The Most Misinterpreted Study in Medicine

Ah, blood. That sweet nectar of life that quiets angina, abolishes dyspnea, prevents orthostatic syncope, and quells sinus tachycardia. As a cardiologist, I am an unabashed hemophile. 
But we liberal transfusionists are challenged on every request for consideration of transfusion. Whereas the polite may resort to whispered skepticism, vehement critics respond with scorn as if we’d asked them to burn aromatic herbs or fetch a bucket of leeches. And to what do we owe this pathological angst? The broad and persistent misinterpretation of the pesky TRICC trial. You know; the one that should have been published with a black-box warning stating, “Misinterpretation of this trial could result in significant harm.” 
Published in 1999, the TRICC trial enrolled critical anemic patients older than 16 years who were stable after fluid resuscitation and were not actively bleeding. They had a hemoglobin level < 9 g/dL and were expected to stay in the intensive care unit (ICU) for more than 24 hours. They were randomly assigned to either a conservative trigger for transfusion of < 7 g/dL or a liberal threshold of < 10 g/dL. Mortality at 30 days was lower with the conservative approach — 18.7% vs 23.3% — but the difference was not statistically significant (P =.11). The findings were similar for the secondary endpoints of inpatient mortality (22.2% vs 28.1%; P =.05) and ICU mortality (13.9% vs 16.2%; P =.29). 
One must admit that these P values are not impressive, and the authors’ conclusion should have warranted caution: “A restrictive strategy…is at least as effective as and possibly superior to a liberal transfusion strategy in critically ill patients, with the possible exception of patients with acute myocardial infarction and unstable angina.” 
Another criticism of TRICC is that only 13% of those assessed and 26% of those eligible were enrolled, mostly owing to physician refusal. Only 26% of enrolled patients had cardiac disease. This makes the TRICC population highly selected and not representative of typical ICU patients. 
To prove my point that the edict against higher transfusion thresholds can be dangerous, I’ll describe my most recent interface with TRICC trial misinterpretation 
The patient, Mrs Kemp*, is 79 years old and has been on aspirin for years following coronary stent placement. One evening, she began spurting bright red blood from her rectum, interrupted only briefly by large clots the consistency of jellied cranberries. When she arrived at the hospital, she was hemodynamically stable, with a hemoglobin level of 10 g/dL, down from her usual 12 g/dL. That level bolstered the confidence of her provider, who insisted that she be managed conservatively. 
Mrs Kemp was transferred to the ward, where she continued to bleed briskly. Over the next 2 hours, her hemoglobin level dropped to 9 g/dL, then 8 g/dL. Her daughter, a healthcare worker, requested a transfusion. The answer was, wait for it — the well-scripted, somewhat patronizing oft-quoted line, “The medical literature states that we need to wait for a hemoglobin level of 7 g/dL before we transfuse.” 
Later that evening, Mrs Kemp’s systolic blood pressure dropped to the upper 80s, despite her usual hypertension. The provider was again comforted by the fact that she was not tachycardic (she had a pacemaker and was on bisoprolol). The next morning, Mrs Kemp felt the need to defecate and was placed on the bedside commode and left to her privacy. Predictably, she became dizzy and experienced frank syncope. Thankfully, she avoided a hip fracture or worse. A stat hemoglobin returned at 6 g/dL. 
Her daughter said she literally heard the hallelujah chorus because her mother’s hemoglobin was finally below that much revered and often misleading threshold of 7 g/dL. Finally, there was an order for platelets and packed red cells. Five units later, Mr. Kemp achieved, a hemoglobin of 8 g/dL and survived. Two more units and she was soaring at 9 g/dL! 
There are many lessons here. 
The TRICC study found that hemodynamically stable, asymptomatic patients who are not actively bleeding may well tolerate a hemoglobin level of 7 g/dL. But a patient with bright red blood actively pouring from an orifice and a rapidly declining hemoglobin level isn’t one of those people. Additionally, a patient who faints from hypovolemia is not one of those people. 
Patients with a history of bleeding presenting with new resting sinus tachycardia (in those who have chronotropic competence) should be presumed to be actively bleeding, and the findings of the TRICC study do not apply to them. Patients who have bled buckets on anticoagulant or antiplatelet therapies and have dropped their hemoglobin will probably continue to ooze and should be subject to a low threshold for transfusion. 
Additionally, anemic people who are hemodynamically stable but can’t walk without new significant shortness of air or new rest angina need blood, and sometimes at hemoglobin levels higher than generally accepted by conservative strategists. Finally, failing to treat or at least monitor patients who are spontaneously bleeding as aggressively as some trauma patients is a failure to provide proper medical care. 
The vast majority of my healthcare clinician colleagues are competent, compassionate individuals who can reasonably discuss the nuances of any medical scenario. One important distinction of a good medical team is the willingness to change course based on a change in patient status or the presentation of what may be new information for the provider. 
But those proud transfusion conservatives who will not budge until their threshold is met need to make certain their patient is truly subject to their supposed edicts. Our blood banks should not be more difficult to access than Fort Knox, and transfusion should be used appropriately and liberally in the hemodynamically unstable, the symptomatic, and active brisk bleeders. 
I beg staunch transfusion conservatives to consider how they might feel if someone stuck a magic spigot in their brachial artery and acutely drained their hemoglobin to that magic threshold of 7 g/dL. When syncope, shortness of air, fatigue, and angina find them, they may generate empathy for those who need transfusion. Might that do the TRICC? 
*Some details have been changed to conceal the identity of the patient, but the essence of the case has been preserved. 
Melissa Walton-Shirley, MD, is a native Kentuckian who retired from full-time invasive cardiology. She enjoys locums work in Montana and is a champion of physician rights and patient safety. In addition to opinion writing, she enjoys spending time with her husband and daughters, and she sidelines as a backing vocalist for local rock bands. Her Heartfelt column was the 2022 northeast regional gold and national silver Azbee award winner.