Serum Total Cholesterol Level is Associated With Increased Short- and Long-Term Mortality in Patients Hospitalized With Non-Ischemic Acute Heart Failure: A Report from Korean Heart Failure (KorHF) Registry

S92 Journal of Cardiac Failure Vol. 16 No. 8S August 2010 have changed in the past 20 years, we calculated the hazard ratios for cardiac, non cardiac and total mortality compared to the 2003 Life Tables for the control arms of 69 published heart failure trials from major general medical or cardiology journals. For this study, we utilized a competing risk model for mortality utilizing a probabilistic programming language. We then evaluated how mortality changed over time, from 1991 until the present. We present the data in 5 year increments based on the date the study was published. Of the 69 selected studies, 54 provided separate tallies for cardiac and noncardiac mortality, while 15 only reported total mortality. The population average age was 64 with 76% male. The average follow up was 2 years (range 0.2 to 4.75). The placebo arm averaged 1274 patients (110-5438) with an average of 286 deaths per study. The following table demonstrates the reduction hazard ratio for death from cardiac, noncardiac and all causes during the 4 time frames. Results

Time Frame 1991-1995 1996-2000 2001-2005 2006-2010

# of Studies

NYHA Class

Cardiac HR

Noncardiac HR

Total Mortality HR

13 15 23 18

2.4 2.6 2.4 2.5

33.1 20.7 14.2 9.9

0.82 1.27 0.99 1.04

10.3 7.2 5.1 3.8

Over the past 20 years, overall mortality rates for heart failure patients have decreased by 63%, while cardiac mortality in heart failure trials has decreased by almost 70%. This data suggests that historical mortality rates for heart failure patients are not representative of current expectations and may explain the repeated failure of recent heart failure trials to further impact these, already low, mortality rates. Because our analysis was based on mortality rates found in the control arm of published clinical trials, where patients are likely to be receiving contemporary care, the data suggests that optimal use of our evidence based therapies has significantly reduced the risk of death in patients with moderate heart failure.

303 Serum Total Cholesterol Level is Associated With Increased Short- and LongTerm Mortality in Patients Hospitalized With Non-Ischemic Acute Heart Failure: A Report from Korean Heart Failure (KorHF) Registry Dong-Ju Choi1, Sung-Ji Park2, Chang-Hwan Yoon1, Gu-Yung Cho1, Jin-Ho Choi2, Eu-Seok Jeon2; 1Cardiovascular Center, Seoul National University Hospital, Seongnam, Gyeonggi-Do, Korea; 2Cardiovascular Image Center, Sunggunkwan University Samsung Hospital, Seoul, Korea Background: Hypercholesterolemia is a major risk factor for incident coronary artery disease, mortality, and prevalence of heart failure (HF). However, patients with advanced HF often have low cholesterol, which is associated with a paradoxical poor prognosis. The relationship between lipid levels and outcomes in non-ischemic acute HF has not been studied. This study investigates the relationship between cholesterol levels and short- and long-term mortality in patients hospitalized with acute HF. Methods: We analyzed a cohort of 1,561 patients with HF of non-ischemic origin. Fasting lipid panels were measured at time of admission. Patients were stratified into quartiles of TC (Q1 !131, Q2 132-158, Q3 159-189, and Q4 O190 mg/dL). Results: The cohort was 842 (47.1%) male, mean age was 65.3 + / 15.8 years, and LVEF was 36.9 + / 14.7%. Patients with lower serum TC had worse hemodynamic profiles, lower LVEF, and higher cardiac natriuretic peptide level. Low TC was associated with increased risk of 1-year all-cause mortality (24%, 20%, 15%, 14% for Q1-Q4 respectively, P ! .0001) as well as long-term all-cause mortality (69%, 29%, 22%, 20% for Q1-Q4 respectively, P ! .0001). On multivariate analysis, adjusting for multiple HF prognostic factors, low TC proved to be an independent predictor of worse outcomes, with hazard ratios for all-cause mortality of 1.163 (95% C.I.;1.085-1.227, p ! 0.001). Conclusions: Low TC is strongly associated with increased mortality in patients with nonischemic, HF. Further study is warranted to determine the nature of this relationship, optimal lipid levels, and the role of statins in patients with non-ischemic HF.

304 Venous Thromboembolism Prophylaxis among Elderly Heart Failure Patients and 60-Day Outcomes: An Analysis from ADHERE Linked to Medicare Claims Robb D. Kociol1, Bradley G. Hammill1, Adrian F. Hernandez1, Winslow Klaskala2, Roger M. Mills3, Lesley H. Curtis1, Gregg C. Fonarow4; 1DCRI, Durham, NC; 2J&J Pharmaceuticals, Titusville, NJ; 3Ortho-McNeil Janssen, Raritan, NJ; 4UCLA Cardiomyopathy Center, Los Angeles, CA Background: Hospitalized medically ill patients are at increased risk for venous thromboembolism (VTE). While antithrombotic prophylaxis regimens have reduced VTE risk in medically ill patients, their effectiveness for post-discharge outcomes among HF patients is unknown. Methods: We used ADHERE registry data linked to Medicare claims. We included patients hospitalized from 2001 to 2004, discharged alive. Patients on warfarin or IV heparin were excluded. We estimated observed 60-day rates of mortality, thromboembolic events, major adverse cardiovascular events (MACE), and rehospitalization for HF using estimates based on the cumulative incidence function. We used propensity score methods to estimate the effect of VTE prophylaxis on each outcome. Results: Of 39,995 patients in 265 hospitals, 13,194 (33%) received

antithrombotic VTE prophylaxis. The VTE prophylaxis cohort was younger (79.2 vs. 79.7 years) and included more African-Americans (16.9% vs. 12.5%). They were less likely to have CAD (57.5% vs. 60.1%), renal insufficiency (25.7% vs. 29.2%), atrial fibrillation (19.7% vs. 21.3%), or anemia on admission (Hb !9 g/dl; 4.3% vs. 6.8%). Observed events rates by VTE prophylaxis were similar (Figure). In unadjusted analysis, VTE prophylaxis was not associated with outcomes. After weighting by the inverse probability of treatment, VTE prophylaxis was not associated with mortality, thromboembolic events, or HF rehospitalization, but was associated with a slight increase in MACE. Conclusion: Antithrombotic VTE prophylaxis is underutilized among hospitalized HF patients. The impact of thromboprophylaxis with currently available inpatient regimens on post-discharge outcomes is minimal. Research is needed to develop VTE prophylaxes which are clinically effective and improve outcomes.

305 Does Change in Serum Sodium Concentration Predict Outcomes in Patients Hospitalized With Heart Failure and Hyponatremia? Vinay D. Madan, Michael W. Rich; Cardiology, Washington University School of Medicine, Saint Louis, MO Background: Hyponatremia is a common electrolyte abnormality among patients hospitalized with heart failure (HF) and it is a marker for increased short-term and longterm mortality. However, little is known about the time-course of hyponatremia and whether changes in serum sodium (Na) levels impact clinical outcomes. Objective: To determine the relationship between change in serum Na levels and mortality in patients hospitalized with HF and hyponatremia. Methods: 322 patients hospitalized with decompensated HF and serum Na !135 mmol/L were evaluated retrospectively. Following hospital discharge, the first follow-up Na value obtained within a 60-270 day period was recorded, and patients were classified into 3 groups based on whether the Na value increased (O1 mmol/L), decreased (!1 mmol/L), or remained unchanged (+ /-1 mmol/L) relative to the baseline value. Kaplan-Meier survival curves were constructed to illustrate mortality as a function of change in Na concentration over time, and multiple linear regression and Cox-proportional hazards analyses were performed to determine if change in serum Na concentration was an independent predictor of mortality after adjusting for relevant covariates. Results: The mean age was 66 years, 45% were women, and 55% were Caucasian. The mean baseline Na level was 131 mmol/L and the mean ejection fraction was 32.5%. 222 patients (68.9%) exhibited an increase in Na during follow-up; in 57 patients (17.7%) the level was unchanged and in 43 patients (13.4%) there was a decrease in Na level. Figure 1 depicts survival curves of these groups using the Cox-proportional hazards model adjusted for all