Tissue Doppler imaging (TDI) has improved the usefulness

Transcription

1 Reference Values and Distribution of Conventional Echocardiographic Doppler Measures and Longitudinal Tissue Doppler Velocities in a Population Free From Cardiovascular Disease Havard Dalen, MD; Anders Thorstensen, MD; Lars J. Vatten, MD, PhD; Svein A. Aase, MSc, PhD; Asbjorn Stoylen, MD, PhD Background This study aimed to investigate the distribution of conventional Doppler measurements, pulsed wave tissue Doppler imaging (pwtdi)- and color tissue Doppler imaging-derived velocities, by age and sex in a healthy population. Methods and Results Longitudinal tissue Doppler velocities were determined in 1266 healthy individuals from standard apical 4- and 2-chamber views. By the pwtdi method, mean SD systolic mitral annular velocities were cm/s in women and cm/s in men, and by color tissue Doppler, they were cm/s in women and cm/s in men. With pwtdi, diastolic early mitral annular velocities were cm/s in women and cm/s in men, with corresponding ratios between mitral early flow velocity and early diastolic tissue velocity of in women and in men. By pwtdi, tricuspid annular systolic and early diastolic velocities were and , respectively, in women and and , respectively, in men. There was a significant decrease in left and right ventricular function with increasing age. Conclusions The longitudinal mitral and tricuspid annular velocities from this population study are widely applicable as reference values. Reference values for annular velocities should be specified by sex and age. The average of inferoseptal and anterolateral wall velocities may be the preferred index of left ventricular performance. (Circ Cardiovasc Imaging. 2010;3: ) Key Words: aging diastole myocardium sex systole Tissue Doppler imaging (TDI) has improved the usefulness of echocardiography in detecting subclinical heart failure. 1 3 Annular tissue velocities based on spectral pulsed wave TDI (pwtdi) and color tissue TDI (ctdi) can be used to quantify left ventricular (LV) and right ventricular (RV) longitudinal function, and the ratio between mitral early flow velocity (E) and early diastolic tissue velocity (e ) is used as an indicator of LV filling pressure. 4 The feasibility of annular tissue velocities is superior to deformation imaging, but both pwtdi and ctdi are susceptible to translational myocardial motion, and determination of regional dysfunction is not feasible. 2,5 Clinical Perspective on p 622 Small clinical studies have been used to determine pathological thresholds of myocardial function, but reference values of myocardial function in healthy individuals require data from large, population-based studies. Mitral annular velocities using ctdi recently have been reported from healthy control groups, 6,7 but mitral and tricuspid annular velocities obtained by pwtdi only have been reported from small studies The distribution of mitral annular velocities by myocardial walls has not been described in population studies, and comparisons of ctdi- and pwtdi-derived velocities by age have not shown consistent results Therefore, we used echocardiography to describe the distribution of ctdi- and pwtdi- derived longitudinal mitral and pwtdi tricuspid annular velocities in a population of 1266 subjects without known cardiovascular disease, hypertension, or diabetes. We also assessed the relationship between LV velocities and conventional Doppler measurements by sex and age. The main aim of the study was to provide population-based reference values for tricuspid and mitral annular velocities, including the corresponding E/e ratios. Methods Study Population The Nord-Trøndelag Health (HUNT) study 11 in Nord-Trøndelag County, Norway, was initiated in the 1980s, and the third wave was Received November 24, 2009; accepted June 17, From the Department of Circulation and Medical Imaging (H.D., A.T., S.A.A., A.S.) and Department of Public Health (L.J.V.), Norwegian University of Science and Technology, Trondheim, Norway; Department of Medicine (H.D.), Levanger Hospital, Nord-Trøndelag Health Trust, Nord-Trøndelag, Norway; and Department of Cardiology (A.S.), St Olavs Hospital/Trondheim University Hospital, Trondheim, Norway. The online-only Data Supplement is available with this article at Correspondence to Havard Dalen, MD, Department of Circulation and Medical Imaging, NTNU, MTFS, N-7489 Trondheim, Norway. havard.dalen@ntnu.no 2010 American Heart Association, Inc. Circ Cardiovasc Imaging is available at DOI: /CIRCIMAGING

2 Dalen et al The HUNT Study in Norway 615 Figure 1. A, The pwtdi recording from the inferoseptal mitral annular site. B, The ctdi velocity curves assessed from the mitral annular inferoseptal (yellow curve and arrows) and anterolateral (green curve and arrows) sites. Peak S, peak e, and a are marked. conducted from 2006 to A total of people were invited, and (54%) participated. Within the third wave, we conducted an echocardiography study among a random sample of participants in predetermined communities in the county. 12 The participants had to be free from known cardiovascular disease, diabetes, and hypertension. They then were selected by random sampling. Exclusion criteria were arrhythmias, valvular pathology, or any kind of myocardial pathology. 12 Among the selected participants who consented, 30 were excluded later due to significant pathology that was disclosed during the echocardiographic examination. 12 Thus, this echocardiography study is based on examinations of 1266 men and women. The study was approved by the Regional Committee for Medical Research Ethics and conducted according to the second Declaration of Helsinki. Written informed consent was obtained. Image Acquisition One experienced physician-echocardiographer (H.D.) conducted all examinations according to American Society of Echocardiography and European Society of Echocardiography (ASE/EAE) recommendations. 13,14 Participants were examined in the left-lateral decubitus position with a Vivid 7 scanner version BT06 using a phased-array transducer (M3S and M4S). The echo-doppler examination included parasternal long- and short-axis views and 3 standard apical views. For each view, at least 3 consecutive cardiac cycles were recorded during quiet respiration. From the 3 apical planes, separate grayscale and ctdis were recorded. Color tissue Doppler mode was recorded at a mean frame rate of 100 seconds 1 with underlying gray-scale images at a mean frame rate of 25 seconds 1. Separate gray-scale recordings were optimized for evaluation of the LV at a mean frame rate of 44 seconds 1. Color tissue Doppler recordings were performed to exclude valvular insufficiencies. Echocardiographic data were stored digitally and subsequently analyzed. Blood flow Doppler recordings were done as described in the ASE/EAE recommendations. 15 The pwtdi mitral annular velocities were acquired from the base of the inferoseptal, anterolateral, inferior, and anterior walls. Tricuspid annular velocity was obtained from the RV free wall in the 4-chamber view after alignment to the direction of the ultrasound beam. Sample volume was positioned within 1 cm of insertion sites of the valve leaflets and adjusted to cover the longitudinal excursion of the mitral and tricuspid annulus in both systole and diastole. All measurements reflected the average of 3 cardiac cycles during quiet respiration. 14 Analysis of Conventional Echocardiography The unique aspects of the image acquisition and analysis have been published recently. 12,16 One experienced physician-echocardiographer (H.D.) performed all analyses according to the ASE/EAE recommendations. 13,14 Interventricular septal and posterior wall thickness, fractional shortening, and LV internal dimension were analyzed on parasternal M-mode echocardiograms with the ultrasound beam at the tip of the mitral leaflet. Pulmonary venous Doppler recordings were analyzed for systolic and diastolic velocities, and the systole/diastole ratio was calculated. Mitral flow was analyzed for E and atrial (A) diastolic filling and deceleration time (DT) of the E wave, and the E/A ratio was calculated. Isovolumic relaxation time (IVRT) was measured from the start of the aortic valve closure signal to the start of mitral flow. LV outflow tract (LVOT) velocities were measured with tracing of the pwtdi spectrum with low gain setting. Analyses of TDI Velocities LV longitudinal function was assessed by myocardial velocities in the inferoseptal, anterolateral, inferior, and anterior mitral annular positions. All echocardiograms were analyzed for pwtdi-derived mitral and tricuspid annular systolic velocity (S ), e, and late diastolic velocity (a ). 14 Measurements were done at the peak of the upper edge of the solid Doppler curve, 17 with scale optimized and low gain settings (Figure 1). Color tissue Doppler recordings from the standard apical 4- and 2-chamber views also were analyzed to assess mitral annular systolic velocities in all echocardiograms using customized semiautomatic software that runs on a Matlab platform. The method calculates the mean velocities of the sample volume by the same algorithm as the commercial software (EchoPAC). TDI velocity curves were extracted from regions of 5 mm 1 beam positioned within 1 cm of the insertion of the mitral leaflets and tracked through the cardiac cycle by tissue Doppler along the ultrasound beams and speckle tracking perpendicular to the ultrasound beams. Peak systolic velocity was set as the highest velocity obtained during the ejection phase. Peak diastolic TDI velocities were minimum negative values during early and late diastole but converted to absolute values. Comparison of TDI Methods The ctdi recordings also were analyzed by the EchoPAC software in a subset of 100 participants for mitral annular S, e, and a. Circular sample volume (6 mm) was positioned as described earlier, and peak velocities were measured (Figure 1). Temporal smoothing was 30 milliseconds. Comparisons of mitral annular S assessed by pwtdi and ctdi were done for the population as a whole, where ctdi was analyzed by customized software (GcMat), and in the subset of 100 participants, where ctdi was analyzed by both customized and commercial software. Thus, comparisons of diastolic velocities between pwtdi and ctdi software (both EchoPAC) were done within the subset of participants. Data Reproducibility Two physician-echocardiographers (H.D., A.T.) were blinded to each other s recordings and conducted separate echocardiographic

3 616 Circ Cardiovasc Imaging September 2010 Table 1. Characteristics of the Study Population Female Sex Male Sex No Age, y Height, cm Weight, kg Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg Heart rate, bpm Daily and occasional smokers, % Conventional echocardiographic measurements Interventricular septum diastolic thickness, mm End-diastolic LV internal diameter, mm Ejection fraction, Teicholtz, % LV posterior wall diastolic thickness, mm Fractional shortening, % Data are presented as mean SD, unless otherwise indicated. There was a significant difference between sexes for all displayed parameters with P 0.001, except for ejection fraction (P 0.04) and fractional shortening (nonsignificant). acquisitions (a total of 20 echocardiograms) on 10 healthy volunteers (3 women; age range, 24 to 36 years), all with adequate echocardiographic images. Recordings were analyzed for both conventional and tissue Doppler measurements. Data from these acquisitions were used to test interobserver variability, and the echocardiographers reanalyzed their own recordings in scrambled order after approximately 3 weeks to test intraobserver variability. Additional data on reproducibility of both conventional echocardiographic data and tissue Doppler indices have been published recently. 12,16 Statistical Analysis Mean differences were tested by Student t test, and differences between age groups and between myocardial walls were tested by linear mixed-effects models using post hoc Bonferroni corrections. Mitral annular velocities obtained by different methods were compared and tested by linear mixed-effects models with a post hoc least significant difference method to enable detection of even smaller differences among the 3 methods. P values are reported to indicate degree of significance. Coefficients of repeatability (COR) were calculated according to the Bland-Altman method, 18 and coefficients of variation (COV) were calculated as the within-subject SD divided by the mean of the observations. Association between mitral and tricuspid annular velocities (dependent variables) and age was tested in a multivariate regression analysis with adjustment for body surface area, end-diastolic LV internal diameter, heart rate, and systolic or diastolic blood pressure. All statistical analyses were performed with SPSS for Windows versions 15.0 and Results Study Population Table 1 and online-only Data Supplement table I show basic characteristics of the 1266 participants (663 women and 603 men) who were included in the analyses. Additional basic characteristics have been published recently. 12 Mean SD age was years among women and years among men (Figure 2). Study participants were divided according to age group ( 40, 40 to 59, and 60 years) and sex (Table 2). Women in the middle age group (40 to 59 years) were Figure 2. Age and sex of the study participants. Red columns refer to women and blue to men. slightly older than the men (mean difference, 1.0 year; P 0.03), but age did not differ by sex in the 2 other groups. Image Acquisition A total of 1266 echocardiograms were analyzed for conventional Doppler and TDI measurements. The feasibility of all displayed measurements was at least 96% (Tables 2 and 3). Doppler Flow Table 2 shows Doppler flow indices. Differences were highly significant between the youngest and oldest age groups for all displayed parameters and in both sexes (all P 0.002, except for LVOT measurements in both sexes). The pulmonary venous systole/diastole ratio did not differ by sex, but other indices showed consistently higher velocities in women (LVOT peak velocity and pulmonary diastolic velocity, P 0.05; others, P 0.001). Mitral Annular Velocities Mitral annular velocities were normally distributed (Figure 3), and Table 3 shows mean mitral annular S,e, and a by pwtdi and S by ctdi. All parameters differed by sex and age (all P 0.001), with a consistent decrease in S and e by age and increase in a by age. In multivariate regression analysis, age was a significant predictor of both mitral and tricuspid annular velocities after adjusting for body surface area, end-diastolic LV internal diameter, heart rate, and systolic or diastolic blood pressure. S and a were higher in men than in women (P 0.001), and e was higher in women. Supplemental figure I shows 95% CIs of S and e by age. Table 4 shows S, e, and a overall and by each wall. S values were highest in the anterolateral and inferior wall, and e was highest in the anterolateral wall. S and e were lowest in the inferoseptal annular site. The age and sex dependency of systolic and diastolic mitral annular velocities was consistent for all walls (all P 0.001), and S and e correlated in both the LV (r 0.58; P 0.001) and the RV (r 0.52; P 0.001).

4 Dalen et al The HUNT Study in Norway 617 Table 2. Conventional Doppler Measurements by Age and Sex Mitral E, cm/s Mitral A, cm/s E/A Ratio DT, ms IVRT, ms Pulm S, cm/s Pulm D, cm/s Pulm S/D Ratio LVOT Vmax, m/s LVOT VTI, cm Female sex Feasibility, no. (%) 657 (99) 657 (99) 657 (99) 657 (99) 653 (98) 646 (98) 637 (96) 635 (96) 654 (99) 654 (99) 40y(n 208) y (n 336) y(n 119) All (n 663) Male sex Feasibility, no. (%) 599 (99) 599 (99) 599 (99) 599 (99) 597 (99) 583 (97) 583 (97) 578 (96) 591 (98) 591 (98) 40y(n 126) y (n 327) y(n 150) All (n 603) Data are presented as mean SD for Doppler flow velocities from 1266 healthy individuals by sex and age, unless otherwise indicated. Level of significance specified in the Doppler Flow subsection of the Results section. Pulm D indicates peak diastolic pulmonary velocity; Pulm S, peak systolic pulmonary velocity; Vmax, peak velocity; VTI, velocity time integral. Table 5 shows that the E/e ratio increased by age overall and for each myocardial wall (all P 0.001). Averaging the E/e ratio from the inferoseptal and anterolateral wall, or from all 4 walls, gave similar results. However, as e differed between walls, the E/e ratio was significantly lower if e was measured in the anterolateral wall only and significantly higher when measured in the inferoseptal wall only (both P 0.001). Overall, the E/e ratio based on pwtdi measurements was 6.7 in women and 6.4 in men (P 0.001). E/e ratios increased with age ( 40, 40 to 59, and 60 years) in both sexes (all P 0.001); the respective ratios (SD) among age groups were 5.6 (1.3), 6.8 (1.8), and 8.7 (2.8) in women and 5.5 (1.5), 6.2 (1.6), and 7.7 (2.3) in men. The differences by age were significant for each wall (all P 0.001). Tricuspid Annular Velocities Data were normally distributed, and Table 3 shows mean tricuspid S, e, and a as assessed by pwtdi. In relation to Table 3. Age- and Sex-Specific Mean Annular Velocities by pwtdi and ctdi LV (Mean of 4 Walls) LV, S and e decreased by age, and there was an increase by age in a (all P 0.03). Mean tricuspid annular S values were higher in men than in women, and e values were higher in women (both P 0.003). However, there was no sex difference for a. The sex differences for RV velocities were not consistent across age groups. Supplemental figure II shows the 95% CIs of S and e by age. For all pwtdi indices, the velocities were higher in the RV free wall than in the mitral annulus, and these differences were consistent across age groups (all P 0.001). Comparison of Methods Table 6 shows that the ctdi methods (EchoPAC and GcMat) measured lower mitral annular velocities than the pwtdi for all myocardial walls (all P 0.001), but there were no differences between the 2 ctdi methods (P 0.17). However, there were slight differences between the 2 ctdi methods when data were analyzed per wall (P 0.08). RV (Free Wall) S (pwtdi) S (ctdi) e (pwtdi) a (pwtdi) S (pwtdi) e (pwtdi) a (pwtdi) Female sex Feasibility, no. (%) 652 (98%) 657 (99%) 652 (98%) 652 (98%) 648 (98%) 648 (98%) 648 (98%) 40 y, cm/s y, cm/s y, cm/s All, cm/s Male sex Feasibility, no. (%) 590 (98%) 601 (99%) 590 (98%) 590 (98%) 586 (97%) 586 (97%) 586 (97%) 40 y, cm/s y, cm/s y, cm/s All, cm/s Data are presented as mean SD tissue Doppler velocities from 1266 healthy individuals for the stratified age groups, unless otherwise indicated. Level of significance is specified in the Mitral Annular Velocities and Tricuspid Annular Velocities subsections of the Results section.

5 618 Circ Cardiovasc Imaging September 2010 Figure 3. A, Distribution of mitral annular S by ctdi. B, Distribution of mitral annular S by pwtdi. C, Distribution of mitral annular e by pwtdi. D, Distribution of mitral annular a by pwtdi. Table 5 (2 right columns) shows the influence of the TDI method on the E/e ratio, with a higher estimate (mean difference, 2.2; P 0.001) by ctdi than by pwtdi. These differences were consistent across the 4 LV walls (all P 0.001). Measuring mean mitral annular velocities by averaging the inferoseptal and anterolateral wall and by averaging velocities for the 4 myocardial walls yielded similar results both in the population as a whole and in the subset of 100 study participants. The correlation between pwtdi and ctdi was high in our data (r 0.85; P 0.001). Data Reproducibility Interobserver COV for the Doppler flow measurements ranged from 6% to 14% when analyzed on separate echocardiograms, and intraobserver variability ranged from 1% to 4%. Late mitral inflow, DT, IVRT, and pulmonary vein systole and diastole showed the highest variability. Interobserver COV was 6% for mitral S,e, and a averaged for the 4 walls and 6% to 13% for each wall separately. The intraobserver COV was 2% and 2% to 4%, respectively. Inter- and intraobserver COV for tricuspid annular velocities were similar to the single-wall measurements presented for LV. Interobserver COR for mitral annular S,e, and a was 1.7, 3.4, and 2.1 cm/s, respectively. The tricuspid annulus interobserver COR for S, e, and a was 4.2, 5.0, and 2.2 cm/s, respectively. The pwtdi and the ctdi method had similar reproducibility. The complete data and graphs of the reproducibility study have been published recently. 12,16 Discussion In this population-based study of 1266 randomly selected men and women without cardiovascular disease, hypertension, and diabetes, we have described and compared the distribution of conventional and TDI measurements by age

6 Dalen et al The HUNT Study in Norway 619 Table 4. Tissue Doppler Velocities by LV Myocardial Wall S (pwtdi), cm/s* S (ctdi), cm/s e (pwtdi), cm/s* a (pwtdi), cm/s Mean Inferoseptum Anterolateral Inferior Anterior Mean SD velocities overall and by myocardial walls of the LV from 1266 study participants for S by pwtdi and ctdi and e and a by pwtdi. Significance (linear mixed-effects models, post hoc Bonferroni correction) of difference between walls. *All P All P 0.001, except lateral versus inferior (nonsignificant). All P 0.001, except lateral versus anterior (nonsignificant). and sex. Apart from the wide applicability as reference values, the results show a clear age-related reduction in cardiac performance measured by pwtdi, ctdi, and conventional Doppler. A conventional definition of normal values in a healthy population is to use 2 SD of the mean (corresponding to approximately 95% of the population). Based on our results for pwtdi S, the normal range would be 5.6 to 10.8 cm/s in women and 5.8 to 11.4 cm/s in men. Measured by ctdi, the corresponding ranges would be 4.4 to 8.8 cm/s for women and 4.3 to 9.5 cm/s for men. For e, the normal ranges by pwtdi would be 5.4 to 18.2 cm/s in women and 4.8 to 16.8 cm/s in men. Tricuspid ranges measured by pwtdi for S and e would be 8.7 to 16.3 cm/s and 7.3 to 19.3 cm/s, respectively, in women and 8.4 to 17.2 cm/s and 5.9 to 19.1 cm/s, respectively, in men. Table 5. E/e Ratio by Age, Wall, and Method pwtdi* Doppler Indices in the Study Population The distribution of blood flow and tissue Doppler indices followed a normal distribution, and tissue Doppler indices gave significantly higher values in the RV than in the mitral annulus. The results showed a decrease in cardiac function by increasing age as measured by both blood flow and tissue Doppler in both sexes. There was a general decrease in annular S and e and a consistent increase in a. These findings apply to both ventricles. The reduction in atrioventricular annular velocities by increasing age was most evident in e. It might be suggested that e values are more closely related to maximal cardiac function than the systolic resting velocities, 19,20 and it is possible that stress testing would have given more consistent results between the different indices. The difference by sex was less clear in tricuspid annular velocities compared to mitral annular velocities. Heart rate was significantly higher in women, and this might have influenced the results. However, the difference by sex remained significant after adjustment for size, heart rate, and blood pressure in multivariate regression analysis. The lower number of subjects within the oldest age group might also contribute to a less-pronounced sex difference in tricuspid annular velocities in that age group. Previous studies of RV systolic function have not shown a consistent reduction with age, but these studies were not powered to detect small differences. 8,9,21,22 A recent publication showed a significant decrease of RV systolic function as assessed by pwtdi by age. 23 To our knowledge, sex- and age-dependent data on tricuspid annular velocities in a large population have not been reported previously. The agedependent decrease in LV systolic and diastolic function was highly significant across all myocardial walls, using both TDI methods. Previous studies also have suggested that diastolic flow indices depend on age. 10,22,24 26 In the Tromsø Heart Study, one of the largest population-based echocardiography studies, mitral inflow velocities and IVRT were similar to the findings of the present study, but DT differed slightly between the studies. 25,26 The ctdi values were in line with those of others, 6,7,22,27 but the pwtdi-assessed mitral annular velocities were lower than those reported from other smaller studies. 8,14,28 However, diastolic pwtdi indices were in line with those reported previously in 80 healthy individuals. 10 Systolic and late diastolic TDI values were lower in women (except for a in RV), but early diastolic values were lower in men than in women, both in the LV and in the RV. Similar results have been reported for LV, 7 but data for RV are scarce. 22 Gain settings and localization of the measurement in the Doppler spectrum are the most crucial points regarding pwtdi analysis. Further, misalignment of the ultrasound beam will cause underestimation of the tissue velocity. In earlier studies, gain 40 y y 60 y All pwtdi ctdi No Mean Mean septum-lateral Inferoseptum Anterolateral Inferior Anterior Level of significance specified in the Results section. *Data presented as mean SD E/e with separate e values by the 4 different myocardial walls and mean of 2 and all 4 walls separated by age, unless otherwise indicated. Data are presented as mean SD E/e for the 2 methods compared in the subset of 100 study participants.

7 620 Circ Cardiovasc Imaging September 2010 Table 6. Mitral Annular Velocities From pwtdi and ctdi in a Subset of 100 Study Participants Method S, cm/s pwtdi S, cm/s ctdi (EP) S, cm/s ctdi (Gc) e, cm/s pwtdi e, cm/s ctdi (EP) a, cm/s pwtdi a, cm/s ctdi (EP) Mean Inferoseptum Anterolateral Inferior Anterior Data are presented as mean SD. Level of significance specified in the Comparison of Methods subsection of the Results section. Gc indicates GcMat software; EP, EchoPAC software. settings and localization in the Doppler spectrum used for measurement were rarely completely specified. 8,14 We used low gain settings and measured at the outer edge of the solid spectrum. 17 Different gain settings may be one reason why other pwtdi studies have reported higher mitral annular velocities than we have in the present study. 8,28,29 In accordance with previous findings, the results differed between LV walls. 14,22 It has earlier been hypothesized that orientation of myocardial fibers are most longitudinal in the lateral and inferior wall, and this explains the finding of highest myocardial velocities in these regions. 22 However, wall stress and fibroelastic properties also might be relevant. These 2 walls are not that affected by RV and septal function, which also might play a role. Accordingly, the E/e ratio depended on the site of the mitral annulus where e was measured. Thus, single-wall measurements should be related to wall-specific normal values. However, because the septal and lateral wall gave the same mean as the average of the 4 walls, normal values may be interchangeable. In other studies, only overall mitral annular velocities have been reported. 6,7 In a study by Kuznetsova et al, 6 it was noted that e was highest in the anterolateral wall, but data were not shown. In our study, LV e values were highest in the anterolateral site and lowest in the inferoseptal site. Thus, the estimated E/e ratio was highest when e was assessed from the inferoseptal site and lowest when e was assessed from the anterolateral site. The results of averaging mitral annular velocities obtained from the 4 myocardial walls did not differ from the average of using velocities measured at the inferoseptal and anterolateral sites. This finding was consistent for both the systolic and the diastolic indices as well as for the E/e ratio estimates. In the evaluation of diastolic LV function, the ASE/EAE recommended averaging the indices from at least the inferoseptal and anterolateral sites. 14 Using 2 sites is attractive because of feasibility and time efficiency, but averaging indices from 4 instead of 2 walls will increase test-retest reproducibility. The pwtdi velocities and E/e ratio have been validated against cardiac catheterization, resulting in the recommended E/e ratio limits of 8 for normal LV filling pressures and 15 for elevated LV filling pressures. 4,30,31 The validation studies were small, and the use of the E/e ratio alone as an index of filling pressure may not be appropriate in healthy individuals. 14,32 In our study, a substantial proportion had an E/e ratio above the lower limit of 8 but did not fulfill other echocardiographic criteria of diastolic dysfunction, 14,32 suggesting that 8 may not be a useful limit in this setting. The E/e ratio also is age dependent, and in women, reference limits based on the definitions presented earlier would give a normal range of 3.1 to 8.1 in the youngest age group, 3.2 to 10.3 in the middle age group, and 3.2 to 14.3 in the oldest age group in the present study. In men, the corresponding limits would be 2.5 to 8.4, 2.9 to 9.5, and 3.1 to 12.4, respectively. This is in line with the findings of De Sutter et al. 33 Using the ctdi method would result in larger ratios, and although both E and e will decrease with age, the ratio will increase, and recommended normal ranges, therefore, need to be stratified by age. In our data, SDs were only modestly reduced when averaging TDI measurements of all 4 walls compared to using TDI wall by wall. In the reproducibility study, the COV was 6% to 13% for single walls and 6% for the average of 4 walls. In an earlier study, variability was 25% higher by single walls than for the 4-wall average. 5 The large number of study participants reduces measurement error, and most of the variability may reflect biological variability. However, in daily clinical practice, averaging will yield more-robust measurements. Comparison of Methods The pwtdi and ctdi methods are fundamentally different. On the basis of previous validation studies, the pwtdi method is recommended for diastolic quantification of the LV. 4,14,30,34 However, recent studies of healthy individuals have used the ctdi method, which has shown good correlation against pro-brain natriuretic peptide 6 and mortality. 35,36 Thus, ctdi may be equally valuable, and a revision of the current recommendations on LV diastolic evaluation may be appropriate. 14 Compared to pwtdi, ctdi resulted in 1.7-cm/s lower values corresponding to 20% lower systolic velocities and 2.5-cm/s lower values corresponding to 23% lower diastolic mitral annular velocities. Both methods seem feasible, but measurements should refer to the method-specific normal values. Gain setting is crucial for spectral analysis by the pwtdi method. We minimized gain settings and measured S,e, and a at the peak of the outer edge of the solid spectral Doppler curve, which reduces maximal velocities by 1 to 2 cm/s compared to using high gain. 29 By all Doppler methods, inappropriate alignment of the ultrasound beam will cause underestimation. CORs for test-retest analyses ranged 1.7 to 3.4 cm/s for averaged systolic mitral annular velocities. The corresponding COR was significantly higher in single-wall analyses and

8 Dalen et al The HUNT Study in Norway 621 significantly lower in similar analyses performed on the same echocardiograms. 16 Thus, all presented analyses show clinically acceptable reproducibility. Study Limitations The study sample was a healthy population free from known cardiovascular disease, hypertension, and diabetes, and subjects with significant echocardiographic pathology were excluded from the analyses. All participants were of white northern European descent. However, myocardial deformation by race was reported from the Multi-Ethnic Study of Atherosclerosis, and showed that race was not associated with LV performance measured by cardiac-tagged MRI. 37 Conclusions In this large, population-based study of healthy individuals, we present reference values for conventional Doppler indices, pwtdi and ctdi velocities of LV function and pwtdi velocities from the tricuspid annulus according to age and sex. The absolute velocities differ between Doppler methods, but results were highly correlated, and using average values from the inferoseptal and anterolateral myocardial walls will give a fair global average function. The reference values obtained from the present study suggest that recommendations of normal limits should be specified by age group and sex. Acknowledgments We thank the participants of the echocardiography study, our colleagues at Levanger Hospital, and the personnel at the HUNT study centers who have been of invaluable assistance. We gratefully acknowledge the statistical expertise of Oyvind Salvesen. The HUNT study is collaboration among the HUNT Research Centre (Faculty of Medicine, Norwegian University of Science and Technology), Nord-Trøndelag County Council, and The Norwegian Institute of Public Health. Sources of Funding This study was funded by grants from the Norwegian University of Science and Technology. None. Disclosures References 1. Executive summary: HFSA 2006 comprehensive heart failure practice guideline. J Card Fail. 2006;12: Marwick TH. Should we be evaluating the ventricle or the myocardium? Advances in tissue characterization. J Am Soc Echocardiogr. 2004;17: Yu CM, Sanderson JE, Marwick TH, Oh JK. Tissue Doppler imaging a new prognosticator for cardiovascular diseases. J Am Coll Cardiol. 2007; 49: Ommen SR, Nishimura RA, Appleton CP, Miller FA, Oh JK, Redfield MM, Tajik AJ. Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures: a comparative simultaneous Doppler-catheterization study. Circulation. 2000;102: Stoylen A, Skjaerpe T. Systolic long axis function of the left ventricle. Global and regional information. Scand Cardiovasc J. 2003;37: Kuznetsova THL, López B, Jin Y, Richart T, Thijs L, Gonsáles A, Herregods MC, Fagard RH, Diez J, Staessen JA. Prevalence of left ventricular diastolic dysfunction in a general population. Circ Heart Fail. 2009;2: Mogelvang R, Sogaard P, Pedersen SA, Olsen NT, Schnohr P, Jensen JS. Tissue Doppler echocardiography in persons with hypertension, diabetes, or ischaemic heart disease: the Copenhagen City Heart Study. Eur Heart J. 2009;30: Alam M, Wardell J, Andersson E, Samad BA, Nordlander R. Characteristics of mitral and tricuspid annular velocities determined by pulsed wave Doppler tissue imaging in healthy subjects. J Am Soc Echocardiogr. 1999;12: Lindqvist P, Waldenstrom A, Henein M, Morner S, Kazzam E. Regional and global right ventricular function in healthy individuals aged years: a pulsed Doppler tissue imaging study: Umea General Population Heart Study. Echocardiography. 2005;22: Wierzbowska-Drabik K, Krzeminska-Pakula M, Chrzanowski L, Plewka M, Waszyrowski T, Drozdz J, Kurpesa M, Trzos E, Kasprzak JD. Age-dependency of classic and new parameters of diastolic function. Echocardiography. 2008;25: Holmen J, Midthjell K, Krüger O, Langhammer A, Holmen TL, Bratberg GH, Vatten L, Lund-Larsen PG. The Nord-Trøndelag Health Study (HUNT 2): objectives, contents, methods and participation. Nor Epidemiol. 2003;13: Dalen H, Thorstensen A, Aase SA, Ingul CB, Torp H, Vatten LJ, Stoylen A. Segmental and global longitudinal strain and strain rate based on echocardiography of 1266 healthy individuals: the HUNT study in Norway. Eur J Echocardiogr. 2010;11: Gottdiener JS, Bednarz J, Devereux R, Gardin J, Klein A, Manning WJ, Morehead A, Kitzman D, Oh J, Quinones M, Schiller NB, Stein JH, Weissman NJ. American Society of Echocardiography recommendations for use of echocardiography in clinical trials. J Am Soc Echocardiogr. 2004;17: Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA, Waggoner AD, Flachskampf FA, Pelikka PA, Evangelisa A. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. Eur J Echocardiogr. 2009;10: Quinones MA, Otto CM, Stoddard M, Waggoner A, Zoghbi WA. Recommendations for quantification of Doppler echocardiography: a report from the Doppler Quantification Task Force of the Nomenclature and Standards Committee of the American Society of Echocardiography. J Am Soc Echocardiogr. 2002;15: Thorstensen A, Dalen H, Amundsen BH, Aase SA, Stoylen A. Reproducibility in echocardiographic assessment of the left ventricular global and regional function, the HUNT study. Eur J Echocardiogr. 2010;11: Aase STH, Amundsen BH, Bjaastad TG, Stoylen A. Peak systolic velocity by upper edge of pulsed wave tissue Doppler spectrums cause overestimation compared to M-mode in normal subjects [Abstract]. Eur J Echocardiogr. 2008;9(suppl 1):S7 S Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1: Thorstensen A, Dalen H, Amundsen BH, Aase SA, Stoylen A. Peak systolic annulus velocity is the best index for measurement of contractility changes [Abstract]. Eur J Echocardiogr. 2009;10:ii Goebel B, Arnold R, Koletzki E, Ulmer HE, Eichhorn J, Borggrefe M, Figulla HR, Poerner TC. Exercise tissue Doppler echocardiography with strain rate imaging in healthy young individuals: feasibility, normal values and reproducibility. Int J Cardiovasc Imaging. 2007;23: Kukulski T, Hubbert L, Arnold M, Wranne B, Hatle L, Sutherland GR. Normal regional right ventricular function and its change with age: a Doppler myocardial imaging study. J Am Soc Echocardiogr. 2000;13: Nikitin NP, Witte KK, Thackray SD, de Silva R, Clark AL, Cleland JG. Longitudinal ventricular function: normal values of atrioventricular annular and myocardial velocities measured with quantitative twodimensional color Doppler tissue imaging. J Am Soc Echocardiogr. 2003;16: Innelli P, Esposito R, Olibet M, Nistri S, Galderisi M. The impact of ageing on right ventricular longitudinal function in healthy subjects: a pulsed tissue Doppler study. Eur J Echocardiogr. 2009;10: Bryg RJ, Williams GA, Labovitz AJ. Effect of aging on left ventricular diastolic filling in normal subjects. Am J Cardiol. 1987;59: Schirmer H, Lunde P, Rasmussen K. Mitral flow derived Doppler indices of left ventricular diastolic function in a general population; the Tromsø study. Eur Heart J. 2000;21: Schirmer H, Lunde P, Rasmussen K. Prevalence of left ventricular hypertrophy in a general population: the Tromsø study. Eur Heart J. 1999; 20: Madler CF, Payne N, Wilkenshoff U, Cohen A, Derumeaux GA, Pierard LA, Engvall J, Brodin LA, Sutherland GR, Fraser AG. Non-invasive

9 622 Circ Cardiovasc Imaging September 2010 diagnosis of coronary artery disease by quantitative stress echocardiography: optimal diagnostic models using off-line tissue Doppler in the MYDISE study. Eur Heart J. 2003;24: Pai RG, Gill KS. Amplitudes, durations, and timings of apically directed left ventricular myocardial velocities: I. Their normal pattern and coupling to ventricular filling and ejection. J Am Soc Echocardiogr. 1998;11: Manouras A, Shahgaldi K, Winter R, Brodin LA, Nowak J. Measurements of left ventricular myocardial longitudinal systolic displacement using spectral and colour tissue Doppler: time for a reassessment? Cardiovasc Ultrasound. 2009;7: Kasner M, Westermann D, Steendijk P, Gaub R, Wilkenshoff U, Weitmann K, Hoffmann W, Poller W, Schultheiss HP, Pauschinger M, Tschope C. Utility of Doppler echocardiography and tissue Doppler imaging in the estimation of diastolic function in heart failure with normal ejection fraction: a comparative Doppler-conductance catheterization study. Circulation. 2007;116: Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MA. Doppler tissue imaging: a noninvasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol. 1997;30: Oh JK, Hatle L, Tajik AJ, Little WC. Diastolic heart failure can be diagnosed by comprehensive two-dimensional and Doppler echocardiography. J Am Coll Cardiol. 2006;47: De Sutter J, De Backer J, Van de Veire N, Velghe A, De Buyzere M, Gillebert TC. Effects of age, gender, and left ventricular mass on septal mitral annulus velocity (E ) and the ratio of transmitral early peak velocity to E (E/E ). Am J Cardiol. 2005;95: Rivas-Gotz C, Manolios M, Thohan V, Nagueh SF. Impact of left ventricular ejection fraction on estimation of left ventricular filling pressures using tissue Doppler and flow propagation velocity. Am J Cardiol. 2003; 91: Hillis GS, Moller JE, Pellikka PA, Gersh BJ, Wright RS, Ommen SR, Reeder GS, Oh JK. Noninvasive estimation of left ventricular filling pressure by E/e is a powerful predictor of survival after acute myocardial infarction. J Am Coll Cardiol. 2004;43: Mogelvang R, Sogaard P, Pedersen SA, Olsen NT, Marott JL, Schnohr P, Goetze JP, Jensen JS. Cardiac dysfunction assessed by echocardiographic tissue Doppler imaging is an independent predictor of mortality in the general population. Circulation. 2009;119: Fernandes VR, Edvardsen T, Rosen BD, Carvalho B, Campos O, Cordeiro MA, Kronmal R, Bluemke DA, Lima JA. The influence of left ventricular size and global function on regional myocardial contraction and relaxation in an adult population free of cardiovascular disease: a tagged CMR study of the MESA cohort. J Cardiovasc Magn Reson. 2007;9: CLINICAL PERSPECTIVE Mitral and tricuspid annular velocities are used for quantification of left and right ventricular function in everyday clinical practice. This echocardiographic study of 1266 healthy subjects free from known cardiovascular disease, hypertension, and diabetes provides normal ranges of conventional blood flow Doppler and mitral and tricuspid annular pulsed wave and color tissue Doppler velocities according to age and sex. We suggest that the data presented in this study can be used as reference values in an everyday clinical setting.