Small Bowel Lesions and Bleeding Risk in Hemodialysis Patients: A Narrative Review

ABSTRACT

Background

Gastrointestinal bleeding is very common among hemodialysis patients. This high bleeding risk is caused by uremic platelet dysfunction, vascular fragility, intradialytic hemodynamic instability, and widespread antithrombotic therapy. Small bowel lesions, especially angiodysplasias, constitute a significant but often overlooked source of recurrent or occult hemorrhage while conventional endoscopy frequently fails to identify these lesions.

Objectives

Investigation of small bowel lesions in hemodialysis patients and identification of bleeding risk of this vulnerable population.

Design

Narrative review.

Methods

English language studies in the last 25 years using PubMed and Google Scholar databases up to November 2025. Search terms included “hemodialysis,” “small bowel lesions,” “angiodysplasia,” and “gastrointestinal bleeding.” The review incorporates different types of research including observational cohorts, randomized trials, cross‐sectional studies, systematic reviews, and narrative reviews involving hemodialysis adult or end stage renal disease populations.

Results

Across the analysis of 26 studies, capsule endoscopy revealed a significantly higher prevalence of small bowel lesions in hemodialysis patients compared to controls, with angiodysplasias being the most frequent finding. Large‐scale epidemiological analyses identified hemodialysis as an independent risk factor for gastrointestinal hemorrhage compared to peritoneal dialysis, providing the clinical background for the increased susceptibility of these patients to small bowel‐specific lesions. Overall, patients have a higher risk of rebleeding tendency than controls and exhibited a high‐risk phenotype characterized by recurrent angiodysplasia‐related bleeding and elevated one‐year mortality rates following the first bleeding episode.

Conclusion

Small bowel lesions may constitute a major and underrecognized cause of gastrointestinal bleeding in hemodialysis patients. Capsule endoscopy offers the highest diagnostic yield, yet optimal treatment strategies remain undefined, and recurrent bleeding episodes are common. Thus, effective management requires early detection, individualized therapeutic planning, and careful treatment with anticoagulant and antiplatelet drugs.

1. Introduction

Hemodialysis patients are at increased risk for gastrointestinal bleeding, which remains a major cause of morbidity and mortality in this population [1] and it is an issue, which continues to challenge nephrologists. Multiple factors contribute to this situation, including platelet dysfunction, uremic toxins, anticoagulant and antiplatelet therapy, and vascular changes related to chronic kidney disease (CKD) associated with uremic vasculopathy and capillary fragility characteristic of end‐stage renal disease patients [2, 3, 4, 5]. These pathophysiologic mechanisms can lead to mucosal injury throughout the gastrointestinal tract, with the small intestine being a particularly challenging site for diagnosis and management because it is difficult to approach [6]. Gastrointestinal bleeding is one of the most reported causes of mortality in CKD and end‐stage renal disease accounting for 3%–7% mortality of hemodialysis patients [7]. While anemia in this population is typically normocytic due to erythropoietin deficiency, concomitant iron deficiency frequently arises from hemodialysis‐related blood loss or occult gastrointestinal bleeding. Consequently, patients exhibiting refractory iron deficiency despite replacement or sudden hemoglobin drops warrant a comprehensive workup, including iron studies and prompt evaluation for gastrointestinal bleeding sources. Several studies have demonstrated that obscure gastrointestinal bleeding occurs more frequently in hemodialysis patients and accounts for a considerable proportion of mortality within this population [8, 9].

Among small bowel lesions observed in hemodialysis patients, angiodysplasias/diminutive angiectasias, erosions—ulcers and red spots lesions are the most frequently reported. Angiodysplasias are abnormally dilated blood vessels with a diameter of less than 1 cm, in most cases, and represent the most common vascular malformation that may cause recurrent or occult gastrointestinal bleeding. Prevalence of angiodysplasia as a cause of hemorrhage in chronic renal disease ranges from 19% to 32%, and they are responsible for approximately 5% of all gastrointestinal bleeding cases, particularly among older adults, and for 60% of cases of small bowel hemorrhage [10, 11]. The rising trend in gastrointestinal bleeding hospitalizations cannot be fully explained by improved diagnostic capabilities alone. It is plausible that the increasing prevalence of CKD probably plays a significant role. This issue is further compounded by the widespread and escalating use of antiplatelet and anticoagulant therapies. Notably, Medicare spending on oral anticoagulants increased 16‐fold between 2011 and 2019, and aspirin use more than doubled from 2000 to 2021 [12, 13]. Although these data refer to the general population, this trend heavily impacts hemodialysis patients, given that atrial fibrillation affects approximately 20%–30% of this group [14]. Consequently, the widespread popularity of direct oral anticoagulants may set new difficulties, because there are no established protocols for monitoring these drugs in CKD, negatively influencing the bleeding risk in this vulnerable population. Erosions and ulcers, on the other hand, may stem from different causes like ischemic injury, inflammation, or medication‐related toxicity [15]. With an incidence rate 19.4–57/100,000 per year complicated by hemorrhage, may lead to hospitalization [16]. Finally, red spots are common among patients but their clinical significance is not fully understood. Unfortunately, much remains uncertain about red spots regarding their frequency and bleeding potential. Despite their clinical importance, the detection of all these lesions remains challenging, as they often escape conventional upper and lower endoscopy, because small bowel is difficult to be approached, and require specialized modalities such as capsule endoscopy for visualization and investigation [17].

Capsule endoscopy is a well‐established, non‐invasive method for assessing disorders of the small bowel. In patients with obscure small bowel bleeding, selecting appropriate candidates and performing capsule endoscopy soon after the latest hemorrhage episode may enhance its diagnostic effectiveness. According to recent European Society of Gastrointestinal Endoscopy ESGE guidelines, physicians should consider capsule endoscopy as a first‐line procedure for small bowel evaluation after excluding upper and lower gastrointestinal bleeding sources [18]. Furthermore, gastrointestinal bleeding further deteriorates the already low quality of life in hemodialysis patients and imposes a substantial financial strain on healthcare systems, as these individuals often undergo repeated hospital admissions, diagnostic evaluations, and blood transfusions. Therefore, beyond achieving the initial diagnosis, attention should also be given to the risk of rebleeding. De Sousa Magalhães et al. reported the use of the Rhemitt score to estimate and predict the risk of small bowel rebleeding after capsule endoscopy, incorporating factors like renal or heart failure, major bleeding, smoking, endoscopic findings, and endoscopic intervention. This index enables clinicians to focus on rebleeding risk patients and identify who may benefit from closer follow‐up [19, 20].

The aim of the present review is to summarize current evidence regarding the epidemiology, diagnostic approaches, and therapeutic strategies for small bowel lesions in patients undergoing hemodialysis, and to highlight the existing gap in knowledge that warrants further investigation.

2. Methods

An extensive literature search was conducted in the PubMed and Google Scholar databases up to November 2025. The following search terms were used, individually and in combination with Boolean operators (AND, OR): “hemodialysis,” “small bowel lesions,” “angiodysplasias,” “obscure gastrointestinal bleeding.”

Different types of studies were reviewed, including observational retrospective or prospective studies, systematic reviews of randomized controlled trials, and narrative reviews over the past 25 years. Inclusion criteria comprised studies involving human subjects, published in English, and addressing one or more of the following aspects:

• The prevalence or incidence of small bowel lesions in hemodialysis patients.

• Pathophysiological associations between dialysis and gastrointestinal mucosal injury.

Studies were excluded if they involved animal models, pediatric population (defined as patients < 18 years of age), consisted solely of isolated case reports without confirmed diagnoses, or lacked full‐text availability. Reference lists of relevant articles were also manually screened to identify additional publications.

This review was designed as a narrative review, aiming to synthesize and compare available data rather than perform a quantitative meta‐analysis. Two reviewers independently screened the titles and abstracts.

3. Results

3.1. Gastrointestinal Lesions in Hemodialysis Patients

3.1.1. Prevalence and Characteristics of Small Bowel Lesions (Capsule Endoscopy Findings)

Multiple observational studies utilizing capsule endoscopy have consistently demonstrated a significantly higher prevalence of small bowel pathology in hemodialysis patients compared to non‐hemodialysis controls (Table 1). The diagnostic yield of capsule endoscopy in this population is substantial; Nakajima et al. and Ohmori et al. reported that erosive and vascular lesions were significantly more frequent in hemodialysis patients than in controls. Specifically, vascular lesions—predominantly angiodysplasias—were identified as the most common pathological finding in patients with obscure gastrointestinal bleeding, occurring in up to 61.5% of cases [21, 22].

TABLE 1.

Studies investigating small bowel lesions in hemodialysis patients using capsule endoscopy.

Name of the article and first author	Journal and years of publication	Type of study	Type/number of patients and number of HD	Conclusions of the study	Main limitations
Investigation of Small Intestinal Lesions in Dialysis Patients Using Capsule Endoscopy [21] Nakajima et al.	Hemodialysis International 2019	Cross sectional study	HD and non‐HD patients with fecal occult blood‐positive taking CE N = 36 (HD 16)	Erosive and vascular lesions were significantly more prevalent in HD patients compared to non‐HD controls (50% vs. 25% and 62.5% vs. 25%, respectively).	Small number of patients
Abnormalities of the Small Intestine Detected by Capsule Endoscopy in Hemodialysis Patients [22] Ohmori et al.	Internal Medicine 2012	Cohort study	Patients with progressive anemia N = 90 (HD = 13)	Vascular lesions were the predominant finding in HD patients (61.5%) compared to ulcerative lesions (30.7%).	Small number of HD patients
Small Intestinal Lesions in Patients with Anemia on Hemodialysis: A Two‐Center, Cross‐Sectional Study [23] Ozaki et al.	Digestion 2020	Cross‐sectional study	HD patients with anemia N = 39 (HD = 39)	High prevalence of small intestinal abnormalities (angiodysplasia and erosions) contributes to chronic anemia in HD patients.	Lack of a control group Direct comparison with non‐HD patients was not possible
Capsule Endoscopy in Hemodialysis Versus Non‐Hemodialysis Patients With Suspected Small Bowel Bleeding: a Prospective Cross‐Sectional Study [6] Smyrlis et al.	Hippokratia 2024	Prospective cross‐sectional study	100 patients with OGIB undergoing CE N = 100 (HD = 25) (non‐HD = 75)	HD patients exhibited a higher burden of vascular lesions and an increased risk of rebleeding based on RHEMITT score.	Small number of HD patients Single center study
Wireless Capsule Endoscopy in the Investigation of Patients With Chronic Renal Failure and Obscure Gastrointestinal Bleeding (Preliminary Data) [24] Karagiannis et al.	World Journal of Gastroenterology 2006	Prospective study	Compare 17 patients with advanced CKD, including some on HD, and compared them with 51 non‐CKD patients with OGIB N = 67 (HD = 4)	SB angiodysplasia has an increased prevalence among CKD patients with obscure GI bleeding.	Very small number of HD patients Single center study
Use of Small Bowel Capsule Endoscopy in Patients With Chronic Kidney Disease: Experience From a University Referral Center [8] Docherty et al.	Annals of Gastroenterology 2015	Retrospective study	Patients with CKD and anemia or OGIB N = 69 (HD = 4)	SB angiectasias were the most frequent pathological finding (33%). Fecal calprotectin did not predict positive findings.	Retrospective study Very small number of HD patients The difference in group sizes between CKD 3 and CKD 4/5
Cross‐Sectional Small Intestinal Surveillance of Maintenance Hemodialysis Patients Using Video Capsule Endoscopy: SCHEMA Study [25] Hosoe et al.	Endoscopy 2016	Prospective cross‐sectional study	Asymptomatic HD patients N = 56 (HD = 56)	High prevalence of small bowel abnormalities (64.8%) in asymptomatic patients, though the incidence of acute bleeding events was low during 1‐year follow‐up.	Comorbid conditions in HD patients were not captured, since they were asymptomatic
Comparison of Risk Factors Between Small Intestinal Ulcerative and Vascular Lesions in Occult Versus Overt Obscure Gastrointestinal Bleeding [26] Okamoto et al.	Digestive Diseases and Sciences 2016	Retrospective cohort study	390 OGIB cases (occult, n = 101, overt, n = 289) N = 390 (HD = 22)	Low hemoglobin, hematologic disease, and liver cirrhosis were linked to vascular lesions in occult OGIB, whereas hemodialysis and cirrhosis were associated with vascular lesions in overt OGIB.	Retrospective cohort study; Comparisons restricted to OGIB subtypes Small number of HD patients
Characteristics of the Small Bowel Lesions Detected by Capsule Endoscopy in Patients With Chronic Kidney Disease [27] Kawamura et al.	Gastroenterology Research and Practice 2013	Retrospective cross sectional study	Compared the incidence of the small‐bowel lesions among two groups Ν = 42 (HD = 19)	SB vascular lesions were found at a higher incidence in CKD patients than in non‐CKD patients.	Retrospective study Small number of HD patients
Long‐Term Prognosis of Patients With Obscure Gastrointestinal Bleeding: A Retrospective Cohort Study [28] Arai et al.	Digestion 2019	Retrospective cohort	Patients with OGIB who underwent CE N = 240 (HD = 13)	Mortality in OGIB patients is frequently associated with severe comorbidities rather than GI hemorrhage itself especially those who were elderly or who had liver cirrhosis.	Retrospective study Small number of HD patients
Gastrointestinal Angiectasia in Patients With Chronic Kidney Disease: A Matched Case‐Control Study [29] Azancot et al.	Clinics and Research in hepatology and gastroenterology 2024	Retrospective study	CDK patients with an endoscopic diagnosis of at least one GI angiectasia lesion N = 55 (HD = 17)	CKD patients with GI angiectasia exhibit a distinct cardiovascular profile, including higher rates of valvular disease and anticoagulant use.	Retrospective study Single center Small number of HD patients
Diagnostic and Therapeutic Considerations for Obscure Gastrointestinal Bleeding in Patients With Chronic Kidney Disease [30] Muftah et al.	Annals of Gastroenterology 2018	Review	Review of diagnostic and therapeutic strategies	The role of endoscopic therapies (for accessible lesions, and pharmacological approaches—including octreotide and thalidomide too).

Abbreviations: CKD: chronic kidney disease, GI: gastrointestinal, HD: hemodialysis, OGIB: obscure gastrointestinal bleeding, PPIs: proton pump inhibitors.

In addition, Ozaki et al. assessed the association between small bowel endoscopic findings and anemia in hemodialysis patients. Capsule endoscopy was successfully completed in 39 hemodialysis patients, revealing a high incidence of small intestinal lesions, such as angiodysplasia and erosions. These findings suggested that small bowel abnormalities may contribute to chronic anemia in hemodialysis population even in the absence of visible bleeding. The authors underscore the importance of evaluating the small bowel in hemodialysis patients with unexplained anemia [23].

Recent data reinforce these findings. In 2024, we conducted a prospective cross‐sectional study, which confirmed that hemodialysis patients (N = 25) with suspected small bowel bleeding exhibited higher rates of red spots (96%), angiodysplasias (32%), and erosions compared to patients without CKD (N = 75). Furthermore, this study highlighted a higher risk of rebleeding in the hemodialysis group, as assessed by the RHEMITT score [6]. These results support the concept that uremia and dialysis‐related factors predispose to small‐bowel mucosal injury and occult bleeding.

Notably, Karagiannis et al. conducted a prospective study (among 67 patients with obscure gastrointestinal bleeding, of whom 4 were on hemodialysis) using capsule endoscopy in 17 patients with advanced CKD, including some on hemodialysis, and compared them with 51 non‐CKD patients with obscure gastrointestinal bleeding. They found a significantly higher diagnostic yield in the CKD group (70.6% vs. 41.2%, p < 0.05), with angiodysplasia detected in 47% of chronic renal failure patients versus 17.6% of controls. Univariate analysis identified CKD as a significant independent predictor of venous malformations (OR = 4.1). Therapeutic interventions based on capsule endoscopy findings were applied in about two‐thirds of CKD patients, including argon plasma coagulation for angiodysplasia, pharmacologic therapy, or withdrawal of NSAIDs. The authors conclude that capsule endoscopy is both safe and clinically useful in CKD patients, and should be integrated into the evaluation of obscure gastrointestinal bleeding in this high‐risk group [24].

Additionally, in a retrospective study (March 2005–August 2012) from a university referral center, Docherty et al. evaluated the diagnostic yield of small‐bowel capsule endoscopy in patients with chronic kidney disease. Among 69 CKD patients (including 4 on hemodialysis), mostly stage 3, the main indication was unexplained anemia or obscure gastrointestinal bleeding. The overall diagnostic yield was modest, with small‐bowel angiectasias representing the most frequent finding, while malignant or inflammatory lesions were rare. Fecal calprotectin did not predict positive findings. The authors concluded that although small bowel capsule endoscopy can identify vascular lesions in CKD, its yield for clinically significant pathology is limited, and non‐enteric causes of anemia should be carefully considered before referral [6]. Another prospective study, by using capsule endoscopy, investigated the prevalence of small bowel disease among hemodialysis patients. Although asymptomatic maintenance hemodialysis patients had a high prevalence of small bowel abnormalities (64.8%), they did not have a high incidence of small bowel‐associated events during the 1 year follow‐up [25].

While angiodysplasias are the hallmark finding, erosive and ulcerative lesions are also frequent. Okamoto et al. conducted a retrospective study between January 2005 and March 2011, examining 390 patients (including 22 on hemodialysis) with obscure gastrointestinal bleeding to identify risk factors for ulcerative and vascular lesions in the small intestine. Using capsule and balloon endoscopy, the researchers found that ulcerative lesions were more common in cases of occult small bowel bleeding, while vascular lesions were more frequent in overt small bowel bleeding. Non‐steroidal anti‐inflammatory drug use was a significant risk factor for ulcerative lesions in overt bleeding cases. Risk factors appear to vary by lesion type. Okamoto et al. noted that non‐steroidal anti‐inflammatory drug use was associated with ulcerative lesions in overt gastrointestinal bleeding, whereas hemodialysis itself was strongly linked to vascular lesions in cases of obscure gastrointestinal bleeding [26].

Moreover, in a cross‐sectional observational study, Kawamura et al. used capsule endoscopy to systematically evaluate mucosal lesions in patients with chronic kidney disease, including those undergoing hemodialysis. They demonstrated a significantly higher incidence of vascular lesions, particularly angiodysplasias, in CKD patients compared with matched non‐CKD controls. These findings highlight the underestimated burden of microvascular small bowel pathology in the CKD/hemodialysis population and its potential role in recurrent or occult gastrointestinal bleeding. The study emphasized the superiority of capsule endoscopy over conventional endoscopic techniques in detecting these subclinical vascular abnormalities, supporting its inclusion in the diagnostic approach for hemodialysis patients with suspected small bowel bleeding [27].

Regarding long‐term outcomes, Arai et al. investigated the long‐term prognosis of patients undergoing capsule endoscopy for obscure gastrointestinal bleeding. In their retrospective cohort, they observed that while rebleeding rates were notable, the overall mortality was mainly due to comorbidities, such as malignancies and cardiovascular diseases, rather than the gastrointestinal hemorrhage itself. Although hemodialysis was identified as a risk factor for poor prognosis in univariate analysis, it did not remain significant in the multivariate model, suggesting that the severe underlying conditions this population have play a more critical role in survival than the bleeding events alone [28].

Also, in a recent matched case–control analysis, Azancot et al. focused on the specific profile of gastrointestinal angiectasia within the CKD population. By matching CKD patients with angiectasia against the control group, they identified a clear “cardiovascular phenotype.” Specifically, the data showed that angiectasia is closely linked to valvular heart disease, arrhythmias, and anticoagulant use, a finding that supports the theory which connects uremic angiodysplasia with acquired von Willebrand syndrome. Consequently, the authors suggest that cardiovascular burden is a more potent predictor of angiectasia than the stage of renal failure itself [29]. However, as summarized by Muftah et al., while diagnostic modalities like capsule endoscopy have improved lesion detection, the management of recurrent bleeding remains challenging. The authors advocate for a multimodal approach, combining endoscopic intervention with pharmacological therapies like thalidomide for refractory angiodysplasias, although caution is advised due to renal clearance issues [30].

3.1.2. Patterns of Upper and Lower GI Bleeding (Conventional Endoscopy)

Since small bowel bleeding (obscure gastrointestinal bleeding) is defined by negative findings on conventional endoscopy, understanding the patterns of upper and lower gastrointestinal lesions in hemodialysis patients is essential for context and differential diagnosis. Regarding upper gastrointestinal bleeding, erosive and ulcerative lesions remain the predominant causes. A 5‐year follow‐up study indicated that upper gastrointestinal bleeding affected about 21% of hemodialysis patients. Nakayama et al. noted that these ulcers and erosions often develop even without Helicobacter pylori infection. On the other hand, lower gastrointestinal bleeding was less common but associated with significantly higher mortality (71.4%) and was strongly associated with non‐steroid anti‐inflammatory drugs (NSAID) or aspirin use. These patterns highlight that despite preventive efforts, the risk and prognosis remain poor in chronic hemodialysis patients [31]. Laeeq et al. reported that approximately half of hemodialysis patients presenting with upper gastrointestinal bleeding required endoscopic therapeutic intervention, with mortality being significantly higher in those who did not receive early treatment. They identified specific predictors for needing endoscopic therapeutic intervention, such as the combination of hematemesis and melena, ulcerative lesions, or coexisting chronic liver disease. Mortality was higher in patients who did not undergo endoscopic therapeutic intervention, underscoring the clinical importance of timely intervention and highlighting the distinct hemorrhage profile and elevated procedural needs in this high‐risk population [32].

Interestingly, routine pre‐transplant endoscopy has revealed pathological findings in up to 50.6% of asymptomatic hemodialysis patients (Pakfetrat et al.), suggesting a high prevalence of silent mucosal damage. The most frequent lesions were mild gastritis (35.6%) and gastro‐esophageal reflux disease (16.7%). Ulcers accounted for 11% of cases, with duodenal ulcers being most common (6.8%). The authors concluded that, given the high prevalence of gastrointestinal lesions, even among asymptomatic candidates, endoscopic evaluation of upper gastrointestinal should be considered as part of pre‐transplant work‐up in hemodialysis patients, especially those with risk factors for development of ulcers [33].

Regarding lower gastrointestinal bleeding, angiodysplasia is a frequent source, particularly in the right colon (≈31.8%). In 36.3% of cases, multiple lesions were found in combinations of stomach, duodenum, and right colon. Zajjari et al. demonstrated that endoscopic treatment with Argon Plasma Coagulation was effective in achieving hemostasis, although multiple sessions were often required. The study highlights angiodysplasia as a common source of gastrointestinal hemorrhage in hemodialysis patients, emphasizing the utility of endoscopic coagulation as a safe and effective therapeutic approach [11].

Moreover, a critical recent finding regarding procedural safety comes from Kang et al., who demonstrated that the timing of endoscopy relative to the hemodialysis session significantly impacts outcomes. Procedures performed on the same day as dialysis were associated with a threefold increase in mortality risk compared to those performed 1–2 days after dialysis, highlighting the need for careful scheduling. Patients in the same‐day group were more often inpatients and were frequently treated for gastrointestinal bleeding or anemia, indicating higher baseline illness severity. Interestingly, the study did not find significant differences in the use of hemodynamic support during the procedures [34].

Marinescu et al. conducted a retrospective descriptive study of 30 CKD patients admitted for upper gastrointestinal bleeding, of whom 16 (53.3%) were on hemodialysis. Most patients presented with melena (70%) and all were anemic at admission. Only five patients underwent upper endoscopy, which revealed erosive gastritis, duodenitis, and a bleeding duodenal ulcer in one patient; two had no visible lesion. Management was entirely conservative, primarily using proton pump inhibitors, and none required endoscopic or surgical intervention. The authors concluded that although CKD patients—particularly those on hemodialysis—are at increased risk of upper gastrointestinal bleeding, such episodes are often mild and respond well to conservative therapy [35]. The studies described above are presented in the Table 2.

TABLE 2.

Studies investigating small bowel lesions in hemodialysis patients using conventional endoscopy.

Name of the article and first author	Journal and years of publication	Type of study	Type/number of patients and number of HD	Conclusions of the study	Main limitations
Gastrointestinal Bleeding Among 151 Patients Undergoing Maintenance Hemodialysis for End‐Stage Renal Failure: A 5‐Year Follow‐Up Study [31] Nakayama et al.	Medicine (Baltimore) 2024	Retrospective follow‐up study	151 HD patients, 51 with upper GI bleeding, 7 with lower bleeding, 119 without bleeding N = 151 (HD = 151)	Annual incidence of GI bleeding was 4.2%. Ulcers were frequent even in H. pylori ‐negative patients. Prophylactic PPI's did not significantly prevent bleeding.	Retrospective study No assessment for small bowel
Upper Gastrointestinal Bleeding in Patients with End Stage Renal Disease: Causes, Characteristics and Factors Associated With Need for Endoscopic Therapeutic Intervention [32] Laeeq e al.	Journal of Translational Internal Medicine 2017	Prospective study	59 patients with UGIB and ESRD who presented in endoscopy unit N = 59 (HD = 59)	Endoscopic therapeutic intervention was required in 55.9% of cases. Mortality was significantly higher in patients who did not receive timely endoscopic hemostasis.	Small number of patients Single center study
Endoscopic Findings in Hemodialysis Patients Upon Workup for Kidney Transplantation [33] Pakfetrat et al.	Saudi Journal of Kidney Diseases Transplantation 2020	Retrospective review	Endoscopic findings in hemodialysis patients upon workup for kidney transplantation Ν = 1256 (HD 1256)	High prevalence of mucosal lesions (50.6%) in asymptomatic patients suggests the need for pre‐transplant endoscopic screening.	Retrospective Single center study No assessment for SB Administrative databases may underreport/misclassify
Relationship Between Timing of Hemodialysis and Procedure‐Related Complications Following Endoscopy in Patients With End‐Stage Renal Disease [34] Kang et al.	Digestive Diseases and Science 2025	Retrospective cohort study	191 ESRD patients undergoing 252 endoscopic procedures Ν = 191 (HD 191)	Endoscopy performed on the same day as hemodialysis session is associated with a threefold increase in mortality risk.	Retrospective study No assessment for small bowel
Gastrointestinal Bleeding due to Angiodysplasia in Patients on Hemodialysis: A Single‐Center Study [11] Zajjari et al.	Saudi Journal of Kidney Diseases Transplantation 2016	Retrospective study	HD patients who diagnosed with angiodysplasia N = 22 (HD 22)	Angiodysplasia is a frequent source of GI bleeding. Endoscopic coagulation as a safe and effective therapeutic approach.	Small number of patients Single center study Retrospective
Upper Gastrointestinal Bleeding in Chronic Kidney Disease Patients [35] Marinescu et al.	Current Health Sciences Journal 2016	Retrospective study	30 CKD patients admitted and treated for upper GI bleeding N = 30 (HD = 16)	Although CKD patients have a high risk of upper GI bleeding, the conservative treatment applied has a very high rate of success.	Retrospective study Single center study No assessment for SB

Abbreviations: CKD: chronic kidney disease, ESDR: end stage renal disease, GI: gastrointestinal, HD: hemodialysis, PD: peritoneal dialysis, PPIs: proton pump inhibitors, SB: small bowel, UGIB: upper gastrointestinal bleeding.

3.1.3. Epidemiological Data and Risk Factors Based on Registries

Large‐scale cohort studies confirm that hemodialysis is an independent risk factor for gastrointestinal bleeding (Table 3). Moreover, in a retrospective cohort study comparing 417 hemodialysis patients with 14,034 non‐hemodialysis patients over a mean follow‐up of 3.5 years, to evaluate the incidence of upper and lower gastrointestinal bleeding. The cumulative gastrointestinal bleeding incidence in hemodialysis patients was 4.44% at 1 year, 7.15% at 3 years and 10.40% at 5 years, compared with 2.35%, 2.98%, and 3.79%, respectively, in the non‐hemodialysis group. Hemodialysis was independently associated with increased hemorrhage risk [hazard ratio (HR) = 1.67]. The most common bleeding sources identified were gastric ulcers (upper gastrointestinal) and colonic diverticular bleeding (lower gastrointestinal). Summarized, patients on hemodialysis have a significantly higher risk of gastrointestinal bleeding and warrant careful monitoring for both upper and lower gastrointestinal bleeding events [36].

TABLE 3.

Epidemiological data and risk factors based on registries.

Name of the article and first author	Journal and years of publication	Type of study	Type/number of patients and number of HD	Conclusions of the study	Main limitations
Natural History of Upper and Lower Gastrointestinal Bleeding in Hemodialysis Patients: A Dual‐Center Long‐Term Cohort Study [36] Niikura et al.	Journal Gastroenterology and Hepatology 2021	Retrospective cohort	HD patients and non‐HD patients who had GI bleeding N = 14.451 (417 HD)	Cumulative 5‐year GI bleeding incidence was 10.4% in HD vs. 3.79% in controls. HD was identified as an independent risk factor (HR = 1.67).	Retrospective study No assessment of small bowel
In‐Hospital Mortality Risk Estimation in Patients With Acute Nonvariceal Upper Gastrointestinal Bleeding Undergoing Hemodialysis: A Retrospective Cohort Study [37] Weng et al.	Renal Failure 2013	Retrospective cohort	HD patients diagnosed with UGIB N = 322 (HD = 322)	High 1‐year mortality (27%). Hypoalbuminemia, hospitalization, infection, and single episodic bleeding were identified as independent predictors of all‐cause mortality.	Retrospective study Νo assessment for SB
Gastrointestinal Bleeding in Patients With Renal Failure Under Hemodialysis Treatment: A Single‐Center Experience [38] Can et al.	International Urology and Nephrology 2017	Retrospective study	51 uremic bleeding patients N = 51 (HD = 51)	HD patients presenting with GI bleeding should be evaluated regarding use of prescriptions and efforts should be done.	Retrospective study No assessment of SB Small number of HD and non‐HD patients
Gastrointestinal Bleeding in Patients on Long‐Term Dialysis [39] Trivedi et al.	Journal of Nephrology 2015	Retrospective cohort study	N = 406.836 (HD 374.891) (PD 31.945)	GI bleeding represents an increasingly significant complication among HD patients.	Administrative databases, which may underreport or misclassify gastrointestinal bleeding events and comorbidities
Effect of Dialysis Modalities on Risk of Hospitalization for Gastrointestinal Bleeding [1] Huang et al.	Scientific Reports 2023	Retrospective cohort study	Compare the risk differences in incident GI bleeding between patients receiving HD and PD N = 12,592 (HD 6296) (PD 6296)	HD patients have a 13% higher risk of hospitalization for GI bleeding compared to propensity‐matched PD patients (HR 1.13), likely due to heparin use.	Retrospective study No data for endoscopic examination Based on ICD 10 diagnoses or prescription
Bleeding Risk of Hemodialysis and Peritoneal Dialysis Patients [40] Sluijs et al.	International Urology and Nephrology 2017	Prospective cohort study	Dialysis patients were prospectively followed for major bleeding events over 3 years N = 1745 (HD = 1211) (PD = 534)	HD is associated with a 1.5‐fold higher risk of major bleeding compared to PD, particularly in patients with a previous history of bleeding.	No reports for small bowel
Hemodialysis Increases the Risk of Lower Gastrointestinal Bleeding and Angiodysplasia Bleeding: A Nationwide Population Study [41] Tsai et al.	Gastroenterology Research and Practice 2020	Retrospective cohort study	Patients with any admission diagnosis with gastrointestinal disorder N = 2870 (HD = 574)	HD patients have higher risks of lower GI bleeding and angiodysplasia bleeding than the non‐CKD subjects and dialysis‐free patients.	Retrospective study No assessment of SB
Impact of CKD/Kidney Failure on Angiodysplasia‐Related Gastrointestinal Bleeding: Incidence, Outcomes, and Readmission Risk in a National Cohort [42] He et al.	Kidney360 2025	Retrospective cohort	Adults who hospitalized with AGIB N = 46,670 (HD = 5227)	AGIB is more common among individuals with CKD, particularly those maintained on HD.	Retrospective study No assessment for small bowel Based on ICD 10 diagnoses

Abbreviations: CKD: chronic kidney disease, GI: gastrointestinal, HD: hemodialysis, HR: hazard ratio, ICD 10: International Classification of Diseases, 10th Revision, PD: peritoneal dialysis, SB: small bowel.

Upper gastrointestinal bleeding in hemodialysis patients is associated with high mortality and frequent recurrence. Weng et al. reported cumulative mortality rates of 13.7% at 1 month and 27.0% at 1 year in a cohort of 322 hemodialysis patients. Multivariate analysis identified hypoalbuminemia (≤ 3 g/dL, HR = 2.67), infection during hospitalization (HR = 1.85), female sex, and a single episodic upper gastrointestinal bleeding (adjusted HR = 2.00) as independent predictors of death. Hemodialysis patients admitted with primary upper gastrointestinal bleeding had a profoundly increased risk of all‐cause in‐hospital mortality during the follow‐up period [37]. Similarly, Can et al. observed that peptic ulcers were the primary bleeding etiology and linked in‐hospital mortality to inflammatory markers, specifically elevated C‐reactive protein (CRP) and hypoalbuminemia. Patients with bleeding also exhibited longer dialysis vintage and higher NSAID use compared to controls. Therefore, careful medications management and efforts to reduce hospitalization are crucial for improving outcomes in hemodialysis patients with gastrointestinal bleeding [38]. Trivedi et al. emphasized the prognostic significance of these events, noting that gastrointestinal bleeding nearly doubles the risk of mortality and that a single episode significantly increases the likelihood of recurrence, necessitating enhanced surveillance. These findings underscore the clinical importance of early risk stratification, preventive measures, and optimized management pathways for high‐risk hemodialysis populations [39].

Comparative studies indicate a high bleeding risk in hemodialysis compared to peritoneal dialysis, probably due to the need for heparin and hemodynamic stress. Huang et al., using a propensity‐matched cohort (n = 6296 pairs), demonstrated a 13% higher risk of gastrointestinal bleeding hospitalization in hemodialysis patients (HR ~ 1.13) while this gap was even more noticeable in patients who were receiving anticoagulants [1]. Validating these findings, the prospective NECOSAD [3] cohort reported a significantly higher incidence density of major hemorrhage in hemodialysis versus peritoneal dialysis (60.8 vs. 34.6 events/1000 person‐years). After adjustment, hemodialysis was linked to a 1.5‐fold increase in major bleeding risk (HR = 1.5), especially among patients with a bleeding history. Consequently, peritoneal dialysis may be a safer dialysis method for patients with an increased initial risk of bleeding. So, this risk should be incorporated into shared decision‐making when selecting dialysis modality [40].

Furthermore, epidemiological data from Tsai et al. show that hemodialysis is a strong, independent predictor of lower gastrointestinal bleeding, with incidence rates significantly exceeding those of CKD not undergoing dialysis (12.9% vs. 3.6%) [41]. Similarly, He et al. demonstrated a progressive, stage‐dependent correlation specifically between renal impairment and angiodysplasia‐related bleeding, which culminates in a “high‐risk phenotype” among dialysis patients. This study emphasized the severe clinical consequences, reporting significantly increased in‐hospital mortality, greater need for blood transfusion, and frequent readmissions within 30 days. Collectively, these findings highlight that the specific vascular and hemostatic abnormalities, and general risk of hemorrhage, in this population require close monitoring and personalized care [42].

4. Discussion

Until the last decade, small intestinal pathology in hemodialysis patients had been studied in only a limited number of case series [22, 24, 43, 44], leaving a blind spot in our understanding of obscure gastrointestinal bleeding. Small bowel lesions represent a significant but underestimated cause of gastrointestinal bleeding in patients undergoing hemodialysis. The evidence synthesized in this review highlights that vascular abnormalities, particularly angiodysplasias, constitute the predominant small bowel pathology in this population, followed by erosions and ulcers. These lesions frequently remain undetected by conventional upper and lower endoscopy and require specialized diagnostic tools such as capsule endoscopy or balloon‐assisted enteroscopy for accurate evaluation and localization. As illustrated in our proposed algorithm (Figure 1), the role of capsule endoscopy remains crucial. The utility is not only restricted after negative upper and lower gastrointestinal endoscopy but also after findings of hemorrhage lesions in order to exclude concomitant lesions.

FIGURE 1.

Flow chart for early detection and management of small bowel lesions in hemodialysis patients.

Several mechanisms have been proposed to explain the high prevalence of small bowel lesions among hemodialysis patients. First of all, uremia itself causes platelet dysfunction and accumulation of uremic toxins, and vascular fragility secondary to CKD contributes to mucosal vulnerability and hemorrhage tendency. Moreover, the widespread use of antiplatelet and anticoagulant therapy, which is common in this population due to coexisting cardiovascular comorbidities, further grows the bleeding risk. Ischemic injury and inflammation from intradialytic hemodynamic instability may also play contributory roles, particularly in the appearance and growth of erosive or ulcerative lesions. Additionally, a potential, but under‐investigated mechanism involves chronic tissue hypoxia driving angiogenesis through Hypoxia‐Inducible Factors, which may explain the recurrent formation of vascular ectasias in these anemic patients (Heyde's syndrome‐like mechanism) [45].

From a diagnostic perspective, capsule endoscopy has emerged as the preferred method to evaluate obscure gastrointestinal bleeding in hemodialysis patients. The studies reviewed demonstrate a high diagnostic yield of capsule endoscopy in identifying lesions such as erosions and ulcers or angiectasias, particularly when performed soon after a hemorrhage episode. However, the correlation between lesion detection and clinical outcomes such as rebleeding or anemia correction remains incompletely defined. The introduction of a rebleeding index, such as the RHEMITT score [16], may assist in identifying patients at high risk, like the end‐stage renal disease population, for hemorrhage and optimizing follow‐up strategies.

Clinically, small bowel bleeding has major implications for hemodialysis patients, contributing to recurrent anemia, consecutive transfusion, and frequent hospitalizations, all of which deteriorate the quality of life and increase global healthcare costs. Early detection and appropriate management are essential. Furthermore, endoscopic therapy, including argon plasma coagulation, has shown efficacy for hemostasis of angiodysplastic lesions; the problem of rebleeding remains stubborn. In these cases, pharmacologic options, such as somatostatin analogues, thalidomide [30], or anti‐angiogenic agents, represent potential adjuncts, though evidence in hemodialysis populations is still limited. A significant advance in pharmacotherapy was recently reported by Chen et al. regarding recurrent episodes of bleeding from small‐bowel angiodysplasia. Their randomized trial showed that thalidomide (50–100 mg/day) offered advanced efficacy over placebo, with approximately 69% of treated patients achieving a ≥ 50% reduction in bleeding episodes compared to just 4% in the control group. These findings make thalidomide a very promising drug for difficult‐to‐treat cases [46].

Despite growing interest, several gaps remain in the current literature. There is a lack of large, prospective studies that directly compare the prevalence and natural history of small bowel lesions between hemodialysis and non‐hemodialysis populations. Given that heparin‐based anticoagulation is used during hemodialysis, the question arises whether the growing hemodialysis population is at increased risk of gastrointestinal hemorrhage. Furthermore, standardized diagnostic algorithms and treatment protocols are lacking, and the long‐term effectiveness of endoscopic or medical therapies in preventing relapse has not been established. Future research should focus on elucidating the mechanisms linking dialysis parameters to mucosal damage, improving noninvasive diagnostic tools, and identifying preventive strategies tailored to the CKD patient population.

Evolving data also point to the phenomenon of “renalism,” which means the reluctance to perform invasive procedures in CKD patients, as a barrier to effective care [47]. A 2023 systematic review by Lin et al. highlighted that delayed endoscopy contributes to the higher mortality observed in patients with end stage renal disease, emphasizing the need for early intervention [48]. To solve this problem, new risk management methods are being developed. Larkin et al. recently introduced a machine learning model (XGBoost), which is capable of predicting the risk of gastrointestinal bleeding hospitalization in hemodialysis patients over a 180‐day window with superior accuracy compared to traditional logistic regression. Such AI‐based tools could facilitate personalized monitoring and early intervention, potentially mitigating the high morbidity associated with these lesions [49].

Finally, this narrative review has some limitations. First, most of the available studies are retrospective, single‐center, and include a small number of samples, which may contribute to selection bias. Second, there is considerable heterogeneity of study designs and definitions of small bowel lesion which can complicate direct comparison and quantitative synthesis. Nevertheless, the current evidence consistently highlights the clinical importance of gastrointestinal mucosal injury in hemodialysis patients. In order to fill the knowledge gap about gastrointestinal tract lesions, particularly small bowel, in hemodialysis patients, future research should prioritize large, prospective, multicenter studies to: (a) clarify the specific impact of different anticoagulation therapy on bleeding risk, (b) validate diagnostic algorithms that incorporate capsule endoscopy early in the workup, and (c) investigate the long‐term efficacy of combined endoscopic and drug therapies in preventing recurrence of bleeding.

5. Conclusion

Small bowel lesions, particularly angiodysplasias, are an important but often overlooked cause of gastrointestinal bleeding in hemodialysis patients. This group of patients has several comorbidities and, in addition, several causes that increase gastrointestinal injury and the risk of bleeding. Capsule endoscopy has proven to be the most valuable diagnostic tool, enabling the detection of lesions that are usually missed by conventional endoscopy. However, optimal management strategies remain unclear, and rebleeding is common despite endoscopic and medical interventions. To improve outcomes, early diagnosis, individualized therapeutic approaches, and careful adjustment of medications affecting coagulation are crucial. Future prospective, multicenter studies are warranted to establish standardized diagnostic pathways and evaluate new pharmacologic and endoscopic therapies aimed at reducing relapse and improving quality of life in hemodialysis patients.

Funding

The authors have nothing to report.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.