Investigators from the Acute Decompensated Heart Failure National Registry (ADHERE) have derived a tool to predict high, intermediate, and low risk for death in hospitalized heart failure patients from an observational evaluation of 39 clinical variables in more than 65,000 patients hospitalized with heart failure. The tool demonstrates that the risk of in-hospital death can be easily predicted in these patients using the results of two simple blood tests and a single blood pressure measurement.
According to the American Heart Association, heart failure is a serious condition that affects nearly 5 million Americans, with 550,000 new cases each year. It is the most frequent cause of hospitalization in patients older than 65 years.
Heart failure, typically a chronic and progressive condition, occurs when the heart muscle becomes weakened or damaged and lacks the strength to pump enough blood to meet the body's needs. Symptoms may include shortness of breath, swelling of extremities, weakness or tiredness, and coughing when lying down.
A heart failure patient whose condition suddenly or rapidly deteriorates may be described as "acutely decompensated," with symptoms that warrant care in a hospital or acute care setting.
Assessing a patient's risk for death on hospital admission helps guide medical decision making and outcomes for patients with acute coronary syndromes such as heart attack and unpredictable bouts of chest pain. Such assessments can help clinicians determine which patients need to be treated in a more closely monitored setting and may be helpful in designing clinical trials where risk is balanced across treatment groups or only patients of specific mortality risk are enrolled. Until now, useful risk assessment tools for acutely decompensated heart failure (ADHF) patients did not exist.
Using a type of statistical analysis known as CART (Classification And Regression Tree), ADHERE study investigators analyzed a multitude of clinical variables to develop a tool to assess mortality risk in heart failure patients. They were able to define three clinical characteristics that put hospitalized heart failure patients at greatest risk of mortality: high levels of blood urea nitrogen (BUN 43 mg/dL), a measurement of kidney function; low systolic blood pressure (SBP <115 mm Hg); and high serum creatinine (Cr>2.75 mg/dL) levels, another measure of kidney function, were also used to accurately predict high, low, and intermediate risks for mortality. The validity of the tool was then tested using data from additional hospitalizations.
The overall mortality risk for patients hospitalized with acute heart failure was 4.1%. The tool determined mortality risk levels starting from low risk at 2.1%, up to 21.9% for patients at the highest mortality risk.
On the basis of this initial evaluation, investigators created a decision tree that graphically represented the best predictors of in-hospital mortality in these patients. To validate the model, investigators tested the ability of the decision tree to accurately predict mortality in 32,229 subsequent hospitalizations of heart failure patients between March and July 2003.
Additional variables to identify mortality risk beyond high BUN level, low SBP, and high serum Cr level were identified but did not add sufficient discrimination to merit inclusion in the decision tree. By identifying two measures of kidney function as primary predictors of in-hospital mortality in heart failure patients, investigators raised the importance of monitoring renal function in these patients.
The ADHERE registry collects observational data from across the United States to track and study the clinical characteristics and medical management of patients hospitalized with acute heart failure. ADHERE is designed to help the medical community better understand acute heart failure, improve its management, and enhance quality of care. The ADHERE national registry is sponsored by Scios Inc and overseen by an independent scientific advisory committee of nationally recognized heart failure experts.
Scios Inc, owned by Johnson& Johnson, is a biopharmaceutical company headquartered in Fremont, CA. Scios is developing novel treatments for cardiovascular disease, inflammatory disease, and cancer by integrating expertise in protein biology with computational and medicinal chemistry to identify novel targets and rationally design small-molecule compounds and peptides.
For more information, visit http://www.adhereregistry.com or e-mail [email protected].