Acute lower gastrointestinal bleeding

A 74-year-old man on apixaban for atrial fibrillation arrives at 02:00 with six hours of painless, large-volume hematochezia, heart rate 112, blood pressure 98/60. Every branch of the next twelve hours — crystalloid before scope, CTA before colonoscopy, clip before coil, and never an unlocalized colectomy — is mapped by the ACG 2016 and 2023 guidelines, the BSG 2019 guideline, and a small but decisive trial literature. This page gives you the pathway with its numbers attached.^[1][2][3]

Definition, and the upper-GI trap

Acute lower gastrointestinal bleeding is the acute onset of hematochezia originating from the colon or anorectum. The definition carries a built-in trap: hematochezia associated with hemodynamic instability may be indicative of an upper gastrointestinal bleeding source, and an upper endoscopy should be performed when suspicion is high — brisk duodenal bleeding presents exactly this way.^[1][2] Severity frameworks define severe bleeding pragmatically: transfusion of 2 or more units of packed red cells and/or a haematocrit fall of 20% or more from baseline in the first 24 hours, or recurrent bleeding after 24 hours of stability.^[12][13]

The stigmata of recent hemorrhage (SRH) language matters for everything downstream: active bleeding, non-bleeding visible vessel and adherent clot are the high-risk stigmata; clean base and flat spot are the low-risk findings.^[2][19]

Epidemiology and the etiology hierarchy

The population numbers come from Longstreth's Kaiser cohort: an annual incidence of 20.5 per 100,000 (24.2 male versus 17.2 female), rising more than 200-fold from the third to the ninth decade of life.^[6] US hospitalization data show lower GI bleeding admissions actually declining (41.8 to 35.7 per 100,000 from 2001 to 2009, with colonic diverticular bleeding 30.4 to 23.9) against a 2009 case fatality of 1.47% for LGI bleeding.^[7]

The etiology league table: colonic diverticulosis 41.6%, colorectal malignancy 9.1%, ischemic colitis 8.7%, miscellaneous 28.8%, unknown 11.9% — and the ACR confirms diverticulosis remains the commonest cause.^[6][5]

^{[6] [8] [9] [10]}

Mortality is patterned, not random. In the 227,022-patient national cohort, 3.9% died in hospital; independent predictors were age over 70 (OR 4.91), intestinal ischemia (OR 3.47), two or more comorbidities (OR 3.00), bleeding that began while hospitalized for another process (OR 2.35), coagulation defects, hypovolemia, transfusion and male sex.^[8] The community-versus-inpatient onset contrast is stark: 2.4% versus 23.1% mortality.^[6] Drugs matter: aspirin (HR 2.75) and NSAIDs (HR 8.61) are independent LGIB risk factors, and vitamin K antagonists treble all-cause GI bleeding versus placebo.^[1]

Initial assessment and triage

Initial patient assessment and hemodynamic resuscitation proceed simultaneously — a strong ACG recommendation.^[2] The history sorts the differential fast: painless large-volume hematochezia in an elderly patient is diverticular until proven otherwise; iron deficiency with occult loss suggests angioectasia; pain with bloody diarrhoea points to ischemic colitis or IBD; the post-polypectomy bleeder declares itself by timing. Crystalloid resuscitation to normalize blood pressure and heart rate precedes any diagnostic or therapeutic intervention.^[1] Thrombocytopenia: maintain platelets above 30 × 10⁹/L in clinically significant bleeding, above 50 × 10⁹/L if an invasive procedure is required.^[1]

Risk scores — Oakland first, judgment always

Oakland score (2017) — derived and validated in two prospective UK cohorts. Seven variables: age, sex, previous admission for LGIB, rectal examination findings, heart rate, systolic blood pressure and haemoglobin. It discriminated safe discharge at C statistic 0.84 (development) and 0.79 (validation), outperformed Rockall, Blatchford, Strate, BLEED, AIMS65 and NOBLADS, and a score of 8 or less predicts a 95% probability of safe discharge.^[9] The US external validation (46,128 patients) reported AUROC 0.87; the ≤8 threshold identified 8.7% of patients with sensitivity 98.4% and specificity 16.0% for safe discharge — nearly half of all admitted patients met safe-discharge criteria overall.^[10] The 2024 Spanish validation found Oakland (AUC 0.85) outperformed SHA2PE (0.797), with the ≤8 cutoff giving 100% PPV and specificity at the cost of low sensitivity.^[11]

The supporting cast: the Strate rule (2005) stratifies severe-bleeding risk into three bands — low 6–9%, moderate 43%, high 79–84% — validated prospectively with AUROC ~0.75.^[12] NOBLADS (NSAIDs, no diarrhoea, no abdominal tenderness, BP ≤100, antiplatelet use, albumin under 3.0 g/dL, disease score ≥2, syncope): a score of 5 or more meant 75.7% developed severe LGIB versus 2% with no factors (AUC 0.77).^[13]

ACG 2023 position: use risk stratification tools (e.g., Oakland ≤8) to identify low-risk patients for early discharge and outpatient evaluation — but scores supplement, never replace, clinician judgment (conditional, low-quality).^[1]

Colonoscopy — the timing evidence you must quote precisely

Colonoscopy is the initial diagnostic procedure for nearly all patients with acute LGIB (ACG 2016, strong recommendation) — its value is detecting the source, and its therapeutic reach is what separates it from every imaging test.^[2][1] The controversy is timing, and the trial record is now clear:

Green 2005 (Duke). 100 patients randomized to urgent purge and immediate colonoscopy versus standard care (elective colonoscopy, or angiographic pathway if ongoing bleeding). Urgent colonoscopy found a definite source more often (OR 2.6, 95% CI 1.1–6.2) — but no difference in mortality (2% vs 4%), stay (5.8 vs 6.6 days), transfusion (4.2 vs 5 units), early rebleeding (22% vs 30%), surgery (14% vs 12%) or late rebleeding (16% vs 14%).^[14]

Niikura 2020 (15 Japanese hospitals). 159 patients: stigmata found in 21.5% early versus 21.3% elective (P = .967); rebleeding within 30 days 15.3% versus 6.7%; no significant differences in successful endoscopic treatment, transfusion, stay, thrombotic events or death.^[15]

The syntheses. Tsay and Laine's RCT-only review (4 trials): further bleeding with early colonoscopy RR 1.57 (95% CI 0.74–3.31) — colonoscopy within 24 hours does not reduce further bleeding or mortality.^[16] Kherad's meta-analysis adds the confounding lesson: among RCTs there was no rebleeding difference, while observational studies alone showed early colonoscopy associated with lower mortality, surgery and transfusion — association, not causation.^[17]

The guideline moved. ACG 2016 conditionally recommended colonoscopy within 24 hours for high-risk patients with ongoing bleeding after a rapid purge.^[2] ACG 2023, on the new RCT evidence, recommends a nonemergent inpatient colonoscopy for hospitalized patients (strong recommendation, moderate-quality) — and exempts patients whose bleeding has subsided after a high-quality colonoscopy within 12 months showing diverticulosis without neoplasia.^[1] Bowel preparation: 4–6 L PEG historically, split-dose or low-volume regimens acceptable.^[1]

CTA — the unstable patient's test

CTA's diagnostic credentials: the meta-analysis of 22 studies (672 patients) puts sensitivity at 85.2% and specificity 92.1% (AUC 0.935) for active acute GI hemorrhage^[21]; the ACG quotes 90% and 92% across 14 observational studies specific to LGIB.^[1] Its virtues are speed and no bowel preparation — it can be completed within minutes even in unstable patients; its vices are radiation, contrast, and a low yield once bleeding has stopped.^[1]

The positioning is explicit. ACG 2023: CTA as the initial diagnostic test in patients with ongoing hemodynamically significant hematochezia (conditional, low-quality) — and of low yield in minor bleeding or when bleeding has clinically subsided.^[1] A negative CTA is genuinely reassuring: nearly 80% of patients with an initial negative CTA have no further clinical or radiologic rebleeding and settle spontaneously.^[1] One practical number for the viva: in a head-to-head series, time to first diagnostic test was 3 hours with CTA versus 22 hours with colonoscopy, with active bleeding found more often by CTA (31% vs 15%).^[1]

Tagged RBC scintigraphy is fading: long acquisition, imprecise localization (accurate localization in only 65–80%), and CTA uptake displaced it (3.8% to 57% at one academic centre); the ACG/SAR 2024 imaging consensus and the CTA-versus-scintigraphy meta-analysis both place CTA first.^[2][4][22]

Angiography and superselective embolization

A positive CTA is a referral, not an answer: ACG 2023 recommends that extravasation on CTA prompts interventional radiology referral for transcatheter arteriography and possible embolization (strong, moderate quality) — with colonoscopy a reasonable alternative in expert centres.^[1] Timing decays fast: arteriography within 90 minutes of a positive angiogram is nine times more likely to enable embolization than delayed arteriography.^[1] And embolization is only possible when extravasation is seen — which is more likely exactly when the patient is least stable (ACR).^[5]

Quote the outcome envelope honestly. The Helsinki series (123 angiographies, 55 embolization attempts): technical success 96%, rebleeding 26%, overall complications 36% (major 17%), post-embolization ischemia 17%, bowel resection 13%, mortality 6%.^[24] The ACG-summarised figures: 98% embolization success, ulcer or infarction in nearly 5%, and even after embolization 30-day rebleeding 16% and mortality 13% — a reminder that this is a moribund population.^[1] Does angiography-first beat colonoscopy-first? The Japanese nationwide study (propensity-matched, 3,220 versus 805): no mortality difference, though fewer patients went to surgery within a day with angiography.^[23] Practice varies accordingly: tachycardia, syncope and early recurrent bleeding push towards radiography, while post-polypectomy bleeding and weekday presentation push towards colonoscopy.^[25]

Endoscopic hemostasis — diverticular stigmata

Why treatment matters: the natural history of untreated stigmata is bad. In Jensen's prospective data, major stigmata managed medically rebled in 65.8% and needed hemostatic intervention in 44.7% — while clean bases and flat spots never rebled; a Doppler probe heard arterial flow under 92% of major stigmata, silenced by hemostasis.^[19] Jensen's NEJM series established urgent colonoscopy with endoscopic therapy for severe diverticular hemorrhage: definite stigmata in about a fifth of patients, and endoscopic treatment reduced rebleeding and the need for surgery compared with medical management.^[18]

The modality menu (ACG 2023, strong recommendation): through-the-scope clips, endoscopic band ligation (EBL), or coagulation for diverticular SRH.^[1] Across 16 studies and 384 patients, bipolar coagulation, clipping and EBL all achieved initial hemostasis in 99–100% — but EBL was more effective than clipping at avoiding transarterial embolization or surgery.^[1]

The long view: presumptive diverticular hemorrhage rebled in 24.5% over a median 73 months, with 56% of rebleeders converting to definitive (stigmata-proven) disease — and all-cause mortality of 42.8%, none from bleeding. These are elderly patients dying with diverticulosis, not of it.^[20]

Angiodysplasia

Most angiodysplasia lesions (54–81.9%) sit in the caecum and ascending colon, formed by chronic low-grade intermittent obstruction of submucosal veins with VEGF-driven proliferation; they present with occult loss and iron deficiency as often as with overt bleeding.^[26] Argon plasma coagulation resolves bleeding in about 85% of colonic angiodysplasia^[26] — in the 100-patient Buenos Aires series, freedom from rebleeding was 98% at one year and 90% at two.^[27] Set expectations on recurrence: the pooled recurrence risk is 34% across uncontrolled studies.^[1] Clips versus APC for angiodysplasia: unadjusted rebleeding favoured clips (32.6% vs 46.5%), but the difference evaporated on multivariate regression — modality is the endoscopist's choice.^[28] Salvage for failures: thalidomide or octreotide give clinically meaningful responses in 71.4% and 77% respectively.^[26]

Blood, anticoagulants and antiplatelets — the playbook

This is the most quotable section for the written papers.^[1]

• Transfusion: restrictive strategy, threshold 7 g/dL in hemodynamically stable LGIB (conditional, low-quality); a higher threshold of 8 g/dL can be considered with known cardiovascular disease or acute coronary syndrome.^[1]
• VKA reversal: only for life-threatening bleeding with INR substantially above range; 4-factor PCC preferred to FFP for rapidity of INR reduction (90% vs 75% effective hemostasis in the surgical RCT).^[1]
• DOAC reversal: the small subset with life-threatening bleeding failing resuscitation plus cessation — idarucizumab for dabigatran, andexanet alfa for apixaban and rivaroxaban, if taken within 24 hours.^[1]
• Aspirin after diverticular hemorrhage: discontinue for primary prevention; continue for secondary prevention — continuation raises recurrent bleeding (HR 2.76) but protects against cardiovascular events (HR 0.59) and death (HR 0.33).^[1]
• Anticoagulation: resume after cessation of LGIB (strong, moderate-quality) — typically within 7 days, because resumption reduces post-bleeding thromboembolism and mortality.^[1]
• NSAIDs: discontinue nonaspirin NSAIDs after diverticular hemorrhage (strong recommendation).^[1]
• Tranexamic acid: avoid — thromboembolism and seizure without outcome benefit.^[1]

Surgery — localized, last

Surgery's role is deliberately small: in the UK audit it was needed in 0.2% of all LGIB cases, and ACG 2023 states it should be considered only after endoscopic and radiologic interventions have failed.^[1] ACG 2016 adds the referral discipline — a surgical consultation for high-risk patients with ongoing bleeding — and the cardinal rule: localize the source before resecting, because a nonlocalized hemicolectomy carries a higher risk of recurrent bleeding than a targeted resection.^[2][1] When resection happens, 85% are partial colectomies and 15% total colectomies; total colectomy brings more cardiac and renal complications and postoperative ileus.^[1]

Revision summary

• Resuscitate and assess simultaneously; hematochezia plus instability → EGD first.^[1][2]
• Oakland ≤8 discharges (95% safe); scores supplement judgment.^[9][1]
• Urgent colonoscopy within 24 h does not improve rebleeding or mortality (Niikura 21.5% vs 21.3%; Tsay RR 1.57) — nonemergent inpatient colonoscopy is the 2023 strong recommendation.^[15][16][1]
• CTA first for ongoing significant bleeding (sens ~85-90%); negative CTA → ~80% settle; extravasation → IR for embolization within 90 minutes.^[21][1]
• Embolization: 96-98% technical success, but quote the price — 17% ischemia, 13% resection, 6-13% mortality.^[24][1]
• Diverticular SRH: clips, EBL or coagulation (99-100% initial hemostasis; EBL best at avoiding TAE/surgery); untreated major stigmata rebleed 65.8%.^[1][19]
• Angiodysplasia: right colon, APC ~85% resolution, 34% pooled recurrence.^[26][1]
• The drug playbook: 7 g/dL threshold; PCC for VKA; targeted DOAC reversal; stop primary-prevention aspirin, continue secondary; resume anticoagulation within ~7 days; no tranexamic acid.^[1]
• Surgery in 0.2%: localize first, partial over total colectomy.^[1][2]

≥2 units PRBC and/or ≥20% haematocrit fall in the Strate/NOBLADS frameworks)?.^[12] 47.9% met safe-discharge criteria; Green/Niikura placebo-arm outcomes)?.^[10]

References28ShowHide

1. [1]Sengupta N, Feuerstein JD, Jairath V, Shergill AK, et al. Management of Patients With Acute Lower Gastrointestinal Bleeding: An Updated ACG Guideline. Am J Gastroenterol, 2023.PMID 36735555
2. [2]Strate LL, Gralnek IM ACG Clinical Guideline: Management of Patients With Acute Lower Gastrointestinal Bleeding. Am J Gastroenterol, 2016.PMID 26925883
3. [3]Oakland K, Chadwick G, East JE, Guy R, et al. Diagnosis and management of acute lower gastrointestinal bleeding: guidelines from the British Society of Gastroenterology. Gut, 2019.PMID 30792244
4. [4]Sengupta N, Kastenberg DM, Bruining DH, Latorre M, et al. The Role of Imaging for Gastrointestinal Bleeding: Consensus Recommendations From the American College of Gastroenterology and Society of Abdominal Radiology. Am J Gastroenterol, 2024.PMID 38857483
5. [5]Karuppasamy K, Kapoor BS, Fidelman N, Abujudeh H, et al. ACR Appropriateness Criteria® Radiologic Management of Lower Gastrointestinal Tract Bleeding: 2021 Update. J Am Coll Radiol, 2021.PMID 33958109
6. [6]Longstreth GF Epidemiology and outcome of patients hospitalized with acute lower gastrointestinal hemorrhage: a population-based study. Am J Gastroenterol, 1997.PMID 9068461
7. [7]Laine L, Yang H, Chang SC, Datto C Trends for incidence of hospitalization and death due to GI complications in the United States from 2001 to 2009. Am J Gastroenterol, 2012.PMID 22688850
8. [8]Strate LL, Ayanian JZ, Kotler G, Syngal S Risk factors for mortality in lower intestinal bleeding. Clin Gastroenterol Hepatol, 2008.PMID 18558513
9. [9]Oakland K, Jairath V, Uberoi R, Guy R, et al. Derivation and validation of a novel risk score for safe discharge after acute lower gastrointestinal bleeding: a modelling study. Lancet Gastroenterol Hepatol, 2017.PMID 28651935
10. [10]Oakland K, Kothiwale S, Forehand T, Jackson E, et al. External Validation of the Oakland Score to Assess Safe Hospital Discharge Among Adult Patients With Acute Lower Gastrointestinal Bleeding in the US. JAMA Netw Open, 2020.PMID 32633766
11. [11]Gonzalez-Gonzalez L, Iborra I, Fortuny M, Mañosa M, et al. External validation of the SHA(2)PE score and its comparison to the Oakland score for the prediction of safe discharge in patients with lower gastrointestinal bleeding. Surg Endosc, 2024.PMID 38902406
12. [12]Strate LL, Saltzman JR, Ookubo R, Mutinga ML, Syngal S Validation of a clinical prediction rule for severe acute lower intestinal bleeding. Am J Gastroenterol, 2005.PMID 16086720
13. [13]Aoki T, Nagata N, Shimbo T, Niikura R, et al. Development and Validation of a Risk Scoring System for Severe Acute Lower Gastrointestinal Bleeding. Clin Gastroenterol Hepatol, 2016.PMID 27311620
14. [14]Green BT, Rockey DC, Portwood G, Tarnasky PR, et al. Urgent colonoscopy for evaluation and management of acute lower gastrointestinal hemorrhage: a randomized controlled trial. Am J Gastroenterol, 2005.PMID 16279891
15. [15]Niikura R, Nagata N, Yamada A, Honda T, et al. Efficacy and Safety of Early vs Elective Colonoscopy for Acute Lower Gastrointestinal Bleeding. Gastroenterology, 2020.PMID 31563627
16. [16]Tsay C, Shung D, Stemmer Frumento K, Laine L Early Colonoscopy Does Not Improve Outcomes of Patients With Lower Gastrointestinal Bleeding: Systematic Review of Randomized Trials. Clin Gastroenterol Hepatol, 2020.PMID 31843595
17. [17]Kherad O, Restellini S, Almadi M, Strate LL, et al. Systematic review with meta-analysis: limited benefits from early colonoscopy in acute lower gastrointestinal bleeding. Aliment Pharmacol Ther, 2020.PMID 32697886
18. [18]Jensen DM, Machicado GA, Jutabha R, Kovacs TO Urgent colonoscopy for the diagnosis and treatment of severe diverticular hemorrhage. N Engl J Med, 2000.PMID 10631275
19. [19]Jensen DM, Ohning GV, Kovacs TO, Jutabha R, et al. Natural history of definitive diverticular hemorrhage based on stigmata of recent hemorrhage and colonoscopic Doppler blood flow monitoring for risk stratification and definitive hemostasis. Gastrointest Endosc, 2016.PMID 26227931
20. [20]Wangrattanapranee P, Khrucharoen U, Jensen DM, Jensen ME Long-Term Natural History of Presumptive Diverticular Hemorrhage. Am J Gastroenterol, 2024.PMID 38989865
21. [21]García-Blázquez V, Vicente-Bártulos A, Olavarria-Delgado A, Plana MN, et al. Accuracy of CT angiography in the diagnosis of acute gastrointestinal bleeding: systematic review and meta-analysis. Eur Radiol, 2013.PMID 23192375
22. [22]Yaxley KL, Mulhem A, Godfrey S, Oke JL The Accuracy of Computed Tomography Angiography Compared With Technetium-99m Labelled Red Blood Cell Scintigraphy for the Diagnosis and Localization of Acute Gastrointestinal Bleeding: A Systematic Review and Meta-Analysis. Curr Probl Diagn Radiol, 2023.PMID 37271638
23. [23]Miyakuni Y, Nakajima M, Ohbe H, Sasabuchi Y, et al. Angiography versus colonoscopy in patients with severe lower gastrointestinal bleeding: a nation-wide observational study. Acute Med Surg, 2020.PMID 32617165
24. [24]Nykänen T, Peltola E, Kylänpää L, Udd M Transcatheter Arterial Embolization in Lower Gastrointestinal Bleeding: Ischemia Remains a Concern Even with a Superselective Approach. J Gastrointest Surg, 2018.PMID 29549618
25. [25]Strate LL, Syngal S Predictors of utilization of early colonoscopy vs. radiography for severe lower intestinal bleeding. Gastrointest Endosc, 2005.PMID 15672055
26. [26]Sami SS, Al-Araji SA, Ragunath K Review article: gastrointestinal angiodysplasia - pathogenesis, diagnosis and management. Aliment Pharmacol Ther, 2014.PMID 24138285
27. [27]Olmos JA, Marcolongo M, Pogorelsky V, Herrera L, et al. Long-term outcome of argon plasma ablation therapy for bleeding in 100 consecutive patients with colonic angiodysplasia. Dis Colon Rectum, 2006.PMID 17024322
28. [28]Ismail B, Alayoubi MS, Abdelwadoud M, Castro FJ Rebleeding after hemoclip versus argon plasma coagulation for gastrointestinal angiodysplasias: a retrospective multicenter study. Eur J Gastroenterol Hepatol, 2022.PMID 33731586