A restrictive approach to blood transfusion reduced blood use by half but, more importantly, caused no harm to 998 critically ill patients with septic shock, according to results from a recent study published in the New England Journal of Medicine. A lower threshold also meant more patients were able to avoid transfusion.
"Blood transfusions are frequently given to patients with septic shock," the authors, led by Lars B. Holst, MD, from University of Copenhagen in Denmark, wrote. "However, the benefits and harms of different hemoglobin thresholds for transfusion have not been established."
Guidelines have generally advised transfusion to a hematocrit of 30 percent or hemoglobin of 10 g/dL during the first six hours of septic shock – if hypoperfusion is present despite fluids and vasopressors. The evidence supporting these recommendations, though, has been weak.
The reports from the current analysis of the TRISS trial may put an end to those practices. In this multicenter, parallel-group trial, Dr. Holst and investigators enrolled 998 patients with septic shock from 32 intensive care units (ICU) throughout Denmark.
Patients were randomized to transfusion throughout their ICU stay to maintain a hemoglobin concentration of â‰¥7 g/dL (lower threshold) or â‰¥9 g/dL (higher threshold). The primary endpoint was mortality at 90 days after randomization.
Ninety-nine percent of patients in the higher-threshold group were transfused, receiving a median of four units of blood while in the ICU; two-thirds of the lower-threshold patients were transfused, but received only a median of one unit. At 90 days after randomization, mortality was statistically similar in both groups: 216 of 502 patients (43.0%) assigned to the lower-threshold group had died, compared with 223 of 496 (45.0%) assigned to the higher-threshold group (p=0.44).
Secondary endpoints, including adverse events and the need for life support, were also equivalent in both groups:
- Ischemic events (lower vs. higher threshold): 7.2 percent versus 8 percent (relative risk [RR] = 0.90; 95% CI 0.58–1.39; p=0.64)
- Requiring life support at day 14: 36.8 percent versus 36.8 percent (RR=0.99; 95% CI 0.81–1.19; p=0.95)
- Alive without renal-replacement therapy (mean % of days): 85 percent versus 83 percent (p=0.54)
- Alive and out of the hospital (mean % of days): 30 percent versus 31 percent (p=0.89)
Where the groups differed was the amount of blood used in transfusion: There was a significant reduction in the number of blood transfusions between each group (low vs. high, 1,545 vs. 3,088; p<0.001), as well as the number of patients who received no transfusions at all (low vs. high, 176 vs. 6; p<0.001).
Future studies will focus on identifying safe transfusion thresholds for specific subgroups in this study – for instance, patients with cardiovascular comorbidities or myocardial ischemia (who were excluded from this analysis), Dr. Holst and authors concluded.
- Holst LB, Haase N, Wetterslev J, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med. 2014;371:1381-91.