Advanced Cardiac Therapies Improving Outcomes Network Research Articles

Intracorporeal VAD Outcomes in the ACTION Quality Improvement Network

The initial project of ACTION (Advanced Cardiac Therapies Improving Outcomes Network) is focused on improving stroke rates in patients implanted with a ventricular assist devices (VAD) in participating centers. The initial interventions targeted improving communication, anticoagulation, and blood pressure management. We sought to describe the characteristics and outcomes of patients implanted with intracorporeal, continuous flow VADs (cfVAD). The ACTION registry was used to determine the stroke frequency in cfVAD implanted patients between April 2018 and September 2019, the time from development of the quality improvement bundle to the present. Sixty two patients (33 Heartmate3 [HM3] and 29 Heartware [HVAD]) were identified (Table). The majority had dilated cardiomyopathy (76%). Of the 12 (20%) patients with congenital heart disease (CHD), 9 (75%) had single ventricle (SV) physiology. The HM3 patients were older (16.5 yr vs 12.8 yr) and larger (64.2 kg vs 51.9 kg) than the HVAD patients. No significant difference was observed in diagnosis, duration of support, device strategy, or clinical outcome. One patient (2%) experienced a stroke during 6,920 days of support. Forty of the 45 (89%) patients to reach a clinical endpoint experienced a positive outcome (transplant or explant for recovery) while 17 (27%) patients remain on device. The stroke rate is low and survival good among patients implanted with an intracorporeal cfVAD at ACTION centers using the clinical protocols developed within the network. The HVAD is currently being used more frequently in younger and smaller patients. There has been an appreciable increase in the number and frequency of implants in patients with CHD, especially SV disease, as compared to previous reports. Future efforts within ACTION will focus on improving clinical outcomes as devices are increasingly used at individual centers and as the number of VAD-supported patients with congenital heart disease continues to grow.

Discharge and Readmissions after VAD Placement in US Pediatric Hospitals: A Collaboration in ACTION

There is a concerted effort to increase discharges of children on ventricular assist device (VAD) support. Optimizing this process represents a quality improvement target of Advanced Cardiac Therapies Improving Outcomes Network (ACTION). We sought to describe discharge and readmission frequency in children on VAD support using a multicenter administrative database. All VAD-implanted patients aged 10-21 years at ACTION centers were identified from the Pediatric Health Information System database (2009-2018). Discharge frequency on VAD was calculated. Patients discharged on VAD were compared to those not discharged. Freedom from readmission was assessed using the Kaplan Meier method, censoring at readmission for transplantation. The indications for readmission were determined using ICD diagnosis codes. A total of 298 patients from 25 centers were identified, of which 163 (54.7%) were discharged. The discharge frequency ranged from 8% to 100% across centers. The frequency of discharge increased over time (36.9% [2009-2012] vs. 59.7% [2013-2018], p=0.001). Of 144 discharged patients with follow-up data, 96 (66.7%) were readmitted for reasons other than transplantation with a median time to readmission of 45 days (IQR 18-79) [Figure]. Heart failure was the most common reason for readmission (27.7%), followed by infection (25.8%), hematologic concerns (16.8%), neurologic events (8.6%), arrhythmias (7.8%), and fluid/electrolyte problems (4.7%). In-hospital mortality on readmission was uncommon (1.8%) and the median length of stay was 6 days (IQR 2-19 days). Patients discharged with VADs from ACTION centers has increased over time, with high variability across centers. Readmissions are common with diverse indications; however, the risk of mortality during a readmission encounter is low. Further collaboration in ACTION is critical to optimize the outpatient management of this vulnerable population.

Contemporary Berlin Heart EXCOR Outcomes in North America: Report from the ACTION Registry

Berlin Heart EXCOR was introduced in North America almost two decades ago. Field of pediatric ventricular assist device, as it relates to available devices, anticoagulation strategies and patient population, has changed since then. This study reports the outcomes of most recent cohort of EXCOR patients. Advanced Cardiac Therapies Improving Outcomes Network (ACTION) is a learning collaborative of over 40 pediatric heart failure programs worldwide, which collects data for all ventricular assist device (VAD) implants as one of its initiatives. All EXCOR implants reported to ACTION (n=52) from April 2018 - Sep 2019 in North America were included in the analysis and compared to previously reported EXCOR study (IDE and Post Approval - n=187) and database (also include compassionate use patients at non-IDE sites - n=320) cohorts (2007-2015). ACTION cohort was younger, smaller in weight/BSA, more likely to have congenital heart disease (CHD) diagnosis and less likely to receive a BiVAD compared to EXCOR study and database cohorts. [Table1] Patients in the ACTION cohort were less likely to be in INTERMACS profile 1 and were supported for longer duration. Of the total, 89% (46) of the ACTION cohort was on Bivalirudin for anticoagulation. Success, defined as transplant, weaned or alive on device was 83% for the ACTION cohort. Incidence of stroke reduced by 51% and pump thrombosis reduced by 65% in the ACTION cohort compared to the EXCOR study cohort. Similarly, all other adverse events including major bleeding reduced significantly in the ACTION cohort. Use and outcome of Berlin Heart EXCOR has also evolved over time. EXCOR has experienced early implantation, more utilization in CHD, longer support duration and more importantly, improved outcomes.

Variation in Cardiac Rehabilitation for Pediatric Ventricular Assist Device Recipients across North America: An ACTION Network Survey

Despite increasing utilization of ventricular assist devices (VAD) in pediatric heart failure, data on exercise testing and training through cardiac rehabilitation (CR) programs have not been reported to date in this population. We aimed to describe variation in attitudes and practice and understand barriers to CR among pediatric VAD centers in North America. An online survey was performed through the Advanced Cardiac Therapies Improving Outcomes Network (ACTION) representing pediatric VAD centers across North America. Descriptive statistics were performed. For multiple-choice questions, more than one answer choice was allowed. A multidisciplinary cohort of 52 respondents from 28 North American pediatric VAD centers completed the survey. Although 52% of respondents reported conducting exercise tests on VAD patients, most of those were six minute walk tests (89%) instead of formal stress tests (32% stationary bike and 16% treadmill). In regards to exercise training, while all centers utilize physical therapy early post-implant, only half (53%) refer to a formal CR program. If CR is performed, it is usually done at the ACTION center (84%), a local specialized center (21%), or a mix of the two (5%). A few (5/19, 26%) reported utilizing a home-based CR program. The most commonly reported barriers to exercise testing in pediatric VAD patients were inadequate center resources and expertise, and a perception of low clinical utility. The dominant reported barriers to standardized CR were inadequate staff to supervise sessions (70%), inability of patients/caregivers to travel to the CR facility (55%), and inadequate knowledge of how to implement a CR program (45%). Safety concerns were relatively rare (5%). Over 90% of respondents were interested in implementing a standardized CR program at their center. There is significant variation in attitudes and practice regarding exercise testing and training for pediatric VAD patients among North American centers. Despite perceived barriers, there is broad interest in a standardized CR program. Based on these survey data, we are implementing a CR toolkit across ACTION centers with the aims of harmonizing practice, as well as demonstrating feasibility, safety, and early efficacy.

“How are You Doing?”: Patient and Family Support Resources for Children on Ventricular Assist Device

While psychiatric morbidity is associated with poor outcomes in adults on ventricular assist device (VAD), psychiatric concerns are often under recognized in pediatric patients and families. To optimize mental health support of these groups, we sought to first assess the availability and utilization of ancillary services to patients and families within a large pediatric VAD network. This was a descriptive cross-sectional survey of ACTION (Advanced Cardiac Therapies Improving Outcomes Network) centers. Because no applicable survey instrument exists, content experts within ACTION developed the survey. Questionnaires were distributed to 31 centers. Only one completed survey per center was included in the analysis. Descriptive statistics as well as response distributions (numbers and percentages) were calculated for key variables of interest. Between March and April 2019, data were collected from 24/31 (77%) centers. There were 28 completed responses (4 duplicates eliminated). Psychiatry and psychology services are available at 90% of centers with 57% of centers consulting one or both of these services for all patients unless declined by patient and family. One-third (35%) of centers consult these services only when mental health concerns are raised. Antidepressants are seldom used by centers (14%; 3/24) for pediatric VAD patients. Antidepressant use is guided by mental health specialists at the majority (82%) of centers. Other patient and family support resources available at participating centers include: palliative care (96%), family advisory council (55%), family resource center (70%), and parent support groups (37%). Interestingly, the practice of meeting parents of other VAD patients was reported by 66% of centers. Primary nursing assignments for pediatric VAD patients was reported to be "often" or "almost always" in only 35%, and 21% respectively. Typical resources to patients and family are available but may be underutilized. This survey identifies targets for care standardization and process improvement to allow broader access to these resources, as well as implementation of novel parent/patient support approaches used at other ACTION centers.

Understanding Pediatric VAD Center Practices for Establishing and Maintaining Optimal Support

Ventricular assist device (VAD) interrogations allow clinicians to establish and maintain optimal VAD settings through the support course. There is currently no data regarding standards for establishing and modifying pediatric VAD settings, despite a heterogeneous population with variable diagnoses, anatomy and physiology. The Advanced Cardiac Therapies Improving Outcomes Network (ACTION), a 37-center North American learning network, was surveyed in July 2019 regarding current practices for establishing, monitoring and modifying pediatric VAD settings. A total of 25 (68%) centers responded to the survey. Criteria for establishing intraoperative device settings varied between solely surgeon determined (36%) versus global imaging, surgeon, and VAD team driven (60%). Post-operative inpatient and outpatient interrogations were performed predominantly (80%) by a VAD physician or advanced practice provider (APP), but frequency ranged from daily (64%) to weekly (24%) to even monthly (4%). There was consensus (100%) on including device trends and alarm assessment for VAD interrogation, but many centers (88%) include log file interpretation in each interrogation as well. Factors used for optimization of post-operative settings were mixed. Several centers (28%) rely only on device interrogation and clinical course while many (24%) utilize echocardiogram only. The majority (44%) of sites use a combination of echocardiogram and device interrogation to determine optimal support. Outpatient practices were similar amongst sites with the exception of device wean trials which had significant variation with regards to use of echocardiogram (16%), hemodynamic catheterization (8%) or a combination of both studies (76%). Billing practices for VAD interrogation were inconsistent amongst centers, particularly regarding billing frequency. VAD interrogation and optimization practices were highly variable amongst ACTION centers, most notably during inpatient hospitalizations. Given the majority of pediatric VAD patients remain hospitalized while on their device, a standardized approach to this process may minimize unnecessary variation. Furthermore, defining "standard" management approaches and device settings in varied pediatric populations may help define the optimal clinical approach to support.

Current Discharge Practices in Pediatric VAD Patients: A Survey of the ACTION Collaborative

The use of intracorporeal continuous flow ventricular assist devices (CF-VAD) in pediatric patients with advanced heart failure has allowed for patients to be discharged home. However, currently only about 50% of pediatric patients are discharged from the hospital. We sought to examine the discharge practices across the Advanced Cardiac Therapies Improving Outcomes Network (ACTION). Surveys were sent to the members of the ACTION collaborative in September 2018 to determine discharge practices in preparation for a quality improvement project. Surveys were sent to 28 sites, with respondents from 22 sites (response rate: 78.5%). The majority of centers had discharged 1-5 patients (59.1%, 13/22) in the preceding 24 months. The remaining centers' discharge experience in the preceding 24 months was the following: 9.1% (2/22) with >5-15 patients, 9.1% (2/22) with >15 patients, and 22.7% (5/22) had not discharged a patient. The majority of patients discharged were on the HVAD (81.8%), followed by HeartMate 3 (13.6%), HeartMate II (9.1%), and SynCardia TAH (9.1%). Barriers to discharge identified were family readiness (72.7%), patient too sick (59.9%) and distance from VAD center to home (36.4%). At the time of the survey, 45.5% of sites were following 1-5 outpatients with 50% having no outpatients. Half of the programs manage patients from remote locations with most centers (63.6%) not sharing responsibility of care with a local cardiologist. A small number of centers, 27% (6/22), had VAD patients in the outpatient setting who were not listed for transplant due to reasons such as awaiting further rehabilitation, assessment of myocardial recovery, and family preference. The majority of centers allowed patients to return to school post discharge (95.5%), to drive (63.6%), and travel nationally (77.3%) but not internationally (36.4%). Barriers perceived by health care providers and families continue to limit pediatric centers from discharging CF-VAD patients from the hospital. As discharge and maintenance of outpatient care for this unique population is a low frequency but resource intense process, collaboration among centers to gain knowledge of best practices is valuable.

Preparing for Discharge in Pediatric Ventricular Assist Device Supported Patients

While the use of durable ventricular assist device (VAD) support with potential for discharge in pediatric patients has increased over time, there is limited knowledge of individual centers' approach to preparation and education for patients and families around discharge. The Advanced Cardiac Therapies Improving Outcomes Network (ACTION) was surveyed in September 2018 to determine current discharge practices in preparation for a quality improvement project. Surveys were sent to 28 sites, with 22 sites responding (response rate 78.5%). The majority of programs (86.4%, 18/22) reported having a standardized discharge protocol. Discharge teaching was performed in a majority of centers by a VAD coordinator (95.5%). Frequently utilized practices in preparation for discharge included field trips with and without medical chaperones, assessment of home environment (without an actual visit), rooming-in process, establishment of an EMS plan, and education about using a shower bag. The three main topics that patient and families received education on was: dressing change teaching, VAD equipment teaching and emergency planning. In terms of community outreach, programs offered education to EMS (72.7%), school staff (63.6%) and local fire stations (50%). In addition, 50% notified the local utility company in case of an electricity outage. The majority of programs caring for pediatric patients on durable VADs have a standardized approach to discharge with extensive education of patients, families, and community providers. As most programs discharge a small number of patients, a collaborative effort can allow for the development of shared educational materials that best address the needs of patients and their families.

Barriers to Establishing and Maintaining Therapeutic Anticoagulation in Pediatric Ventricular Assist Device Support: Advanced Cardiac Therapies Improving Outcomes Network Initiative

Establishing and maintaining target anticoagulation (AC) levels is perceived as a way to mitigate the risk of bleeding and thrombotic complications in pediatric ventricular assist device (VAD) support. To better understand potential reasons of having monitoring labs outside institutionally-set goal ranges, we surveyed centers to understand perceived barriers. As part of the Advanced Cardiac Therapies Improving Outcomes Network (ACTION), a 37-center collaborative, we surveyed participating VAD centers across North America. The survey was comprised of multiple choice questions addressing reasons for out of range coagulation labs. Specifically, asking centers to rate the frequency of occurrences; low, happens rarely, medium, happens often, high, happens almost daily and not applicable, meaning this barrier does not exists. We then weighted the average responses. Of the 49 responses received from the pediatric VAD-ACTION centers that responded to a 4-question multiple choice survey, the most commonly reported barrier was contamination of lab samples by sampling from a line infusing an anticoagulant through it in over 65% of respondents. This was followed by multiple providers involved in managing AC and micro-managing lab results as the second and third chosen as barriers reported in just under 50%. Interestingly, lack of documentation and obscure guidelines were not perceived as frequent causes of out of range values. While achieving and maintaining stable anticoagulation is the goal of therapy, understanding perceived obstacles is vital to targeted intervention. Targeted interventions to decrease the most common barriers, such as contaminated labs, will guide future ACTION quality improvement initiatives.

Early experience with the HeartMate 3 continuous-flow ventricular assist device in pediatric patients and patients with congenital heart disease: A multicenter registry analysis

The HeartMate 3 ventricular assist device (VAD) is a newer centrifugal continuous-flow VAD used for bridge-to-transplant and destination therapy in adults. However, there is limited experience regarding its use in children and adults with complex congenital heart disease (CHD). The Advanced Cardiac Therapies Improving Outcomes Network (ACTION) is a multicenter learning network comprised of pediatric hospitals implanting VADs in children and adults with complex CHD. We examined the outcomes of patients undergoing HeartMate 3 implantation at an ACTION center between December 2017 and September 2019. The HeartMate 3 was implanted in 35 patients at 9 ACTION centers, with a median age of 15.7 (8.8-47.3) years, median weight of 65.7 (19.1-114.1) kg, and median body surface area (BSA) of 1.74 (0.78-2.36) m2. Of the cohort, 14 patients (40%) weighed <60 kg. Diagnoses included dilated cardiomyopathy (63%), dilated cardiomyopathy in neuromuscular disease (20%), and CHD (17%). Of those with CHD, most had a Fontan circulation. With a median 78 days of follow-up, there was 1 death on device (97% survival); 20 out of 35 (57%) underwent transplantation with no post-transplantation mortality. There were no episodes of stroke or pump thrombosis. Use of the HeartMate 3 in ACTION centers was associated with a low incidence of mortality and adverse events. Patients as small as 19 kg (BSA 0.78 m2) were successfully implanted and supported, indicating that this device may be appropriate for older children and small adults.

Collaboration and new data in ACTION: a learning health care system to improve pediatric heart failure and ventricular assist device outcomes

Outcomes in pediatric heart failure and ventricular assist devices (VADs) remain suboptimal. Given the complexity and limited numbers, centers cannot be expected to acquire mastery and generalizable knowledge independently. Recognizing this problem, the community has formed Advanced Cardiac Therapies Improving Outcomes Network (ACTION), a collaborative learning health care system committed to unrelenting collaboration and continuous learning to improve critical outcomes in pediatric heart failure. ACTION is inclusive of all interested parties and believes the fastest and most effective way to make progress is through working together. ACTION's approaches to collaboration and data sharing will be highlighted.

It Takes a Village: Pediatric Ventricular Assist Device Training and Education Practices among ACTION Centers

Purpose The last decade has seen a dramatic increase in the number of ventricular assist devices (VADs) implanted in children. Despite the increases, the procedure remains uncommon at individual centers making patient, family and staff education challenging. ACTION ( Advanced Cardiac Therapies Improving Outcomes Network) is the first pediatric VAD quality improvement network. The initial goal of ACTION is stroke reduction. We sought to describe the education and care delivery structure among members of the ACTION pediatric VAD quality improvement collaborative. Methods A 26 question electronic survey assessing education infrastructure, resource use, methods and gaps in knowledge was distributed among the primary VAD education providers (VEPs) at ACTION sites. Results Seventeen centers replied, the majority of respondents were VAD coordinators (71%). The majority of VEPs train >100 providers/staff yearly (53%) and devote >6 hours/month on education (75%). There was significant variability in the effectiveness of VAD education programs, competence of care providers, patient/family discharge preparation and ongoing competence assessment among centers. The majority of respondents felt that the existing, educational materials and resources were insufficient for pediatric VAD patients/families and care providers, leading centers to center specific VAD education resources (figure). Conclusion Pediatric VEPs currently spend a significant amount of time educating patients and patient care providers on VAD management and care. The existing VAD educational materials and resources are insufficient and have led programs to create highly variable education/discharge protocols and center specific educational materials. The rapid growth of the field of pediatric VAD has created the urgent need for pediatric specific, high quality VAD educational materials and protocols.

Speaking the Same Language? Assessing Blood Pressure Measurement and Effectiveness in Pediatric VAD Patients

Purpose Elevated blood pressure (BP) is a risk factor for stroke in adult ventricular assist device (VAD) patients. There are currently no data or BP management guidelines for pediatric VAD patients. ACTION ( Advanced Cardiac Therapies Improving Outcomes Network) is the first pediatric VAD quality improvement network. The initial goal of ACTION is stroke reduction. This report describes the methods used to obtain BP and effectiveness of BP control among ACTION members. We also describe the development of a BP clinical decision support (CDS) tool to standardize BP reporting and goals. Methods The BP protocols among ACTION members were assessed. Process measures that may impact stroke rates were identified via anonymous surveys beginning June 2018. Statistical process control charts were used to assess baseline metrics. Results A total of 28 patients from 11 centers were identified with BP data. Including 12 Heartware, 10 Berlin, 2 Heartmate3, and 4 paracorporeal continuous flow (CF) devices. Height/age BP %iles were used to guide BP goal in 19 (68%) patients. Among CFVAD patients, BP measurement method (outside the perioperative period) and BP goal value varied, especially among the 11 patients (55%) with a palpable pulse (Figure 1). Patients had well controlled BP on 69% of support days (Figure 2). Conclusion The BP measurement methods and goals varied across ACTION, especially among patients on a CF-VAD with a pulse. A CDS tool has been developed to standardize BP measurement and management among ACTION members and will guide subsequent collaborative studies to assess the role of BP in stroke among pediatric VAD patients.

Anti-Coagulation Management in Pediatric Ventricular Assist Device: A Quality Improvement Target

Purpose Pediatric Ventricular Assist Device (VAD) therapy is associated with high rates of stroke and bleeding. There have been limited efforts to date, to define and implement optimal anticoagulation (AC) strategies in this population. ACTION (Advanced Cardiac Therapies Improving Outcomes Network) is the first pediatric VAD quality improvement network, with an initial goal of stroke reduction. This report describes baseline AC data from ACTION centers, including agents used and aggregate levels of performance, to inform development of specific interventions. Methods The AC practices of ACTION centers were assessed. Data collected included VAD type, weekly data characterizing AC agents used; time to achieve desired levels; and percentage of levels within the desired range. Results Data were collected from 11 centers, with a total of 30 patients. AC data were collected from 6/1-9/1/18, while strokes were collected retrospectively for the entire VAD support interval. Devices included: HVAD (12); EXCOR (10); Heartmate 3 (3); and 5 paracorporeal continuous flow devices. There were 5 strokes (16.7%). Intravenous AC included unfractionated heparin (36%), and bivalirudin (64%). There was wide variability between patients in the time to achieve first therapeutic AC level (Figure 1), with 15 patients at goal within 20 hours, while others required up to 170 hours. Heparin levels varied widely, with the percent of labs within range falling below 50%, in 50% of the weeks reported. For bivalirudin, levels were less variable, with percent of labs within range falling below 50% in only 14% of weeks reported. Conclusion At baseline, stroke rates remain high in the ACTION centers. Intravenous AC includes both heparin and bivalirudin, with high variability in time until therapeutic levels are achieved, and with high variability in subsequent levels, particularly for heparin. These data will inform interventions designed to reduce variability in AC performance as a basis for stroke reduction.

Real World Data from the ACTION Quality Improvement Network - Preliminary Experience with a Magnetically Levitated Ventricular Assist Device in US Pediatric Centers

Purpose We sought to describe the off-label use to date of an intracorporeal, centrifugal, continuous flow ventricular assist device (cfVAD) characterized by a magnetically levitated impeller in US pediatric cardiac centers, in order to further inform device selection quality improvement initiatives. Methods Members of the ACTION (Advanced Cardiac Therapies Improving Outcomes Network) collaborative whose centers implanted ≥1 intracorporeal magnetically levitated cfVAD were identified. Centers submitted de-identified data on each implant through 10/21/18. Results are reported as count (percentage) or median (range) for summary statistics. Results Fourteen implants occurred at 6 centers; the number of implants per center was 1 - 4. Thirteen patients (93%) had dilated cardiomyopathy (DCM); one patient (7%) had congenital heart disease (CHD) with a Fontan circulation. Of the DCM patients, 3 had neuromuscular disease (2 with Duchenne and 1 with Becker muscular dystrophy). Device strategy included bridge to transplant in 8/14 (57%), bridge to candidacy in 4/14 (29%), and destination therapy in 2/14 (14%). Median age at implant was 15 y (12 y - 24 y), with median weight 70 kg (31 kg - 118 kg) and median body surface area (BSA) 1.8 m2 (1.3 m2 - 2.4 m2). Most patients (10/14, 71%) were INTERMACS category 2 at implant; 3/14 (21%) were category 3 and 1/14 (7%) was category 4. No biVADs were placed. Median support time was 32 d (5 d - 315 d). To date, 5/14 (36%) patients were discharged, 1/14 (7%) underwent heart transplantation, and 1/14 (7%) died on VAD support; the remainder are hospitalized with the device in place. For discharged patients, median hospital length of stay after implant was 26 d (12 d - 37 d). There were no episodes of stroke or pump thrombosis. Conclusion The intracorporeal, centrifugal, magnetically levitated cfVAD is a newer device being used with increasing frequency in the pediatric and CHD population. Utilization of the device has largely been limited to older, larger children with DCM. This preliminary data will contribute to our understanding of device selection in order to improve outcomes.