Hypovolaemic phlebotomy in patients undergoing hepatic resection at higher risk of blood loss (PRICE-2): a randomised controlled trial.

Background: Blood loss and subsequent red blood cell transfusions are common in liver surgery. Hypovolaemic phlebotomy is associated with decreased red blood cell transfusion in observational studies. This trial aimed to investigate whether hypovolaemic phlebotomy is superior to usual care in reducing red blood cell transfusions in patients undergoing liver resection.

Methods: PRICE-2 was a multicentre, single-blind, superiority randomised controlled trial. Patients at a higher risk of blood loss undergoing liver resection for any indication at four Canadian academic tertiary-care hospitals were randomised to receive hypovolaemic phlebotomy or usual care. Hypovolaemic phlebotomy consisted of the removal of 7-10 mL/kg of whole blood, without volume replacement, before liver transection. Patients were randomised centrally using permuted blocks of randomly variable length, stratified by centre. The randomisation sequence was computer-generated by an independent statistician. Surgeons, patients, and outcome assessors were masked to treatment allocation. The primary outcome was perioperative red blood cell transfusion to 30 days post-randomisation, analysed in all randomly assigned patients who underwent liver resection. PRICE-2 trial was registered with ClinicalTrials.gov (NCT03651154) and is completed.

Findings: Between Oct 1, 2018, and Jan 13, 2023, 486 individuals were randomly assigned to receive hypovolaemic phlebotomy (n=245) or usual care (n=241). 22 individuals in the hypovolaemic phlebotomy group and 18 in the usual care group did not undergo liver resection and were thus excluded from the primary analysis population. 223 patients were included in the hypovolaemic phlebotomy group (mean age 61·4 years [SD 13·0]; 137 [61%] men) and 223 in the control group (62·1 years [12·1]; 114 [51%]). 17 (8%) of 223 patients allocated to hypovolaemic phlebotomy and 36 (16%) of 223 patients allocated to usual care had a perioperative red blood cell transfusion by 30 days (difference -8·8 percentage points [95% CI -14·8 to -2·8]; adjusted risk ratio [aRR] 0·47 [95% CI 0·27 to 0·82]). Severe complications to 30 days occurred in 37 (17%) patients allocated to hypovolaemic phlebotomy and 36 (16%) allocated to usual care (aRR 1·06 [95% CI 0·70-1·61]). Overall complications to 30 days occurred in 135 (61%) of 223 patients allocated to hypovolaemic phlebotomy and 116 (52%) of 223 patients allocated to usual care (1·08 [0·92-1·25]). There was no postoperative mortality to 90 days.

Interpretation: In patients undergoing liver resection, hypovolaemic phlebotomy reduced perioperative red blood cell transfusion and improved operative conditions, with no statistically significant increase in the incidence of complications compared with usual care. Hypovolaemic phlebotomy should be considered for routine use in patients undergoing liver resection at higher risk of bleeding.

Funding: Canadian Institutes of Health Research (PJT-156108).