Send to

Choose Destination
Semin Thorac Cardiovasc Surg. 2019 Winter;31(4):783-793. doi: 10.1053/j.semtcvs.2019.05.006. Epub 2019 May 11.

Hemoadsorption to Reduce Plasma-Free Hemoglobin During Cardiac Surgery: Results of REFRESH I Pilot Study.

Author information

University of Pittsburgh Medical Center, Department of Cardiothoracic Surgery, Pittsburgh, Pennsylvania.
Columbia University Medical Center-New York Presbyterian Hospital, Department of Surgery, New York, New York.
Hospital of the University of Pennsylvania, Department of Surgery, Philadelphia, Pennsylvania.
Texas Children's Hospital, Department of Surgery, Houston, Texas.
Texas Heart Institute, Department of Surgery, Houston, Texas.
Baystate Medical Center, Department of Surgery, Springfield, Massachusetts.
University of Maryland School of Medicine, Department of Anesthesiology, Baltimore, Maryland.
Cooper University Hospital, Camden, New Jersey.
University of Kentucky College of Medicine, Department of Surgery, Lexington, Kentucky.
University of Kentucky College of Medicine, Department of Surgery, Lexington, Kentucky. Electronic address:


Generation of plasma-free hemoglobin (pfHb) and activated complement during complex cardiac surgery contributes to end-organ dysfunction. This prospective, multicenter REFRESH I (REduction in FREe Hemoglobin) randomized controlled trial evaluated the safety and feasibility of CytoSorb hemoadsorption therapy to reduce these factors during prolonged cardiopulmonary bypass (CPB). Eligible patients underwent elective, nonemergent complex cardiac surgery with expected CPB duration ≥3 hours. Exclusions included single procedures including primary coronary artery bypass graft, single valves, transplant, and left ventricular assist device extraction. TREATMENT used 2 parallel 300 mL CytoSorb hemoadsorption cartridges in a side circuit during CPB. CONTROL was standard of care. Of 52 enrolled patients, 46 underwent surgery (Safety group, n = 23 vs Control, n = 23), and 38 were evaluated for pfHb reduction (EFFICACY group, n = 18 vs CONTROL, n = 20). Type and number of serious adverse events (44 vs 43 CONTROL) were similar, as was 30-day mortality. Transient reduction in platelets during CPB was observed in both groups, especially TREATMENT, but returned to pretreatment levels after CPB without bleeding. Peak pfHb was positively correlated with CPB length (P = 0.01) but the high variability of pfHb, due to the broad surgical procedure mix, prevented detection of changes in pfHb in the overall EFFICACY population. However, the valve replacement surgery subgroup (8 vs 10 CONTROL) had the highest peak pfHb levels, and TREATMENT demonstrated significant pfHb reductions vs CONTROL (P ≤ 0.05) in CPB ≥3 hours. In the EFFICACY group, C3a and C5a were significantly reduced by treatment throughout surgery. Intraoperative hemoadsorption with CytoSorb was safe and feasible in this randomized, controlled pilot study during complex cardiac surgery. Treatment with CytoSorb resulted in significant reductions in pfHb during valve replacement surgery and reductions in C3a and C5a in the overall EFFICACY group. Future studies will target complex cardiac surgery patients with prolonged CPB to assess hemoadsorption effect on end-organ dysfunction and outcomes.


Cardiopulmonary bypass; Hemoadsorption; Plasma-free hemoglobin

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center