Sleep Disordered Breathing Detection in Congestive Heart Failure Subjects with Chronicle® Implantable Hemodynamic Monitors

The 9th Annual Scientific Meeting

•

HFSA

S191

378

380

Anemia in a Hospitalized Heart Failure Population: Characteristics and Relationship to Outcomes. A Report from OPTIMIZE-HF J. Young1, W. Stough2, N. M. Albert1, L. She2, G. Fonarow3; 1Dept of Cardiovascular Med, Cleveland Clinic Foundation; 2Duke Clinical Research Institue; 3University of California Los Angeles

Continuous Hemodynamic Monitoring in the Management of Advanced Heart Failure Patients Jose A. Tallaj1, Robert C. Bourge1, Mark F. Aaron2, William T. Abraham3, Juan M. Aranda4, Anthony Magalski5, Amy D. Roettger6, Sue Dale6, Jason Grimes1; 1 Cardiology, University of Alabama, Birmingham, AL; 2Cardiology, Saint Thomas Hospital/ The Heart Group, Nashville, TN; 3Cardiology, The Ohio State University Heart Center, Columbus, OH; 4Cardiovascular Medicine, University of Florida Shands, Gainsville, FL; 5Cardiology, Mid America Hospital, Kansas City, MO; 6Medtronic Inc, Minneapolis, MN

Introduction: Anemia is associated w/ increased symptoms and adverse outcomes in heart failure(HF), but little is known regarding the prevalence of anemia and its relationship to outcomes in patients(pts) hospitalized with HF. Hypothesis: Low hemoglobin(hgb) is correlated with HF severity and adverse outcomes. Methods: OPTIMIZE-HF is a registry and performance improvement program for hospitalized HF pts and includes a subgroup w/ 60–90 day follow-up (f/u) data. Admit hgb was analyzed by quartiles. Comparisons were analyzed by Pearson chi-square test and analysis of variance. Results: Data includes 48612 pts at 259 hospitals. Pts w/ low hgb tended to be older, female, Caucasian, have preserved systolic function, and elevated serum Cr. There was no difference in troponin. They were less likely to receive ACEI and BB, but more likely to receive ARB. Pts w/ lower hgb had higher in-hospital and f/u mortality, longer LOS and more readmissions. Conclusions: Data reveal a higher prevalence of low hgb than in randomized HF trials. Low hgb is associated w/ worse outcomes in hospitalized HF pts. Admit Hgb, g/dL Characteristic % (SD)

5.0–ⱕ10.7 ⬎10.7–ⱕ12.1 ⬎12.1–ⱕ13.5 ⬎13.5–ⱕ20.0 N ⫽ 12099 N ⫽ 12277 N ⫽ 11723 N ⫽ 11507

Mean Age, y Female Caucasian LVSD Mean EF Mean HR Admit Mean SBP Admit Ischemic Etiology Mean Hgb Mean BNP Mean Troponin I Mean serum Cr ACEI DC ARB DC BB DC All Cause Mortality Inhosp All Cause Mortality f/u Mean LOS d Readmissions

74.1 (13.4) 58.9 72.8 45.5 43.2 (16.8) 83.7 (19.5)

74.8 (13.4) 57.9 74.5 49.5 40.4 (17.6) 84.9 (20.6)

141.4 (32.1) 142.8 (32.5) 47.1

46.8

9.6 (.95) 11.5 (.40) 1372 (1362) 1326 (1369) .50 (1.44) .46 (1.43) 2.1 (1.8) 1.8 (1.6) 45.3 49.8 12.0 13.1 62.2 64.4 4.8 3.9 11.9 6.5 (6.0) 33.1

9.9 5.7 (5.4) 33.0

73.7 (13.8) 51.3 74.0 54.4 37.9 (17.7) 87.4 (21.7)

P

70.1 (14.8) ⬍ .0001 37.8 ⬍ .0001 75.0 .0101 59.4 ⬍ .0001 34.9 (17.3) ⬍ .0001 90.9 (23.3) ⬍ .0001

143.1 (33.0) 144.1 (34.3) ⬍ .0001 45.4 12.8 (.40) 1250 (1321) .48 (1.45) 1.6 (1.4) 55.9 12.1 64.6 3.2 9.0 5.3 (5.0) 28.6

43.0 14.8 (1.04) 1152 (1221) .50 (1.43) 1.5 (1.1) 60.5 11.3 66.5 3.0 8.6 5.3 (5.4) 24.2

⬍ .0001 ⬍ .0001 ⬍ .0001 NS ⬍ .0001 ⬍ .0001 .0003 ⬍ .0001 ⬍ .0001 .0166 ⬍ .0001 ⬍ .0001

379 Economic Implications of Continuous Hemodynamic Monitoring on Heart Failure Care William T. Abraham1, Mark F. Aaron2, Juan M. Aranda3, Mark A. O’Shaughnessy4, Mariell L. Jessup5, Krista Calfee1, Sergio Cavaglia6, Brandon Sparks6, Robert C. Bourge7, COMPASS-HF Investigators; 1Cardiology, The Ohio State University Heart Center, Columbus, OH; 2Cardiology, Saint Thomas Hospital/ The Heart Group, Nashville, TN; 3Cardiovascular Medicine, University of Florida Shands, Gainsville, FL; 4Cardiology, Parkview Memorial, Fort Wayne, IN; 5 Cardiology, Hospital of the University of Pennsylvania, Philadelphia, PA; 6Medtronic Inc, Minneapolis, MN; 7Cardiology, University of Alabama, Birmingham, AL Background: Over the past 2–3 decades, the number of hospitalizations for heart failure (HF) has relentlessly increased resulting in severe economic consequences. Hospitalizations for HF account for approximately two-thirds of the total cost of HF care, estimated between $28 billion and more than $60 billion annually. This growing burden of heart failure can be expected to increase during the next decade unless innovative interventions are implemented. Methods: The COMPASS-HF trial randomized 274 moderate to severe HF patients (85% NYHA III) to standard medical therapy (SMT) complemented by continuous hemodynamic monitoring (Chronicle group) versus SMT alone (control group). HF-related hospitalizations were evaluated in the whole population and as pre-specified in the protocol, in the NYHA class III patients. Results: Overall the event rate in the control group was 0.8 per 6 months and was reduced by 25% (p ⫽ 0.22) in the Chronicle group. In the Class III patients the hospitalizations were reduced by 43% (p ⫽ 0.028). The percentage of patients experiencing HF-related hospitalizations during the study was 39% in the control group and 26% (p ⫽ 0.029) in the Chronicle group (38% vs. 25% respectively when NYHA class III pts only are considered, p ⫽ 0.023). A 5-year model based on the result obtained for NYHA class III patients would suggest $18.8M saving in hospitalizations for every 1,000 patients monitored (considering an average DRG of $5,456 per discharge). Conclusion: A HF management strategy based on continuous hemodynamic monitoring on top of standard HF care results in significant reduction of HF-related hospitalizations. Given the high costs associated to HF hospitalizations, continuous hemodynamic monitoring could favorably impact health care expenses.

Background: COMPASS-HF is a single blind randomized trial in 274 NYHA Class III and IV patients designed to prove safety and clinical effectiveness on a heart failure (HF) management strategy based on continuous hemodynamic monitoring on top of optimized HF care (Chronicle group) vs. optimized HF care (control group). In the pre-specified subgroup analysis of NYHA class III patients (85% of the entire population) a HF management strategy guided by Chronicle resulted in 41% reduction of HF-related events (HFRE ⫽ Admission or Emergency Visit Rx with IV therapy) (p ⫽ 0.03). In contrast to all other sub-groups, the trend of improved outcomes in the Chronicle Group was not seen in Class IV patients (8 of the 22 Chronicle and 8 of the 18 Control patients had a HFRE). Aim: NYHA class IV patients in COMPASSHF were analyzed to better characterize this study population and the impact of continuous hemodynamic monitoring on this group. Results: Out of 274 patients enrolled in the COMPASS-HF study, a total of 40 patients (15%) were in NYHA class IV at baseline. Conclusions: In this subset of COMPASS-HF, the Chronicle and the control groups were significantly different at baseline with sicker patients assigned, by chance, to the Chronicle group as demonstrated by higher creatinine levels and lower exercise tolerance. This may partially explain the unexpected outcome in the NYHA Class IV patients than in other pre-specified subgroups. Additional analysis of this subgroup is ongoing, including an analysis of hemodynamics at baseline and leading to admission.

Baseline Creatinine Baseline 6min HWT (m) Baseline Minnesota QoL Event rate/ 6-month Clinical composite response Worsened Unchanged Improved

Control Group

Chronicle Group

P Value

1.2 ⫾ 0.4 227 ⫾ 137 68.6 ⫾ 23.5 0.88

1.9 ⫾ 0.7 107 ⫾ 84 79.2 ⫾ 23.6 1.66

0.0014 0.005 0.166 0.23 0.60

69% 0% 31%

57% 7% 36%

HWT ⫽ hall walk test; QoL ⫽ quality of life questionnaire

381 Sleep Disordered Breathing Detection in Congestive Heart Failure Subjects with Chronicle쑓 Implantable Hemodynamic Monitors Mark F. Aaron1, Yong K. Cho2, Jennifer A. Vance2, Tom Bennett2, Brooke H. Yorke1, Beth Davidson1, Jon Tumen1, Sue Dale2; 1Tennessee Cardiovascular Research Institute, St. Thomas Heart Institute, Nashville, TN; 2New Therapies and Diagnostics, Medtronic, Inc., Minneapolis, MN Background: A significant fraction of patients (pts) with heart failure (HF) suffer from undiagnosed sleep apnea (SA). The hemodynamic effects of SA can cause HF disease progression. Recently, the Chronicle has been demonstrated to decrease hospitalizations in HF pts through continuous hemodynamic monitoring. An algorithm to detect SA with the Chronicle could decrease HF morbidity further. Methods: In the ongoing Chronicle SA study, 7 pts underwent diagnostic sleep study (SS). Six pts were diagnosed with SA (Apnea Hypopnea Index ⬎5) and subsequently underwent a second SS for continuous positive airway pressure (CPAP) titration. The Chronicle hemodynamic data were simultaneously recorded with polysomnography (PSG) during SS. Twelve paired (PSG/Chronicle) datasets were used to develop a SA episode (⬎5 minutes of periodic apnea-recovery cycles)detection algorithm for the Chronicle. The algorithm using the development dataset was then evaluated against three metrics: (a) diagnosis of SA using apnea episode index (AEI) ⬎5, (b) episodeby-episode comparison (PSG/Chronicle) of detected apnea episodes, and (c) detection

S192

Journal of Cardiac Failure Vol. 11 No. 6 Suppl. 2005

of differences in the diagnostic and the CPAP titration SS in the same pts. Results: Using an AEI threshold (equivalent to AHI) of 5, the Chronicle algorithm correctly diagnosed 7 of 7 datasets with SA (sensitivity ⫽ 100%) and 4 of 5 datasets without SA (specificity ⫽ 80%). In an episode-by-episode comparison, the algorithm correctly identified 28 of 38 episodes, but also falsely detected 16 episodes. The algorithm detected CPAP correction of SA in 4 of 5 (80%) of pts. Conclusion: Hemodynamic data from the Chronicle can reliably detect and quantify periodic apnea-recovery cycles. The agreement with reference PSG demonstrates the feasibility of SA detection and monitoring using the Chronicle.

382 A Motivational Approach Effectively Improves Heart Failure Self-Care Victoria V. Dickson1, Janet McMahon1, Brendali F. Reis2, Steven L. Sayers3, Linda Hoke4, Barbara Riegel1; 1School of Nursing, University of Pennsylvania, Philadelphia, PA; 2Institute for Graduate Clinical Psychology, Widener University, Chester, PA; 3Department of Psychiatry, Philadelphia Veterans Medical Center, Philadelphia, PA; 4Department of Nursing, Hospital of the University of Pennsylvania, Philadelphia, PA Despite extensive patient education, many heart failure (HF) patients are unable to master self-care. Purpose: The purpose of this study was to identify 1) patients who improved in HF self-care or knowledge after receipt of an intervention by an APN trained in motivational interviewing and couples counseling, and 2) specific strategies that were effective. Methods: The intervention was provided during home visits (M ⫽ 3 1.5, range 1–6) following a HF hospitalization. Patients (N ⫽ 15) were followed for 3-months. Using a comparative mixed method design, one investigator categorized patients according to whether they improved in HF self-care (SelfCare of HF Index (SCHFI) .73 this sample) or HF knowledge (Representations, .75). A second investigator, blinded to the quantitative results, analyzed qualitative data obtained from audiotaped intervention sessions augmented with field notes and personal stories using Atlas software. Improvement was judged by consensus based on quantitative and qualitative results. Effective strategies were coded from the qualitative data. Results: Of the 15 patients who completed the study, sufficient data were available on 12 (66.7% female, 50% African American, age M ⫽ 57.75 (13.4) years, 66.7% functional class II, 75% systolic HF). Eleven of 12 (92%) improved in self-care, defined as ⱖ8 point improvement in SCHFI maintenance, management, or confidence scores or ⱖ5 points in knowledge plus qualitative evidence of behavioral change. Techniques used by the APN to improve self-care included motivational enhancement strategies such as linking self-care behaviors to symptoms, skill building, therapeutic listening, correcting misconceptions, activating resources and support, negotiating an action plan, and bridging the transition from hospital to home. Conclusions: A multifaceted counseling intervention that considers the unique needs of individual patients and actively targets ambivalence and motivation may be essential in promoting HF self-care.

383 Disparity in Emergency Department Treatment of Acute Decompensated Heart Failure by Gender Deborah B. Diercks1, W. Franklin Peacock2, J. Douglas Kirk1, Jim E. Weber3; 1 Emergency Medicine, University of California, Davis Medical Center, Sacramento, CA; 2Emergency Medicine, Cleveland Clinic, Cleveland, OH; 3Emergency Medicine, University of Michigan, Ann Arbor, MI Background: Previous research suggests that there is a disparity in treatment between women and men with congestive heart failure. Objective: To treatment differences noted between men and women who present to the emergency department (ED) with acute decompensated heart failure (ADHF). Methods: Prospective, convenience sample of patients presenting to the ED patients with a complaint shortness of breath or leg swelling and an ED diagnosis of ADHF from 1/1/2003 to 12/1/2004. Acute treatments were defined as diuretics, Angiotenisn Converting Enzyme (ACE) inhibitors, and vasodilators (nitrates, nesiritide, nitroprusside). Only medications administered during ED care were evaluated. Descriptive statistics were utilized to describe catergorical variables, a p value ⬍ 0.05 was significant. A multivariate analysis including variables of age, radiographic evidence of heart failure, physical findings (rales, jugular venous distention), past medical history (diabetes, hypertension, coronary artery disease, heart failure, COPD), social history (current or remote smoking), if BNP testing was done, and current medications (Diuretics, ACE Inhibitors, Nitrates, B-blockers) was preformed determine if gender was an independent predictor of acute therapy. Results: The study cohort consisted of 506 patients was compromised of 282 (56%) men and 224 (44%) women. Although women were older than men (59 yrs vs 55 yrs, p ⫽ 0.026), there was no difference in admission rates by gender (81% vs 80%, p ⫽ NS). There was no difference in the presence of JVD, S4, rales on exam, or radiographic evidence of pulmonary edema between women and men. Women were less likely to receive a diuretic (58% vs 67%, OR 0.6, 95% CI 0.4–.8). There was no difference by sex in vasodilator use (52% vs 56%, p ⫽ .6), and ACE inhibitor administration (p ⫽ 0.8). In multivariate analysis female sex was the only independent predictors of receiving a diuretic in the ED (OR 0.6, 95% CI 0.4–2.5). Conclusion: Women with ADHF were treated less often with a diuretic in the ED when compared to men. Further study is required to identify barriers to acute pharmacologic therapy in women.

384 Audioelectric Cardiographic Parameters for the Bedside Detection of Heart Failure Kimberly L. Dulaney1, Patti Adams1, Lukas Jantac1, Sean P. Collins2, W. Frank Peacock3, E. Magnus Ohman1, Patricia Chang1, Carla Sueta1; 1Cardiology, University of North Carolina, Chapel Hill, NC; 2Emergency Medicine, University of Cincinnati, Cincinnati, OH; 3Emergency Medicine, Cleveland Clinic, Cleveland, OH The third heart sound (S3) often provides acoustical evidence of heart failure (HF). Using new audioelectric technology, we tested the ability of the electronically recorded S3 and other computerized audioelectric cardiographic parameters to detect HF in a variety of clinical settings. Hypothesis: A combination of ECG and cardiac acoustical data can distinguish between subjects with and without HF. Methods: We obtained recordings from 482 age and heart rate matched subjects (272 men, ages 25–90, mean 60 years) using Audicor, a device that records and algorithmically interprets simultaneous 12-lead ECG and heart sound data using acoustical sensors attached to the V3 and V4 leads. Of the 482 subjects, 131 had clinical evidence of HF and 351 did not. Of the 131 patients with HF, 58 had presented to an emergency department (ED) with acute HF and 73 were being managed in a chronic HF clinic. Of the 351 subjects without HF, 198 were ambulatory and asymptomatic, and 153 had presented to an ED with dyspnea, but were found to have a non-HF etiology. The mean ⫾ SD BNP in pg/mL was 1247 ⫾ 1269 in ED patients with heart failure, and 251 ⫾ 595 pg/mL in ED patients without HF. The mean ⫾ SD pro-BNP in pg/mL was 2771 ⫾ 5264 in chronic HF clinic. The clinical diagnoses were made independent of the Audicor findings. We evaluated the following computerized parameters: the S3, the Electromechanical Activation Time (EMAT) (the time from the onset of the Q wave to the accoustical detection of closure of the mitral valve), and EMAT as a proportion of the cardiac cycle (EMAT/RR). Unpaired t-test and chi square analysis were utilized to test statistical significance between the two groups. Results: The following table shows the results (mean ⫾ SD) Parameter

No HF (n ⫽ 351)

HF (n ⫽ 131)

P

9% 97 ⫾ 19 ms 0.11 ⫾ .03 ms

34% 124 ⫾ 26 ms 0.15 ⫾ .04 ms

⬍ 0.001 ⬍ 0.00001 ⬍ 0.00001

S3 Prevalance EMAT EMAT/RR

For these parameters there was no significant difference between ambulatory patients who were asymptomatic and those who had dyspnea without heart failure. Conclusions: HF patients demonstrated a prolonged EMAT when compared to patients without HF. Audioelectric cardiography may provide a new, non-invasive method for diagnosing patients with heart failure.

385 Functional Class, Quality of Life, and 6 Minute Walk Test in HF Patients with Low vs. Preserved Systolic Function: Baseline Data from HearT-I Mark E. Dunlap1,2, Katherine S. Thweatt1, Michelle Casedonte1, Ileana L. Pina1,2, David Baker3, David Aron1,2; 1Medicine/Cardiology, Louis Stokes VA Medical Center, Cleveland, OH; 2Medicine, Case Western Reserve University, Cleveland, OH; 3 Medicine, Northwestern University Fineberg School of Medicine, Chicago, IL Background: Patients with HF and preserved systolic function (PSF) account for a large proportion of HF hospitalizations, although their overall prognosis appears to be more favorable compared to HF patients with reduced EF. Less is known about health status and functional status between these 2 groups of HF patients. We sought to compare symptoms, functional class (NYHA), quality of life (QOL), and 6 Minute Walk Test (6-MWT) in HF patients with low EF vs. PSF. Methods: We studied patients at baseline who had been enrolled into a Heart failure Telephone Intervention study, HearT-I. All patients had a current or past history of HF, and were enrolled regardless of EF. Two health status measures, the Kansas City Cardiomyopathy Questionnaire (KCCQ) and the SF-36V were completed at baseline in all patients, and 6-MWT was performed in all patients who had a KCCQ composite score ⬎ 32. Results: Out of 469 patients, 379 had assessment of EF within 1 year prior to enrollment. HF patients enrolled in this study had marked limitation of 6-MTW, demonstrated decreased QOL, and showed reduced functional capacity, with similar scores on most measures between groups (see table). Conclusions: HF patients with PSF have similar functional capacity and health status compared to patients with reduced EF indicating similar degrees of limitations. While mortality may be lower in this group, these patients remain highly symptomatic and perform as poorly as do patients with low EF. Health Status and EF

n NYHA 6-MWT (m) KCCQ Symptom change Symptom severity Self efficacy QOL Total symptom score Funtional status Clinical summary SF-36V Physical functioning General health Vitality Social Mental Emotional

Low EF (ⱕ40%)

PSF(EF⬎40%)

p-value

263 3.50 ⫾ 0.04 257 ⫾ 7

116 3.54 ⫾ 0.05 253 ⫾ 10

– 0.51 0.96

50.2 ⫾ 1.3 59.7 ⫾ 1.4 67.3 ⫾ 1.5 48.2 ⫾ 1.5 56.7 ⫾ 1.3 53.9 ⫾ 1.2 50.4 ⫾ 1.2

46.2 ⫾ 1.8 55.5 ⫾ 2.1 66.5 ⫾ 2.3 52.8 ⫾ 2.1 52.9 ⫾ 2.1 51.0 ⫾ 1.8 50.6 ⫾ 1.8

0.08 0.10 0.78 0.09 0.13 0.19 0.91

61.9 ⫾ 1.5 47.7 ⫾ 0.9 36.1 ⫾ 1.1 61.2 ⫾ 1.7 65.6 ⫾ 1.3 53.7 ⫾ 1.8

58.2 ⫾ 2.5 48.4 ⫾ 1.3 37.1 ⫾ 1.8 65.7 ⫾ 2.5 66.9 ⫾ 1.8 60.3 ⫾ 2.7

0.76 0.69 0.64 0.14 0.58 0.04