ADVERTISEMENT

Lower hemoglobin threshold did not affect outcomes in septic shock

Author and Disclosure Information
Adopt threshold of 7 g/dL

Key clinical point: A hemoglobin transfusion threshold of 7 g/dL versus 9 g/dL did not affect outcomes in patients with septic shock.

Major finding: Ninety-day mortality was 43% for the 7-g/dL threshold group, compared with 45% for the 9-g/dL group.

Data source: TRISS, a multicenter, partially blinded, parallel-group randomized trial of 998 ICU patients with septic shock.

Disclosures: The study was funded by the Copenhagen University Hospital, Rigshospitalet, and several Scandinavian research foundations. Two coauthors reported financial relationships with Pharmacosmos, CSL Behring, and other companies. The other authors reported no conflicts of interest.

Patients with septic shock had similar outcomes when transfused at hemoglobin thresholds of 7 g/dL as compared with 9 g/dL, investigators reported online Oct. 1 in the New England Journal of Medicine.

Mortality at 90 days, use of life support, number of days alive and out of the hospital, number of ischemic events, and severe adverse reactions to blood in the intensive care unit (ICU) all were similar between the two groups, said Dr. Lars B. Holst at Rigshospitalet and the University of Copenhagen and his associates. At the same time, patients treated at the lower threshold received three times as many units of blood as did those treated at the higher threshold (median of 4 units and 1 unit, respectively), the researchers reported.

For the multicenter Transfusion Requirements in Septic Shock (TRISS) trial, the investigators randomized 998 ICU patients with septic shock to receive 1 unit of leukoreduced red cells when their hemoglobin level was either 9 g/dL or less (higher threshold) or 7 g/dL or less (lower threshold). Ninety days later, 43% of patients in the 7-g/dL threshold group had died, compared with 45% of the higher-threshold group, a nonsignificant difference, the investigators said. Controlling for baseline risk factors did not affect the finding, they reported (N. Engl. J. Med. 2014 Oct. 1 [doi: 10.1056/NEJMoa1406617]).

Rates of ischemic events and the need for life support also did not differ significantly between the two groups, Dr. Holst and his associates reported.

The study was funded by the Copenhagen University Hospital, Rigshospitalet, and several Scandinavian research foundations. Two coauthors reported financial relationships with Pharmacosmos, CSL Behring, and other companies. The other authors reported no conflicts of interest.