An 86-year-old man was referred to the hypertension clinic because of difficult-to-control systolic hypertension and low diastolic blood pressure. He stated that he had been on antihypertensive medication for approximately 5 years. A review of his outpatient clinic record revealed a mean blood pressure of 163/57 mm Hg between April 1998 and August 2004. From August 2004 until January 2007, the mean blood pressure was 168/51 mm Hg (Figure 1).
Echocardiography in 1993 showed moderate aortic regurgitation (AR) with a normal-appearing aortic valve and a dilated aortic root. The left ventricle was mildly dilated with normal systolic function. Follow-up echocardiography in January 2006 showed progression to severe AR, mild to moderate aortic stenosis, and an estimated ejection fraction of 60%.
When seen in consultation, the patient was lean and described himself as active. He denied chest discomfort, breathing difficulty, dizziness, or loss of consciousness. Some ankle swelling was noted while taking long-acting nifedipine. He stated that he was taking nifedipine extended-release 30 mg/d, losartan 50 mg BID, and furosemide 20 mg/d. An intermittent dry daily cough had persisted despite changing lisinopril to losartan.
Blood pressure was 162/50 mm Hg in both arms supine and 156/52 mm Hg in the right arm standing. Cardiac examination revealed a regular rate and rhythm with I–II/VI systolic and II/VI diastolic murmurs throughout the precordium. The diastolic murmur was sustained through most of diastole. Heart tones were soft and there were no gallops. Bounding carotid and femoral pulses were noted without bruits or hepatojugular reflux. There was 1+ bilateral pretibial edema. Pertinent laboratory results showed the following values: creatinine, 1.0 mg/dL (normal, 0.7–1.3 mg/dL); potassium, 5.2 mmol/L (normal, 3.5–5.0 mmol/L); sodium, 140 mmol/L (normal, 135–145 mmol/L); low-density lipoprotein cholesterol, 79 mg/dL (normal, <100 mg/dL); and hemoglobin, 12.1 mg/dL (normal, 14.0–18.0 g/dL). Electrocardiography showed left ventricular hypertrophy (LVH) by voltage criteria with a heart rate of 72 beats/min.
The patient’s edema was felt to be a calcium channel blocker effect, and furosemide did not reduce it. The patient was instructed to stay on nifedipine extended-release 30 mg/d, to switch from furosemide to hydrochlorothiazide (HCTZ), and to use daytime compression stockings to control his mild edema. HCTZ was increased to 50 mg/d and losartan was increased to 100 mg/d. On this 3-drug combination, follow-up blood pressure values were 142 to 144/50 to 52 mm Hg.
Care of this elderly patient with systolic hypertension, severe asymptomatic AR, and low diastolic blood pressure raises clinical questions: (1) What is the relationship between hypertension and AR? (2) Are there particular drugs to favor or avoid in treating hypertension in patients with AR? (3) Should low diastolic blood pressure limit the systolic blood pressure treatment goal?
Examinations of the Framingham cohort have shown that the most important determinant of all stages of AR is age (Table I).1,2 Male sex is a far less important contributor, and the presence of hypertension is unrelated to AR. AR is uncommon in patients with hypertension younger than 60.3 Surveys generally show that idiopathic age-related aortic root dilatation, unknown causes, and degenerative valve disease are the leading causes of AR, predominating over congenital abnormalities, rheumatic heart disease, and infective endocarditis.4 In developed countries, age-related dilatation of the aortic root is the most common cause of severe AR.5 Chronic severe AR contributes to increased dilatation of the aortic root and AR progression.
Table I. Prevalence of Aortic Regurgitation in the Framingham Offspring Study
Values are expressed in percentages. Reprinted with permission from Singh et al, 1999.2
However, there is an important interplay between AR and systolic hypertension. Systolic hypertension increases wall stress and the left ventricular volume:mass ratio.4 Prolonged asymptomatic stability occurs as the left ventricle adapts to the volume overload with increasing end-diastolic volume, compliance, and hypertrophy. As compensatory preload increases, the slope of the pressure volume relationship or elastance diminishes. When the limits of compensation are reached, further increases in wall stress lead to myocardial dysfunction and systolic failure. Simultaneous LVH and fibrosis cause diastolic dysfunction. At this stage, symptoms include those of heart failure as well as angina, which may occur with normal coronary arteries due to the loss of coronary reserve with LVH as well as diminished coronary perfusion pressure.6 The natural history of AR is described in Table II.7 While heart failure symptoms are the most common indication for aortic valve replacement, declining ejection fraction is also a surgical indication in asymptomatic patients.7
Table II. Natural History of Aortic Regurgitation
Abbreviation: LV, left ventricular. Reprinted with permission from Bonow et al, 2008.7
Asymptomatic patients with normal LV systolic function
Progression to symptoms and/or LV dysfunction
<6% per year
Progression to asymptomatic LV dysfunction
<3.5% per year
<0.2% per year
Asymptomatic patients with LV systolic dysfunction
Progression to cardiac symptoms
>25% per year
>10% per year
In a prospective cohort study of 80 asymptomatic patients with chronic severe AR, the presence of systolic hypertension was the major predictor of adverse clinical outcomes, independent of age, sex, diastolic blood pressure, and ejection fraction.8 All patients had diastolic blood pressure levels <90 mm Hg, systolic hypertension was found in 33%, and 27 patients had a pulse pressure ≥82 mm Hg, representing the highest tercile. Cardiac events were 3 times more likely in patients with systolic hypertension than in those without, and the prognostic effect of systolic hypertension was greatest in those with pulse pressure values ≥82 mm Hg.8
The major goal of medical therapy in chronic AR is to treat systolic hypertension. The evidence that vasodilator therapy acts to reduce wall stress, improve left ventricular function, and reduce progression of AR is mixed. With the chronic volume overload seen in advanced AR, the most consistent vasodilator effect in several studies has been a decrease in left ventricular filling pressures.9 A compilation of hemodynamic effects seen in long-term vasodilator therapy of AR is seen in Table III.9–17 Improvements occurred in ejection fraction, regurgitant fraction, end-diastolic and end-systolic volumes, end-diastolic and end-systolic dimensions, and fractional shortening. In a prospective randomized study of 143 patients with severe asymptomatic AR, patients were selected for treatment with either digoxin or nifedipine 20 mg BID. After 6 years, there was a significant difference in the incidence of aortic valve replacement favoring the nifedipine group when the chief indication for valve replacement was symptoms (Figure 2).18 Ejection fraction, left ventricular mass, as well as end-diastolic and end-systolic volumes were significantly improved in the nifedipine group.
Table III. Vasodilators in Aortic Regurgitation: Long-Term Effects
The results of the trial by Scognamiglio and associates18 showing clinical benefit of afterload reduction with nifedipine were not reproducible. In a smaller trial of 95 patients with asymptomatic severe AR randomized to open-label nifedipine 20 mg BID, enalapril 20 mg/d, and placebo, there was no difference among the 3 groups after a mean of 7 years of follow-up in the rate of aortic valve replacement, ejection fraction, regurgitant volume, or left ventricular mass. The trial by Evangelista and colleagues19 had the advantage of a placebo group compared with the trial by Scognamiglio and associates,18 but it was not blinded to the prescribed treatment and had insufficient statistical power to conclude that vasodilator therapy was without benefit.19 The population in Scognamiglio and colleagues’18 study had higher systolic blood pressure and greater pulse pressure values. Due to the large overlap of patient types in the 2 trials, a subset of patients for whom vasodilator therapy might be beneficial could not be defined.
After evaluating the conflicting data, the 2008 American College of Cardiology/American Heart Association (ACC/AHA) valvular heart disease guidelines downgraded prior class I recommendations for long-term vasodilator therapy in severe asymptomatic AR to class IIb (recommendation's usefulness/efficacy less well established) and gave vasodilator therapy a class III indication (general agreement that treatment is not useful/effective) for asymptomatic patients with mild to moderate AR.7 They concluded that the sole indication of vasodilator therapy in AR patients was treatment of hypertension.
Are there any drugs to be avoided in patients with systolic hypertension and AR? As previously noted, aortic regurgitant volume is a determinant of AR progression. Because aortic regurgitant volume is directly proportional to the duration of diastole, heart rate–slowing drugs would be expected to increase regurgitant volume and therefore be deleterious. However, in a retrospective review of 20 patients with moderate or severe AR associated with Takayasu arteritis who were followed up to almost 2 years, the use of β-blockers in patients with severe refractory hypertension and already on other agents was associated with a decrease in left ventricular mass in 4 cases.20 The authors conjectured that reduction of sympathetic nervous system activity by β-blockade negated the effect of increased regurgitant volume in these patients. 2008 ACC/AHA valvular heart disease guidelines make no mention of a contraindication for β-blockade in the treatment of hypertension associated with AR.7 Consistent with a growing consensus of the third or fourth place of using β-blockers in managing hypertension in elderly patients lacking a compelling indication,21 the placement of β-blockers in managing refractory hypertension in patients with severe AR should be the same. In the absence of a compelling indication, treatment-related chronic bradycardia should be avoided in severe AR.
Reducing diastolic blood pressure may be of significant concern in patients with AR due to the possible J-shape relation increasing risk of coronary events and stroke. How low should a clinician feel comfortable lowering diastolic blood pressure in order to reduce systolic blood pressure? Analyses of a possible J-point effect in the 2 large treatment trials of isolated systolic hypertension, the Systolic Hypertension in the Elderly Program (SHEP) and the Systolic Hypertension in Europe Trial (Syst-Eur), are beset by the small numbers of patients with very low diastolic blood pressure and the consequent large and overlapping confidence intervals.22,23 SHEP subset analysis suggested a J-point diastolic blood pressure of approximately 55 to 60 mm Hg,22 and evaluation of the Syst-Eur subsets suggested a diastolic J-point threshold of 55 mm Hg for patients without known coronary artery disease.23
Four English-language long-term studies of vasodilator therapy for AR contain blood pressure data at baseline and following intervention (Table IV).14,16,17,19 Of interest, the 2 studies with baseline diastolic blood pressure values from 60 to 68 mm Hg did not show a significant diastolic decrease despite significant reductions in systolic blood pressure with treatment.16,17 Given dramatic reductions in stroke, mortality, and heart failure seen in octogenarians in whom goal systolic blood pressure levels (<150 mm Hg) are reached,24 diastolic blood pressure lowering did not constitute a barrier in managing this patient. Systolic coronary flow velocity is more than doubled in patients with severe AR compared with control patients due to reduced resistance.25 There is some evidence that coronary perfusion improves with pulsatile flow into a low-compliance vascular system and that systolic blood pressure may therefore play an important role in coronary perfusion in elderly patients with low diastolic blood pressures.26
Table IV. Long-Term Vasodilator Studies Containing Blood Pressure Data in Patients With Aortic Regurgitation
No. of patients
mBPb, mm Hg
mBP6, mm Hg
mBPe, mm Hg
Abbreviations: ENAL, enalapril; HDZ, hydralazine; mBPb, mean blood pressure at baseline; mBP6, mean blood pressure after 6 months’ intervention; mBP12, mean blood pressure at end of study; NIF, nifedipine; NS, nonsignificant; QUIN, quinapril; S, significant.
The strategy for the current case was to use a maximized dose of a thiazide diuretic, maximized angiotensin receptor blocker therapy, and a dihydropyridine calcium channel blocker to achieve systolic blood pressure values <150 mm Hg in a patient with severe AR. The systolic blood pressure reduction was accomplished without further reduction in the diastolic blood pressure of 50 to 52 mm Hg.