Early red blood cell transfusion strategy in septic shock among patients with cancer: the TRANSPORT randomized controlled trial
In brief
Liberal transfusion triples thrombotic risk to 13% in cancer septic shock
In a French RCT of 187 cancer patients with septic shock, targeting a hemoglobin of 9 g/dL did not improve 12-hour lactate clearance or 28-day mortality compared with a 7 g/dL target. However, arterial or venous clots occurred in 13% of the liberal group versus 1% with restrictive transfusion. The findings suggest no mortality benefit and raise safety concerns about routine liberal transfusion in this high-risk population.
- Journal
- Intensive care medicine (Q1)
- Published
- 19 August 2026
- Study design
- Randomized controlled trial
- Evidence level
- Level 1, High (CEBM 1b)
- Authors
- Frédéric Pène, Clara Vigneron, Antoine Lafarge, Djamel Mokart, Yoann Zerbib, Emmanuel Canet, et al.
- PMID
- 42616082
- DOI
- 10.1007/s00134-026-08582-4
Why clinicians should know about it
- Picked for Critical Care and Intensive Care Medicine (top studies of the week, 23 August 2026): Early RBC transfusion RCT in septic shock cancer patients
Abstract
PURPOSE: Septic shock in cancer patients remains associated with a grim prognosis. With regard to the high prevalence of anemia, the optimal hemoglobin target to restore tissue oxygenation remains uncertain. METHODS: This was a multicenter superiority randomized controlled trial carried out in 18 centers in France. Adult patients with hematological or solid malignancies presenting with septic shock with lactate level > 2.0 mmol/L and hemoglobin level < 9.0 g/dL were randomly assigned to liberal or restrictive red blood cell (RBC) transfusion directed by hemoglobin thresholds of 9.0 g/dL or 7.0 g/dL during the first 48 h of resuscitation. The primary endpoint was 12-h lactate reduction, defined as normalization ≤ 2.0 mmol/L or relative decrease by 30% from baseline. Key secondary endpoints included ICU-acquired infections, arterial and venous thrombotic events, and 28-day mortality. RESULTS: The trial was designed to enroll 260 patients but was prematurely stopped for safety concerns. Finally, 187 patients were enrolled, distributed into 93 in the liberal arm and 94 in the restrictive arm. Twelve-hour lactate reduction was achieved in 52% and 50% of patients in the liberal and restrictive arms, respectively (odds ratio 1.07; 95% confidence interval 0.60 to 1.90; P = 0.82). Arterial or venous thrombotic events occurred more frequently in the liberal group (13% vs. 1%; P = 0.001), and 28-day mortality rates were 46% and 53% (P = 0.34), respectively. CONCLUSIONS: In cancer patients with septic shock, this trial did not demonstrate superiority of liberal over restrictive transfusion strategy. The absence of mortality benefit and the uncertainty surrounding the observed imbalance in adverse events do not support the routine preference of either transfusion strategy over the other.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.