Author + information
- Paolo Palatini, MD∗ ( and )
- Stevo Julius, MD
- ↵∗Dipartimento di Medicina, University of Padova, via Giustiniani 2, 35128 Padua, Italy
Opdahl et al. (1) recently reported some interesting findings from the MESA (Multi-Ethnic Study of Atherosclerosis) trial showing that elevated resting heart rate (HR) was associated with increased risk for incident heart failure (HF) during a 7-year follow-up. Subjects in the top HR quartile had a more than 3-fold greater adjusted relative risk for incident HF than subjects in the bottom HR quartile. The authors are to be commended for their interesting analysis, but when commenting on their findings, they did not take into account the recent results of the VALUE (Valsartan Antihypertensive Long-term Use Evaluation) study (2) which made a more extensive analysis of the HR–HF association in a larger sample (n = 15,245 subjects). Our analysis showed that both the baseline HR and the HR values recorded after 1 year (in-trial HR) were strong predictors of incident HF (2). In a multivariate Cox model, the hazard ratios for the highest HR quintile were 2.48 (95% confidence interval [CI]: 1.91 to 3.21) for baseline HR and 1.93 (95% CI: 1.45 to 2.55) for in-trial HR, compared to the bottom HR quintile.
In their article, the MESA investigators claimed that the association between elevated HR and incident HF may be explained by mechanisms other than an underlying coronary artery disease because in their study HR was not a predictor of coronary events (1). However, this negative finding was probably due to their study being underpowered to detect this relationship because in the VALUE study we found a significant independent association between HR and incident myocardial infarction (2). The hazard ratios for the highest HR quintile were 1.48 (95% CI: 1.15 to 1.90) for baseline HR and 1.38 (95% CI: 1.03 to 1.85) for in-trial HR.
As the MESA investigators stated in their article (1), HR might not be a causative factor and the connection between HR and HF could be explained by a latent left ventricular dysfunction accompanied by a compensatory increase in resting HR (3). To exclude this possibility, in the VALUE study, we evaluated the effect of baseline HR on incident HF occurring separately in each year of the trial (2). If tachycardia were a marker of subclinical congestive HF, a stronger association would be expected in the early phases of the follow-up with an attenuation in later years. However, the association between high HR and incident HF was present throughout the trial and was significant even at the fifth year.
To see whether this finding obtained in a hypertensive sample also holds true within a general population, we encourage the MESA investigators to analyze the predictive power of baseline HR by using the same statistical procedure.
- American College of Cardiology Foundation