- Email: [email protected]
Left Ventricular Diastolic Dysfunction Late After Aortic Valve Replacement in Patients With Aortic Stenosis
Left Ventricular Diastolic Dysfunction Late After Aortic Valve Replacement in Patients With Aortic Stenosis Peter Gjertsson, MD, PhDa,*, Kenneth Caidahl, MD, PhDa,b, and Odd Bech-Hanssen, MD, PhDa Patients with severe aortic stenosis (AS) are known to have increased left ventricular (LV) mass and diastolic dysfunction. It has been suggested that LV mass and diastolic function normalize after aortic valve replacement (AVR). In the present study, change in LV mass index and diastolic function 10 years after AVR for AS was evaluated. Patients who underwent AVR from 1991 to 1993 (n ⴝ 57; mean age 67 ⴞ 8.6 years at AVR, 58% men) were investigated with Doppler echocardiography preoperatively and 2 and 10 years postoperatively. Diastolic function was evaluated by integrating mitral and pulmonary venous flow data. Expected values for each patient, taking age into consideration, were defined using a control group (n ⴝ 71; age range 18 to 83 years). Patients were classified into 4 types: normal diastolic function (type A), mild diastolic dysfunction (type B), moderate diastolic dysfunction (type C), and severe diastolic dysfunction (type D). There was a reduction in LV mass index between the preoperative (161 ⴞ 39 g/m2) and 2-year follow-up (114 ⴞ 28 g/m2) examinations (p <0.0001), but no further reduction was seen at 10 years (119 ⴞ 49 g/m2). The percentage of patients with increased LV mass index decreased from 83% preoperatively to 29% at 2-year follow-up (p <0.001). The percentage of patients with moderate to severe LV diastolic dysfunction (types C and D) was unchanged between the preoperative (7%) and 2-year follow-up (13%) examinations (p ⴝ 0.27). The percentage of patients increased at 10-year follow-up to 61% (p <0.0001). In conclusion, this reveals the development of moderate to severe diastolic dysfunction 10 years after AVR, despite a reduction in the LV mass index. © 2005 Elsevier Inc. All rights reserved. (Am J Cardiol 2005;96:722–727)
Patients with aortic stenosis (AS) are known to have left ventricular (LV) hypertrophy and diastolic dysfunction. It has previously been shown that regression of LV hypertrophy after aortic valve replacement (AVR) is seen at intermediate follow-up, 2 to 4 years after AVR.1–3 LV hypertrophy in patients with AS consists of increased muscle fiber diameter and interstitial fibrosis,4,5 and this can explain the observed diastolic dysfunction. Endomyocardial biopsy and functional studies at intermediate and long-term follow-up have shown that the reduction of interstitial fibrosis and the reversal of diastolic dysfunction after AVR for AS is a slow process, continuing for many years.4,5 Previous reports indicate that patients after AVR experience the normalization of LV mass and diastolic function in the long term.3,4 These
a
Department of Clinical Physiology, Cardiovascular Institute, Sahlgrenska University Hospital, Göteborg; and bKarolinska Institute, Stockholm, Sweden. Manuscript received January 11, 2005; revised manuscript received and accepted April 13, 2005. This study was supported by grants from the Göteborg Medical Society, Göteborg; Sahlgrenska University Hospital, Göteborg; the Swedish Heart and Lung Foundation, Stockholm; the Västra Götaland Region, Vänersborg; the Swedish Medical Research Council, Stockholm; and the Swedish Medical Society, Stockholm, Sweden. * Corresponding author: Tel: 46-31-3421800; fax: 46-31-411735. E-mail address: [email protected] (P. Gjertsson). 0002-9149/05/$ – see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2005.04.052
studies, however, were small, and most important, the patients were relatively young and therefore not representative of the population of patients who undergo AVR. In the present study, we used Doppler echocardiography to evaluate the long-term effect on LV mass and diastolic function after AVR in a representative group of patients with severe AS. Methods Patients: From January 1991 to August 1993, a previously described population of 239 patients underwent AVR at Sahlgrenska University Hospital (Göteborg, Sweden).6 Patients with severe aortic regurgitation (grade ⬎2/4) or concomitant mitral surgery were excluded. Two years after surgery, 137 patients participated in an investigation.1 The 77 patients who were still alive 10 years after surgery were asked to participate in a new investigation; 20 of these patients were not willing to participate. The study population therefore comprised 57 patients who were prospectively investigated using echocardiography preoperatively and 2 and 10 years after surgery. At all 3 investigations, patients were asked about their symptoms (dyspnea and angina). Preoperative coronary angiographic reports were collected. Informed consent was obtained from all patients. The www.AJConline.org
Valvular Heart Disease/Diastolic Dysfunction in AS Patients
human ethics committee at Sahlgrenska University Hospital approved the study. M-mode and 2-dimensional echocardiography: Mmode measurements were made according to the American Society of Echocardiography.7 LV mass was calculated according to the corrected cube formula,8 and the LV ejection fraction was calculated either according to Simpson’s rule9 or according to Teichholz et al.10 Signs of hypertrophy were defined as increased LV mass. In patients in whom the LV ejection fraction could not be calculated because of uncertain outlining or an angulated chamber, an experienced observer made a visual estimation of the LV ejection fraction. The same method of estimating the LV ejection fraction for patients was used at all 3 examinations. Planimetry of the left atrium was performed from a late systolic stop frame with the maximum atrial area. Doppler measurements: Blood flow velocity in the LV outflow tract was estimated by pulse-wave Doppler from an apical 4-chamber view with a sample size of 5 mm. Mitral flow was recorded between the mitral leaflets in the 4-chamber view. From the mitral velocity tracings, early flow velocity (E), the deceleration time of E waves, and peak velocity during atrial systole (A) were measured. The E/A ratio was calculated. Pulmonary venous flow velocities were obtained from the upper right pulmonary vein. Peak velocities during systole (S) and diastole (D) were measured. The S/D ratio was calculated. Continuous-wave Doppler signals were recorded from multiple windows with a 2-MHz nonimaging probe. The stroke volume was calculated as the product of the crosssectional area of the LV outflow tract and the velocity–time integral. Pressure gradients were calculated according to the simplified Bernoulli equation. The systolic tricuspid gradient was estimated using continuous-wave Doppler. At the 10-year follow-up investigation, the LV isovolumic relaxation time was calculated using pulsed Doppler, with a sample size of 8 mm and the sample volume positioned anteriorly and medially from the mitral valve toward the LV outflow tract.11 Patterns describing diastolic function: Diastolic function was evaluated by integrating mitral flow and pulmonary venous information.11–13 Because the normal range for the E/A ratio is wide, deceleration time was also used to distinguish between normal diastolic function and mild diastolic dysfunction. Four different filling patterns were identified: type A: normal diastolic function (normal E/A ratio, S/D ratio, and deceleration time); type B: mild diastolic dysfunction (decreased E/A ratio or increased deceleration time and normal S/D ratio); type C: moderate diastolic dysfunction (normal E/A ratio [pseudonormalization] and decreased S/D ratio); and type D: severe diastolic dysfunction (increased E/A ratio and decreased S/D ratio). The patients were further divided into 2 groups: those with moderate to severe diastolic dysfunction (types C and D),
723
who often have increased LV filling pressure, and those with normal diastolic function or mild diastolic dysfunction (types A and B), who are likely to have normal filling pressures.11 Mitral and pulmonary venous flow patterns are age dependent.14 A healthy control group (n ⫽ 71; age range 18 to 83 years, 59% men) without hypertension or diabetes mellitus, with normal electrocardiographic results at rest and without a history of heart disease, was compared with the study group. Patients with atrial fibrillation or pacemaker rhythm were not included in the classification of diastolic function type, as well as patients from whom E/A or S/D ratios were not obtained. Statistical analysis: Continuous variables are summarized using the mean and SD. Categorical variables are summarized using absolute and relative frequencies. The mean of 3 M-mode and Doppler measurements was used in patients with sinus rhythm, and the mean of 5 measurements was used in patients with atrial fibrillation. Differences among investigations were evaluated using a paired Student’s t test. Differences among groups were evaluated using an unpaired Student’s t test. Proportions among patient groups were tested using the chi-square test and Fisher’s exact test when the sample size was small. A p value of ⬍0.05 was considered statistically significant. Diastolic function parameters are known to be age dependent, and LV mass is known to be dependent on body surface area. From the healthy controls, regression equations were calculated, as well as residuals. For each patient in the study group, the expected LV mass and E/A and S/D ratios were predicted. The observed values in patients were regarded as being reduced or increased if they differed by ⬎1.96 SD from the predicted value using the Z score.
Results Clinical characteristics: The patients were 67 ⫾ 8.6 years old at the time of surgery, 58% were men, 58% received mechanical prostheses (St. Jude Medical, Inc., St. Paul, Minnesota), and 42% received stented biologic prostheses (Biocor, St. Jude Medical). Normal findings on coronary angiography were seen in 71% of the patients. Combined coronary artery bypass grafting was performed in 25%. In relation to the AVR, echocardiography was performed a median of 1.0 days before (range 1 to 84), 2.1 years after (range 1.0 to 3.2), and finally 9.8 years after (range 7.0 to 11.7) the procedure. Severe dyspnea (New York Heart Association classes III and IV) was present in 52% preoperatively, in 4% at 2-year follow-up (p ⬍0.0001), and in 7% at 10-year follow-up (p ⫽ 0.43). Angina was present in 29% preoperatively, in 2% at 2-year follow-up (p ⬍0.0001), and in 2% at 10-year follow-up. Atrial fibrillation was present in 7% preoperatively, in 6% at 2-year follow-up (p ⫽ 0.77), and in 14% at 10-year follow-up (p ⫽ 0.12). Clinical and echocardiographic characteristics at the 3 examinations are listed in Table 1.
724
The American Journal of Cardiology (www.AJConline.org)
Table 1 Clinical and echocardiographic characteristics preoperatively and 2 and 10 years after AVR Variable
Preoperative (n ⫽ 57)
Follow-Up
142 ⫾ 22 82 ⫾ 13 69 ⫾ 13 53 (93%) 99 ⫾ 29 63 ⫾ 19 0.56 ⫾ 0.13 2.8 ⫾ 0.8 23 ⫾ 6 21 ⫾ 4 272 ⫾ 107
156 ⫾ 20* 160 ⫾ 23‡ 80 ⫾ 20 79 ⫾ 11 72 ⫾ 13* 71 ⫾ 14 52 (92%) 46 (81%) 26 ⫾ 9* 24 ⫾ 10 15 ⫾ 5* 13 ⫾ 5† 0.59 ⫾ 0.10 0.57 ⫾ 0.09 3.1 ⫾ 0.9* 2.8 ⫾ 1.1 21 ⫾ 4 22 ⫾ 6 22 ⫾ 5 26 ⫾ 7† 248 ⫾ 73 236 ⫾ 88
(p ⫽ 0.04) and the LV ejection fraction (p ⫽ 0.07). At 10-year follow-up, the LV ejection fraction and cardiac index had decreased slightly (Table 1).
Left ventricular mass: Changes in LV dimensions and LV mass index are shown in Figure 1. Preoperatively, 83% of the patients showed signs of LV hypertrophy. This had decreased to 29% (p ⬍0.0001) at 2-year follow-up. No further decrease was seen at 10-year follow-up, when 34% showed signs of LV hypertrophy (p ⫽ 0.61).
Left ventricular diastolic function: Deceleration time decreased during 10-year follow-up (Table 1), whereas the E/A ratio increased and the S/D ratio decreased (Figure 2). Figure 3 shows the distribution of diastolic types (A to D) at the 3 examinations. Forty-five patients were included in the diastolic function type classification at the preoperative investigation, 40 patients were included at 2-year follow-up, and 43 patients were included at 10-year follow-up. Only 7% of the patients had moderate to severe diastolic dysfunction (types C and D) preoperatively. There was no significant change in the percentage of estimated diastolic type between the preoperative examination and 2-year follow-up (p ⫽ 0.27). At 10-year followup, there was a marked increase in patients with moderate to severe diastolic dysfunction (61%, p ⬍ 0.0001). The tricuspid systolic gradient was increased at 10-year follow-up compared with 2-year follow-up, but the size of the left atrium was unchanged (Table 1). The characteristics of patients with moderate to severe diastolic dysfunction and normal diastolic function or mild dysfunction at 10-year follow-up are listed in Table 2. The regression equation for the E/A ratio, calculated from the control group, was 2.32 ⫺ 0.019 ⫻ age (in years), R ⫽ 0.79, and the SD for the residuals was 0.26. The regression equation for the S/D ratio was 0.46 ⫹ 0.014 ⫻ age (in years), R ⫽ 0.74, and SD for the residuals was 0.21.
Left ventricular systolic function: LV systolic function was improved at 2-year follow-up compared with the preoperative examination, with an increase in cardiac index
Gender differences: Impaired diastolic function was seen predominantly in the women (Table 2). A further gender analysis was therefore carried out (Table 3). There was no signif-
Systolic BP (mm Hg) Diastolic BP (mm Hg) Heart rate (beats/min) Sinus rhythm Maximum gradient (mm Hg) Mean gradient (mm Hg) LV ejection fraction Cardiac index (L/min/m2) Left atrium size (cm2) Tricuspid gradient (mm Hg) Deceleration time (ms)
At 2 Years (n ⫽ 57)
At 10 Years (n ⫽ 57)
Values are expressed as mean ⫾ SD. * p ⬍0.05 comparing preoperative and 2-year follow-up. † p ⬍0.05 comparing 2-year and 10-year follow-up. ‡ p ⬍0.05 comparing preoperative and 10-year follow-up. BP ⫽ blood pressure.
Figure 1. Left, LV diastolic diameter (open circles), interventricular septum (boxes), and posterior wall thickness (closed circles) preoperatively and at the 2- and 10-year follow-up. Right, LV mass index preoperatively and at the 2- and 10-year follow-up. Error bars show SDs.
Valvular Heart Disease/Diastolic Dysfunction in AS Patients
725
Figure 2. E/A ratio (left) and S/D ratio (right) preoperatively and 2 and 10 years postoperatively. Error bars show SDs.
Discussion
Figure 3. Distribution of different diastolic function types preoperatively and 2 and 10 years postoperatively: type A: normal diastolic function; type B: mild diastolic dysfunction; type C: moderate diastolic dysfunction; type D: severe diastolic dysfunction.
icant difference between men and women in terms of LV mass index or maximum gradient preoperatively or at follow-up, and there was no difference in age. A smaller percentage of women than men tended to be subjected to concomitant coronary bypass grafting. Women were more symptomatic preoperatively (61% with severe dyspnea compared with 45% of men; p ⫽ 0.25). At follow-up, women had greater systolic blood pressure.
To our knowledge, this is the first longitudinal study of a representative group of patients with AS to evaluate the regression of LV mass and change in LV diastolic function by Doppler echocardiography at intermediate and late follow-up. The most important finding was that in patients with preoperatively normal LV diastolic function or mild dysfunction, we observed moderate to severe diastolic dysfunction at 10-year follow-up, despite a reduction in LV mass index. We describe changes in heart function and LV mass occurring over a 10-year period after AVR. It is known that LV diastolic function changes with normal aging, and we therefore included a control group to define the correlation between diastolic function parameters and age. From a regression equation, we predicted the expected values and classified an observed value in a patient as normal, reduced, or increased. In the changes in type and degree of diastolic impairment described here, we have therefore corrected for the normal changes occurring with aging. We think that this is an important point that is unique to the present study and strengthens the conclusion that diastolic function actually worsens from the preoperative investigation to late follow-up after AVR. Our findings regarding the level of LV mass reduction after AVR agree with those of other investigators,1,3,5,15,16 and 2/3 of the patients achieved normal LV mass after 10 years. It was thus unexpected to find that most of the patients developed more severe diastolic dysfunction. LV hypertrophy in patients with pressure overload is due to an increase in muscle fiber diameter and an increase in fibrous content.4 Villari et al4 showed that the reduction in LV mass
726
The American Journal of Cardiology (www.AJConline.org)
Table 2 Patients with normal diastolic function or mild diastolic dysfunction (types A and B) or moderate to severe diastolic dysfunction (types C and D) 10 years after valve replacement Variable
Types A and B (n ⫽ 17)
Types C and D (n ⫽ 26)
p Value
Isovolumic relaxation time (ms) Left atrium area/BSA (cm2/m2) LV mass index (g/m2) Tricuspid gradient (mm Hg) Prosthesis maximum gradient (mm Hg) Prosthesis mean gradient (mm Hg) LV ejection fraction Normal coronary angiography Concomitant coronary bypass Bioprosthesis Distribution of valve sizes: 19 21 23 25 27 (n) Men Systolic BP preoperative (mm Hg) Systolic BP at 2 yrs (mm Hg) Systolic BP at 10 yrs (mm Hg)
107 ⫾ 17 10 ⫾ 1.7 127 ⫾ 69 23 ⫾ 5 30 ⫾ 12 15 ⫾ 6 0.57 ⫾ 0.09 10 (57%) 7 (41%) 10 (59%) 0/0/7/8/2 14 (82%) 135 ⫾ 22 152 ⫾ 22 157 ⫾ 22
89 ⫾ 21 12 ⫾ 2.6 112 ⫾ 33 28 ⫾ 7 23 ⫾ 8 12 ⫾ 4 0.59 ⫾ 0.09 21 (79%) 5 (19%) 9 (35%) 1/5/11/8/1 10 (38%) 143 ⫾ 20 159 ⫾ 19 167 ⫾ 26
0.01 0.005 0.47 0.08 0.04 0.05 0.45 0.15 0.12 0.12 0.23 0.005 0.26 0.32 0.22
Values are expressed as mean ⫾ SD. BP ⫽ blood pressure; BSA ⫽ body surface area.
Table 3 Gender differences Variable
Men (n ⫽ 33)
Women (n ⫽ 24)
p Value
Preoperative diastolic types C and D Ten-yr follow-up diastolic types C and D Preoperative severe dyspnea (NYHA class III or IV) Ten-yr follow-up severe dyspnea (NYHA class III or IV) Age at valve replacement (yrs) Normal coronary angiography Concomitant coronary bypass Preoperative LV mass index (g/m2) Ten-yr follow-up LV mass index (g/m2) Preoperative aortic maximum gradient (mm Hg) Ten-yr follow-up aortic maximum gradient (mm Hg) Preoperative systolic BP (mm Hg) Two-yr follow-up systolic BP (mm Hg) Ten-yr follow-up systolic BP (mm Hg)
2 (8%) 11 (42%) 14 (45%) 2 (6%) 66 ⫾ 9.5 20 (67%) 11 (33%) 165 ⫾ 36 126 ⫾ 52 103 ⫾ 27 26 ⫾ 11 139 ⫾ 20 152 ⫾ 22 154 ⫾ 22
1 (5%) 16 (84%) 14 (61%) 2 (10%) 68 ⫾ 7.3 17 (78%) 3 (12%) 155 ⫾ 44 103 ⫾ 38 94 ⫾ 31 21 ⫾ 8 147 ⫾ 25 163 ⫾ 16 167 ⫾ 24
0.99 0.005 0.25 0.99 0.42 0.40 0.11 0.44 0.20 0.25 0.06 0.26 0.04 0.04
Values are expressed as mean ⫾ SD. NYHA ⫽ New York Heart Association; other abbreviation as in Table 2.
observed at intermediate follow-up is due to a reduction in muscle fiber diameter. The reduction in fibrous content is a slow process, and at intermediate follow-up, the relative fibrous content increases, which can explain the observed increase in stiffness using pressure measurements.4 At longterm follow-up, however, the investigators found a reduction in fibrous content, and the diastolic function had normalized. We can see 2 possible reasons for the more adverse outcome in our patient group. First, our patients were older (67 vs 44 years), and they had probably been exposed to increased pressure overload for a longer period. It has been shown that older patients with severe AS have more pronounced hypertrophy, increased interstitial fibrosis, and disturbed diastolic function compared with younger patients,17 and the degree of structural change at AVR has been shown
to determine postoperative function and prognosis.18 Second, we can speculate that in elderly patients, the remodeling capacity of the fibrous content is limited, and the relative percentage of fibrous tissue will therefore increase when the muscle fiber diameter decreases. To our knowledge, there are no biopsy or autopsy data to support this. Overall, systolic blood pressure increased during the study, but the change was most pronounced during the first 2 years. The change in blood pressure was not significant in the period when the diastolic dysfunction occurred, and therefore, it is unlikely that hypertension contributed significantly to the results. We found that women were more likely to develop moderate to severe diastolic dysfunction at late follow-up than men. This was another unexpected finding. From stud-
Valvular Heart Disease/Diastolic Dysfunction in AS Patients
ies of gender differences in myocardial structure and the effects of aging19 and pressure overload,20 we would have expected a more favorable outcome in women compared with men. In a study by Villari et al20 of middle-aged subjects with AS, men were found to have more pronounced diastolic dysfunction, probably because of more disrupted myocardial structure. Studies of the normal aging process in the heart reveal a loss of muscle cells in the male but not the female heart.19 The women in the present study were more symptomatic at AVR compared with the men, and they had greater blood pressure at 2- and 10-year follow-up. This observed gender difference is interesting, but with the present study design, we were not able to elucidate whether this is an expression of the later diagnosis or referral of women to AVR, which is a known general problem in the management of cardiovascular disease,21,22 caused by untreated hypertension or gender differences in remodeling capacity. Study limitations: LV diastolic function was not evaluated using invasive catheter measurements, which is the gold standard. We used different mitral and pulmonary vein filling patterns to grade the severity of diastolic dysfunction, and although well documented, this is an indirect and categorical method.11 Because of technical difficulties, we were not able to measure the isovolumetric relaxation time or the duration of the flow reversal during atrial systole in the pulmonary vein. It is possible that pseudonormal patients have been wrongly classified as normal, and adding these parameters might have improved the grading of severity.
Acknowledgment: We would like to thank Mahmood Farasati, MSc, for his technical expertise in performing echocardiograms. The collaboration with the hospitals in western Sweden, which made this study possible, is greatly appreciated. 1. Bech-Hanssen O, Caidahl K, Wall B, Myken P, Larsson S, Wallentin I. Influence of aortic valve replacement, prosthesis type, and size on functional outcome and ventricular mass in patients with aortic stenosis. J Thorac Cardiovasc Surg 1999;118:57– 65. 2. Ikonomidis I, Tsoukas A, Parthenakis F, Gournizakis A, Kassimatis A, Rallidis L, Nihoyannopoulos P. Four year follow up of aortic valve replacement for isolated aortic stenosis: a link between reduction in pressure overload, regression of left ventricular hypertrophy, and diastolic function. Heart 2001;86:309 –316. 3. Monrad ES, Hess OM, Murakami T, Nonogi H, Corin WJ, Krayenbuehl HP. Time course of regression of left ventricular hypertrophy after aortic valve replacement. Circulation 1988;77:1345–1355. 4. Villari B, Vassalli G, Monrad ES, Chiariello M, Turina M, Hess OM. Normalization of diastolic dysfunction in aortic stenosis late after valve replacement. Circulation 1995;91:2353–2358. 5. Villari B, Vassalli G, Betocchi S, Briguori C, Chiariello M, Hess OM. Normalization of left ventricular nonuniformity late after valve replacement for aortic stenosis. Am J Cardiol 1996;78:66 –71.
727
6. Bech-Hanssen O, Wallentin I, Larsson S, Caidahl K. Reference Doppler echocardiographic values for St. Jude Medical, Omnicarbon and Biocor prosthetic valves in the aortic position. J Am Soc Echocardiogr 1998;11:466 – 477. 7. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantification in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072–1083. 8. Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol 1986;57:450 – 458. 9. Starling MR, Crawford MH, Sorensen SG, Levi B, Richards KL, O Rourke RA. Comparative accuracy of apical biplane cross-sectional echocardiography and gated equilibrium radionuclide angiography for estimating left ventricular size and performance. Circulation 1981;63: 1075–1084. 10. Teichholz LE, Kreulen T, Herman MV, Gorlin R. Problems in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence or absence of asynergy. Am J Cardiol 1976;37:7–11. 11. Rakowski H, Appelton C, Chan C-L, Dumesnil JG, Honos G, Jue J, Koilpillai C, Lepage S, Martin RP, Mercier L-A, et al. Canadian consensus recommendations for the measurement and reporting of diastolic dysfunction by echocardiography. J Am Soc Echocardiogr 1996;9:736 –760. 12. Appleton CP, Hatle LK, Popp RL. Relation of transmitral flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J Am Coll Cardiol 1988;12:426 – 440. 13. Jensen JL, Williams FE, Beilby BJ, Johnson BL, Miller LK, Ginter TL, Tomaselli-Martin G, Appleton CP. Feasibility of obtaining pulmonary venous flow velocity in cardiac patients using transthoracic pulsed wave Doppler technique. J Am Soc Echocardiogr 1997;10:60 – 66. 14. Arakawa M, Akamatsu S, Terazawa E, Dohi S, Miwa H, Kagawa K, Nishigaki K, Ito Y, Hirakawa S. Age-related increase in systolic fraction of pulmonary vein flow velocity-time integral from transesophageal Doppler echocardiography in subjects without cardiac disease. Am J Cardiol 1992;70:1190 –1194. 15. Lund O, Erlandsen M. Changes in left ventricular function and mass during serial investigations after valve replacement for aortic stenosis. J Heart Valve Dis 2000;9:583–593. 16. Kühl HFA, Puschmann D, Schöndube F, Hoffmann R, Hanrath P. Regression of left ventricular mass one year after aortic valve replacement for pure severe aortic stenosis. Am J Cardiol 2002;89: 408 – 413. 17. Villari B, Vassalli G, Schneider J, Chiariello M, Hess OM. Age dependency of left ventricular diastolic function in pressure overload hypertrophy. J Am Coll Cardiol 1997;29:181–186. 18. Lund O, Kristensen LH, Baandrup U, Hansen OK, Nielsen TT, Emertsen K, Jensen FT, Flö C, Rasmussen BS, Pilegaard HK. Myocardial structure as a determinant of pre- and postoperative ventricular function and long-term prognosis after valve replacement for aortic stenosis. Eur Heart J 1998;19:1099 –1108. 19. Olivetti G, Giordano G, Corradi D, Melissari M Lagrasta C, Gambert SR, Anversa P. Gender differences and aging: effects on the human heart. J Am Coll Cardiol 1995;26:1068 –1079. 20. Villari B, Campbell SE, Schneider J, Vassalli G, Chiariello M, Hess OM. Sex-dependent differences in left ventricular function and structure in chronic pressure overload. Eur Heart J 1995;16:1410 – 1419. 21. Bergelson B, Tommaso C. Gender differences in clinical evaluation and triage in coronary artery disease. Chest 1995;108:1510 – 1513. 22. Bowling A, Band M, McKee D, McClay M, Banning A, Dudley N, Edler A, Martin A, Blackman I. Equity in access to exercise tolerance testing, coronary angiography, and coronary bypass grafting by age, sex and clinical indications. Heart 2001;85:680 – 686.
Patients with aortic stenosis (AS) are known to have left ventricular (LV) hypertrophy and diastolic dysfunction. It has previously been shown that regression of LV hypertrophy after aortic valve replacement (AVR) is seen at intermediate follow-up, 2 to 4 years after AVR.1–3 LV hypertrophy in patients with AS consists of increased muscle fiber diameter and interstitial fibrosis,4,5 and this can explain the observed diastolic dysfunction. Endomyocardial biopsy and functional studies at intermediate and long-term follow-up have shown that the reduction of interstitial fibrosis and the reversal of diastolic dysfunction after AVR for AS is a slow process, continuing for many years.4,5 Previous reports indicate that patients after AVR experience the normalization of LV mass and diastolic function in the long term.3,4 These
a
Department of Clinical Physiology, Cardiovascular Institute, Sahlgrenska University Hospital, Göteborg; and bKarolinska Institute, Stockholm, Sweden. Manuscript received January 11, 2005; revised manuscript received and accepted April 13, 2005. This study was supported by grants from the Göteborg Medical Society, Göteborg; Sahlgrenska University Hospital, Göteborg; the Swedish Heart and Lung Foundation, Stockholm; the Västra Götaland Region, Vänersborg; the Swedish Medical Research Council, Stockholm; and the Swedish Medical Society, Stockholm, Sweden. * Corresponding author: Tel: 46-31-3421800; fax: 46-31-411735. E-mail address: [email protected] (P. Gjertsson). 0002-9149/05/$ – see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2005.04.052
studies, however, were small, and most important, the patients were relatively young and therefore not representative of the population of patients who undergo AVR. In the present study, we used Doppler echocardiography to evaluate the long-term effect on LV mass and diastolic function after AVR in a representative group of patients with severe AS. Methods Patients: From January 1991 to August 1993, a previously described population of 239 patients underwent AVR at Sahlgrenska University Hospital (Göteborg, Sweden).6 Patients with severe aortic regurgitation (grade ⬎2/4) or concomitant mitral surgery were excluded. Two years after surgery, 137 patients participated in an investigation.1 The 77 patients who were still alive 10 years after surgery were asked to participate in a new investigation; 20 of these patients were not willing to participate. The study population therefore comprised 57 patients who were prospectively investigated using echocardiography preoperatively and 2 and 10 years after surgery. At all 3 investigations, patients were asked about their symptoms (dyspnea and angina). Preoperative coronary angiographic reports were collected. Informed consent was obtained from all patients. The www.AJConline.org
Valvular Heart Disease/Diastolic Dysfunction in AS Patients
human ethics committee at Sahlgrenska University Hospital approved the study. M-mode and 2-dimensional echocardiography: Mmode measurements were made according to the American Society of Echocardiography.7 LV mass was calculated according to the corrected cube formula,8 and the LV ejection fraction was calculated either according to Simpson’s rule9 or according to Teichholz et al.10 Signs of hypertrophy were defined as increased LV mass. In patients in whom the LV ejection fraction could not be calculated because of uncertain outlining or an angulated chamber, an experienced observer made a visual estimation of the LV ejection fraction. The same method of estimating the LV ejection fraction for patients was used at all 3 examinations. Planimetry of the left atrium was performed from a late systolic stop frame with the maximum atrial area. Doppler measurements: Blood flow velocity in the LV outflow tract was estimated by pulse-wave Doppler from an apical 4-chamber view with a sample size of 5 mm. Mitral flow was recorded between the mitral leaflets in the 4-chamber view. From the mitral velocity tracings, early flow velocity (E), the deceleration time of E waves, and peak velocity during atrial systole (A) were measured. The E/A ratio was calculated. Pulmonary venous flow velocities were obtained from the upper right pulmonary vein. Peak velocities during systole (S) and diastole (D) were measured. The S/D ratio was calculated. Continuous-wave Doppler signals were recorded from multiple windows with a 2-MHz nonimaging probe. The stroke volume was calculated as the product of the crosssectional area of the LV outflow tract and the velocity–time integral. Pressure gradients were calculated according to the simplified Bernoulli equation. The systolic tricuspid gradient was estimated using continuous-wave Doppler. At the 10-year follow-up investigation, the LV isovolumic relaxation time was calculated using pulsed Doppler, with a sample size of 8 mm and the sample volume positioned anteriorly and medially from the mitral valve toward the LV outflow tract.11 Patterns describing diastolic function: Diastolic function was evaluated by integrating mitral flow and pulmonary venous information.11–13 Because the normal range for the E/A ratio is wide, deceleration time was also used to distinguish between normal diastolic function and mild diastolic dysfunction. Four different filling patterns were identified: type A: normal diastolic function (normal E/A ratio, S/D ratio, and deceleration time); type B: mild diastolic dysfunction (decreased E/A ratio or increased deceleration time and normal S/D ratio); type C: moderate diastolic dysfunction (normal E/A ratio [pseudonormalization] and decreased S/D ratio); and type D: severe diastolic dysfunction (increased E/A ratio and decreased S/D ratio). The patients were further divided into 2 groups: those with moderate to severe diastolic dysfunction (types C and D),
723
who often have increased LV filling pressure, and those with normal diastolic function or mild diastolic dysfunction (types A and B), who are likely to have normal filling pressures.11 Mitral and pulmonary venous flow patterns are age dependent.14 A healthy control group (n ⫽ 71; age range 18 to 83 years, 59% men) without hypertension or diabetes mellitus, with normal electrocardiographic results at rest and without a history of heart disease, was compared with the study group. Patients with atrial fibrillation or pacemaker rhythm were not included in the classification of diastolic function type, as well as patients from whom E/A or S/D ratios were not obtained. Statistical analysis: Continuous variables are summarized using the mean and SD. Categorical variables are summarized using absolute and relative frequencies. The mean of 3 M-mode and Doppler measurements was used in patients with sinus rhythm, and the mean of 5 measurements was used in patients with atrial fibrillation. Differences among investigations were evaluated using a paired Student’s t test. Differences among groups were evaluated using an unpaired Student’s t test. Proportions among patient groups were tested using the chi-square test and Fisher’s exact test when the sample size was small. A p value of ⬍0.05 was considered statistically significant. Diastolic function parameters are known to be age dependent, and LV mass is known to be dependent on body surface area. From the healthy controls, regression equations were calculated, as well as residuals. For each patient in the study group, the expected LV mass and E/A and S/D ratios were predicted. The observed values in patients were regarded as being reduced or increased if they differed by ⬎1.96 SD from the predicted value using the Z score.
Results Clinical characteristics: The patients were 67 ⫾ 8.6 years old at the time of surgery, 58% were men, 58% received mechanical prostheses (St. Jude Medical, Inc., St. Paul, Minnesota), and 42% received stented biologic prostheses (Biocor, St. Jude Medical). Normal findings on coronary angiography were seen in 71% of the patients. Combined coronary artery bypass grafting was performed in 25%. In relation to the AVR, echocardiography was performed a median of 1.0 days before (range 1 to 84), 2.1 years after (range 1.0 to 3.2), and finally 9.8 years after (range 7.0 to 11.7) the procedure. Severe dyspnea (New York Heart Association classes III and IV) was present in 52% preoperatively, in 4% at 2-year follow-up (p ⬍0.0001), and in 7% at 10-year follow-up (p ⫽ 0.43). Angina was present in 29% preoperatively, in 2% at 2-year follow-up (p ⬍0.0001), and in 2% at 10-year follow-up. Atrial fibrillation was present in 7% preoperatively, in 6% at 2-year follow-up (p ⫽ 0.77), and in 14% at 10-year follow-up (p ⫽ 0.12). Clinical and echocardiographic characteristics at the 3 examinations are listed in Table 1.
724
The American Journal of Cardiology (www.AJConline.org)
Table 1 Clinical and echocardiographic characteristics preoperatively and 2 and 10 years after AVR Variable
Preoperative (n ⫽ 57)
Follow-Up
142 ⫾ 22 82 ⫾ 13 69 ⫾ 13 53 (93%) 99 ⫾ 29 63 ⫾ 19 0.56 ⫾ 0.13 2.8 ⫾ 0.8 23 ⫾ 6 21 ⫾ 4 272 ⫾ 107
156 ⫾ 20* 160 ⫾ 23‡ 80 ⫾ 20 79 ⫾ 11 72 ⫾ 13* 71 ⫾ 14 52 (92%) 46 (81%) 26 ⫾ 9* 24 ⫾ 10 15 ⫾ 5* 13 ⫾ 5† 0.59 ⫾ 0.10 0.57 ⫾ 0.09 3.1 ⫾ 0.9* 2.8 ⫾ 1.1 21 ⫾ 4 22 ⫾ 6 22 ⫾ 5 26 ⫾ 7† 248 ⫾ 73 236 ⫾ 88
(p ⫽ 0.04) and the LV ejection fraction (p ⫽ 0.07). At 10-year follow-up, the LV ejection fraction and cardiac index had decreased slightly (Table 1).
Left ventricular mass: Changes in LV dimensions and LV mass index are shown in Figure 1. Preoperatively, 83% of the patients showed signs of LV hypertrophy. This had decreased to 29% (p ⬍0.0001) at 2-year follow-up. No further decrease was seen at 10-year follow-up, when 34% showed signs of LV hypertrophy (p ⫽ 0.61).
Left ventricular diastolic function: Deceleration time decreased during 10-year follow-up (Table 1), whereas the E/A ratio increased and the S/D ratio decreased (Figure 2). Figure 3 shows the distribution of diastolic types (A to D) at the 3 examinations. Forty-five patients were included in the diastolic function type classification at the preoperative investigation, 40 patients were included at 2-year follow-up, and 43 patients were included at 10-year follow-up. Only 7% of the patients had moderate to severe diastolic dysfunction (types C and D) preoperatively. There was no significant change in the percentage of estimated diastolic type between the preoperative examination and 2-year follow-up (p ⫽ 0.27). At 10-year followup, there was a marked increase in patients with moderate to severe diastolic dysfunction (61%, p ⬍ 0.0001). The tricuspid systolic gradient was increased at 10-year follow-up compared with 2-year follow-up, but the size of the left atrium was unchanged (Table 1). The characteristics of patients with moderate to severe diastolic dysfunction and normal diastolic function or mild dysfunction at 10-year follow-up are listed in Table 2. The regression equation for the E/A ratio, calculated from the control group, was 2.32 ⫺ 0.019 ⫻ age (in years), R ⫽ 0.79, and the SD for the residuals was 0.26. The regression equation for the S/D ratio was 0.46 ⫹ 0.014 ⫻ age (in years), R ⫽ 0.74, and SD for the residuals was 0.21.
Left ventricular systolic function: LV systolic function was improved at 2-year follow-up compared with the preoperative examination, with an increase in cardiac index
Gender differences: Impaired diastolic function was seen predominantly in the women (Table 2). A further gender analysis was therefore carried out (Table 3). There was no signif-
Systolic BP (mm Hg) Diastolic BP (mm Hg) Heart rate (beats/min) Sinus rhythm Maximum gradient (mm Hg) Mean gradient (mm Hg) LV ejection fraction Cardiac index (L/min/m2) Left atrium size (cm2) Tricuspid gradient (mm Hg) Deceleration time (ms)
At 2 Years (n ⫽ 57)
At 10 Years (n ⫽ 57)
Values are expressed as mean ⫾ SD. * p ⬍0.05 comparing preoperative and 2-year follow-up. † p ⬍0.05 comparing 2-year and 10-year follow-up. ‡ p ⬍0.05 comparing preoperative and 10-year follow-up. BP ⫽ blood pressure.
Figure 1. Left, LV diastolic diameter (open circles), interventricular septum (boxes), and posterior wall thickness (closed circles) preoperatively and at the 2- and 10-year follow-up. Right, LV mass index preoperatively and at the 2- and 10-year follow-up. Error bars show SDs.
Valvular Heart Disease/Diastolic Dysfunction in AS Patients
725
Figure 2. E/A ratio (left) and S/D ratio (right) preoperatively and 2 and 10 years postoperatively. Error bars show SDs.
Discussion
Figure 3. Distribution of different diastolic function types preoperatively and 2 and 10 years postoperatively: type A: normal diastolic function; type B: mild diastolic dysfunction; type C: moderate diastolic dysfunction; type D: severe diastolic dysfunction.
icant difference between men and women in terms of LV mass index or maximum gradient preoperatively or at follow-up, and there was no difference in age. A smaller percentage of women than men tended to be subjected to concomitant coronary bypass grafting. Women were more symptomatic preoperatively (61% with severe dyspnea compared with 45% of men; p ⫽ 0.25). At follow-up, women had greater systolic blood pressure.
To our knowledge, this is the first longitudinal study of a representative group of patients with AS to evaluate the regression of LV mass and change in LV diastolic function by Doppler echocardiography at intermediate and late follow-up. The most important finding was that in patients with preoperatively normal LV diastolic function or mild dysfunction, we observed moderate to severe diastolic dysfunction at 10-year follow-up, despite a reduction in LV mass index. We describe changes in heart function and LV mass occurring over a 10-year period after AVR. It is known that LV diastolic function changes with normal aging, and we therefore included a control group to define the correlation between diastolic function parameters and age. From a regression equation, we predicted the expected values and classified an observed value in a patient as normal, reduced, or increased. In the changes in type and degree of diastolic impairment described here, we have therefore corrected for the normal changes occurring with aging. We think that this is an important point that is unique to the present study and strengthens the conclusion that diastolic function actually worsens from the preoperative investigation to late follow-up after AVR. Our findings regarding the level of LV mass reduction after AVR agree with those of other investigators,1,3,5,15,16 and 2/3 of the patients achieved normal LV mass after 10 years. It was thus unexpected to find that most of the patients developed more severe diastolic dysfunction. LV hypertrophy in patients with pressure overload is due to an increase in muscle fiber diameter and an increase in fibrous content.4 Villari et al4 showed that the reduction in LV mass
726
The American Journal of Cardiology (www.AJConline.org)
Table 2 Patients with normal diastolic function or mild diastolic dysfunction (types A and B) or moderate to severe diastolic dysfunction (types C and D) 10 years after valve replacement Variable
Types A and B (n ⫽ 17)
Types C and D (n ⫽ 26)
p Value
Isovolumic relaxation time (ms) Left atrium area/BSA (cm2/m2) LV mass index (g/m2) Tricuspid gradient (mm Hg) Prosthesis maximum gradient (mm Hg) Prosthesis mean gradient (mm Hg) LV ejection fraction Normal coronary angiography Concomitant coronary bypass Bioprosthesis Distribution of valve sizes: 19 21 23 25 27 (n) Men Systolic BP preoperative (mm Hg) Systolic BP at 2 yrs (mm Hg) Systolic BP at 10 yrs (mm Hg)
107 ⫾ 17 10 ⫾ 1.7 127 ⫾ 69 23 ⫾ 5 30 ⫾ 12 15 ⫾ 6 0.57 ⫾ 0.09 10 (57%) 7 (41%) 10 (59%) 0/0/7/8/2 14 (82%) 135 ⫾ 22 152 ⫾ 22 157 ⫾ 22
89 ⫾ 21 12 ⫾ 2.6 112 ⫾ 33 28 ⫾ 7 23 ⫾ 8 12 ⫾ 4 0.59 ⫾ 0.09 21 (79%) 5 (19%) 9 (35%) 1/5/11/8/1 10 (38%) 143 ⫾ 20 159 ⫾ 19 167 ⫾ 26
0.01 0.005 0.47 0.08 0.04 0.05 0.45 0.15 0.12 0.12 0.23 0.005 0.26 0.32 0.22
Values are expressed as mean ⫾ SD. BP ⫽ blood pressure; BSA ⫽ body surface area.
Table 3 Gender differences Variable
Men (n ⫽ 33)
Women (n ⫽ 24)
p Value
Preoperative diastolic types C and D Ten-yr follow-up diastolic types C and D Preoperative severe dyspnea (NYHA class III or IV) Ten-yr follow-up severe dyspnea (NYHA class III or IV) Age at valve replacement (yrs) Normal coronary angiography Concomitant coronary bypass Preoperative LV mass index (g/m2) Ten-yr follow-up LV mass index (g/m2) Preoperative aortic maximum gradient (mm Hg) Ten-yr follow-up aortic maximum gradient (mm Hg) Preoperative systolic BP (mm Hg) Two-yr follow-up systolic BP (mm Hg) Ten-yr follow-up systolic BP (mm Hg)
2 (8%) 11 (42%) 14 (45%) 2 (6%) 66 ⫾ 9.5 20 (67%) 11 (33%) 165 ⫾ 36 126 ⫾ 52 103 ⫾ 27 26 ⫾ 11 139 ⫾ 20 152 ⫾ 22 154 ⫾ 22
1 (5%) 16 (84%) 14 (61%) 2 (10%) 68 ⫾ 7.3 17 (78%) 3 (12%) 155 ⫾ 44 103 ⫾ 38 94 ⫾ 31 21 ⫾ 8 147 ⫾ 25 163 ⫾ 16 167 ⫾ 24
0.99 0.005 0.25 0.99 0.42 0.40 0.11 0.44 0.20 0.25 0.06 0.26 0.04 0.04
Values are expressed as mean ⫾ SD. NYHA ⫽ New York Heart Association; other abbreviation as in Table 2.
observed at intermediate follow-up is due to a reduction in muscle fiber diameter. The reduction in fibrous content is a slow process, and at intermediate follow-up, the relative fibrous content increases, which can explain the observed increase in stiffness using pressure measurements.4 At longterm follow-up, however, the investigators found a reduction in fibrous content, and the diastolic function had normalized. We can see 2 possible reasons for the more adverse outcome in our patient group. First, our patients were older (67 vs 44 years), and they had probably been exposed to increased pressure overload for a longer period. It has been shown that older patients with severe AS have more pronounced hypertrophy, increased interstitial fibrosis, and disturbed diastolic function compared with younger patients,17 and the degree of structural change at AVR has been shown
to determine postoperative function and prognosis.18 Second, we can speculate that in elderly patients, the remodeling capacity of the fibrous content is limited, and the relative percentage of fibrous tissue will therefore increase when the muscle fiber diameter decreases. To our knowledge, there are no biopsy or autopsy data to support this. Overall, systolic blood pressure increased during the study, but the change was most pronounced during the first 2 years. The change in blood pressure was not significant in the period when the diastolic dysfunction occurred, and therefore, it is unlikely that hypertension contributed significantly to the results. We found that women were more likely to develop moderate to severe diastolic dysfunction at late follow-up than men. This was another unexpected finding. From stud-
Valvular Heart Disease/Diastolic Dysfunction in AS Patients
ies of gender differences in myocardial structure and the effects of aging19 and pressure overload,20 we would have expected a more favorable outcome in women compared with men. In a study by Villari et al20 of middle-aged subjects with AS, men were found to have more pronounced diastolic dysfunction, probably because of more disrupted myocardial structure. Studies of the normal aging process in the heart reveal a loss of muscle cells in the male but not the female heart.19 The women in the present study were more symptomatic at AVR compared with the men, and they had greater blood pressure at 2- and 10-year follow-up. This observed gender difference is interesting, but with the present study design, we were not able to elucidate whether this is an expression of the later diagnosis or referral of women to AVR, which is a known general problem in the management of cardiovascular disease,21,22 caused by untreated hypertension or gender differences in remodeling capacity. Study limitations: LV diastolic function was not evaluated using invasive catheter measurements, which is the gold standard. We used different mitral and pulmonary vein filling patterns to grade the severity of diastolic dysfunction, and although well documented, this is an indirect and categorical method.11 Because of technical difficulties, we were not able to measure the isovolumetric relaxation time or the duration of the flow reversal during atrial systole in the pulmonary vein. It is possible that pseudonormal patients have been wrongly classified as normal, and adding these parameters might have improved the grading of severity.
Acknowledgment: We would like to thank Mahmood Farasati, MSc, for his technical expertise in performing echocardiograms. The collaboration with the hospitals in western Sweden, which made this study possible, is greatly appreciated. 1. Bech-Hanssen O, Caidahl K, Wall B, Myken P, Larsson S, Wallentin I. Influence of aortic valve replacement, prosthesis type, and size on functional outcome and ventricular mass in patients with aortic stenosis. J Thorac Cardiovasc Surg 1999;118:57– 65. 2. Ikonomidis I, Tsoukas A, Parthenakis F, Gournizakis A, Kassimatis A, Rallidis L, Nihoyannopoulos P. Four year follow up of aortic valve replacement for isolated aortic stenosis: a link between reduction in pressure overload, regression of left ventricular hypertrophy, and diastolic function. Heart 2001;86:309 –316. 3. Monrad ES, Hess OM, Murakami T, Nonogi H, Corin WJ, Krayenbuehl HP. Time course of regression of left ventricular hypertrophy after aortic valve replacement. Circulation 1988;77:1345–1355. 4. Villari B, Vassalli G, Monrad ES, Chiariello M, Turina M, Hess OM. Normalization of diastolic dysfunction in aortic stenosis late after valve replacement. Circulation 1995;91:2353–2358. 5. Villari B, Vassalli G, Betocchi S, Briguori C, Chiariello M, Hess OM. Normalization of left ventricular nonuniformity late after valve replacement for aortic stenosis. Am J Cardiol 1996;78:66 –71.
727
6. Bech-Hanssen O, Wallentin I, Larsson S, Caidahl K. Reference Doppler echocardiographic values for St. Jude Medical, Omnicarbon and Biocor prosthetic valves in the aortic position. J Am Soc Echocardiogr 1998;11:466 – 477. 7. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantification in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072–1083. 8. Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol 1986;57:450 – 458. 9. Starling MR, Crawford MH, Sorensen SG, Levi B, Richards KL, O Rourke RA. Comparative accuracy of apical biplane cross-sectional echocardiography and gated equilibrium radionuclide angiography for estimating left ventricular size and performance. Circulation 1981;63: 1075–1084. 10. Teichholz LE, Kreulen T, Herman MV, Gorlin R. Problems in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence or absence of asynergy. Am J Cardiol 1976;37:7–11. 11. Rakowski H, Appelton C, Chan C-L, Dumesnil JG, Honos G, Jue J, Koilpillai C, Lepage S, Martin RP, Mercier L-A, et al. Canadian consensus recommendations for the measurement and reporting of diastolic dysfunction by echocardiography. J Am Soc Echocardiogr 1996;9:736 –760. 12. Appleton CP, Hatle LK, Popp RL. Relation of transmitral flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J Am Coll Cardiol 1988;12:426 – 440. 13. Jensen JL, Williams FE, Beilby BJ, Johnson BL, Miller LK, Ginter TL, Tomaselli-Martin G, Appleton CP. Feasibility of obtaining pulmonary venous flow velocity in cardiac patients using transthoracic pulsed wave Doppler technique. J Am Soc Echocardiogr 1997;10:60 – 66. 14. Arakawa M, Akamatsu S, Terazawa E, Dohi S, Miwa H, Kagawa K, Nishigaki K, Ito Y, Hirakawa S. Age-related increase in systolic fraction of pulmonary vein flow velocity-time integral from transesophageal Doppler echocardiography in subjects without cardiac disease. Am J Cardiol 1992;70:1190 –1194. 15. Lund O, Erlandsen M. Changes in left ventricular function and mass during serial investigations after valve replacement for aortic stenosis. J Heart Valve Dis 2000;9:583–593. 16. Kühl HFA, Puschmann D, Schöndube F, Hoffmann R, Hanrath P. Regression of left ventricular mass one year after aortic valve replacement for pure severe aortic stenosis. Am J Cardiol 2002;89: 408 – 413. 17. Villari B, Vassalli G, Schneider J, Chiariello M, Hess OM. Age dependency of left ventricular diastolic function in pressure overload hypertrophy. J Am Coll Cardiol 1997;29:181–186. 18. Lund O, Kristensen LH, Baandrup U, Hansen OK, Nielsen TT, Emertsen K, Jensen FT, Flö C, Rasmussen BS, Pilegaard HK. Myocardial structure as a determinant of pre- and postoperative ventricular function and long-term prognosis after valve replacement for aortic stenosis. Eur Heart J 1998;19:1099 –1108. 19. Olivetti G, Giordano G, Corradi D, Melissari M Lagrasta C, Gambert SR, Anversa P. Gender differences and aging: effects on the human heart. J Am Coll Cardiol 1995;26:1068 –1079. 20. Villari B, Campbell SE, Schneider J, Vassalli G, Chiariello M, Hess OM. Sex-dependent differences in left ventricular function and structure in chronic pressure overload. Eur Heart J 1995;16:1410 – 1419. 21. Bergelson B, Tommaso C. Gender differences in clinical evaluation and triage in coronary artery disease. Chest 1995;108:1510 – 1513. 22. Bowling A, Band M, McKee D, McClay M, Banning A, Dudley N, Edler A, Martin A, Blackman I. Equity in access to exercise tolerance testing, coronary angiography, and coronary bypass grafting by age, sex and clinical indications. Heart 2001;85:680 – 686.