Search

Background Continuous flow left ventricular assist devices (CF -LVAD) improve survival in patients with advanced heart failure, but confer risk of bleeding complications. Serotonergic antidepressants (SA) are commonly used in heart failure patients receiving LVADs, but their inhibitory effect on platelet function may contribute to bleeding risk. Methods We performed a retrospective analysis of LVAD patients at our institution from 2016 –2019 comparing patients treated with SA after LVAD to those without SA. Demographic and clinical variables related to bleeding were collected on discharge from index hospitalization for CF-LVAD implantation and on admission for any bleeding event. The primary endpoint was incidence of bleeding requiring hospitalization after discharge . Secondary endpoints included overall number of admissions for bleeding, time to first hospitalization for a bleeding event, and incidence rate of hospitalizations for bleeding per patient year. Results 100 patients met inclusion criteria for the study. A total of 5 patients without a history of SA use and 31 patients who were prescribed SA after CF –LVAD implant were readmitted for a bleeding event after initial implant hospitalization (15% vs 46%, p = 0.004). Bleeding rate per person year (0.3 vs 0.61, p = 0.01) were significantly less in patients without SA use. Age-adjusted multivariable analysis found SA use to be associated with a hospitalization for bleeding (HR 2.3, 95% CI 0.99 –5.4). The higher incidence of hospitalization for bleeding was driven by non-gastrointestinal anatomical sites (6% vs 28%, p = 0.02) with a HR 7.7 (95% CI 0.96 –62). Conclusions SA treatment after CF-LVAD implantation was associated with an increased risk for bleeding complications requiring hospitalization, particularly non-gastrointestinal bleeding.

Purpose Right heart failure (RHF) after left ventricular assist device implant is a common. The EUROMACS right-sided heart failure risk score is a novel and more simplified model for predicting early RHF after LVAD implantation. We sought to evaluate its accuracy after Off-pump left ventricular assist device implantation (OPLVADI). Methods This is a retrospective analysis of 82 patients (93% males) who underwent OPLVADI from 2011 to 2019. Forty eight patient were implanted with a HM2 (58%), 31 (38%) with HM3 and 3 (4%) with an HVAD. Using the EUROMACS 5-item scoring system (RA/PWCP >0.54 = 2 pts., Hgb 3 = 2.5 pts., severe RV dysfunction = 2pts) low RHF was defined as score 0-2, moderate >2-4 and severe > 4points. We compared the predicted with actual (post implant) RVF rate in our cohort. Results Based on EUROMACS RHF score pre implant RHF was predicted as low in 57% of patients, moderate in 27% and severein 10%. Actual RHF was mild in 76% of patients, moderate in 22% and severe in 2% of patients. Conclusion In patients who underwent off pump LVAD implantation, the EUROMACS RHF risk score system appears to underestimate the prevalence of mild RHF and overestimates moderate and severe RHF. Larger studies are needed to better characterize factors associated with RHF in this patient population.

Background— In hypertrophic cardiomyopathy (HCM), heart transplant has been predominantly confined to patients with systolic dysfunction. An underappreciated HCM subset comprises patients with preserved left ventricular (LV) systolic function who may also require consideration for transplantation. Therefore, we sought to define the clinical profile and occurrence of advanced heart failure among patients with nonobstructive HCM and preserved systolic function. Methods and Results— Databases from 2 referral centers comprising 2100 HCM patients were interrogated. Forty-six nonobstructive HCM patients (2.2%) either received or were listed for heart transplant, including 20 with normal systolic function (ejection fraction ≥50%). At transplant listing, these 20 patients were 42±13 years old, each in New York Heart Association functional class III/IV with ejection fraction of 62±7%. LV was hypertrophied with maximum wall thickness of 22±4 mm and nondilated (end-diastolic dimension, 39±7 mm). Cardiovascular magnetic resonance in 10 (of 15) patients showed no or minimal fibrosis (≤5% LV mass). Elevated LV end-diastolic or pulmonary capillary wedge pressure, consistent with diastolic dysfunction, was present in 15 patients (75%). LV filling was impaired by echocardiographic measures in all patients, including a restrictive inflow pattern in 8 (40%). In 2 patients, traditional criteria for transplant were absent, including peak V O 2 >14 mL/kg/min. Heart transplantation was performed in 12 patients with each alive and without cardiovascular symptoms, 2.3±1.7 years later. Conclusions— A previously under-recognized segment of the broad HCM clinical spectrum consists of nonobstructive patients with advanced heart failure, in the presence of preserved systolic function, for whom heart transplant is the sole definitive therapeutic option.

Purpose Serotonergic antidepressants (SA) are commonly used for a variety of indications in patients supported with durable left ventricular assist devices (LVAD). Reduced serotonin reuptake with SAs leads to platelet dysfunction and is associated with an increased risk of bleeding across diverse populations. There is limited data on the bleeding risk with SAs in the LVAD population prone to hemorrhagic complications due to other mechanisms. Methods We performed a retrospective analysis of LVAD patients (pts) managed at our institution from January 2016 to August 2018. Pertinent demographics and clinical variables related to bleeding were collected at the time of discharge from LVAD implantation and on admission for a bleeding event. Pts were divided into those prescribed an SA at discharge from LVAD implantation or admitted for a bleeding event with documented SA use prior to admission (Group 1) and those without SA exposure after implant (Group 2). Primary and secondary endpoints included incidence of bleeding requiring hospitalization, time from implant to first hospitalization for a bleeding event, and incidence rate of hospitalizations for bleeding per patient years. Results A preliminary analysis of 20 randomly selected pts out of 95 total implants performed. Ten pts were prescribed an SA (Group 1) and 10 were not (Group 2). Baseline demographics and bleeding risk factors were similar between the (two) 2 groups. Trazodone and sertraline were the most commonly prescribed SAs (35% each) and multiple SA use was common (50%). Acid suppressive therapy was used in all pts at discharge. In Group 1, 50% of the pts were hospitalized for a bleeding event compared to 10% of pts in Group 2 (p=0.1). At 1 year, 60% of pts remained free from hospitalization for a bleeding event in Group 1 compared to 100% in Group 2 (p=0.16). The number of bleeds per patient-year in Group 1 vs. those in Group 2 was 0.79 vs 0.4, respectively (p=0.26). Gastrointestinal (6 vs. 3 events) and intracranial (2 vs. 0 events) bleeding was more common in Group 1. Conclusion A preliminary analysis of patients with an LVAD treated with SA therapy demonstrated a trend towards an increased rate of hospitalization due to bleeding and a shorter time to first hospitalization compared to those who were not treated with an SA. Our final analysis will expand to all 95 patients to determine the significance of these findings.

Institutional Affiliations Co-chairs Feldman D: Minneapolis Heart Institute, Minneapolis, Minnesota, Georgia Institute of Technology and Morehouse School of Medicine; Pamboukian SV: University of Alabama at Birmingham, Birmingham, Alabama; Teuteberg JJ: University of Pittsburgh, Pittsburgh, Pennsylvania Task force chairs Birks E: University of Louisville, Louisville, Kentucky; Lietz K: Loyola University, Chicago, Maywood, Illinois; Moore SA: Massachusetts General Hospital, Boston, Massachusetts; Morgan JA: Henry Ford Hospital, Detroit, Michigan Contributing writers Arabia F: Mayo Clinic Arizona, Phoenix, Arizona; Bauman ME: University of Alberta, Alberta, Canada; Buchholz HW: University of Alberta, Stollery Children’s Hospital and Mazankowski Alberta Heart Institute, Edmonton, Alberta, Canada; Deng M: University of California at Los Angeles, Los Angeles, California; Dickstein ML: Columbia University, New York, New York; El-Banayosy A: Penn State Milton S. Hershey Medical Center, Hershey, Pennsylvania; Elliot T: Inova Fairfax, Falls Church, Virginia; Goldstein DJ: Montefiore Medical Center, New York, New York; Grady KL: Northwestern University, Chicago, Illinois; Jones K: Alfred Hospital, Melbourne, Australia; Hryniewicz K: Minneapolis Heart Institute, Minneapolis, Minnesota; John R: University of Minnesota, Minneapolis, Minnesota; Kaan A: St. Paul’s Hospital, Vancouver, British Columbia, Canada; Kusne S: Mayo Clinic Arizona, Phoenix, Arizona; Loebe M: Methodist Hospital, Houston, Texas; Massicotte P: University of Alberta, Stollery Children’s Hospital, Edmonton, Alberta, Canada; Moazami N: Minneapolis Heart Institute, Minneapolis, Minnesota; Mohacsi P: University Hospital, Bern, Switzerland; Mooney M: Sentara Norfolk, Virginia Beach, Virginia; Nelson T: Mayo Clinic Arizona, Phoenix, Arizona; Pagani F: University of Michigan, Ann Arbor, Michigan; Perry W: Integris Baptist Health Care, Oklahoma City, Oklahoma; Potapov EV: Deutsches Herzzentrum Berlin, Berlin, Germany; Rame JE: University of Pennsylvania, Philadelphia, Pennsylvania; Russell SD: Johns Hopkins, Baltimore, Maryland; Sorensen EN: University of Maryland, Baltimore, Maryland; Sun B: Minneapolis Heart Institute, Minneapolis, Minnesota; Strueber M: Hannover Medical School, Hanover, Germany Independent reviewers Mangi AA: Yale University School of Medicine, New Haven, Connecticut; Petty MG: University of Minnesota Medical Center, Fairview, Minneapolis, Minnesota; Rogers J: Duke University Medical Center, Durham, North Carolina

Purpose: Development of aortic insufficiency (AI) in patients (pts) during continuous flow LVAD support is a well known phenomenon that can lead to ineffective LVAD output. Current literature reports an average AI progression by 0.7 grade at 6 months (mo). We hypothesized that optimal medical therapy (MT) together with lowering the LVAD speed to allow aortic valve (AV) opening delays development of significant AI. Methods and Materials: Charts and echocardiograms (echos) of 40 LVAD pts were reviewed for demographics and medications. Medications were uptitrated to MAP of 60-80mmHg. VAD speed (RPM) was reduced according to our outpatient VAD protocol and collected at 1 and 6 mo. Echos were reviewed at baseline (pre LVAD), 1 and 6 mo for left ventricular end diastolic dimensions (LVEDD)(cm), AV opening, presence/ absence of AI. AI was graded: O-none, 0.5-trace,1-mild,1.5-mild-moderate, 2-moderate, 2.5-moderate-severe, 3-severe. Results: Forty pts, mean age 58, were included in the study. At 6 mo 38/40 (93%) pts were receiving beta blocker (BB), 30/40 (73%) ACE-I/ARB, 27/40 (66%) aldosterone receptor blocker (ALDOBL). Fifty six percent of pts achieved half of maximal daily dose of ACE-I/ARB, 45% BB and 53% ALDOBL. Twenty nine out of 40 (71%) pts were on furosemide, mean dose 32 mg. At baseline mean LVEDD was 6.8, AI present in 12/40 pts (30%), mean grade 0.3. At 1 mo LVEDD decreased to a mean 5, AI present in 6/22 (27%) pts, mean grade 0.27. At 6 mo LVEDD was the same, mean 5.1, AI present in 8/19 (42%) pts. Mean grade increased from 0.27 to 0.4 (p 0.3). Mean LVAD speed at 1 mo was 9190 and decreased to 8840 at 6 mo (p 0.001). At 1 mo AV opened with every beat in 33% pts, partially opened in 8%, did not open in 54%. At 6 mo AV opened with every beat in 36% pts, partially opened in 21%, did not open in 43%. Conclusions: Our results suggest slower progression in the AI grade at 6 mo in LVAD pts, than reported to date. An early strategy of decreasing VAD speed combined with optimization of MT may account for this difference and possibly improve long term survival.