rebleeding vs liberal
	Meta-analysis 5 RCTs (n = 1965)	Restrictive: RR mortality 0.65 (0.44-0.97); Rebleeding RR 0.58 (0.40-0.84)
	Teutsch et al[24] meta-analysis (7 RCTs)	Restrictive did not increase in-hospital or 30-day mortality
LGIB data	United Kingdom audit post hoc analysis	No significant difference in rebleeding or mortality between strategies; post hoc, underpowered
Guidelines	AABB 2023 International Guidelines[27]	Restrictive strategy recommended across clinical settings
	ACG, ESGE, BSG	Hb 7 g/dL stable patients; 8 g/dL in CVD; early transfusion if hemodynamically unstable

Evidence for lower gastrointestinal bleeding is largely extrapolated from upper gastrointestinal bleeding trials; dedicated lower gastrointestinal bleeding transfusion randomized controlled trials are needed. Hb: Hemoglobin; UGIB: Upper gastrointestinal bleeding; RCT: Randomized controlled trial; LGIB: Lower gastrointestinal bleeding; AABB: American Association of Blood Banks; ACG: American College of Gastroenterology; ESGE: European Society of Gastrointestinal Endoscopy; BSG: British Society of Gastroenterology; CVD: Cardiovascular disease; RR: Relative risk.

Diagnostic evaluation

The overall management algorithm for acute CDB is summarized in Figure 2. Individual components are discussed in detail in the sections that follow.

Figure 2 Stepwise clinical management algorithm for acute colonic diverticular bleeding. Initial resuscitation and hemodynamic assessment using the shock index (heart rate divided by systolic blood pressure) guide the primary diagnostic pathway. Hemodynamically unstable patients (shock index > 1) should proceed directly to computed tomographic angiography; a positive result prompts interventional radiology consultation and transcatheter arterial embolization within 60-90 minutes. Hemodynamically stable patients are risk-stratified using the Oakland score; a score ≤ 8 predicts a ≥ 95% probability of safe discharge. For low-risk patients discharged early, outpatient colonoscopy within 14 days is recommended; however, colonoscopy is not mandatory in all low-risk outpatients and the decision should be individualized based on clinical findings and the likelihood of identifying a treatable source. For admitted patients, colonoscopy with bowel preparation at the next available non-urgent opportunity remains the cornerstone of diagnosis and treatment. When stigmata of recent hemorrhage or active bleeding are identified, endoscopic hemostasis should be performed with endoscopic band ligation preferred over through-the-scope clipping, acknowledging that this recommendation is based predominantly on Japanese retrospective cohort data. Transcatheter arterial embolization and surgery are reserved for refractory or recurrent cases. All patients require individualized recurrence prevention planning at discharge. GI: Gastrointestinal; pts: Patients; CBC: Complete blood count; BMP: Basic metabolic panel; HR: Heart rate; SBP: Systolic blood pressure; SI: Shock index; CT: Computed tomography; STAT: Immediately; CTA: Computed tomographic angiography; TAE: Transcatheter arterial embolization; IR: Interventional radiology; SRH: Stigmata of recent hemorrhage; CDB: Colonic diverticular bleeding; EBL: Endoscopic band ligation; TTS: Through-the-scope; OTSC: Over-the-scope clip; NSAIDs: Non-steroidal anti-inflammatory drugs; f/u: Follow up.

Colonoscopy: Colonoscopy is considered the diagnostic test of choice for patients admitted with LGIB, given its ability to identify a specific etiology, obtain mucosal tissue samples, and provide therapeutic options for bleeding control[5,18]. The diagnostic yield of colonoscopy varies widely, ranging from 68% to 79% depending on the cohort and timing of the procedure[5]. Adequate bowel preparation is essential for endoscopic visualization, diagnosis, and treatment, and typically requires at least 4 L of polyethylene glycol solution administered over approximately 4 hours for urgent colonoscopy[18].

Timing of colonoscopy-early vs elective: The optimal timing of inpatient colonoscopy for LGIB has been one of the most debated topics in the field. Niikura et al[9] conducted a multicenter RCT at 15 hospitals in Japan of 170 patients comparing early colonoscopy (within 24 hours) to elective colonoscopy (24-96 hours). Early colonoscopy did not increase the rate of detecting stigmata of recent hemorrhage (SRH). Notably, the early colonoscopy group had a numerically higher 30-day rebleeding rate (15.3% vs 6.7%), although this difference did not reach statistical significance. There were no significant differences in successful endoscopic treatment, transfusion requirements, length of stay, thrombotic events, or mortality[9]. In a second RCT, early colonoscopy was associated with reduced hospital length of stay (2 days vs 3 days) but an increased risk of recurrent bleeding (13% vs 3%) and hospital readmission (11% vs 3%) compared with standard colonoscopy[5]. Based on these data, the updated ACG guideline concludes that urgent colonoscopy should likely only be performed in select, high-risk patients with a high pretest probability of detecting SRH and performing endoscopic intervention[5,28,29].

CTA: CTA has an increasing role in the diagnosis and management of LGIB because of its ability to rapidly obtain images without the need for bowel preparation[5,14]. In a meta-analysis of 14 observational studies, the sensitivity and specificity of CTA in the diagnosis of LGIB were 90% and 92%, respectively[5]. CTA can detect bleeding rates as low as 0.3 mL/minute and can precisely localize the source of arterial and venous bleeding as well as delineate vascular anatomy before embolization[13]. The ACG-SAR consensus recommendations state that CTA should be performed as the first diagnostic study in hemodynamically unstable patients[14]. For patients with a positive CTA, prompt angiography is critical; transcatheter angiography performed within 90 minutes of a positive CTA was 9 times more likely to detect extravasation[5].

Endoscopic management

Definitions and SRH: To ensure consistent interpretation across studies, this review uses standardized definitions of CDB-specific terms, summarized in Table 4. Endoscopic SRH categories are summarized in Table 5.

Table 4 Key definitions used throughout this narrative review.

Term	Definition
Definitive CDB	Colonic diverticular bleeding with endoscopic identification of stigmata of recent hemorrhage (SRH) at a diverticulum, or active bleeding observed during colonoscopy or angiography
Presumptive CDB	Suspected diverticular hemorrhage in a patient with diverticulosis identified at colonoscopy but without SRH or active bleeding at the diverticulum, after exclusion of other LGIB sources
Stigmata of recent hemorrhage (SRH)	Endoscopic findings indicative of recent bleeding, including active spurting/oozing bleeding, non-bleeding visible vessel, adherent clot, or flat pigmented spot at the diverticulum
Successful endoscopic hemostasis	Cessation of active bleeding and absence of recurrent bleeding during the index colonoscopy, with no clinical or laboratory evidence of ongoing hemorrhage
Early rebleeding	Recurrent hematochezia or hemodynamic instability within 30 days of initial hemostasis, requiring repeat intervention or hospitalization
Late rebleeding	Recurrent diverticular hemorrhage occurring beyond 30 days of the index episode, typically reported at 90 days, 365 days, or longer cumulative intervals
Technical success (TAE)	Successful catheterization of the bleeding vessel with deployment of embolic material and angiographic cessation of extravasation
Clinical success (TAE)	Absence of recurrent bleeding within 30 days following the procedure, with no requirement for additional intervention
Hemodynamic instability	Shock index (HR/SBP) > 1, persistent hypotension despite resuscitation, or ongoing transfusion requirement

CDB: Colonic diverticular bleeding; SRH: Stigmata of recent hemorrhage; TAE: Transcatheter arterial embolization; LGIB: Lower gastrointestinal bleeding; HR: Heart rate; SBP: Systolic blood pressure.

Table 5 Classification of stigmata of recent hemorrhage in colonic diverticular bleeding.

Stigmata category	Endoscopic appearance	Rebleeding risk	Management implications
Active spurting bleeding	Pulsatile arterial bleeding from the diverticular neck or dome	Highest	Active bleeding present; mandatory endoscopic hemostasis
Active oozing bleeding	Non-pulsatile slow bleeding from the diverticulum	High	Endoscopic hemostasis indicated
Non-bleeding visible vessel	Pigmented protuberant lesion within the diverticulum without active bleeding	High	Endoscopic hemostasis recommended
Adherent clot	Clot tightly adherent to the diverticulum despite irrigation	Moderate	Consider removal and treatment; controversial
Flat pigmented spot	Black or red discoloration without protrusion, no active bleeding or clot	Lower	No proven benefit from endoscopic therapy; observation appropriate
Clean-based diverticulum (no SRH)	Diverticulum without stigmata in a patient with presumptive CDB	Lowest	Conservative management; consider further investigation

This classification was synthesized by the authors from criteria and outcome data reported in the CODE-BLUE J-study[7] and current society guidelines[5]; it is an original synthesis, not a reproduction of a single previously published table, and therefore does not require copyright permission. SRH: Stigmata of recent hemorrhage; CDB: Colonic diverticular bleeding.

Rationale for endoscopic treatment of SRH: When colonoscopy identifies active bleeding (spurting or oozing) or SRH (non-bleeding visible vessel or adherent clot), endoscopic treatment should be performed to reduce the risk of ongoing and recurrent bleeding[5,12]. The landmark CODE-BLUE J-study, a large multicenter Japanese cohort of 5823 patients with CDB who underwent colonoscopy at 49 hospitals, demonstrated that treating definitive CDB endoscopically was most effective in reducing recurrent bleeding over the short and long term compared with conservative management[8]. Both early and late recurrent bleeding rates were significantly lower in patients with definitive CDB treated endoscopically than in those with presumptive CDB treated conservatively (30-day: 19.6% vs 26.0%, P < 0.001; 365-day: 33.7% vs 41.6%, P < 0.001)[8].

TTS clips: TTS clips are the most widely used endoscopic hemostatic tool for CDB worldwide. Direct clipping of the bleeding vessel at the diverticular neck or dome achieves initial hemostasis rates of 95%-100%[12]. The technique of clip placement significantly affects outcomes: Direct clipping of a vessel is associated with a reduced risk of early and late rebleeding compared with indirect clipping techniques[5,12]. TTS clipping has the largest overall patient volume in the literature and remains the most widely used technique globally, supported by multiple meta-analyses and the CODE-BLUE J-study[7,30].

EBL: EBL has emerged as a highly effective technique for CDB, with initial hemostasis rates of 93%-100% and a low risk of complications (0%-2%)[12]. In a meta-analysis of 16 studies and 790 patients by Nagata et al[31], the pooled frequency of early rebleeding was significantly lower for EBL than clipping (8% vs 19%; P = 0.012), as was late rebleeding (9% vs 29%; P = 0.024)[31]. In the CODE-BLUE J-study, Kobayashi et al[7] analyzed 1679 patients treated with EBL (n = 638) or clipping (n = 1041); EBL was independently associated with reduced early rebleeding (adjusted OR = 0.46; P < 0.001) and late rebleeding (adjusted OR = 0.62; P < 0.001), reduced need for interventional radiology (adjusted OR = 0.37; P = 0.006), and shorter hospital stay (adjusted OR = 0.35; P < 0.001)[7].

Current evidence suggests that EBL may reduce early and late rebleeding compared with clipping; however, these findings derive predominantly from retrospective Japanese cohorts in which right-sided diverticulosis predominates. The superiority of EBL over TTS clipping has not yet been firmly established in Western populations, where left-sided sigmoid diverticulosis predominates and the anatomical characteristics of diverticula may differ substantially. Prospective multicenter studies especially in Western populations are needed before strong recommendations can be made. The most recent comprehensive meta-analysis by Meine et al[30] of 21 studies and 3161 participants confirmed that TTS clipping was consistently associated with worse rebleeding outcomes compared with other modalities; however, no RCTs have directly compared EBL and TTS clipping in a randomized fashion.

OTSCs: OTSC have emerged as another therapeutic option for CDB. In a multicenter retrospective propensity score-matched cohort analysis by Ohashi et al[32] comparing 68 patients treated with OTSC and 68 treated with EBL, OTSC was associated with significantly lower early recurrent bleeding rates (OR = 0.16; 95%CI: 0.02-0.77; P = 0.017) and shorter length of hospital stay (P = 0.026) compared with EBL. No significant differences in initial hemostasis, requirements for interventional radiology or surgery, or adverse events were observed[32]. These data suggest that OTSC may represent a promising alternative to EBL; however, the evidence base for OTSC remains very limited (predominantly single-center case series and one propensity-matched retrospective cohort study), and no RCTs have been conducted. OTSC should be considered an emerging modality and its routine adoption over EBL is not yet supported by the current evidence.

Endoscopic detachable snare ligation and topical hemostatic agents: Endoscopic detachable snare ligation (EDSL) is an additional ligation technique for CDB. Yamauchi et al[33] analyzed 77 EDSL and 441 EBL cases from the CODE-BLUE J cohort using propensity score matching and found no significant differences in initial hemostasis, 30-day recurrent bleeding, need for interventional radiology or surgery, mortality, or adverse events between EDSL and EBL. Sigmoid colon involvement was an independent risk factor for 30-day recurrent bleeding after EDSL (OR = 1.87; P = 0.042).

Topical hemostatic agents represent an emerging adjunctive modality for CDB. In a 2025 ESGE Technical and Technology Review, Gralnek et al[34] reported that TC-325 (hemospray) achieved primary hemostasis in 85%-98.5% of upper and lower gastrointestinal bleeding cases, with a pooled hemostasis rate of 93.1% and rebleeding rate of 8.9%. Nexpowder reduced rebleeding in lower gastrointestinal bleeds to 5.5% at 28 days. These agents are particularly useful as bridge therapy in diffuse or malignancy-related bleeding. In a feasibility study by Ng et al[35], hemospray achieved 100% immediate hemostasis in 10 patients with severe diverticular bleeding with no rebleeding and no procedure-related adverse events at median follow-up of 9.5 months. Topical hemostatic powders may be considered as an adjunct or bridge to definitive therapy when standard endoscopic hemostasis is technically challenging.

No RCTs exist directly comparing any of these modalities for CDB. All evidence derives from observational studies, retrospective cohorts, and meta-analyses of non-randomized data. EBL data come predominantly from Japanese cohorts with right-sided diverticulosis. TTS clipping remains the most widely used technique globally. OTSC is an emerging modality with very limited evidence. Overall evidence quality: Moderate for EBL and TTS clipping; very low for OTSC (Table 6).

Table 6 Comparison of endoscopic hemostatic modalities for colonic diverticular bleeding (revised).

Outcome	EBL	TTS clip (direct)	OTSC	Evidence base and quality
Initial hemostasis	93%-100%	95%-100%	Approximately 95%-100%	All modalities equivalent; pooled 98.9% (Meine et al[30], 21 studies, 3161 pts)
Early rebleeding (≤ 30 days)	8%	19%	OR 0.16 vs EBL (PSM cohort, n = 303)	EBL vs TTS: Nagata et al[31], 16 studies/790 pts, P = 0.012; CODE-BLUE J[7]: 1679 pts, aOR = 0.46; OTSC: Ohashi et al[32]
Late rebleeding (≤ 1 year)	9%	29%	Not yet reported	EBL vs TTS: Nagata et al[31], P = 0.024; CODE-BLUE J[7]: AOR = 0.62
Need for TAE/surgery	1%	2% (P = 0.031 vs EBL)	0% (single-center, n = 36)	CODE-BLUE J[7]: AOR = 0.37, P = 0.006; Meine et al[30]
Adverse events	Perforation 0.31%; diverticulitis 0.16%	Perforation 0%; diverticulitis 0.19%	0% (limited data)	CODE-BLUE J[7]: 1679 pts; OTSC: Very limited n
Length of stay	Shorter (aOR = 0.35 vs clip, P < 0.001)	Reference	Shorter vs EBL (P = 0.026)	CODE-BLUE J[7]; Ohashi et al[32]
Number of supporting studies	2 meta-analyses + CODE-BLUE J (638 EBL pts)[5,7,35]	Largest volume: CODE-BLUE J 1041 clip pts[7]; all meta-analyses	1 PSM cohort (n = 303)[32]; 1 single-center (n = 36); case reports
Overall quality of evidence	Moderate: Meta-analyses of observational data + large multicenter retrospective cohort; No RCTs; Predominantly Japanese data	Moderate: Same evidence base; Globally established technique	Very low: 1 small PSM retrospective study; No RCTs

No randomized controlled trials exist comparing these modalities. All evidence derives from observational studies and retrospective cohorts. Endoscopic band ligation data derive predominantly from Japanese cohorts with right-sided diverticulosis predominance; generalizability to Western populations with left-sided disease requires prospective validation. EBL: Endoscopic band ligation; TTS: Through-the-scope; OTSC: Over-the-scope clip; TAE: Transcatheter arterial embolization; aOR: Adjusted odds ratio; pts: Patients; RCT: Randomized controlled trial; PSM: Propensity score matched.

Transcatheter arterial embolization

Technique and efficacy: Transcatheter arterial embolization (TAE) has emerged as an effective salvage approach when endoscopic treatment fails and as a potential first-line treatment in patients with hemodynamically significant bleeding and extravasation on CTA[5,12]. Super-selective embolization using a microcatheter allows targeted embolization of a single vasa recta. Embolic agents include microcoils, N-butyl cyanoacrylate, gelatin sponge particles, and ethylene-vinyl alcohol copolymer[5,12]. In the largest meta-analysis to date by Sun et al[11] of 58 clinical trials involving 3530 patients, the pooled technical success rate was 90.4%, clinical success 73.5%, 30-day rebleeding 19.9%, major complications 7.8%, and 30-day mortality 9.6% (Table 7). Super selective embolization achieved higher technical success (93.4% vs 85.3%) and lower rebleeding rates (16.9% vs 26.6%) compared with non-super selective embolization[11].

Table 7 Pooled outcomes from meta-analyses of transcatheter arterial embolization for lower gastrointestinal bleeding.

Study/agent	Technical success	Clinical success	30-day rebleeding	30-day mortality	Notes
NBCA embolization (SR/MA, n = 243)	98.8%		15.7%	12.7%	Highest technical success of agent subgroups
Matsumoto et al[10] (NBCA or coils)	91.4%	76.5%	17.1%	11.4%	Major complication 10.0%; LGIB subgroup
Sun et al[11] (n = 58 trials, 3530)	90.4%	73.5%	19.9%	9.6%	Largest meta-analysis to date
Superselective TAE	93.4%		16.9%		vs non-super selective: TS 93.4% vs 85.3%; rebleed 16.9% vs 26.6%
Kitamura et al[36] (n = 132)	72% CTA positive	67% TAE performed	3.4% vs 30% (P < 0.001)	0%	Right-sided TAE particularly effective; no deaths

NBCA: N-butyl cyanoacrylate; SR/MA: Systematic review/meta-analysis; TAE: Transcatheter arterial embolization; CTA: Computed tomographic angiography; LGIB: Lower gastrointestinal bleeding; TS: Technical success.

A recent single-center study by Kitamura et al[36] evaluated TAE outcomes in 132 patients with contrast-enhanced computed tomography-confirmed diverticular bleeding. Active bleeding was detected on angiography in 72%, and 67% underwent TAE. The rebleeding rate was significantly lower in embolized patients (3.4% vs 30%; P < 0.001). Importantly, right-sided colonic bleeding had higher overall rebleeding rates than left-sided bleeding (16% vs 2.6%; P = 0.039), but TAE was particularly effective in right-sided cases (3.0% vs 46% rebleeding in those not treated; P < 0.001). No serious adverse events or in-hospital deaths occurred, suggesting that TAE is both safe and effective across anatomical distributions of diverticular disease.

Surgical management

Surgery for CDB is reserved for patients with refractory hemorrhage who have failed or are not candidates for endoscopic and radiologic hemostasis, as well as those with recurrent diverticular hemorrhage who are at high risk of future episodes[5,6]. Given the availability of less invasive alternatives, surgical intervention now accounts for fewer than 5% of CDB cases in contemporary series[5]. When the bleeding source is confirmed by CTA or colonoscopy, segmental colectomy is preferred; total abdominal colectomy is reserved for unlocalized bleeding, with 0% rebleeding but substantially higher rates of diarrhea and incontinence (Table 8)[5,6].

Table 8 Indications for surgical intervention in colonic diverticular bleeding and operative strategy selection.

Surgical indication/operative strategy	Clinical context/selection principle
Hemodynamic instability refractory to resuscitation	Requires massive transfusion; endoscopic/IR hemostasis not feasible or failed
Transfusion > 10 units pRBC with identified colonic source	Ongoing hemorrhage despite non-operative management
Failed endoscopic and TAE hemostasis	Salvage colectomy; segment identification by intraoperative colonoscopy if feasible
Recurrent life-threatening hemorrhage (localized source)	Elective laparoscopic segmental colectomy preferred
Perforation or surgical complication	Emergent laparotomy; bowel viability assessment required
Segmental colectomy	Preferred when source confirmed by CTA or colonoscopy; lower morbidity, better functional outcome
Total abdominal colectomy (TAC)	Reserved for unlocalized bleeding; 0% rebleeding but significant diarrhea/incontinence rates

pRBC: Packed red blood cells; TAE: Transcatheter arterial embolization; TAC: Total abdominal colectomy; CTA: Computed tomographic angiography; IR: Interventional radiology.

Antithrombotic management

Withholding antithrombotic agents during acute bleeding: The management of antithrombotic agents in patients with acute CDB requires individualized risk-benefit assessment, balancing the risk of ongoing or recurrent hemorrhage against the thromboembolic risk of drug discontinuation[5,12] (Table 9). For most patients with active significant bleeding, temporary interruption of antiplatelet and anticoagulant therapy is recommended, with duration guided by the severity of bleeding, the underlying indication for therapy, and achievement of hemostasis[5]. For patients on warfarin with supratherapeutic international normalized ratio, reversal with four-factor prothrombin complex concentrate is preferred over fresh frozen plasma[5]. For patients on DOACs with life-threatening hemorrhage, specific reversal agents include idarucizumab for dabigatran and andexanet alfa for factor Xa inhibitors[5,12]. It should be noted that andexanet alfa (Andexxa) was withdrawn from the United States market following Food and Drug Administration and AstraZeneca’s inability to agree on a feasible pathway to convert its accelerated approval to traditional approval, driven in part by concerns over excess thrombotic events and insufficient confirmatory evidence. Clinicians managing diverticular bleeding in patients on factor Xa inhibitors should be aware of this regulatory change and consult with their pharmacy and hematology services regarding currently available reversal options.

Table 9 Framework for antithrombotic management after colonic diverticular bleeding.

Agent/class	TE risk if stopped	Resumption timing	Clinical considerations
Anticoagulation (high TE risk)	High	7-14 days post-hemostasis	Earlier (< 7 days) for mechanical valves, recent stroke
Anticoagulation (AF)	High	≤ 7 days preferred	Lower MACE without significant increase in GI rebleeding in AF
Warfarin	Varies	After INR therapeutic	Use 4F-PCC (not FFP) for urgent reversal
Rivaroxaban	Moderate	Consider switch to apixaban	HR 1.53 higher LGIB risk vs apixaban; class switch reasonable
Aspirin (2° CV prevention)	High if stopped	Resume as soon as feasible	Discontinuation increases mortality; rebleeding risk outweighed
DAPT (recent PCI/ACS)	Very high if stopped	Continue aspirin; liaise cardiology re P2Y12	P2Y12 pause per stent age; never stop both without cardiology input
NSAIDs	Low (TE)	Substitute if possible	Replace with acetaminophen or COX-2 inhibitor where feasible
Andexanet alfa (factor Xa reversal)			Withdrawn from United States market; confirm availability with pharmacy/hematology

Resumption timing recommendations are based on observational data only; no randomized controlled trials exist. TE: Thromboembolic; AF: Atrial fibrillation; CV: Cardiovascular; DAPT: Dual antiplatelet therapy; PCI: Percutaneous coronary intervention; ACS: Acute coronary syndrome; NSAIDs: Non-steroidal anti-inflammatory drugs; INR: International normalized ratio; MACE: Major adverse cardiovascular events; GI: Gastrointestinal; 4F-PCC: Four-factor prothrombin complex concentrate; FFP: Fresh frozen plasma; HR: Hazard ratio; COX-2: Cyclooxygenase-2; LGIB: Lower gastrointestinal bleeding.

Resumption of antithrombotic therapy: The decision to resume antithrombotic therapy after CDB is one of the most clinically challenging aspects of management, and the optimal timing remains incompletely defined[5]. It is important to emphasize that the available evidence on antithrombotic resumption timing is derived predominantly from retrospective observational cohort studies, and no RCTs have specifically evaluated this question in the CDB population. The certainty of evidence supporting current timing recommendations is therefore low to moderate, and clinical judgment must supplement guideline recommendations on a case-by-case basis. In a retrospective cohort study by Sengupta et al[5] of 1329 patients hospitalized with LGIB who were on antithrombotic therapy, early resumption of anti-thrombotics (within 90 days) was associated with lower all-cause mortality (adjusted HR = 0.67; 95%CI: 0.49-0.91) and a lower composite outcome of mortality and thromboembolic events (adjusted HR = 0.71; 95%CI: 0.54-0.94), without a significant increase in rehospitalization for gastrointestinal bleeding (adjusted HR = 1.18; 95%CI: 0.82-1.70)[5]. Current ACG guidelines recommend resuming anticoagulation within 7-14 days after endoscopic hemostasis in patients with a high thromboembolic risk, with earlier resumption considered in those with the highest risk[5]. For patients requiring long-term aspirin for secondary cardiovascular prevention, continuation is generally preferred over discontinuation given the established mortality benefit[5,12].

Recurrence prevention

Magnitude of recurrence risk: Long-term recurrence of CDB is a significant clinical problem, with cumulative rebleeding rates of approximately 25%-40% within 4 years of an index episode[1,5]. In the CODE-BLUE J-study, 365-day rebleeding rates ranged from 32%-42% depending on the treatment strategy, with endoscopic hemostasis yielding the lowest rates[8]. Risk factors for recurrence include bilateral diverticulosis, continued use of antithrombotic agents (particularly aspirin and NSAIDs), and failure to achieve definitive endoscopic hemostasis at the index admission[5,8].

Predictors of recurrent bleeding and long-term outcomes after elective colectomy: Several clinical factors have been consistently identified as independent predictors of recurrent diverticular hemorrhage. Age ≥ 65 years, ongoing use of antiplatelet agents or NSAIDs, bilateral diverticulosis, hemodynamic instability at presentation, and a high comorbidity burden have all been associated with increased rates of recurrence across multiple cohort studies[37-39]. Cumulative recurrence rates have been reported at approximately 5% at 1 year, 8% at 2 years, and 16% at 5 years, with continued longitudinal risk thereafter[38,39]. A nationwide cohort study by Sato et al[39] of patients discharged after acute LGIB demonstrated that antithrombotic use, age, and the severity of the index bleeding episode were independently associated with long-term recurrence risk, highlighting the importance of individualized secondary prevention planning at discharge.

Elective colectomy significantly reduces but does not eliminate the risk of recurrent diverticular bleeding. Contemporary series report that 5-year recurrent diverticulitis or rebleeding rates are substantially lower after resection compared with conservative management; however, 22%-25% of patients report persistent abdominal symptoms postoperatively, and the overall functional burden of colectomy should be carefully weighed. Patient selection for elective surgical resection should therefore balance the demonstrated reduction in recurrence risk against operative morbidity, the possibility of persistent symptoms, and the patient’s overall comorbidity burden and functional status. Individualized risk stratification guided by the number and severity of prior episodes, medication use, and patient preference is essential when counseling patients regarding elective colectomy[5,37-39].

Role of mesalamine: Mesalamine has been proposed as a potential strategy for secondary prevention of CDB based on its anti-inflammatory properties. However, evidence specifically supporting its use for prevention of diverticular bleeding recurrence is limited and largely derived from retrospective studies. The updated ACG guideline does not recommend mesalamine for secondary prevention of CDB owing to insufficient evidence[5].

Modification of antithrombotic regimens: The modification or de-escalation of antithrombotic therapy represents one of the most impactful recurrence prevention strategies in appropriate patients. Substitution of non-selective NSAIDs with acetaminophen or selective cyclooxygenase-2 inhibitors where clinically feasible may reduce diverticular rebleeding risk[2]. Switching from rivaroxaban to apixaban may also be considered in patients with atrial fibrillation who experience LGIB[5].

Elective colectomy for recurrent diverticular bleeding: Elective surgical resection may be appropriate for select patients with recurrent, localized diverticular bleeding who are good operative candidates and in whom the bleeding segment has been definitively localized[5,6]. Current guidelines do not provide a specific threshold for the number of recurrent episodes that mandates surgical referral[5].

DISCUSSION

A summary of key clinical questions addressed in this review, the supporting evidence base, and the corresponding strength of recommendation is provided in Table 10. Several important limitations of the available evidence on CDB management warrant emphasis. First, as discussed in the epidemiology section, the preponderance of high-quality endoscopic outcome data particularly for EBL derives from Japanese cohorts in which right-sided diverticulosis predominates. The generalizability of these findings to Western populations with left-sided sigmoid diverticulosis remains uncertain, and no large prospective Western trials of EBL vs TTS clipping have yet been published.

Table 10 Summary of key clinical questions, supporting evidence, study types, quality of evidence, and strength of recommendation.

Clinical question	Supporting evidence	Study types	Quality of evidence	Strength of recommendation
Transfusion threshold (Hb 7 g/dL stable LGIB)	Villanueva RCT; 5-RCT meta-analysis; United Kingdom LGIB audit post hoc	2 RCTs (UGIB); 1 post hoc LGIB analysis	Moderate extrapolated from UGIB; LGIB-specific data very limited	Conditional 7 g/dL; 8 g/dL in CVD
CTA first in hemodynamically unstable patients	Meta-analysis 14 studies (90% sensitivity, 92% specificity); ACG-SAR consensus[14]	Systematic review; consensus guideline	Moderate	Strong
Elective colonoscopy (not urgent) for stable patients	Niikura et al[9] RCT (n = 170); second RCT (increased rebleeding with early scope)	2 RCTs	Moderate	Conditional urgent only in selected high-risk patients
EBL preferred over TTS clipping	Nagata et al[31] meta-analysis; CODE-BLUE J[7]; Meine et al[30]	Meta-analyses of observational data; large multicenter retrospective cohort	Moderate no RCTs; predominantly Japanese data	Conditional EBL may be preferred; prospective Western trials needed
OTSC as alternative to EBL	Ohashi et al[32]: PSM cohort n = 303; 1 single-center series	1 PSM retrospective cohort; 1 case series	Very low	Emerging insufficient evidence for routine use
TAE for failed endoscopy or active CTA extravasation	Sun et al[11]: 58 trials/3530 pts; Kitamura et al[36]	Meta-analysis + retrospective cohort	Moderate	Strong as salvage; conditional as primary
Antithrombotic resumption within 7-14 days (high TE risk)	Sengupta et al[5] retrospective cohort, 1329 pts	Retrospective cohort only; no RCTs	Low	Conditional individualize TE vs rebleeding risk
Elective colectomy for recurrent localized CDB	ACG guideline[5]; retrospective comparative series; You et al[37]; Vajravelu et al[38]; Sato et al[39]	Guideline; retrospective series; cohort studies	Low	Conditional after 2 + episodes, localized source, suitable candidate

Hb: Hemoglobin; LGIB: Lower gastrointestinal bleeding; CTA: Computed tomographic angiography; EBL: Endoscopic band ligation; TTS: Through-the-scope; OTSC: Over-the-scope clip; TAE: Transcatheter arterial embolization; TE: Thromboembolic; CDB: Colonic diverticular bleeding; RCT: Randomized controlled trial; ACG: American College of Gastroenterology; SAR: Society of Abdominal Radiology; PSM: Propensity score matched; pts: Patients; UGIB: Upper gastrointestinal bleeding; CVD: Cardiovascular disease.

Second, transfusion threshold recommendations in LGIB are extrapolated from upper gastrointestinal bleeding RCTs. The only LGIB-specific data derive from a post hoc analysis of the United Kingdom audit, which was not powered to detect modest differences in mortality or rebleeding. Dedicated LGIB transfusion RCTs are needed.

Third, definitions used across studies are inconsistent the threshold defining “early” rebleeding (often 30 days but ranging from 7 days to 90 days) varies, as does the distinction between “definitive” and “presumptive” CDB. Fourth, evidence for OTSC, mesalamine, and topical hemostatic agents remains predominantly retrospective and single-center. Fifth, most TAE outcome data derive from retrospective cohort studies with heterogeneous embolic agents and technique definitions[11]. Sixth, evidence on optimal antithrombotic resumption timing derives largely from a single retrospective cohort[5]. Seventh, cost-effectiveness data comparing the available diagnostic and therapeutic pathways are limited.

Finally, this review itself has inherent limitations. It is a narrative review with systematic literature search rather than a fully compliant systematic review with formal risk-of-bias assessments and meta-analytic pooling of data. This design was chosen given the substantial clinical and methodological heterogeneity across the included studies, which precluded meaningful statistical pooling. We restricted inclusion to English-language publications, potentially missing relevant evidence published in other languages.

CONCLUSION

CDB is a common and potentially serious clinical entity whose management has evolved substantially over the past two decades. The paradigm has shifted from reflexive surgical intervention to a structured, stepwise approach anchored in risk stratification, precise diagnostic evaluation, and targeted endoscopic or radiologic hemostasis. The Oakland score provides a validated and practical tool for early risk stratification, enabling safe early discharge in low-risk patients, and should be used in conjunction with clinical judgment rather than as a standalone instrument. CTA has emerged as a rapid, high-yield first-line diagnostic modality in hemodynamically unstable patients, while colonoscopy with bowel preparation remains the cornerstone diagnostic and therapeutic tool for stable patients. Among endoscopic modalities, current evidence derived predominantly from Japanese multicenter retrospective cohorts and meta-analyses suggests that EBL may reduce early and late rebleeding compared with TTS clipping. However, prospective multicenter studies in Western populations are needed before strong recommendations can be made. OTSC represents a promising emerging alternative warranting prospective evaluation. TAE with super selective microcatheter embolization is an effective salvage modality when endoscopic hemostasis fails, and may serve as the primary therapeutic approach in patients with hemodynamically significant bleeding and active extravasation on CTA. Surgery remains reserved for refractory or recurrent cases with a clearly localized bleeding source and should be performed electively whenever possible. The high rate of long-term recurrence approaching 40% at 4 years mandates individualized antithrombotic medication review, targeted modification of modifiable risk factors, careful patient selection for elective colectomy, and thoughtful counseling regarding the risks and benefits of surgical intervention. Future research should prioritize prospective trials of EBL and OTSC in Western populations, LGIB-specific transfusion threshold RCTs, prospective evaluation of mesalamine for secondary prevention, optimal antithrombotic resumption timing trials, and the integration of CTA into algorithmic care pathways.