Introduction. Aims. Keywords

Transcription

1 European Heart Journal Cardiovascular Imaging (2015) 16, doi: /ehjci/jev126 Added value of pulmonary venous flow Doppler assessment in patients with preserved ejection fraction and its contribution to the diastolic grading paradigm Eric Buffle 1, Judith Kramarz 1, Esther Elazar 1, Galit Aviram 2, Meirav Ingbir 1, Nahum Nesher 3, Simon Biner 1, Gad Keren 1, and Yan Topilsky 1 * 1 The Division of Cardiovascular Diseases and Internal Medicine, Tel Aviv Medical Center, 6 Weizmann Street, Tel Aviv, Israel; 2 The Division of Radiology, Tel Aviv Medical Center, Tel Aviv, Israel; and 3 Division of Cardiovascular Surgery, Tel Aviv Medical Center, Tel Aviv, Israel Received 26 November 2014; accepted after revision 19 April 2015; online publish-ahead-of-print 1 June 2015 Aims To evaluate the prognostic role of pulmonary venous flow parameters and their role in patients with preserved ejection fraction (EF).... Methods Pulmonary venous flow parameters were measured in 365 patients in sinus rhythm, without significant mitral disease, and and results EF.50% (age ; 52% female) by a single sonographer. Survival, time to re-admission for heart failure, and to a combined cardiac end point (cardiac death, heart failure, and atrial fibrillation) were retrospectively analysed and correlated to echo parameters. Systolic (S) and diastolic (D) pulmonary vein flow were obtainable in 73% of patients and Ar in 65%. The lower peak S/D ratio and higher DAr-A time were associated with higher rate of heart failure readmission (P ¼ 0.03 for both). The S/D integral ratio was the best pulmonary vein flow predictor of heart failure readmissions (P ¼ ), better than the peak S/D ratio, or DA-Ar time (P, 0.01 for both), and independently predicted worse outcome even when adjusted for diastolic grading (using recent guidelines), left ventricle mass index, E/e, and left atrial volume index (P, 0.05 for all). The addition S/D ratio to diastolic grading recognized patients with pseudonormal filling pattern and S/D ratio.1 with similar clinical outcomes to grade I (P. 0.5), but worse clinical outcomes than in the pseudo-normal patients with lower S/D ratio (P, 0.01).... Conclusions PVFP are obtainable in most patients, add prognostic information on top of routine diastolic parameters, and define an early stage of diastolic dysfunction resembling the pseudo-normal pattern in which S/D ratio is.1, and outcome is excellent Keywords Introduction pulmonary venous flow parameters ejection fraction difference from mitral A-wave duration systolic to diastolic ratio left atrium left ventricle Pulmonary venous flow measurements include the peak systolic (S) velocity, peak diastolic (D) velocity, the duration of the atrial regurgitant (Ar) velocity, the S/D ratio (peak and integral) and the time difference between the Ar velocity and mitral A-wave duration. The recent Joint of Diastology Working Group has granted only a minor and supporting role for pulmonary venous flow parameters in diastolic grading in patients with preserved ejection fraction (EF). 1 The secondary role ascribed to the pulmonary venous flow measurements was attributed to several factors including difficulty in obtaining high-quality recordings suitable for measurements, and the small number of studies showing the prognostic role of pulmonary venous flow, 2 4 all in patients with low EF. In this study, we aimed to analyse the additive prognostic value of assessment of pulmonary venous flow measurements performed as * Corresponding author. Tel: ; fax: ; topilskyyan@gmail.com Published on behalf of the European Society of Cardiology. All rights reserved. & The Author For permissions please journals.permissions@oup.com.

2 1192 E. Buffle et al. part of routine echo practice in a large group of patients with preserved EF and normal sinus rhythm. Furthermore, we analysed whether the inclusion of those measurements may influence the diastolic grading process. Material and methods Study design The study population included only in-hospital patients admitted for numerous causes including cardiological and systemic diagnoses to a tertiary hospital (Tel Aviv Medical Center) between January 2012 and January The frequency of treated hypertension was 41%, of known heart failure 20%, chronic kidney disease 21%, chronic lung disease 6%, and coronary artery disease 25%. The study cohort included only patients with preserved EF (.50%), normal sinus rhythm, without significant mitral disease (mitral regurgitation moderate, any degree of mitral stenosis), pericardial or endocardial disease, severe co-morbid or life-limiting conditions (metastatic cancer, end-stage lung disease, cirrhosis, or end-stage renal disease) at presentation in which routine diastolic assessment including pulmonary venous flow estimation was performed by a single sonographer (E.E.) during routine echo practice (not part of research protocols). Thus, all patients had sinus rhythm, preserved EF, and no cardiac or systemic co-morbidities that may interfere with diastolic assessment or outcome analysis. Outcome was analysed from the date of echo evaluation until death or last follow-up. Baseline clinical assessment and management Patients symptoms and co-morbid conditions summarized as the Charlson age-adjusted co-morbidity index 5 were evaluated by the patients Tel Aviv Medical Centre personal physician. Congestive heart failure (CHF) at baseline, or heart failure hospitalizations were defined based on the following: presence of at least two major criteria or one major criterion with two minor criteria. Major criteria included paroxysmal nocturnal dyspnoea, neck vein distention, rales, cardiomegaly, pulmonary oedema, S3 gallop, hepatojugular reflux, or laboratory evidence of visceral congestion. Minor criteria included ankle oedema, dyspnea on exertion, hepatomegaly, pleural effusion, or tachycardia (,120/min). 6,7 Clinical management was determined independently by the patients and their personal physicians using all information available. Follow-up and outcomes Clinical follow-up was obtained by review of medical records, and surveys. Cause of death was determined by review of all medical records and death certificates. Events used as end points were admission for heart failure exacerbation and cardiovascular events. Cardiovascular events comprised cardiac death including sudden death, CHF, new onset of atrial fibrillation but not death due to other causes. Doppler-echocardiography Ejection fraction was calculated by the Quinones method, and LV diameters, inter-ventricular septal and posterior wall width, LV mass, relative wall thickness, were calculated as recommended. 8 Forward stroke volume was calculated from left ventricular outflow tract with subsequent calculation of cardiac output and index. Valvular function was assessed by standard qualitative assessment using jet size according to the guidelines (normal/trivial ¼ 1, mild ¼ 2, moderate ¼ 3, severe ¼ 4). 9 Early (E) and atrial (A) trans-mitral flow velocities, and early diastolic mitral annular velocity (e ) were measured in the apical 4-chamber view. 1,10 A wave duration was measured at the level of the mitral annulus as previously recommended. 11 The e was measured from septal and lateral annulus. The ratio of peak E to peak e (septal, lateral, and average) was calculated (mitral E/e ratio) from the average of at least three cardiac cycles. Left atrial volume was calculated using the biplane area length method at end systole. 8 Diastolic function was assessed by integrating measurements of the mitral inflow, LA volume and Doppler tissue imaging of the mitral annulus, based on recent guidelines 1,10 and classified into four categories: normal diastolic function ¼ 0, impaired relaxation ¼ 1, pseudo-normal ¼ 2, and restrictive pattern ¼ 3. Pulmonary venous flow assessment PW Doppler of pulmonary venous flow was performed as part of routine practice using the apical 4-chamber view as recommended by recent guidelines. Colour flow imaging was used for the proper location of the sample volume in the right upper pulmonary vein usually by angulating the transducer superiorly such that the aortic valve was seen. A 2 3-mm sample volume was placed 0.5 cm into the pulmonary vein for optimal recording of the spectral waveforms. Wall filter settings were set low enough to display the onset and cessation of the atrial reversal (Ar) velocity waveform. The spectral recordings were obtained at a sweep speed of mm/s at end-expiration and included the average of three consecutive cardiac cycles. Measurements of pulmonary venous waveforms included peak and velocity time integral (VTI), systolic (S) velocity, peak and VTI antero-grade early diastolic (D) velocity, the calculated S/D peak, and S/D VTI ratio, the duration of the Ar velocity, and the time difference between it and mitral A-wave duration. Statistics Descriptive results were expressed as mean + SD for continuous variables and as percentages for categorical variables. Group comparisons used ANOVA, the Fisher s exact test, or x 2 test, as appropriate. Multiple comparisons for continuous and categorical parameters used the Tukey Kramer honestly significant difference test and the Bonferroni correction, respectively. Clinical end points were time to heart failure readmissions and time to combined cardiac event (cardiac death, or heart failure readmission). Cox proportional hazards models for the end points allowed calculation of hazard ratios (HR) attached to pulmonary vein flow parameters and were used unadjusted and adjusted for diastolic grade, age, left atrial volume index and E/e (average and septal). Incremental value of pulmonary vein flow parameters vs. diastolic grading and comparison of the different pulmonary venous parameters for assessment of time to clinical end points were tested by comparing nested models with F-tests. P values,0.05 were considered significant. All authors participated in designing the study, collecting and analysing data, and drafting and revising the manuscript. Results Baseline characteristics All echo exams were performed as part of the routine echo practice and the median duration of exams was 25 [19, 34] min. Out of 541 patients with preserved EF, 50 patients were excluded due to atrial fibrillation, 57 excluded due to significant mitral diseases (interfering with diastolic assessment), 26 due to metastatic cancer, and 15 due to end-stage lung disease, cirrhosis, or end-stage renal disease at presentation. Out of the remaining 393 patients, 284 (73%) had high-quality systolic and diastolic pulmonary venous recordings suitable for measurements and were included in our analysis. In 255/393 (65%) of the patients duration of the Ar velocity and the difference between it and

3 Added value of pulmonary venous flow Doppler assessment 1193 Table 1 Baseline characteristics of patients overall and stratified by severity of diastolic dysfunction All Normal (N 5 36) Grade I (N 5 87) Grade II (N 5 141) Grade III (N 5 20) P... Age, years *, * *,, Male sex, n (%) 48% 60% 47% 47% 57% 0.2 NYHA grade (%) I 72; II 8 III 13; IV 7 I 90; II 6 III 4; IV 0 I 81; II 6 III 10; IV 3 I 76; II 8 III 11; IV 5 I 24; II 19 III 38; IV 19, Coronary heart disease (%) 25% 8% 25% 29% 33% CABG (%) 17% 7% 18% 19% 19% 0.1 COPD (%) 6% 2% 6% 6% 1% 0.2 NIDDM (%) 25% 8% 27% 24% 28% 0.02 Height Weight * * * BSA BMI * * * SBP * * * DPB Haemoglobin, g/l Creatinine, mg/dl Bilirubin, mg/dl Co-morbidity index *, *, *,,, LVEDD * LVESD LV mass , LV mass index * * *, LV mass for height * * * 0.01 LV EF (%) E wave , *,, *, *,, A wave , *,, *, , E/A , *,, *,, *,,, E septal * *, *, E lateral * *, *, E average * *, *, E/e ,, *,, *,, *,,, Cardiac output LA volume index ,, *,, *,, *,,, Systolic PV peak *,, Systolic PV VTI *,,, Diastolic PV peak * , *,,, Diastolic PV TVI , *,,, PV Ar peak * Mitral A time *,,, PV Ar time * *,, PV S/D peak *,, , *,,, PV S/D TVI *,, , *,,, PV S TVI/D peak *,, , *,,, DAr-A time *,,, PV S/D peak,1 33% 25% 12% 23% 100%, NYHA, New York Heart Association; CABG, Coronary Artery Bypass Graft; COPD, Chronic Obstructive Pulmonary Disease; NIDDM, Non Insulin Dependent Diabetes Mellitus; BMI, Body Mass Index; SBP, Systolic Blood Pressure; DPB, Diastolic Blood pressure; LVEDD, Left Ventricular End Diastolic Diameter; LVESD, Left Ventricular End Systolic Diameter; EF, Ejection Fraction; LA, Left Atrium; PV, Pulmonary Vein. *P, 0.05 vs. normal. P, 0.05 vs. grade I. P, 0.05 vs. grade II. P, 0.05 vs. grade III.

4 1194 E. Buffle et al. the mitral inflow A wave duration could be assessed as well. Table 1 shows the baseline characteristics of the 284 patients in the series, overall, and stratified by diastolic grade. The comparison shows that sex, body surface area (BSA), haemoglobin, creatinine, EF, and cardiac output were equally distributed between the groups. Systolic pulmonary vein peak velocity and integral decreased with worsening diastolic dysfunction, but diastolic flow (peak and integral) followed a bimodal trend, decreasing in mild diastolic dysfunction, normalizing in grade II diastolic dysfunction, and becoming abnormally elevated in grade III dysfunction. The combined effect of the trends in systolic and diastolic pulmonary venous flows resulted in an increase in S/D ratio (peak/peak; integral/integral) in mild diastolic dysfunction, normalization with grade II diastolic dysfunction, and markedly low ratios with severe diastolic dysfunction. The time difference between the mitral A-wave and Ar duration (DAr-A time) did not change in grades 0, I, and II but increased significantly with grade III diastolic dysfunction. Outcome after diagnosis There were 30 readmissions for heart failure during follow-up ( months) after diagnosis. Thus, freedom from heart failure rates overall was and % at 6 and 12 months, respectively. Lower peak systolic/peak diastolic pulmonary venous flow ratio and higher DAr-A time (ms) were both associated with rate of heart failure readmission (0.37 [ ], P ¼ 0.03; 1.02 [ ], P ¼ 0.03, respectively). The ratio of pulmonary systolic to diastolic flow integral was the best pulmonary vein flow predictor of heart failure readmission (0.3 [ ], for each decrease in ratio, P ¼ ), better than peak systolic/peak diastolic pulmonary venous flow ratio, or DAr-A time (P, 0.01 for both). Table 2 Impact of pulmonary systolic to diastolic flow integral, and DAr-A time on mortality and heart failure readmissions adjusted for diastolic grade, LV mass and heart rate HR 95% interval P value for PV P value for x 2 for model parameter entire model... Heart failure readmission Diastolic grade alone With S/D integral With D Ar-A time With S/D integral, HR, LV mass index With S/D integral, HR, LV mass index height Survival Diastolic grade alone With S/D integral With D Ar-A time With S/D integral, heart rate, LV mass index With S/D integral, HR, LV mass index height Figure 1 (A H ) Traces of left atrial (LA) area, mitral inflow, tissue Doppler septal e, and pulmonary vein flow in a patient with Grade II with low S/D ratio (A D), and Grade II with high S/D ratio (E H). The patient with low S/D ratio was 65 years old, had LA volume index of 38 cc/m 2 and LV mass index of 81 g/m 2. She was re-admitted for heart failure 84 days after the echo. The patient with high S/D ratio was 60 years old, had hypertension with mild LV hypertrophy, had LA volume index of 39 cc/m 2, LV mass index 112 g/m 2, and is free of events 3 years post-baseline echo.

5 Added value of pulmonary venous flow Doppler assessment 1195 There were 48 cardiac events during follow-up. Thus, the freedom from the combined event overall was and % at 6 and 12 months, respectively. Higher DAr-A time was significantly associated with higher rate of combined event (1.02 [ ], P ¼ 0.01), but not peak systolic/peak diastolic pulmonary venous flow (P ¼ 0.08). The ratio of pulmonary systolic to diastolic flow integral was the best pulmonary venous predictor of the combined event rate (0.05 [ ], P ¼ 0.008), superior to peak systolic/ peak diastolic pulmonary venous flow, or DA-Ar time (P, 0.01 for both). The addition of ratio of pulmonary systolic to diastolic flow integral to diastolic grade (based on recent guidelines) showed that it has independent incremental predicting value for survival (P ¼ 0.05) and readmission for heart failure but not for the combined event (Table 2). Because low S/D may be a proxy for heart rate and/or LV hypertrophy we added heart rate and for LV mass index (for BSA and for height 1.7 ). The addition of heart rate and LV mass index (for BSA and height 1.7 ) to the models showed that the ratio of pulmonary systolic to diastolic flow integral has independent incremental predicting value for heart failure (P ¼ 0.05; 0.05), the combined event (P ¼ 0.05; 0.05), but not for survival (P ¼ 0.2; 0.1). The addition of DAr-A time to diastolic grade showed that it has independent incremental predicting value for readmission for heart failure but not for survival or the combined event (Table 2). Effect of pulmonary venous flow on diastolic grading A total of 108 patients (77%) out of the 141 patients classified as grade II diastolic dysfunction, based on recent guidelines, had an S/D integral ratio.1. We divided the grade II diastolic dysfunction group into patients with normal S/D peak ratio (.1), and those with lower ratio (N ¼ 33; S/D, 1) (Figure 1). Table 3 shows the baseline characteristics of the patients with grade II diastolic dysfunction with high S/ D ratio (S/D. 1) and grade II diastolic dysfunction and low S/D ratio (S/D, 1). The comparison shows that sex, co-morbidities, blood pressure, BSA, haemoglobin, creatinine, EF, and cardiac output were equally distributed between the groups. Interestingly LV size and mass increased and other diastolic parameters assessing left atrial filling pressure (E/e average and septal, left atrial volume index) were worse in patients with grade II diastolic dysfunction and low S/D ratio, although the LA volume index was 34 cc/m 2 in all patients in grade II irrespective of S/D ratio. Grade II with S/D. 1 showed no difference in readmission for heart failure rate vs. grade I diastolic dysfunction ( vs % at 1 year, P ¼ 0.4). On the other hand grade II with S/D,1 patients had higher readmission for heart failure rate compared with grade I ( vs , P, 0.001) or grade II with S/D. 1 diastolic dysfunction ( vs , P ¼ 0.002) (Figure 2A). Grade II with S/D. 1 showed no difference in the combined cardiac outcome vs. grade I diastolic dysfunction ( vs % at 1 year, P ¼ 0.3). On the other hand, grade II with S/D, 1 patients had higher combined cardiac outcome rate compared with grade I ( vs , P ¼ 0.02) or grade II with S/D. 1( vs , P ¼ 0.001) (Figure 2B). The higher heart failure readmission rate in grade II with S/D, 1 compared with grade II with S/D. 1 patients was independent of age Table 3 Baseline characteristics of patients with pseudo-normal inflow pattern stratified by systolic to diastolic flow ratio Grade II with S/D Grade II with S/ P >1(N 5 108) D <1(N 5 33)... Age, years NS Male sex, n (%) 51% 40% NS NYHA grade (%) I 82; II 6 III 9; IV 2 Coronary heart disease (%) I 66; II 11 III 11; IV 11 NS 32% 33% NS CABG (%) 19% 18% NS COPD (%) 5% 7% NS NIDDM (%) 23% 27% NS Height NS Weight NS BSA NS BMI NS SBP NS DPB NS Haemoglobin, g/l NS Creatinine, mg/dl NS Bilirubin, mg/dl NS Co-morbidity NS index LVEDD NS LVESD NS LV mass LV mass index LV mass index for height 1.7 LV EF (%) NS E wave NS A wave NS E/A E septal NS E lateral NS E average NS E/e Cardiac output NS LA volume index , Systolic PV peak Systolic PV VTI , Diastolic PV peak Diastolic PV TVI , PV Ar peak NS Mitral A time NS PV Ar time NS PV S/D peak PV S/D TVI , PV S TVI/D peak , DAr-A time NYHA, New York Heart Association; CABG, Coronary Artery Bypass Graft; COPD, Chronic Obstructive Pulmonary Disease; NIDDM, Non Insulin Dependent Diabetes Mellitus; BMI, Body Mass Index; SBP, Systolic Blood Pressure; DPB, Diastolic Blood pressure; LVEDD, Left Ventricular End Diastolic Diameter; LVESD, Left Ventricular End Systolic Diameter; EF, Ejection Fraction; LA, Left Atrium; PV, Pulmonary Vein.

6 1196 E. Buffle et al. Figure 2 (A) Readmission for heart failure in patients with preserved EF, stratified according to the diastolic grade including grade of IIa (pseudonormal E/A ratio, e, 8 cm/s, LA volume index 34 cc/m 2, but S/D.1). The values indicated for each line are freedom from admission for heart failure from baseline to end of follow-up. Note that there is no difference in readmission for heart failure rates between patients with normal, grade I and grade IIa diastolic dysfunction but that classic grade II and grade III are associated with markedly increased cardiac event rates. (B) Cardiac events (cardiac death or CHF or new onset of atrial fibrillation) in patients with preserved EF, stratified according to the diastolic grade including grade of IIa (pseudo-normal E/A ratio, e, 8 cm/s, LA volume index 34 cc/m 2, but S/D.1). The values indicated for each line are freedom from cardiac events from baseline to end of follow-up. Note that there is no difference in cardiac event rates between patients with normal, grade I and grade IIa diastolic dysfunction but that classic grade II and grade III are associated with markedly increased cardiac event rates. (P ¼ 0.01), LV mass index (P ¼ 0.04), E/e average (P ¼ 0.03), E/e septal (P ¼ 0.05), LA volume index (P ¼ 0.04) or all of the above (P ¼ 0.05). The higher combined cardiac outcome rate in grade II with S/D, 1 compared with grade II with S/D.1 patients was independent of age (P ¼ 0.01), LV mass index (P ¼ 0.03), E/e average (P ¼ 0.02), LA volume index (P ¼ 0.03), or all of them together (P ¼ 0.05), but not for E/e septal (P ¼ 0.3). On the other hand, S/D ratio remained an independent predictor of survival when adjusted for LV mass index (P ¼ 0.008), but not for LA volume index (P ¼ 0.2), E/e average (P ¼ 0.2), or E/e septal (0.7). Discussion Our study shows that pulmonary vein flow parameters are obtainable in most patients even in a busy clinical echo practice. It is the first to link pulmonary vein flow parameters to outcome in patients with preserved EF. We found that the decreased systolic to diastolic pulmonary vein flow ratio is associated with higher rates of readmission for heart failure, or a combined cardiac event even when adjusted for age, LV mass index, E/e, and LA volume. The association of a grade II (pseudo-normal) mitral inflow pattern with S/D ratio.1 allows us to identify a group with a better prognostic outcome, similar to patients with grade I diastolic dysfunction and better than patients with pseudo-normal mitral inflow pattern with S/D ratio,1. Pulmonary vein flow parameters and outcome Pulmonary venous findings can be divided into two groups: those related to the antegrade S and D waves, and those related to the retrograde atrial reversal wave. We found that although DAr-A time is associated with higher rate of cardiac events, the ratio of pulmonary systolic to diastolic flow integral was the best pulmonary vein flow predictor of outcome. Furthermore, only the S/D flow integral ratio had independent incremental predicting value when adjusted for LV mass index, E/e, LA volume, or conventional diastolic grading. A possible reason for the superiority of the S/D flow integral ratio in our study is that interpreting DAr-A time is difficult and time consuming because of the need to obtain high-quality recordings suitable for measurements. This is especially true for Ar velocity, for which atrial contraction can create low-velocity wall motion artefacts that obscure the pulmonary flow velocity signal. 1,10 Furthermore, sinus tachycardia and first-degree AV block often result in the start of atrial contraction occurring before diastolic mitral and pulmonary venous flow velocity has declined to the zero baseline, increasing the width of the mitral A-wave velocity and decreasing that of the reversal in the pulmonary vein, making the Ar A relationship difficult to interpret. 1,10 All these combined result in non-satisfactory Doppler recordings for the assessment of the Ar flow in a significant portion of patients when the exams are performed as part of a routine clinical busy echo practice. On the other hand, assessment of systolic and diastolic pulmonary vein flow is easy and was achievable in the majority of patients, increasing its statistical power. Pulmonary vein flow and diastolic grading The recent Joint of Diastology has recommended that an evaluation of diastolic function in patients with preserved EF begins with e,la volume index, and mitral inflow velocities, and pulmonary venous flow parameters have received only a supporting role. 1,10 The secondary role ascribed to the pulmonary venous flow measurements was attributed to several factors including the difficulty in obtaining high-quality recordings suitable for measurements, 10,12 and the

7 Added value of pulmonary venous flow Doppler assessment 1197 small number of studies showing the prognostic role of pulmonary venous flow, 2 4 all performed in patients with systolic dysfunction. In this study, we describe that the S/D ratio is obtainable in most patients with preserved EF, even when its evaluation is performed as part of a busy clinical practice. The pulmonary venous S-wave is generated not only by atrial relaxation which creates a pulmonary venous-to-left atrial pressure gradient, but also by forward propagation of the right ventricle (RV) pressure pulse and systolic descent of the AV plane. 13,14 D velocity follows the E velocity, thus is positively correlated with LA end systolic pressure (V wave) and LV active relaxation. 15 Thus, although in healthy patients a low S/D ratio may represent super-normal LV relaxation, in our cohort any decrease in S/D ratio was probably related to decrease in LA relaxation, decrease in LA compliance, elevated LA pressure, decrease in LV compliance, or decrease of forward propagation of the RV pressure pulse and systolic descent of the AV plane. Based on our data we believe that Doppler pulmonary venous flow assessment has added value in the grading of diastolic dysfunction. In the majorityof patients with pseudo-normal mitral inflow,although e is already reduced, and LA volume is increased, S pulmonary venous velocity is still higher than D velocity, and outcome is excellent suggesting that these patients represent an intermediate stage that occurs at an early asymptomatic stage of diastolic dysfunction, with preserved LA relaxation and reservoir function. Recent physiological studies may propose a possible explanation to these novel findings. It has been shown that e becomes reduced years earlier than the reduction of mitral inflow velocity 16 so that the pattern of reduced e and normal E/A ratio in a patient with minimal diastolic dysfunction and normal LV filling pressure may resemble the grade II diastolic dysfunction pattern. Both situations have reduced e and normal-appearing mitral inflow velocities. 17 The only difference might be the higher S/D ratio in this initial stage of reduced myocardial relaxation without change in the normal inflow pattern. Conclusions Pulmonary vein flow parameters have added prognostic value in patients with preserved EF. Diastolic dysfunction grade should take into account the systolic to diastolic pulmonary vein ratio. Patients with pseudo-normal E/A, e, 8cm/s, LA volume index 34 cc/m 2, but S/D. 1 constitute a group in the diastolic dysfunction grading system with good mid-term prognosis similar to patients with Grade I diastolic dysfunction. Study limitations The study was not powered to assess differences in overall mortality, new onset of atrial fibrillation, or cardiac death which are rare occurrences, and therefore the analysis of this end point was limited. Therefore, we focused on heart failure re-admissions and combined cardiac events as measures of outcome because these events are associated with poor survival and represent crucial end points. Cardiac arrhythmias, especially atrial fibrillation, are a problem because of the loss of the mitral A wave, pulmonary vein atrial contraction wave, and decrease in pulmonary venous systolic flow, thus analysis of pulmonary vein flow is effective only in patients with sinus rhythm. The duration of pulmonary venous flow velocity at atrial contraction was not optimally recorded in 35% of patients and may account for the limited predictive value of the DAr-A time. Our study was a cross-sectional epidemiological study and was not designed to investigate the issue of mechanism for the favourable prognosis afforded by high S/D ratio. This is an important areafor future investigation that would require a haemodynamic prospective study. Some of the patients in the group with high S/D had no, or only mild heart disease, which could have resulted in bias in the study. Conflict of interest: none declared. References 1. Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. Eur J Echocardiogr 2009;10: Rossi A, Cicoira M, Golia G, Anselmi M, Zardini P. Mitral regurgitation and left ventricular diastolic dysfunction similarly affect mitral and pulmonary vein flow Doppler parameters: the advantage of end-diastolic markers. J Am Soc Echocardiogr 2001;14: Dini FL, Dell Anna R, Micheli A, Michelassi C, Rovai D. Impact of blunted pulmonary venousflowontheoutcomeof patientswith leftventricularsystolicdysfunctionsecondary to either ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 2000; 85: Dini FL, Michelassi C, Micheli G, Rovai D. Prognosticvalue of pulmonary venous flow Doppler signal in left ventricular dysfunction: contribution of the difference in duration of pulmonary venous and mitral flowat atrial contraction. J Am Coll Cardiol 2000; 36: Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol 1994;47: Ho KK, Anderson KM, Kannel WB, Grossman W, Levy D. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation 1993;88: McMurrayJJ, Adamopoulos S, AnkerSD, AuricchioA, Bohm M, DicksteinKet al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The TaskForce for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012;33: Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA et al. Recommendations for chamber quantification. Eur J Echocardiogr 2006;7: Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD, Levine RA et al. Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr 2003;16: Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. J Am Soc Echocardiogr 2009;22: Rossvoll O, Hatle LK. Pulmonary venous flow velocities recorded by transthoracic Doppler ultrasound: relation to left ventricular diastolic pressures. J Am Coll Cardiol 1993;21: Jensen JL, Williams FE, Beilby BJ, Johnson BL, Miller LK, Ginter TL et al. Feasibility of obtaining pulmonary venous flow velocity in cardiac patients using transthoracic pulsed wave Doppler technique. J Am Soc Echocardiogr 1997;10: Appleton CP. Hemodynamic determinants of Doppler pulmonary venous flow velocity components: new insights from studies in lightly sedated normal dogs. J Am Coll Cardiol. 1997;30: Smiseth OA, ThompsonCR, Lohavanichbutr K, LingH, Abel JG, MiyagishimaRTet al. The pulmonary venous systolic flow pulse its origin and relationship to left atrial pressure. J Am Coll Cardiol 1999;34: NishimuraRA, Abel MD, HatleLK, TajikAJ. Relation of pulmonary veinto mitralflow velocities by transesophageal Doppler echocardiography. Effect of different loading conditions. Circulation. 1990;81: Sohn DW, Chai IH, Lee DJ, Kim HC, Kim HS, Oh BH et al. Assessment of mitral annulus velocity by Doppler tissue imaging in the evaluation of left ventricular diastolic function. J Am Coll Cardiol 1997;30: Oh JK, ParkSJ, Nagueh SF. Established andnovel clinical applications of diastolic function assessment by echocardiography. Circ Cardiovasc Imaging 2011;4: