Transfusing to 9 g/dL did not help cancer patients in septic shock, and clots rose

The edition in brief

TRANSPORT randomised 187 adults with cancer, septic shock and haemoglobin below 9 g/dL to a liberal (9 g/dL) or restrictive (7 g/dL) transfusion threshold for the first 48 hours. Lactate clearance at 12 hours was the same (52% vs 50%), arterial or venous thrombosis was more frequent with the liberal strategy (13% vs 1%), and 28-day mortality did not differ significantly (46% vs 53%). The trial stopped early for safety, so the restrictive threshold remains a reasonable default. In one US emergency department, a multicomponent programme raised outpatient management of low-risk pulmonary embolism (PESI 85 or less) from 5% to 33% over three years, and to 75% among those also negative on Hestia criteria, with no recurrent clots or deaths at 30 days after discharge. In a target-trial emulation of 3,810 septic shock patients from two US ICU databases, starting vasopressin within three hours of norepinephrine reaching 0.25 µg/kg/min was associated with lower 28-day mortality (48.1% vs 53.0%) and less renal replacement therapy than starting at three to six hours. A meta-analysis of seven trials found automated closed-loop oxygen control kept patients in their target saturation range 21% more of the time than manual adjustment. The pearl covers euglycaemic ketoacidosis in patients on SGLT2 inhibitors.