If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The prognostic role of E/e’ and e’ in patients without heart failure is undefined.
•
Varying E/e’ and e’ are associated with death and cardiovascular hospitalizations.
•
After adjusting for covariates, E/e’ is not an independent prognosticator.
•
At multivariable analysis only e’ independently and incrementally predicted the outcome.
Abstract
Background
Tissue Doppler imaging (TDI) indices of left ventricular (LV) diastolic function provide incremental prognostic information on mortality and morbidity in the general population and in several clinical scenarios. Their independent, additional role in outpatients with normal LV ejection fraction (LVEF) and without heart failure (HF) is undefined.
Methods
We reviewed clinical and echocardiographic records of 2628 consecutive outpatients 52.8% male, median age 71 years) with LVEF > 50% without concurrent or prior HF, from the Cardiovascular Center of Trieste. We analyzed septal early mitral annular velocity (e’) and its combination with mitral peak early filling velocity (E/e’) in relation to the composite end-point of death and cardiovascular hospitalizations.
Results
During follow-up of 26 months (interquartile range: 12–41), 392 (15%) patients experienced the endpoint (88 deaths). Increasing E/e’ showed an overall association with the clinical end-point (log rank p < 0.02), but with no prognostic difference between the middle and upper tertile. Decreasing e’ also showed an association with the end-point, with a more balanced stepwise risk increase for increasing tertiles (log rank p < 0.01 for all contrasts). At multivariable analysis, E/e’ (either in tertiles or dichotomized according to the threshold of 15) was no longer associated with clinical outcome, whereas e’ independently predicted the combined endpoint [hazard ratio 0.73 (0.53-0.94), p = 0.04]. The prognostic value of e’ was incremental to that of other clinical and echocardiographic variables (p = 0.04).
Conclusions
In outpatients with normal LVEF and without HF, e’ and E/e’ are both associated with clinical end-points, though only e’ is an independent and incremental predictor of outcome. These findings suggest a potential role for e’ as a prognosticator, and spread a cautionary word about the utilization of septal E/e’ alone as a surrogate for a comprehensive assessment of diastolic function in this context.
Tissue Doppler imaging (TDI) indices of left ventricular (LV) diastolic function provide incremental prognostic information on cardiovascular mortality and morbidity in the general population and in several clinical scenarios [
Clinical relevance of the spectral tissue Doppler E/e’ ratio in the management of patients with atrial fibrillation: a comprehensive review of the literature.
]. Early mitral annular velocity assessed by TDI (e’) is a reproducible and relatively load-independent measure of LV myocardial relaxation, and is also determined by restoring forces and filling pressure. TDI e' velocity has been proven to predict fatal and nonfatal cardiovascular events in a general population [
]. When used in conjunction with mitral peak early filling velocity (E), the E/e’ ratio is a widely used non-invasive estimate of LV filling pressure (LVFP) that plays a pivotal role in evaluating diastolic dysfunction [
Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
]. Although the prognostic value of E/e’ is established, a number of studies have raised concerns about its diagnostic accuracy in various cardiovascular conditions [
Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
New echocardiographic predictors of clinical outcome in patients presenting with heart failure and a preserved left ventricular ejection fraction: a subanalysis of the Ka (Karolinska) Ren (Rennes) Study.
Is echocardiographic evaluation of diastolic function useful in determining clinical care? Doppler echocardiography yields dubious estimates of left ventricular diastolic pressures.
Diagnostic accuracy of tissue doppler index E/e’ for evaluating left ventricular filling pressure and diastolic dysfunction/heart failure with preserved ejection fraction: a systematic review and meta-analysis.
Risk stratification in chronic heart failure: independent and incremental prognostic value of echocardiography and brain natriuretic peptide and its N-terminal fragment.
Optimal noninvasive assessment of left ventricular filling pressures: a comparison of tissue Doppler echocardiography and B-type natriuretic peptide in patients with pulmonary artery catheters.
Echocardiographic indices do not reliably track changes in left-sided filling pressure in healthy subjects or patients with heart failure with preserved ejection fraction.
Correlation with invasive left ventricular filling pressures and prognostic relevance of the echocardiographic diastolic parameters used in the 2016 ESC heart failure guidelines and in the 2016 ASE/EACVI recommendations: a systematic review in patients with heart failure with preserved ejection fraction.
Thus, we questioned whether TDI indices of diastolic function might be of clinical relevance in the heterogeneous case mix of outpatients without heart failure (HF) and with normal LV ejection fraction (LVEF), commonly encountered in clinical practice. We explored if e’ and E/e’ portend independent prognostic information in outpatients with normal LVEF and without prevalent HF, referred for echocardiographic and clinical evaluation in an outpatient primary care clinic.
Materials and methods
Study design and population
This retrospective study was based on clinical and echocardiographic records of the Cardiovascular Center of Trieste, the territorial outpatient clinic local health services, performing the majority (87%) of the ambulatory cardiovascular clinical and instrumental examinations offered by the local facilities of the National Health Care System. The Cardiovascular Center has direct access and contributes to the Regional Health Data Warehouse, which includes several regional databases such as the Registry of Births and Deaths, Hospital Discharge, District Healthcare Services (intermediate and home care), and Public Drug Distribution System database. This study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the Institutional Review Board of the University Hospital and Health Services of Trieste, Italy. We thus used the resources of the database of the Cardiovascular Center to identify all patients resident in the Trieste territory who underwent a clinical and echocardiographic evaluation between November 1st, 2009 and October 31st, 2014.
Patients were included if they were 50 years or older, in sinus rhythm, with an adequate echocardiographic examination showing a value of LVEF > 50% at the inclusion echocardiogram, and with no previous or concurrent diagnosis of HF. HF was ruled out by purposely trained clinical cardiologists with years of expertise in the field, implementing clinical evaluation, biomarkers, and multiple cardiac and non-cardiac diagnostic tests, as needed according to the European Society of Cardiology guidelines and based on an organizational approach recently resumed in a national consensus document [
2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC)developed with the special contribution of the Heart Failure Association (HFA) of the ESC.
]. Exclusion criteria included unstable angina or indication for coronary revascularization within 30 days before or after the inclusion echocardiogram, sinus tachycardia (>100 bpm), complete left bundle branch block, history of atrial fibrillation or flutter, either complex or frequent supra-ventricular or ventricular arrhythmias and permanent pacemaker stimulation, mitral stenosis of any degree, any other valvular disease of more than mild degree, severe mitral annulus calcification, previous mitral valvuloplasty, valve prosthesis, any other previous surgical or interventional procedures but coronary revascularization. In all patients, height, weight, and blood pressure were measured in the echocardiographic laboratory using standard procedures. Comorbidities and therapies were collected from the medical records of the Cardiovascular Center and the Charlson Comorbidity index [
] was calculated. The echocardiogram closest in time to the visit [median 30 days, interquartile range (IQR) 11–90 days] was used.
Echocardiographic evaluation
At the Trieste Cardiovascular Center, all echocardiographic examinations are performed on the basis of a pre-defined protocol for image acquisition, storage, review, and off-line measurements. All reports include LV dimensions and mass from parasternal views, LV volumes and LVEF by the biplane modified Simpson’s rule, left atrial area measured from the apical four-chamber view, LV inflow indices [peak early (E) and late diastolic (A) velocities, E/A ratio, and E wave deceleration time (DT)] measured by pulsed Doppler at the level of mitral leaflets tips, and peak early diastolic mitral annular velocity (e’) measured by pulsed-wave TDI at the junction of the interventricular septum with the mitral annulus in the apical 4-chamber view [
Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures: a comparative simultaneous Doppler-catheterization study.
]. From the values of E and septal e’, the septal E/e’ ratio is routinely calculated. When recording TDI velocities, special attention is paid to adequately adjust the Doppler spectral gain settings and the velocity scale. An angulation <20° is maintained between the ultrasound beam and the direction of cardiac motion during the sampling of the septal annulus. All measurements are averaged over 3–5 consecutive beats.
Inter- and intra-observer variability of echocardiographic measures was assessed by randomly selecting a sample of 24 patients. For each echocardiographic index, 12 measurements per subject (6 different operators, double evaluation each) were performed. This allowed us to achieve 80% power to detect an intra-class correlation coefficient (ICC) of 0.8 under the null hypothesis of ICC = 0.60, using an F-test at a significance level of 0.05 [
]. The intra-observer variability was evaluated by computing the ICC between the two repeated measures of the same operator. All ICCs were ≥0.80 and ≥0.90 for the inter-observer and intra-observer analysis, respectively (all p < 0.001). When averaged across the six observers, all intra-observer coefficients of variation (calculated for each echocardiographic index between the two repeated measures on the same patients) ranged between 4.1% and 6.3%.
Study endpoint
The endpoint was a composite of all-cause mortality and hospital admission for cardiovascular causes. The end of follow-up was December 31, 2014. For each patient, only the first event was considered in the analysis. Hospitalizations were classified based on the International Classification of Disease 9 (ICD-9 CM) codes of Hospital Discharge Reports, and vital status was extracted from Regional Register of Birth and Death. All this information is automatically transferred in the Regional Health Data Warehouse, which is updated monthly. Cardiovascular hospitalization was defined as hospitalization for unstable angina, non-fatal myocardial infarction, heart failure, interventional cardiology including percutaneous coronary intervention or coronary artery bypass grafting, peripheral vascular intervention, or transient ischemic attack/stroke. Causes of death were ascertained by evaluation of hospital records and autopsy reports. Each clinical event was diagnosed and classified by two expert clinicians who formally generated the information which migrated into the database. For the purpose of this study two expert cardiologists, working in the Cardiovascular Center, retrospectively analyzed in detail the clinical reports to validate the endpoint [
]. To explore the prognostic relevance of TDI parameters, we a priori decided to categorize septal e’ in tertiles, as estimated in our population, and E/e’ both in tertiles and as a dichotomous variable based on the threshold of E/e’ > 15.
Statistical analysis
Data were expressed as mean ± SD, median (IQR), or counts and percentage, as appropriate. Continuous variables were compared using ANOVA followed by the post-hoc Bonferroni or Tamhane’s T2 correction, according to the variance equality assumption. The chi-square or Fisher’s exact tests were used to compare discrete variables. Multivariable Cox regression was used to evaluate predictors of outcome. A nested-model strategy was employed, starting from a first model including relevant clinical parameters, adding standard echocardiographic measures, and finally adding TDI indices. The choice of the covariates was decided according to subject matter knowledge. The model chi-square increase was used to assess the incremental prognostic value of the variables added at each step. An internal bootstrap validation procedure was performed to evaluate the final model performance in terms of the discrimination index Dxy (a measure of correlation between predicted and observed responses that is strictly related to the C-statistic) and the slope shrinkage factor [
]. Kaplan-Meier survival curves were plotted to assess differences in the clinical outcome. All analyses were performed using the SPSS software version 19 [IBM SPSS Statistics 19, IBM corporation, Armonk, New York] and the R statistical package version 3.1.2 [R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.].
Results
Baseline characteristics
During the study period, 11,254 individuals aged >50 years were referred for a comprehensive clinical and echocardiographic evaluation, mostly by their general practitioners. Referral was due to a variety of reasons: assessment of LV remodeling, size, and function in hypertensive, diabetic, and/or patients with known coronary artery disease CAD; presence or concerns for heart failure, rhythm disorders, CAD, or valvular heart disease; shortness of breath; screening and miscellaneous reasons. A total of 2628 subjects met the study criteria and are included in Table 1). The study population was characterized by a slight male preponderance and a significant difference in age between genders [median age (IQR): male 69 (62–75) years vs female 72 (64–78) years, p < 0.001]. Median (IQR) values of septal TDI e’ at each tertile were 5 (4.2–6), 7 (7–8), 10 (9–11) cm/s respectively, while median values of E/e’ at each tertile were 6.8 (5.8–7.6), 10 (8.8–10), and 14 (12–18). Severe diastolic dysfunction, defined as septal E/e’ > 15, was present in only 313 patients (11.9%) (Supplemental Table 1, Table 2).
Table 1Baseline characteristics stratified by events.
No events (n = 2236)
Events (n = 392)
p-value
Clinical
Male gender, No. (%)
1153 (52)
234 (60)
0.003
Age, median (IQR), y
70 (62–76)
74 (68–80)
<0.001
Body mass index, median (IQR), kg/m2
26.2 (23.7–29.2)
26.7 (24.5–29.4)
.054
Body surface area, median (IQR), m2
1.9 (1.7–2)
1.9 (1.8–2)
.079
Systolic blood pressure, median (IQR), mmHg
140 (125–150)
140 (130–150)
.10
Diastolic blood pressure, median (IQR), mmHg
80 (75–90)
80 (72–85)
.40
Heart rate, median (IQR), beats/min
70 (61–78)
70 (60–78)
.65
History of systemic hypertension, No. (%)
1567(70)
309 (79)
<.001
Diabetes mellitus, No. (%)
576 (26)
140 (36)
<.001
History of dyslipidemia, No. (%)
1438 (64)
278 (71)
0.01
Current smoking, No. (%)
323 (14)
62 (16)
.47
eGFR (MDRD < 60 ml/min/1.73), No. (%)
236 (11)
70 (18)
<.001
End-stage kidney disease, No. (%)
20 (0.9)
7 (1.8)
0.1
History of coronary artery disease, No. (%)
671 (30)
207 (52.8)
<0.001
History of vasculopathy, No. (%)
216 (10)
89 (23)
<0.001
History of stroke/TIA, No. (%)
311 (14)
106 (27)
.20
Charlson Comorbidity Index (CCI)
<0.001
CCI < 2, No. (%)
1531 (68.4)
184 (46.9)
>2 CCI < 5, No. (%)
551 (24.7)
146 (37.3)
CCI > 5, No. (%)
154 (6.9)
62 (15.8)
Therapy
ACEI or ARB, No. (%)
1282 (57)
248 (63)
0.02
Beta-blockers, No. (%)
863 (39)
179 (46)
0.08
Calcium channel blockers, No. (%)
410 (18)
111 (28)
<.001
Methyldopa-clonidine-doxazosin, No. (%)
106 (5)
32 (8)
0.005
Nitrates, No. (%)
412 (18)
128 (33)
<.001
Other vasodilators, No. (%)
474 (21)
92 (23)
0.3
Loop diuretics, No. (%)
162 (7)
54 (14)
<.001
Aldosterone receptor antagonists, No. (%)
14 (0.6)
10 (2.6)
<.001
Other diuretics, No. (%)
380 (21.7)
186 (21.2)
.75
Amiodarone/Antiarrhythmic Agents, No. (%)
70 (3.2)
26 (6.7)
<0.001
Ivabradine, No. (%)
23 (1)
11 (3)
0.004
Echocardiography
Left atrial area, median (IQR), cm2
19 (16–22)
20 (17–24)
.001
LV end-diastolic septal wall thickness, median (IQR), mm
1.01 (0. 82-1.21)
1.12 (0.91–1.34)
<0.001
LV end-diastolic posterior wall thickness, median (IQR), mm
1.1 (1.0–1.2)
1.1 (0.9–1.3)
<0.001
LV end-diastolic diameter, median (IQR), cm
4.9 (4.4–5.3)
5.0 (4.5–5.4)
.007
LV end-systolic diameter, median (IQR), cm
2.9 (2.5–3.3)
3.0 (2.6–3.3)
.15
LV mass index, median (IQR), g/m2.7
46.4 (37.2–56.6)
52.6 (40.6–63.3)
<.001
LV end-diastolic volume index, median (IQR), ml/m2
41.0 (32.7–49.5)
41.9 (34.1–51.1)
0.15
LV ejection fraction, median (IQR), %
65 (60–69)
63 (59–68)
0.018
Transmitral E peak, median (IQR), m/s
0.7 (0.58-0.8)
0.65 (0.52-0.8)
0.67
Mitral deceleration time, median (IQR), ms
221 (189–260)
220 (189–261)
.75
Transmitral A peak, median (IQR), m/s
0.8 (0.66-0.9)
0.8 (0.7-0.94)
.003
Mitral E/A ratio, median (IQR)
0.85 (0.71–1.0)
0.81 (0.66–1.0)
.004
Septal TDI e′ peak, median (IQR), cm/s
7 (6–9)
6 (5–8)
<0.001
TDI E/e′ ratio, median (IQR)
9.4 (7.5–12)
10 (8.3–13.3)
<0.001
eGFR, estimated glomerular filtration rate; ACEI, angiotensin-converting enzyme inhibitors; ARB, angiotensin receptor blockers; IQR, interquartile range; LV, left ventricle; MDRD, modification of the diet in renal disease; TIA, transient ischemic attack; TDI, tissue Doppler imaging.
During an average follow-up of 26 months (IQR: 12–41), 392 (15%) patients experienced the endpoint (88 deaths and 304 cardiovascular hospitalizations). These patients showed significant differences for multiple clinical and echocardiographic characteristics compared to those who did not undergo the endpoint (Table 1). The cause of death was cardiovascular in 24 subjects (27%) and undefined in 8. Hospitalizations were related to acute coronary syndrome or coronary revascularization (either surgical or percutaneous) in 211 (69%) cases, stroke/transient ischemic attack in 67 (22%), and heart failure in 26 (8%). The 1-, 2-, and 3-year event-free survival rates were 90%, 86%, and 83%, respectively. Among clinical variables age, male gender, history of CAD, cerebrovascular accident, and Charlson Comorbidity index were independent predictors of outcome in multivariable analysis (Table 2, upper panel). These variables maintained their independent prognostic value after inclusion of standard echocardiographic indices in the model, among which LV mass index was the only independent predictor of outcome (Table 2, second panel).
Table 2Multivariable Cox models considering clinical characteristics and adding step-by-step baseline standard echocardiographic measures and diastolic variables E/e’ and e’ tertiles (for sake of simplicity only significant covariates are reported in tables, except for E/e’ tertiles).
IQR = interquartile range [58 g/m2.7 vs 38 g/m2.7].
1.21
1.04-1.41
.02
e’ tertiles
0.73
0.3-0.94
.04
List of clinical covariates included: Age, male gender, body mass index, systolic blood pressure, diastolic blood pressure, heart rate, history of systemic hypertension, diabetes mellitus, history of dyslipidemia, current smoking, estimated glomerular filtration rate (modification of the diet in renal disease <60 ml/min/1.73), history of coronary artery disease, history of vasculopathy, history of stroke/TIA, Charlson index.
List of standard echocardiographic measures included: Left ventricular ejection fraction, left ventricular end diastolic volume index, left ventricular mass index (g/m2.7), left atrial area.
LV, left ventricular; TIA, transient ischemic attack.
a IQR = interquartile range [76 vs 62 years old].
b Charlson Index classes are ≤2, 3–5 and >5.
c IQR = interquartile range [58 g/m2.7 vs 38 g/m2.7].
The event-rate was significantly higher in patients with E/e’>15 than in those with E/e’≤15 (log rank p = 0.004, Fig. 1a). After dichotomizing the population based on its median age (71 years) this relationship maintained its significance for patients aged <71 years (p = 0.002), but not in those aged >71 years (p = 0.289), and was stronger in males (p = 0.008) than in females (p = 0.06).
Fig. 1Kaplan-Meier Curves. (a) E/e’ ratio ≤ or > 15; (b) E/e’ ratio in tertiles; (c) e’ in tertiles, all showing survival free from overall mortality and cardiovascular (CV) hospitalization (each panel displays numbers of subjects at risk for each category).
Increasing E/e’ tertiles showed an overall association with the endpoint (log-rank p < 0.02), although the contrast between the second and third tertiles was not significant (p = 0.18, Fig. 1b). Again, this relationship maintained its significance in younger (p = 0.008) but not in older patients (p = 0.171), in males (p = 0.001) but not in females (p = 0.098).
Overall, decreasing e’ tertiles were significantly associated with the end-point (log-rank p < 0.01 for all contrasts, Fig. 1c) and maintained its significance in patients aged <71 years (p = 0.005) and in those aged >71 years (p = 0.004), in males (p < 0.001), and in females (p = 0.001).
When TDI diastolic indices were added to the models including clinical variables and LV mass index, adjusted for age and sex, only e’ was independently linked to the composite endpoint, whereas the association of E/e’ (either in tertiles or dichotomized according to the threshold of 15) with the clinical outcome did not achieve significance (Table 2). Similar findings were obtained if only the 24 cardiovascular deaths were used as an end-point together with cardiovascular hospitalization (data not shown). Adding e’ also significantly improved the overall model chi-square, with an incremental prognostic value over clinical, demographic, and standard echocardiographic variables (Fig. 2). The internal validation procedure showed an unadjusted C index of 0.72 and a bootstrap-corrected C index of 0.70, suggesting good discrimination.
Fig. 2Barplot of the Chi-square values corresponding to the nested multivariable models.
In this study, we aimed at investigating the prognostic value of TDI diastolic indices for the prediction of death or cardiovascular hospitalization in a large population of stable outpatients in sinus rhythm, with normal LVEF, and a substantial prevalence of cardiovascular diseases, risk factors, and LV remodeling but without prevalent HF, significant valvular disease, or previous surgical or interventional procedures but remote coronary revascularization. These patients were referred for a comprehensive cardiovascular evaluation by their general practitioners thus reflecting a primary care setting. The clinical question was to assess whether the available evidence reported in multiple, different experimental settings [
Clinical relevance of the spectral tissue Doppler E/e’ ratio in the management of patients with atrial fibrillation: a comprehensive review of the literature.
Correlation with invasive left ventricular filling pressures and prognostic relevance of the echocardiographic diastolic parameters used in the 2016 ESC heart failure guidelines and in the 2016 ASE/EACVI recommendations: a systematic review in patients with heart failure with preserved ejection fraction.
] may be duplicated in the heterogeneous clinical scenario encountered in every day clinical practice. During a midterm follow-up, e’ and E/e’ were both associated with clinical outcome in these patients, but only e’ maintained its independent prognostic information in multivariable analysis. In fact, prognostic value of e’ was incremental to established clinical and echocardiographic variables.
Previous studies have shown that TDI indices of diastolic function provide information on outcome which is incremental to conventional 2D-DE in general population and multiple conditions [
Clinical relevance of the spectral tissue Doppler E/e’ ratio in the management of patients with atrial fibrillation: a comprehensive review of the literature.
], in 793 individuals from the general population (mean age 51 years), showed that e' velocity was a significant predictor of fatal and nonfatal cardiovascular and cardiac events. Noteworthy, in that study, E/e’ ratio was borderline associated with an increased risk of cardiac events. The role of e’ is plausible from a pathophysiologic point of view, since e’ reflects LV relaxation, a highly energy consuming process which is susceptible to ischemia, is relatively preload independent when LV relaxation is abnormal, is inversely related to arterial load, reflects LV restoring forces, decreases early in the course of diastolic impairment, and is abnormal at all stages of diastolic dysfunction [
Determinants of left ventricular early-diastolic lengthening velocity: independent contributions from left ventricular relaxation, restoring forces, and lengthening load.
] but are of practical interest since they are obtained in a real-world cohort, composed mostly of elderly individuals with a substantial prevalence of cardiovascular diseases, risk factors, and LV remodeling but without prevalent HF. It is well known that age, together with sex and physical fitness, affect e’ [
Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
Age- and sex-related influences on left ventricular mechanics in elderly individuals free of prevalent heart failure. The ARIC Study (Atherosclerosis Risk in Communities).
]. However, current guidelines do not provide age-adjusted limit for TDI values. Importantly, the use of age-based normative values was recently proven to improve the risk discrimination of diastolic measures for incident HF or death in 2 independent studies [
]. Nonetheless, the cut-off values derived from these two studies to define abnormalities were different, raising the need for further studies to resolve these discrepancies [
]. Moreover, other considerations suggest that additional work is warranted to determine whether age‐ and sex‐specific reference limits should be used to assess diastolic dysfunction in clinical practice [
]. Interestingly, in our cohort, age (dichotomized at median age) and sex influenced the relationship between E/e’ but not e’ and cumulative events. However, our paper was neither aimed nor powered to detect these age- and sex-specific values. Thus, future, purposely designed studies, based on larger populations, covering wider age categories for both sexes, are needed for such an ambitious aim. Importantly, however, our multivariable analysis is adjusted for sex and age distribution detected in our population. Noteworthy, the median lower and upper tertiles, respectively, of septal e’ and E/e’ in the present study resemble those reported by others [
Age- and sex-related influences on left ventricular mechanics in elderly individuals free of prevalent heart failure. The ARIC Study (Atherosclerosis Risk in Communities).
], algorithms proposed for the clinical evaluation of diastolic dysfunction. E/e’ ratio is also utilized as a tool for risk stratification in patients with the preclinical stage of HF as well as in multiple clinical scenario [
Echocardiographic estimation of left ventricular and pulmonary pressures in patients with heart failure and preserved ejection fraction: a study utilizing simultaneous echocardiography and invasive measurements.
Role of diastolic stress testing in the evaluation for heart failure with preserved ejection fraction: a simultaneous invasive-echocardiographic study.
]. Interestingly, in individuals at high risk for HF due to the presence of hypertensive heart disease or diabetes mellitus, conflicting findings regarding the independent role of E/e’ ratio as prognosticator have been reported [
Left ventricular diastolic function, assessed by echocardiography and tissue Doppler imaging, is a strong predictor of cardiovascular events, superior to global left ventricular longitudinal strain, in patients with type 2 diabetes.
]. Noteworthy, it has been shown that, indices of diastolic dysfunctions (including E, e’, E/e’ ratio, and LA volume) display substantial overlap between normal individuals and patients with HF and preserved EF [
Role of diastolic stress testing in the evaluation for heart failure with preserved ejection fraction: a simultaneous invasive-echocardiographic study.
] despite significant differences. Indeed, a comprehensive meta-analysis of the diagnostic accuracy of E/e’ demonstrates that the evidence for using E/e’ for estimating LV filling pressure in routine clinical practice is limited in patients with normal LVEF [
Diagnostic accuracy of tissue doppler index E/e’ for evaluating left ventricular filling pressure and diastolic dysfunction/heart failure with preserved ejection fraction: a systematic review and meta-analysis.
]. Thus, our findings, showing that septal E/e’ is not an independent marker of prognosis in the midterm in patients with normal LVEF and without HF can be interpreted in the light of these studies. Moreover, our findings support the notion that E/e’ should not be used as a stand-alone tool for the non-invasive estimation of LVFP when LVEF is normal and HF absent, but rather integrated in a multiparametric approach [
Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
This study has other limitations beyond its retrospective design. The risk profile of our study group is heterogeneous, with different contribution of cardiovascular diseases, of comorbid conditions, risk factors, and LV remodeling but without prevalent HF. Furthermore, study exclusion criteria (i.e. patients with history of atrial fibrillation, or significant valvular heart disease, or previous, well treated, and stabilized HF), favored the reported low prevalence of severe diastolic dysfunction. Moreover, the investigated subjects were only about one-fourth of screened individuals. Thus, the clinical impact of our findings is likely referable to the specific, relatively narrow, profile of our study group, and might not be generalizable to other populations with different demographic and clinical characteristics. Finally, in our cohort HF was carefully ruled out by cardiologists specifically trained in HF, thus resulting in the relatively low rate of HF events in the follow-up. Moreover, the follow-up duration might be underpowered to fully explore the role of septal E/e’ ratio as a prognosticator in such a clinical scenario. Nonetheless, the significant additive role played by e’ despite the midterm observation, add evidence to the utility of TDI indices of diastolic function in clinical practice. Consistently with multiple previous independent studies [
], at the time in which the echocardiographic examinations were performed, we used only septal e’ and E/e’. Thus we cannot make any assumption on the prognostic meaning of lateral and/or average e’ and E/e’ in individuals with normal LVEF without HF. However, while future studies are needed to re-evaluate our findings during longer follow-up periods, adopting an updated approach to assessment of TDI diastolic indices in the context of a comprehensive, stepwise, multi-parametric diagnostic work-out of diastolic dysfunction, we do not feel that the above-mentioned limitations significantly flaw our data or their interpretation.
Conclusions
In a large cohort of primary care patients with normal LVEF, free from HF, but with significant prevalence of cardiovascular diseases, risk factors, and comorbid conditions, low septal e’ is an independent and incremental predictor of overall mortality and cardiovascular hospitalizations while septal E/e’ is not. These findings might have implications for clinical practice, reinforcing previous results obtained in relatively smaller series of younger individuals from the general population [
], since e’ represents a feasible echocardiographic measure which could be easily implemented for refining cardiovascular risk for practical purposes. Moreover, our data spread a cautionary word about the utilization of E/e’ alone for prognostic purposes and as a shortcut of comprehensive assessment of diastolic dysfunction in patients without HF.
Conflict of interest
Declarations of interest: none
Appendix A. Supplementary data
The following are Supplementary data to this article:
Clinical relevance of the spectral tissue Doppler E/e’ ratio in the management of patients with atrial fibrillation: a comprehensive review of the literature.
Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
New echocardiographic predictors of clinical outcome in patients presenting with heart failure and a preserved left ventricular ejection fraction: a subanalysis of the Ka (Karolinska) Ren (Rennes) Study.
Is echocardiographic evaluation of diastolic function useful in determining clinical care? Doppler echocardiography yields dubious estimates of left ventricular diastolic pressures.
Diagnostic accuracy of tissue doppler index E/e’ for evaluating left ventricular filling pressure and diastolic dysfunction/heart failure with preserved ejection fraction: a systematic review and meta-analysis.
Risk stratification in chronic heart failure: independent and incremental prognostic value of echocardiography and brain natriuretic peptide and its N-terminal fragment.
Optimal noninvasive assessment of left ventricular filling pressures: a comparison of tissue Doppler echocardiography and B-type natriuretic peptide in patients with pulmonary artery catheters.
Echocardiographic indices do not reliably track changes in left-sided filling pressure in healthy subjects or patients with heart failure with preserved ejection fraction.
Correlation with invasive left ventricular filling pressures and prognostic relevance of the echocardiographic diastolic parameters used in the 2016 ESC heart failure guidelines and in the 2016 ASE/EACVI recommendations: a systematic review in patients with heart failure with preserved ejection fraction.
2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC)developed with the special contribution of the Heart Failure Association (HFA) of the ESC.
Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures: a comparative simultaneous Doppler-catheterization study.
Determinants of left ventricular early-diastolic lengthening velocity: independent contributions from left ventricular relaxation, restoring forces, and lengthening load.
Age- and sex-related influences on left ventricular mechanics in elderly individuals free of prevalent heart failure. The ARIC Study (Atherosclerosis Risk in Communities).
Echocardiographic estimation of left ventricular and pulmonary pressures in patients with heart failure and preserved ejection fraction: a study utilizing simultaneous echocardiography and invasive measurements.
Role of diastolic stress testing in the evaluation for heart failure with preserved ejection fraction: a simultaneous invasive-echocardiographic study.
Left ventricular diastolic function, assessed by echocardiography and tissue Doppler imaging, is a strong predictor of cardiovascular events, superior to global left ventricular longitudinal strain, in patients with type 2 diabetes.
30034-4&id=gr1.jpg){kind=link}
30034-4&id=gr2.jpg){kind=link}