Surgically Corrected Mitral Regurgitation During Left Ventricular Assist Device Implantation Is Associated With Low...

Indications for concomitant intervention for mitral regurgitation (MR) during left ventricular assist device (LVAD) implantation remain controversial. The objective of this study was to determine the impact of the surgical correction of MR during LVAD implantation.

From July 2008 to December 2014, 164 patients with significant preoperative MR underwent LVAD (HeartMate II; Thoratec, Pleasanton, CA) implantation. The MR resolved after LVAD implantation in 110 of 164 patients (67.1%) with either surgical or spontaneous correction. The cohort (n = 110) without significant postoperative MR was divided into two groups: a spontaneous correction group (n = 54, MR spontaneously resolved after LVAD implantation); and a surgical correction group (n = 56, MR surgically corrected). Patients who received aortic valve procedures (n = 17) were excluded from this study.

Patient demographics, perioperative outcomes including bleeding, prolonged intubation, and stroke, and inhospital mortality did not differ in the two groups except for significantly longer cardiopulmonary bypass time in the surgical correction group (spontaneous correction 123 minutes [interquartile range (IQR): 107 to 150] versus surgical correction 177 minutes [IQR: 132 to 198], p < 0.001). During follow-up, pulmonary wedge pressure (spontaneous correction 17 mm Hg [IQR: 12 to 23 mm Hg] versus surgical correction 12 mm Hg [IQR: 4 to 17 mm Hg], p = 0.015) and pulmonary vascular resistance (spontaneous correction 2.0 Wood units [IQR: 1.5 to 2.4] versus surgical correction 1.7 Wood units [IQR: 0.8 to 2.1], p = 0.047) were significantly improved in the surgical correction group compared with the spontaneous correction group. Overall survival rate and freedom from recurrent MR were significantly better in the surgical correction group compared with the spontaneous correction group (1-year survival, spontaneous correction 59.4% ± 6.9% versus surgical correction 69.6% ± 6.4%, log rank p = 0.030; 1-year freedom from recurrent MR, spontaneous correction 76.2% ± 7.5% versus surgical correction 95.0% ± 3.5%, log rank p = 0.028).

The LVAD patients with surgically corrected MR had improved midterm hemodynamics and survival compared with spontaneously resolved MR, along with low recurrence of MR. Aggressive surgical mitral valve intervention during LVAD implantation may be recommended.