The Use of a Viscoelastic-Based Transfusion Algorithm Significantly Reduces Non-red Blood Cell Transfusion in Patients Undergoing Left Ventricular Assist Device Placement or Heart Transplantation: A Single-Center Observational Study.

OBJECTIVES

Does point-of-care viscoelastic testing in patients undergoing left ventricular assist device implantation or orthotopic heart transplantation reduce non-red blood cell transfusion or improve postoperative outcomes?

DESIGN

A retrospective observational study.

SETTING

At a single-center tertiary university hospital.

PARTICIPANTS

Patients undergoing left ventricular assist device placement or heart transplantation INTERVENTIONS: The authors implemented a TEG-based transfusion algorithm to reduce non-red cell transfusion rates compared with historical controls.

MEASUREMENTS AND MAIN

RESULTS

From May 15, 2019, through March 20, 2020, 68 patients underwent left ventricular assist device placement or heart transplantation. Algorithm adherence was 49.2%. After adjusting for relevant variables, platelet (odds ratio [OR] 0.58 [0.39-0.84]; p = 0.004) and cryoprecipitate (OR 0.37 [0.19-0.72]; p = 0.004) transfusion rates and time to extubation (OR -14.1 [-25.8 to -2.3]; p = 0.020) were significantly reduced compared with historical controls. After adjusting for relevant clinical variables, there was a statistically significant reduction in plasma (median [interquartile range] 0.16 [0.07-0.36], p < 0.001), platelets (0.06 [0.02-0.21], p < 0.001), and cryoprecipitate (0.06 [0.01-0.47], p = 0.007) transfusion rates and time to extubation (-16.95 [-27.20 to -6.71], p = 0.002) compared with historical controls.

CONCLUSIONS

The authors report a statistically significant reduction in transfusion of platelets and cryoprecipitate and time to extubation after adjusting for relevant clinical variables compared with historical controls and a significant reduction in the transfusion of plasma, platelets, and cryoprecipitate and time to extubation in those patients for whom the transfusion algorithm was followed. Their results suggest the importance of implementing transfusion algorithms for patients undergoing heart transplantation and left ventricular assist device placement and of accounting for adherence.