C 2007, the Authors
C 2007, Blackwell Publishing, Inc. Journal compilation DOI: 10.1111/j.1540-8175.2007.00501.x
Left Ventricular Response to Exercise in Aortic Stenosis: An Exercise Echocardiographic Study Sylvestre Mar´echaux, M.D.,∗ Pierre-Vladimir Ennezat, M.D.,∗ Thierry H. LeJemtel, M.D.,§ Anne-Sophie Polge, M.D.,∗ Pascal de Groote, M.D., Ph.D.,∗ Philippe Asseman, M.D.,∗ R´emi Nevi`ere, M.D., Ph.D.,∗ Thierry Le Tourneau, M.D., Ph.D.,∗ and Ghislaine Deklunder, M.D., Ph.D.∗ ∗
Cardiology Hospital, Lille University Medical Center, CRHU Lille, France, §Division of Cardiology, Tulane School of Medicine, New Orleans, LA [Correction added after online publication 30-May-2007: Author affiliations have been corrected.] Background: While normal at rest, left ventricular (LV) systolic function may become abnormal during exercise in patients with aortic stenosis. Once contraindicated in patients with aortic stenosis, exercise testing is now recommended in asymptomatic patients with aortic stenosis to elicit symptoms and thereby ascertain the need for aortic valve replacement. However, the clinical significance of an abnormal LV response to exercise in asymptomatic patients with aortic stenosis remains unknown. Objective: The aim of this study was to evaluate the clinical implications of an abnormal LV response during exercise in the setting of aortic stenosis. Methods: We monitored the LV response to exercise by 2D-Doppler echocardiography during a symptom limited semirecumbent bicycle exercise in 50 patients with tight aortic stenosis (aortic valve area ≤ 1.0 cm2 ) and a normal LV systolic function (LV ejection fraction, EF ≥ 50%) and followed them for an average of 11 months. Results: Twenty patients had an abnormal LV response to exercise with a mean decrease in LV EF from 64 ± 10 to 53 ± 12% while 30 patients had a normal LV response to exercise with a mean increase in LV EF from 62 ± 7 to 70 ± 8%. Patients with an abnormal LV response during exercise were more likely to develop symptoms during exercise than patients with a normal LV response: 80% versus 27% (P < 0.0001). The survival free of cardiac events was significantly lower in patients with abnormal LV response to exercise than in patients with a normal response (P = 0.03). Conclusion: Exercise echocardiography provides objective data that facilitate interpretation of exercise elicited symptoms in asymptomatic patients with severe aortic stenosis. In addition, an abnormal LV response to exercise may predict a poor outcome. (ECHOCARDIOGRAPHY, Volume 24, October 2007) aortic stenosis, left ventricular function, exercise echocardiography, prognosis As left ventricular (LV) mass increases, LV systolic function may remain normal at rest in patients with severe aortic stenosis despite a substantial increase in afterload and patients may remain asymptomatic.1 However, LV systolic function may become abnormal during exercise in patients with aortic stenosis.2 Exercise testing that was once contraindicated in patients with severe aortic stenosis, is now rec-
Address for correspondence and reprint requests: Pierre V. Ennezat, M.D., Cardiology Hospital, Bd Pr J Leclerc, Lille, France. Fax: +33 20 44 65 04; E-mail:
[email protected] Sylvestre Mar´echaux and Pierre-Vladimir Ennezat contributed equally to the preparation of the manuscript.
Vol. 24, No. 9, 2007
ommended to elicit symptoms in aortic stenosis patients and thereby to ascertain the need for aortic valve replacement.3–5 The aim of the present study was to evaluate LV response to exercise in patients with severe aortic stenosis who do not volunteer symptoms during daily activities. Accordingly we prospectively monitored by Doppler echocardiography the LV response to exercise in asymptomatic patients with severe aortic stenosis and normal resting LV systolic function. All patients performed a symptom-limited exercise on a semirecumbent bicycle exercise and were followed up thereafter. The decision to replace the aortic valve was left to the patient and the referring cardiologist.
ECHOCARDIOGRAPHY: A Jrnl. of CV Ultrasound & Allied Tech.
955
´ MARECHAUX, ET AL.
Methods Study Population Patients who were referred to the cardiovascular noninvasive laboratory of Lille University Medical Center for assessment of aortic stenosis were eligible for the study when they did not volunteer symptoms during daily activities, had an aortic valve area ≤1 cm2 , and a preserved LV ejection fraction (≥50%). Patients underwent coronary angiography when considered for aortic valve replacement. Patients with chronic obstructive pulmonary or peripheral arterial disease, heart failure, atrial fibrillation or flutter, technically inadequate echocardiograms, and mitral or aortic regurgitation greater than 1+ were ineligible for the study. All patients signed an informed consent before participating in the study.
chamber view at the level of the LV outflow tract and at the level of valve leaflets, respectively by pulsed- and continuous-wave Doppler. Mean and peak transvalvular gradients were derived using the modified Bernoulli equation. Aortic valve area was determined using the continuity equation using the ratio of the VTI across the valve and in the LVOT. Resting LV diastolic function was assessed by peak velocities of the E-wave (early diastole) the A-wave (late diastole) the deceleration time of E-wave, the E’-wave (early diastolic mitral annular velocity) and the E/A and the E/E’ ratios. For the E’-wave, the sample volume was placed sequentially at the lateral and medial mitral annulus. Both velocities were averaged. Systolic tricuspid pressure gradient was estimated from the maximal tricuspid regurgitation velocity.
Exercise Testing
Follow-up
Patients exercised on a semirecumbent bicycle ergometer to exhaustion or until they experienced incapacitating symptoms. The initial workload was 25 Watts that was increased by 25 Watts increment every 3 minutes. Blood pressure was measured at rest while sitting on the semirecumbent bicycle and during the last minute of each workload. Patients discontinue exercising because of incapacitating symptoms or because of exhaustion. The exercise test was considered to be abnormal in the presence of angina, shortness of breath, near syncope or syncope, ≥2 mm ST segment depression 80 ms after the J point, fall or no increase in systolic blood pressure at peak exercise when compared with baseline level, and ventricular arrhythmias. None of the patients were receiving βadrenergic blockers or calcium antagonists at the time of exercise testing.
After exercise echocardiography, the clinical status of patients was monitored by telephone calls to referring cardiologists and primary care physicians. The decision and timing of aortic valve replacement was left to referring cardiologists who were aware of the exercise echocardiography findings.
Exercise Echocardiography Continuous 2-dimensional Doppler echocardiography was performed with a SONOS 7500 (Philips Andover, MA, USA) and a 2–4 MHz transducer by the same experienced echocardiographer. Measurements were made over at least 3 cardiac cycles and the averaged value taken. Volumes and ejection fraction (EF) were calculated using the Simpson biplane method. Parasternal long-axis view was used to measure the LV outflow diameter that assumed to be constant at different flow states. Velocities time integrals (VTI) were measured in the apical 5956
Statistical Analysis Variables are expressed as mean ± SD or absolute numbers and percentages as appropriate. Abnormal LV response to exercise was predefined as a decrease in LV ejection fraction (LV ejection fraction from rest to peak exercise < 0%). Paired and unpaired continuous variables were compared by Wilcoxon’s and Mann–Whitney’s tests, respectively. Chisquare and Fischer’s exact tests were used as appropriate for categorical variables. Eventfree survival was defined as survival without cardiovascular death or spontaneous symptoms (dyspnea, angina, syncope, and heart failure). Patients who underwent AVR without preceding spontaneous symptoms on the basis of exercise testing and patient’s or physician’s decision were censored at the time of surgery because of the nonblinded nature of the study. Survival curves according to LV response to exercise established by the Kaplan–Meier estimation method were compared using the log rank test. P values
