Right Heart Failure in Left Ventricular Assist Device Recipients Michael Kiernan, MD, MS Assistant Professor of Medicine, Tufts University School of Medicine Medical Director, Ventricular Assist Device Program, Tufts Medical Center

The problem with right heart failure following LVAD implantation 12 mo 78% Total RVF 20% 6 % RVAD 7% early extended inotropes 7% late inotropes Survival (N=484) 59% Historically RVAD 5-25%, RVF 20-40% 423 pts in BTT trial RVF strong independent determinants of outcomes Kormos. J Thorac Cardiovasc Surg 2010;39:1316

Identifying risk for RVF Fukamachi1 Michigan RVFRS2 Kormos3 Device Type Pulsatile Pulsatile (84%) & cf-LVAD (16%) cf-LVAD # of LVAD patients 100 197 484 RVF Definition RVAD RVAD, inotropes, iNO RVAD, inotropes Incidence of RVF 11% 35% 20% Incidence RVAD 15% 6% Institution Single center Multicenter Predictors Univariate Multivariable RVSWI < 300 Vasopressor support Ventilator support Mean PAP < 40 Bilirubin > 2.0 mg/dL BUN > 39 mg/dL AST AST > 80 IU/L RA/PCWP > 0.63 Creatinine > 2.3 mg/dL Fukamachi: Cleveland Clinic: 64 Thoratec HM 1000 IP, 36, VE HM Matthews: HMI IP 1000 15 pts, VE 65, XVE 77, HMII 28 (14%), micromed Lack of consistently identified predictors of rvf across studies made identification of pts who would benefit from biventricular support difficult Planned use of bivads improve clinical outcomes compared to delayed Risk factors, multifactorial nature of RVF difficult to predict Other risk factors: female gender, NICM 1. Fukamachi. Ann Thorac Surg 1999;68:2181. 3. Kormos. J Thorac Cardiovasc Surg 2010;139:1316 2. Mathews. J Am Coll Cardiol 2008;51:2163

Complex etiology of post-LVAD RV failure RVD High flow Chronic Intra-op ischemia LVAD & IVS RVF RAP PVR CPB PRBCs Hypoxia Acidosis PRBCs TR Hepatic/renal congestion Many patients have some degree of preoperative right ventricular dysfunction then develop the syndrome of rv failure Septum contributes > 1/3 systolic function of RV Transfusion associated with pulmonary insufficiency. TRALI, complement activation with CPB RH dysfunction worsening congestive hepatopathy, coagulopathy, bleeding Historically overall RVF 10-40%, rvad 5-25% Rv less susceptible to ischemic damage. Less myocardial mass and more favorable oxygen supply demand ratio Excessive lvad flows may result in excessive rv preload RV dilates and rv output is initially maintained at expense of elevated ra and RV end diastolic pressures Maintain LVAD flows high enough to maintain end organ perfusion but low enough to avoid RV volume overload, adjust flows to minimize distortion of RV Intraoperative RV protection RV lower stroke work than LV, oxygen extraction 50 versus 75% Coupled to highly compliant pulmonary vascular tree Diagnosis is complex Ventricular interdependence implies shape and size and compliance of one ventricle affect size shape and pressure volume relationship of other through direct mechanical interactions. Systolic inter-dependence is mediated primarily though IVS, while diastolic inter-dependence occurs mainly via pericardium Successful filling of LVAD requires rv to increase output to match lvad flow Global contractility of rv impaired during lvad support changes in ivs position and motion when lv unladed and ivs buldge into lv jeopardizing efficient rv contraction, rv free wall must work harder and cannot compensate for loss of septal function Stroke Volume RV LV McDonald. Curr Opin Card. 2009;24 Meineri. Best Pract & Res Clin Anesth 2012 (26):217 Pressure vessel (mmHg)

Severe RV Failure in INTERMACS Continuous Flow LVAD N =2900 RVAD at time of LVAD Implant N = 84 (3%) Return to OR for RVAD N = 26 (1%) 4% of 2900 patients required RV support. 24% returned to OR, 76% in OR 9% durable RVAD: Durable N = 5 (6%) RVAD: Temporary N = 79 (94%) RVAD: Durable N = 5 (20%) RVAD: Temporary N = 21 (80%) Kiernan. ISHLT 2012

56% Cleveland. J Heart Lung Transplant 2011;30(8):862 Bivads have higher rates of infection, bleeding, 4-fold increase in neurologic complications Cleveland. J Heart Lung Transplant 2011;30(8):862

Early management: Outcomes with planned versus delayed BiVAD Survival until Discharge P = 0.046 P = 0.001 P = 0.054 Median time to delayed RVAD 2 days Mortality results primarily from MOSF, hemorrhage, thrombotic events, pulmonary complications Need for RVAD most significant risk factor for LVAD recipients ASAIO 2014: German experience. No difference in survival for thoratec bivad, HMII and r centrimag, LVAD with inotropes Fitzpatrick (UPenn). J Thorac Cardiovasc Surg 2009;137(4):971

Risk Factors for Early RVAD Following LVAD Surgery Adult Primary Continuous Flow Implants (N=2900) Multivariable Logistic Regression (Event=RVAD) Risk Factors (pre-implant) Odds Ratio p – value INTERMACS Patient Profile Level 1 3.11 < 0.0001 PaPi (per 1 unit larger) 0.71 0.0003 LVEDD (per 1 mm increase) 0.78 0.01 RV dysfunction by echo (any) 3.17 0.01 Primary Diagnosis CAD 1.78 0.03 Hemoglobin (per 1 gm/dl increase) 0.89 0.03 Concomitant surgery 1.55 0.03 INTERMACS Patient Profile Level 2 1.81 0.03 RVD by echo sensitive by not specific. If looks normal than probably okay, but dysfunction not helpful in determine need for bivad Adjustment for multiple corrections with bon-fiori p < 0.001 Kiernan. ISHLT 2012

PA Systolic Pressure – PA Diastolic Pressure Pulmonary Artery Pulsatility Index PA Systolic Pressure – PA Diastolic Pressure Right Atrial Pressure PaPi = Korabathina. Catheter Cardiovasc Interv. 2011 Sep 27.

Prevention?: Pre-op optimization Class I (LOE C) Therapy (Preload, Afterload, Inotropy): Diuresis Renal Replacement Vasodilators Inotropes IABP/short term MCS Vitamin K Surgical technique ? RCA/LAD revascularization Goals: RA < 15 mmHg Euvolemia Correction metabolic derangements and end-organ function Prevention?: Pre-op optimization Class I (LOE C) Two things to elicit RHF: decreased contractility or increased afterload FS mechanism less important in RV hemodynamics Larger surface to volume ratio: takes large increase in volume to stretch RV RV more restrained by pericardium: increase fluid flattened septum Recommended to admit preoperatively for tuning of RV Patrick McCarthy’s at Cleveland clinicgroup demonstrated Vitamin K prevad 10mg sq daily for three days decreased post-operative nonsurgical (> 48 hrs) bleeding Adequate BP necessary for coronary perfusion of RV Piazza. Chest 2005;128:1836-52 ISHLT 2013 MCS Guidelines. JHLT 2013;32:157-187 10

Implants June 2006 – March 2011: RHF Analysis Adult Primary Continuous Flow Implants N=2900 By Right Heart Failure Level Severe Right Heart Failure, n= 110 Moderate or Worse Right Heart Failure, n= 398 % Freedom RHF Mild or Worse Right Heart Failure, n=1284 Overall p < 0.0001 Event: Right Heart Failure Months post implant

Readmission rate by cause following LVAD implantation Hospitalization/(patient*yr) Readmission rate Hasin (Mayo). JACC 2013;61(2):153

Impact of Tricuspid Valve Repair at time of LVAD LVAD Only N=81 LVAD + TVR N=34 P-value Baseline Characteristics SCr 1.78 + 0.8 1.98 + 0.9 0.32 BUN 36 + 21 48 + 30 0.06 CVP/PCWP .57 + 0.2 0.75 + 0.3* <0.01 CVP 16 + 8 19 + 7 0.09 Severe TR 33% 62% OUTCOMES RVAD 10% 3% 0.27 Inotrope 10d (8,17) 8d (7,12) 0.04 Post-op renal insuff 39% 21% 0.05 Hosp LOS 23d (16,46) 19d (14,25) 0.02 LVAD + TVR Survival LVAD Not consistent in some other cohorts. Piacentino (Duke). Ann Thorac Surg 2011;92:1414

2196 patients with mod-severe TR--27% TVR TVR associated with: No difference in death or RVAD Increased renal failure Greater transfusion requirement Increased LOS Concomitant TVP did not affect risk post-operative right VAD insertion (risk ratio[RR],0.81;95%confidenceinterval [CI],0.49–1.36; p ¼ 0.4310)or death(RR,0.95;95%CI,0.68–1.33; p ¼ 0.7658). However, TVP was associated with increased risk post-operative renal failure(RR,1.53;95%CI,1.13–2.08; p ¼ 0.0061), dialysis(RR,1.49;95%CI,1.03–2.15; p ¼ 0.0339), reoperation(RR,1.24;95%CI,1.07–1.45; p ¼ 0.0056), Greater total transfusion requirement (RR,1.03;95%CI,1.01–1.05; p ¼ 0.0013), hospital length of stay 421 days(RR,1.29;95%CI,1.16–1.43; p o 0.0001). Time on the ventilator and intensive care unit length of stay were also significantly prolonged for the LVAD þTVP group JHLT 2014

Effect of digoxin on RV function in primary pulmonary hypertension with symptomatic heart failure 17 pts at Rush No change in HR, no change in PCWP or RA or systemic BP 10% increase CO Positive inotrope and sympotholytic properties (Decreased NE levels) 1mg IV digoxin given with baseline and 2h hemodynamics recorded Reduction in NE levels Recent presentation at ACC regarding effect of digoxin on improving long-term outcomes among patients with PAH Class IIb (LOE C) Rich (Rush). Chest 1998;114:787-792. 15

Effect of PDE-5A inhibition on PVR and RV hemodynamics post LVAD Control Sildenafil PVR (WU) dP/dtmax/IP 26 pts with persistently elevated pvr following lvad placement PDE5 improve PVR in pts with LVADs may be started and help to wean iNO Does not have large effect on mean arterial pressure Class IIb (LOE C) Tedford (Hopkins). Circ Heart Fail 2008;1:213

Effect of RV pacing on RV function in model of pulmonary hypertension induced RVD Control PHTN RV dP/dt max Effects of RV pacing in rat model of pulmonary hypertension induced RV dysfunction PH induced by injection of MCT resulted in PHTN phenotype with RH failure 28 days later RV pacing improves rv function and diminishes adverse diastolic interaction in rat model Onset diastolic relaxation impaired. Delay in peak myocardial shortening and delayed onset of rv diastolic relaxation contributes to diminished efficiency of rv contraction and reduced stroke volume, Rv to lv delay in onset of diastolic relaxation. Hardziyenk europace 2011; cteph with rvd. Improved stroke volume with rv pacing RVSP Class IIb (LOE C) Handoko. Am J Physiol Heart Circ Physiol 2009:297:H1752

Conclusions & Future Directions RVF post LVAD remains common Need multi-disciplinary pre-op evaluation Need data/trials investigate operative techniques Need trials investigate strategies for management of chronic RVF Trials of temporary RV MCS support ongoing pre- and post-LVAD Emerging biventricular mechanical support devices