The Society of Thoracic Surgeons Intermacs Database Annual Report: Evolving Indications, Outcomes, and Scientific Partnerships

Background: The Interagency Registry for Mechanically Assisted Circulatory Support (Intermacs), a joint effort among the National Heart, Lung, and Blood Institute, the Food and Drug Administration, the Centers for Medicare and Medicaid Services, and others, was established in 2005 at the University of Alabama at Birmingham. The registry examined clinical outcomes and quality-of-life metrics of patients who received an Food and Drug Administration-approved durable mechanical circulatory support (MCS) device to treat advanced heart failure. On January 1, 2018, the Intermacs Database became part of The Society of Thoracic Surgeons National Database, providing additional resources for quality assessment and improvement and scientific advancement.

Methods: The Intermacs Database Annual Report summarizes outcomes in patients (≥19 years of age) who underwent durable MCS implant between June 23, 2006, and December 31, 2017. Outcomes are presented for patients who underwent isolated continuous flow left ventricular assist device (CF LVAD) support, CF LVAD support with concomitant right ventricular assist device (RVAD) implant, or total artificial heart implant. Analyses of patients with CF LVADs are stratified by axial flow and centrifugal flow configurations. Because of the association of era with outcomes, the survival analyses are restricted to isolated CF LVADs implanted in the 2012 to 2016 era.

Results: There were 25,145 adult patients with MCS reported to Intermacs, of whom 18,539 (74%) received CF LVADs, 667 (2.6%) had an RVAD with CF LVAD, 339 received a total artificial heart (1.3%), and 20 (0.07%) received an isolated RVAD. Of the CF LVADs, mean age was 57 ± 1 years, 26% were listed for transplantation, and 51% were in cardiogenic shock (profile 1 to 2) preoperatively. CF LVADs included 14,527 axial flow (78%) and 4,012 centrifugal flow (22%) devices. Intermacs patient phenotype has evolved over time to include more patients with profile 3 (26% in 2006 to 2011 versus 35% in 2012 to 2016) and fewer patients with profile 2 (40% versus 35%), patients with better markers of preoperative renal and hepatic function, and more patients who received implants for destination therapy (29% versus 48%) indication. In 2017, centrifugal flow implants (51%) approximated that of axial flow devices (49%). Mean CF LVAD support duration was 20 months (31,563 patient-years). One-year survival for isolated CF LVADs was 83% and 5-year survival was 46%. One-year survivals for centrifugal versus axial flow devices were 85% and 84%, respectively. Patients who required concomitant RVAD support had 1- and 5-year survivals of 58% and 28%, respectively. Freedom from all-cause readmission was 70% at 1 month and 20% at 1 year. At 1 year, stroke occurred in 20% of patients on centrifugal flow and 13% of patients on axial flow support (p < 0.001), gastrointestinal bleeding affected 20% of patients with centrifugal flow devices and 25% of patients with axial flow devices (p < 0.001), and pump-related infection occurred in 28% of patients with centrifugal flow devices versus 25% of patients with axial flow devices (p = 0.01). Neurologic dysfunction (19% of deaths) and multisystem organ dysfunction (15%) were the most common causes of death.

Conclusions: With the evolution of MCS, patient phenotype and outcomes are also changing over time. CF LVAD support is increasingly being used in the less ill patient phenotype and more patients are supported for destination therapy. Mean survival is now approaching 5 years, but adverse events, especially neurologic events, continue to have a detrimental impact on the success of CF LVAD support.