on behalf of the INTRINSIC RV Study InvestigatorsBackground-Elevated heart rate (HR) is associated with adverse cardiovascular events and total mortality in the general population and in individuals with heart disease. Our hypothesis was that mean HR predicts total mortality and heart failure hospitalization in patients undergoing implantable cardioverter-defibrillator (ICD) implantation. Methods and Results-The Inhibition of Unnecessary RV Pacing With AV Search Hysteresis in ICDs (INTRINSIC RV) trial included 1530 patients undergoing ICD implantation. After implantation of a dual-chamber ICD, patients were followed for a mean of 10.4 months. The mean HR for 1436 patients over the follow-up period was determined from device histograms. Patients were grouped into strata by mean HR, and the relationship between the primary end point and mean HR was analyzed with Mantel-Haenszel ordinal 2 tests. Higher intrinsic (unpaced) HR was associated with greater risk of achieving the primary end point of death or heart failure hospitalization (PϽ0.001). Of patients with a mean HR Ͻ75 bpm, 5.8% died or were hospitalized for heart failure, whereas 20.9% with a mean HR Ͼ90 bpm achieved the same end point, a 3.6-fold difference (PϽ0.0001). In a multivariate model with the use of Cox regression, HR was a significant predictor with a hazard ratio of 1.34 (Pϭ0.0001; 95% confidence interval, 1.19 to 1.50), as were age, New York Heart Association functional class, and percent right ventricular pacing, but it was independent of gender and ␤-blocker dosing. When considered as continuous or discrete variables grouped by 5-bpm increments, HR remained a significant predictor. Conclusions-In this ICD population, the mean intrinsic HR was strongly associated with outcomes. Clinical Trial Registration-http://www.clinicaltrials.gov.

No abstract

On the basis of our findings, 1 Drs Fazio and Carlomagno and Dr Lin and colleagues considered that a J-shaped relationship may exist between heart rate and outcomes. Like the U-shaped relationship we saw with an increasing percentage of right ventricular pacing and outcomes, 2 we suspected that a J-or U-shaped relationship may exist between heart rate and outcomes. However, we could not discern with certainty that the slowest heart rates were associated with worse outcomes even though a trend in that direction appeared.Dr Lin and colleagues also suspected that a J-shaped relationship may be present in those with mild heart failure in lieu of a linear heart rate/outcome relationship that may occur in those with advanced heart failure. Dividing data into these small subgroups did not provide definitive answers. We agree, as Dr Lin and colleagues cautioned, that there is need to interpret the data carefully. Nevertheless, this issue remains worthy of further study because some, 3 but not all, 4,5 data regarding heart rate support a J-shaped relationship, albeit in different populations and for different outcomes.Dr Lin and colleagues wondered if slow heart rates are related to increasing amounts of right ventricular pacing. Heart rate was a predictor of outcomes independently of right ventricular pacing. Although right ventricular pacing predicted outcomes, heart rate was the stronger predictor.The pacemaker programming we used would not necessarily promote high amounts of right ventricular pacing at slow rates. Indeed, some right ventricular pacing in the Inhibition of Unnecessary Right Ventricular Pacing With Atrioventricular Search Hysteresis in Implantable Cardioverter Defibrillators (INTRINSIC RV) study was associated with favorable outcomes. 2 Despite this, the relationship between right ventricular pacing and outcomes in the context of slow heart rates requires further study.We agree with Drs Fazio and Carlomagno that important relationships between lifestyle, physical activity, and heart rate may exist and that they can be difficult to discern in assessments of heart rate over prolonged periods. Nevertheless, mean heart rate effectively predicted outcomes. Transtelephonic implantable cardioverterdefibrillator monitoring may better define relationships between level of exercise and its correlation with heart rate during exercise and at rest. Although our study did not permit investigation of resting heart rate and its change under exercise, understanding the relationships between heart rate and specific daily activities or lifestyles and even circadian variations could provide new insight into how heart rate is associated with outcomes.