Eastern Cardiothoracic Surgical Society

Prophylactic Application of Hemostatic Agents in Left Ventricular Assist Device Does Not Reduce Chest Tube Output nor Transfusion Requirements
Mahim A. Malik1, Arman Kilic2, William Rothstein1, Peter H. Lee1, Bryan A. Whitson1, Robert S. Higgins1,
Ahmet Kilic1. 1The Ohio State University, Columbus, OH USA, 2Johns Hopkins University, Baltimore, MD USA.


Objective: The objective of this study was to evaluate the impact of using topical hemostatic agents on postoperative bleeding in patients undergoing left ventricular assist device (LVAD) implantation.

Methods: Patients undergoing LVAD implantation between 2012 – 2014 at our institution were reviewed. Primary stratification was by use of hemostatic agent on the inflow suture line, and secondary stratification was by type of hemostatic agent. The primary outcome was requiring a blood transfusion in the first 48 hours postoperatively. Secondary outcomes included cumulative chest tube output in the first 48 hours postoperatively, number of red blood cell units transfused, and re-exploration for bleeding. Multivariable analyses were also conducted incorporating univariate predictors for risk adjustment.

Results: 90 LVAD implants were performed, and hemostatic agents were utilized in 56 (62.2%), including 29 (32.2%) with fibrin sealant and 27 (30.0%) with an albumin and glutaraldehyde based sealant. The groups were well matched with no between-group differences. There was no difference in blood transfusion rates in the first 48 hours (58.8% no sealant versus 51.8% sealant; p=0.52) (Table SA8-1). Multivariable analyses adjusting for potential confounders including use of heparin postoperatively confirmed that sealant use did not reduce the odds of requiring a blood transfusion or reduce chest tube output (each p>0.05).

Conclusions: In this series of LVAD implants, the use of hemostatic agents on the inflow suture line did not reduce the occurrence of clinically significant bleeding. In the setting of rising cost of healthcare, judicious use of hemostatic agents seems most prudent.

Table 1 – Outcome Comparison Between Sealant and No Sealant Groups

  Sealant group No sealant p-value
(n=34) (n=57)

Chest tube output (ml per patient) 2539.9+/-1197.2 2880.1+/-1921.1 0.36
Blood transfusion at 48 hrs (n/%) 20(58.8%) 29 (51.1%) 0.52
RBC units transfused (units per patient) 1.94 +/-2.49 2.12 +/-3.44 0.80
Re-exploration for bleeding (n/%) 5 (14.7%) 7 (12.5%) 0.77