International Differences in Acute Heart Failure (AHF) Patients: Insight from the PROTECT Trial

The 15th Annual Scientific Meeting device based on transverse distribution of electromagnetic energy through the chest. Changes in the clinical status of patients, LI and NT-proBNP levels were concurrently recorded at each outpatient heart failure clinic visit (20618 days). Results: 150 CHF patients (72610 years) at NYHA II/III/IV (60/60/30) were followed for 31614 months in an outpatient clinic. Patients were treated with diuretics, beta blockers and ACE/ARB/aldosterone. An LI decreaseO15% from normal baseline was used to initiate early preventive therapy since we have shown previously that clinical decompensation occurred at this level of LI decrease. 75 of 150 patients were treated by clinical evaluation (Group1) and 75 according to LI (Group 2). LVEF and NT-proBNP in groups 1 and 2 were 2367%, 582062434 pg/ml, and 2366% and 586862532 pg/ml, respectively (p5NS). 140 episodes of LI decreaseO15% occurred in group 1 with treatment administered according to clinical signs only. These LI decrease episodes included 124 hospitalizations and 35 deaths. Positive predictive value for hospitalization at level of LI decreaseO15% was 89%. In group 2, 149 episodes of LI decreaseO15% were recorded. Treatment was immediately intensified. In 124 cases LI increased as the result of treatment intensification and only 25 hospitalizations were required (p!0.01) and 14 patients died (p! 0.01). LI decrease at admission in group 2 (2865.3%) was similar to that in group 1 (2765.6%). Time elapsed between LI decrease O 15% and hospitalization in both groups was 1666 days. Conclusions: Non invasive lung impedance monitoring of CHF patients in an outpatient clinic allows predict decompensation in 89%. Early preventive LI-guided treatment is effective for prevention of hospitalizations (in 83%) and reduces deaths by 2.5times.

190 Design of the Investigations in Pregnancy Associated Cardiomyopathy (IPAC) Trial: A Multicenter Investigation of Immune Pathogenesis and Myocardial Recovery in Peripartum Cardiomyopathy Dennis M. McNamara1, Uri Elkayam2, Leslie T. Cooper3, Jessica Pisarcik1, John Gorcsan1, Erik B. Schelbert1, Charles McTiernan1, James D. Fett4, IPAC Investigators; 1Cardiovascular Institute, University of Pittsburgh Medical Center, Pittsburgh, PA; 2University of Southern California, Los Angeles, CA; 3Mayo Clinic Foundation, Rochester, MN; 4Peripartum Cardiomyopathy Consultant, Seattle, WA Introduction: Peripartum cardiomyopathy (PPCM) is a rare complication of pregnancy which occurs in 1 to 3000-4000 births in North America but remains a major cause of maternal morbidity and mortality. Women with PPCM classically present with left ventricular dysfunction in the last month of pregnancy or the first five months post partum. While 50% of women experience dramatic myocardial recovery, others develop chronic cardiomyopathy and progress toward poorer outcomes including death or cardiac transplantation. The etiology is unknown, however an autoimmune pathogenesis is suspected. We sought to investigate the immune pathogenesis of PPCM and the potential for myocardial recovery in a prospective multicenter study Method: Thirty centers in the North American Peripartum Cardiomyopathy Network (PCN), are participating in the NHLBI sponsored IPAC (Investigations in Pregnancy Associated Cardiomyopathy) study. One hundred women with PPCM, all with an LVEF # 0.45 and presenting less than 2 months post partum, will be enrolled. At entry, 2, and 6 months post partum, the systemic immune profile will be assessed by peripheral blood sampling including cellular immune activation (T cells subsets by FACS, flourscence-activated cell sorting), antibody mediated immunity (by in vivo assay of anti-cardiac antibodies), transcriptional expression profile (paxgene collection and targeted microarray), and serum markers of T cell activation (IL2 soluble receptor and neopterin). LVEF will be assessed by transthoracic echocardiogram at entry 2, 6 and 12 months postpartum. Cardiac Magnetic Resonance (cMR) assessment of cardiac inflammation will also be investigated. The investigation will define the role of autoimmunity in myocardial injury and recovery in PPCM. Conclusion: IPAC will investigate the immune pathogenesis of PPCM and develop clinical and biomarker predictors of subsequent myocardial recovery. By defining the role of autoimmunity in pathogenesis, and the potential to predict myocardial recovery, IPAC will set the stage for future interventional trials in PPCM.

191 The Relationship between Red Cell Distribution Width with Clinical Characteristics: Analysis Based on 16681 Chronic Systolic Heart Failure Patients with Different Causation Shengbo Yu, Qingyan Zhao, He Huang, Hongying Cui, Mu Qin, Congxin Huang; Renmin Hospital of Wuhan University, Wuhan, China Objective: To determinate the prognostic value of red cell distribution (RDW) and the relationships between RDW and clinical characteristics in patients with CHF caused by different etiologies. Material and Methods: 16681 in-hospital patients from 12 hospital of Hubei province, China, with diagnosis of chronic systolic HF and LVEF!50% were enrolled. All patients were followed up by telephone contact. Patients were divided into RDW!13.3%, 13.3%-14.1%, 14.1%-14.8% andO14.8% groups; Mean corpuscular volume (MCV) decline(MCV!82FL),MCV elevation(MCVO92FL)and MCV normal(92FLOMCVO82FL)groups in patients with RDWO16%; death and survival groups according to the result of follow-up. Compared with RDW!13.3% group, adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of all-cause mortality for RDW 13.3%-14.1%,14.1%-14.8% and O14.8% were 0.892 (95% CI 0.818-0.973;P50.01), 0.859 (95% CI 0.793-0.931;



HFSA

S59

P!0.001) and 1.034 (95% CI 0.961-1.111;P50.373) respectively. Compared with MCV normal group, the adjusted HR of MCV elevation and MCV decline group was 1.351 (95% CI 1.063-1.718;P!0.001) and 1.316 (95% CI 1.034-1.675;P! 0.001), respectively. Compared with patients with RHD, the adjusted HR for allcause mortality those with CHD, DCM and HHD with RDWO16% were 1.437 (95% CI 1.141-1.810; P50.002), 1.651 (95% CI 1.276-2.138; P!0.001) and 1.276 (95% CI 1.004-1.621; P50.007). The RDW is independently correlated with BMI (b5e0.019; P!0.001), diastolic blood pressure (b5e0.008; P!0.001), albumin (b5e0.019; P!0.001), blood urine nitrogen (b50.559; P!0.001), right ventricular end-diastolic diameter (b50.01; P!0.001), LVEF (b5e0.013, P!0.001) and heart rate (b50.005; P!0.001). There was difference in the correlation factors with RDW among different etiologies. Conclusions: The relationship between allcause mortality and RDW is J shape while not linear. The elevation or decline of MCV in CHF patients with RDW increase suffered higher all-cause death risk. RDW has closely relationships with the clinical characteristics and the relationships are associated with different etiologies.

192 Intervention Research for Heart Failure Patients and Caregivers in Hospice e Is It Even Feasible? Cheryl H. Zambroski1, Chris Garrison1, Harleah G. Buck2, Susan C. McMillan1; 1 Nursing, University of South Florida, Tampa, FL; 2Nursing, Pennsylvannia State University, University Park, PA Objective/Background: National clinical practice guidelines suggest that hospice is a valuable alternative for patients with heart failure (HF) requiring end of life care. Yet, there are few evidence-based nursing interventions for hospice nurses caring for HF patients. This poster will illustrate the challenges of recruitment to a NINR-funded randomized clinical trial (RCT) of a psychoeducational intervention to assist caregivers of patients with HF to better manage symptoms. Method: The study was conducted in one of the largest hospices in the US. For inclusion, patients needed to have a primary diagnosis of HF and a caregiver who provided at least four hours of care/day. Patients were excluded if they had PPS scores of #30 to increase likelihood that patients could report their own symptoms through all three data collection points. Because the study focused primarily on management of five common symptoms (dyspnea, edema, pain, depression and constipation) patients were excluded if they did not experience at least two of these symptoms. Furthermore, based on the literature and our pilot work, multiple strategies were included to support recruitment and retention of dyads. Results: During the first 16 months of the study, 648 patient/caregiver dyads were screened. Only 15% were eligible. The most common reason for ineligibility was that patients were residing in a nursing home/ECF or that they had died within the first 1-2 weeks of admission to hospice. For dyads that were eligible, only 32% consented. The most common reasons for refusal were that they were either not interested or overwhelmed. Overall, of all dyads screened, only 3.5% completed the study. Conclusion: Similar to previously reported challenges with HF patient recruitment in non-hospice settings, the number of eligible patients was far lower than anticipated from the potential pool. Refusal rates were higher in hospice than in many previously reported studies of heart failure patients. The importance of feasibility studies with this population cannot be understated. As we evaluate further research to improve nursing care of HF patients, methods to increase the desirability of participation in hospice research in concert with examination of protocols for studies must be balanced with consideration of cost/benefit of the RCT in hospice.

193 International Differences in Acute Heart Failure (AHF) Patients: Insight from the PROTECT Trial M. Fiuzat1, B. Massie2, K. Chiswell1, M. Givertz3, B. Davison4, G. Cotter4, J. Cleland5, H. Dittrich6, G. Mansoor6, P. Ponikowski7, A. Voors8, J.R. Teerlink2, R. Mentz1, M. Metra9, C. O’Connor1; 1Duke Univ, Durham; 2 SFVAMC; 3Brigham & Womens; 4Momentum; 5Univ of Hull, United Kingdom; 6 Merck; 7Med. Univ, Wroclaw, Poland; 8Univ Med. Ctr, Groningen, Netherlands; 9 Univ of Brescia, Brescia, Italy Background: With the globalization of clinical trials come unique challenges. We explored international differences in the PROTECT trial and whether these may have influenced outcomes. Methods: The PROTECT trial enrolled patients admitted with AHF with CrCl 20-80 mls/min, randomized within 24 hrs to rolofylline vs placebo. Pts were enrolled in 17 countries from 2005-2008. For the purpose of this analysis, we grouped countries into 5 global regions: North America (NA), Argentina (AR), Israel (IS), Western Europe (WE), and Eastern Europe (EE). Unadjusted Kaplan-Meier estimates for death or CV/renal hospitalization at 60 days and death at 180 days were generated by region. Results: 2033 pts were enrolled: 15% from NA, 3% AR, 16% IS, 19% WE, and 47% EE. At baseline, NA pts were younger and had lower SBP. EE had lower BUN, higher serum sodium, and better renal function. BIVP or AICD was highest in NA. Aldo-blockers and ACE-inhibitors were more often used at discharge in EE. In-hosp inotrope use was slightly higher in NA. Importantly, length of stay was shorter in NA, AR and IS vs. EE and WE (median of 5, 5, 5, 10, and 10 days, respectively). The 60-day outcomes varied significantly across the regions; however death at 180-days did not show a significant difference. Conclusion: There were significant differences across the 5 global

S60 Journal of Cardiac Failure Vol. 17 No. 8S August 2011 regions in baseline characteristics, baseline treatments, patient management, and outcomes. Length of stay varied significantly across regions, which may influence rehospitalization rates. However, there was no difference in mortality at 180 days.

Fig. 1. Kaplan-Meier estimates of probability of death or CV/renal rehospitalization at 60 days by region.

characteristics when compared to healthy controls. Methods: Muscle function tests were performed on 41 HF patients and 12 controls. Peak torque (corrected for body weight), muscle work and acceleration and deceleration characteristics of muscle performance were assessed. The muscle function tests were performed using a Biodex System 4 Pro with an Isokinetic protocol, 25 repetitions at a speed of 180 deg/ sec. Clinical data were obtained from institutional records. Statistical significance was set a p#0.05 using an unpaired t-test. Results: The patient cohort had an average age of 61623 years in HF and 51618 in controls. Total muscle work was reduced in HF patients compared to controls (p!0.05). Patients with HF showed prolonged acceleration in knee extension (1.360.2 nm/msec in HF vs. 0.560.1 nm/msec in controls; p!0.02) and flexion (3.360.4 nm/msec in HF vs. 1.560.3 nm/msec in controls; p!0.007). Deceleration time was also prolonged in HF patients (knee extension: 2.360.3 nm/msec in HF vs. 1.260.1 nm/msec in controls; p!0.01; flexion: 3.760.3 nm/msec in HF vs. 2.160.2 nm/msec in controls; p!0.01). Both acceleration and deceleration times were corrected for peak torque and body weight. Identical Results were found when acceleration and deceleration time was corrected for total muscle work (both p!0.05). Conclusion: In conclusion, patients with HF show impairments of muscle fiber recruitment as indicated by prolonged acceleration and deceleration. Further, HF patients show reduced total muscle work during defined exercise intervals. These findings suggest a complex neuromuscular dysfunction contributing to peripheral muscle changes in patients with advanced HF.

196 Is Radial Strain by 2D Speckle Tracking Ready for Prime Time in Multicenter Echocardiographic Study of Heart Failure Patients? Salima Qamruddin1, Reza Rafie1, Kyungmoo Ryu2, Allen J. Keel2, Tasneem Z. Naqvi1; 1Non Invasive Lab, Division of Cardiology, Department of Medicine, University of Southern California, Los Angeles, CA; 2Research Division, St. Jude Medical, Inc., Sylmar, CA

Fig. 2. Kaplan-Meier estimates of probability of death at 180 days by region.

194 The Prognostic Value of Right Ventricular End-Diastolic Diameter in Patients with Chronic Systolic Heart Failure Shengbo Yu, Qingyan Zhao, He Huang, Hongying Cui, Mu Qin, Congxin Huang; Renmin Hospital of Wuhan University, Wuhan, China Aims: To determinate whether right ventricular end-diastolic diameter (RVDD) was an independent and specific predictor of chronic systolic heart failure (CSHF) prognosis. Methods and Results: 16681 patients, diagnosed with CSHF, from 12 hospitals were analyzed. Patients were categorized into four RVDD groups: !25 mm (n58801), 25 to 30mm (n51149), 30 to 38 mm (n51959) and O38 mm (n54772). Univariate and multivariate Cox proportional hazards analyses were performed to examine the risk of all-cause death, death because of heart failure (HF) progress and sudden cardiac death (SCD) in different RVDD groups. The sensitivity and specificity of RVDD in predicting the prognosis were examined by 2 multivariate logistic models and receiver operating characteristic (ROC) curves. Over media 3 yeras follow-up, compared with patients with RVDD!25 mm, adjusted Hazard Ratio (HR) for those with RVDD 25 to 30 mm, 30 to 38 mm and O38 mm were 1.87 (1.64 to 2.13; P!0.001), 2.41 (2.06-2.80; P!0.001) and 3.95 (3.61 to 4.32; P!0.001), for all-cause mortality, respectively; 1.32 (1.12 to 1.55; P50.001), 1.48 (1.10 to 1.91; P!0.001) and 2.89 (2.53 to 3.28; P!0.001), for HF progress mortality, respectively; 3.17 (2.11 to 4.61; P5 0.016), 1.85 (1.09 to 3.16; P5 0.024) and 3.82 (2.27 to 5.94; P!0.001), for SCD, respectively. The ROC curves showed the RVDD increased sensitivity and specificity of predicting model for all-cause mortality and the best model including RVDD did discriminate between the HF and SCD mortality.

We evaluated the feasibility of 2D radial strain measurement in a multicenter study. Methods: Raw data was received by a core lab from 6 expert European sites in 27 patients with heart failure enrolled in a biventricular (biv) pacemaker study. Echocardiographic protocol training was provided via webinar, sites were certified after qualifying studies and feedback on study quality was provided to sites within 72 hours after data receipt at the core lab. Radial strain was measured in 6 mid short-axis (SAX) left ventricular (LV) segments per patient using 2D speckle tracking analysis software (GE EchoPac) and wall motion (WM) of these segments was scored (15normal, 25hyokinetic, 35akinetic, 45dyskinetic and 55aneursymal) Blinded measurement of radial strain was performed in triplicate. Results: Of the 162 segments collected in 27 patients, 89 mid SAX segments could be analyzed for radial strain. 73 segments were excluded due to inability for the speckle tracking software to track the segment. Intraobserver variability of radial strain measurement was 5% in 10 randomly selected patients. Mean global peak radial strain was 8.665%, and LVEF was 2768%. Peak radial strain in mid SAX segments correlated with WM (r50.7, p 50.0001). Peak radial strain was 19.869% in 20 segments with normal WM, 8.864.7% in 53 segments with WM score of 2, and 1.263.4% in 16 segments with WM of 3, 4,or 5, (p!0.0001, ANOVA). The average global peak strain correlated with LVEF (r50.64, p50.0001). Time to peak radial strain was earlier for segments with peak strain O 10% than for segments with peak strain !10% (546670 vs. 5816119ms, p50.014). Conclusion: Radial strain by speckle tracking is feasible in a multicenter setting, has low intraobserver variability and correlates well with WM and LVEF in Biv paced patients. However 45% of LV segments could not be measured by speckle tracking. Factors that may improve measurable radial strain by speckle tracking need to be addressed. Our findings have implications on the use of radial strain in a multicenter setting.

195 Impairment of Muscle Fiber Recruitment and Decreased Total Muscle Work Capacity during Isokinetic Exercise in Patients with Advanced Heart Failure Meaghan L. Jones1, Aalap Chokshi1, J. Elias Collado1, Aimee Layton1, Hilary F. Armstrong1, Daniel E. Forman2, Matthew Bartels1, Donna Mancini1, P. Christian Schulze1; 1Cardiology, Columbia University, New York, NY; 2Cardiology, VA Boston Health Care System, Boston, MA Introduction: Skeletal muscle dysfunction and exercise intolerance develop in patients with advanced heart failure (HF). We hypothesized that HF patients show changes in muscle strength and power as well as abnormal fiber recruitment

Fig. 1. Correlation of Global Peak Radial Strain with LV Ejection Fraction.