Share this article on:

Use of a Discriminant Analysis Model to Evaluate the Protective Effect of Exercise Against DOX Mediated Cardiotoxicity: 600June 2 10:15 AM - 10:30 AM

Jensen, Brock T.; Lin, Yen-Kuang; Hydock, David S.; Lien, Chia-Ying; Takach, Oliver P.; Schneider, Carole M. FACSM; Hayward, Reid

Medicine & Science in Sports & Exercise: May 2010 - Volume 42 - Issue 5 - p 5
doi: 10.1249/01.MSS.0000384241.12040.64
A-14 Free Communication/Slide - Cardiac: JUNE 2, 2010 9:30 AM - 11:30 AM: ROOM: 331

University of Northern Colorado and Rocky Mountain Cancer Rehabilitation Institute, Greeley, CO.


(No disclosure reported)

Doxorubicin (DOX) is a commonly prescribed chemotherapeutic drug. However, DOX is associated with a cumulative dose-dependent cardiotoxicity. It has been repeatedly shown that exercise can offer protection against a variety of myocardial injuries.

PURPOSE: The purpose of this investigation was to evaluate the effects of exercise on DOX-induced cardiac dysfunction using discriminant analysis. With this statistical procedure we used data from three previously completed studies to determine group classification and determine which cardiac function variable is the best predictor of cardiac function.

METHODS: Male Sprague-Dawley rats were randomly assigned to one of three primary experimental groups: sedentary (SED), wheel running (WR) or treadmill (TM). Following the exercise treatment, animals in each group were randomly assigned to receive a 10 mg·kg-1 i.p. bolus injection of DOX (TM+DOX, WR+DOX, SED+DOX) or saline (SED+SAL). Left ventricle function was assessed ex vivo using an isolated working heart model 5, 10, or 28 days post injection. A multivariate analysis of variance (MANOVA) was used to determine if a significant difference existed between groups for several measures of cardiac function (SP, DP, LVDP, dP/dtmax, and dP/dtmin). Stepwise discriminant analysis was performed to determine which cardiac function variable was the best predictor of cardiac function. Descriptive discriminant analysis was then used as a linear classification tool to categorize control and experimental groups.

RESULTS: A significant difference existed across the 5 cardiac function variables (F(12, 344.24) = 8.73, p<0.001). Stepwise discriminant analysis revealed that LVDP was the most indicative cardiac function variable remaining in the model (R 2 = 0.39, F = 30.34, p<0.0001). 5 days post DOX treatment, 92% of TM+DOX and 70% of WR+DOX were categorized as SED+SAL. 10 days post DOX exposure, 91% of TM+DOX and 86% of WR+DOX were categorized as SED+SAL. Furthermore, 88% of TM+DOX and 93% of WR+DOX 28 days post DOX treatment were categorized as SED+SAL.

CONCLUSIONS: LVDP contributed the most to the explanation of differences in ex vivo cardiac function between the control and experimental groups. Cardiac function of the majority of trained rats treated with DOX was categorized as SED+SAL up to 28 days post DOX treatment.

© 2010 American College of Sports Medicine