“A Multicenter, Randomized, Controlled Clinical Trial of Transfusion Requirements in Critical Care” – Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group; February 11, 1999, The New England Journal of Medicine
Ask any physician when he initiates blood transfusions and you will get many different answers. The most appropriate one (and the one medical students should memorize for their wards services) is when it matters – when the patient has lost enough red blood cells (RBC) to cause symptoms. However, there are many situations where symptoms related to RBC loss or destruction are confounded by numerous other medical conditions. Patients who go to the Intensive Care Unit (ICU) are usually the most difficult to assess due to the severity and complexity of the illnesses that they present with. For this reason, this Canadian group attempted to set a standard threshold for when to initiate blood transfusions based on a simple, routine laboratory blood study – the hemoglobin concentration (Hgb). The Hgb is a study of the concentration of oxygen-carrying molecules in RBCs, and normal values for adults run from 13-15 g/dL.
Blood transfusions save lives. In the general sense, if your blood cannot carry oxygen, then there is no point in breathing it in. If your Hgb drops too low, it is just as bad as suffocating. However, blood transfusions are also high-risk and harmful interventions. When you receive blood from another individual, your own immune system recognizes the blood as foreign. This leads to an immune response, whether severe or mild, which may divert resources away from fighting the pathogen that was originally the reason for the illness and further weakening the body. Additionally, there is always the risk of acquiring blood borne illnesses, such as Hepatitis C or HIV, despite the screeningefforts of blood banks. Due to the value and risk, blood transfusions must be used carefully, and the TRICC trial was vital to understanding how to do so.
This trial split patients into two groups. The first, the restricted group, was given blood transfusions only when their Hgb dropped below 7, and their Hgb was maintained in the 7 to 9 range. The second, the liberal group, was transfused any time the Hgb fell below 10, and their Hgb was maintained in the 10 to 12 range. The primary outcome, all cause mortality at 30 days, showed no statistical difference between the two groups. However, when the analysis was repeated for patients who were less than 55 years old, there was significantly less mortality in the restrictive (transfuse at 7) group. The analysis was also repeated for patients who were less sick, as defined by an APACHEII score of <20 (APACHEII is a score calculated from multiple physiologic factors (vital signs, lab values, etc.), on a range from 0-71 that indicates increasing severity of disease and risk of death with an increasing score). In this sub-analysis as well, there was reduced mortality in the restrictive group. For patients older than 55 and with an APACHEII > 20, there was no difference in 30 day mortality between the restrictive and liberal groups.
Why We Do What We Do
The authors of this study presented their results in a very objective and humble manner. Even though there was a clear trend towards reduced mortality in the restrictive group for the entire study (18.7 percent mortality in the restrictive group vs. 23.3 percent mortality in the liberal group), they refused to acknowledge it due to lack of statistical significance (P = 0.11). At the least, this validates that there is no additional harm done if transfusions are restricted to patients with Hgb < 7. Patients in the restricted group received less blood overall in this study, reducing their risk for contracting transfusion-associated infectious diseases and major transfusion-associated complications, such as lung injury (TRALI) or cardiac overload (TACO). For younger patients and those who were not as severely ill, it is clearly apparent that over-transfusion is actually deleterious, and this is statistically significant. So the next time you approach a patient with possible blood requirements, it is not a bad idea to use the Hgb of 7 as a starting point to guide management.
Hébert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, Tweeddale M, Schweitzer I, Yetisir E. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med. 1999 Feb 11;340(6):409-17. Erratum in: N Engl J Med 1999 Apr 1;340(13):1056. PubMed PMID: 9971864. http://www.ncbi.nlm.nih.gov/pubmed/9971864