Ratio of transmitral early filling velocity to early diastolic strain rate predicts long-term risk of cardiovascular morbidity and mortality in the general population

Ratio of transmitral early filling velocity to early diastolic strain rate predicts long-term... Abstract Aims It has previously been demonstrated that the ratio of early mitral inflow velocity to global diastolic strain rate (E/e′sr) is a significant predictor of cardiac events in specific patient populations. The utility of this measurement to predict cardiovascular events in a general population has not been evaluated. Methods and results A total of 1238 participants in a general population study underwent a health examination including echocardiography where global longitudinal strain (GLS) and E/e′sr were determined. The primary endpoint was the composite of incident heart failure (HF), acute myocardial infarction (AMI) or cardiovascular death (CVD). During follow-up (median 11 years), 140 (11.3%) participants reached the composite endpoint. E/e′sr was associated with adverse outcome [HR 1.17 95% CI (1.13–1.21); P < 0.001, per 10 cm increase]. After multivariable adjustment for echocardiographic and clinical parameters, E/e′sr remained an independent predictor of the composite endpoint [HR 1.08, 95% CI (1.02–1.13); P = 0.003] as opposed to E/e′ [HR 1.03, 95% CI (0.99–1.06); P = 0.11 per 1 unit increase]. Global longitudinal strain modified the relationship between E/e′sr and outcome (P for interaction = 0.015). E/e′sr was a stronger predictor in participants with good systolic function as determined by GLS (GLS > 18%) after multivariable adjustment, when compared to participants with reduced systolic function (GLS < 18%) [HR 1.28 95% CI (1.06–1.54); P = 0.011, and HR 1.08 95% CI (1.02–1.14); P = 0.012, respectively). E/e′sr provided incremental information [Harrell’s C-index: 0.839 (0.81–0.87) vs. 0.844 (0.82–0.87); P = 0.045] beyond the SCORE risk chart. Conclusion In the general population, E/e′sr provides independent and incremental prognostic information regarding cardiovascular morbidity and mortality. Additionally, E/e′sr is a stronger predictor of cardiac events than E/e′. Two-dimensional speckle tracking echocardiography, Global longitudinal strain , General population , Diastolic function , Long-term outcome Introduction Cardiovascular death (CVD) is a major cause of mortality, accounting for 31% of all deaths worldwide.1 Hence, early detection of left ventricular (LV) dysfunction in persons at high-risk of developing cardiac events is important. E/e′ as a measure of LV filling pressure has been used to asses impaired diastolic function following acute myocardial infarction (MI) and has proved to be an important predictor of major adverse events.2,3 Two-dimensional (2D) echocardiographic speckle tracking can correctly measure myocardial strain and strain rate. The measurements of global diastolic strain rate (e′sr) and ratio of early mitral inflow velocity (E) to global e′sr (E/e′sr) have been suggested as new measures of elevated LV filling pressure.4–6 These measures have the advantage that global early diastolic strain rate represents the diastolic performance of all segments of the myocardium, whereas tissue Doppler echocardiography may only consider a single myocardial wall. Furthermore, speckle tracking has overcome several technical limitations which tissue Doppler imaging is subject to (e.g. angle dependency). Several studies have demonstrated a close correlation between invasively measured LV filling pressure and E/e′sr.4–6 Myocardial ischaemia leads to impaired active relaxation during diastole and to reduced early diastolic untwisting and ventricular suction. These events affect E/e′sr. Furthermore, E/e′sr is also influenced by co-morbid conditions such as diabetes and hypertension, which affect myocardial relaxation through myocardial hypertrophy and deposition of triglycerides.8,9E/e′sr has not previously been investigated as a prognosticator in addition to existing echocardiographic indices in a large-scale study on a general population. Hence, the aim of this study was to determine the prognostic significance of E/e′sr in the general population. Methods Study population This longitudinal cohort study included 2154 participants who underwent echocardiographic examination including colour tissue Doppler imaging (TDI) and 2D speckle tracking analysis. Exclusion criteria were prevalent diagnosis of heart failure (HF), atrial fibrillation, and inadequate image quality for 2D speckle tracking. After exclusion, 1238 participants remained. Exclusion process is displayed in Figure 1 and has been described in detail in the Supplementary material online, Page 2. The study was performed in accordance to the second Helsinki Declaration and approved by the regional ethics committee. Figure 1 View largeDownload slide Flow diagram showing the exclusion process of the study. Flow diagram of the exclusion process for the Copenhagen City Heart study ending with the included participants in this study. 2D, two-dimensional; TDI, tissue Doppler imaging; E/e′sr, ratio of transmitral early filling velocity to early diastolic strain rate. Figure 1 View largeDownload slide Flow diagram showing the exclusion process of the study. Flow diagram of the exclusion process for the Copenhagen City Heart study ending with the included participants in this study. 2D, two-dimensional; TDI, tissue Doppler imaging; E/e′sr, ratio of transmitral early filling velocity to early diastolic strain rate. Health examination Detailed information is available in the Supplementary material online and has also previously been described elsewhere.10,11 Echocardiography Echocardiography was performed using Vivid 5 ultrasound systems (GE Healthcare, Horten, Norway) with a 2.5-MHz transducer by three experienced sonographers. All participants were examined with complete 2D, colour, and spectral Doppler echocardiography and colour TDI. All echocardiograms were stored on magneto-optical disks and an external FireWire hard drive (LaCie, France) and analysed offline with commercially available post-processing software (EchoPac version 8, GE Medical, Horten, Norway). Conventional echocardiography and colour tissue Doppler imaging Conventional echocardiography and colour TDI was performed as has been previously described10,11. An elaborate description is available in the Supplementary material online, Page 2. Speckle tracking echocardiography Two-dimensional speckle tracking was performed in the 4-chamber, 2-chamber, and 3-chamber apical projections with an average of 57 frames per second (FPS), standard deviation (SD) of 4 FPS. Speckle tracking methodology is described in detail in the Supplementary material online, page 3. The mean values of all LV segments from all three apical views were used to calculate GLS and e′sr. E/e′sr ratio was calculated as E velocity (m/s) divided by the absolute value of global e′sr (s−1). In cases where speckle tracking was deemed unsatisfactory from one of the chamber views, it was obtained from the mean value of the remaining chamber projections. A total of 1094 subjects had 4 chamber views adequate for speckle tracking analysis, 770 had adequate 2 chamber views, and 586 subjects had adequate 3 chamber views for speckle tracking analysis. Thus, all 3 apical views were adequate for strain analysis in 396 participants, 2 in 527, and 1 in 315 participants. Previously, our lab has shown good intra- and interobserver variability of GLS. In the same study, we also found a good intra- and interobserver variability of strain rate e with a small bias (for strain rate e: mean difference ± 1.96 SDs was −0.06 ± 0.25 for the intraobserver analysis and 0.06 ± 0.28 for the interobserver analysis).12 One investigator who was blinded to all other information analysed all 2D echocardiograms by speckle tracking. Follow-up and outcome Follow-up was 100%. The primary endpoints were defined as CVD or admission due to either acute MI or incident HF. The Danish National Board of Health’s National Patient Registry was used to retrieve the endpoints in the Copenhagen City Heart Study by using International classification of diseases codes 10th revision (ICD-10). This is described in more detail elsewhere.10 Furthermore, the National Danish Causes of Death Registry was used to obtain follow-up data regarding CVD. ICD-10 codes I00-I99 were used to define CVD. Statistics STATA statistics/data analysis, SE 12.0 (StataCorp, College Station, TX, USA) was used for all analyses. Statistical significance was defined as P ≤ 0.05. Baseline clinical and echocardiographic data across E/e′sr quartiles (Table 1) were compared by trend tests using linear regression for continuous Gaussian distributed variables and with an extension of the Wilcoxon rank-sum test13 for continuous non-Gaussian distributed variables. The χ2 for trend for proportions were used when relevant. Rates of all events were calculated as the number of events divided by person-time at risk. Figure 2 was constructed using a Poisson model to estimate incidence rates. The association between E/e′sr and the composite outcome was tested using restricted cubic splines with the number of knots selected according to the value associated with the lowest Akaike information criterion value. Both linear and non-linear interaction terms were tested using this method. Cox proportional hazards regression models were used to examine the associations of baseline variables, echocardiographic parameters, and E/e′sr with the risk of the composite endpoint and are displayed in Tables 2 and 3. Univariable Cox proportional hazards regression models were used to obtain Harrell’s c-statistics14 in order to test the prognostic value of E/e′sr for predicting the composite endpoint. Left ventricular ejection fraction (LVEF) and proBNP were dichotomized in high/low categories using LVEF cut-off of 50% and proBNP > 150 pmol/L and included as dichotomous terms when used in the multivariable Cox regression models. Furthermore, Harrell’s C-statistics was used to explore the incremental value of adding E/e′sr and the other variables to the clinical predictors from the SCORE risk chart15 (Supplementary material online, Table S3). Calibration of the model was evaluated with Groennesby–Borgan goodness-of-fit χ2-estimates using deciles. Competing-risks regression models were used to account for non-cardiovascular death as a competing risk to the composite outcome (our event of interest). Table 1 Baseline clinical characteristics for the participants stratified according to quartiles of E/e′sr E/e′sr (m)  All 0.69 ± 0.3 n = 1238  1. Quartile <0.52 n = 310  2. Quartile 0.52–0.63 n = 310  3. Quartile 0.63–0.79 n = 309  4. Quartile >0.79 n = 309  P-value for trend  Demographics   Age (years)  56.9 ± 16.2  53.4 ± 15.7  53.6 ± 16.3  57.3 ± 15.8  63.3 ± 15.0  <0.001   Male gender, n (%)  522 (42.2)  121 (39.0)  133 (42.9)  123 (39.8)  145 (46.9)  0.10  Clinical   Systolic blood pressure (mmHg)  134 ± 23  127 ± 22  131 ± 23  134 ± 21  143 ± 23  <0.001   Diastolic blood pressure (mmHg)  78 ± 12  76 ± 11  78 ± 14  77 ± 12  80 ± 12  <0.001   Heart rate (bpm)  65 ± 10  66 ± 10  65 ± 11  65 ± 10  67 ± 11  0.37   BMI (kg/m2)  25.0 ± 3.6  24.3 ± 3.4  24.1 ± 3.3  25.2 ± 3.4  26.4 ± 3.8  <0.001   Hypertension, n (%)  466 (37.8)  82 (26.6)  97 (31.4)  107 (34.7)  180 (58.3)  <0.001   Diabetes, n (%)  113 (9.1)  29 (9.4)  24 (7.7)  28 (9.1)  32 (10.4)  0.55   Previous ischaemic heart disease, n (%)  58 (4.7)  8 (2.6)  11 (3.5)  10 (3.2)  29 (9.4)  0.001   Smoking status  0.038    Never, n (%)  410 (33.3)  120 (39.1)  98 (31.9)  96 (31.4)  96 (31.4)    Previous, n (%)  401 (33.2)  104 (33.9)  96 (29.6)  97 (31.7)  103 (33.7)    Current, n (%)  415 (33.5)  90 (29.3)  105 (34.2)  113 (36.9)  107 (35.0)  Lab work   Cholesterol (mmol)  5.5 ± 1.2  5.5 ± 1.1  5.4 ± 1.2  5.6 ± 1.2  5.7 ± 1.1  0.007   eGFR (mL/min/1.73 m2)  75.2 (66.0, 86.1)  75.3 (66.2, 85.2)  76.8 (67.3, 88.4)  75.5 (64.9, 87.0)  73.0 (64.3, 83.8)  0.06   Plasma proBNP (pmol/L)  16.0 (8.0, 29.0)  15.0 (5.0, 27.0)  15.0 (7.0, 24.0)  17.0 (8.0, 32.0)  19.0 (9.0, 38.0)  0.001  Echocardiography   LVEF < 50%, n (%)  9 (0.7)  0 (0.0)  0 (0.0)  2 (0.7)  7 (2.3)  0.001   Hypertrophy, n (%)  176 (15.8)  28 (9.7)  30 (10.5)  42 (15.4)  76 (28.4)  0.001   Dilatation, n (%)  43 (3.8)  15 (5.2)  9 (3.1)  10 (3.6)  21 (7.7)  0.16   LVIDd/height (cm/m)  2.8 ± 0.3  2.8 ± 0.3  2.8 ± 0.2  2.8 ± 0.3  2.9 ± 0.3  0.14   Left atrial dimension (cm)  3.4 ± 0.4  3.3 ± 0.4  3.3 ± 0.4  3.4 ± 0.4  3.5 ± 0.4  <0.001   E/A ratio  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.0 (0.8, 1.2)  <0.001   DT (ms)  166.9 ± 41.0  162.1 ± 36.0  166.3 ± 40.2  166.1 ± 40.1  173.2 ± 45.0  0.008   E/e′  9.7 (7.9, 12.2)  8.3 (7.0, 10.0)  8.9 (7.6, 11.1)  10.1 (8.4, 12.4)  11.9 (9.8, 15.6)  <0.001   LVMI (g/m2)  82.2 (70.9, 96.9)  78.6 (68.0, 91.6)  81.2 (71.3, 95.3)  81.4 (70.3, 97.7)  87.0 (74.8, 104.5)  <0.001   LVEF (%)  59.8 ± 1.7  59.9 ± 0.6  59.8 ± 1.1  59.9 ± 1.1  59.4 ± 2.9  <0.001   GLS (%)  18.1 ± 3.7  20.7 ± 3.1  19.0 ± 2.8  17.5 ± 2.9  15.0 ± 3.1  <0.001  Events during follow-up   AMI admissions during follow-up, n (%)  41 (3.3)  5 (1.6)  8 (2.6)  5 (1.6)  23 (7.4)  <0.001   Heart failure admission during follow-up, n (%)  72 (5.8)  6 (1.9)  12 (3.9)  12 (3.9)  42 (13.6)  <0.001   Cardiovascular death during follow-up, n (%)  71 (5.7)  9 (2.9)  15 (4.8)  14 (4.5)  33 (10.7)  <0.001   Composite endpoint during follow-up, n (%)  140 (11.3)  15 (4.8)  27 (8.7)  25 (8.1)  73 (23.6)  <0.001  E/e′sr (m)  All 0.69 ± 0.3 n = 1238  1. Quartile <0.52 n = 310  2. Quartile 0.52–0.63 n = 310  3. Quartile 0.63–0.79 n = 309  4. Quartile >0.79 n = 309  P-value for trend  Demographics   Age (years)  56.9 ± 16.2  53.4 ± 15.7  53.6 ± 16.3  57.3 ± 15.8  63.3 ± 15.0  <0.001   Male gender, n (%)  522 (42.2)  121 (39.0)  133 (42.9)  123 (39.8)  145 (46.9)  0.10  Clinical   Systolic blood pressure (mmHg)  134 ± 23  127 ± 22  131 ± 23  134 ± 21  143 ± 23  <0.001   Diastolic blood pressure (mmHg)  78 ± 12  76 ± 11  78 ± 14  77 ± 12  80 ± 12  <0.001   Heart rate (bpm)  65 ± 10  66 ± 10  65 ± 11  65 ± 10  67 ± 11  0.37   BMI (kg/m2)  25.0 ± 3.6  24.3 ± 3.4  24.1 ± 3.3  25.2 ± 3.4  26.4 ± 3.8  <0.001   Hypertension, n (%)  466 (37.8)  82 (26.6)  97 (31.4)  107 (34.7)  180 (58.3)  <0.001   Diabetes, n (%)  113 (9.1)  29 (9.4)  24 (7.7)  28 (9.1)  32 (10.4)  0.55   Previous ischaemic heart disease, n (%)  58 (4.7)  8 (2.6)  11 (3.5)  10 (3.2)  29 (9.4)  0.001   Smoking status  0.038    Never, n (%)  410 (33.3)  120 (39.1)  98 (31.9)  96 (31.4)  96 (31.4)    Previous, n (%)  401 (33.2)  104 (33.9)  96 (29.6)  97 (31.7)  103 (33.7)    Current, n (%)  415 (33.5)  90 (29.3)  105 (34.2)  113 (36.9)  107 (35.0)  Lab work   Cholesterol (mmol)  5.5 ± 1.2  5.5 ± 1.1  5.4 ± 1.2  5.6 ± 1.2  5.7 ± 1.1  0.007   eGFR (mL/min/1.73 m2)  75.2 (66.0, 86.1)  75.3 (66.2, 85.2)  76.8 (67.3, 88.4)  75.5 (64.9, 87.0)  73.0 (64.3, 83.8)  0.06   Plasma proBNP (pmol/L)  16.0 (8.0, 29.0)  15.0 (5.0, 27.0)  15.0 (7.0, 24.0)  17.0 (8.0, 32.0)  19.0 (9.0, 38.0)  0.001  Echocardiography   LVEF < 50%, n (%)  9 (0.7)  0 (0.0)  0 (0.0)  2 (0.7)  7 (2.3)  0.001   Hypertrophy, n (%)  176 (15.8)  28 (9.7)  30 (10.5)  42 (15.4)  76 (28.4)  0.001   Dilatation, n (%)  43 (3.8)  15 (5.2)  9 (3.1)  10 (3.6)  21 (7.7)  0.16   LVIDd/height (cm/m)  2.8 ± 0.3  2.8 ± 0.3  2.8 ± 0.2  2.8 ± 0.3  2.9 ± 0.3  0.14   Left atrial dimension (cm)  3.4 ± 0.4  3.3 ± 0.4  3.3 ± 0.4  3.4 ± 0.4  3.5 ± 0.4  <0.001   E/A ratio  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.0 (0.8, 1.2)  <0.001   DT (ms)  166.9 ± 41.0  162.1 ± 36.0  166.3 ± 40.2  166.1 ± 40.1  173.2 ± 45.0  0.008   E/e′  9.7 (7.9, 12.2)  8.3 (7.0, 10.0)  8.9 (7.6, 11.1)  10.1 (8.4, 12.4)  11.9 (9.8, 15.6)  <0.001   LVMI (g/m2)  82.2 (70.9, 96.9)  78.6 (68.0, 91.6)  81.2 (71.3, 95.3)  81.4 (70.3, 97.7)  87.0 (74.8, 104.5)  <0.001   LVEF (%)  59.8 ± 1.7  59.9 ± 0.6  59.8 ± 1.1  59.9 ± 1.1  59.4 ± 2.9  <0.001   GLS (%)  18.1 ± 3.7  20.7 ± 3.1  19.0 ± 2.8  17.5 ± 2.9  15.0 ± 3.1  <0.001  Events during follow-up   AMI admissions during follow-up, n (%)  41 (3.3)  5 (1.6)  8 (2.6)  5 (1.6)  23 (7.4)  <0.001   Heart failure admission during follow-up, n (%)  72 (5.8)  6 (1.9)  12 (3.9)  12 (3.9)  42 (13.6)  <0.001   Cardiovascular death during follow-up, n (%)  71 (5.7)  9 (2.9)  15 (4.8)  14 (4.5)  33 (10.7)  <0.001   Composite endpoint during follow-up, n (%)  140 (11.3)  15 (4.8)  27 (8.7)  25 (8.1)  73 (23.6)  <0.001  A, peak transmitral late diastolic inflow velocity; AMI, acute myocardial infarction; BMI, body mass index; DT, deceleration time of early diastolic inflow; E, peak transmitral early diastolic inflow velocity; e′, average peak early diastolic longitudinal mitral annular velocity determined by colour TDI; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; LVIDd, left ventricular internal diameter at end-diastole; LVMI, left ventricular mass index; proBNP, pro B-type natriuretic peptide. Table 1 Baseline clinical characteristics for the participants stratified according to quartiles of E/e′sr E/e′sr (m)  All 0.69 ± 0.3 n = 1238  1. Quartile <0.52 n = 310  2. Quartile 0.52–0.63 n = 310  3. Quartile 0.63–0.79 n = 309  4. Quartile >0.79 n = 309  P-value for trend  Demographics   Age (years)  56.9 ± 16.2  53.4 ± 15.7  53.6 ± 16.3  57.3 ± 15.8  63.3 ± 15.0  <0.001   Male gender, n (%)  522 (42.2)  121 (39.0)  133 (42.9)  123 (39.8)  145 (46.9)  0.10  Clinical   Systolic blood pressure (mmHg)  134 ± 23  127 ± 22  131 ± 23  134 ± 21  143 ± 23  <0.001   Diastolic blood pressure (mmHg)  78 ± 12  76 ± 11  78 ± 14  77 ± 12  80 ± 12  <0.001   Heart rate (bpm)  65 ± 10  66 ± 10  65 ± 11  65 ± 10  67 ± 11  0.37   BMI (kg/m2)  25.0 ± 3.6  24.3 ± 3.4  24.1 ± 3.3  25.2 ± 3.4  26.4 ± 3.8  <0.001   Hypertension, n (%)  466 (37.8)  82 (26.6)  97 (31.4)  107 (34.7)  180 (58.3)  <0.001   Diabetes, n (%)  113 (9.1)  29 (9.4)  24 (7.7)  28 (9.1)  32 (10.4)  0.55   Previous ischaemic heart disease, n (%)  58 (4.7)  8 (2.6)  11 (3.5)  10 (3.2)  29 (9.4)  0.001   Smoking status  0.038    Never, n (%)  410 (33.3)  120 (39.1)  98 (31.9)  96 (31.4)  96 (31.4)    Previous, n (%)  401 (33.2)  104 (33.9)  96 (29.6)  97 (31.7)  103 (33.7)    Current, n (%)  415 (33.5)  90 (29.3)  105 (34.2)  113 (36.9)  107 (35.0)  Lab work   Cholesterol (mmol)  5.5 ± 1.2  5.5 ± 1.1  5.4 ± 1.2  5.6 ± 1.2  5.7 ± 1.1  0.007   eGFR (mL/min/1.73 m2)  75.2 (66.0, 86.1)  75.3 (66.2, 85.2)  76.8 (67.3, 88.4)  75.5 (64.9, 87.0)  73.0 (64.3, 83.8)  0.06   Plasma proBNP (pmol/L)  16.0 (8.0, 29.0)  15.0 (5.0, 27.0)  15.0 (7.0, 24.0)  17.0 (8.0, 32.0)  19.0 (9.0, 38.0)  0.001  Echocardiography   LVEF < 50%, n (%)  9 (0.7)  0 (0.0)  0 (0.0)  2 (0.7)  7 (2.3)  0.001   Hypertrophy, n (%)  176 (15.8)  28 (9.7)  30 (10.5)  42 (15.4)  76 (28.4)  0.001   Dilatation, n (%)  43 (3.8)  15 (5.2)  9 (3.1)  10 (3.6)  21 (7.7)  0.16   LVIDd/height (cm/m)  2.8 ± 0.3  2.8 ± 0.3  2.8 ± 0.2  2.8 ± 0.3  2.9 ± 0.3  0.14   Left atrial dimension (cm)  3.4 ± 0.4  3.3 ± 0.4  3.3 ± 0.4  3.4 ± 0.4  3.5 ± 0.4  <0.001   E/A ratio  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.0 (0.8, 1.2)  <0.001   DT (ms)  166.9 ± 41.0  162.1 ± 36.0  166.3 ± 40.2  166.1 ± 40.1  173.2 ± 45.0  0.008   E/e′  9.7 (7.9, 12.2)  8.3 (7.0, 10.0)  8.9 (7.6, 11.1)  10.1 (8.4, 12.4)  11.9 (9.8, 15.6)  <0.001   LVMI (g/m2)  82.2 (70.9, 96.9)  78.6 (68.0, 91.6)  81.2 (71.3, 95.3)  81.4 (70.3, 97.7)  87.0 (74.8, 104.5)  <0.001   LVEF (%)  59.8 ± 1.7  59.9 ± 0.6  59.8 ± 1.1  59.9 ± 1.1  59.4 ± 2.9  <0.001   GLS (%)  18.1 ± 3.7  20.7 ± 3.1  19.0 ± 2.8  17.5 ± 2.9  15.0 ± 3.1  <0.001  Events during follow-up   AMI admissions during follow-up, n (%)  41 (3.3)  5 (1.6)  8 (2.6)  5 (1.6)  23 (7.4)  <0.001   Heart failure admission during follow-up, n (%)  72 (5.8)  6 (1.9)  12 (3.9)  12 (3.9)  42 (13.6)  <0.001   Cardiovascular death during follow-up, n (%)  71 (5.7)  9 (2.9)  15 (4.8)  14 (4.5)  33 (10.7)  <0.001   Composite endpoint during follow-up, n (%)  140 (11.3)  15 (4.8)  27 (8.7)  25 (8.1)  73 (23.6)  <0.001  E/e′sr (m)  All 0.69 ± 0.3 n = 1238  1. Quartile <0.52 n = 310  2. Quartile 0.52–0.63 n = 310  3. Quartile 0.63–0.79 n = 309  4. Quartile >0.79 n = 309  P-value for trend  Demographics   Age (years)  56.9 ± 16.2  53.4 ± 15.7  53.6 ± 16.3  57.3 ± 15.8  63.3 ± 15.0  <0.001   Male gender, n (%)  522 (42.2)  121 (39.0)  133 (42.9)  123 (39.8)  145 (46.9)  0.10  Clinical   Systolic blood pressure (mmHg)  134 ± 23  127 ± 22  131 ± 23  134 ± 21  143 ± 23  <0.001   Diastolic blood pressure (mmHg)  78 ± 12  76 ± 11  78 ± 14  77 ± 12  80 ± 12  <0.001   Heart rate (bpm)  65 ± 10  66 ± 10  65 ± 11  65 ± 10  67 ± 11  0.37   BMI (kg/m2)  25.0 ± 3.6  24.3 ± 3.4  24.1 ± 3.3  25.2 ± 3.4  26.4 ± 3.8  <0.001   Hypertension, n (%)  466 (37.8)  82 (26.6)  97 (31.4)  107 (34.7)  180 (58.3)  <0.001   Diabetes, n (%)  113 (9.1)  29 (9.4)  24 (7.7)  28 (9.1)  32 (10.4)  0.55   Previous ischaemic heart disease, n (%)  58 (4.7)  8 (2.6)  11 (3.5)  10 (3.2)  29 (9.4)  0.001   Smoking status  0.038    Never, n (%)  410 (33.3)  120 (39.1)  98 (31.9)  96 (31.4)  96 (31.4)    Previous, n (%)  401 (33.2)  104 (33.9)  96 (29.6)  97 (31.7)  103 (33.7)    Current, n (%)  415 (33.5)  90 (29.3)  105 (34.2)  113 (36.9)  107 (35.0)  Lab work   Cholesterol (mmol)  5.5 ± 1.2  5.5 ± 1.1  5.4 ± 1.2  5.6 ± 1.2  5.7 ± 1.1  0.007   eGFR (mL/min/1.73 m2)  75.2 (66.0, 86.1)  75.3 (66.2, 85.2)  76.8 (67.3, 88.4)  75.5 (64.9, 87.0)  73.0 (64.3, 83.8)  0.06   Plasma proBNP (pmol/L)  16.0 (8.0, 29.0)  15.0 (5.0, 27.0)  15.0 (7.0, 24.0)  17.0 (8.0, 32.0)  19.0 (9.0, 38.0)  0.001  Echocardiography   LVEF < 50%, n (%)  9 (0.7)  0 (0.0)  0 (0.0)  2 (0.7)  7 (2.3)  0.001   Hypertrophy, n (%)  176 (15.8)  28 (9.7)  30 (10.5)  42 (15.4)  76 (28.4)  0.001   Dilatation, n (%)  43 (3.8)  15 (5.2)  9 (3.1)  10 (3.6)  21 (7.7)  0.16   LVIDd/height (cm/m)  2.8 ± 0.3  2.8 ± 0.3  2.8 ± 0.2  2.8 ± 0.3  2.9 ± 0.3  0.14   Left atrial dimension (cm)  3.4 ± 0.4  3.3 ± 0.4  3.3 ± 0.4  3.4 ± 0.4  3.5 ± 0.4  <0.001   E/A ratio  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.0 (0.8, 1.2)  <0.001   DT (ms)  166.9 ± 41.0  162.1 ± 36.0  166.3 ± 40.2  166.1 ± 40.1  173.2 ± 45.0  0.008   E/e′  9.7 (7.9, 12.2)  8.3 (7.0, 10.0)  8.9 (7.6, 11.1)  10.1 (8.4, 12.4)  11.9 (9.8, 15.6)  <0.001   LVMI (g/m2)  82.2 (70.9, 96.9)  78.6 (68.0, 91.6)  81.2 (71.3, 95.3)  81.4 (70.3, 97.7)  87.0 (74.8, 104.5)  <0.001   LVEF (%)  59.8 ± 1.7  59.9 ± 0.6  59.8 ± 1.1  59.9 ± 1.1  59.4 ± 2.9  <0.001   GLS (%)  18.1 ± 3.7  20.7 ± 3.1  19.0 ± 2.8  17.5 ± 2.9  15.0 ± 3.1  <0.001  Events during follow-up   AMI admissions during follow-up, n (%)  41 (3.3)  5 (1.6)  8 (2.6)  5 (1.6)  23 (7.4)  <0.001   Heart failure admission during follow-up, n (%)  72 (5.8)  6 (1.9)  12 (3.9)  12 (3.9)  42 (13.6)  <0.001   Cardiovascular death during follow-up, n (%)  71 (5.7)  9 (2.9)  15 (4.8)  14 (4.5)  33 (10.7)  <0.001   Composite endpoint during follow-up, n (%)  140 (11.3)  15 (4.8)  27 (8.7)  25 (8.1)  73 (23.6)  <0.001  A, peak transmitral late diastolic inflow velocity; AMI, acute myocardial infarction; BMI, body mass index; DT, deceleration time of early diastolic inflow; E, peak transmitral early diastolic inflow velocity; e′, average peak early diastolic longitudinal mitral annular velocity determined by colour TDI; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; LVIDd, left ventricular internal diameter at end-diastole; LVMI, left ventricular mass index; proBNP, pro B-type natriuretic peptide. Table 2 E/e′sr as a predictor of long-term outcome in the general population (n = 1238)   Composite endpoint (140 events)  P-value  HF (72 events)  P-value  AMI (41 events)  P-value  Cardiovascular death (71 events)  P-value  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)    E/e′sr per 10 cm increase  1.17 (1.13–1.21) C-stat 0.67  <0.001  1.16 (1.11–1.21) C-stat 0.70  <0.001  1.17 (1.11–1.25) C-stat 0.69  <0.001  1.12 (1.07–1.18) C-stat 0.63  <0.001  Model 1  E/e′sr per 10 cm increase  1.10 (1.06–1.15)  <0.001  1.10 (1.05–1.15)  <0.001  1.12 (1.05–1.19)  0.001  1.05 (0.99–1.12)  0.069  Model 2  E/e′sr per 10 cm increase  1.08 (1.03–1.13)  0.001  1.08 (1.03–1.15)  0.003  1.10 (1.03–1.18)  0.007  1.00 (0.92–1.08)  0.98  Model 3  E/e′sr per 10 cm increase  1.06 (1.01–1.12)  0.018  1.08 (1.01–1.15)  0.027  1.10 (1.01–1.20)  0.021  0.99 (0.90–1.09)  0.82    Composite endpoint (140 events)  P-value  HF (72 events)  P-value  AMI (41 events)  P-value  Cardiovascular death (71 events)  P-value  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)    E/e′sr per 10 cm increase  1.17 (1.13–1.21) C-stat 0.67  <0.001  1.16 (1.11–1.21) C-stat 0.70  <0.001  1.17 (1.11–1.25) C-stat 0.69  <0.001  1.12 (1.07–1.18) C-stat 0.63  <0.001  Model 1  E/e′sr per 10 cm increase  1.10 (1.06–1.15)  <0.001  1.10 (1.05–1.15)  <0.001  1.12 (1.05–1.19)  0.001  1.05 (0.99–1.12)  0.069  Model 2  E/e′sr per 10 cm increase  1.08 (1.03–1.13)  0.001  1.08 (1.03–1.15)  0.003  1.10 (1.03–1.18)  0.007  1.00 (0.92–1.08)  0.98  Model 3  E/e′sr per 10 cm increase  1.06 (1.01–1.12)  0.018  1.08 (1.01–1.15)  0.027  1.10 (1.01–1.20)  0.021  0.99 (0.90–1.09)  0.82  Model 1 is adjusted for age and gender. Model 2 is adjusted for age, gender, heart rate, BMI, diabetes, eGFR, smoking, previous ischaemic heart disease, systolic blood pressure, and proBNP (>150 pmol/L). Model 3 is adjusted for the same variables as Model 2 and additionally for left ventricular (LV) ejection fraction (<50%), LV mass index, LV dilatation, and left atrium dimension. E/e′sr is per 10 cm increase. AMI, acute myocardial infarction; CI, confidence interval; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; HF, heart failure. Table 2 E/e′sr as a predictor of long-term outcome in the general population (n = 1238)   Composite endpoint (140 events)  P-value  HF (72 events)  P-value  AMI (41 events)  P-value  Cardiovascular death (71 events)  P-value  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)    E/e′sr per 10 cm increase  1.17 (1.13–1.21) C-stat 0.67  <0.001  1.16 (1.11–1.21) C-stat 0.70  <0.001  1.17 (1.11–1.25) C-stat 0.69  <0.001  1.12 (1.07–1.18) C-stat 0.63  <0.001  Model 1  E/e′sr per 10 cm increase  1.10 (1.06–1.15)  <0.001  1.10 (1.05–1.15)  <0.001  1.12 (1.05–1.19)  0.001  1.05 (0.99–1.12)  0.069  Model 2  E/e′sr per 10 cm increase  1.08 (1.03–1.13)  0.001  1.08 (1.03–1.15)  0.003  1.10 (1.03–1.18)  0.007  1.00 (0.92–1.08)  0.98  Model 3  E/e′sr per 10 cm increase  1.06 (1.01–1.12)  0.018  1.08 (1.01–1.15)  0.027  1.10 (1.01–1.20)  0.021  0.99 (0.90–1.09)  0.82    Composite endpoint (140 events)  P-value  HF (72 events)  P-value  AMI (41 events)  P-value  Cardiovascular death (71 events)  P-value  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)    E/e′sr per 10 cm increase  1.17 (1.13–1.21) C-stat 0.67  <0.001  1.16 (1.11–1.21) C-stat 0.70  <0.001  1.17 (1.11–1.25) C-stat 0.69  <0.001  1.12 (1.07–1.18) C-stat 0.63  <0.001  Model 1  E/e′sr per 10 cm increase  1.10 (1.06–1.15)  <0.001  1.10 (1.05–1.15)  <0.001  1.12 (1.05–1.19)  0.001  1.05 (0.99–1.12)  0.069  Model 2  E/e′sr per 10 cm increase  1.08 (1.03–1.13)  0.001  1.08 (1.03–1.15)  0.003  1.10 (1.03–1.18)  0.007  1.00 (0.92–1.08)  0.98  Model 3  E/e′sr per 10 cm increase  1.06 (1.01–1.12)  0.018  1.08 (1.01–1.15)  0.027  1.10 (1.01–1.20)  0.021  0.99 (0.90–1.09)  0.82  Model 1 is adjusted for age and gender. Model 2 is adjusted for age, gender, heart rate, BMI, diabetes, eGFR, smoking, previous ischaemic heart disease, systolic blood pressure, and proBNP (>150 pmol/L). Model 3 is adjusted for the same variables as Model 2 and additionally for left ventricular (LV) ejection fraction (<50%), LV mass index, LV dilatation, and left atrium dimension. E/e′sr is per 10 cm increase. AMI, acute myocardial infarction; CI, confidence interval; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; HF, heart failure. Table 3 E/e′sr as a predictor of long-term outcome in persons with good systolic function (global longitudinal strain > 18%) or reduced systolic function (global longitudinal strain < 18%) from the general population   Good systolic cardiac function (GLS > 18%) (n = 617)  P-value  Reduced systolic function (GLS < 18%) (n = 621)  P-value  Composite endpoint (46 events)  Composite endpoint (94 events)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  E/e′sr per 10 cm increase  1.41 (1.19–1.66)  <0.001  1.13 (1.08–1.19)  <0.001  C-stat 0.62  C-stat 0.64  Model 1  E/e′sr per 10 cm increase  1.21 (1.04–1.41)  0.015  1.08 (1.03–1.13)  0.001  Model 2  E/e′sr per 10 cm increase  1.16 (1.01–1.37)  0.037  1.08 (1.02–1.14)  0.004  Model 3  E/e′sr per 10 cm increase  1.28 (1.06–1.54)  0.011  1.08 (1.02–1.14)  0.012    Good systolic cardiac function (GLS > 18%) (n = 617)  P-value  Reduced systolic function (GLS < 18%) (n = 621)  P-value  Composite endpoint (46 events)  Composite endpoint (94 events)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  E/e′sr per 10 cm increase  1.41 (1.19–1.66)  <0.001  1.13 (1.08–1.19)  <0.001  C-stat 0.62  C-stat 0.64  Model 1  E/e′sr per 10 cm increase  1.21 (1.04–1.41)  0.015  1.08 (1.03–1.13)  0.001  Model 2  E/e′sr per 10 cm increase  1.16 (1.01–1.37)  0.037  1.08 (1.02–1.14)  0.004  Model 3  E/e′sr per 10 cm increase  1.28 (1.06–1.54)  0.011  1.08 (1.02–1.14)  0.012  Model 1 is adjusted for age and gender. Model 2 is adjusted for age, gender, heart rate, BMI, diabetes, eGFR, smoking, previous ischaemic heart disease, systolic blood pressure, and proBNP (>150 pmol/L). Model 3 is adjusted for the same variables as Model 2 and additionally for left ventricular (LV) ejection fraction (<50%), LV mass index, LV dilatation, and left atrium dimension. E/e′sr is per 10 cm increase. Composite endpoint: Heart failure, acute myocardial infarction, and cardiovascular death. CI, confidence interval; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; GLS, global longitudinal strain. Table 3 E/e′sr as a predictor of long-term outcome in persons with good systolic function (global longitudinal strain > 18%) or reduced systolic function (global longitudinal strain < 18%) from the general population   Good systolic cardiac function (GLS > 18%) (n = 617)  P-value  Reduced systolic function (GLS < 18%) (n = 621)  P-value  Composite endpoint (46 events)  Composite endpoint (94 events)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  E/e′sr per 10 cm increase  1.41 (1.19–1.66)  <0.001  1.13 (1.08–1.19)  <0.001  C-stat 0.62  C-stat 0.64  Model 1  E/e′sr per 10 cm increase  1.21 (1.04–1.41)  0.015  1.08 (1.03–1.13)  0.001  Model 2  E/e′sr per 10 cm increase  1.16 (1.01–1.37)  0.037  1.08 (1.02–1.14)  0.004  Model 3  E/e′sr per 10 cm increase  1.28 (1.06–1.54)  0.011  1.08 (1.02–1.14)  0.012    Good systolic cardiac function (GLS > 18%) (n = 617)  P-value  Reduced systolic function (GLS < 18%) (n = 621)  P-value  Composite endpoint (46 events)  Composite endpoint (94 events)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  E/e′sr per 10 cm increase  1.41 (1.19–1.66)  <0.001  1.13 (1.08–1.19)  <0.001  C-stat 0.62  C-stat 0.64  Model 1  E/e′sr per 10 cm increase  1.21 (1.04–1.41)  0.015  1.08 (1.03–1.13)  0.001  Model 2  E/e′sr per 10 cm increase  1.16 (1.01–1.37)  0.037  1.08 (1.02–1.14)  0.004  Model 3  E/e′sr per 10 cm increase  1.28 (1.06–1.54)  0.011  1.08 (1.02–1.14)  0.012  Model 1 is adjusted for age and gender. Model 2 is adjusted for age, gender, heart rate, BMI, diabetes, eGFR, smoking, previous ischaemic heart disease, systolic blood pressure, and proBNP (>150 pmol/L). Model 3 is adjusted for the same variables as Model 2 and additionally for left ventricular (LV) ejection fraction (<50%), LV mass index, LV dilatation, and left atrium dimension. E/e′sr is per 10 cm increase. Composite endpoint: Heart failure, acute myocardial infarction, and cardiovascular death. CI, confidence interval; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; GLS, global longitudinal strain. Figure 2 View largeDownload slide Incidence rate of heart failure, acute myocardial infarction or cardiovascular mortality according to E/e′sr and high/reduced global longitudinal strain groups. Displaying the unadjusted incidence rate of heart failure, acute myocardial infarction or cardiovascular mortality (with 95% confidence intervals) per 1000 person years for the population according to high/reduced global longitudinal strain. A significant linear association was found between E/e′sr and the composite outcome in participants with both reduced and high global longitudinal strain, P < 0.001 for both, with an overall P-value for linear association being 0.015. Non-linear association was tested but was not significant in participants with high global longitudinal strain, P = 0.10 nor for reduced global longitudinal strain, P = 0.16. Hazard rates are calculated per 10 cm increase of E/e′sr. GLS, global longitudinal strain; HR, hazard rate; E/e′sr, ratio of transmitral early filling velocity to early diastolic strain rate. Figure 2 View largeDownload slide Incidence rate of heart failure, acute myocardial infarction or cardiovascular mortality according to E/e′sr and high/reduced global longitudinal strain groups. Displaying the unadjusted incidence rate of heart failure, acute myocardial infarction or cardiovascular mortality (with 95% confidence intervals) per 1000 person years for the population according to high/reduced global longitudinal strain. A significant linear association was found between E/e′sr and the composite outcome in participants with both reduced and high global longitudinal strain, P < 0.001 for both, with an overall P-value for linear association being 0.015. Non-linear association was tested but was not significant in participants with high global longitudinal strain, P = 0.10 nor for reduced global longitudinal strain, P = 0.16. Hazard rates are calculated per 10 cm increase of E/e′sr. GLS, global longitudinal strain; HR, hazard rate; E/e′sr, ratio of transmitral early filling velocity to early diastolic strain rate. Figure 3 View largeDownload slide Title: number of participants meeting the composite outcome during follow-up. Bar diagram displaying the number of participants meeting the composite outcome during follow-up stratified according to quartiles of E/e′sr. Figure 3 View largeDownload slide Title: number of participants meeting the composite outcome during follow-up. Bar diagram displaying the number of participants meeting the composite outcome during follow-up stratified according to quartiles of E/e′sr. Results Mean age of the study sample was 56.9 ± 16.2 years and 42.2% were male (Table 1). During the follow-up period (median 11.0 years, Interquartile range (IQR): 9.9–11.2 years), 72 (5.8%) participants were admitted due to HF, 41 (3.3%) were admitted due to a MI, and 71 (5.7%) of the participants died due to cardiovascular causes. In total, 140 (11.3%) of the participants reached the composite outcome (Figure 3). The median E/e′sr ratio was 0.63 m, and higher E/e′sr was significantly associated with higher age, male sex, higher BMI, higher proportion of smokers, higher cholesterol levels, higher systolic and diastolic blood pressure, reduced LVEF, increased LVMI, longer deceleration time, LV hypertrophy, left atrial diameter, and lower GLS (Table 1). Diastolic measurements (deceleration time, E/e′, and E/A ratio were not included in the models with E/e′sr due to the collinearity of these predictors). Differences in variables between excluded and included participants are shown in Supplementary material online, Table S1. Relationship between E/e′sr and outcome In univariable Cox regression analysis, E/e′sr proved to be significantly associated with the composite outcome [HR 1.17, 95% CI (1.13–1.21); P < 0.001, per 10 cm increase] (Table 2), which was also the case for E/e′ [HR 1.13, 95% CI (1.11–1.15); P < 0.001, per 1 unit increase]. Univariable Cox regression analysis for all variables included in the fully adjusted model is displayed in Supplementary material online, Table S2. After multivariable adjustment for echocardiographic and clinical parameters being age, gender, heart rate, BMI, smoking, previous ischaemic heart disease, systolic blood pressure, diabetes, eGFR, proBNP, LVEF (<50%), LVMI, LV dilatation and left atrium dimension, E/e′sr remained an independent predictor [HR 1.08, 95% CI (1.02–1.13); P = 0.003, per 10 cm increase], whereas E/e′ did not remain an independent predictor [HR 1.03, 95% CI [0.99–1.06] P = 0.11, per 1 unit increase). In competing-risks regression analysis with the composite outcome being the event of interest and non-cardiovascular death the competing event, similar results were found (Supplementary material online, Table S4). Global longitudinal strain modified the relationship between E/e′sr and outcome (P = 0.015) (Figure 2). E/e′sr was a stronger prognosticator in participants with high GLS (≥18%; participants with GLS values above the median in the population) [HR 1.41, 95% CI (1.19–1.66); P < 0.001, per 10 cm increase] when compared to participants with reduced GLS (<18%; participants with GLS values below the median in the population) [HR 1.13 95% CI (1.08–1.19); P < 0.001, per 10 cm increase] (Table 3). After multivariable adjustment E/e′sr was found to be a stronger predictor of the composite outcome in participants with high GLS [HR 1.28, 95% CI (1.06–1.54); P = 0.011 per 10 cm increase], than in participants with reduced GLS [HR 1.08, 95% CI (1.02–1.14); P = 0.012, per 10 cm increase] (Table 3). After the interaction term between E/e′sr and GLS was entered in the fully adjusted model 3 E/e′sr remained an independent predictor of the composite outcome [HR 1.11, 95% CI (1.03–1.20); P = 0.008, per 10 cm increase]. Incremental value of E/e′sr in relation to predicting cardiovascular morbidity and mortality in the general population The SCORE risk chart is currently the primary risk stratification model used for evaluating risk of cardiovascular morbidity and mortality in the general population. In order to assess the incremental value of E/e′sr we added the measurement to the SCORE risk chart prediction model (age, gender, cholesterol level, smoking status, and systolic blood pressure). E/e′sr provided incremental prognostic information in predicting the composite endpoint. Difference in Harrell’s C-statistics was calculated using Somers’ D transformation; the SCORE risk chart without addition of E/e′sr: 0.839 (0.81–0.87) vs. the SCORE risk chart with the addition of E/e′sr: 0.844 (0.82–0.87); P = 0.045. C-statistics index for all variables included are displayed in Supplementary material online, Table S3. Calibration of the model with E/e′sr included was evaluated and found good (P = 0.61) (Supplementary material online, Figure S1). In order to test the additive prognostic value of E/e′sr to GLS in the study population, E/e′sr was added to a univariable Cox model including GLS. This resulted in a significant increase in Harrel’s C-statistics; [0.62 (0.57–0.67) vs. 0.67 (0.62–0.72); P = 0.005]. Discussion This is the first study to assess the prognostic usefulness of E/e′sr in a general population. In this prospective study of a general population who participated in an extensive cardiac examination including echocardiography with speckle tracking and long-term outcome ascertainment, we demonstrate that: (i) LV filling pressure as assessed by E/e′sr is a significant predictor of the composite outcome; MI, HF, and/or CVD independent of clinical and other echocardiographic predictors, (ii) E/e′sr is a stronger prognosticator in participants with good systolic function as opposed to participants with reduced systolic function, (iii) E/e′sr provides incremental prognostic value in predicting the composite cardiovascular (CV) outcome beyond the SCORE risk chart, and (iv) E/e′sr is a superior predictor of cardiovascular morbidity and mortality when compared to E/e′ in the population. The prognostic value of E/e′sr compared with E/e′ It may be hard to fully capture the early regional and global myocardial relaxation abnormalities with annular myocardial velocities assessed by TDI due to various factors. This may be due to the limitations of Doppler methods such as the angle dependency with the risk of significant errors with angulations >20°.16 Using e′ as a velocity-based assessment of the early relaxation sampled from the lateral and septal mitral annulus might not correctly reflect global diastolic LV function. Instead, a global measurement should be used to assess the overall LV function to avoid a misrepresentation of the overall LV function. Our discoveries can be interpreted in the context that deformation-based diastolic evaluation not only superiorly reflects all LV myocardial regions but also potentially offers better discriminative information on pathological injury of the LV. Consequently, indexing E to e′sr possibly yields more information on the global relaxation properties of the myocardium when compared to annular velocity-based measures. This indicates that E/e′sr might be a more sensitive and physiological approach to retrieve information on myocardial relaxation properties and haemodynamics in a large general population. Similarly to the study by Ersbøll et al.,7 we demonstrated the incremental importance of assessing the global myocardial relaxation properties with the use of E/e′sr as a superior alternative to E/e′ in participants with good systolic function assessed by GLS. Additionally, we showed that E/e′sr provides incremental prognostic information over and beyond GLS alone. Left ventricular filling pressure assessed by E/e′sr Previous studies have shown a closer association between E/e′sr and invasively measured LV filling pressure compared with E/e′ in various populations.4,–6,E/e′sr has previously been demonstrated as an incremental prognosticator superior to LVEF and GLS in a large STEMI population.7 In an atrial fibrillation population E/e′sr was demonstrated to be a superior prognosticator to E/e′.17,E/e′sr was found to be superior when compared to E/e′ in predicting outcomes in a population of patients with systolic HF. Furthermore, they found that E/e′sr provided incremental prognostic information when compared with GLS.18 We found similar results in the present report. However, the evidence is not clear at the current moment since a recent study19 did not find E/e′sr to have prognostic value in STEMI patients, which is in contrast to the findings of Ersbøll et al.7 In our study we found that, when examining a large general population, E/e′sr was a superior prognosticator to E/e′, especially in persons with good systolic function as assessed by GLS. This suggests that E/e′sr may not be appropriate as a predictor of cardiac events in persons with overt systolic dysfunction. It seems that E/e′sr could work as an early marker of cardiac pathology in the general population. This may be due to the slowly progressive impairment of active myocardial relaxation in the diastole where the ATP hydrolysis is required for the actin–myosin separation and the calcium dissociation from troponin-C. Furthermore, inappropriate ADP/ATP ratio, changes in cytosolic Ca+ and changed expression in extracellular matrix components of the myocardium have been suggested to affect early diastolic dysfunction.20 These are subtle changes where diastolic function is probably affected before the systolic function. In this way, E/e′sr might be a sensitive marker of early cardiac dysfunction. It is important to identify high-risk individuals in the general population. This may improve risk stratification and assist in identifying subjects which could benefit from early intervention to reduce progression of cardiac dysfunction.21 The range of E/e′sr in our study correlates well with previous studies investigating the prognostic value of E/e′sr. Thus, we believe our observed difference in E/e′sr between those who met the composite outcome and those who did not to be of clinical significance.22 Limitations There are several limitations to this study. The study sample was almost exclusively Caucasian, which limits the generalizability of our results to other ethnicities and races. Furthermore, e′ was measured with colour TDI instead of pulsed wave TDI which is the clinically used modality. Global longitudinal strain and/or E/e′sr was not obtained in 916 participants due to either low frame rate and/or inadequate images. However, these images were obtained with outdated ultrasound systems (Vivid 5, GE Healthcare) between 2001 and 2003. This proportion would have been substantially lower, had the study been conducted today with more recent ultrasound systems. E/e′sr proved to be an independent predictor even though the included group of participants had better mean baseline data than participants excluded due to low frame rate and/or inadequate images (Supplementary material online, Table S1), thus underscoring the sensitive value of the measurement. Conclusion In the general population, E/e′sr is an independent predictor of cardiovascular morbidity and mortality. E/e′sr provides incremental prognostic information over and above the current risk assessment model for a composite cardiovascular endpoint. In addition, E/e′sr seems to be a stronger predictor of cardiac events than E/e′ in the general population. Supplementary material Supplementary material is available at European Heart Journal online. Funding Lundbeck Foundation to fund Lundbeck Foundation Clinical Research Fellowship for Mats Højbjerg Lassen. Furthermore, Mats Højbjerg Lassen received a research grant from Gentofte & Herlev Hospital. The sponsors had no role in the study design, data collection, analysis, interpretation, or writing of the article. Conflict of interest: none declared. References 1 WHO | Cardiovascular diseases (CVDs) [Internet]. http://www.who.int/mediacentre/factsheets/fs317/en/ (14 February 2017). 2 Hillis GS, Møller 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: 360– 367. Google Scholar CrossRef Search ADS PubMed  3 Jons C, Joergensen RM, Hassager C, Gang UJ, Dixen U, Johannesen A, Olsen NT, Hansen TF, Messier M, Huikuri HV, Thomsen PE. Diastolic dysfunction predicts new-onset atrial fibrillation and cardiovascular events in patients with acute myocardial infarction and depressed left ventricular systolic function: a CARISMA substudy. Eur J Echocardiogr  2010; 11: 602– 607. Google Scholar CrossRef Search ADS PubMed  4 Dokainish H, Sengupta R, Pillai M, Bobek J, Lakkis N. Usefulness of new diastolic strain and strain rate indexes for the estimation of left ventricular filling pressure. Am J Cardiol  2008; 101: 1504– 1509. Google Scholar CrossRef Search ADS PubMed  5 Wang J, Khoury DS, Thohan V, Torre-Amione G, Nagueh SF. Global diastolic strain rate for the assessment of left ventricular relaxation and filling pressures. Circulation  2007; 115: 1376– 1383. Google Scholar CrossRef Search ADS PubMed  6 Kimura K, Takenaka K, Ebihara A, Okano T, Uno K, Fukuda N, Ando J, Fujita H, Morita H, Yatomi Y, Nagai R. Speckle tracking global strain rate E/E’ predicts LV filling pressure more accurately than traditional tissue Doppler E/E’. Echocardiogr Mt Kisco N  2012; 29: 404– 410. Google Scholar CrossRef Search ADS   7 Ersbøll M, Andersen MJ, Valeur N, Mogensen UM, Fakhri Y, Fahkri Y, Thune JJ, Møller JE, Hassager C, Søgaard P, Køber L. Early diastolic strain rate in relation to systolic and diastolic function and prognosis in acute myocardial infarction: a two-dimensional speckle-tracking study. Eur Heart J  2014; 35: 648– 656. Google Scholar CrossRef Search ADS PubMed  8 Høfsten DE, Løgstrup BB, Møller JE, Pellikka PA, Egstrup K. Abnormal glucose metabolism in acute myocardial infarction: influence on left ventricular function and prognosis. JACC Cardiovasc Imaging  2009; 2: 592– 599. Google Scholar CrossRef Search ADS PubMed  9 Ng ACT, Delgado V, Bertini M, van der Meer RW, Rijzewijk LJ, Hooi Ewe S, Siebelink HM, Smit JW, Diamant M, Romijn JA, de Roos A, Leung DY, Lamb HJ, Bax JJ. Myocardial steatosis and biventricular strain and strain rate imaging in patients with type 2 diabetes mellitus. Circulation  2010; 122: 2538– 2544. Google Scholar CrossRef Search ADS PubMed  10 Biering-Sørensen T, Mogelvang R, Pedersen S, Schnohr P, Sogaard P, Jensen JS. Usefulness of the myocardial performance index determined by tissue Doppler imaging m-mode for predicting mortality in the general population. Am J Cardiol  2011; 107: 478– 483. Google Scholar CrossRef Search ADS PubMed  11 Biering-Sørensen T, Biering-Sørensen SR, Olsen FJ, Sengeløv M, Jørgensen PG, Mogelvang R, Shah AM, Jensen JS. Global longitudinal strain by echocardiography predicts long-term risk of cardiovascular morbidity and mortality in a low-risk general population: the Copenhagen City Heart Study. Circ Cardiovasc Imaging  2017; 10: e005521. Google Scholar CrossRef Search ADS PubMed  12 Biering-Sørensen T, Hoffmann S, Mogelvang R, Zeeberg Iversen A, Galatius S, Fritz-Hansen T, Bech J, Jensen JS. Myocardial strain analysis by 2-dimensional speckle tracking echocardiography improves diagnostics of coronary artery stenosis in stable angina pectoris. Circ Cardiovasc Imaging  2014; 7: 58– 65. Google Scholar CrossRef Search ADS PubMed  13 Cuzick J. A Wilcoxon-type test for trend. Stat Med  1985; 4: 87– 90. Google Scholar CrossRef Search ADS PubMed  14 Harrell FE, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med  1996; 15: 361– 387. Google Scholar CrossRef Search ADS PubMed  15 Conroy RM, Pyörälä K, Fitzgerald AP, Sans S, Menotti A, De Backer G, De Bacquer D, Ducimetière P, Jousilahti P, Keil U, Njølstad I, Oganov RG, Thomsen T, Tunstall-Pedoe H, Tverdal A, Wedel H, Whincup P, Wilhelmsen L, Graham IM. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J  2003; 24: 987– 1003. Google Scholar CrossRef Search ADS PubMed  16 De Boeck BWL, Meine M, Leenders GE, Teske AJ, van Wessel H, Kirkels JH, Prinzen FW, Doevendans PA, Cramer MJ. Practical and conceptual limitations of tissue Doppler imaging to predict reverse remodelling in cardiac resynchronisation therapy. Eur J Heart Fail  2008; 10: 281– 290. Google Scholar CrossRef Search ADS PubMed  17 Lee W-H, Hsu P-C, Chu C-Y, Chen S-C, Su H-M, Lin T-H, Lee C-S, Yen H-W, Voon W-C, Lai W-T, Sheu S-H. The ratio of early mitral inflow velocity to global diastolic strain rate as a useful predictor of cardiac outcomes in patients with atrial fibrillation. J Am Soc Echocardiogr  2014; 27:454–425. 18 Chan YH, Lee HF, Wu LS, Wang CL, Wu CT, Yeh YH, Ho YW, Hsu LA, Chu PH, Kuo CT. Ratio of transmitral early filling velocity to early diastolic strain rate predicts outcomes in patients with systolic heart failure. Eur Heart J Cardiovasc Imaging  2017; 18: 79– 85. Google Scholar CrossRef Search ADS PubMed  19 Shanks M, Ng ACT, van de Veire NRL, Antoni ML, Bertini M, Delgado V, Nucifora G, Holman ER, Choy JB, Leung DY, Schalij MJ, Bax JJ. Incremental prognostic value of novel left ventricular diastolic indexes for prediction of clinical outcome in patients with ST-elevation myocardial infarction. Am J Cardiol  2010; 105: 592– 597. Google Scholar CrossRef Search ADS PubMed  20 Maharaj R. Diastolic dysfunction and heart failure with a preserved ejection fraction: relevance in critical illness and anaesthesia. J Saudi Heart Assoc  2012; 24: 99– 121. Google Scholar CrossRef Search ADS PubMed  21 Cooney MT, Dudina A, D'Agostino R, Graham IM. Cardiovascular risk-estimation systems in primary prevention: do they differ? Do they make a difference? Can we see the future? Circulation  2010; 122: 300– 310. Google Scholar CrossRef Search ADS PubMed  22 Chen SC, Lee WH, Hsu PC, Lee CS, Lee MK, Yen HW, Lin TH, Voon WC, Lai WT, Sheu SH, Su HM. Association of the ratio of early mitral inflow velocity to the global diastolic strain rate with a rapid renal function decline in atrial fibrillation. PLoS One  2016; 11: e0147446. Google Scholar CrossRef Search ADS PubMed  Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Heart Journal Oxford University Press

Ratio of transmitral early filling velocity to early diastolic strain rate predicts long-term risk of cardiovascular morbidity and mortality in the general population

Loading next page...
 
/lp/ou_press/ratio-of-transmitral-early-filling-velocity-to-early-diastolic-strain-YVP0xuE71W
Publisher
Oxford University Press
Copyright
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com.
ISSN
0195-668X
eISSN
1522-9645
D.O.I.
10.1093/eurheartj/ehy164
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims It has previously been demonstrated that the ratio of early mitral inflow velocity to global diastolic strain rate (E/e′sr) is a significant predictor of cardiac events in specific patient populations. The utility of this measurement to predict cardiovascular events in a general population has not been evaluated. Methods and results A total of 1238 participants in a general population study underwent a health examination including echocardiography where global longitudinal strain (GLS) and E/e′sr were determined. The primary endpoint was the composite of incident heart failure (HF), acute myocardial infarction (AMI) or cardiovascular death (CVD). During follow-up (median 11 years), 140 (11.3%) participants reached the composite endpoint. E/e′sr was associated with adverse outcome [HR 1.17 95% CI (1.13–1.21); P < 0.001, per 10 cm increase]. After multivariable adjustment for echocardiographic and clinical parameters, E/e′sr remained an independent predictor of the composite endpoint [HR 1.08, 95% CI (1.02–1.13); P = 0.003] as opposed to E/e′ [HR 1.03, 95% CI (0.99–1.06); P = 0.11 per 1 unit increase]. Global longitudinal strain modified the relationship between E/e′sr and outcome (P for interaction = 0.015). E/e′sr was a stronger predictor in participants with good systolic function as determined by GLS (GLS > 18%) after multivariable adjustment, when compared to participants with reduced systolic function (GLS < 18%) [HR 1.28 95% CI (1.06–1.54); P = 0.011, and HR 1.08 95% CI (1.02–1.14); P = 0.012, respectively). E/e′sr provided incremental information [Harrell’s C-index: 0.839 (0.81–0.87) vs. 0.844 (0.82–0.87); P = 0.045] beyond the SCORE risk chart. Conclusion In the general population, E/e′sr provides independent and incremental prognostic information regarding cardiovascular morbidity and mortality. Additionally, E/e′sr is a stronger predictor of cardiac events than E/e′. Two-dimensional speckle tracking echocardiography, Global longitudinal strain , General population , Diastolic function , Long-term outcome Introduction Cardiovascular death (CVD) is a major cause of mortality, accounting for 31% of all deaths worldwide.1 Hence, early detection of left ventricular (LV) dysfunction in persons at high-risk of developing cardiac events is important. E/e′ as a measure of LV filling pressure has been used to asses impaired diastolic function following acute myocardial infarction (MI) and has proved to be an important predictor of major adverse events.2,3 Two-dimensional (2D) echocardiographic speckle tracking can correctly measure myocardial strain and strain rate. The measurements of global diastolic strain rate (e′sr) and ratio of early mitral inflow velocity (E) to global e′sr (E/e′sr) have been suggested as new measures of elevated LV filling pressure.4–6 These measures have the advantage that global early diastolic strain rate represents the diastolic performance of all segments of the myocardium, whereas tissue Doppler echocardiography may only consider a single myocardial wall. Furthermore, speckle tracking has overcome several technical limitations which tissue Doppler imaging is subject to (e.g. angle dependency). Several studies have demonstrated a close correlation between invasively measured LV filling pressure and E/e′sr.4–6 Myocardial ischaemia leads to impaired active relaxation during diastole and to reduced early diastolic untwisting and ventricular suction. These events affect E/e′sr. Furthermore, E/e′sr is also influenced by co-morbid conditions such as diabetes and hypertension, which affect myocardial relaxation through myocardial hypertrophy and deposition of triglycerides.8,9E/e′sr has not previously been investigated as a prognosticator in addition to existing echocardiographic indices in a large-scale study on a general population. Hence, the aim of this study was to determine the prognostic significance of E/e′sr in the general population. Methods Study population This longitudinal cohort study included 2154 participants who underwent echocardiographic examination including colour tissue Doppler imaging (TDI) and 2D speckle tracking analysis. Exclusion criteria were prevalent diagnosis of heart failure (HF), atrial fibrillation, and inadequate image quality for 2D speckle tracking. After exclusion, 1238 participants remained. Exclusion process is displayed in Figure 1 and has been described in detail in the Supplementary material online, Page 2. The study was performed in accordance to the second Helsinki Declaration and approved by the regional ethics committee. Figure 1 View largeDownload slide Flow diagram showing the exclusion process of the study. Flow diagram of the exclusion process for the Copenhagen City Heart study ending with the included participants in this study. 2D, two-dimensional; TDI, tissue Doppler imaging; E/e′sr, ratio of transmitral early filling velocity to early diastolic strain rate. Figure 1 View largeDownload slide Flow diagram showing the exclusion process of the study. Flow diagram of the exclusion process for the Copenhagen City Heart study ending with the included participants in this study. 2D, two-dimensional; TDI, tissue Doppler imaging; E/e′sr, ratio of transmitral early filling velocity to early diastolic strain rate. Health examination Detailed information is available in the Supplementary material online and has also previously been described elsewhere.10,11 Echocardiography Echocardiography was performed using Vivid 5 ultrasound systems (GE Healthcare, Horten, Norway) with a 2.5-MHz transducer by three experienced sonographers. All participants were examined with complete 2D, colour, and spectral Doppler echocardiography and colour TDI. All echocardiograms were stored on magneto-optical disks and an external FireWire hard drive (LaCie, France) and analysed offline with commercially available post-processing software (EchoPac version 8, GE Medical, Horten, Norway). Conventional echocardiography and colour tissue Doppler imaging Conventional echocardiography and colour TDI was performed as has been previously described10,11. An elaborate description is available in the Supplementary material online, Page 2. Speckle tracking echocardiography Two-dimensional speckle tracking was performed in the 4-chamber, 2-chamber, and 3-chamber apical projections with an average of 57 frames per second (FPS), standard deviation (SD) of 4 FPS. Speckle tracking methodology is described in detail in the Supplementary material online, page 3. The mean values of all LV segments from all three apical views were used to calculate GLS and e′sr. E/e′sr ratio was calculated as E velocity (m/s) divided by the absolute value of global e′sr (s−1). In cases where speckle tracking was deemed unsatisfactory from one of the chamber views, it was obtained from the mean value of the remaining chamber projections. A total of 1094 subjects had 4 chamber views adequate for speckle tracking analysis, 770 had adequate 2 chamber views, and 586 subjects had adequate 3 chamber views for speckle tracking analysis. Thus, all 3 apical views were adequate for strain analysis in 396 participants, 2 in 527, and 1 in 315 participants. Previously, our lab has shown good intra- and interobserver variability of GLS. In the same study, we also found a good intra- and interobserver variability of strain rate e with a small bias (for strain rate e: mean difference ± 1.96 SDs was −0.06 ± 0.25 for the intraobserver analysis and 0.06 ± 0.28 for the interobserver analysis).12 One investigator who was blinded to all other information analysed all 2D echocardiograms by speckle tracking. Follow-up and outcome Follow-up was 100%. The primary endpoints were defined as CVD or admission due to either acute MI or incident HF. The Danish National Board of Health’s National Patient Registry was used to retrieve the endpoints in the Copenhagen City Heart Study by using International classification of diseases codes 10th revision (ICD-10). This is described in more detail elsewhere.10 Furthermore, the National Danish Causes of Death Registry was used to obtain follow-up data regarding CVD. ICD-10 codes I00-I99 were used to define CVD. Statistics STATA statistics/data analysis, SE 12.0 (StataCorp, College Station, TX, USA) was used for all analyses. Statistical significance was defined as P ≤ 0.05. Baseline clinical and echocardiographic data across E/e′sr quartiles (Table 1) were compared by trend tests using linear regression for continuous Gaussian distributed variables and with an extension of the Wilcoxon rank-sum test13 for continuous non-Gaussian distributed variables. The χ2 for trend for proportions were used when relevant. Rates of all events were calculated as the number of events divided by person-time at risk. Figure 2 was constructed using a Poisson model to estimate incidence rates. The association between E/e′sr and the composite outcome was tested using restricted cubic splines with the number of knots selected according to the value associated with the lowest Akaike information criterion value. Both linear and non-linear interaction terms were tested using this method. Cox proportional hazards regression models were used to examine the associations of baseline variables, echocardiographic parameters, and E/e′sr with the risk of the composite endpoint and are displayed in Tables 2 and 3. Univariable Cox proportional hazards regression models were used to obtain Harrell’s c-statistics14 in order to test the prognostic value of E/e′sr for predicting the composite endpoint. Left ventricular ejection fraction (LVEF) and proBNP were dichotomized in high/low categories using LVEF cut-off of 50% and proBNP > 150 pmol/L and included as dichotomous terms when used in the multivariable Cox regression models. Furthermore, Harrell’s C-statistics was used to explore the incremental value of adding E/e′sr and the other variables to the clinical predictors from the SCORE risk chart15 (Supplementary material online, Table S3). Calibration of the model was evaluated with Groennesby–Borgan goodness-of-fit χ2-estimates using deciles. Competing-risks regression models were used to account for non-cardiovascular death as a competing risk to the composite outcome (our event of interest). Table 1 Baseline clinical characteristics for the participants stratified according to quartiles of E/e′sr E/e′sr (m)  All 0.69 ± 0.3 n = 1238  1. Quartile <0.52 n = 310  2. Quartile 0.52–0.63 n = 310  3. Quartile 0.63–0.79 n = 309  4. Quartile >0.79 n = 309  P-value for trend  Demographics   Age (years)  56.9 ± 16.2  53.4 ± 15.7  53.6 ± 16.3  57.3 ± 15.8  63.3 ± 15.0  <0.001   Male gender, n (%)  522 (42.2)  121 (39.0)  133 (42.9)  123 (39.8)  145 (46.9)  0.10  Clinical   Systolic blood pressure (mmHg)  134 ± 23  127 ± 22  131 ± 23  134 ± 21  143 ± 23  <0.001   Diastolic blood pressure (mmHg)  78 ± 12  76 ± 11  78 ± 14  77 ± 12  80 ± 12  <0.001   Heart rate (bpm)  65 ± 10  66 ± 10  65 ± 11  65 ± 10  67 ± 11  0.37   BMI (kg/m2)  25.0 ± 3.6  24.3 ± 3.4  24.1 ± 3.3  25.2 ± 3.4  26.4 ± 3.8  <0.001   Hypertension, n (%)  466 (37.8)  82 (26.6)  97 (31.4)  107 (34.7)  180 (58.3)  <0.001   Diabetes, n (%)  113 (9.1)  29 (9.4)  24 (7.7)  28 (9.1)  32 (10.4)  0.55   Previous ischaemic heart disease, n (%)  58 (4.7)  8 (2.6)  11 (3.5)  10 (3.2)  29 (9.4)  0.001   Smoking status  0.038    Never, n (%)  410 (33.3)  120 (39.1)  98 (31.9)  96 (31.4)  96 (31.4)    Previous, n (%)  401 (33.2)  104 (33.9)  96 (29.6)  97 (31.7)  103 (33.7)    Current, n (%)  415 (33.5)  90 (29.3)  105 (34.2)  113 (36.9)  107 (35.0)  Lab work   Cholesterol (mmol)  5.5 ± 1.2  5.5 ± 1.1  5.4 ± 1.2  5.6 ± 1.2  5.7 ± 1.1  0.007   eGFR (mL/min/1.73 m2)  75.2 (66.0, 86.1)  75.3 (66.2, 85.2)  76.8 (67.3, 88.4)  75.5 (64.9, 87.0)  73.0 (64.3, 83.8)  0.06   Plasma proBNP (pmol/L)  16.0 (8.0, 29.0)  15.0 (5.0, 27.0)  15.0 (7.0, 24.0)  17.0 (8.0, 32.0)  19.0 (9.0, 38.0)  0.001  Echocardiography   LVEF < 50%, n (%)  9 (0.7)  0 (0.0)  0 (0.0)  2 (0.7)  7 (2.3)  0.001   Hypertrophy, n (%)  176 (15.8)  28 (9.7)  30 (10.5)  42 (15.4)  76 (28.4)  0.001   Dilatation, n (%)  43 (3.8)  15 (5.2)  9 (3.1)  10 (3.6)  21 (7.7)  0.16   LVIDd/height (cm/m)  2.8 ± 0.3  2.8 ± 0.3  2.8 ± 0.2  2.8 ± 0.3  2.9 ± 0.3  0.14   Left atrial dimension (cm)  3.4 ± 0.4  3.3 ± 0.4  3.3 ± 0.4  3.4 ± 0.4  3.5 ± 0.4  <0.001   E/A ratio  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.0 (0.8, 1.2)  <0.001   DT (ms)  166.9 ± 41.0  162.1 ± 36.0  166.3 ± 40.2  166.1 ± 40.1  173.2 ± 45.0  0.008   E/e′  9.7 (7.9, 12.2)  8.3 (7.0, 10.0)  8.9 (7.6, 11.1)  10.1 (8.4, 12.4)  11.9 (9.8, 15.6)  <0.001   LVMI (g/m2)  82.2 (70.9, 96.9)  78.6 (68.0, 91.6)  81.2 (71.3, 95.3)  81.4 (70.3, 97.7)  87.0 (74.8, 104.5)  <0.001   LVEF (%)  59.8 ± 1.7  59.9 ± 0.6  59.8 ± 1.1  59.9 ± 1.1  59.4 ± 2.9  <0.001   GLS (%)  18.1 ± 3.7  20.7 ± 3.1  19.0 ± 2.8  17.5 ± 2.9  15.0 ± 3.1  <0.001  Events during follow-up   AMI admissions during follow-up, n (%)  41 (3.3)  5 (1.6)  8 (2.6)  5 (1.6)  23 (7.4)  <0.001   Heart failure admission during follow-up, n (%)  72 (5.8)  6 (1.9)  12 (3.9)  12 (3.9)  42 (13.6)  <0.001   Cardiovascular death during follow-up, n (%)  71 (5.7)  9 (2.9)  15 (4.8)  14 (4.5)  33 (10.7)  <0.001   Composite endpoint during follow-up, n (%)  140 (11.3)  15 (4.8)  27 (8.7)  25 (8.1)  73 (23.6)  <0.001  E/e′sr (m)  All 0.69 ± 0.3 n = 1238  1. Quartile <0.52 n = 310  2. Quartile 0.52–0.63 n = 310  3. Quartile 0.63–0.79 n = 309  4. Quartile >0.79 n = 309  P-value for trend  Demographics   Age (years)  56.9 ± 16.2  53.4 ± 15.7  53.6 ± 16.3  57.3 ± 15.8  63.3 ± 15.0  <0.001   Male gender, n (%)  522 (42.2)  121 (39.0)  133 (42.9)  123 (39.8)  145 (46.9)  0.10  Clinical   Systolic blood pressure (mmHg)  134 ± 23  127 ± 22  131 ± 23  134 ± 21  143 ± 23  <0.001   Diastolic blood pressure (mmHg)  78 ± 12  76 ± 11  78 ± 14  77 ± 12  80 ± 12  <0.001   Heart rate (bpm)  65 ± 10  66 ± 10  65 ± 11  65 ± 10  67 ± 11  0.37   BMI (kg/m2)  25.0 ± 3.6  24.3 ± 3.4  24.1 ± 3.3  25.2 ± 3.4  26.4 ± 3.8  <0.001   Hypertension, n (%)  466 (37.8)  82 (26.6)  97 (31.4)  107 (34.7)  180 (58.3)  <0.001   Diabetes, n (%)  113 (9.1)  29 (9.4)  24 (7.7)  28 (9.1)  32 (10.4)  0.55   Previous ischaemic heart disease, n (%)  58 (4.7)  8 (2.6)  11 (3.5)  10 (3.2)  29 (9.4)  0.001   Smoking status  0.038    Never, n (%)  410 (33.3)  120 (39.1)  98 (31.9)  96 (31.4)  96 (31.4)    Previous, n (%)  401 (33.2)  104 (33.9)  96 (29.6)  97 (31.7)  103 (33.7)    Current, n (%)  415 (33.5)  90 (29.3)  105 (34.2)  113 (36.9)  107 (35.0)  Lab work   Cholesterol (mmol)  5.5 ± 1.2  5.5 ± 1.1  5.4 ± 1.2  5.6 ± 1.2  5.7 ± 1.1  0.007   eGFR (mL/min/1.73 m2)  75.2 (66.0, 86.1)  75.3 (66.2, 85.2)  76.8 (67.3, 88.4)  75.5 (64.9, 87.0)  73.0 (64.3, 83.8)  0.06   Plasma proBNP (pmol/L)  16.0 (8.0, 29.0)  15.0 (5.0, 27.0)  15.0 (7.0, 24.0)  17.0 (8.0, 32.0)  19.0 (9.0, 38.0)  0.001  Echocardiography   LVEF < 50%, n (%)  9 (0.7)  0 (0.0)  0 (0.0)  2 (0.7)  7 (2.3)  0.001   Hypertrophy, n (%)  176 (15.8)  28 (9.7)  30 (10.5)  42 (15.4)  76 (28.4)  0.001   Dilatation, n (%)  43 (3.8)  15 (5.2)  9 (3.1)  10 (3.6)  21 (7.7)  0.16   LVIDd/height (cm/m)  2.8 ± 0.3  2.8 ± 0.3  2.8 ± 0.2  2.8 ± 0.3  2.9 ± 0.3  0.14   Left atrial dimension (cm)  3.4 ± 0.4  3.3 ± 0.4  3.3 ± 0.4  3.4 ± 0.4  3.5 ± 0.4  <0.001   E/A ratio  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.0 (0.8, 1.2)  <0.001   DT (ms)  166.9 ± 41.0  162.1 ± 36.0  166.3 ± 40.2  166.1 ± 40.1  173.2 ± 45.0  0.008   E/e′  9.7 (7.9, 12.2)  8.3 (7.0, 10.0)  8.9 (7.6, 11.1)  10.1 (8.4, 12.4)  11.9 (9.8, 15.6)  <0.001   LVMI (g/m2)  82.2 (70.9, 96.9)  78.6 (68.0, 91.6)  81.2 (71.3, 95.3)  81.4 (70.3, 97.7)  87.0 (74.8, 104.5)  <0.001   LVEF (%)  59.8 ± 1.7  59.9 ± 0.6  59.8 ± 1.1  59.9 ± 1.1  59.4 ± 2.9  <0.001   GLS (%)  18.1 ± 3.7  20.7 ± 3.1  19.0 ± 2.8  17.5 ± 2.9  15.0 ± 3.1  <0.001  Events during follow-up   AMI admissions during follow-up, n (%)  41 (3.3)  5 (1.6)  8 (2.6)  5 (1.6)  23 (7.4)  <0.001   Heart failure admission during follow-up, n (%)  72 (5.8)  6 (1.9)  12 (3.9)  12 (3.9)  42 (13.6)  <0.001   Cardiovascular death during follow-up, n (%)  71 (5.7)  9 (2.9)  15 (4.8)  14 (4.5)  33 (10.7)  <0.001   Composite endpoint during follow-up, n (%)  140 (11.3)  15 (4.8)  27 (8.7)  25 (8.1)  73 (23.6)  <0.001  A, peak transmitral late diastolic inflow velocity; AMI, acute myocardial infarction; BMI, body mass index; DT, deceleration time of early diastolic inflow; E, peak transmitral early diastolic inflow velocity; e′, average peak early diastolic longitudinal mitral annular velocity determined by colour TDI; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; LVIDd, left ventricular internal diameter at end-diastole; LVMI, left ventricular mass index; proBNP, pro B-type natriuretic peptide. Table 1 Baseline clinical characteristics for the participants stratified according to quartiles of E/e′sr E/e′sr (m)  All 0.69 ± 0.3 n = 1238  1. Quartile <0.52 n = 310  2. Quartile 0.52–0.63 n = 310  3. Quartile 0.63–0.79 n = 309  4. Quartile >0.79 n = 309  P-value for trend  Demographics   Age (years)  56.9 ± 16.2  53.4 ± 15.7  53.6 ± 16.3  57.3 ± 15.8  63.3 ± 15.0  <0.001   Male gender, n (%)  522 (42.2)  121 (39.0)  133 (42.9)  123 (39.8)  145 (46.9)  0.10  Clinical   Systolic blood pressure (mmHg)  134 ± 23  127 ± 22  131 ± 23  134 ± 21  143 ± 23  <0.001   Diastolic blood pressure (mmHg)  78 ± 12  76 ± 11  78 ± 14  77 ± 12  80 ± 12  <0.001   Heart rate (bpm)  65 ± 10  66 ± 10  65 ± 11  65 ± 10  67 ± 11  0.37   BMI (kg/m2)  25.0 ± 3.6  24.3 ± 3.4  24.1 ± 3.3  25.2 ± 3.4  26.4 ± 3.8  <0.001   Hypertension, n (%)  466 (37.8)  82 (26.6)  97 (31.4)  107 (34.7)  180 (58.3)  <0.001   Diabetes, n (%)  113 (9.1)  29 (9.4)  24 (7.7)  28 (9.1)  32 (10.4)  0.55   Previous ischaemic heart disease, n (%)  58 (4.7)  8 (2.6)  11 (3.5)  10 (3.2)  29 (9.4)  0.001   Smoking status  0.038    Never, n (%)  410 (33.3)  120 (39.1)  98 (31.9)  96 (31.4)  96 (31.4)    Previous, n (%)  401 (33.2)  104 (33.9)  96 (29.6)  97 (31.7)  103 (33.7)    Current, n (%)  415 (33.5)  90 (29.3)  105 (34.2)  113 (36.9)  107 (35.0)  Lab work   Cholesterol (mmol)  5.5 ± 1.2  5.5 ± 1.1  5.4 ± 1.2  5.6 ± 1.2  5.7 ± 1.1  0.007   eGFR (mL/min/1.73 m2)  75.2 (66.0, 86.1)  75.3 (66.2, 85.2)  76.8 (67.3, 88.4)  75.5 (64.9, 87.0)  73.0 (64.3, 83.8)  0.06   Plasma proBNP (pmol/L)  16.0 (8.0, 29.0)  15.0 (5.0, 27.0)  15.0 (7.0, 24.0)  17.0 (8.0, 32.0)  19.0 (9.0, 38.0)  0.001  Echocardiography   LVEF < 50%, n (%)  9 (0.7)  0 (0.0)  0 (0.0)  2 (0.7)  7 (2.3)  0.001   Hypertrophy, n (%)  176 (15.8)  28 (9.7)  30 (10.5)  42 (15.4)  76 (28.4)  0.001   Dilatation, n (%)  43 (3.8)  15 (5.2)  9 (3.1)  10 (3.6)  21 (7.7)  0.16   LVIDd/height (cm/m)  2.8 ± 0.3  2.8 ± 0.3  2.8 ± 0.2  2.8 ± 0.3  2.9 ± 0.3  0.14   Left atrial dimension (cm)  3.4 ± 0.4  3.3 ± 0.4  3.3 ± 0.4  3.4 ± 0.4  3.5 ± 0.4  <0.001   E/A ratio  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.0 (0.8, 1.2)  <0.001   DT (ms)  166.9 ± 41.0  162.1 ± 36.0  166.3 ± 40.2  166.1 ± 40.1  173.2 ± 45.0  0.008   E/e′  9.7 (7.9, 12.2)  8.3 (7.0, 10.0)  8.9 (7.6, 11.1)  10.1 (8.4, 12.4)  11.9 (9.8, 15.6)  <0.001   LVMI (g/m2)  82.2 (70.9, 96.9)  78.6 (68.0, 91.6)  81.2 (71.3, 95.3)  81.4 (70.3, 97.7)  87.0 (74.8, 104.5)  <0.001   LVEF (%)  59.8 ± 1.7  59.9 ± 0.6  59.8 ± 1.1  59.9 ± 1.1  59.4 ± 2.9  <0.001   GLS (%)  18.1 ± 3.7  20.7 ± 3.1  19.0 ± 2.8  17.5 ± 2.9  15.0 ± 3.1  <0.001  Events during follow-up   AMI admissions during follow-up, n (%)  41 (3.3)  5 (1.6)  8 (2.6)  5 (1.6)  23 (7.4)  <0.001   Heart failure admission during follow-up, n (%)  72 (5.8)  6 (1.9)  12 (3.9)  12 (3.9)  42 (13.6)  <0.001   Cardiovascular death during follow-up, n (%)  71 (5.7)  9 (2.9)  15 (4.8)  14 (4.5)  33 (10.7)  <0.001   Composite endpoint during follow-up, n (%)  140 (11.3)  15 (4.8)  27 (8.7)  25 (8.1)  73 (23.6)  <0.001  E/e′sr (m)  All 0.69 ± 0.3 n = 1238  1. Quartile <0.52 n = 310  2. Quartile 0.52–0.63 n = 310  3. Quartile 0.63–0.79 n = 309  4. Quartile >0.79 n = 309  P-value for trend  Demographics   Age (years)  56.9 ± 16.2  53.4 ± 15.7  53.6 ± 16.3  57.3 ± 15.8  63.3 ± 15.0  <0.001   Male gender, n (%)  522 (42.2)  121 (39.0)  133 (42.9)  123 (39.8)  145 (46.9)  0.10  Clinical   Systolic blood pressure (mmHg)  134 ± 23  127 ± 22  131 ± 23  134 ± 21  143 ± 23  <0.001   Diastolic blood pressure (mmHg)  78 ± 12  76 ± 11  78 ± 14  77 ± 12  80 ± 12  <0.001   Heart rate (bpm)  65 ± 10  66 ± 10  65 ± 11  65 ± 10  67 ± 11  0.37   BMI (kg/m2)  25.0 ± 3.6  24.3 ± 3.4  24.1 ± 3.3  25.2 ± 3.4  26.4 ± 3.8  <0.001   Hypertension, n (%)  466 (37.8)  82 (26.6)  97 (31.4)  107 (34.7)  180 (58.3)  <0.001   Diabetes, n (%)  113 (9.1)  29 (9.4)  24 (7.7)  28 (9.1)  32 (10.4)  0.55   Previous ischaemic heart disease, n (%)  58 (4.7)  8 (2.6)  11 (3.5)  10 (3.2)  29 (9.4)  0.001   Smoking status  0.038    Never, n (%)  410 (33.3)  120 (39.1)  98 (31.9)  96 (31.4)  96 (31.4)    Previous, n (%)  401 (33.2)  104 (33.9)  96 (29.6)  97 (31.7)  103 (33.7)    Current, n (%)  415 (33.5)  90 (29.3)  105 (34.2)  113 (36.9)  107 (35.0)  Lab work   Cholesterol (mmol)  5.5 ± 1.2  5.5 ± 1.1  5.4 ± 1.2  5.6 ± 1.2  5.7 ± 1.1  0.007   eGFR (mL/min/1.73 m2)  75.2 (66.0, 86.1)  75.3 (66.2, 85.2)  76.8 (67.3, 88.4)  75.5 (64.9, 87.0)  73.0 (64.3, 83.8)  0.06   Plasma proBNP (pmol/L)  16.0 (8.0, 29.0)  15.0 (5.0, 27.0)  15.0 (7.0, 24.0)  17.0 (8.0, 32.0)  19.0 (9.0, 38.0)  0.001  Echocardiography   LVEF < 50%, n (%)  9 (0.7)  0 (0.0)  0 (0.0)  2 (0.7)  7 (2.3)  0.001   Hypertrophy, n (%)  176 (15.8)  28 (9.7)  30 (10.5)  42 (15.4)  76 (28.4)  0.001   Dilatation, n (%)  43 (3.8)  15 (5.2)  9 (3.1)  10 (3.6)  21 (7.7)  0.16   LVIDd/height (cm/m)  2.8 ± 0.3  2.8 ± 0.3  2.8 ± 0.2  2.8 ± 0.3  2.9 ± 0.3  0.14   Left atrial dimension (cm)  3.4 ± 0.4  3.3 ± 0.4  3.3 ± 0.4  3.4 ± 0.4  3.5 ± 0.4  <0.001   E/A ratio  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.1 (0.9, 1.4)  1.0 (0.8, 1.2)  <0.001   DT (ms)  166.9 ± 41.0  162.1 ± 36.0  166.3 ± 40.2  166.1 ± 40.1  173.2 ± 45.0  0.008   E/e′  9.7 (7.9, 12.2)  8.3 (7.0, 10.0)  8.9 (7.6, 11.1)  10.1 (8.4, 12.4)  11.9 (9.8, 15.6)  <0.001   LVMI (g/m2)  82.2 (70.9, 96.9)  78.6 (68.0, 91.6)  81.2 (71.3, 95.3)  81.4 (70.3, 97.7)  87.0 (74.8, 104.5)  <0.001   LVEF (%)  59.8 ± 1.7  59.9 ± 0.6  59.8 ± 1.1  59.9 ± 1.1  59.4 ± 2.9  <0.001   GLS (%)  18.1 ± 3.7  20.7 ± 3.1  19.0 ± 2.8  17.5 ± 2.9  15.0 ± 3.1  <0.001  Events during follow-up   AMI admissions during follow-up, n (%)  41 (3.3)  5 (1.6)  8 (2.6)  5 (1.6)  23 (7.4)  <0.001   Heart failure admission during follow-up, n (%)  72 (5.8)  6 (1.9)  12 (3.9)  12 (3.9)  42 (13.6)  <0.001   Cardiovascular death during follow-up, n (%)  71 (5.7)  9 (2.9)  15 (4.8)  14 (4.5)  33 (10.7)  <0.001   Composite endpoint during follow-up, n (%)  140 (11.3)  15 (4.8)  27 (8.7)  25 (8.1)  73 (23.6)  <0.001  A, peak transmitral late diastolic inflow velocity; AMI, acute myocardial infarction; BMI, body mass index; DT, deceleration time of early diastolic inflow; E, peak transmitral early diastolic inflow velocity; e′, average peak early diastolic longitudinal mitral annular velocity determined by colour TDI; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; LVIDd, left ventricular internal diameter at end-diastole; LVMI, left ventricular mass index; proBNP, pro B-type natriuretic peptide. Table 2 E/e′sr as a predictor of long-term outcome in the general population (n = 1238)   Composite endpoint (140 events)  P-value  HF (72 events)  P-value  AMI (41 events)  P-value  Cardiovascular death (71 events)  P-value  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)    E/e′sr per 10 cm increase  1.17 (1.13–1.21) C-stat 0.67  <0.001  1.16 (1.11–1.21) C-stat 0.70  <0.001  1.17 (1.11–1.25) C-stat 0.69  <0.001  1.12 (1.07–1.18) C-stat 0.63  <0.001  Model 1  E/e′sr per 10 cm increase  1.10 (1.06–1.15)  <0.001  1.10 (1.05–1.15)  <0.001  1.12 (1.05–1.19)  0.001  1.05 (0.99–1.12)  0.069  Model 2  E/e′sr per 10 cm increase  1.08 (1.03–1.13)  0.001  1.08 (1.03–1.15)  0.003  1.10 (1.03–1.18)  0.007  1.00 (0.92–1.08)  0.98  Model 3  E/e′sr per 10 cm increase  1.06 (1.01–1.12)  0.018  1.08 (1.01–1.15)  0.027  1.10 (1.01–1.20)  0.021  0.99 (0.90–1.09)  0.82    Composite endpoint (140 events)  P-value  HF (72 events)  P-value  AMI (41 events)  P-value  Cardiovascular death (71 events)  P-value  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)    E/e′sr per 10 cm increase  1.17 (1.13–1.21) C-stat 0.67  <0.001  1.16 (1.11–1.21) C-stat 0.70  <0.001  1.17 (1.11–1.25) C-stat 0.69  <0.001  1.12 (1.07–1.18) C-stat 0.63  <0.001  Model 1  E/e′sr per 10 cm increase  1.10 (1.06–1.15)  <0.001  1.10 (1.05–1.15)  <0.001  1.12 (1.05–1.19)  0.001  1.05 (0.99–1.12)  0.069  Model 2  E/e′sr per 10 cm increase  1.08 (1.03–1.13)  0.001  1.08 (1.03–1.15)  0.003  1.10 (1.03–1.18)  0.007  1.00 (0.92–1.08)  0.98  Model 3  E/e′sr per 10 cm increase  1.06 (1.01–1.12)  0.018  1.08 (1.01–1.15)  0.027  1.10 (1.01–1.20)  0.021  0.99 (0.90–1.09)  0.82  Model 1 is adjusted for age and gender. Model 2 is adjusted for age, gender, heart rate, BMI, diabetes, eGFR, smoking, previous ischaemic heart disease, systolic blood pressure, and proBNP (>150 pmol/L). Model 3 is adjusted for the same variables as Model 2 and additionally for left ventricular (LV) ejection fraction (<50%), LV mass index, LV dilatation, and left atrium dimension. E/e′sr is per 10 cm increase. AMI, acute myocardial infarction; CI, confidence interval; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; HF, heart failure. Table 2 E/e′sr as a predictor of long-term outcome in the general population (n = 1238)   Composite endpoint (140 events)  P-value  HF (72 events)  P-value  AMI (41 events)  P-value  Cardiovascular death (71 events)  P-value  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)    E/e′sr per 10 cm increase  1.17 (1.13–1.21) C-stat 0.67  <0.001  1.16 (1.11–1.21) C-stat 0.70  <0.001  1.17 (1.11–1.25) C-stat 0.69  <0.001  1.12 (1.07–1.18) C-stat 0.63  <0.001  Model 1  E/e′sr per 10 cm increase  1.10 (1.06–1.15)  <0.001  1.10 (1.05–1.15)  <0.001  1.12 (1.05–1.19)  0.001  1.05 (0.99–1.12)  0.069  Model 2  E/e′sr per 10 cm increase  1.08 (1.03–1.13)  0.001  1.08 (1.03–1.15)  0.003  1.10 (1.03–1.18)  0.007  1.00 (0.92–1.08)  0.98  Model 3  E/e′sr per 10 cm increase  1.06 (1.01–1.12)  0.018  1.08 (1.01–1.15)  0.027  1.10 (1.01–1.20)  0.021  0.99 (0.90–1.09)  0.82    Composite endpoint (140 events)  P-value  HF (72 events)  P-value  AMI (41 events)  P-value  Cardiovascular death (71 events)  P-value  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  Hazard ratio (95% CI)    E/e′sr per 10 cm increase  1.17 (1.13–1.21) C-stat 0.67  <0.001  1.16 (1.11–1.21) C-stat 0.70  <0.001  1.17 (1.11–1.25) C-stat 0.69  <0.001  1.12 (1.07–1.18) C-stat 0.63  <0.001  Model 1  E/e′sr per 10 cm increase  1.10 (1.06–1.15)  <0.001  1.10 (1.05–1.15)  <0.001  1.12 (1.05–1.19)  0.001  1.05 (0.99–1.12)  0.069  Model 2  E/e′sr per 10 cm increase  1.08 (1.03–1.13)  0.001  1.08 (1.03–1.15)  0.003  1.10 (1.03–1.18)  0.007  1.00 (0.92–1.08)  0.98  Model 3  E/e′sr per 10 cm increase  1.06 (1.01–1.12)  0.018  1.08 (1.01–1.15)  0.027  1.10 (1.01–1.20)  0.021  0.99 (0.90–1.09)  0.82  Model 1 is adjusted for age and gender. Model 2 is adjusted for age, gender, heart rate, BMI, diabetes, eGFR, smoking, previous ischaemic heart disease, systolic blood pressure, and proBNP (>150 pmol/L). Model 3 is adjusted for the same variables as Model 2 and additionally for left ventricular (LV) ejection fraction (<50%), LV mass index, LV dilatation, and left atrium dimension. E/e′sr is per 10 cm increase. AMI, acute myocardial infarction; CI, confidence interval; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; HF, heart failure. Table 3 E/e′sr as a predictor of long-term outcome in persons with good systolic function (global longitudinal strain > 18%) or reduced systolic function (global longitudinal strain < 18%) from the general population   Good systolic cardiac function (GLS > 18%) (n = 617)  P-value  Reduced systolic function (GLS < 18%) (n = 621)  P-value  Composite endpoint (46 events)  Composite endpoint (94 events)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  E/e′sr per 10 cm increase  1.41 (1.19–1.66)  <0.001  1.13 (1.08–1.19)  <0.001  C-stat 0.62  C-stat 0.64  Model 1  E/e′sr per 10 cm increase  1.21 (1.04–1.41)  0.015  1.08 (1.03–1.13)  0.001  Model 2  E/e′sr per 10 cm increase  1.16 (1.01–1.37)  0.037  1.08 (1.02–1.14)  0.004  Model 3  E/e′sr per 10 cm increase  1.28 (1.06–1.54)  0.011  1.08 (1.02–1.14)  0.012    Good systolic cardiac function (GLS > 18%) (n = 617)  P-value  Reduced systolic function (GLS < 18%) (n = 621)  P-value  Composite endpoint (46 events)  Composite endpoint (94 events)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  E/e′sr per 10 cm increase  1.41 (1.19–1.66)  <0.001  1.13 (1.08–1.19)  <0.001  C-stat 0.62  C-stat 0.64  Model 1  E/e′sr per 10 cm increase  1.21 (1.04–1.41)  0.015  1.08 (1.03–1.13)  0.001  Model 2  E/e′sr per 10 cm increase  1.16 (1.01–1.37)  0.037  1.08 (1.02–1.14)  0.004  Model 3  E/e′sr per 10 cm increase  1.28 (1.06–1.54)  0.011  1.08 (1.02–1.14)  0.012  Model 1 is adjusted for age and gender. Model 2 is adjusted for age, gender, heart rate, BMI, diabetes, eGFR, smoking, previous ischaemic heart disease, systolic blood pressure, and proBNP (>150 pmol/L). Model 3 is adjusted for the same variables as Model 2 and additionally for left ventricular (LV) ejection fraction (<50%), LV mass index, LV dilatation, and left atrium dimension. E/e′sr is per 10 cm increase. Composite endpoint: Heart failure, acute myocardial infarction, and cardiovascular death. CI, confidence interval; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; GLS, global longitudinal strain. Table 3 E/e′sr as a predictor of long-term outcome in persons with good systolic function (global longitudinal strain > 18%) or reduced systolic function (global longitudinal strain < 18%) from the general population   Good systolic cardiac function (GLS > 18%) (n = 617)  P-value  Reduced systolic function (GLS < 18%) (n = 621)  P-value  Composite endpoint (46 events)  Composite endpoint (94 events)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  E/e′sr per 10 cm increase  1.41 (1.19–1.66)  <0.001  1.13 (1.08–1.19)  <0.001  C-stat 0.62  C-stat 0.64  Model 1  E/e′sr per 10 cm increase  1.21 (1.04–1.41)  0.015  1.08 (1.03–1.13)  0.001  Model 2  E/e′sr per 10 cm increase  1.16 (1.01–1.37)  0.037  1.08 (1.02–1.14)  0.004  Model 3  E/e′sr per 10 cm increase  1.28 (1.06–1.54)  0.011  1.08 (1.02–1.14)  0.012    Good systolic cardiac function (GLS > 18%) (n = 617)  P-value  Reduced systolic function (GLS < 18%) (n = 621)  P-value  Composite endpoint (46 events)  Composite endpoint (94 events)  Hazard ratio (95% CI)  Hazard ratio (95% CI)  E/e′sr per 10 cm increase  1.41 (1.19–1.66)  <0.001  1.13 (1.08–1.19)  <0.001  C-stat 0.62  C-stat 0.64  Model 1  E/e′sr per 10 cm increase  1.21 (1.04–1.41)  0.015  1.08 (1.03–1.13)  0.001  Model 2  E/e′sr per 10 cm increase  1.16 (1.01–1.37)  0.037  1.08 (1.02–1.14)  0.004  Model 3  E/e′sr per 10 cm increase  1.28 (1.06–1.54)  0.011  1.08 (1.02–1.14)  0.012  Model 1 is adjusted for age and gender. Model 2 is adjusted for age, gender, heart rate, BMI, diabetes, eGFR, smoking, previous ischaemic heart disease, systolic blood pressure, and proBNP (>150 pmol/L). Model 3 is adjusted for the same variables as Model 2 and additionally for left ventricular (LV) ejection fraction (<50%), LV mass index, LV dilatation, and left atrium dimension. E/e′sr is per 10 cm increase. Composite endpoint: Heart failure, acute myocardial infarction, and cardiovascular death. CI, confidence interval; E/e′sr, ratio of peak transmitral early diastolic inflow velocity to global early diastolic strain rate; GLS, global longitudinal strain. Figure 2 View largeDownload slide Incidence rate of heart failure, acute myocardial infarction or cardiovascular mortality according to E/e′sr and high/reduced global longitudinal strain groups. Displaying the unadjusted incidence rate of heart failure, acute myocardial infarction or cardiovascular mortality (with 95% confidence intervals) per 1000 person years for the population according to high/reduced global longitudinal strain. A significant linear association was found between E/e′sr and the composite outcome in participants with both reduced and high global longitudinal strain, P < 0.001 for both, with an overall P-value for linear association being 0.015. Non-linear association was tested but was not significant in participants with high global longitudinal strain, P = 0.10 nor for reduced global longitudinal strain, P = 0.16. Hazard rates are calculated per 10 cm increase of E/e′sr. GLS, global longitudinal strain; HR, hazard rate; E/e′sr, ratio of transmitral early filling velocity to early diastolic strain rate. Figure 2 View largeDownload slide Incidence rate of heart failure, acute myocardial infarction or cardiovascular mortality according to E/e′sr and high/reduced global longitudinal strain groups. Displaying the unadjusted incidence rate of heart failure, acute myocardial infarction or cardiovascular mortality (with 95% confidence intervals) per 1000 person years for the population according to high/reduced global longitudinal strain. A significant linear association was found between E/e′sr and the composite outcome in participants with both reduced and high global longitudinal strain, P < 0.001 for both, with an overall P-value for linear association being 0.015. Non-linear association was tested but was not significant in participants with high global longitudinal strain, P = 0.10 nor for reduced global longitudinal strain, P = 0.16. Hazard rates are calculated per 10 cm increase of E/e′sr. GLS, global longitudinal strain; HR, hazard rate; E/e′sr, ratio of transmitral early filling velocity to early diastolic strain rate. Figure 3 View largeDownload slide Title: number of participants meeting the composite outcome during follow-up. Bar diagram displaying the number of participants meeting the composite outcome during follow-up stratified according to quartiles of E/e′sr. Figure 3 View largeDownload slide Title: number of participants meeting the composite outcome during follow-up. Bar diagram displaying the number of participants meeting the composite outcome during follow-up stratified according to quartiles of E/e′sr. Results Mean age of the study sample was 56.9 ± 16.2 years and 42.2% were male (Table 1). During the follow-up period (median 11.0 years, Interquartile range (IQR): 9.9–11.2 years), 72 (5.8%) participants were admitted due to HF, 41 (3.3%) were admitted due to a MI, and 71 (5.7%) of the participants died due to cardiovascular causes. In total, 140 (11.3%) of the participants reached the composite outcome (Figure 3). The median E/e′sr ratio was 0.63 m, and higher E/e′sr was significantly associated with higher age, male sex, higher BMI, higher proportion of smokers, higher cholesterol levels, higher systolic and diastolic blood pressure, reduced LVEF, increased LVMI, longer deceleration time, LV hypertrophy, left atrial diameter, and lower GLS (Table 1). Diastolic measurements (deceleration time, E/e′, and E/A ratio were not included in the models with E/e′sr due to the collinearity of these predictors). Differences in variables between excluded and included participants are shown in Supplementary material online, Table S1. Relationship between E/e′sr and outcome In univariable Cox regression analysis, E/e′sr proved to be significantly associated with the composite outcome [HR 1.17, 95% CI (1.13–1.21); P < 0.001, per 10 cm increase] (Table 2), which was also the case for E/e′ [HR 1.13, 95% CI (1.11–1.15); P < 0.001, per 1 unit increase]. Univariable Cox regression analysis for all variables included in the fully adjusted model is displayed in Supplementary material online, Table S2. After multivariable adjustment for echocardiographic and clinical parameters being age, gender, heart rate, BMI, smoking, previous ischaemic heart disease, systolic blood pressure, diabetes, eGFR, proBNP, LVEF (<50%), LVMI, LV dilatation and left atrium dimension, E/e′sr remained an independent predictor [HR 1.08, 95% CI (1.02–1.13); P = 0.003, per 10 cm increase], whereas E/e′ did not remain an independent predictor [HR 1.03, 95% CI [0.99–1.06] P = 0.11, per 1 unit increase). In competing-risks regression analysis with the composite outcome being the event of interest and non-cardiovascular death the competing event, similar results were found (Supplementary material online, Table S4). Global longitudinal strain modified the relationship between E/e′sr and outcome (P = 0.015) (Figure 2). E/e′sr was a stronger prognosticator in participants with high GLS (≥18%; participants with GLS values above the median in the population) [HR 1.41, 95% CI (1.19–1.66); P < 0.001, per 10 cm increase] when compared to participants with reduced GLS (<18%; participants with GLS values below the median in the population) [HR 1.13 95% CI (1.08–1.19); P < 0.001, per 10 cm increase] (Table 3). After multivariable adjustment E/e′sr was found to be a stronger predictor of the composite outcome in participants with high GLS [HR 1.28, 95% CI (1.06–1.54); P = 0.011 per 10 cm increase], than in participants with reduced GLS [HR 1.08, 95% CI (1.02–1.14); P = 0.012, per 10 cm increase] (Table 3). After the interaction term between E/e′sr and GLS was entered in the fully adjusted model 3 E/e′sr remained an independent predictor of the composite outcome [HR 1.11, 95% CI (1.03–1.20); P = 0.008, per 10 cm increase]. Incremental value of E/e′sr in relation to predicting cardiovascular morbidity and mortality in the general population The SCORE risk chart is currently the primary risk stratification model used for evaluating risk of cardiovascular morbidity and mortality in the general population. In order to assess the incremental value of E/e′sr we added the measurement to the SCORE risk chart prediction model (age, gender, cholesterol level, smoking status, and systolic blood pressure). E/e′sr provided incremental prognostic information in predicting the composite endpoint. Difference in Harrell’s C-statistics was calculated using Somers’ D transformation; the SCORE risk chart without addition of E/e′sr: 0.839 (0.81–0.87) vs. the SCORE risk chart with the addition of E/e′sr: 0.844 (0.82–0.87); P = 0.045. C-statistics index for all variables included are displayed in Supplementary material online, Table S3. Calibration of the model with E/e′sr included was evaluated and found good (P = 0.61) (Supplementary material online, Figure S1). In order to test the additive prognostic value of E/e′sr to GLS in the study population, E/e′sr was added to a univariable Cox model including GLS. This resulted in a significant increase in Harrel’s C-statistics; [0.62 (0.57–0.67) vs. 0.67 (0.62–0.72); P = 0.005]. Discussion This is the first study to assess the prognostic usefulness of E/e′sr in a general population. In this prospective study of a general population who participated in an extensive cardiac examination including echocardiography with speckle tracking and long-term outcome ascertainment, we demonstrate that: (i) LV filling pressure as assessed by E/e′sr is a significant predictor of the composite outcome; MI, HF, and/or CVD independent of clinical and other echocardiographic predictors, (ii) E/e′sr is a stronger prognosticator in participants with good systolic function as opposed to participants with reduced systolic function, (iii) E/e′sr provides incremental prognostic value in predicting the composite cardiovascular (CV) outcome beyond the SCORE risk chart, and (iv) E/e′sr is a superior predictor of cardiovascular morbidity and mortality when compared to E/e′ in the population. The prognostic value of E/e′sr compared with E/e′ It may be hard to fully capture the early regional and global myocardial relaxation abnormalities with annular myocardial velocities assessed by TDI due to various factors. This may be due to the limitations of Doppler methods such as the angle dependency with the risk of significant errors with angulations >20°.16 Using e′ as a velocity-based assessment of the early relaxation sampled from the lateral and septal mitral annulus might not correctly reflect global diastolic LV function. Instead, a global measurement should be used to assess the overall LV function to avoid a misrepresentation of the overall LV function. Our discoveries can be interpreted in the context that deformation-based diastolic evaluation not only superiorly reflects all LV myocardial regions but also potentially offers better discriminative information on pathological injury of the LV. Consequently, indexing E to e′sr possibly yields more information on the global relaxation properties of the myocardium when compared to annular velocity-based measures. This indicates that E/e′sr might be a more sensitive and physiological approach to retrieve information on myocardial relaxation properties and haemodynamics in a large general population. Similarly to the study by Ersbøll et al.,7 we demonstrated the incremental importance of assessing the global myocardial relaxation properties with the use of E/e′sr as a superior alternative to E/e′ in participants with good systolic function assessed by GLS. Additionally, we showed that E/e′sr provides incremental prognostic information over and beyond GLS alone. Left ventricular filling pressure assessed by E/e′sr Previous studies have shown a closer association between E/e′sr and invasively measured LV filling pressure compared with E/e′ in various populations.4,–6,E/e′sr has previously been demonstrated as an incremental prognosticator superior to LVEF and GLS in a large STEMI population.7 In an atrial fibrillation population E/e′sr was demonstrated to be a superior prognosticator to E/e′.17,E/e′sr was found to be superior when compared to E/e′ in predicting outcomes in a population of patients with systolic HF. Furthermore, they found that E/e′sr provided incremental prognostic information when compared with GLS.18 We found similar results in the present report. However, the evidence is not clear at the current moment since a recent study19 did not find E/e′sr to have prognostic value in STEMI patients, which is in contrast to the findings of Ersbøll et al.7 In our study we found that, when examining a large general population, E/e′sr was a superior prognosticator to E/e′, especially in persons with good systolic function as assessed by GLS. This suggests that E/e′sr may not be appropriate as a predictor of cardiac events in persons with overt systolic dysfunction. It seems that E/e′sr could work as an early marker of cardiac pathology in the general population. This may be due to the slowly progressive impairment of active myocardial relaxation in the diastole where the ATP hydrolysis is required for the actin–myosin separation and the calcium dissociation from troponin-C. Furthermore, inappropriate ADP/ATP ratio, changes in cytosolic Ca+ and changed expression in extracellular matrix components of the myocardium have been suggested to affect early diastolic dysfunction.20 These are subtle changes where diastolic function is probably affected before the systolic function. In this way, E/e′sr might be a sensitive marker of early cardiac dysfunction. It is important to identify high-risk individuals in the general population. This may improve risk stratification and assist in identifying subjects which could benefit from early intervention to reduce progression of cardiac dysfunction.21 The range of E/e′sr in our study correlates well with previous studies investigating the prognostic value of E/e′sr. Thus, we believe our observed difference in E/e′sr between those who met the composite outcome and those who did not to be of clinical significance.22 Limitations There are several limitations to this study. The study sample was almost exclusively Caucasian, which limits the generalizability of our results to other ethnicities and races. Furthermore, e′ was measured with colour TDI instead of pulsed wave TDI which is the clinically used modality. Global longitudinal strain and/or E/e′sr was not obtained in 916 participants due to either low frame rate and/or inadequate images. However, these images were obtained with outdated ultrasound systems (Vivid 5, GE Healthcare) between 2001 and 2003. This proportion would have been substantially lower, had the study been conducted today with more recent ultrasound systems. E/e′sr proved to be an independent predictor even though the included group of participants had better mean baseline data than participants excluded due to low frame rate and/or inadequate images (Supplementary material online, Table S1), thus underscoring the sensitive value of the measurement. Conclusion In the general population, E/e′sr is an independent predictor of cardiovascular morbidity and mortality. E/e′sr provides incremental prognostic information over and above the current risk assessment model for a composite cardiovascular endpoint. In addition, E/e′sr seems to be a stronger predictor of cardiac events than E/e′ in the general population. Supplementary material Supplementary material is available at European Heart Journal online. Funding Lundbeck Foundation to fund Lundbeck Foundation Clinical Research Fellowship for Mats Højbjerg Lassen. Furthermore, Mats Højbjerg Lassen received a research grant from Gentofte & Herlev Hospital. The sponsors had no role in the study design, data collection, analysis, interpretation, or writing of the article. Conflict of interest: none declared. References 1 WHO | Cardiovascular diseases (CVDs) [Internet]. http://www.who.int/mediacentre/factsheets/fs317/en/ (14 February 2017). 2 Hillis GS, Møller 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: 360– 367. Google Scholar CrossRef Search ADS PubMed  3 Jons C, Joergensen RM, Hassager C, Gang UJ, Dixen U, Johannesen A, Olsen NT, Hansen TF, Messier M, Huikuri HV, Thomsen PE. Diastolic dysfunction predicts new-onset atrial fibrillation and cardiovascular events in patients with acute myocardial infarction and depressed left ventricular systolic function: a CARISMA substudy. Eur J Echocardiogr  2010; 11: 602– 607. Google Scholar CrossRef Search ADS PubMed  4 Dokainish H, Sengupta R, Pillai M, Bobek J, Lakkis N. Usefulness of new diastolic strain and strain rate indexes for the estimation of left ventricular filling pressure. Am J Cardiol  2008; 101: 1504– 1509. Google Scholar CrossRef Search ADS PubMed  5 Wang J, Khoury DS, Thohan V, Torre-Amione G, Nagueh SF. Global diastolic strain rate for the assessment of left ventricular relaxation and filling pressures. Circulation  2007; 115: 1376– 1383. Google Scholar CrossRef Search ADS PubMed  6 Kimura K, Takenaka K, Ebihara A, Okano T, Uno K, Fukuda N, Ando J, Fujita H, Morita H, Yatomi Y, Nagai R. Speckle tracking global strain rate E/E’ predicts LV filling pressure more accurately than traditional tissue Doppler E/E’. Echocardiogr Mt Kisco N  2012; 29: 404– 410. Google Scholar CrossRef Search ADS   7 Ersbøll M, Andersen MJ, Valeur N, Mogensen UM, Fakhri Y, Fahkri Y, Thune JJ, Møller JE, Hassager C, Søgaard P, Køber L. Early diastolic strain rate in relation to systolic and diastolic function and prognosis in acute myocardial infarction: a two-dimensional speckle-tracking study. Eur Heart J  2014; 35: 648– 656. Google Scholar CrossRef Search ADS PubMed  8 Høfsten DE, Løgstrup BB, Møller JE, Pellikka PA, Egstrup K. Abnormal glucose metabolism in acute myocardial infarction: influence on left ventricular function and prognosis. JACC Cardiovasc Imaging  2009; 2: 592– 599. Google Scholar CrossRef Search ADS PubMed  9 Ng ACT, Delgado V, Bertini M, van der Meer RW, Rijzewijk LJ, Hooi Ewe S, Siebelink HM, Smit JW, Diamant M, Romijn JA, de Roos A, Leung DY, Lamb HJ, Bax JJ. Myocardial steatosis and biventricular strain and strain rate imaging in patients with type 2 diabetes mellitus. Circulation  2010; 122: 2538– 2544. Google Scholar CrossRef Search ADS PubMed  10 Biering-Sørensen T, Mogelvang R, Pedersen S, Schnohr P, Sogaard P, Jensen JS. Usefulness of the myocardial performance index determined by tissue Doppler imaging m-mode for predicting mortality in the general population. Am J Cardiol  2011; 107: 478– 483. Google Scholar CrossRef Search ADS PubMed  11 Biering-Sørensen T, Biering-Sørensen SR, Olsen FJ, Sengeløv M, Jørgensen PG, Mogelvang R, Shah AM, Jensen JS. Global longitudinal strain by echocardiography predicts long-term risk of cardiovascular morbidity and mortality in a low-risk general population: the Copenhagen City Heart Study. Circ Cardiovasc Imaging  2017; 10: e005521. Google Scholar CrossRef Search ADS PubMed  12 Biering-Sørensen T, Hoffmann S, Mogelvang R, Zeeberg Iversen A, Galatius S, Fritz-Hansen T, Bech J, Jensen JS. Myocardial strain analysis by 2-dimensional speckle tracking echocardiography improves diagnostics of coronary artery stenosis in stable angina pectoris. Circ Cardiovasc Imaging  2014; 7: 58– 65. Google Scholar CrossRef Search ADS PubMed  13 Cuzick J. A Wilcoxon-type test for trend. Stat Med  1985; 4: 87– 90. Google Scholar CrossRef Search ADS PubMed  14 Harrell FE, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med  1996; 15: 361– 387. Google Scholar CrossRef Search ADS PubMed  15 Conroy RM, Pyörälä K, Fitzgerald AP, Sans S, Menotti A, De Backer G, De Bacquer D, Ducimetière P, Jousilahti P, Keil U, Njølstad I, Oganov RG, Thomsen T, Tunstall-Pedoe H, Tverdal A, Wedel H, Whincup P, Wilhelmsen L, Graham IM. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J  2003; 24: 987– 1003. Google Scholar CrossRef Search ADS PubMed  16 De Boeck BWL, Meine M, Leenders GE, Teske AJ, van Wessel H, Kirkels JH, Prinzen FW, Doevendans PA, Cramer MJ. Practical and conceptual limitations of tissue Doppler imaging to predict reverse remodelling in cardiac resynchronisation therapy. Eur J Heart Fail  2008; 10: 281– 290. Google Scholar CrossRef Search ADS PubMed  17 Lee W-H, Hsu P-C, Chu C-Y, Chen S-C, Su H-M, Lin T-H, Lee C-S, Yen H-W, Voon W-C, Lai W-T, Sheu S-H. The ratio of early mitral inflow velocity to global diastolic strain rate as a useful predictor of cardiac outcomes in patients with atrial fibrillation. J Am Soc Echocardiogr  2014; 27:454–425. 18 Chan YH, Lee HF, Wu LS, Wang CL, Wu CT, Yeh YH, Ho YW, Hsu LA, Chu PH, Kuo CT. Ratio of transmitral early filling velocity to early diastolic strain rate predicts outcomes in patients with systolic heart failure. Eur Heart J Cardiovasc Imaging  2017; 18: 79– 85. Google Scholar CrossRef Search ADS PubMed  19 Shanks M, Ng ACT, van de Veire NRL, Antoni ML, Bertini M, Delgado V, Nucifora G, Holman ER, Choy JB, Leung DY, Schalij MJ, Bax JJ. Incremental prognostic value of novel left ventricular diastolic indexes for prediction of clinical outcome in patients with ST-elevation myocardial infarction. Am J Cardiol  2010; 105: 592– 597. Google Scholar CrossRef Search ADS PubMed  20 Maharaj R. Diastolic dysfunction and heart failure with a preserved ejection fraction: relevance in critical illness and anaesthesia. J Saudi Heart Assoc  2012; 24: 99– 121. Google Scholar CrossRef Search ADS PubMed  21 Cooney MT, Dudina A, D'Agostino R, Graham IM. Cardiovascular risk-estimation systems in primary prevention: do they differ? Do they make a difference? Can we see the future? Circulation  2010; 122: 300– 310. Google Scholar CrossRef Search ADS PubMed  22 Chen SC, Lee WH, Hsu PC, Lee CS, Lee MK, Yen HW, Lin TH, Voon WC, Lai WT, Sheu SH, Su HM. Association of the ratio of early mitral inflow velocity to the global diastolic strain rate with a rapid renal function decline in atrial fibrillation. PLoS One  2016; 11: e0147446. Google Scholar CrossRef Search ADS PubMed  Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Heart JournalOxford University Press

Published: Apr 5, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off