The Curbsiders podcast

#355 GI Bleed with Dr. Tanvi Dhere

September 19, 2022 | By


Upper and Lower GI Bleed Diagnosis and Management

Handle the gush of upper and lower GI bleeds with ease thanks to Dr. Tanvi Dhere (@DhereMdEmory)! Learn diagnosis, initial management as well as post-procedure management including when to resume anticoagulation and antithrombotic therapy.

Claim free CME for this episode at!

Episodes | Subscribe | Spotify | Swag! | Top Picks | Mailing List | | Free CME!


  • Producers and Writers: Meredith Trubitt MD, MPH; Monee Amin MD
  • Show Notes:  Monee Amin, MD
  • Infographics and Cover Art: Caroline Coleman, MD
  • Hosts: Meredith Trubitt MD, MPH; Monee Amin MD  
  • Reviewer: Emi Okamoto MD
  • Showrunner: Matthew Watto MD, FACP
  • Technical Production: PodPaste
  • Guest: Tanvi Dhere, MD

Sponsor: Birch

Birch is giving $400 off all mattresses and 2 free eco-resy pillows at

Sponsor: BetterHelp

Visit to get 10% off your first month. 

Sponsor: Medmastery

Go to to claim a discount on any of your subscriptions. 

Sponsor: Indeed

Visit to start hiring now.

CME Partner: VCU Health CE

The Curbsiders are partnering with VCU Health Continuing Education to offer FREE continuing education credits for physicians and other healthcare professionals. Visit and search for this episode to claim credit. 

Show Segments

  • Intro, disclaimer, guest bio
  • Guest one-liner, Picks of the Week*
  • Case from Kashlak;
  • Upper GI bleed
  • Labs, Stool Studies
  • Initial Management of GI bleed
  • Endoscopy
  • Post-procedure management
  • Lower GI bleed
  • Colonoscopy
  • Imaging/Interventional Radiology
  • Outro

GI Bleed Pearls

  1. Glasgow Blatchford scoring system can help risk stratify patients presenting with upper GI bleed
  2. Erythromycin is a pro-motility agent that can be given prior to endoscopic evaluation to help improve quality of the evaluation
  3. H. pylori stool testing as an inpatient is low yield. It should be done in the outpatient setting.
  4. Follow-up EGD may be indicated, especially for gastric ulcers which can commonly be signs of underlying malignancy. 
  5. Pill endoscopy is an option for further evaluation if EGD, colonoscopy and push enteroscopy have yielded no source for bleed. 
  6. CTA and tagged red blood cell scans can be helpful for brisker bleeds with unclear source or difficult to reach source. 

Gastrointestinal (GI) Bleed Show Notes

Upper GI Bleed

Initial Presentation History and Physical

Patients can present with abdominal pain, light-headedness, dark, tarry stool, hematemesis (frank blood or coffee ground emesis). However, dark stools can be misleading and history can help determine if a bleed is not the cause of dark stools. Iron, black licorice or bismuth subsalicylate are substances that can cause dark colored stools. While melena and hematemesis are clear symptoms of a GI bleed, a thorough history can also help narrow the differential for the etiology of the bleed (Wilkins, 2020).

Symptoms include epigastric pain, nausea/vomiting. Reflux symptoms can suggest reflux esophagitis. Weight loss, appetite loss, dysphagia can suggest underlying malignancy (Wilkins, 2020).

Risk factors for peptic ulcer disease (PUD) include NSAIDs, aspirin, OTC pain relievers with high dose aspirin. Variceal bleeds should also be considered–take a thorough history for cirrhosis risk factors (Wilkins, 2020).

Past medical history can reveal other less common causes of bleed. For example, renal disease or  aortic stenosis can predispose patients to arteriovenous malformations (AVMs). Similarly,  previous abdominal surgeries, specifically upper GI tract surgeries like gastric bypass may result in anastomotic ulcers. History of AAA repair can cause aortoenteric fistulas which can lead to life threatening bleeds.  These may present with a self-limited minor GI bleed referred to as a herald bleed prior to presenting with a more catastrophic GI bleed (Jones, 2012).

Once a thorough history is obtained, this can be used to calculate a Glasgow Blatchford score , which is used to determine risk of adverse outcomes in a bleed.  A score of zero means low risk for adverse outcomes and a patient with this score can be discharged directly from the emergency department (Laine, 2021). An example of this would be a patient with a clear viral gastroenteritis presentation who has had ongoing non-bloody emesis that eventually becomes bloody but is otherwise stable. This is consistent with a Mallory-Weiss tear. In cases when a patient is discharged home from the emergency department, they should have close follow up.

Stool Testing

Guaiac stool tests are more appropriate in the outpatient setting in colon cancer screening in someone who is average risk and should not be used in the inpatient setting. False positives can occur in several instances–epistaxis, and can lead to inadvertent other invasive testing. False negatives may occur with slow bleeds or intermittent bleeds. Not really needed given history yields high positive predictive value in GI bleed (Lee, 2020).

Regarding H. pylori, biopsies are often taken during endoscopy when an ulcer is found. However in active bleeds and PPI use, there can be false negatives. Dr. Dhere recommends outpatient follow up, after the acute ulcer bleed, and consider repeat H.pylori testing if inpatient testing is negative. PPI  must be held for at least 2 weeks before a repeat stool H.pylori antigen test. Serum antibody testing should also be avoided (Wang, 2015).  If H.pylori testing is positive, documentation of eradication should be performed after completion of treatment to prevent recurrence of ulcer. This is further reviewed on the Curbsiders Episode #322: H. pylori Infection!


While a decreased hemoglobin can solidify concerns for GI bleed, much like history, other labs can help narrow the likely etiology of a bleed. A BUN/Cr ratio of greater than 36 is more suggestive of upper GI bleed than lower GI bleed (Wilkins, 2020). Thrombocytopenia, leukopenia and abnormal coagulation studies may suggest cirrhosis, which may change approach to work-up and initial management compared to PUD.

Initial Management

First step in management is to consider the ABCs. Ensure adequate access with 2 large bore IVs, type and screen, IV fluids if needed. Fluids are indicated in patients with hemodynamic instability. In addition to addressing instability, starting an IV PPI is key–these can help downgrade severity of ulcer bleed and need for intervention (Wilkins, 2020).

PPIs can also be beneficial in non-ulcer related upper GIB causes by  increasing the pH and increase clotting ability as a result.  Notably, there does not appear to be a difference in outcomes between continuous drip of PPI versus intermittent bolus dosing (BID) of 80mg (Laine, 2021).

In patients who have a suspected bleeding peptic ulcer, endoscopy is recommended within 24 hours of presentation. If the patient is hemodynamically unstable, endoscopy should take place sooner once the patient is stabilized. 

Consideration for administering IV erythromycin prior to the procedure should be made. Erythromycin is a pro motility agent and can help empty the stomach of blood debris/clot to improve endoscopic visualization (Laine, 2021).

Patients with cirrhosis who present with suspected variceal bleeds, initial management should include octreotide and prophylactic antibiotics. Mortality in such cases is as high as 20%. For this reason, Dr. Dhere recommends endoscopy within 12 hours after adequate resuscitation.

Transfusion Thresholds

In the Villanueva study in 2013, outcomes in patients with more restrictive strategies had better outcomes than those with liberal transfusion thresholds. Specifically, the target hemoglobin is a hemoglobin of 7 g/dL. This study did not include patients with massive hemorrhage or patients with cardiovascular events within 90 days of the bleed (Villanueva, et al 2013).

Patients with underlying cardiovascular disease, the threshold is more liberal to a hemoglobin of 8g/dL (Laine, 2021). While the Villanueva study excluded active coronary disease only, Dr. Dhere recommends a higher threshold in any patient with significant coronary history.

Post-Procedure Management

Management following endoscopy depends on if there was an intervention during the endoscopy. Interventions are performed when there is an active bleed or exposed blood vessel. The interventions can include injection of epinephrine followed by cautery. In cases where intervention is performed, it is recommended that the patient remain on IV PPI twice a day for another 72 hours. If a patient has a clean-based ulcer, PPI can be switched from IV to oral. (Laine, 2021)

The duration of PPI course depends on the location of ulcers. Dr. Dhere recommends 8-12 weeks for patients with gastric ulcers and 4-8 weeks for duodenal ulcers. She also says this may change if a patient has other risk factors for peptic ulcers. For instance, if they are unable to stop using NSAIDs or require aspirin or anticoagulation, PPI should be used indefinitely.  Similarly, if no clear risk factor or etiology is identified, PPI is also recommended indefinitely.

According to Dr. Dhere, the need for repeat endoscopy depends on the kind of ulcer. Gastric ulcers in particular should have repeat EGD as they can be malignant. In duodenal ulcers, malignancy is less likely so repeat scope is only indicated in patients with ongoing symptoms.

Post-Procedure Anticoagulation/Antithrombotic Therapy

It can be difficult to determine when to resume anticoagulation and/or antithrombotic therapy in patients with upper GI bleed as there is limited data. Expert opinion suggests that for patients on AC for atrial fibrillation, it should be resumed 7 days after bleed if possible. For higher risk conditions, it should be done sooner. For patients with CAD, aspirin should be resumed immediately after endoscopic intervention. (Wilkins, 2020)

Lower GI Bleed

Patients with lower GI bleed can present with painless hematochezia in an acute situation but may also have melena if the bleed is coming from the distal small bowel or right colon. Similarly, up to 10-15% of patients presenting with hematochezia actually have brisk upper GI bleeds (Strate, 2016).
Much like upper GI bleeds, a good history can help determine etiology of the bleed. NSAIDs, systemic antithrombotic therapies, and aspirin are all risk factors for lower GI bleed (Strate, 2016).
Indications for admission and subsequent work up are based on risk stratification including the following risk factors: hemodynamic instability at presentation (tachycardia, hypotension, syncope), ongoing bleeding (gross blood on initial digital rectal examination, recurrent hematochezia), comorbid illnesses, age > 60 years, a history of diverticulosis or angioectasia, an elevated creatinine, and anemia (initial hematocrit ≤35%) (Strate, 2016).
Dr. Dhere says asking about when the patient’s last bowel movement was can help determine the acuity of the bleed. Changes in bowel patterns prior to the bleed such as diarrhea can suggest inflammatory bowel disease.
Medical Management
Determining a patient’s hemodynamic stability is key. Intravenous fluids can be used to achieve this. Blood transfusions can also help attain stability. Like upper GI bleeds,a hemoglobin above 7g/dL is recommended. This recommendation changes if the patient is having a massive hemorrhage, in which case 9g/dL is the goal (Strate, 2016).
Correction of coagulopathy may also be required to stabilize and prepare a patient for endoscopy. In patients with INR >2.5, a reversal agent should be considered. In patients with platelet count of less than 50 × 10^9/L, consider transfusing platelets to that count (Strate, 2016).
Endoscopic Evaluation
Stability or instability of the patient dictates the urgency and need for endoscopic evaluation. For patients who are unstable and have risk factors for upper GI bleed, urgent EGD is recommended once the patient is stabilized. In situations where patients are stable, colonoscopy should be performed within 24 hours. Dr. Dhere recommends waiting to speak to the GI consultant before beginning a colonoscopy prep.
Pill camera/small bowel video capsule endoscopy is utilized in the inpatient setting in patients who continue to bleed and have had negative EGD, colonoscopy and push enteroscopy. If the patient stops bleeding and there is concern for small bowel bleed, a video capsule can be performed in the outpatient setting (Wu, 2021).
Imaging Studies and Interventional Radiology
Sometimes imaging before endoscopic evaluation is required. Both computed tomographic angiography (CTA) and tagged red blood cell scans can be used in the diagnosis of GI bleeds. Generally, they should be utilized in the acute setting when a patient is actively bleeding. Moreover, imaging is of particular use in patients who are unstable from their active bleed. In these situations, the need to identify the location of the bleed quickly supersedes the need for any potential prep for colonoscopy. They are particularly helpful in patients who have a known history of a diverticular bleed. In such patients, colonoscopy is less useful as the etiology is already known. If a patient is not actively bleeding, neither study has utility (Strate, 2016).
Dr. Dhere tells us that CTAs are faster and better at giving an anatomic location, whereas tagged red blood cells scans give a less specific potential location of bleed. However, these scans can be useful in patients with renal insufficiency as CTA delivers a high contrast load.
Interventional radiology (IR) can also be helpful in acute, unstable situations. In addition to being able to identify the location of the bleed, IR can also stop the bleed via embolization. An important risk to consider when deciding if IR is an appropriate strategy for a patient is up to 4% of patients can develop bowel ischemia/necrosis after embolization of a vessel (Strate, 2016).
Surgical Intervention
Most diverticular bleeds stop without intervention. However, in some scenarios, lower GI bleeds can be recurrent and intermittent and difficult to localize. In these situations, general surgery should be consulted for possible surgical interventions such as colectomy. Importantly, confirmatory testing regarding the existence and location of active bleed should be conducted before consulting surgery (Strate, 2016).
Transitions of Care
Dr. Dhere suggests that while direct medical therapies for lower GI bleed are not currently available, a thorough review of risk factors and home medication regimen should be performed. If a patient is on systemic antiplatelet or anticoagulation, it is important to assess the utility of continuing it long term. In some cases, such as recent coronary event or saddle pulmonary embolism, antiplatelet/anticoagulation cannot safely be held. In other situations, a full risk benefit analysis should be conducted.  For those hospitalized patients who need to continue anticoagulation, Dr. Dhere recommends trialing patients on a heparin drip to assess for development of a bleed before resuming oral anticoagulation. For those patients whose anticoagulation was held during the hospitalization, consideration for  resumption of home anticoagulation 7-14 days days following the inciting event should be made (Kido, 2017).


  1. Welcome To Our Show (New Girl rewatch podcast)
  2. Meredith’s 2000s Peloton ride


Listeners will develop a framework to approach evaluation and management of GI bleeds.

Learning objectives

After listening to this episode listeners will…  

  1. Risk stratify patients with GI bleed for admission versus discharge from emergency department
  2. Compare and contrast the different presentations commonly associated with upper and lower bleeds
  3. Identify when CTA or tagged red blood cell scans can assist with diagnosis and treatment of GI bleeds


Dr. Dhere reports no relevant financial disclosures. The Curbsiders report no relevant financial disclosures. 


Dhere T, Trubitt ME, Amin M. “355 GI Bleed”. The Curbsiders Internal Medicine Podcast. September 19, 2022.


  1. September 19, 2022, 1:45pm Tim Bresnahan writes:

    Outstanding review! Is there any role for adding sucralfate to PPI for gastric or duodenal ulcer? At my hospital, general surgeons perform the bulk of our endoscopy and typically add sucralfate to the PPI. Although I'm familiar with head to head comparisons favoring PPI over sucralfate, I'm not aware of trials comparing combination to PPI only. I usually end up stopping the sucralfate since my patients don't like the frequency or, depending on the preparation, the cost.

  2. October 14, 2022, 11:30am Terry Shaneyfelt writes:

    Pre-endoscopic PPI therapy was not recommended in the ACG guideline from 2021 on upper GI bleeding. They report a paucity of evidence supporting it and the evidence was poorly designed. Not sure how to handle this as its seems pretty much SOC to do this.

  3. October 15, 2022, 9:48am Gabriel Gustin writes:

    What is the mortality rate of patients that come to the Ed with an Acute Gi bleed and are hospitalized?

    • November 23, 2022, 10:25am Ask Curbsiders writes:

      Hi! Great question. The number varies as underlying cause of the bleed is elucidated. For example, variceal bleed being higher mortality than gastritis. Additionally, mortality goes up with increasing co-morbidities. For UGIB, they have asked the question using observational studies, so not very robust data, but that rate is ~10%. For LGIB, similar data but rate is lower at < 4%.

CME Partner


The Curbsiders are partnering with VCU Health Continuing Education to offer FREE continuing education credits for physicians and other healthcare professionals. Visit and search for this episode to claim credit.

Contact Us

Got feedback? Suggest a Curbsiders topic. Recommend a guest. Tell us what you think.

Contact Us

We love hearing from you.


We and selected third parties use cookies or similar technologies for technical purposes and, with your consent, for other purposes as specified in the cookie policy. Denying consent may make related features unavailable.

Close this notice to consent.