- Email: [email protected]
Risk stratification for in-hospital mortality in acutely decompensated heart failure. Classification and regression tree analysis
Heart Failure/Transplant
Effect of Candesartan on New York Heart Association Functional Class. Results of the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) Programme
Abstracts Risk Stratification for In-Hospital Mortality in Acutely Decompensated Heart Failure. Classification and Regression Tree Analysis
O’Meara E, Solomon S, McMurray J, et al. Eur Heart J 2004;25:1920 – 6.
Fonarow GC, Adams KF Jr, Abraham WT, Yancy CW, Boscardin WJ, for the ADHERE Scientific Advisory Committee, Study Group, and Investigators. JAMA 2005;293:572– 80.
Study Question: What is the effect of the angiotensin receptor blocker candesartan on New York Heart Association (NYHA) functional class in chronic heart failure (CHF)? Methods and Results: Patients in the CHARM Programme with symptomatic CHF were randomized either to placebo (n⫽3796) or candesartan (n⫽3803) and followed for a median of 38 months. The NYHA functional class was assessed at baseline, at 2 weekly intervals during dose titration and 4 monthly intervals thereafter. Patients were classified as “better,” “unchanged” or “worse” at the end of the study compared to baseline. Both a simple “last visit carried forward” (LVCF) analysis and “worst rank carried forward” (WRCF) analysis (where patients who died were allocated NYHA functional class V) were used. In the LVCF analysis, compared to placebo, more candesartan patients improved (35.4% vs. 32.5%) and fewer worsened (9.0% vs. 10.3%) in NYHA functional class (p⫽0.003). The WRCF analysis also showed a better overall change in NYHA functional class with candesartan compared to placebo. There was no heterogeneity in the response to candesartan between the CHARM component trials. Conclusions: The investigators conclude that candesartan improves NYHA functional class to a similar extent as do other proven treatments for CHF when added to these other treatments. Perspective: Given the favorable effects of candesartan on the combined outcome of death and hospitalization in the overall analysis of the CHARM study, a corresponding improvement in NYHA functional class is not surprising. However, in light of the far more substantial benefits of beta blockade (BB) on mortality and hospitalization when added to an ACE inhibitor (ACEI), a BB and an ACEI (or an ARB in patients with ACEI-induced cough) remain the two-drug regimen of choice in CHF with low EF. The big remaining question is whether an ACEI, an ARB and a BB would be better still: it was with candesartan in CHARM, but wasn’t with valsartan in VAL-HeFT. KA
Study Question: Can a user-friendly bedside tool effectively stratify mortality risk in patients hospitalized with acute decompensated heart failure (ADHF)? Methods: The investigators queried the Acute Decompensated Heart Failure National Registry (ADHERE) of patients hospitalized with a primary diagnosis of ADHF in 263 hospitals in the United States to develop a risk-stratification model for in-hospital mortality. Patient data for 39 variables were evaluated by recursive partitioning (CART). Patients had a mean age of 72.5 years, 52% were female, 58% had coronary artery disease and 46% had preserved left ventricular systolic function. The first 33,046 hospitalizations were analyzed to develop the model, and then the validity of the model was prospectively tested using data from 32,229 subsequent hospitalizations. Results: In-hospital mortality was similar in the derivation (4.2%) and validation (4.0%) cohorts. CART analysis for 39 variables indicated that the best single predictor for mortality was high admission levels of blood urea nitrogen (ⱖ43 mg/dL) followed by low admission systolic blood pressure (SBP) (⬍115 mm Hg) and then by high levels of serum creatinine (ⱖ2.75 mg/dL [243.1 mol/L]). A simple risk tree identified five patient groups with mortality ranging from 2.1% to 21.9%. The odds ratio for mortality between patients identified as high and low risk was 12.9 (95% CI, 10.4 –15.9), and similar results were seen when this risk stratification was applied prospectively to the validation cohort. Conclusions: The researchers conclude that ADHF patients at low, intermediate and high risk for in-hospital mortality can be easily identified using vital sign and laboratory data obtained on hospital admission. The ADHERE risk tree provides clinicians with a validated, practical bedside tool for mortality risk stratification. Perspective: The research team provided two useful tools to clinicians managing patients with ADHF. The diagnostic accuracy of the 3-variable CART model was significantly lower than that of a 4-variable logistic regression model including BUN, SBP, serum creatinine and age (66.8% vs. 75.7% in the validation cohort) that they also developed. The researchers prefer “the ease of use” of the CART model, but now that we are all carrying personal digital assistants (PDAs) in our coat pocket, the more accurate model result is literally at one’s fingertips. Either way, the next step is to use this information to test strategies that more effectively target the intensity of hospital care to patients at highest risk. KA
Statin Use and Survival Outcomes in Elderly Patients With Heart Failure Ray JG, Gong Y, Sykora K, Tu JV. Arch Intern Med 2005;165:62–7. Study Question: What are the effects of statins on death, nonfatal myocardial infarction (MI) and nonfatal stroke following hospitalization for newly diagnosed heart failure?
ACC CURRENT JOURNAL REVIEW April 2005
34
Effect of Candesartan on New York Heart Association Functional Class. Results of the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) Programme
Abstracts Risk Stratification for In-Hospital Mortality in Acutely Decompensated Heart Failure. Classification and Regression Tree Analysis
O’Meara E, Solomon S, McMurray J, et al. Eur Heart J 2004;25:1920 – 6.
Fonarow GC, Adams KF Jr, Abraham WT, Yancy CW, Boscardin WJ, for the ADHERE Scientific Advisory Committee, Study Group, and Investigators. JAMA 2005;293:572– 80.
Study Question: What is the effect of the angiotensin receptor blocker candesartan on New York Heart Association (NYHA) functional class in chronic heart failure (CHF)? Methods and Results: Patients in the CHARM Programme with symptomatic CHF were randomized either to placebo (n⫽3796) or candesartan (n⫽3803) and followed for a median of 38 months. The NYHA functional class was assessed at baseline, at 2 weekly intervals during dose titration and 4 monthly intervals thereafter. Patients were classified as “better,” “unchanged” or “worse” at the end of the study compared to baseline. Both a simple “last visit carried forward” (LVCF) analysis and “worst rank carried forward” (WRCF) analysis (where patients who died were allocated NYHA functional class V) were used. In the LVCF analysis, compared to placebo, more candesartan patients improved (35.4% vs. 32.5%) and fewer worsened (9.0% vs. 10.3%) in NYHA functional class (p⫽0.003). The WRCF analysis also showed a better overall change in NYHA functional class with candesartan compared to placebo. There was no heterogeneity in the response to candesartan between the CHARM component trials. Conclusions: The investigators conclude that candesartan improves NYHA functional class to a similar extent as do other proven treatments for CHF when added to these other treatments. Perspective: Given the favorable effects of candesartan on the combined outcome of death and hospitalization in the overall analysis of the CHARM study, a corresponding improvement in NYHA functional class is not surprising. However, in light of the far more substantial benefits of beta blockade (BB) on mortality and hospitalization when added to an ACE inhibitor (ACEI), a BB and an ACEI (or an ARB in patients with ACEI-induced cough) remain the two-drug regimen of choice in CHF with low EF. The big remaining question is whether an ACEI, an ARB and a BB would be better still: it was with candesartan in CHARM, but wasn’t with valsartan in VAL-HeFT. KA
Study Question: Can a user-friendly bedside tool effectively stratify mortality risk in patients hospitalized with acute decompensated heart failure (ADHF)? Methods: The investigators queried the Acute Decompensated Heart Failure National Registry (ADHERE) of patients hospitalized with a primary diagnosis of ADHF in 263 hospitals in the United States to develop a risk-stratification model for in-hospital mortality. Patient data for 39 variables were evaluated by recursive partitioning (CART). Patients had a mean age of 72.5 years, 52% were female, 58% had coronary artery disease and 46% had preserved left ventricular systolic function. The first 33,046 hospitalizations were analyzed to develop the model, and then the validity of the model was prospectively tested using data from 32,229 subsequent hospitalizations. Results: In-hospital mortality was similar in the derivation (4.2%) and validation (4.0%) cohorts. CART analysis for 39 variables indicated that the best single predictor for mortality was high admission levels of blood urea nitrogen (ⱖ43 mg/dL) followed by low admission systolic blood pressure (SBP) (⬍115 mm Hg) and then by high levels of serum creatinine (ⱖ2.75 mg/dL [243.1 mol/L]). A simple risk tree identified five patient groups with mortality ranging from 2.1% to 21.9%. The odds ratio for mortality between patients identified as high and low risk was 12.9 (95% CI, 10.4 –15.9), and similar results were seen when this risk stratification was applied prospectively to the validation cohort. Conclusions: The researchers conclude that ADHF patients at low, intermediate and high risk for in-hospital mortality can be easily identified using vital sign and laboratory data obtained on hospital admission. The ADHERE risk tree provides clinicians with a validated, practical bedside tool for mortality risk stratification. Perspective: The research team provided two useful tools to clinicians managing patients with ADHF. The diagnostic accuracy of the 3-variable CART model was significantly lower than that of a 4-variable logistic regression model including BUN, SBP, serum creatinine and age (66.8% vs. 75.7% in the validation cohort) that they also developed. The researchers prefer “the ease of use” of the CART model, but now that we are all carrying personal digital assistants (PDAs) in our coat pocket, the more accurate model result is literally at one’s fingertips. Either way, the next step is to use this information to test strategies that more effectively target the intensity of hospital care to patients at highest risk. KA
Statin Use and Survival Outcomes in Elderly Patients With Heart Failure Ray JG, Gong Y, Sykora K, Tu JV. Arch Intern Med 2005;165:62–7. Study Question: What are the effects of statins on death, nonfatal myocardial infarction (MI) and nonfatal stroke following hospitalization for newly diagnosed heart failure?
ACC CURRENT JOURNAL REVIEW April 2005
34