Author + information
- Received January 30, 2015
- Revision received February 20, 2015
- Accepted February 20, 2015
- Published online May 1, 2015.
- Matteo Lisi, MD∗,†∗ (, )
- Matteo Cameli, MD∗,
- Francesca Maria Righini, MD∗,
- Angela Malandrino, MD∗,
- Damiana Tacchini, MD‡,
- Marta Focardi, MD, PhD∗,
- Charilaos Tsioulpas, MD§,
- Sonia Bernazzali, MD§,
- Piero Tanganelli, MD‡,
- Massimo Maccherini, MD§,
- Sergio Mondillo, MD∗ and
- Michael Y. Henein, MD, PhD†
- ∗Department of Cardiovascular Disease, University of Siena, Siena, Italy
- †Department of Public Health and Clinical Medicine, Umeå University and Heart Centre, Umeå, Sweden
- ‡Department of Pathology, University of Siena, Siena, Italy
- §Department of Cardiac Surgery, University of Siena, Siena, Italy
- ↵∗Reprint requests and correspondence:
Dr. Matteo Lisi, Department of Cardiovascular Disease, University of Siena, Policlinico Le Scotte, Viale Bracci 1, 53100 Siena, Italy.
Objectives This study was performed to determine the accuracy of right ventricular (RV) longitudinal strain (LS) in predicting myocardial fibrosis in patients with severe heart failure (HF) undergoing heart transplantation.
Background RVLS plays a key role in the evaluation of its systolic performance and clinical outcome in patients with refractory HF.
Methods We studied 27 patients with severe systolic HF (ejection fraction ≤25% and New York Heart Association functional class III to IV, despite full medical therapy and cardiac resynchronization therapy) using echocardiography before heart transplantation. RV free wall LS, right atrial LS, sphericity index (SI), and tricuspid annular plane systolic excursion (TAPSE) were all measured. Upon removal of the heart, from the myocardial histologic analysis, the ratio of the fibrotic to the total sample area determined the extent of fibrosis (%).
Results RV myocardial fibrosis correlated with RV free wall LS (r = 0.80; p < 0.0001), SI (r = 0.42; p = 0.01) and VO2 max (r = –0.41; p = 0.03), with a poor correlation with TAPSE (r = –0.34; p = 0.05) and right atrial LS (r = –0.37; p = 0.03). Stepwise multivariate analysis showed that RV free wall LS (β = 0.701, p < 0.0001) was independently associated with RV fibrosis (overall model R2 = 0.64, p < 0.0001). RV free wall LS was the main determinant of myocardial fibrosis. In the subgroup of patients with severe RV fibrosis, RV free wall LS had the highest diagnostic accuracy for detecting severe myocardial fibrosis (area under the curve = 0.87; 95% confidence interval: 0.80 to 0.94).
Conclusions In late-stage HF patients, the right ventricle is enlarged, with reduced systolic function due to significant myocardial fibrosis. RV free wall myocardial deformation is the most accurate functional measure that correlates with the extent of RV myocardial fibrosis and functional capacity.
Right ventricular (RV) systolic function is an important prognostic factor in heart failure (HF) with longitudinal systolic amplitude of motion and myocardial strain predicting exercise capacity and survival (1–4). Such measurements of RV free wall have proved essential in contributing to 80% of its stroke volume and are able to explain the cavity adaptive mechanisms to volume and pressure overload (5). RV longitudinal and transverse shortening decline also occurs in RV failure in pulmonary arterial hypertension, which in late stages is associated with progressive leftward septal displacement (6,7). The aim of this study was to determine the value of right ventricular longitudinal strain (RVLS) in predicting the extent of RV myocardial fibrosis in patients with end stage HF requiring heart transplantation (HT).
We enrolled 27 patients with end-stage HF (left ventricular [LV] ejection fraction ≤25%; New York Heart Association [NYHA] functional class III to IV despite full medical therapy and cardiac resynchronization therapy [CRT]), referred to Le Scotte Hospital of Siena for a simultaneous right heart catheterization and echocardiographic evaluation before HT (8,9).
Tissue samples of the RV free wall were obtained from explanted hearts. Patients were excluded if they were not in sinus rhythm, on mechanical ventilation, had severe mitral/tricuspid regurgitation, other valve disease, or suboptimal echocardiographic image quality. Patients’ echocardiographic results were compared with those from 25 controls, none of whom had a cardiac condition or history of any systemic disease. Patients and controls gave written informed consent before participating in the study. The project complied with the declaration of Helsinki and had been approved by the local ethics committee.
All echocardiographic examinations were performed according to the recommendations of the American and European Society of Echocardiography (10,11). Studies were performed using Vivid 7, GE Medical System echocardiograph (Horten, Norway), equipped with an adult 1.5- to 4.3-MHz phased array transducer, and an echocardiogram continuously displayed. LV ejection fraction was calculated from the apical views using the biplane modified Simpson method (10) and LV mass using the Penn formula (12) and was indexed to body surface area (BSA).
LV diastolic function
Pulsed-wave Doppler velocities of LV filling were recorded from the apical 4-chamber view by placing the sample volume by the tips of the mitral leaflets. Early (E) and late (A) diastolic LV filling velocities were registered and E/A ratio was calculated (13) and used as standard index of LV diastolic function (14). The LV filling pattern was considered restrictive when the E/A ratio was >2.0, E-wave deceleration time <140 ms, and the left atrium dilated, and >40 mm in diameter (15). Raised E/e’ was also taken as a marker of raised filling pressures (16). The same method was used to obtain RV filling velocities and measurements (17).
RV volume was measured from the 4-chamber view at end-diastole. The mid-cavity diameter was measured in the middle third at the level of the LV papillary muscles (18).
RV longitudinal function
RV longitudinal function was studied using the pulsed tissue Doppler imaging technique with the sample volume at the level of tricuspid lateral annulus from the apical 4-chamber view (19). Peak systolic (s’), early diastolic (e’), and late diastolic (a’) tricuspid annular velocities were obtained. s’ is considered to be a relatively load-independent index of RV longitudinal systolic function, and e’ and e’/a’ ratio are considered to be load-independent markers of diastolic relaxation. Tricuspid annular plane systolic excursion (TAPSE) was measured with the cursor placed at the lateral angle of the annulus from the apical 4-chamber view (20).
LV and RV longitudinal myocardial intrinsic function
LV and RV myocardial function was studied using speckle tracking echocardiography of the apical long axis, obtained from the apical 4- and 2-chamber views, during a quiet breath hold, with a frame rate of 60 to 80 per s. Three consecutive heart cycles were recorded and averaged. The off-line analysis was performed using a commercially available semiautomated 2-D strain software (EchoPac, GE, Waukesha, Wisconsin).
The peak LV longitudinal strain (LS) was defined as the peak negative value on the strain curve. The LV cavity was traced manually from the innermost endocardial edge at end-systole, and the software automatically defined the LS throughout the cardiac cycle. LS was analyzed first in the apical long axis view where the closure of the aortic valve was defined with respect to the R wave to mark end-systole. Then the same process was used for the 4-chamber and 2-chamber views. The automated algorithm provided the longitudinal peak systolic strain value for each LV segment from a 17-segment model polar plot and an average value for the 3 apical views (LV global LS) (21).
Calculation of RV free wall LS was obtained with the same method delineating a region of interest composed of three segments of the lateral wall: basal, mid-cavity, and apical. After segmental tracking, the RV LS curves were generated by the software and the average value of LS was calculated (22) (RV free wall LS [Figure 1]).
RV systolic function
RV sphericity index (SI), was calculated at end-diastole by dividing the short axis by the long-axis diameters (18,23).
Right atrial structure and function
Right atrial (RA) area and volume were measured using the biplane method of disks in the apical 4-chamber view at end-systole (10) and were indexed to BSA. RA myocardial function was studied from the apical 4-chamber view using the same methods applied for the RV and same frame rate. The RA endocardial border was manually traced, thus delineating a region of interest composed of 6 segments. RA peak atrial longitudinal strain (RALS), measured at the end of the reservoir phase, was calculated by averaging values obtained from all segments (global RALS) (24).
Right heart catheterization
The hemodynamic measurements were performed before the HT. The pressure transducers were balanced before data acquisition with the zero level at mid-axillary line. Pulmonary artery (PA) catheters were used to measure PA pressure (PAP), central venous pressure and RV stroke volume (RVSV), derived by the thermodilution technique (average of 5 cardiac cycles with <10% variation) and by the Fick equation through the sampling of a mixed central venous blood gas obtained from the PA and an arterial blood gas. RV stroke volume index (RVSVI) was calculated by dividing the RVSV by BSA. RV stroke work index (RVSWI) was derived as:where PAPm is mean PAP and RAPm is mean RA pressure. A RVSWI value of 0.25 mm Hg/l · m2 detected impaired RV function (25).
N-terminal pro-hormone of brain natriuretic peptide measurement
N-terminal pro-hormone of brain natriuretic peptide (NT-proBNP) was assessed at the time of the right heart catheterization from a peripheral venous blood sample collected in heparin-containing tubes and measured using an electrochemiluminescence immunoassay and the analytic range of the NT-proBNP assay spanned from 5 to 35,000 pg/ml.
Cardiopulmonary exercise test
Before HT, all patients underwent a maximum cardiopulmonary exercise testing in a lower-limb cycle ergometer (Inbrasport CG-04, Porto Alegre, Brazil), using a graded exercise test with an incremental protocol of 25 watts every minute until exhaustion or symptom development (American Thoracic Society/American College of Chest Physicians statement). Oxygen saturation was monitored with pulse oximeter and 12-lead electrocardiogram monitoring was recorded throughout the test. Blood pressure was taken before and at peak exercise. Furthermore, using arterial and venous blood gas analysis we obtained oxygen concentrations at rest and at peak exercise.
During the cardiopulmonary exercise, the exhaled gases were collected by a pneumotachograph attached to a mouthpiece, with concomitant nasal occlusion, and was quantified by a metabolic analyzer (MedGraphics, St. Paul, Minnesota). Gas analysis measurements; oxygen consumption (VO2), carbon dioxide production, tidal volume, respiratory rate, and minute ventilation were all made. Maximal VO2 (VO2 max) was defined by the Fick equation as the product of cardiac output and arterio-venous oxygen difference [C(a-v)O2] at peak exercise:where HR is heart rate and SV is stroke volume. VO2 max was measured every minute of exercise, as average of 6 readings and was presented as liter/minute.
Samples of the RV free wall were obtained after removing the heart. For each patient, 3 full-thickness slices of the RV (basal, mid-cavity, and apical) were obtained and the extent of myocardial fibrosis was calculated as the average of the 3 slices. This analysis was performed to better compare the extent of myocardial fibrosis with RV whole free wall LS strain. Tissue samples were subsequently prepared for histopathological examination. The average sample dimension was 1 × 0.5 cm which was subsequently fixed in 10% buffered formalin, embedded in paraffin, and cut into slices of 4-μm thick for hematoxylin-eosin and Masson’s trichrome staining. For Masson’s staining, slices were de-waxed with xylol (2 steps, 2 minutes each and soaked into a series of a gradient concentrations from 99% to 95% of alcohol). All slices were washed in distilled water and put in a solution of hematoxylin for 3 min. Subsequently, color change was performed with lithium carbonate (0.1%) to steer the hematoxylin staining. Slices were washed in pure water and colored with Red Ponceau staining (oven at 30°C for 20 s at 45 KW) then put in acid water and phosphomolybdic acid (1 min). The last stage included the addition of green-light and washing with acid water. The ratio of the fibrosis area to the total sample area of each section was used to assess the degree of RV fibrosis (%) as (fibrosis area-total area) × 100 (26).
RV strain measurement reproducibility
The software was able to track 146 of 156 (94%) acquired RV segments. Ten patients were randomly selected and Bland-Altman analysis was performed to evaluate the intraobserver and interobserver variability, demonstrating a good agreement with small bias of 0.6 ± 2.9% and 0.8 ± 4.1%, respectively.
Data are shown as mean ± SD. A p value <0.05 was considered statistically significant. Analyses were performed using the SPSS (Statistical Package for the Social Sciences, Chicago, Illinois) software Release 11.5. Pearson’s correlation coefficients were calculated to assess the relationship between continuous variables. The sensitivity and specificity were calculated using standard definition, receiver operating characteristic curves were constructed, and area under the curve was calculated for the prediction of severe RV myocardial fibrosis. Multiple regression analysis was performed to explore the independent determinants of RV fibrosis.
Patient’s mean peak VO2 was 11.2 ± 1.7 ml/kg/min (I° quartile: 9.8 ml/kg/min; II° quartile: 11.2 ml/kg/min [also the median], III° quartile: 12.8 ml/kg/min; IV° quartile: 14.6 ml/kg/min). They had reduced exercise capacity (NYHA functional class III to IV) and elevated NT-proBNP value (I° quartile: 1,545 pg/ml; II° quartile: 3,362 pg/ml [also the median], III° quartile: 5,145 pg/ml; IV° quartile: 10,479 pg/ml) (Table 1). Right heart catheterization data are listed in Table 2. Patients underwent HT 22 ± 14 days after the echocardiographic examination. Seventeen patients (63%) had severe RV myocardial fibrosis (>50%).
Patients versus controls
Compared to controls, patients had higher LV mass index (p < 0.0001), reduced ejection fraction (p < 0.0001) (inclusion criterion) and reduced LV global LS (p < 0.0001). LV e′ was lower (p = 0.002), whereas E/A and E/e′ were higher (p < 0.0001 for both) (Table 3).
Patients had increased RV inlet diameter (p = 0.002). RV SI was increased (p < 0.001), and E/A and e′/a′ (p = 0.008 and p = 0.03, respectively) were lower than controls. They also had reduced TAPSE and s′ (p = 0.001 and p = 0.003, respectively) and RV free wall LS (p < 0.0001 for both) and increased RA area (p = 0.006) with severely reduced RALS (p < 0.0001).
RV function versus fibrosis
RV myocardial fibrosis correlated with free wall LS (r = 0.80; p < 0.0001), SI (r = 0.42; p = 0.01) and VO2 max (r = –0.41; p = 0.03), with a poor correlation with TAPSE (r = –0.34; p = 0.05) and RALS (r = –0.37; p = 0.03) (Figures 2 to 5⇓⇓⇓⇓, Table 4).
Stepwise multivariate analysis showed that RV free wall LS (β = 0.701, p < 0.0001) was independently associated with RV fibrosis (overall model R2 = 0.64, p < 0.0001). RV free wall LS was the main determinant of myocardial fibrosis.
In the subgroup of patients with severe RV fibrosis, myocardial fibrosis correlated with free wall LS (r = 0.72; p < 0.0001), SI (r = 0.47; p < 0.0001), RALS (r = –0.46; p = 0.005), and VO2 max (r = –0.44, p = 0.01) with a poor correlation with TAPSE (r = –0.32; p = 0.01). RV free wall LS had the highest diagnostic accuracy for detecting severe myocardial fibrosis (area under the curve [AUC] = 0.87; 95% confidence interval: 0.80 to 0.94).
LV function versus exercise capacity
There was no relationship between LV myocardial deformation (global LS) and exercise capacity (VO2 max) (R = –0.29; p = ns).
The findings of this study can be summarized as follows: 1) in patients with end-stage HF, the RV is significantly enlarged with reduced systolic function; and 2) RV free wall LS is the best echocardiographic measure that correlates with the extent of myocardial fibrosis.
RV function in the form of reduced TAPSE has been shown to correlate with exercise capacity and to predict survival in patients with dilated cardiomyopathy and reduced LV systolic function (5,27). In the current study we have confirmed these findings in a group of patients with end-stage HF requiring HT. This gave us a unique opportunity to assess the extent of RV free wall fibrosis histologically, having considered >50% of the myocardial sample as severe fibrosis and compared it with the pattern and extent of RV dysfunction. Among a full battery of RV structure and function measurements, we identified free wall LS as the best predictor of the extent of myocardial fibrosis. Finally, the same parameter predicted the limited exercise tolerance in these patients. Thus, the reduced free wall strain, usually described as impaired intrinsic myocardial function, in fact reflects the severity of segmental fibrosis. Interestingly, there was no similar relationship between LV function and the extent of its myocardial fibrosis.
End-stage myocardial adaptation process is mostly in the form of collagen formation and fibrosis, which when mild, disturbs organ function but when severe it may have significant systemic implications (28). RV response to preload, afterload, or primary myocardial pathology is by cavity enlargement due to its remarkable thin walls compared to the left ventricle (5,6). A behavior that results in significant disturbances to myocardial fiber architecture and interstitial tissue remodeling, localized inflammatory processes, then eventually collagen formation (29) and fibrosis as we have shown. When in a small percentage, the viable segments take over the function, but when severe it results in raised end-diastolic pressure, as was the case in our patients. Because fibrosis is irreversible, it explains the intractable clinical picture of our patients who needed transplantation. Furthermore, in patients with severe myocardial fibrosis, the reduction of functional capacity (VO2) was closely related to the compromised myocardial performance in the form of RV free wall LS, again supporting the essential role of the RV longitudinal myocardial function in determining symptoms and prognosis of HF (30,31). Unfortunately, our results were unable to ascertain the role of RV free wall LS in establishing the exact levels of function impairment that determine disease progression, from early symptom development to a need for cardiac transplantation, because of the selection bias.
This study shows that RVLS is able to estimate RV systolic performance and to predict the extent of myocardial fibrosis in patients with end-stage HF requiring cardiac transplantation. These data can be useful to guide towards optimal medical therapy in HF patients and to identify patients with RV dysfunction before they develop severe irreversible myocardial fibrosis. Having such sensitivity, RV free wall LS could be used to identify patients likely to respond to more aggressive medical treatment, i.e., pressure offloading, compared to those needing an LV assist device implantation (32). Even in the latter, RV free wall LS might stratify patients needing cardiac transplantation rather than a hope for recovery by LV assist device (30).
The measurements of global and free wall RVLS depend on adequate 2-dimensional apical views to permit an easy delineation of the RV endocardial border and to permit the visualization of the entire RV free wall. Despite these limitations, the RV speckle tracking echocardiography analysis was performed in all patients with good reproducibility. Our study has a limited patient number, despite being a homogenous population. Eleven patients (41%) had CRT and RV leads, and although this could affect RV systolic and diastolic time relations, it is unlikely to alter intrinsic myocardial function. This claim is supported by previous studies showing that RV apical pacing did not affect RV systolic function, despite induction of electromechanical dyssynchrony (33). Finally, our results cannot propose a cut-off value for RVLS because absolute measurements may be vendor dependent.
In late-stage HF patients, the right ventricle is enlarged with reduced systolic function due to significant myocardial fibrosis. RV free wall myocardial deformation is the most accurate functional measure that correlates with the extent of RV myocardial fibrosis and functional capacity.
COMPETENCY IN MEDICAL KNOWLEDGE: In patients with end-stage HF the RV is enlarged and its systolic function is severely compromised due to significant fibrosis. RV free wall longitudinal strain is the most accurate echocardiographic measure that correlates with the extent of myocardial fibrosis.
TRANSLATIONAL OUTLOOK: If confirmed in large future longitudinal studies, RV free wall longitudinal strain might assist in stratifying patients who need transplantation contrary to those who might benefit from assist devices as a bridge to recovery.
All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- cardiac resynchronization therapy
- heart failure
- N-terminal pro-hormone of brain natriuretic peptide
- New York Heart Association
- receiver operating characteristic
- tricuspid annular plane systolic excursion
- Received January 30, 2015.
- Revision received February 20, 2015.
- Accepted February 20, 2015.
- American College of Cardiology Foundation
- Zhao Y.,
- Henein M.Y.,
- Mörner S.,
- Gustavsson S.,
- Holmgren A.,
- Lindqvist P.
- Francis G.S.,
- Greenberg B.H.,
- Hsu D.T.
- Lang R.M.,
- Bierig M.,
- Devereux R.B.,
- et al.
- Devereux R.B.,
- Reicheck N.
- Quinones M.A.,
- Otto C.M.,
- Stoddard M.,
- Waggoner A.,
- Zoghbi W.A.
- Nishimura R.A.,
- Tajik A.J.
- Ommen S.R.,
- Nishimura R.A.,
- Appleton C.P.,
- et al.
- Sade L.E.,
- Gulmez O.,
- Eroglu S.,
- Sezgin A.,
- Muderrisoglu H.
- Rudski L.G.,
- Lai W.W.,
- Afilao J.,
- et al.
- Yu C.M.,
- Sanderson J.E.,
- Oh J.K.
- Mondillo S.,
- Galderisi M.,
- Mele D.,
- et al.,
- Echocardiography study group of the Italian Society of Cardiology (Rome, Italy)
- Cameli M.,
- Lisi M.,
- Righini F.M.,
- Mondillo S.
- Matthews J.C.,
- Koelling T.M.,
- Pagani F.D.,
- Aaronson K.D.
- Cameli M.,
- Lisi M.,
- Righini F.M.,
- et al.
- Tan J.L.,
- Prati D.,
- Gatzoulis M.A.,
- Gibson D.,
- Henein M.Y.,
- Li W.
- Munkhammar P.,
- Carlsson M.,
- Arheden H.,
- Pesonen E.
- Grant A.D.,
- Smedira N.G.,
- Starling R.C.,
- Marwich T.H.