Background/aim: The traditional approach to postcardiotomy shock includes inotropic support followed by the application of an intra-aortic balloon pump (IABP). Consideration toward insertion of a ventricular assist device (VAD) becomes necessary when these maneuvers fail to restore hemodynamic stability. The definition of maximal inotropic support, however, is lacking such that a standard formula for VAD insertion remains problematic. The purpose of this paper is to define the pharmacological thresholds for VAD implantation in the setting of postcardiotomy cardiogenic shock.
Methods: The medical records of all adult open-heart operations performed at Hahnemann University Hospital, Philadelphia, PA, from 1 July 1996 through 1 July 1999 were reviewed. Specific data were collected on the hemodynamics and inotrope levels upon separation from cardiopulmonary bypass (CPB). The hospital course was reviewed with attention toward documenting hospital mortality. Cardiogenic shock was defined as systolic blood pressure (SBP) < 100 mmHg, mean pulmonary artery blood pressure (mPAP) > 25 mmHg, central venous pressure (CVP) > 15 mmHg, and cardiac index (CI) < 2.0 L/min/per m2. Inotrope dosages were defined as low, moderate, and high according to assigned values. A formula for VAD insertion was established if cardiogenic shock parameters were present in the setting of two or more high dose inotropes. Early VAD insertion was defined as implantation within three hours of the first attempt to wean from CPB. The VAD recipients were divided into two groups. Group A were VADs placed in conjunction with the formula. Group B was VADs placed in violation (excess) of the formula. The results of these two groups were compared. [table: see text]
Results: From 1 July 1996 to 1 July 1999, there were 3462 adult open-heart operations performed at Hahnemann University Hospital, Philadelphia, Pa. The hospital mortality for patients successfully separating from CPB on no inotropes, low-dose, moderate-dose, one high-dose, two high-dose, and three high-dose inotropes were approximately 2.0%, 3.0%, 7.5%, 21%, 42%, and 80% respectively. During this time there were 29 patients supported with the Abiomed BVS (Danvers, Mass) system for postcardiotomy cardiogenic shock. For the entire group of VAD recipients, there were 18 (62%) who were successfully weaned and 8 (28%) who were discharged from the hospital. For the 20 VAD recipients in group A, there were 16 (80%) who were successfully weaned and 8 (40%) who were discharged from the hospital. For the nine VAD recipients in group B, there were two (22%) who were successfully weaned and zero (0%) who were discharged from the hospital. Multiple organ system failure occurred in three (15%) in group A versus seven (78%) in group B patients, respectively. Early VAD insertion was accomplished in 17 (85%) group A patients and 2 (22%) group B patients.
Conclusions: Hospital mortality correlates with the number and level of inotropic support necessary to separate from CPB following adult open heart surgery. The application of a standard pharmacological formula together with hemodynamic criteria for VAD insertion after postcardiotomy cardiogenic shock results in earlier insertion, lower incidence of postoperative MOSF, and improved wean and discharge rates.