Rectal bleeding

EM-CCM Conference: February 2014 (Sorry posting this a little late!)

Presented by Dr. Freedman

Summary by Dr. Youn

CASE: 84M with CAD (recent NSTEMI) on ASA and Clopidogrel, CHF, HTN, DM p/w 4 episodes of BRBPR with dizziness and SOB. Pt was hypotensive, tachycardic with pale conjunctiva and gross red blood on rectal examination. GI, Surgery, and Critical Care Medicine were consulted. Labs were notable for H/H of 7/22 and Troponin 2.94. The patient’s primary problem list included GI bleed (lower vs. brisk upper), shock/hypotension, Type 2 MI, and platelet dysfunction.

DISCUSSION:

LGIB DEFINITION: any bleeding distal to Ligament of Treitz

ETIOLOGY:

Diverticulosis: 17-40%
AV Malformation: 2-30%
Colitis: 9-21%
Neoplasm: 11-14%
Anorectal: 4-10%
Brisk UGIB: 0-11%
Small Bowel: 2-9%

RISK STRATIFICATION: no scoring system, but characteristics with increased risk of severe LGIB

HR > 100
SBP < 115
Syncope
Nontender abdomen
Bleeding noted during initial 4 hours of course
History of aspirin use
Two or more comorbidities
Strate LL, Orav E, Syngal S. Early Predictors of Severity in Acute Lower Intestinal Tract Bleeding. Arch Intern Med. 2003;163(7):838-843.

GOALS OF LGIB MANAGEMENT:

Resuscitation
- ABCs
- Vascular access
- Blood products
- Hemodynamic monitoring
Identification/Localization of bleeding source
- NG lavage (NG = NasoGastric to some, NG = No Good to others)
- Endoscopy
- Angiogram/IR
- CT Angiography
- Bleeding scan
Appropriate intervention
- Endoscopy
- Surgery
- IR embolization/vasopressin infusion

EBM#1: ANTI-PLATELET REVERSAL

No RCTs, numerous retrospective, observational studies: No difference in outcomes
Desmopression (DDAVP): No outcome data
Nearly 2-fold increase in risk of UGIB with use of low-dose aspirin; further increased risk of UGIB with use of low-dose aspirin along with clopidogrel, oral anticoagulants, NSAIDs, or high-dose corticosteroids (Circulation. 2011; 123:1108-1115)

EBM#2: EMERGENT ENDOSCOPY/COLONOSCOPY

Green et al. Urgent colonoscopy for evaluation and management of acute lower gastrointestinal hemorrhage: a randomized controlled trial. Am J Gastroenterol 2005; 100:2395–2402.
- Prospective, randomized
- N = 100
- Purge and colonoscopy within 8 hours vs. elective colonoscopy
- Urgent colonoscopy identified definite source of LGIB more than standard care algorithm
- No difference in outcome measures: mortality, hospital stay, ICU stay, transfusion requirements, early rebleeding, late rebleeding, surgery
Laine L, Shah A. Randomized trial of urgent vs. elective colonoscopy in patients hospitalized with lower GI bleeding. Am J Gastroenterol 2010; 105: 2636–2641.
- Randomized, prospective study
- N = 72 (after EGD)
- All received EGD within 6 hours. If no upper bleeding source identified, randomized to colonoscopy within 12 hours or elective colonoscopy (36-60hours)
- No difference in outcome measures: further bleeding, units of blood, hospital days, subsequent diagnostic or therapeutic interventions for bleeding, hospital charges

EBM#3: ANGIOGRAPHY – requires > 1cc/min bleeding

Localization: 40-80% in literature
Embolization: 80-100% success rate, with 14-29% rebleeding
Complication rate: up to 23%

EBM#4: SURGERY

Subtotal colectomy: bleeding source not identified, massive blood loss, hemodynamic instability (last resort?); high mortality 25-33%
Targeted resection: refractory bleeding after localization, 7% mortality if localized; avoid blind resection

EBM #5: CT ANGIOGRAPHY

Yoon W, Jeong Y Y, Shin S S. et al. Acute massive gastrointestinal bleeding: detection and localization with arterial phase multi-detector row helical CT. Radiology. 2006; 239(1):160–167.
- Prospective, observational, non-randomized
- All patients for IR underwent CTA prior to exam
- Sens 90%, Spec 99%, 95% PPV, 99%NPV
CTA predicts those who will benefit from IR

RESIDENT/FACULTY DISCUSSION:

– Do not hesitate to initiate massive transfusion protocol for unstable, bleeding patients upon initial evaluation and management

– Is there a target Hgb in LGIB? Can we apply the 2013 NEJM study (restrictive vs. liberal transfusion strategies in UGIB) to LGIB patients?

– Consider Cordis catheter and Level 1 for fast (and warm), large volume resuscitation in hemodynamically unstable bleeding patients

– Don’t fret if GI delays endoscopy, just make sure they see the patient

– Ultimately, sick bleeders need ICU (MICU vs. SICU depends on your institution, administrative protocols)

– Premarin (conjugated estrogens) for GIB? Some case studies on use of conjugated estrogens in GI bleeding patients with CRF or ESRD, no randomized studies.

– 1g drop in hemoglobin ~200cc blood loss. Therefore, if patient has significant bleed with drop in hemoglobin, guaiac status (positive or negative) is irrelevant. There should be gross blood or melena on rectal exam. If you’re finding only guaiac positive stool, then that’s not the source of the Hgb drop. Do the rectal exam. Leave the guaiac card behind.

– Tranexamic acid (TXA): Cochrane review 2012 in UGIB

7 double-blinded, randomized studies on TXA versus placebo, cimetidine, or lansoprazole
Primary outcome of all-cause mortality showed benefit of TXA compared with placebo, but no difference in subgroup analyses
No difference between TXA versus placebo in bleeding, surgery, or transfusion requirements
Overall, number of patients with any thrombotic event was not significantly increased in TXA group
No literature on TXA in LGIB