- Email: [email protected]
Adverse Renal Effects of Intravenous Diuretics in Decompensated Heart Failure. Data from the ADHERE Registry
The 9th Annual Scientific Meeting
•
HFSA
S167
Outcomes 288
290
The Development of a Database of Patients Admitted with Primary and Secondary Heart Failure from 2.5 Million Admissions in 350 U.S. Hospitals: Patient Characteristics and Outcomes Joseph F. Dasta1, Amy J. Durtschi2, Trent McLaughlin3, Rob Padley2, David S. Feldman1; 1Colleges of Pharmacy and Medicine, The Ohio State University, Columbus, OH; 2Health Economics and Outcomes Research, Abbott Laboratories, Abbott Park, IL; 3Outcomes Research, NDC Health, Phoenix, AZ
Metabolic Syndrome Predicts a Higher Risk for Heart Failure Hospitalizations in the Elderly Joann Goring, Jennifer Listerman, Robert Huang, Mary Alice Nading, Carrie Geisberg, Javed Butler; Cardiology, Vanderbilt University, Nashville, TN
Background: Acute decompensated heart failure (HF) is recognized as a serious public health problem. An evaluation of patients admitted for (primary) HF and patients developing (secondary) HF after admission is needed to characterize the patient population and the effectiveness of current management strategies. The purpose of this study is to describe the development of a database of patients with both primary and secondary HF to understand the characteristics, resource usage, and outcomes resulting from the care of these two patient populations. Methods: All inpatient admissions in 2003 from a database of 350 geographically diverse hospitals (NDCHealth) were reviewed for a discharge diagnosis of primary and secondary HF (ICD9 428.xxx). Patient demographcs, drugs prescribed, resources used, and clinical outcomes were obtained. Results: Of the 2,515,872 admissions, 498,713 (19.8%) had HF, with the following characteristics: 55.4% female, 22.3% (40–64 y), 21.9% (65–74 y), and 53.4% (over 75 y). Most admissions had secondary HF 367,656 (73.7%), compared with primary HF 131,057 (26.3%). Patients were admitted from the ED (65.9%), routine admission (14%), transfer from another facility (8.6%), and referral (7.8%). 9.8% received inotropes, 8.8% received vasodilators, and 70% received iv furosemide, often in combination. 49.7% were discharged to home, 20.2% to a treatment facility, 15.9% to a skilled nursing facility, and 5.8% to a home care service. The average hospital cost was $18,589 and length of hospital stay was 8.7 days. Overall, 32% of admissions required ICU/CCU care, and in-hospital mortality was 7.0%. However, secondary HF patients had a higher mortality rate (8.0 vs 4.3%) and higher hospital costs ($20,084 vs $14,395). Cumulative hospital costs of primary HF patients were $1.88 billion while secondary HF patients totaled $7.38 billion. Conclusion: This database describes the care provided to nearly a half-million patients with HF, representing 20% of 2.5 million admissions. Secondary HF patients not only account for 74% of HF admissions, they account for 79.6% of total HF costs during the study period. The high mortality rate and large economic burden on the health care system suggest improvements in HF care are needed.
Background: The importance of the metabolic syndrome (MetSyn) is largely derived from data in the middle-aged and younger populations. Whether MetSyn predicts a higher risk for adverse mortality and heart failure (HF) hospitalizations among the elderly is not known. Methods: We studied the impact of MetSyn in 3035 participants in the Health ABC study (51% women, 42% black, aged 70–79 yrs). All cause mortality, and HF and all-cause hospitalization were assessed during a six year followup period. Results: Prevalence of MetSyn at baseline was 38%. There were a total of 434 deaths and 231 HF hospitalizations; 1784 (59%) of the subjects had at least one hospitalization for any cause. Heart failure hospitalizations occurred significantly more in subjects with MetSyn (10.0% vs. 6.1%, p ⬍ 0.001). Similarly, overall hospitalizations (63.1% vs. 56.1%, p ⬍ 0.001) were more common in subjects with MetSyn. No significant difference in overall mortality was seen between the two groups (14.5% vs. 14.2%, p ⫽ 0.85). After adjusting for demographics, weight, smoking status and diabetes, patients with MetSyn were at a significantly higher risk for HF hospitalization (OR 1.49, 95% CI 1.10–2.00). Conclusion: MetSyn is associated with significant risk for HF hospitalizations in the elderly.
289
291
Admission B-Type Natriuretic Peptide Levels in Acute Decompensated Heart Failure Predict In-Hospital Mortality: An Analysis of 48,629 Hospitalizations in ADHERE Gregg C. Fonarow1, Lynne W. Stevenson2, William F. Peacock3, Christopher O. Phillips4, Michael M. Givertz2, Margarita Lopatin5; 1Ahmanson-UCLA Cardiomyopathy Center, University of California, Los Angeles Medical Center, Los Angeles, CA; 2Department of Medicine, Brigham & Women’s Hospital, Boston, MA; 3 Department of Emergency Medicine, The Cleveland Clinic Foundation, Cleveland, OH; 4Department of General Internal Medicine, The Cleveland Clinic Foundation, Cleveland, OH; 5Department of Biostatistics, Scios Inc., Fremont, CA
Adverse Renal Effects of Intravenous Diuretics in Decompensated Heart Failure. Data from the ADHERE Registry Franklin W. Peacock1, Charles L. Emerman2, Maria Rosa Costanzo3, For the ADHERE Scientific Advisory Committee; 1Emergency Medicine, Cleveland Clinic Foundation, Cleveland, OH; 2Emergency Medicine, Case Western Reserve University, Cleveland, OH; 3Midwest Heart Specialists, Naperville, IL
Background: Levels of B-type natriuretic peptide (BNP) have been demonstrated to facilitate the diagnosis of heart failure (HF) and predict mortality in chronic HF. Whether admission BNP levels are predictive of in-hospital mortality in acutely decompesated HF has not been well studied. Methods: Hospitalizations for acutely decompensated HF from April 2003 to December 2004 entered into ADHERE were analyzed. BNP levels on admission were obtained in 48,629 (63%) out of 77,467 hospitalization episodes. In-hospital mortality was assessed by BNP quartiles in the entire cohort and in patients with reduced (19,544) as well as preserved (18,164) left ventricular systolic function using Chi-Square and logistic regression models. Results: Quartiles of BNP were Q1 ( ⬍ 430), Q2 (430–839), Q3 (840–1729), and Q4 (ⱖ1730). BNP levels were ⬍100 pg/mL in 3.3%. Patients in Q1 vs. Q4 were younger, more likely to be female, had lower creatinine, and higher LVEF. In-hospital mortality rates increased by BNP levels: Q1 1.9%, Q2 2.8%, Q3 3.9%, Q4 6.0%, P ⬍0.0001. BNP quartile remained highly predictive of mortality even after adjustment for age, gender, SBP, BUN, creatinine, sodium, pulse, dyspnea at rest. Q4 vs. Q1, adjusted OR 2.23 (1.91–2.62), P ⬍0.0001. BNP quartiles independently predicted mortality in patients with LVEF ⬍0.40 and LVEF ⱖ0.40 (Figure). Conclusions: Admission BNP is a significant predictor of in-hospital mortality in acutely decompensated HF, independent of clinical and other laboratory variables. BNP levels are predictive of mortality risk in HF patients with either reduced or preserved left ventricular ejection fraction.
Background: Intravenous diuretics are commonly given to patients with acute decompensated heart failure. Prior studies have shown an association between worsening renal function and adverse outcomes. The purpose of this study was to utilize the results of the ADHEREaˆ heart failure registry to explore the association between intravenous diuretic use and worsening renal function. Methods: 61,207 ADHERE enrollments from 245 hospitals with both admission and discharge creatinine, not on chronic dialysis and w/o inhospital vasoactive use were analyzed. Pts with creatinine ⬎6 mg/dL and LOS ⬍ 24 hrs were excluded. Change in Cr clearance was assessed in 36, 571 of 61,207 episodes with admission and discharge weight. Chi-square, ANOVA and Wilcoxon tests were used in univariate analyses. Multiple logistic regression was used to adjust comparisons for clinically meaningful covariates. Results: 5,594 (9.1%) out of 61,207 patient episodes did not receive iv diuretics at any point during hospitalization. IV diuretic patient episodes (Group 1) were slightly older (73.8 vs 72.1) and less likely to be male (44.7% vs 49.1%) compared to no iv diuretic episodes (Group 2), both P ⬍ .0001. The median BNP level was higher (726 vs 493), the mean systolic BP was higher (145 vs 137) and the EF was higher (40.6 vs 37.1) for Group 1, all P ⬍ .0001. Group 1 was more likely to have an increase in Cr ⬎0.5 compared to Group 2 (6.2% vs 4.0%; p ⬍ .0001). Group 1 was more likely to have a significant decrease in Cr clearance of more than 10 ml/min (23.4% vs 15.0%; p ⬍ .0001). Group 1 had a greater median LOS (4.1 vs 3.8 days), greater likelihood of a LOS⬎ 4 days (52.6% vs 44.9%) and a greater likelihood of LOS in the ICU⬎3 days (31.6% vs 21.9%), all P ⬍ .0001. After adjustment for race, gender, age, systolic BP, sodium, creatinine, history of diabetes, hypertension, or hemoglobin ⬍ 12, Group 1 was significantly more likely to have an increase in Cr ⬎ 0.5 compared to group 2 (OR ⫽ 1.51,95% CI ⫽ (1.31–1.75) p ⬍ .0001) and more likely to have a decrease in CrCl ⬎ 10 ml/min (OR ⫽ 1.87, 95% CI ⫽ (1.66–2.11) p ⬍ .0001). Conclusions: Intravenous diuretics are associated with worsening renal function in hospitalized patients with decompensated heart failure. These results add to the evidence of adverse effects of intravenous diuretics in this patient population.
•
HFSA
S167
Outcomes 288
290
The Development of a Database of Patients Admitted with Primary and Secondary Heart Failure from 2.5 Million Admissions in 350 U.S. Hospitals: Patient Characteristics and Outcomes Joseph F. Dasta1, Amy J. Durtschi2, Trent McLaughlin3, Rob Padley2, David S. Feldman1; 1Colleges of Pharmacy and Medicine, The Ohio State University, Columbus, OH; 2Health Economics and Outcomes Research, Abbott Laboratories, Abbott Park, IL; 3Outcomes Research, NDC Health, Phoenix, AZ
Metabolic Syndrome Predicts a Higher Risk for Heart Failure Hospitalizations in the Elderly Joann Goring, Jennifer Listerman, Robert Huang, Mary Alice Nading, Carrie Geisberg, Javed Butler; Cardiology, Vanderbilt University, Nashville, TN
Background: Acute decompensated heart failure (HF) is recognized as a serious public health problem. An evaluation of patients admitted for (primary) HF and patients developing (secondary) HF after admission is needed to characterize the patient population and the effectiveness of current management strategies. The purpose of this study is to describe the development of a database of patients with both primary and secondary HF to understand the characteristics, resource usage, and outcomes resulting from the care of these two patient populations. Methods: All inpatient admissions in 2003 from a database of 350 geographically diverse hospitals (NDCHealth) were reviewed for a discharge diagnosis of primary and secondary HF (ICD9 428.xxx). Patient demographcs, drugs prescribed, resources used, and clinical outcomes were obtained. Results: Of the 2,515,872 admissions, 498,713 (19.8%) had HF, with the following characteristics: 55.4% female, 22.3% (40–64 y), 21.9% (65–74 y), and 53.4% (over 75 y). Most admissions had secondary HF 367,656 (73.7%), compared with primary HF 131,057 (26.3%). Patients were admitted from the ED (65.9%), routine admission (14%), transfer from another facility (8.6%), and referral (7.8%). 9.8% received inotropes, 8.8% received vasodilators, and 70% received iv furosemide, often in combination. 49.7% were discharged to home, 20.2% to a treatment facility, 15.9% to a skilled nursing facility, and 5.8% to a home care service. The average hospital cost was $18,589 and length of hospital stay was 8.7 days. Overall, 32% of admissions required ICU/CCU care, and in-hospital mortality was 7.0%. However, secondary HF patients had a higher mortality rate (8.0 vs 4.3%) and higher hospital costs ($20,084 vs $14,395). Cumulative hospital costs of primary HF patients were $1.88 billion while secondary HF patients totaled $7.38 billion. Conclusion: This database describes the care provided to nearly a half-million patients with HF, representing 20% of 2.5 million admissions. Secondary HF patients not only account for 74% of HF admissions, they account for 79.6% of total HF costs during the study period. The high mortality rate and large economic burden on the health care system suggest improvements in HF care are needed.
Background: The importance of the metabolic syndrome (MetSyn) is largely derived from data in the middle-aged and younger populations. Whether MetSyn predicts a higher risk for adverse mortality and heart failure (HF) hospitalizations among the elderly is not known. Methods: We studied the impact of MetSyn in 3035 participants in the Health ABC study (51% women, 42% black, aged 70–79 yrs). All cause mortality, and HF and all-cause hospitalization were assessed during a six year followup period. Results: Prevalence of MetSyn at baseline was 38%. There were a total of 434 deaths and 231 HF hospitalizations; 1784 (59%) of the subjects had at least one hospitalization for any cause. Heart failure hospitalizations occurred significantly more in subjects with MetSyn (10.0% vs. 6.1%, p ⬍ 0.001). Similarly, overall hospitalizations (63.1% vs. 56.1%, p ⬍ 0.001) were more common in subjects with MetSyn. No significant difference in overall mortality was seen between the two groups (14.5% vs. 14.2%, p ⫽ 0.85). After adjusting for demographics, weight, smoking status and diabetes, patients with MetSyn were at a significantly higher risk for HF hospitalization (OR 1.49, 95% CI 1.10–2.00). Conclusion: MetSyn is associated with significant risk for HF hospitalizations in the elderly.
289
291
Admission B-Type Natriuretic Peptide Levels in Acute Decompensated Heart Failure Predict In-Hospital Mortality: An Analysis of 48,629 Hospitalizations in ADHERE Gregg C. Fonarow1, Lynne W. Stevenson2, William F. Peacock3, Christopher O. Phillips4, Michael M. Givertz2, Margarita Lopatin5; 1Ahmanson-UCLA Cardiomyopathy Center, University of California, Los Angeles Medical Center, Los Angeles, CA; 2Department of Medicine, Brigham & Women’s Hospital, Boston, MA; 3 Department of Emergency Medicine, The Cleveland Clinic Foundation, Cleveland, OH; 4Department of General Internal Medicine, The Cleveland Clinic Foundation, Cleveland, OH; 5Department of Biostatistics, Scios Inc., Fremont, CA
Adverse Renal Effects of Intravenous Diuretics in Decompensated Heart Failure. Data from the ADHERE Registry Franklin W. Peacock1, Charles L. Emerman2, Maria Rosa Costanzo3, For the ADHERE Scientific Advisory Committee; 1Emergency Medicine, Cleveland Clinic Foundation, Cleveland, OH; 2Emergency Medicine, Case Western Reserve University, Cleveland, OH; 3Midwest Heart Specialists, Naperville, IL
Background: Levels of B-type natriuretic peptide (BNP) have been demonstrated to facilitate the diagnosis of heart failure (HF) and predict mortality in chronic HF. Whether admission BNP levels are predictive of in-hospital mortality in acutely decompesated HF has not been well studied. Methods: Hospitalizations for acutely decompensated HF from April 2003 to December 2004 entered into ADHERE were analyzed. BNP levels on admission were obtained in 48,629 (63%) out of 77,467 hospitalization episodes. In-hospital mortality was assessed by BNP quartiles in the entire cohort and in patients with reduced (19,544) as well as preserved (18,164) left ventricular systolic function using Chi-Square and logistic regression models. Results: Quartiles of BNP were Q1 ( ⬍ 430), Q2 (430–839), Q3 (840–1729), and Q4 (ⱖ1730). BNP levels were ⬍100 pg/mL in 3.3%. Patients in Q1 vs. Q4 were younger, more likely to be female, had lower creatinine, and higher LVEF. In-hospital mortality rates increased by BNP levels: Q1 1.9%, Q2 2.8%, Q3 3.9%, Q4 6.0%, P ⬍0.0001. BNP quartile remained highly predictive of mortality even after adjustment for age, gender, SBP, BUN, creatinine, sodium, pulse, dyspnea at rest. Q4 vs. Q1, adjusted OR 2.23 (1.91–2.62), P ⬍0.0001. BNP quartiles independently predicted mortality in patients with LVEF ⬍0.40 and LVEF ⱖ0.40 (Figure). Conclusions: Admission BNP is a significant predictor of in-hospital mortality in acutely decompensated HF, independent of clinical and other laboratory variables. BNP levels are predictive of mortality risk in HF patients with either reduced or preserved left ventricular ejection fraction.
Background: Intravenous diuretics are commonly given to patients with acute decompensated heart failure. Prior studies have shown an association between worsening renal function and adverse outcomes. The purpose of this study was to utilize the results of the ADHEREaˆ heart failure registry to explore the association between intravenous diuretic use and worsening renal function. Methods: 61,207 ADHERE enrollments from 245 hospitals with both admission and discharge creatinine, not on chronic dialysis and w/o inhospital vasoactive use were analyzed. Pts with creatinine ⬎6 mg/dL and LOS ⬍ 24 hrs were excluded. Change in Cr clearance was assessed in 36, 571 of 61,207 episodes with admission and discharge weight. Chi-square, ANOVA and Wilcoxon tests were used in univariate analyses. Multiple logistic regression was used to adjust comparisons for clinically meaningful covariates. Results: 5,594 (9.1%) out of 61,207 patient episodes did not receive iv diuretics at any point during hospitalization. IV diuretic patient episodes (Group 1) were slightly older (73.8 vs 72.1) and less likely to be male (44.7% vs 49.1%) compared to no iv diuretic episodes (Group 2), both P ⬍ .0001. The median BNP level was higher (726 vs 493), the mean systolic BP was higher (145 vs 137) and the EF was higher (40.6 vs 37.1) for Group 1, all P ⬍ .0001. Group 1 was more likely to have an increase in Cr ⬎0.5 compared to Group 2 (6.2% vs 4.0%; p ⬍ .0001). Group 1 was more likely to have a significant decrease in Cr clearance of more than 10 ml/min (23.4% vs 15.0%; p ⬍ .0001). Group 1 had a greater median LOS (4.1 vs 3.8 days), greater likelihood of a LOS⬎ 4 days (52.6% vs 44.9%) and a greater likelihood of LOS in the ICU⬎3 days (31.6% vs 21.9%), all P ⬍ .0001. After adjustment for race, gender, age, systolic BP, sodium, creatinine, history of diabetes, hypertension, or hemoglobin ⬍ 12, Group 1 was significantly more likely to have an increase in Cr ⬎ 0.5 compared to group 2 (OR ⫽ 1.51,95% CI ⫽ (1.31–1.75) p ⬍ .0001) and more likely to have a decrease in CrCl ⬎ 10 ml/min (OR ⫽ 1.87, 95% CI ⫽ (1.66–2.11) p ⬍ .0001). Conclusions: Intravenous diuretics are associated with worsening renal function in hospitalized patients with decompensated heart failure. These results add to the evidence of adverse effects of intravenous diuretics in this patient population.