External Validation of the MEESSI Acute Heart Failure Risk Score: A Cohort Study

TY - JOUR

T1 - External Validation of the MEESSI Acute Heart Failure Risk Score: A Cohort Study

AU - Wussler, Desiree

AU - Kozhuharov, Nikola

AU - Sabti, Zaid

AU - Walter, Joan

AU - Strebel, Ivo

AU - Scholl, Letizia

AU - Miró, Oscar

AU - Rossello, Xavier

AU - Martín-Sánchez, Francisco Javier

AU - Pocock, Stuart J

AU - Nowak, Albina

AU - Badertscher, Patrick

AU - Twerenbold, Raphael

AU - Wildi, Karin

AU - Puelacher, Christian

AU - du Fay de Lavallaz, Jeanne

AU - Shrestha, Samyut

AU - Strauch, Olivia

AU - Flores, Dayana

AU - Nestelberger, Thomas

AU - Boeddinghaus, Jasper

AU - Schumacher, Carmela

AU - Goudev, Assen

AU - Pfister, Otmar

AU - Breidthardt, Tobias

AU - Mueller, Christian

PY - 2019/2/19

Y1 - 2019/2/19

N2 - Background: The MEESSI-AHF (Multiple Estimation of risk based on the Emergency department Spanish Score In patients with AHF) score was developed to predict 30-day mortality in patients presenting with acute heart failure (AHF) to emergency departments (EDs) in Spain. Whether it performs well in other countries is unknown.Objective: To externally validate the MEESSI-AHF score in another country.Design: Prospective cohort study. (ClinicalTrials.gov: NCT01831115).Setting: Multicenter recruitment of dyspneic patients presenting to the ED.Participants: The external validation cohort included 1572 patients with AHF.Measurements: Calculation of the MEESSI-AHF score using an established model containing 12 independent risk factors.Results: Among 1572 patients with adjudicated AHF, 1247 had complete data that allowed calculation of the MEESSI-AHF score. Of these, 102 (8.2%) died within 30 days. The score predicted 30-day mortality with excellent discrimination (c-statistic, 0.80). Assessment of cumulative mortality showed a steep gradient in 30-day mortality over 6 predefined risk groups (0 patients in the lowest-risk group vs. 35 [28.5%] in the highest-risk group). Risk was overestimated in the high-risk groups, resulting in a Hosmer-Lemeshow P value of 0.022. However, after adjustment of the intercept, the model showed good concordance between predicted risks and observed outcomes (P = 0.23). Findings were confirmed in sensitivity analyses that used multiple imputation for missing values in the overall cohort of 1572 patients.Limitations: External validation was done using a reduced model. Findings are specific to patients with AHF who present to the ED and are clinically stable enough to provide informed consent. Performance in patients with terminal kidney failure who are receiving long-term dialysis cannot be commented on.Conclusion: External validation of the MEESSI-AHF risk score showed excellent discrimination. Recalibration may be needed when the score is introduced to new populations.Primary Funding Source: The European Union, the Swiss National Science Foundation, the Swiss Heart Foundation, the Cardiovascular Research Foundation Basel, the University of Basel, and University Hospital Basel.

AB - Background: The MEESSI-AHF (Multiple Estimation of risk based on the Emergency department Spanish Score In patients with AHF) score was developed to predict 30-day mortality in patients presenting with acute heart failure (AHF) to emergency departments (EDs) in Spain. Whether it performs well in other countries is unknown.Objective: To externally validate the MEESSI-AHF score in another country.Design: Prospective cohort study. (ClinicalTrials.gov: NCT01831115).Setting: Multicenter recruitment of dyspneic patients presenting to the ED.Participants: The external validation cohort included 1572 patients with AHF.Measurements: Calculation of the MEESSI-AHF score using an established model containing 12 independent risk factors.Results: Among 1572 patients with adjudicated AHF, 1247 had complete data that allowed calculation of the MEESSI-AHF score. Of these, 102 (8.2%) died within 30 days. The score predicted 30-day mortality with excellent discrimination (c-statistic, 0.80). Assessment of cumulative mortality showed a steep gradient in 30-day mortality over 6 predefined risk groups (0 patients in the lowest-risk group vs. 35 [28.5%] in the highest-risk group). Risk was overestimated in the high-risk groups, resulting in a Hosmer-Lemeshow P value of 0.022. However, after adjustment of the intercept, the model showed good concordance between predicted risks and observed outcomes (P = 0.23). Findings were confirmed in sensitivity analyses that used multiple imputation for missing values in the overall cohort of 1572 patients.Limitations: External validation was done using a reduced model. Findings are specific to patients with AHF who present to the ED and are clinically stable enough to provide informed consent. Performance in patients with terminal kidney failure who are receiving long-term dialysis cannot be commented on.Conclusion: External validation of the MEESSI-AHF risk score showed excellent discrimination. Recalibration may be needed when the score is introduced to new populations.Primary Funding Source: The European Union, the Swiss National Science Foundation, the Swiss Heart Foundation, the Cardiovascular Research Foundation Basel, the University of Basel, and University Hospital Basel.

KW - Aged

KW - Aged, 80 and over

KW - Emergency Service, Hospital

KW - Female

KW - Heart Failure/diagnosis

KW - Humans

KW - Logistic Models

KW - Male

KW - Prospective Studies

KW - Reproducibility of Results

KW - Risk Assessment/methods

KW - Spain/epidemiology

KW - Switzerland/epidemiology

U2 - 10.7326/M18-1967

DO - 10.7326/M18-1967

M3 - SCORING: Journal article

C2 - 30690646

VL - 170

SP - 248

EP - 256

JO - ANN INTERN MED

JF - ANN INTERN MED

SN - 0003-4819

IS - 4

ER -