Sage Journals: Discover world-class research

Abstract

Background:

Colonic diverticular bleeding is the most common cause of lower gastrointestinal bleeding in adults and carries a significant risk of recurrence. However, there are many uncertainties regarding the management of the prevention of diverticular rebleeding.

Objectives:

To review the current evidence on the potential role of lifestyle, pharmacological and endoscopic treatments and to discuss the unmet needs in the prevention of colonic diverticular rebleeding.

Design:

A systematic review.

Data sources and methods:

Based on the identified Patients-Interventions-Comparators-Outcomes questions, a detailed and comprehensive literature search was conducted, from inception to 12 January 2024, without language restriction, according to the modified Preferred Reporting Items for Systematic review and Meta-Analyses reporting guidelines.

Results:

We did not find any dietary or lifestyle interventions (fibre intake, smoking, physical activity, alcohol consumption, BMI) to prevent colonic diverticular rebleeding. We also did not find any interventional studies of specific pharmacological treatments (such as rifaximin, mesalazine or probiotics) to prevent diverticular rebleeding. Data comparing endoscopic and conservative approaches used during the index episode come from observational studies and show conflicting results. Finally, there is a paucity of data regarding the timing of resumption of antiplatelet and anticoagulant therapy after an episode of colonic diverticular bleeding, and this remains to be determined.

Conclusion:

This review highlights the paucity of data on the possible role of lifestyle, pharmacological and endoscopic treatments in the prevention of colonic diverticular rebleeding and advocates future studies aimed at finding effective therapeutic strategies.

Keywords

anticoagulant treatment antiplatelet treatment diverticular disease endoscopic management

Introduction

Colonic diverticulosis is a common condition, particularly in developed countries, with high prevalence rates in Europe and the United States.¹ Most patients with diverticulosis remain asymptomatic throughout their lives. The two main diverticular complications, namely acute diverticulitis and diverticular bleeding, occur in 1%–5% of patients.^2,3

Colonic diverticular bleeding (CDB), which represents the most common cause of lower gastrointestinal (GI) bleeding in adult patients, carries a significant risk of recurrence. Recurrence rates for CDB vary widely, ranging from 3.8% to 15.1% and 6.9% to 25% at 1 and 5 years, respectively.^4–6

Although most cases of diverticular bleeding resolve spontaneously (80%–90%), a small proportion progresses to haemorrhagic shock requiring intensive care.² The in-hospital mortality rate due to CDB ranges from 0.7% to 2.5%, with older age and multiple comorbidities being the strongest predictors of mortality.^7,8

Patients with spontaneous bleeding resolution not undergoing urgent Gl evaluation (i.e. colonoscopy, contrast-enhanced angiography, interventional radiology) remain underdiagnosed and therefore the suspected diverticular source of bleeding is only presumed, additionally masking the real epidemiology of this condition.^9,10

Mechanisms hypothesised to be involved in CDB are vascular disease and/or structural weakness leading to arterial rupture. Inflammation does not appear to play a role in CDB since acute diverticulitis is rarely complicated by bleeding and the risk factors associated with these two complications may be different. It can be assumed that an alteration in the function of the colonic muscle may also contribute to the pathogenesis of this complication.¹¹

Although recurrent CDB is a common condition, only a few studies, mostly from Eastern countries, investigated risk factors for preventing recurrence.

This review aims to ascertain the current evidence regarding the potential role of lifestyle, pharmacological and endoscopic treatment and discuss the unmet needs regarding the prevention of colonic diverticular rebleeding.

Materials and methods

An expert panel composed of six gastroenterologists with long-lasting experience in colonic diverticular disease was involved in identifying the most important open questions regarding the management of this condition. In a face-to-face meeting, chaired by a panel moderator experienced in facilitating group discussions and criteria development, the experts were asked to generate relevant clinical questions using the Patients-Interventions-Comparators-Outcomes (PICO) format. Based on the identified PICO questions, a detailed and comprehensive literature search was conducted, from inception to 12th January 2024, without language restriction, according to the modified Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) reporting guidelines.¹² The search strategy is reported in Supplemental Table 1. Following the search strategy, papers published in languages other than English were not considered. The references of the selected papers were also reviewed to identify additional papers of potential interest. The final list of references was evaluated by the panel experts, who were asked to check for any lack of relevant studies. Information on patient population, study design, interventions, control group and outcomes assessed was collected by two authors independently. Discordance regarding the pertinence of the study to address each PICO was resolved in a face-to-face meeting. We have not registered our review in PROSPERO; we plan to do so for future studies.

Results

Three PICO questions were identified regarding the management of diverticular disease after a CDB episode (Table 1). The literature search initially identified 362 papers, of which 45 were considered relevant to address the PICO questions (Figure 1). When available, the results of meta-analyses were used as the primary source of information.

Table 1.

PICO questions identified by the expert panel.

Question	Population	Intervention	Control	Outcomes
Question #1	Patients with a previous episode of diverticular bleeding	Dietary/lifestyle interventions: 1. High-fibre diet/fibre supplementation 2. Lifestyle intervention (smoking cessation, physical activity, alcohol consumption, weight management).	1. Free diet 2. No lifestyle intervention	Recurrence of bleeding Diverticulitis Quality of life Hospitalisations Need for surgical intervention Resource utilisation Mortality
Question #2	Patients with a previous episode of diverticular bleeding	Pharmacological treatment: 1. Rifaximin 2. Mesalazine 3. Probiotics Endoscopic treatment adopted during the index episode	Placebo Usual practice Head-to-head comparison	Recurrence of bleeding Diverticulitis Quality of life Hospitalisations Need for surgical intervention Resource utilisation Mortality
Question #3	Patients on antiplatelet/anticoagulant treatment with an episode of diverticular bleeding	Resumption of antiplatelet/anticoagulant treatment after the episode of diverticular bleeding	Different timing of resumption of antiplatelet/anticoagulant treatment Interruption of antiplatelet/anticoagulant treatment	Recurrence of bleeding thrombosis Cardiovascular events Need for surgery/interventional radiology Mortality

Figure 1.

Study flow chart.

Role of dietary and lifestyle interventions in preventing colonic diverticula rebleeding

The bibliographic search did not allow to identify any dietary or lifestyle intervention studies (dietary fibre intake, smoking, physical activity, alcohol consumption, body weight) to prevent diverticular rebleeding.

Several observational studies have investigated risk factors associated with diverticular rebleeding^13–27 (Table 2). In all, 11 of the 15 studies were conducted in Asian countries, with the majority coming from Japan. Of the remaining studies, two were from the USA, one from Portugal and one from France.

Table 2.

Observational studies on risk factors associated with diverticular rebleeding.

Author (year)^Ref.	Study design	Number of patients	Risk factors evaluated	Independent predictors of rebleeding
Gonai (2022)¹³	Retrospective (re-bleeding within 1 month)	370	Age, sex, systolic blood pressure, hypertension, diabetes, CKD, neoplasms, SRH, haemostatic treatment, performance status, antithrombotic agents (ATs), number of ATs, continuation/discontinuation of ATs, NSAIDs, steroids.	ATs (yes vs no) (HR = 2.38; 95% CI 1.10–2.50; p = 0.03). Multiple ATs versus no ATs (HR = 3.88; 95% CI 1.49–10.0; p = 0.007). Continuation of ATs versus no ATs (HR = 3.30; 95% CI 1.23–8.63; p = 0.02).
Gonai (2021)¹⁴	Retrospective (re-bleeding within 1 year)	324	Age, sex, systolic blood pressure, hypertension, diabetes, CKD, neoplasms, haemostatic treatment, antithrombotic agents (ATs), number of ATs, continuation/discontinuation of ATs, NSAIDs, steroids.	ATs (HR = 3.89; 95% CI 1.53–10.74; p = 0.004
Sato (2021)¹⁵	Retrospective, case–control (late re-bleeding: >30 days)	519	Age, sex, hypertension, diabetes, CKD, cardiovascular disease, cerebrovascular disease, respiratory disease, antithrombotic agents, NSAIDs, bilateral diverticula, bleeding source, type of SRH, CT extravasation, endoscopic haemostasis, shock index, blood transfusion	Chronic kidney disease (OR = 2.22; 95% CI 1.49–3.27; p < 0.001) NSAIDs (OR = 2.27; 95% CI 1.37–3.78; p = 0.002) Bilateral diverticula (OR = 1.98; 95% CI 1.361–2.895; p < 0.001).
Jalil (2018)¹⁶	Retrospective, case–control (re-bleeding within 2 years)	93 cases and 152 controls	Age, sex, race, length of stay, hypertension, Hb at admission, BMI, diabetes, CKD, cardiovascular disease, cerebrovascular disease, obstructive sleep apnoea, gout, alcohol and tobacco consumption, antithrombotic agents, NSAIDs, proton pump inhibitors, ACEi/ARBs, beta-blockers, calcium channel blockers.	No statistically significant predictors of rebleeding were found in multivariate analysis.
Kitagawa (2019)¹⁷	Retrospective (re-bleeding within 90 days)	144	Age, sex, race, length of stay, Hb at admission, BMI, history of diverticular bleeding, blood transfusion, shock on admission, hypertension, diabetes, CKD, hyperlipidaemia, cardiovascular disease, cerebrovascular disease, antiplatelets, aspirin, anticoagulants, NSAIDs, steroids, SRH diverticula	Shock on admission (OR = 5.12; 95% CI, 1.17–22.43, p = 0.030)
Vajravelu (2018)¹⁸	Retrospective (re-bleeding >30 days)	14,925	Anticoagulants and PAIs, aspirin, NSAIDs, SSRIs, age, sex, race, comorbidities	Platelet aggregation inhibitors (OR = 1.47; 95% CI 1.15–1.88; p < 0.05).
Taki (2017)¹⁹	Case-control (re-bleeding at 1 year)	100 cases and 200 controls	Age, gender, location of diverticulosis, BMI, hypertension, dyslipidaemia, diabetes mellitus, cerebrovascular disease, ischaemic heart disease, medication use (NSAIDs, COX-2 inhibitors, low-dose aspirin, non-aspirin antiplatelet drugs, anticoagulants, steroids, proton-pump inhibitors), dialysis, haemostasis, colonoscopy within 24 h.	Steroids (HR = 4.31; 95% CI 2.35–45.84; p = 0.002).
Lorenzo (2017)²⁰	Retrospective (median FUP 3.9 years)	365	Age > 80 years, gender, anticoagulant or antiplatelet therapy prior to re-bleeding, Charlson comorbidity score >2, past history of bleeding.	No statistically significant predictors of rebleeding were found in multivariate analysis.
Joaquim (2017)²¹	Retrospective (re-bleeding at 1 year)	74	Age, gender, location of diverticula, severe bleeding at the index episode, treatment for the first episode, hypertension, hyperlipidaemia, diabetes, CKD, cardiovascular disease, peripheral vascular disease, arrhythmia, ischaemic heart disease, CHF, valvular heart disease, cerebrovascular disease, osteoarticular disease, chronic pulmonary disease, malignant neoplasia, Charlson-age comorbidity index, medication (NSAIDs, anticoagulant, antiaggregant).	No statistically significant predictors of rebleeding were found at multivariate analysis.
Nagata (2015)²²	Prospective (re-bleeding at 1 year)	41	NSAIDs	NSAIDs discontinuation (HR = 0.06; 95% CI 0.01–0.31)
Watanabe (2014)²³	Retrospective (median interval 11.5 months)	151	Age, sex, length of stay, treatment for the first episode, BMI, antiplatelets, aspirin, anticoagulants, NSAIDs, hypertension, ischaemic heart disease, cerebrovascular disease, diabetes, hyperlipidaemia, alcohol and tobacco consumption, systolic blood pressure, heart rate at admission, Hb, creatinine, blood transfusion, the position of haemorrhage.	Hypertension (OR = 6.72; 95% CI 2.33–23.5; p < 0.001) Hyperlipidaemia (OR = 3.55; 95% CI 1.56–8.35; p = 0.002) Heart rate at admission (OR = 1.03; 95% CI 1.01–1.06; p = 0.01) Length of stay (OR = 1.19; 95% CI 1.05–1.35; p = 0.005).
Fujino (2013)²⁴	Retrospective (re-bleeding within 1 month)	90	Age, gender, history of diverticular bleeding, hypertension, NSAIDs, antithrombotic agents, time from the start of bleeding to admission, signs of shock, Hb, active bleeding on the first colonoscopy, transfusion before first re-bleeding	Signs of shock on admission (OR = 5.23; 95% CI 1.84–14.90; p = 0.0019).
Niikura (2012)²⁵	Retrospective (re-bleeding at a median interval of 1535 days).	72	NSAIDs, steroids, antithrombotic agents, hypertension, diabetes mellitus, hyperlipidaemia, cardio/cerebrovascular disease, chronic hepatic dysfunction, chronic renal failure, alcohol and tobacco consumption.	NSAIDs (HR = 2.57; 95% CI 0.89–7.46; p = 0.08); antiplatelet agents (HR = 2.39; 95% CI 1.01–5.67; p = 0.05); hypertension (HR = 4,16; 95% CI 1.22–14.2; p = 0.02)
Tanaka (2012)²⁶	Retrospective (re-bleeding within 1 month and >1 month)	111	Age, sex, BMI, hypertension, diabetes mellitus, hyperlipidaemia, cardio/cerebrovascular disease, steroid, antiplatelets, NSAIDs, anticoagulants, antihypertensives, colonoscopic findings	Multivariate analysis not performed. At univariate analysis, patients with re-bleeding had higher BMI. Colonoscopic findings of actively bleeding or nonbleeding visible vessels in the responsible diverticula were more frequent in the group with re-bleeding.
Okamoto (2012)²⁷	Retrospective (mean FUP 2.4 years)	62	Sex, age, hypertension, hyperlipidaemia, diabetes, cardio/cerebrovascular disease, anticoagulant drugs, NSAIDs, treatment using haemoclips	NSAIDs (HR = 6.3, 95%CI 1.7–20.7; p = 0.007)

BMI, body mass index; CHF, chronic heart failure; CKD, chronic kidney disease; FUP, follow-up; NSAIDs, non-steroidal anti-inflammatory drugs; PAI, platelet aggregation inhibitors; SRH, stigmata of recent haemorrhage; SSRI, selective serotonin reuptake inhibitors.

All the studies were retrospective, and most were small. They evaluated the risk of rebleeding after a variable time from the index episode (ranging from within a month to several years) and reported a considerable number of different risk factors. Some of these studies have evaluated the role of smoking and alcohol consumption,^16,23,25 but an independent role of these risk factors has not been demonstrated. The role of body mass index (BMI) has also been investigated in several studies,^{16,17,19,23,26} but it has never been found to be an independent predictor of diverticular rebleeding in multivariate analysis. In one study,²⁶ patients with rebleeding had a higher BMI on univariate analysis, but no multivariate analysis was performed.

Therefore, based on observational data, there is currently no evidence to suggest that lifestyle or dietary factors are associated with the risk of diverticular bleeding recurrence.

Role of pharmacological management in the prevention of recurrent CDB

The bibliographic search did not allow us to identify any pharmacological treatment intervention studies focusing on drugs commonly used in diverticular disease (rifaximin, mesalazine, probiotics) to prevent diverticular rebleeding.

Role of endoscopic management of the index episode in the prevention of recurrent CDB

Endoscopic treatment versus conservative approach

To the best of our knowledge, there are no studies directly comparing endoscopic versus a conservative and the available data come from observational studies. In a small cohort study,²⁸ 17 and 10 patients with a definite source of bleeding were managed conservatively and endoscopically, respectively. Of the 17 patients treated conservatively, 9 had recurrent or persistent bleeding requiring additional transfusion. Three of these were managed with medical treatment that stopped the bleeding, while the remaining six patients had severe bleeding and required emergency hemicolectomy. None of the 10 patients treated endoscopically had recurrent bleeding or complications or required further red cell transfusions or surgery. No delayed re-bleeding was observed in either group.

Another observational study compared 88 patients who underwent endoscopic (i.e. injection and/or clipping) or conservative treatment.²⁹ In all, 16 (61.5%) of 26 endoscopically treated patients and 24 (38.7%) of 62 conservatively treated patients experienced CDB recurrence during a median follow-up of 42.7 months. Kaplan–Meier analysis showed that the recurrence rate was significantly higher (p < 0.05) in endoscopically treated cases compared to conservatively treated cases, with a mean time to recurrence of 55.3 months (95% CI 30.8–79.9) versus 99.9 months (95% CI 80.7–119.1). These observational studies have important methodological limitations, mainly related to the difficulty in assessing whether patients treated endoscopically and those treated conservatively had similar characteristics at baseline in terms of disease severity and risk factor distribution.

More recently, a large nationwide retrospective study of 5,823 patients with CDB undergoing colonoscopy in 49 Japanese hospitals compared three strategies: (1) find stigmata of recent haemorrhage (SRH) and treat endoscopically, (2) find SRH and treat conservatively and (3) treat conservatively without finding SRH (presumed CDB).³⁰ When conducting pairwise comparisons of outcomes in these groups, the propensity score matching technique was used to balance baseline characteristics between the compared groups. Both early and late recurrent bleeding rates were significantly lower in patients with proven CDB treated endoscopically compared to those with presumed CDB treated conservatively (<30 days, 19.6% vs 26.0% (p < 0.001); <365 days, 33.7% vs 41.6% (p < 0.001), respectively). In patients with established CDB, the rate of early recurrent bleeding was significantly lower in those treated endoscopically compared to those treated conservatively (17.4% vs 26.7%; p = 0.038 for a single hypothesis test). However, this difference was no longer significant after correction for multiple testing. The rate of late recurrent bleeding was also lower in the endoscopically treated group, but the difference was not statistically significant. Patients with established CDB treated endoscopically had significantly lower rates of early and late recurrent bleeding rates than patients with established CDB treated conservatively with active bleeding, non-active bleeding and in the right colon but not the left colon. Promising observations on the usefulness of endoscopic treatment in reducing late bleeding (within 30 days) could be hypotheses, in the light of the conclusions of the recent study by Aoki et al.³¹ where the analysis is restricted to patients with SRH; the authors found that in the presence of diverticular bleeding and HRS, endoscopic therapy does indeed reduce the risk of rebleeding by 40%.

Endoscopic treatment: Clipping versus ligation

The role of the endoscopic approach used for the index episode of CDB in reducing the risk of recurrence has been evaluated in several studies and summarised in a recent systematic review.³² This review assessed the comparative effectiveness of endoscopic clipping versus band ligation for the prevention of CDB recurrence. In all, 16 studies met the eligibility criteria, with a total of 790 participants. Band ligation showed a significantly lower pooled prevalence of both early (within 30 days) and late (within 1 year) rebleeding compared with clipping (0.08 vs 0.19 (test for heterogeneity, p = 0.012), 0.09 vs 0.29 (test for heterogeneity, p = 0.024), respectively). There was no significant difference in the initial haemostasis between the two groups. The overall prevalence of patients requiring transcatheter arterial embolisation or surgery was significantly lower with band ligation than with clipping (0.01 vs 0.02; test for heterogeneity, p = 0.031). Two cases of colonic diverticulitis were reported after band ligation, but none after clipping.

Endoscopic treatment: Direct versus indirect clipping

Regarding the type of endoscopic approach, we found only one large Japanese observational study comparing direct and indirect clipping.³³ In direct clipping, the vessel was clamped directly, whereas in indirect clipping, the diverticular orifice was closed in a zipper-like fashion. The study included 1041 patients with CDB, of whom 360 underwent direct clipping and 681 underwent indirect clipping. Multivariate analysis adjusted for age, sex and important confounders (heart rate ⩾ 100 bpm, modified CCI ⩾ 2, extravasation on computed tomography, active bleeding, use of an endoscopic distal attachment cap and use a waterjet scope) showed that direct clipping was independently associated with a reduced risk of early rebleeding (<30 days; adjusted odds ratio (AOR) 0.592, p = 0.002), late rebleeding (<1 year; AOR 0.707, p = 0.018) and need for blood transfusion requirement (AOR 0.741, p = 0.047) compared to indirect clipping. There was no significant difference in initial haemostasis rates observed between the two groups. Propensity score matching showed a significant reduction in the early and late rebleeding rates with direct clipping compared to indirect clipping. Therefore, when comparing band ligation versus clipping or clipping versus conservative approach, the type of clipping (direct vs indirect) should be considered.

Resumption of antiplatelet and anticoagulant therapy after an episode of diverticular bleeding

Antiplatelet therapy

No studies have evaluated when antiplatelet therapy (i.e. aspirin, thienopyridine, cilostazol, eicosapentaenoic acid, sarpogrelate hydrochloride, dipyridamole and prostaglandin E1 derivatives) should be resumed after an interruption in patients with CDB.

A retrospective, cohort study of 295 patients with lower GI bleeding (not limited to CDB) on aspirin showed that continuation of aspirin was associated with an almost three-fold increased risk of recurrent bleeding but also with a 1.6-fold reduced risk of major cardiovascular events and a more than three-fold reduced risk of death within 5 years, which underlies the net benefit of resuming aspirin after the bleeding event.³⁴

Anticoagulant therapy

Only one study evaluated whether anticoagulant therapy should be restarted after a CDB episode.¹⁸ In CDB patients treated for ischaemic stroke prophylaxis, there was no increased relative risk of recurrent CDB in those who continued any form of anticoagulant therapy compared with those who discontinued (HR 0.98, 95% CI 0.79–1.22). However, discontinuation of anticoagulation was associated with an increased relative risk of ischaemic stroke (HR 1.93, 95% CI 1.17–3.19).

Given the paucity of studies specifically focusing on CDB, we analysed the available data on GI bleeding (not limited to CDB) with appropriate caution to identify possible risk factors for diverticular rebleeding. Several cohort studies have evaluated the effect of resuming anticoagulant therapy after withdrawal due to acute GI bleeding (therefore not specific to diverticular bleeding) and have shown that prolonged interruption of anticoagulant therapy is associated with a risk of thromboembolism^35–37 and subsequent increased mortality at 3³⁶ and 12 months.³⁷

A meta-analysis of three observational studies showed that restarting warfarin was associated with a significant reduction in thromboembolic events (hazard ratio (HR) 0.68; 95% CI 0.52–0.88, p < 0.004, I² = 82%).³⁸ There was no statistically significant increase in recurrent GI bleeding in patients who resumed warfarin compared to those who did not (HR 1.20; 95% CI 0.97–1.48, p = 0.10, I² = 0%). Restarting warfarin was associated with a significant reduction in mortality (HR 0.76; 95% CI 0.66–0.88, p < 0.001, I² = 87%).

A more recent meta-analysis of 10 observational studies included 2080 patients who resumed anticoagulation and 2296 patients who discontinued anticoagulation after a GI bleeding episode.³⁹ The anticoagulant used in seven studies was warfarin alone, and three studies included patients on DOACs. The duration of anticoagulation withdrawal after GI bleeding ranged from 4 days (IQR 2–9 days) to 50 days (IQR 21–78 days). Resumption of anticoagulation was associated with a significant increase in recurrent GI bleeding (OR 1.646; 95% CI 1.035–2.617, p = 0.035). However, a significant reduction in thromboembolic events was observed in patients who resumed anticoagulant therapy compared to those who did not (OR 0.340; 95% CI 0.178–0.652, p = 0.001, I² = 62.7%). Resumption of anticoagulant therapy was also associated with a significant reduction in all-cause mortality (OR 0.499; 95% CI 0.419–0.595, p < 0.0001). Taken together, these data, deriving from observational studies not limited to CDB, suggest that resuming the anticoagulant therapy may increase the risk of GI bleeding but reduce the overall risk of thromboembolic events and mortality. However, selection bias may limit the interpretation of these results. Indeed, we cannot exclude the possibility that anticoagulation was less likely to be resumed in frail patients, who have a higher risk of thromboembolic events and mortality.

The above-mentioned meta-analysis by Tapaskar et al.³⁹ looked specifically at the timing of anticoagulation resumption, which was reported in five studies.^{36,37,40–42} The median time to restart therapy ranged from 4 days³⁶ to 50 days.³⁷ Overall, these studies suggest that restarting anticoagulation between 2 and 6 weeks after the initial discharge or index bleeding episode may be optimal and warranted.³⁹ However, it has been argued that this suggestion should be considered with caution, as it is not based on the results of a meta-analysis, but on an unstructured review that was not subjected to formal statistical analysis.⁴³ In addition, most of the data came from patients treated with vitamin K antagonists (VKAs), with the exception of one study of direct oral anticoagulants (DOACs). Therefore, the faster onset of action of the DOAC must be taken into account and it may be necessary to delay the resumption times of the DOAC longer than the resumption times of warfarin. A larger prospective study of a DOAC or VKA comparing early and late resumption of oral anticoagulants after GI bleeding would be ideal to answer the optimal timing of resumption of oral anticoagulants.⁴³

A longitudinal study⁴⁴ included 948 patients hospitalised for GI bleeding occurring during treatment with vitamin K antagonists (n = 531) or DOACs (n = 417). In a time-dependent analysis, anticoagulant treatment was associated with a higher risk of recurrent clinically relevant bleeding (HR 1.55; 95% CI 1.08–2.22), but a lower risk of thromboembolism (HR 0.34; 95% CI 0.21–0.55) and death (HR 0.50; 95% CI 0.36–0.68). Previous bleeding, index major bleeding and lower glomerular filtration rate were associated with a higher risk of recurrent bleeding. The incidence of recurrent bleeding increased after anticoagulation was resumed, regardless of the timing of the resumption. The study suggests that the risk of recurrent bleeding seems to be influenced by patient characteristics rather than the timing of anticoagulation resumption.

Effect of non-steroidal anti-inflammatory drugs and aspirin use in colonic diverticular rebleeding

In view of the known adverse effects of non-steroidal anti-inflammatory drugs (NSAIDs) and aspirin on the GI tract, we have also analysed the potential effect of these drugs on colonic diverticular rebleeding. As shown in Table 2, data from observational studies indicate that the only modifiable risk factor that may reduce the risk of CDB is the discontinuation of NSAIDs. Indeed, the use of NSAIDs is a strong independent risk factor for rebleeding in some studies,^15,22,25,27 although other studies have found no significant association between the use of NSAIDs and the risk of rebleeding.^{13,14,16–19,21,23,24,26} A possible explanation for this conflicting result may be related to the frequency of use and dosage of NSAIDs.

In Table 3, the risk of diverticular rebleeding in relation to NSAIDs and aspirin dosage and frequency of use has been reported. Unfortunately, most studies do not report data on the duration and/or dosage of treatment.

Table 3.

Risk of diverticular rebleeding with respect to the NSAIDs/aspirin dosage and frequency use.

Author (year)	Study design	Number of patients	Risk of CDB in NSAIDs high dosage/regular use	Risk of CDB in NSAIDs low dosage /occasional use	Risk of CDB in aspirin occasional use	Risk of CDB in aspirin chronic use (as antiplatelets)
Gonai (2022)¹³	Retrospective (re-bleeding within 1 month)	370	Not specified	Not specified	Not specified	Not specified
Gonai (2021)¹⁴	Retrospective (re-bleeding within 1 year)	324	Not specified	Not specified	Not specified	Not specified
Sato (2021)¹⁵	Retrospective, case–control (late re-bleeding: >30 days)	519	Not specified	Not specified	Not specified	Not specified
Jalil (2018)¹⁶	Retrospective, case–control (re-bleeding within 2 years)	93 cases and 152 controls	Not specified	Not specified	Not specified	Not specified
Kitagawa (2019)¹⁷	Retrospective (re-bleeding within 90 days)	144	Not specified	Not specified	Not specified	Not specified
Vajravelu (2018)¹⁸	Retrospective (re-bleeding >30 days)	14,925	Not specified	Not specified	Not specified	Not specified
Taki (2017)¹⁹	Case–control (re-bleeding at 1 year)	300	Regular use HR = 0.87 (0.21–3.58)	Not specified	Not specified	Low-dose aspirin HR = 0.80 (0.27–2.39)
Lorenzo (2017)²⁰	Retrospective (median FUP 3.9 years)	365	Not specified	Not specified	Not specified	Not specified
Joaquim (2017)²¹	Retrospective (re-bleeding at 1 year)	74	Not specified	Not specified	Not specified	Not specified
Nagata (2015)²²	Prospective (re-bleeding at 1 year)	41	NSAIDs discontinuation HR = 0.06 (0.01–0.31)	Not assessed	Not assessed	Low-dose aspirin at discharge was not significant in multivariate analysis
Watanabe (2014)²³	Retrospective (median interval 11.5 months)	151	Not specified	Not specified	Not specified	Not specified
Fujino (2013)²⁴	Retrospective (re-bleeding within 1 month)	90	Not specified	Not specified	Not specified	Not specified
Niikura (2012)²⁵	Retrospective (re-bleeding at a median interval of 1535 days).	72	Not specified	Not specified	Not specified	Not specified
Tanaka (2012)²⁶	Retrospective (re-bleeding within 1 month and >1 month)	111	Not specified	Not specified	Not specified	Not specified
Okamoto (2012)²⁷	Retrospective (mean FUP 2.4 years)	62	Not specified	Not specified	Not assessed	Not assessed

Discussion

It has been estimated that CDB is the most common cause of lower GI bleeding in adult patients,^44–47 accounting for more than 40% of bleeding episodes.

In this review, we aimed to evaluate the current evidence on the prevention of colonic diverticular rebleeding, assessing the potential role of lifestyle, pharmacological and endoscopic treatments.

In assessing dietary and lifestyle risk factors (intake of dietary fibre, smoking, physical activity, alcohol consumption, BMI), we did not identify any interventions to prevent diverticular rebleeding. Several small, retrospective studies have investigated smoking, alcohol consumption and BMI as potential risk factors for CDB recurrence, without identifying them as independent predictors. Despite large prospective studies linking obesity with CDB⁴⁸ and alcohol consumption in general with GI bleeding (not limited to CDB),⁴⁹ there is, unfortunately, no evidence of their effect on CDB recurrence, so this remains an area for further research.

Unfortunately, the bibliographic search did not identify any specific pharmacological treatment (such as rifaximin, mesalazine or probiotics) able to prevent diverticular rebleeding. On the contrary, the role of the endoscopic approach used during an index episode of CDB in reducing the risk of recurrence has been evaluated in several studies.

Data comparing endoscopic and conservative approaches to CDB are from observational studies and show conflicting results. In particular, one cohort study showed that endoscopic management was associated with fewer recurrences, complications and need for surgery compared with conservative management.²⁸ On the other hand, another observational study showed that patients treated conservatively had fewer CDB recurrences during follow-up than those treated endoscopically.²⁹ Another large retrospective study showed that endoscopic management of confirmed CDB was more effective in reducing short- and long-term recurrences than conservative management of confirmed or presumed CDB.³⁰ A possible explanation for these conflicting results is the possible differences in the baseline characteristics of the patients, as all these studies are observational, which may have led to selection bias. Although a more recent study³¹ also provides promising data on the role of endoscopic treatment, the generalisability of the data seems limited. First, HRS is much less common in Western countries than described in the Japanese population, as Western guidelines do not support early endoscopy (within 24 h); second, the benefit of endoscopic treatment is mainly for right colonic bleeding, which is clearly more common in Eastern than in Western populations. Further prospective randomised studies, also in Western countries, are needed to better determine whether endoscopic treatment is superior to observation management in preventing colonic diverticular rebleeding.

Regarding the endoscopic technique that should be used to prevent CDB recurrence, a recent review showed that band ligation had fewer cases of both early and late rebleeding compared with clipping, although there was no significant difference in the rate of initial haemostasis between the two methods.^27,32 Furthermore, in a large observational study comparing direct clipping and indirect clipping, direct clipping was independently associated with a lower risk of early and late rebleeding and the need for blood transfusion.^32,33 Therefore, when comparing endoscopic techniques for the management of CDB (band ligation vs clipping), the type of clipping (direct vs indirect) should also be considered.

The timing of resumption of antiplatelet and anticoagulant therapy remains to be determined. In fact, we found only one study that investigated this aspect in CDB. The only available study that investigated the resumption of anticoagulant therapy after its discontinuation in patients with CDB showed that there was no increased relative risk of CDB recurrence in those who resumed any of the anticoagulant therapies.¹⁸

Instead, most of the available data from observational studies focus on general acute GI bleeding. Evidence from studies not limited to CDB suggests that although resumption of anticoagulant therapy may increase the risk of GI bleeding, it may also reduce the risk of thromboembolic events and mortality. We have assumed that management approaches and findings from GI bleeding studies may be applicable to CDB, but this assumption has not been definitively proven. As a result, clinicians must rely on indirect evidence when making decisions about both the management and timing of the resumption of antiplatelet or anticoagulant therapy in patients with CDB.

Finally, the only modifiable risk factor associated with a reduction in the risk of CDB is the discontinuation of NSAIDs. However, there are inconsistencies in the available data, probably due to differences in dosage and frequency of use. In fact, therapeutic regimens are often poorly characterised in these studies, contributing to the conflicting results. There is also a paucity of data on complications and mortality associated with discontinuing NSAIDs.

Limitation

Factors limiting the generalisability of these data are that most of the evidence comes from Asia, making it difficult to apply the findings to Western countries, due to differences in population, but also to differences in the diverticular disease itself, including the different location of diverticula. Future research perspectives include the need for studies conducted in Western countries to understand whether the location of diverticula might affect the outcome of treatments, especially endoscopic ones. Another aspect to be studied in the future is to evaluate the timing of reintroduction of antiplatelet and anticoagulant therapies, which are widely used in the elderly population, where the prevalence of colonic diverticula is significant.

Conclusion

In conclusion, this review highlights the paucity of data on the possible role of lifestyle, pharmacological and endoscopic treatments in the prevention of colonic diverticular rebleeding and advocates future studies aimed at finding effective therapeutic strategies.

Supplemental Material

sj-docx-1-tag-10.1177_17562848251321695 – Supplemental material for Look inside the management of colonic diverticular rebleeding: a systematic review

Supplemental material, sj-docx-1-tag-10.1177_17562848251321695 for Look inside the management of colonic diverticular rebleeding: a systematic review by Marilia Carabotti, Giovanni Marasco, Franco Radaelli, Giovanni Barbara, Rosario Cuomo and Bruno Annibale in Therapeutic Advances in Gastroenterology

Supplemental Material

sj-docx-2-tag-10.1177_17562848251321695 – Supplemental material for Look inside the management of colonic diverticular rebleeding: a systematic review

Supplemental material, sj-docx-2-tag-10.1177_17562848251321695 for Look inside the management of colonic diverticular rebleeding: a systematic review by Marilia Carabotti, Giovanni Marasco, Franco Radaelli, Giovanni Barbara, Rosario Cuomo and Bruno Annibale in Therapeutic Advances in Gastroenterology

Footnotes

Acknowledgements

Technical and editorial assistance was provided by CORESEARCH SRL (Antonio Nicolucci) through an Alfasigma Italy unconditional grant.

Declarations

ORCID iDs

Marilia Carabotti

Franco Radaelli

Rosario Cuomo

Bruno Annibale

Supplemental material

Supplemental material for this article is available online.

References

Peery

Crockett

Murphy

, et al. Burden and cost of gastrointestinal, liver, and pancreatic diseases in the United States: update 2021. Gastroenterology 2022; 162: 621–644.

Stollman

Raskin

. Diverticular disease of the colon. Lancet 2004; 363: 631–639.

Shahedi

Fuller

Bolus

, et al. Long-term risk of acute diverticulitis among patients with incidental diverticulosis found during colonoscopy. Clin Gastroenterol Hepatol 2013; 11: 1609–1613.

Vajravelu

Mamtani

Scott

, et al. Incidence, risk factors, and clinical effects of recurrent diverticular hemorrhage: a large cohort study. Gastroenterology 2018; 155: 1416–1427.

Poncet

Heluwaert

Voirin

, et al. Natural history of acute colonic diverticular bleeding: a prospective study in 133 consecutive patients. Aliment Pharmacol Ther 2010; 32: 466–471.

Sato

Aoki

Sadashima

, et al. Long-term risks of recurrence after hospital discharge for acute lower gastrointestinal bleeding: a large nationwide cohort study. Clin Gastroenterol Hepatol 2023; 21: 3258–3269.

Niikura

Yasunaga

Yamaji

, et al. Factors affecting in-hospital mortality in patients with lower gastrointestinal tract bleeding: a retrospective study using a national database in Japan. J Gastroenterol 2015; 50: 533–540.

Strate

Ayanian

Kotler

, et al. Risk factors for mortality in lower intestinal bleeding. Clin Gastroenterol Hepatol 2008; 6: 1004–1010.

Camilleri

Sandler

Peery

. Etiopathogenetic mechanisms in diverticular disease of the colon. Cell Mol Gastroenterol Hepatol 2020; 9: 15–32.

10.

Jensen

. Endoscopic diagnosis and treatment of colonic diverticular bleeding. Gastrointest Endosc Clin N Am 2024; 34: 345–361.

11.

Pallotta

Vona

Maselli

, et al. Oxidative imbalance and muscular alterations in diverticular disease. Dig Liver Dis 2022; 54: 1186–1194.

12.

Page

Moher

Bossuyt

, et al. PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ 2021; 372: n160.

13.

Gonai

Toya

Kudara

, et al. Risk factors for short-term re-bleeding in patients with colonic diverticular bleeding: a multicenter retrospective study. Scand J Gastroenterol 2022: 57: 690–695.

14.

Gonai

Toya

Kawasaki

, et al. Risk factors of re-bleeding within a year in colonic diverticular bleeding patients. DEN Open 2021; 2: e22.

15.

Sato

Yasuda

Nakamoto

, et al. Risk factors for late rebleeding of colonic diverticular bleeding in elderly individuals. J Anus Rectum Colon 2021; 5: 48–157.

16.

Jalil

Gorski

Jalil

, et al. Factors associated with diverticular bleeding and re-bleeding: a United States hospital study. North Clin Istanb 2018; 6: 248–253.

17.

Kitagawa

Katayama

Kobori

, et al. Predictors of 90-day colonic diverticular recurrent bleeding and readmission. Intern Med 2019; 58: 2277–2282.

18.

Vajravelu

Mamtani

Scott

, et al. Incidence, risk factors, and clinical effects of recurrent diverticular hemorrhage: a large cohort study. Gastroenterology 2018; 155: 1416–1427.

19.

Taki

Oshima

Tozawa

, et al. Analysis of risk factors for colonic diverticular bleeding and recurrence. Medicine (Baltimore) 2017; 96: e8090.

20.

Lorenzo

Gallois

Lahmek

, et al. Middle-term mortality and re-bleeding after initial diverticular bleeding: a nationwide study of 365 mostly elderly French patients. United European Gastroenterol J 2017; 5: 119–127.

21.

Joaquim

Caldeira

Antunes

, et al. Risk factors for severity and recurrence of colonic diverticular bleeding. Rev Esp Enferm Dig 2017; 109: 3–9.

22.

Nagata

Niikura

Aoki

, et al. Impact of discontinuing non-steroidal antiinflammatory drugs on long-term recurrence in colonic diverticular bleeding. World J Gastroenterol 2015; 214: 1292–1298.

23.

Watanabe

Kodera

Shimura

. Investigation of risk factors predicting recurrence of colonic diverticular hemorrhage and development of a recurrence risk score. Acta Gastroenterol Belg 2014; 77: 408–412.

24.

Fujino

Inoue

Onodera

, et al. Risk factors for early re-bleeding and associated hospitalization in patients with colonic diverticular bleeding. Colorectal Dis 2013; 15: 982–986.

25.

Niikura

Nagata

Yamada

, et al. Recurrence of colonic diverticular bleeding and associated risk factors. Colorectal Dis 2012; 14: 302–305.

26.

Tanaka

Motomura

Akahoshi

, et al. Predictive factors for colonic diverticular rebleeding: a retrospective analysis of the clinical and colonoscopic features of 111 patients. Gut Liver 2012; 6: 334–338.

27.

Okamoto

Watabe

Yamada

, et al. The association between arteriosclerosis related diseases and diverticular bleeding. Int J Colorectal Dis 2012; 27: 1161–1166.

28.

Jensen

Machicado

Jutabha

, et al. Urgent colonoscopy for the diagnosis and treatment of severe diverticular hemorrhage. N Engl J Med 2000; 3422: 78–82.

29.

Mizuki

Tatemichi

Nakazawa

, et al. Long-term clinical course after conservative and endoscopic treatment of colonic diverticular bleeding. Digestion 2016; 94: 186–191.

30.

Gobinet-Suguro

Nagata

Kobayashi

, et al. Treatment strategies for reducing early and late recurrence of colonic diverticular bleeding based on stigmata of recent hemorrhage: a large multicenter study. Gastrointest Endosc 2022; 95:1 210–22.e12.

31.

Aoki

Sadashima

Kobayashi

, et al. High risk stigmata and treatment strategy for acute lower gastrointestinal bleeding: a nationwide study in Japan. Endoscopy 2024; 56: 291–301.

32.

Nagata

Niikura

Ishii

, et al. Cumulative evidence for reducing recurrence of colonic diverticular bleeding using endoscopic clipping versus band ligation: systematic review and meta-analysis. J Gastroenterol Hepatol 2021; 36: 1738–1743.

33.

Kishino

Nagata

Kobayashi

, et al. Endoscopic direct clipping versus indirect clipping for colonic diverticular bleeding: a large multicenter cohort study. United European Gastroenterol J 2022; 10: 93–103.

34.

Chan

Leung Ki

Wong

, et al. Risks of bleeding recurrence and cardiovascular events with continued aspirin use after lower gastrointestinal hemorrhage. Gastroenterology 2016; 151: 271–277.

35.

Sengupta

Feuerstein

Patwardhan

, et al: The risks of thromboembolism vs. recurrent gastrointestinal bleeding after interruption of systemic anticoagulation in hospitalized in patients with gastrointestinal bleeding: a prospective study. Am J Gastroenterol 2015; 102: 328–335.

36.

Witt

Delate

Garcia

, et al. Risk of thromboembolism, recurrent hemorrhage, and death after warfarin therapy interruption for gastrointestinal tract bleeding. Arch Intern Med 2012; 17219: 1484–1491.

37.

Qureshi

Mittal

Patsias

, et al. Restarting anticoagulation and outcomes after major gastrointestinal bleeding in atrial fibrillation. Am J Cardiol 2014; 1134: 662–668.

38.

Chai-Adisaksopha

Hillis

Monreal

, et al. Thromboembolic events, recurrent bleeding and mortality after resuming anticoagulant following gastrointestinal bleeding. A meta-analysis. Thromb Haemost 2015; 1144: 819–825.

39.

Tapaskar

Pang

Werner

, et al. Resuming anticoagulation following hospitalization for gastrointestinal bleeding is associated with reduced thromboembolic events and improved mortality: results from a systematic review and meta-analysis. Dig Dis Sci 2021; 66: 554–656.

40.

Chen

Hsu

, et al. Gastrointestinal hemorrhage in warfarin anticoagulated patients: incidence, risk factor, management, and outcome. BioMed Res Int 2014: 463767.

41.

Majeed

Wallvik

Eriksson

, et al. Optimal timing of vitamin K antagonist resumption after upper gastrointestinal bleeding. A risk modelling analysis. Thromb Haemost 2017; 117: 491–499.

42.

Sengupta

Marshall

Jones

, et al. Rebleeding vs thromboembolism after hospitalization for gastrointestinal bleeding in patients on direct oral anticoagulants. Clin Gastroenterol Hepatol 2018; 16: 1893–1900.

43.

Kido

. Concise Commentary: whether and when to resume anticoagulation after gastrointestinal bleeding. Dig Dis Sci 2021; 66: 567.

44.

Candeloro

van Es

Cantor

, et al. Recurrent bleeding and thrombotic events after resumption of oral anticoagulants following gastrointestinal bleeding: Communication from the ISTH SSC subcommittee on control of anticoagulation. J Thromb Haemost 2021; 19: 2618–2628.

45.

Hreinsson

Palsdottir

Bjornsson

. The association of drugs with severity and specific causes of acute lower gastrointestinal bleeding: a prospective study. J Clin Gastroenterol 2016; 50: 408–413.

46.

Hreinsson

Gumundsson

Kalaitzakis

, et al. Lower gastrointestinal bleeding: incidence, etiology, and outcomes in a population-based setting. Eur J Gastroenterol Hepatol 2013; 25: 37–43.

47.

Hreinsson

Bjarnason

Bjornsson

. The outcome and role of drugs in patients with unexplained gastrointestinal bleeding. Scand J Gastroenterol 2015; 50: 1482–1489.

48.

Strate

Liu

Aldoori

, et al. Obesity increases the risks of diverticulitis and diverticular bleeding. Gastroenterology 2009; 136: 115–e1.

49.

Strate

Singh

Boylan

, et al. A prospective study of alcohol consumption and smoking and the risk of major gastrointestinal bleeding in men. PLoS One 2016; 11: e0165278.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.01 MB