Approximately 6.5 million adults in the United States have heart failure, accounting for nearly 1 million ED visits annually.1 Advanced heart failure is particularly difficult to treat, and is associated with significant morbidity and mortality. While medical therapy is the initial treatment for patients with advanced heart failure, it has limited effectiveness; therefore, at the present time, heart transplant is the most effective treatment for heart failure refractory to medical management.
According to the 2013 Registry of the International Society for Heart and Lung Transplantation, 4,096 cardiac transplants were performed worldwide in 2011, approximately 2,000 of which were done in the United States.2
The average age of a heart transplant recipient in the United States is 55 years.2 In 2017, there were nearly 4,000 patients on the United Network for Organ Sharing, the organization that manages the national transplant waiting list in the United States and matches donors to recipients.3 Unfortunately, the number of patients requiring a heart transplant far exceeds the number of registered donors, and a large number of patients must wait years for transplantation. In addition to those awaiting a heart transplant, there are many patients with advanced heart failure who are not suitable candidates for transplant (usually due to age).
Left Ventricular Assist Devices
As of December 31, 2016, a total of 22,866 US Food and Drug Administration (FDA)-approved devices were listed in the Interagency Registry for Mechanically Assisted Circulatory Support, 17,016 of which were continuous-flow (CF) left ventricular assist devices (LVADs), including the HeartMate II (HMII) (Abbott Laboratories) and the HeartWare Ventricular Assist Device (HVAD) (Medtronic).4 Left ventricular assist devices, which have been in use for over 30 years, have evolved into smaller, quieter, and more durable devices. The current generation of LVADs has a CF design (as opposed to the older pulsatile-flow [PF] design). More importantly, CF LVADs are associated with higher survival rates and increased quality of life than the earlier PF models.5 For these reasons, CF LVADs are being used much more frequently today. As previously noted, LVADs serve as a temporizing measure for patients awaiting a heart transplant (ie, bridge-to-transplant therapy [BTT]) or as the primary treatment for patients who are not suitable candidates for transplant (ie, destination therapy [DT]).
The percentage of patients receiving an LVAD as a DT has increased from around 15% between 2006 to 2007 to nearly 46% in 2014.6Recently, several reports following LVAD patients demonstrated a reverse remodeling of the heart and recovery of native cardiac function that was sufficient enough in some patients as to permit LVAD removal (ie, bridge to recovery).7 In the United States, the number of patients undergoing LVAD removal due to recovery remains fewer than 3%.6With the increase in the number of patients receiving LVADs, there is an increased likelihood of LVAD patients presenting to an ED due to device-related complications. Recognized complications associated with LVADs include thrombosis, infection, bleeding, and issues with volume status.5,7 However, the frequency of LVAD-associated complications and the final disposition of these patients is less well known.
HeartMate II Patient ED Presentation Study
The purpose of our study was to identify the reasons for LVAD patient presentation to the ED, the frequency of these presentations, and the final disposition of these patients. Our institution, Sentara Norfolk General Hospital (SNGH), is a level I trauma and a tertiary care referral center, and it is the only hospital in a large area of Virginia to perform LVAD implantation.
Our study involved only patients implanted with the HMII LVAD.
Patients and Study Design
This was a retrospective study of patients with an HMII LVAD who presented to the SNGH ED between April 1, 2009 and September 9, 2012. All patients implanted with an HMII LVAD during the study period were assigned a study number linking the patient to their medical record number and social security number. Study numbers were assigned at the time of LVAD implantation by one of the investigators. This document was kept in a secure and locked location in the department of emergency medicine and was not accessible to anyone other than study investigators.
The electronic medical records were retrospectively reviewed to identify any HMII LVAD patient presenting to the SNGH ED during the study period. Information abstracted from the ED medical records included patient age, sex, initial complaint, final diagnosis, and disposition. Only the patient’s assigned study number was used on the data collection form, and no personal identifying information was present.
This study was granted approval for human subject research by the Eastern Virginia Medical School Institutional Review Board. Eligible patients included all patients with an HMII LVAD implanted during the study period. Study patients who presented to the SNGH ED between April 1, 2009 and September 9, 2012 were identified by a retrospective chart review. These patients were instructed to specifically seek care at the SNGH ED in the event of an emergency. There were no exclusion criteria.
Data were collected and reported in real numbers and percentages. No formal statistical analysis was used in evaluating the results.
Between April 1, 2009 and September 9, 2012, there were a total of 98 patients with an HMII LVAD that had been implanted during the study period at SNGH. The average patient age was 53.6 years, with a range from age 20 years to 78 years. Sixty-seven (68%) of the patients enrolled in the study required at least one ED visit. The HMII LVAD patients who presented to the ED ranged in age from 20 years to 78 years, with an average age of 53.1 years. The average number of ED visits by these 67 patients was 3.7, with a range of 1 to 12. Approximately 56% of the ED visits were directly LVAD-related. In all, 67 patients were responsible for a total of 248 ED visits.
The two most common reasons for presentation to the ED involved bleeding and volume overload. A total of 37 ED visits (14.9%), were related to bleeding, which included gastrointestinal (GI) bleeding (18/37 or 49%), epistaxis, hematuria, gingival bleeding, and postoperative bleeding following tooth extraction.
Volume overload accounted for 37 ED visits (14.9%), and the most common presenting symptom in these patients was shortness of breath. Other reasons patients presented to the ED were weakness/lightheadedness/dizziness/syncope (24/9.6%), device malfunction (20/8.1%), infection (7/2.8%), and transient ischemic attack/cerebrovascular accident (6/2.4%). For infection-related ED visits, two presentations (2.9%) involved a driveline infection. Common causes for ED visits related to device malfunction included battery failure and device-alarm activation. Overall, 142 of the 248 total ED visits (57.3%) resulted in hospital admission. One patient in the study presented in cardiac arrest and could not be resuscitated.
The remaining 108 LVAD patient ED visits (44%), did not appear to be related to the presence of the LVAD, but rather represented common reasons for presentation to an ED. These other non-LVAD-related reasons for presentation to the ED were due to motor vehicle incidents (3); assault (2); dental pain (3); mechanical fall (5); and upper respiratory tract infection (4), and represented small groupings of patient reasons for an ED visit.
Examples of singular reasons for presentation to the ED included one patient who presented with suicidal ideation, and another patient who presented for evaluation of symptoms suspicious for a sexually transmitted infection.
As the number of patients with advanced heart failure continues to increase, the number of those with an LVAD also increases. Between 2006 and June 2013, nearly 9,000 adult patients in the United States received a durable LVAD.6 In the early years of LVAD implantation, patients were restricted to remain in proximity of geographical areas surrounding academic health care centers. An increased comfort level by both physicians and patients now allows LVAD patients to reside in more distant communities. This increase in LVAD implantation, coupled with the widening patient distribution, make it important for every emergency physician (EP) to have a working knowledge of the device and its associated complications. To date, the characteristics and frequency of LVAD patient presentations to the ED have not been well characterized.
Left ventricular assist devices are considered in patients who have significant symptoms associated with poor LV function or who cannot maintain normal hemodynamics and vital organ function. Continuous-flow LVADs account for almost all devices currently implanted. During our data-collection period, there were two FDA-approved implantable LVADs—the HMII, approved for BTT in 2008 and for DT in 2010; and the HVAD approved for BTT in 2012. In August 2017, HeartMate III (Abbott Laboratories) was approved by the FDA. All patients enrolled in our study were recipients of the HMII device, as this was the only type of LVAD implant performed at our hospital. Current survival with the HMII LVAD is 80% at 1 year and 69% at 2 years, and there has not been shown to be a significant difference when stratified by era of implant.6
Device Designs and Structures
The pump of the HMII is inserted into the abdominal cavity, whereas the HVAD is implanted in the chest cavity, with the inflow cannula in the apex of the LV and the outflow cannula connecting to the proximal aorta. Blood is continuously pumped through the system.8,9 The pump is connected to a driveline that exits the body and connects to a controller. Continuous-flow devices have either an axial or centrifugal blood pump. Axial devices have an impeller that is connected to ball-and-cup bearings that accelerate blood along its axis. Newer axial flow pumps incorporate magnetic levitation of the rotor and do not require the use of bearings. Centrifugal devices accelerate blood circumferentially with a rotor that is suspended within in the blood pool by electromagnetic or hydrodynamic forces.10 The controller is powered by two external batteries or connected to a power base unit where the pump can be interrogated. The controller is usually housed in a garment worn by the patient, one that also includes the batteries. The controller can also be powered by a base unit that can be plugged into an electrical outlet.11