By Michael H. Crawford, MD, Editor

SYNOPSIS: A retrospective observational study of patients with moderate to severe aortic regurgitation showed the independent predictive value of diastolic blood pressure and resting heart rate for all-cause mortality under medical management.

SOURCES: Yang LT, Pellikka PA, Enriquez-Sarano M, et al. Diastolic blood pressure and heart rate are independently associated with mortality in chronic aortic regurgitation. J Am Coll Cardiol 2020;75:29-39.

Chambers J. Aortic regurgitation: The value of clinical signs. J Am Coll Cardiol 2020;75:40-41.

Current guidelines caution against using drugs that lower heart rate or diastolic blood pressure (DBP) in patients with chronic aortic regurgitation (AR). However, these cautions are really hypotheses since there are little data to support them.

Investigators from the Mayo Clinic retrospectively identified 820 patients with chronic moderate to severe AR without significant concomitant valve or myocardial disease from 2006 to 2017. Routine follow-up echocardiograms included measurement of heart rate and BP. The severity of AR was determined by a combination of parameters approach on echocardiography. The primary endpoint was all-cause mortality under medical management. Aortic valve surgery was a secondary endpoint. For their multivariate logistic regression analysis, the hazard ratios (HR) were plotted using a resting heart rate (RHR) of 60 beats per minute (bpm) and a DBP of 70 mmHg as the reference points. The mean age of patients was 59 years, and 82% were men.

The mechanisms of AR were annular or aortic root dilation in 26%, cusp prolapse in 13%, and cusp restriction in 10%. In 43% of patients, there was a mixed mechanism, and it was indeterminate in 7%. Baseline characteristics showed those with lower DBP and higher heart rates often were more symptomatic and had more severe AR. During the mean follow-up of 5.5 years, 49% underwent aortic valve surgery (AVS) and 19% died, about one-third of the latter after AVS. Survival under medical management was 94% at one year, 79% at five years, and 63% at 10 years.

A multivariate model adjusting for baseline characteristics and the presence of guideline triggers for AVS demonstrated the independent predictive value of DBP and RHR for all-cause death under medical management: DBP HR, 0.79; 95% confidence interval (CI), 0.66-0.94 per 10 mmHg increase; P < 0.01, and RHR HR, 1.23; 95% CI, 1.03-1.45 per 10 beat increase; P = 0.01. These results persisted after further adjustment for hypertension, medications, and timing of AVS. Systolic BP was not predictive, but pulse pressure per 10 mmHg increase was weakly predictive: HR, 1.1; 95% CI, 1.01-1.24; P = 0.02.

Further analysis of DBP showed a J-shaped curve with mortality starting to increase at < 70 mmHg and peaking at 55 mmHg. For RHR, there was a linear relationship to mortality, increasing after RHR was > 60 bpm. The authors concluded that in patients with chronic moderate to severe AR, DBP and RHR are robustly associated with all-cause death independent of comorbidities, guideline-based surgical triggers, hypertension, and medications. These measures should be considered in decision-making for AVS.


In patients without heart valve disease, a rapid pulse has been associated with mortality. This finding in AR patients is unsurprising, but the lack of a J-shaped curve perhaps is. In this observational study, there was no discernable adverse effects of pulses down to 40 bpm, although the number of patients with RHR < 50 bpm was small. The authors advanced the theory that the large stroke volume associated with chronic stable AR activates baroreceptors, resulting in increased vagal tone. Investigators observed about 40% of patients who recorded heart rates > 60 bpm. As the left ventricular begins to fail and stroke volume drops, sympathetic tone increases, leading to an acceleration in RHR. The concept that a slower heart rate increases diastolic filling and makes AR worse may not be correct. Thus, avoiding drugs to slow the heart rate may not be necessary unless they also decrease the contractile state of the left ventricle.

Significant AR will lower the aortic diastolic pressure, and the resultant large stroke volume will increase systolic and pulse pressures. This phenomenon is one of the hallmarks of significant AR and is associated with classic physical findings in this condition. The finding that lower DBP values are associated with higher mortality may just reflect more severe AR. However, low DBP was a predictor of mortality independent of markers of AR severity, such as left ventricular size and function.

Pulse pressure was weakly associated with mortality, but systolic BP was not. This is probably because systolic pressure is more related to arterial stiffness in this age group. This weakens the relationship with pulse pressure.

Considering the strong association of mortality with DBP, one must question the recommendation to aggressively treat hypertension in significant AR. In my experience, aggressive treatment of systolic BP in such patients is not well tolerated. If one is considering treating systolic hypertension in such patients, think about AVS. Aggressive treatment of systolic pressure often lowers DBP, too, which reduces coronary artery perfusion pressure. This may help explain why low DBP predicts mortality.

There were limitations to this study, including selection bias (about half the patients underwent AVS). Also, there are no data on the specifics of medical therapy or non-echo parameters such as BNP. Most patients were white men. Cardiovascular death was not analyzed, and patients with coronary artery disease were included. However, current thinking suggests clinicians may be waiting too long to operate on these patients.

Consideration of factors other than symptoms, left ventricular size, and function such as exercise hemodynamic response, global strain, and BNP, should be analyzed. Clearly, DBP and RHR should be added to this list.