Send to

Choose Destination
Ann Thorac Surg. 2013 Oct;96(4):1259-1265. doi: 10.1016/j.athoracsur.2013.05.081. Epub 2013 Aug 20.

Surgical technique influences HeartMate II left ventricular assist device thrombosis.

Author information

Department of Surgery, Temple University Hospital, Philadelphia, Pennsylvania.
Yale University School of Medicine, New Haven, Connecticut.
Biostatistics Consulting Center, Temple University School of Medicine, Philadelphia, Pennsylvania.
Section of Cardiac Surgery, Yale University School of Medicine, New Haven, Connecticut.
Section of Cardiac Surgery, Yale University School of Medicine, New Haven, Connecticut. Electronic address:



Thrombosis of the HeartMate II (HM2 [Thoratec Corporation, Pleasanton, CA]) is a potentially devastating complication. While attention has been focused on anticoagulation strategies to prevent this complication, the impact of surgical technique has not been assessed.


Patients undergoing HM2 implantation at two institutions were reviewed. Pump thrombosis (PT) was defined as a clinical syndrome that included more than 30% elevation in pump power, more than 30% elevation in lactate dehydrogenase, and greater than 20% decrease in hemoglobin with the presence of thrombus in the HM2 stator or rotor, or both, at explant or autopsy. A blinded clinician reviewed dimensions and angles of the HM2 obtained from chest x-ray films. Patients demonstrating PT were compared with patients having normal function.


Of the 49 patients reviewed, 11 (22.4%) displayed evidence of PT at a median of 42 days after HM2 implantation. Patient with PT had greater acute angulation of the HM2 inflow cannula immediately postoperatively (48.2 ± 6.8 versus 65.4 ± 9.2 degrees, p < 0.001) and after 30 days (50.1 ± 8.0 versus 65.1 ± 9.9 degrees, p < 0.001). Pump pocket depth was lower in the PT group immediately after HM2 implantation (107.0 ± 41.9 versus 144.3 ± 20.3 cm, p < 0.001) and after 30 days (86.0 ± 39.1 versus 113.1 ± 25.4 cm, p = 0.02). Patients with evidence of PT did not have a decrease in end-diastolic diameter (76 ± 9 versus 70 ± 15 mm, p = 0.24) whereas patients in the normal function group had effective remodeling of the left ventricle (70 ± 10 versus 56 ± 12 mm, p = 0.01).


Meticulous surgical technique, which necessitates creating an adequately sized pump pocket and appropriately directing the inflow cannula at the time of operation, may reduce the risk of PT.


27; CT; CXR; HM2; HeartMate II; IC; INR; INTERMACS; Interagency Registry for Mechanically Assisted Circulatory Support; LVAD; LVEDD; NF; OC; PP; PT; chest roentgenogram; computed tomography; inflow cannula; international normalized ratio; left ventricular assist device; left ventricular end-diastolic diameter; normal functioning; outflow cannula; pump pocket; pump thrombosis

Comment in

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center