Comparison of the HEART and TIMI risk scores for suspected acute coronary syndrome in the emergency department

Benjamin Sun, Amber Laurie, Rongwei (Rochelle) Fu, Maros Ferencik, Michael Shapiro, Christopher J. Lindsell, Deborah Diercks, James W. Hoekstra, Judd E. Hollander, J. Douglas Kirk, W. Frank Peacock, Venkataraman Anantharaman, Charles V. Pollack

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Objectives: The emergency department evaluation for suspected acute coronary syndrome (AC S) is common, costly, and challenging. Risk scores may help standardize clinical care and screening for research studies. The Thrombolysis in Myocardial Infarction (TIMI) and HEART are two commonly cited risk scores. We tested the null hypothesis that the TIMI and HEART risk scores have equivalent test characteristics. Methods: We analyzed data from the Internet Tracking Registry of Acute Coronary Syndromes (i∗TrAC S) from 9 EDs on patients with suspected AC S, 1999-2001. We excluded patients with an emergency department diagnosis consistent with AC S, or without sufficient data to calculate TIMI and HEART scores. The primary outcome was 30-day major adverse cardiovascular events, including all-cause death, acute myocardial infarction, and urgent revascularization. We describe test characteristics of the TIMI and HEART risk scores. Results: The study cohort included 8255 patients with 508 (6.2%) 30-day major adverse cardiovascular events. Receiver operating curve and reclassification analyses favored HEART [c statistic: 0.753, 95% confidence interval (CI): 0.733-0.773; continuous net reclassification improvement: 0.608, 95% CI: 0.527-0.689] over TIMI (c statistic: 0.678, 95% CI: 0.655-0.702). A HEART score 0-3 [negative predictive value (NPV) 0.982, 95% CI: 0.978-0.986; positive predictive value (PPV) 0.103, 95% CI: 0.094-0.113; likelihood ratio (LR) positive 1.76; LR negative 0.28] demonstrates similar or superior NPV/PPV/LR compared with TIMI = 0 (NPV 0.978, 95% CI: 0.971-0.983; PPV 0.077, 95% CI: 0.071-0.084; LR positive 1.28; LR negative 0.35) and TIMI = 0-1 (NPV 0.963, 95% CI: 0.958-0.968; PPV 0.102, 95% CI: 0.092-0.113; LR positive 1.73; LR negative 0.58). Conclusions: The HEART score has better discrimination than TIMI and outperforms TIMI within previously published "low-risk" categories.

Original languageEnglish (US)
Pages (from-to)1-5
Number of pages5
JournalCritical Pathways in Cardiology
Volume15
Issue number1
DOIs
StatePublished - 2016

Fingerprint

Acute Coronary Syndrome
Hospital Emergency Service
Myocardial Infarction
Confidence Intervals
Internet
Registries
Cause of Death
Cohort Studies

Keywords

  • HEART score
  • Risk score
  • TIMI

ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine

Cite this

Comparison of the HEART and TIMI risk scores for suspected acute coronary syndrome in the emergency department. / Sun, Benjamin; Laurie, Amber; Fu, Rongwei (Rochelle); Ferencik, Maros; Shapiro, Michael; Lindsell, Christopher J.; Diercks, Deborah; Hoekstra, James W.; Hollander, Judd E.; Kirk, J. Douglas; Peacock, W. Frank; Anantharaman, Venkataraman; Pollack, Charles V.

In: Critical Pathways in Cardiology, Vol. 15, No. 1, 2016, p. 1-5.

Research output: Contribution to journalArticle

Sun, B, Laurie, A, Fu, RR, Ferencik, M, Shapiro, M, Lindsell, CJ, Diercks, D, Hoekstra, JW, Hollander, JE, Kirk, JD, Peacock, WF, Anantharaman, V & Pollack, CV 2016, 'Comparison of the HEART and TIMI risk scores for suspected acute coronary syndrome in the emergency department', Critical Pathways in Cardiology, vol. 15, no. 1, pp. 1-5. https://doi.org/10.1097/HPC.0000000000000066
Sun, Benjamin ; Laurie, Amber ; Fu, Rongwei (Rochelle) ; Ferencik, Maros ; Shapiro, Michael ; Lindsell, Christopher J. ; Diercks, Deborah ; Hoekstra, James W. ; Hollander, Judd E. ; Kirk, J. Douglas ; Peacock, W. Frank ; Anantharaman, Venkataraman ; Pollack, Charles V. / Comparison of the HEART and TIMI risk scores for suspected acute coronary syndrome in the emergency department. In: Critical Pathways in Cardiology. 2016 ; Vol. 15, No. 1. pp. 1-5.
@article{c25b95b20d7e4e37a26f4cc150cb31dc,
title = "Comparison of the HEART and TIMI risk scores for suspected acute coronary syndrome in the emergency department",
abstract = "Objectives: The emergency department evaluation for suspected acute coronary syndrome (AC S) is common, costly, and challenging. Risk scores may help standardize clinical care and screening for research studies. The Thrombolysis in Myocardial Infarction (TIMI) and HEART are two commonly cited risk scores. We tested the null hypothesis that the TIMI and HEART risk scores have equivalent test characteristics. Methods: We analyzed data from the Internet Tracking Registry of Acute Coronary Syndromes (i∗TrAC S) from 9 EDs on patients with suspected AC S, 1999-2001. We excluded patients with an emergency department diagnosis consistent with AC S, or without sufficient data to calculate TIMI and HEART scores. The primary outcome was 30-day major adverse cardiovascular events, including all-cause death, acute myocardial infarction, and urgent revascularization. We describe test characteristics of the TIMI and HEART risk scores. Results: The study cohort included 8255 patients with 508 (6.2{\%}) 30-day major adverse cardiovascular events. Receiver operating curve and reclassification analyses favored HEART [c statistic: 0.753, 95{\%} confidence interval (CI): 0.733-0.773; continuous net reclassification improvement: 0.608, 95{\%} CI: 0.527-0.689] over TIMI (c statistic: 0.678, 95{\%} CI: 0.655-0.702). A HEART score 0-3 [negative predictive value (NPV) 0.982, 95{\%} CI: 0.978-0.986; positive predictive value (PPV) 0.103, 95{\%} CI: 0.094-0.113; likelihood ratio (LR) positive 1.76; LR negative 0.28] demonstrates similar or superior NPV/PPV/LR compared with TIMI = 0 (NPV 0.978, 95{\%} CI: 0.971-0.983; PPV 0.077, 95{\%} CI: 0.071-0.084; LR positive 1.28; LR negative 0.35) and TIMI = 0-1 (NPV 0.963, 95{\%} CI: 0.958-0.968; PPV 0.102, 95{\%} CI: 0.092-0.113; LR positive 1.73; LR negative 0.58). Conclusions: The HEART score has better discrimination than TIMI and outperforms TIMI within previously published {"}low-risk{"} categories.",
keywords = "HEART score, Risk score, TIMI",
author = "Benjamin Sun and Amber Laurie and Fu, {Rongwei (Rochelle)} and Maros Ferencik and Michael Shapiro and Lindsell, {Christopher J.} and Deborah Diercks and Hoekstra, {James W.} and Hollander, {Judd E.} and Kirk, {J. Douglas} and Peacock, {W. Frank} and Venkataraman Anantharaman and Pollack, {Charles V.}",
year = "2016",
doi = "10.1097/HPC.0000000000000066",
language = "English (US)",
volume = "15",
pages = "1--5",
journal = "Critical Pathways in Cardiology",
issn = "1535-282X",
publisher = "Lippincott Williams and Wilkins",
number = "1",

}

TY - JOUR

T1 - Comparison of the HEART and TIMI risk scores for suspected acute coronary syndrome in the emergency department

AU - Sun, Benjamin

AU - Laurie, Amber

AU - Fu, Rongwei (Rochelle)

AU - Ferencik, Maros

AU - Shapiro, Michael

AU - Lindsell, Christopher J.

AU - Diercks, Deborah

AU - Hoekstra, James W.

AU - Hollander, Judd E.

AU - Kirk, J. Douglas

AU - Peacock, W. Frank

AU - Anantharaman, Venkataraman

AU - Pollack, Charles V.

PY - 2016

Y1 - 2016

N2 - Objectives: The emergency department evaluation for suspected acute coronary syndrome (AC S) is common, costly, and challenging. Risk scores may help standardize clinical care and screening for research studies. The Thrombolysis in Myocardial Infarction (TIMI) and HEART are two commonly cited risk scores. We tested the null hypothesis that the TIMI and HEART risk scores have equivalent test characteristics. Methods: We analyzed data from the Internet Tracking Registry of Acute Coronary Syndromes (i∗TrAC S) from 9 EDs on patients with suspected AC S, 1999-2001. We excluded patients with an emergency department diagnosis consistent with AC S, or without sufficient data to calculate TIMI and HEART scores. The primary outcome was 30-day major adverse cardiovascular events, including all-cause death, acute myocardial infarction, and urgent revascularization. We describe test characteristics of the TIMI and HEART risk scores. Results: The study cohort included 8255 patients with 508 (6.2%) 30-day major adverse cardiovascular events. Receiver operating curve and reclassification analyses favored HEART [c statistic: 0.753, 95% confidence interval (CI): 0.733-0.773; continuous net reclassification improvement: 0.608, 95% CI: 0.527-0.689] over TIMI (c statistic: 0.678, 95% CI: 0.655-0.702). A HEART score 0-3 [negative predictive value (NPV) 0.982, 95% CI: 0.978-0.986; positive predictive value (PPV) 0.103, 95% CI: 0.094-0.113; likelihood ratio (LR) positive 1.76; LR negative 0.28] demonstrates similar or superior NPV/PPV/LR compared with TIMI = 0 (NPV 0.978, 95% CI: 0.971-0.983; PPV 0.077, 95% CI: 0.071-0.084; LR positive 1.28; LR negative 0.35) and TIMI = 0-1 (NPV 0.963, 95% CI: 0.958-0.968; PPV 0.102, 95% CI: 0.092-0.113; LR positive 1.73; LR negative 0.58). Conclusions: The HEART score has better discrimination than TIMI and outperforms TIMI within previously published "low-risk" categories.

AB - Objectives: The emergency department evaluation for suspected acute coronary syndrome (AC S) is common, costly, and challenging. Risk scores may help standardize clinical care and screening for research studies. The Thrombolysis in Myocardial Infarction (TIMI) and HEART are two commonly cited risk scores. We tested the null hypothesis that the TIMI and HEART risk scores have equivalent test characteristics. Methods: We analyzed data from the Internet Tracking Registry of Acute Coronary Syndromes (i∗TrAC S) from 9 EDs on patients with suspected AC S, 1999-2001. We excluded patients with an emergency department diagnosis consistent with AC S, or without sufficient data to calculate TIMI and HEART scores. The primary outcome was 30-day major adverse cardiovascular events, including all-cause death, acute myocardial infarction, and urgent revascularization. We describe test characteristics of the TIMI and HEART risk scores. Results: The study cohort included 8255 patients with 508 (6.2%) 30-day major adverse cardiovascular events. Receiver operating curve and reclassification analyses favored HEART [c statistic: 0.753, 95% confidence interval (CI): 0.733-0.773; continuous net reclassification improvement: 0.608, 95% CI: 0.527-0.689] over TIMI (c statistic: 0.678, 95% CI: 0.655-0.702). A HEART score 0-3 [negative predictive value (NPV) 0.982, 95% CI: 0.978-0.986; positive predictive value (PPV) 0.103, 95% CI: 0.094-0.113; likelihood ratio (LR) positive 1.76; LR negative 0.28] demonstrates similar or superior NPV/PPV/LR compared with TIMI = 0 (NPV 0.978, 95% CI: 0.971-0.983; PPV 0.077, 95% CI: 0.071-0.084; LR positive 1.28; LR negative 0.35) and TIMI = 0-1 (NPV 0.963, 95% CI: 0.958-0.968; PPV 0.102, 95% CI: 0.092-0.113; LR positive 1.73; LR negative 0.58). Conclusions: The HEART score has better discrimination than TIMI and outperforms TIMI within previously published "low-risk" categories.

KW - HEART score

KW - Risk score

KW - TIMI

UR - http://www.scopus.com/inward/record.url?scp=84959369611&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959369611&partnerID=8YFLogxK

U2 - 10.1097/HPC.0000000000000066

DO - 10.1097/HPC.0000000000000066

M3 - Article

C2 - 26881812

AN - SCOPUS:84959369611

VL - 15

SP - 1

EP - 5

JO - Critical Pathways in Cardiology

JF - Critical Pathways in Cardiology

SN - 1535-282X

IS - 1

ER -