Measure of fitness may improve CVD risk classification
MedWire News: US researchers report that adding a single measurement of fitness to traditional cardiovascular disease (CVD)-related mortality risk prediction models may improve the accuracy of such models when used in low-risk asymptomatic individuals.
"Even after 25 years of follow-up, baseline fitness provided modest improvement in discrimination and reclassification," remark Jarrett Berry (University of Texas Southwestern Medical Center, Dallas) and team.
As reported in the journal Circulation, the team assessed the baseline fitness of 66,371 CVD-free men (n=49,307) and women (n=17,064) by measuring oxygen consumption during a treadmill exercise test. This measurement was expressed as metabolic equivalent tasks (METs; 1 MET=3.5 ml/kg/minute of oxygen consumption).
Berry and team used Cox proportional hazards models to estimate the 10- and 25-year CVD-related mortality risk of each participant, with and without the addition of the measure of fitness.
After CVD-related mortality risk was predicted, all participants were followed-up for a median period of 16 years (during which 1621 CVD deaths occurred) to determine the true accuracy of the predictive model.
The researchers found that adding fitness to a traditional CVD risk model that includes risk factors, such as age and family history of CVD, led to accurate 10- and 25-year CVD-related mortality risk reclassifications in 11.3% and 13.1% of participants, respectively.
The use of fitness as a variable improved the predictive accuracy of the risk model, such that in men, the model's C statistic score rose from 0.84 without fitness to 0.86 with fitness.
A similar C statistic score increase was seen when the model was used in women, with an increase from 0.86 before to 0.88 after adding fitness.
"Risk prediction in the general population remains a challenge because the majority of the general population is low risk," remark Berry and colleagues.
The current study findings "suggest the potential role for fitness as a novel approach for long-term risk prediction," they add.
In a related editorial, Michael Lauer (National Heart, Lung, and Blood Institute, Bethesda, Maryland, USA) commended the study. He said: "the strengths of the study include its use of standardized symptom-limited exercise testing [and] its enormous sample size."
He concluded: "When this study is combined with prior literature, there can be little doubt that exercise capacity is a meaningful predictor of CVD risk."
MedWire (www.medwire-news.md) is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2011
By Lauretta Ihonor