Inpatient Mortality Risk Scores and Postdischarge Events in Hospitalized Heart Failure Patients

Abstract

Background The Acute Decompensated Heart Failure National Registry (ADHERE) and Get with the Guidelines (GWTG) registries have developed simple heart failure (HF) in-hospital mortality risk scores. We hypothesized that HF scores predictive of in-hospital mortality would perform as well for early post-discharge mortality risk stratification. Methods and Results In this single center, community based, retrospective study of all consecutive primary HF hospitalizations (6203 hospitalizations in 3745 patients) from 2000 to 2013, the ADHERE and GWTG risk scores were calculated from admission data. There were 176 (3.0%) and 399 (6.7%), 869 (14.7%) and 1,272 (21.5%) deaths in-hospital and at 30, 90 and 180 days post-discharge respectively. The GWTG but not ADHERE risk score was well-calibrated for in-hospital mortality. Both the ADHERE (c-statistic 0.66 and 0.67, 0.64, 0.64) and GWTG (c-statistic 0.74 and 0.73, 0.71, 0.70) HF risk scores were similarly predictive of in-hospital and 30, 90 and 180-day post-discharge mortality. The ADHERE risk score identified 10% and the GWTG risk score identified 20% of hospitalizations where 180 day post-discharge mortality was 50%, a prognostic bench mark for hospice referral. In contrast, hospitalizations characterized as lowest risk by the ADHERE (57% of hospitalizations; 180 day mortality 16.2%) or GWTG score (20% of hospitalizations; 180 day mortality 8.0%) had substantially lower mortality (odds ratios high versus low risk of 5–8 (ADHERE) and 11–18 (GWTG) across time points, p<0.0001 for all). Conclusion The simple ADHERE and GWTG scores stratify hospitalized HF patients for both in-patient and early post-discharge mortality risk, allowing comprehensive risk assessment on admission.

Predictability of Inpatient Mortality of Different Comorbidities in Both Types of Acute Decompensated Heart Failure: Analysis of National Inpatient Sample

“…In addition to providing prognostic information, such models can play a crucial role in the management of certain patients, especially when outcomes are predicted to be poor and early palliative consult and/or hospice referral become more reasonable. Most of those prediction models shared similar variables, renal function, age, and blood pressure being the most studied variables [ 3 ]. However, these modules have been underutilized in the daily clinical practice due to their limited accuracy in predicting serious events [ 4 ].…”

A Predictive Model for 6-Month Mortality in Elderly Patients with Heart Failure

“…Over 60 risk scores have been previously developed to assess risks of adverse events in HF patients. 6,7) Among those risk scores, the following variables have emerged as the most consistent and strongest predictors: renal func- Values are expressed as median and interquartile range, or n (%). ACE-I indicates angiotensin-converting enzyme inhibitor; ALB, serum albumin; ARB, angiotensin-receptor blocker; BMI, body mass index; BNP, B-type natriuretic peptide; BUN, blood urea nitrogen; COPD, chronic obstructive pulmonary disease; Dd, diastolic dimension; Ds, systolic dimension; e', peak early diastolic velocity of the mitral annulus; EF, ejection fraction; eGFR, estimated glomerular filtration rate; Hb, hemoglobin; HbA1c, hemoglobin A1c; HF, heart failure; LAD, left atrial dimension; LV, left ventricular; MRA, mineralocorticoid receptor antagonist; and NYHA, New York Heart Association.…”

“…6,7) Existing risk scores have focused mainly on long-term outcomes or inhospital mortality and were often complex and therefore, uniformly underutilized. [7][8][9] A recent study reported that the Acute Decompensated Heart Failure National Registry (ADHERE) Classification and Regression Tree (CART) algorithm and the Get With The Guidelines (GWTG) HF risk score, both of which were developed to predict in-hospital mortality, could also predict early post-discharge mortality. 7,8,10) However, models for the prediction of early post-discharge mortality in elderly HF patients are not well-established.…”

Short-Term Change in Gait Speed and Clinical Outcomes in Older Patients With Acute Heart Failure

“…Finally, multivariate analysis may mitigate bias after adjustment for pre-existing prognostic factors, which have been shown to stratify early post-discharge and long-term mortality risks. 18,33 However, unmeasured and unadjusted factors, such as pre-admission physical activity and other complications, including infection, paralysis, cognitive function, and physical disabilities, and changes in baseline variables, all of which may have an effect on gait speed, leave residual bias, and the results must be replicated in future studies.…”