Share this article on:

Risk Models in Mechanical Circulatory Support: A Conflation of Concepts

Kirklin, James K.

doi: 10.1097/MAT.0000000000000588
Invited Commentaries

From the Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama.

Submitted for consideration April 2017; accepted for publication in revised form April 2017.

Declaration: I have served as the Principle Investigator of the NHLBI INTERMACS project for the past 10 years.

Disclosure: The author has no conflicts of interest to report.

Correspondence: James K. Kirklin, MD, Department of Medicine, University of Alabama at Birmingham, 1900 University Building, THT 721, Birmingham, AL 35294. E-mail:

The study by Kanwar et al.1 examines the predictive value/relevance of risk factors identified in the HeartMate II risk score for 90 day and 1 year mortality in the context of Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) data. The authors found poor predictive value of these risk factors in a large contemporary INTERMACS cohort. To better understand the potential reasons for the authors’ findings, it is useful to review the origins and composition of INTERMACS.2 From its inception in 2006, the intent of the National Heart, Lung and Blood Institute (NHLBI), the federal funding agency, was to capture as completely as possible the contemporary “real-life” experience with durable (capable of outpatient care) devices in the United States, with stated goals that include understanding risk factors for poor outcomes, characterizing adverse events associated with this therapy, and identifying the patient cohorts that could experience good as well as poor outcomes with these devices.

The authors’ analysis is particularly cogent as a reminder of critical differences between clinical trial data and real-world experience. Clinical trials in device therapies represent a research model designed to test a specific hypothesis aimed at gaining regulatory approval for a new device or a novel application. Although such trials provide a standard for data completeness and accuracy, the resource and financial investments are high and the patient population is selected. For example, if the medical or surgical community believes that device approval would offer patient care advantages, bias may exist in patient selection for the trial, enrolling patients perceived to have a better chance for good outcomes. Thus, risk factor models generated from such a population may not be generalizable to the population at large.

Realizing this limitation as well as knowledge gaps in longer term circulatory support, the NHLBI sent out a request for proposals in 2005 for the establishment of a national registry for durable mechanical circulatory support (MCS) devices. The goal was to capture detailed, reliable longitudinal data for a large majority of patients in the United States supported with durable devices approved by the U.S. Food and Drug Administration (FDA).

Although patient data were entered into this “registry,” sufficient resources were available through the NHLBI contract to more closely mimic the rigor of a clinical trial with a design to develop a comprehensive web-based data entry system, follow patients for the duration of their MCS support, audit sites for accuracy and completeness of data collection, monitor center compliance with thresholds for adequate performance, develop definitions for adverse events that would comply with FDA standards, and include a formal process for review of adverse events. What began as a partnership among NHLBI, FDA, and the Center for Medicine and Medicaid Services evolved from a research enterprise into a major quality assurance vehicle for U.S. MCS hospitals. The name INTERMACS highlights its origins in this collaborative effort among the federal government, industry, and clinicians working in the field.

By design, INTERMACS has tracked the introduction and application of newly approved devices in the clinical arena. With the approval of the HeartMate II axial flow pump in 2008, the U.S. clinical experience with continuous flow technology was initiated. Following the nascent experiences with such devices, continuous flow technology grew to dominate the U.S. and global MCS market, accounting for greater than 95% of U.S. durable device implants.3 Predictably, the experience and devices have evolved over the past decade, all of which was chronicled and analyzed by numerous INTERMACS investigators.

So, the lack of concordance between risk factors generated 5 years ago from patients in 2 clinical trials4 and those identified in the current era of U.S. implants should not be surprising. After all, the pace of progress and the climate for change and innovation persist, both in device design and patient management. With nearly 20,000 continuous flow devices now entered into INTERMACS, the detail of risk factor identification reflects the subtle impact of numerous preimplant factors on survival and postimplant experiences. Furthermore, current models characterize multiple phases of risk, with differing risk factors operative early after implant and in the longer term (Table 1).3

Another confounding variable in our predictive modeling is the profound impact of adverse events post implant on subsequent survival. The profound negative impact of device malfunction requiring pump exchange,5 devastating strokes, and major infection all belie the simple prediction of survival by risk factors identified before or at the time of pump implant. Furthermore, the link between preimplant risk factors and subsequent adverse events is unpredictable. Patients who are stable at implant are just as susceptible to most adverse events as those who are critically ill.3 The concatenation of adverse events and their impact on survival will be a major focus of future INTERMACS investigations. Preliminary unpublished data indicate that the preoperative patient risk profile may pale in comparison with postimplant events in determining ultimate survival.

As the field matures and greater focus is placed on application of MCS to the ambulatory heart failure patient, neutralization and prevention of serious events becomes an increasing priority. More than the particular preimplant patient profiles, the timing, type, frequency, and sequelae of adverse events will define the future of MCS therapy.

Back to Top | Article Outline


1. Kanwar Manreet K, Lohmueller Lisa C, Kormos Robert L, et al. Low accuracy of the HeartMate risk score for predicting mortality using the INTERMACS registry data. ASAIO J 2017.63: 251–256.
2. Kirklin JK, Naftel DC, Stevenson LW, et al. INTERMACS database for durable devices for circulatory support: first annual report. J Heart Lung Transplant 2008.27: 1065–1072.
3. Kirklin JK, Naftel DC, Pagani FD, et al. Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015.34: 1495–1504.
4. Cowger J, Sundareswaran K, Rogers JG, et al. Predicting survival in patients receiving continuous flow left ventricular assist devices: the HeartMate II risk score. J Am Coll Cardiol 2013.61: 313–321.
5. Kirklin JK, Naftel DC, Pagani FD, et al. Pump thrombosis in the Thoratec HeartMate II device: an update analysis of the INTERMACS registry. J Heart Lung Transplant 2015.34: 1515–1526.

Mechanical circulatory support; HeartMate II; INTERMACS; risks; outcomes

Copyright © 2017 by the American Society for Artificial Internal Organs