Original Article: Upper Gastrointestinal Bleeding in Adults: Evaluation and Management

Issue Date: March 1, 2020

Available at: https://www.aafp.org/afp/2020/0301/p294.html

To the Editor: The article by Dr. Wilkins and colleagues comprehensively addressed this important topic. However, the section on blood transfusions is not consistent with the latest guidelines and does not address some important considerations. The article states that the threshold for blood transfusion should be 7 g per dL (70 g per L) unless there is ischemic heart disease, recent cardiac surgery, or hematologic malignancies (in which case the threshold is 8 g per dL [80 g per L]) based on the 2010 International Consensus Recommendations on the Management of Patients with Nonvariceal Upper Gastrointestinal Bleeding.¹ However, a 2019 update of the guideline recommends a threshold of 8 g per dL for all patients.² In addition, the guideline goes on to note that “the threshold recommendation does not apply to patients with exsanguinating bleeding. In the setting of acute blood loss, hemoglobin values may initially remain unchanged from baseline because of plasma equilibrium times. In such situations, transfusion should not be dictated by current hemoglobin level alone but should take into account the predicted drop in hemoglobin and the patient's clinical status.”

In the two trials that comprise nearly all of the data supporting the 2019 recommendation, exsanguinating bleeding was either not defined³ or included patients with shock (a systolic blood pressure less than 100 mm Hg or pulse greater than 100 beats per minute) who were transfused within two hours of arrival in the emergency department.⁴ Even if the bleeding has stopped, equilibration will almost certainly bring the hemoglobin to less than 8 g per dL even before the effects of hemodilution with crystalloids, which would be necessary to try to achieve hemodynamic stability. In such a situation, clinical judgment is critical, and in our view, opting to immediately transfuse such a patient is a reasonable choice. However, crystalloids should be started while waiting for the type and crossmatch to be done.

In Reply: Thank you for your thoughtful response to our article. Our article referenced the 2010 consensus recommendations regarding transfusions in patients with nonvariceal upper gastrointestinal bleeding.¹ We completed the final literature search for our article in November 2019. The 2010 consensus statement reports that the blood transfusion threshold should be 7 g per dL. The transfusion threshold is 8 g per dL in patients with ischemic heart disease, recent cardiac surgery, or hematologic malignancies. Drs. Ehrlich and Trow's letter to the editor referenced the consensus guidelines published in December 2019.² The 2019 consensus statement reports the new transfusion threshold of 8 g per dL for all patients. We agree with Drs. Ehrlich and Trow that the hemoglobin may lag in patients with ongoing and life-threatening upper gastrointestinal bleeding. Transfusion decisions in these situations must be individualized based on comorbidities, rate of bleeding, and clinical judgment.