Back to Journals » Therapeutics and Clinical Risk Management » Volume 5

Comparison of linear–stochastic and nonlinear–deterministic algorithms in the analysis of 15-minute clinical ECGs to predict risk of arrhythmic death

Authors Skinner J, Meyer M, Nester BA, Geary U, Taggart P, Mangione A, Ramalanjaona G, Terregino C, Dalsey WC

Published 18 August 2009 Volume 2009:5 Pages 671—682


Review by Single anonymous peer review

Peer reviewer comments 2

James E Skinner1, Michael Meyer2, Brian A Nester3, Una Geary4, Pamela Taggart4, Antoinette Mangione4, George Ramalanjaona5, Carol Terregino6, William C Dalsey4

1Vicor Technologies, Inc., Boca Raton, FL, USA; 2Max Planck Institute for Experimental Physiology, Goettingen, Germany; 3Lehigh Valley Hospital, Allentown, PA, USA; 4Albert Einstein Medical Center, Philadelphia, PA, USA; 5North Shore University Hospital, Plainview, NY, USA; 6Cooper Medical Center, Camden, NJ, USA

Objective: Comparative algorithmic evaluation of heartbeat series in low-to-high risk cardiac patients for the prospective prediction of risk of arrhythmic death (AD).

Background: Heartbeat variation reflects cardiac autonomic function and risk of AD. Indices based on linear stochastic models are independent risk factors for AD in post-myocardial infarction (post-MI) cohorts. Indices based on nonlinear deterministic models have superior predictability in retrospective data.

Methods: Patients were enrolled (N = 397) in three emergency departments upon presenting with chest pain and were determined to be at low-to-high risk of acute MI (>7%). Brief ECGs were recorded (15 min) and R-R intervals assessed by three nonlinear algorithms (PD2i, DFA, and ApEn) and four conventional linear-stochastic measures (SDNN, MNN, 1/f-Slope, LF/HF). Out-of-hospital AD was determined by modified Hinkle–Thaler criteria.

Results: All-cause mortality at one-year follow-up was 10.3%, with 7.7% adjudicated to be AD. The sensitivity and relative risk for predicting AD was highest at all time-points for the nonlinear PD2i algorithm (p ≤ 0.001). The sensitivity at 30 days was 100%, specificity 58%, and relative risk >100 (p ≤ 0.001); sensitivity at 360 days was 95%, specificity 58%, and relative risk >11.4 (p ≤ 0.001).

Conclusions: Heartbeat analysis by the time-dependent nonlinear PD2i algorithm is comparatively the superior test.

Keywords: autonomic nervous system, regulatory systems, electrophysiology, heart rate variability, sudden cardiac death, ventricular arrhythmias, ECG, HRV, PD2i, nonlinear, nonlinear, chaos

Creative Commons License © 2009 The Author(s). This work is published and licensed by Dove Medical Press Limited. The full terms of this license are available at and incorporate the Creative Commons Attribution - Non Commercial (unported, v3.0) License. By accessing the work you hereby accept the Terms. Non-commercial uses of the work are permitted without any further permission from Dove Medical Press Limited, provided the work is properly attributed. For permission for commercial use of this work, please see paragraphs 4.2 and 5 of our Terms.