Is A Hemoglobin Concentration As Low As 7 g/dL Adequate For All Critically Ill Patients With Sepsis? Legitimate Doubts Remain!*
Recent clinical trials have focused on more narrow patient populations, including some that may be particularly vulnerable to the effects of anemia (5–9). For the most part, these studies confirm the safety of a restrictive threshold, although the definition of “restrictive” has, in some cases, been increased to 7.5 or 8 g/dL and the definition of “liberal” decreased from 10 to 9 g/dL. The potential adverse effects of transfusion, most notably a higher risk of nosocomial infection and acute respiratory distress syndrome, have also been highlighted in systematic reviews (10).
Nevertheless, some recent randomized studies do not confirm the safety of a restrictive approach. In a large trial of patients undergoing cardiac surgery, those receiving RBCs when the Hb concentration dropped below 9 g/dL had a lower risk of death compared with those transfused below 7.5 g/dL (8). In a study of critically ill surgical oncology patients, mortality was lower, and major complications less frequent, when the Hb concentration was maintained above 9 g/dL (9). A recent meta-analysis involving patients with cardiovascular disease suggested that acute coronary syndromes occur more often when the Hb concentration is permitted to fall below 8 g/dL (11).
Is sepsis a setting in which the optimal transfusion threshold may be different than in other populations? In the TRICC trial, sepsis was the primary diagnosis in only 5% of patients (1). Abnormal oxygen transport has traditionally been considered to be one of several mechanism contributing to organ dysfunction in sepsis, especially early in the disease process. Since the oxygen carrying capacity of blood is highly dependent on Hb concentration, the transfusion of RBCs was utilized as one component of “early goal directed therapy” (EGDT).
The optimal transfusion strategy in sepsis was best assessed in the Transfusion Requirements in Septic Shock (TRISS) trial, which compared thresholds of 7 versus 9 g/dL for the duration of the ICU stay among more than 1,000 patients in four countries and found no difference in mortality or organ dysfunction, including among patients with chronic cardiovascular disease (12). In addition, three multicenter trials found no mortality benefit with EGDT. Use of RBCs in the early hours of sepsis was significantly greater in the intervention arms of these EGDT trials, although transfusion was relatively uncommon in either group (4–14%) (13).
Based largely on these studies, the most recent Surviving Sepsis Campaign Guidelines recommend that “RBC transfusion occur only when Hb concentration decreases to less than 7.0 g/dL in adults in the absence of extenuating circumstances, such as myocardial ischemia, severe hypoxemia, or acute hemorrhage (strong recommendation, high quality of evidence)” (14). With such a definitive statement, one may question whether there is even the need for further studies.
In this issue of Critical Care Medicine, Dupuis et al (15) present a large, multicenter cohort study involving more than 6,000 septic patients over 16 years, to assess the relationship between RBC transfusion and outcomes.