Author + information
- Received November 26, 1996
- Revision received June 20, 1997
- Accepted July 10, 1997
- Published online November 1, 1997.
- Michael Y Henein, MD, PhD, FACCAB (, )
- Christine O’Sullivan, BScAB,
- George C Sutton, MD, FRCP, FACCAB,
- Derek G Gibson, MB, FRCPAB and
- Andrew J.S Coats, MD, FRCP, FACCAB,*
- ↵*Dr. Andrew J. S. Coats, Clinical Cardiology, Royal Brompton Hospital, Sydney Street, London, SW3 6NP, England, United Kingdom.
Objectives. We sought to identify the pattern of disturbed left ventricular physiology associated with symptom development in elderly patients with effort-induced breathlessness.
Background. Limitation of exercise tolerance by dyspnea is common in the elderly and has been ascribed to diastolic dysfunction when left ventricular cavity size and systolic function appear normal.
Methods. Dobutamine stress echocardiography was used in 30 patients (mean [±SD] age 70 ± 12 years; 21 women, 9 men) with exertional dyspnea and negative exercise test results, and the values were compared with those in 15 control subjects.
Results. Before stress, left ventricular end-diastolic and end-systolic dimensions were reduced, fractional shortening was increased, and the basal septum was thickened (2.3 ± 0.5 vs. 1.4 ± 0.2 cm, p < 0.001, vs. control subjects) in the patients, but posterior wall thickness did not differ from that in control subjects. Left ventricular outflow tract diameter, measured as systolic mitral leaflet septal distance, was significantly reduced (13 ± 4.5 vs. 18 ± 2 mm, p < 0.001). Isovolumetric relaxation time was prolonged, and peak left ventricular minor axis lengthening rate was reduced (8.1 ± 3.5 vs. 10.4 ± 2.6 cm/s, p < 0.05), suggesting diastolic dysfunction. Transmitral velocities and the E/A ratio did not differ significantly. At peak stress, heart rate increased from 66 ± 8 to 115 ± 20 beats/min in the control subjects, but blood pressure did not change. Transmitral A wave velocity increased, but the E/A ratio did not change. Left ventricular outflow tract velocity increased from 0.8 ± 0.1 to 2.0 ± 0.2 m/s, and mitral leaflet septal distance decreased from 18 ± 2 to 14 ± 3 mm, p < 0.001. In the patients, heart rate rose from 80 ± 12 to 132 ± 26 beats/min and systolic blood pressure from 143 ± 22 to 170 ± 14 mm Hg (p < 0.001 for each), but left ventricular dimensions did not change. Peak left ventricular outflow tract velocity increased from 1.5 ± 0.5 m/s (at rest) to 4.2 ± 1.2 m/s; mitral leaflet septal distance fell from 13 ± 4.5 to 2.2 ± 1.9 mm (p < 0.001); and systolic anterior motion of mitral valve appeared in 24 patients (80%) but in none of the control subjects (p < 0.001). Measurements of diastolic function did not change. All patients developed dyspnea at peak stress, but none developed a new wall motion abnormality or mitral regurgitation.
Conclusions. Although our patients fulfilled the criteria for “diastolic heart failure,” diastolic dysfunction was not aggravated by pharmacologic stress. Instead, high velocities appeared in the left ventricular outflow tract and were associated with basal septal hypertrophy and systolic anterior motion of the mitral valve. Their appearance correlated closely with the development of symptoms, suggesting a potential causative link.
Limitation of exercise tolerance by breathlessness is common in the elderly. When left ventricular cavity size is increased and left ventricular ejection fraction reduced, a diagnosis of heart failure secondary to systolic left ventricular disease is usually made , and treatment with diuretic drugs and angiotensin-converting enzyme inhibitors instituted [2–4]. However, in a significant minority of patients with a similar presentation, left ventricular cavity size is normal or even reduced, and some degree of basal septal hypertrophy may be present. Systolic function, in terms of cavity size and shortening fraction, is preserved, but measurements of diastolic function at rest are frequently abnormal. A diagnosis of diastolic heart failure is thus often invoked [5, 6], although this diagnosis depends on the assumption that the abnormalities present at rest are those that limit exercise tolerance. The aim of the present study was to examine this hypothesis by using Doppler echocardiography to study the effects of pharmacologic stress in a group of such patients. In particular, we wished to investigate the extent to which worsening diastolic function could be linked to the development of symptoms.
We studied 30 consecutive patients (mean [±SD] age 70 ± 12 years, range 48 to 84; 21 women, 9 men) presenting with reduced exercise tolerance, negative electrocardiographic stress test results and normal left ventricular systolic function, as demonstrated by echocardiography at rest. Left ventricular end-diastolic dimension was <5 cm and fractional shortening >30% in all 30 patients. Patients with significant valvular heart disease, previous cardiac surgery, hypertrophic cardiomyopathy, uncontrolled hypertension or a clinical diagnosis of chronic lung or musculoskeletal disease were excluded. All patients were in stable sinus rhythm. In eight patients, a clinical diagnosis of heart failure had been made, but all had denied symptomatic benefit from an angiotensin-converting enzyme (ACE) inhibitor. Fourteen patients had a documented history of hypertension that was controlled with diuretic drugs or an ACE inhibitor or a beta-adrenergic blocking agent. Chest X ray showed a normal cardiothoracic ratio in all. Rest ECG findings were usually normal but showed inverted T waves on the septal chest leads in two patients, first-degree heart block in two, right bundle branch block in one and left bundle branch block in one. We also studied 15 control subjects (mean age 67 ± 9 years; 10 women, 5 men) who were referred for assessment of nonspecific chest discomfort but did not develop breathlessness or chest pain during stress. All 15 had normal ECG findings and a normal cardiothoracic ratio; none had hypertension, abnormal cardiac signs or a family history of heart disease; and none were taking cardiac medication at the time of the study.
1.2 Exercise Stress Test
The exercise stress test was performed with a treadmill using the standard Bruce protocol, with a 3-min incremental increase in work load. The 12-lead ECG and blood pressure were continuously monitored. Predetermined exercise end points were either significant breathlessness or chest pain, a decrease in systolic blood pressure >20 mm Hg or sustained arrhythmia.
1.3 Rest Echocardiography
Rest echocardiography was performed using a Hewlett-Packard Sonos 1500 echocardiograph with a 2.5-MHz transducer. The pattern of left ventricular wall motion was assessed from standard left parasternal and apical views, with the patient in the left semilateral position. Systolic and diastolic left ventricular dimensions were measured from the M-mode recording of the left ventricular minor axis on the parasternal view, with the cursor by the tips of mitral valve leaflets. Peak rates of change of minor axis were derived from a digitized M-mode trace . Isovolumetric relaxation time was measured as the time interval from the aortic component of the second heart sound (A2), to the onset of mitral cusp separation on the mitral echogram. Septal thickness at the base and at midcavity and posterior wall thickness were measured from the left ventricular minor axis M-mode traces using leading edge methodology. End-diastole was taken as the onset of the q wave of the simultaneous ECG and end-systole as the onset of the first high frequency component of A2 on the phonocardiogram. Rest left ventricular systolic function was assessed from fractional shortening, expressed as a percentage. The distance from the mitral leaflet to the septum was measured in midsystole on the mitral echogram. Peak early (E) and late (A) diastolic transmitral blood flow velocities were obtained using the same transducer with pulsed Doppler from the apical four-chamber view, with the sample volume positioned at the tips of the mitral valve leaflets. Outflow tract velocities were recorded using continuous wave Doppler (Doptek, Colchester, UK), with the transducer positioned at the cardiac apex.
1.4 Dobutamine Echocardiography
Patients and control subjects underwent stress echocardiography; all medications were continued. Twelve-lead ECG electrodes and a blood pressure cuff were applied with the patient in the semilateral resting position. Dobutamine hydrochloride was delivered intravenously at a starting dose of 5 μg/kg body weight per min by an IVAC 770 infusion pump. The infusion rate was increased by 5 μg/kg per min for four stages, each lasting 3 min. During the last minute of each stage, left ventricular filling and outflow tract velocities were measured, as described earlier, on the two-dimensional echocardiographic display. Left ventricular wall motion was analyzed for the appearance of new regional abnormalities. A 12-lead ECG was routinely obtained at rest and during the last minute of each stage, using a Hewlett-Packard electrocardiograph. Blood pressure was measured automatically simultaneously using a Critikon Dynamap 8100 sphygmomanometer. Stress end points, identified prospectively, were significant shortness of breath or chest pain, an increase in systolic blood pressure to >200 mm Hg, symptomatic hypotension, significant ventricular arrhythmia or attainment of >85% of the maximal predicted heart rate (220 beats/min minus age [years]). No patient received atropine, and no patient achieved a significant decrease in systolic blood pressure or an arrhythmia as an end point.
Results are presented as mean value ± SD. Patient and control rest values were compared using the unpaired Student ttest. Interactions between rest and stress values, using two-way analysis of variance in the control and patient groups at each stage were investigated. The incidence of systolic anterior motion of the mitral valve in the patient group was assessed by the chi-square test.
2.1 Exercise Test (Table 1)
Patients were able to exercise for only 4.0 ± 1.2 min during a standard Bruce protocol, achieving a 65% increase in heart rate and an 18% increase in systolic blood pressure. Exercise was limited by shortness of breath in all 15 and additionally by chest pain in 8. No patient developed significant ST segment depression at peak exercise.
2.2 Rest Echocardiography (Table 2)
Similar echocardiographic results were observed in all patients (Fig. 1). Mean left ventricular end-diastolic and end-systolic cavity dimensions were 4.0 ± 0.4 and 2.4 ± 0.5 cm, respectively, significantly less than those in control subjects of similar age (p < 0.001); fractional shortening had a mean value of 41 ± 8%, and the peak shortening rate (9.4 ± 2.2 cm/s) did not differ from that in the control subjects. The interventricular septum showed basal hypertrophy, with a mean thickness of 2.3 ± 0.5 cm over the proximal 1.9 cm; distal to this, septal thickness was normal. Mean posterior wall thickness was not different from that in the control subjects (1.2 ± 0.2 cm). Isovolumetric relaxation time was significantly prolonged in the patient group to a mean of 100 ± 15 ms compared with 80 ± 14 ms in the control group (p < 0.005), and the peak rate of left ventricular minor axis increase was reduced (8.1 ± 3.5 vs. 10.4 ± 2.6 cm/s, p < 0.05). Outflow tract velocity was increased (1.5 ± 0.5 vs. 0.8 ± 0.1 m/s in control subjects, p < 0.001), but no patient had systolic anterior motion of the mitral valve leaflets at rest.
2.3 Dobutamine Stress Echocardiography (Tables 3–5)
Stress echocardiography was successfully completed in both patients and control subjects. All four stages were completed in the control subjects without significant symptoms. In the patients, the test was terminated because of breathlessness accompanied by chest pain at stage 3 ± 0.5 in all eight patients who had experienced these symptoms during exercise. No patient developed ST segment shift on the standard 12-lead ECG recorded at the time of the symptoms, and no patient developed any new wall motion disturbance of left ventricular wall movement suggestive of ischemia. Heart rate increased by 64% and 57%, respectively (p < 0.001) (Table 3). Systolic blood pressure was effectively unaltered; a small decrease in diastolic blood pressure was significant in the patient group. Left ventricular outflow tract velocity increased significantly in both patients and control subjects, reaching 2.0 m/s in the control subjects and 4.2 m/s in the patients. This difference was very significant, as was the interaction between stress level and control subjects versus patients (Table 4). These high velocities were present throughout ejection. In 24 patients (80%), the distance between the anterior cusp of the mitral valve and the basal septum decreased due to anterior motion of the former, with complete apposition in all 15 (Fig. 2). At the termination of the test, peak outflow tract velocity was 4.4 ± 0.2 m/s in patients with versus 3.5 ± 0.2 m/s in those without anterior motion (p < 0.001). Left ventricular isovolumetric relaxation time decreased consistently, from 100 ± 15 to 65 ± 12 ms at peak stress (p < 0.001). Early diastolic transmitral E wave velocity was unchanged, but atrial systolic A wave velocity increased, so there was a small but consistent decrease in the E/A ratio, not seen in the control subjects. Within 1.5 ± 0.3 min of achieving high outflow tract velocities (≥4 m/s), all patients developed symptoms that subsided within 2 ± 0.2 min of stopping the dobutamine infusion.⇓
Limitation of exercise tolerance by breathlessness in the presence of normal left ventricular systolic but disturbed diastolic function is common in the elderly. This combination is generally considered sufficient to establish a diagnosis of diastolic heart failure [5, 6]. Left ventricular function was similar to that in a group previously described , although the incidence of hypertension was lower. Exercise tolerance in our patients was limited to only 4 min of the standard Bruce protocol; left ventricular systolic function was normal, but discrete disturbances in diastole were present . However, during dobutamine stress, diastolic function did not noticeably deteriorate; instead, high blood velocities developed in the left ventricular outflow tract. These velocities were present throughout ejection rather than accelerating during late ejection, indicating that they were due to left ventricular outflow tract obstruction rather than as the result of dynamic end-systolic cavity obliteration or mitral regurgitation (Fig. 2). These velocities increased with increasing stress, and all patients developed symptoms within 1 to 2 min of their reaching an outflow tract velocity of 4 m/s, corresponding to a pressure decrease of 64 mm Hg. By contrast, the echocardiographic evidence of diastolic dysfunction was little affected: isovolumetric relaxation time decreased, and although A wave velocity on transmitral Doppler increased, this increase could be attributed to an increased heart rate. In addition, no wall motion disturbances occurred during stress, thus making myocardial ischemia unlikely. We therefore conclude that high blood velocities, consistent with significant left ventricular outflow tract obstruction, rather than any aggravation of diastolic dysfunction or myocardial ischemia, corresponded most closely with the appearance of symptoms in this group of patients.
3.1 Ventricular Structure and Function
It is normal for left ventricular wall thickness to increase in the elderly , a change usually attributed to an increase in connective tissue . Comparison of our control group (mean age 69 years) with a previous one (mean age 59 years) studied under identical conditions shows that over 10 years, end-diastolic cavity dimension decreases by as much as 1 cm, possibly as the result of involutional processes. In retrospect, left ventricular outflow tract velocities were slightly increased at rest in our patients compared with those in the control group of similar age, but not to the extent that would suggest any significant obstruction. Rest systolic mitral septal distance, however, was consistently lower in our patients than in the control group. Septal thickness was consistently increased at the base of the ventricle but was normal at the midcavity level. Functional obstruction to left ventricular ejection, often associated with systolic anterior motion of mitral leaflets, is well documented in patients in whom the left ventricular end-systolic cavity size is small, such as is seen after aortic valve replacement for aortic stenosis , with circulatory hypovolemia or external compression of the left ventricle by a pericardial effusion as well as with hypertrophic cardiomyopathy . None of our patients had evidence of any of these conditions or significant left ventricular hypertrophy on the rest ECG or a family history of heart disease, so we do not believe that hypertrophic cardiomyopathy was present. Any increase in overall left ventricular muscle mass was small, the localized increase in septal thickness being counteracted by reduced cavity dimensions. The blood pressure response, both systolic and diastolic, to exercise and dobutamine stress was normal, so that significant hypertension can also be excluded.
3.2 Study Limitations
We do not consider dobutamine stress echocardiography is to be justified in asymptomatic and clinically normal elderly patients. We were thus obliged to include a control group of similar age, with nonspecific chest pain, in whom stress echocardiographic results were negative and no significant pathologic abnormalities were found. Our protocol was noninvasive, so we do not have information on the presence or absence of coronary artery disease; its presence seems unlikely in view of the high left ventricular rate–pressure product achieved by our patients in the absence of ST segment shift and the failure of any localized wall motion abnormalities to appear when symptoms developed. Even in systolic heart failure, the exact basis of breathlessness on exertion remains uncertain , so we are not surprised at being unable to prove conclusively that the high systolic outflow tract velocities in our patients were the direct cause of symptoms. However, their uniform appearance shortly before the onset of breathlessness along with chest pain in the eight patients who also developed pain on exercise strongly suggests that high outflow tract velocities were at least part of a sequence of physiologic changes leading to the appearance of symptoms rather than an unrelated phenomenon. This conclusion is also supported by the absence of both symptoms and high velocities in the control group. The lack of significant changes in transmitral Doppler flow velocities and isovolumetric relaxation time seems to exclude any major disturbance of left ventricular diastolic function, although we cannot exclude the possibility that the effects of impaired relaxation were exactly balanced by those of an increase in left atrial pressure.
3.3 Clinical Significance
Our results thus confirm the existence of a group of elderly patients, predominantly female, with marked limitation of exercise tolerance despite normal rest left ventricular systolic function , although the mitral valve septal distance during systole is reduced. Although diastolic function was abnormal at rest, our observations suggest that symptoms during pharmacologic stress are related more to the development of significant left ventricular outflow tract obstruction during systole and to the associated increase in myocardial systolic wall stress and work than to any interference with ventricular diastolic function or the development of ischemia. It was previously demonstrated [8, 18, 19]that such patients respond adversely to vasodilator and diuretic drugs but more satisfactorily to drugs with a negative inotropic effect, such as verapamil or beta-blockers. These observations are compatible with our results and stress the importance of making the diagnosis in individual patients. The beneficial effects of agents with a negative inotropic action are also explicable on the basis of the mechanisms proposed in the present report. If outflow tract obstruction does prove to be the cause of exercise limitation in these patients, then a series of potential therapeutic approaches becomes available for treating a common and disabling condition whose incidence is likely to increase further in an aging population.
☆ This work is supported by the Royal Brompton Hospital Special Cardiac Fund, British Heart Foundation and Viscount Royston Trust, London, England, United Kingdom.
- angiotensin-converting enzyme
- electrocardiogram, electrocardiographic
- Received November 26, 1996.
- Revision received June 20, 1997.
- Accepted July 10, 1997.
- The American College of Cardiology
- The Task Force on Heart Failure of the European Society of Cardiology
- Captopril Multicentre Research Group
- Appleton CP
- Gibson DG,
- Brown D
- Cuocolo A,
- Sax FL,
- Brush JE,
- Maron BJ,
- Bacharach SL,
- Bonow RO
- Pellikka PA,
- Oh JK,
- Bailey KR,
- Nichols BA,
- Monahan KH,
- Tajik AJ
- Henein MY,
- Xiao HB,
- Brecker SJD,
- Gibson DG
- Aurigemma G,
- Battista S,
- Orsinelli D,
- Sweeney A,
- Pape L,
- Cuenoud H
- Coats AJS,
- Poole-Wilson PA