Article Text

PDF
Acute coronary syndromes
Simple point-of-care risk stratification in acute coronary syndromes: the AMIS model
  1. D J Kurz1,
  2. A Bernstein2,
  3. K Hunt2,
  4. D Radovanovic3,
  5. P Erne4,
  6. Z Siudak5,
  7. O Bertel6
  1. 1
    Department of Internal Medicine, Division of Cardiology, Triemli Hospital, Zurich, Switzerland
  2. 2
    Dynamic and Distributed Information Systems Group, Department of Informatics, University of Zurich, Switzerland
  3. 3
    AMIS-PLUS Data Centre, Institute of Social and Preventive Medicine, University of Zurich, Switzerland
  4. 4
    Department of Internal Medicine, Division of Cardiology, Kantonsspital Luzern, Switzerland
  5. 5
    Department of Interventional Cardiology, University Hospital, Krakow, Poland
  6. 6
    Cardio-Vascular Center Zurich, Klinik im Park, Zurich, Switzerland
  1. Dr David J Kurz, Division of Cardiology, Triemli Hospital, Birmensdorferstrasse 497, CH-8063 Zurich, Switzerland; david.kurz{at}triemli.stzh.ch

Abstract

Background: Early risk stratification is important in the management of patients with acute coronary syndromes (ACS).

Objective: To develop a rapidly available risk stratification tool for use in all ACS.

Design and methods: Application of modern data mining and machine learning algorithms to a derivation cohort of 7520 ACS patients included in the AMIS (Acute Myocardial Infarction in Switzerland)-Plus registry between 2001 and 2005; prospective model testing in two validation cohorts.

Results: The most accurate prediction of in-hospital mortality was achieved with the “Averaged One-Dependence Estimators” (AODE) algorithm, with input of seven variables available at first patient contact: age, Killip class, systolic blood pressure, heart rate, pre-hospital cardiopulmonary resuscitation, history of heart failure, history of cerebrovascular disease. The c-statistic for the derivation cohort (0.875) was essentially maintained in important subgroups, and calibration over five risk categories, ranging from <1% to >30% predicted mortality, was accurate. Results were validated prospectively against an independent AMIS-Plus cohort (n = 2854, c-statistic 0.868) and the Krakow-Region ACS Registry (n = 2635, c-statistic 0.842). The AMIS model significantly outperformed established “point-of-care” risk-prediction tools in both validation cohorts. In comparison to a logistic regression-based model, the AODE-based model proved to be more robust when tested on the Krakow validation cohort (c-statistic 0.842 vs 0.746). Accuracy of the AMIS model prediction was maintained at 12-month follow-up in an independent cohort (n = 1972, c-statistic 0.877).

Conclusions: The AMIS model is a reproducibly accurate point-of-care risk stratification tool for the complete range of ACS, based on variables available at first patient contact.

Statistics from Altmetric.com

Footnotes

  • Contributors and guarantor: DJKurz, AB and OB devised the study. KH performed the model development and statistical analyses together with AB and DJK. DR, PE and ZS participated in collection and maintenance of the respective patient registries. DJK wrote the first draft of the manuscript. All authors contributed to the final version of the manuscript. DJK takes overall responsibility for the manuscript.

  • AB and KH contributed equally to this work.

  • Funding: DJK receives financial support from the Swiss Heart Foundation.

  • Competing interests: None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.