Background
The assessment of right ventricular (RV) function with two-dimensional echocardiography can be challenging in patients with pulmonary hypertension, especially in those with chronic pulmonary disease. The aim of the present study was to evaluate the feasibility of measuring RV longitudinal peak systolic strain (LPSS) in the echocardiographic subcostal view in patients with suspected pulmonary hypertension.
Methods
A total of 179 patients evaluated for pulmonary hypertension were included (85 with systemic disorder, 64 with pulmonary disease, and 30 with RV dilatation and dysfunction). Additionally, 30 normal controls were evaluated. The feasibility of RV LPSS speckle-tracking measurements in the apical four-chamber view and in the subcostal view was evaluated. Furthermore, the RV LPSS speckle-tracking measurements performed in these two echocardiographic views were compared.
Results
The feasibility of RV LPSS in the subcostal view was 95.3%, 92.2%, 93.3%, and 93.3% in patients with systemic disorder, with pulmonary disease, with RV dilatation and dysfunction, and controls, respectively. In comparison, the feasibility of RV LPSS in the apical four-chamber view was 92.9%, 82.8%, 90%, and 93.3% in each group, respectively. Bland-Altman analysis showed good agreement between measurements in both echocardiographic views (systemic disorder: mean bias, −0.14; pulmonary disease: mean bias, 0.28; RV dilatation and dysfunction: mean bias, 0.3; and normal controls: mean bias, −0.14).
Conclusions
The subcostal view provides a good alternative for RV strain assessment in patients who are evaluated for pulmonary hypertension. This measurement may be a valuable surrogate of RV function in patients with challenging apical windows.
The performance of the right ventricle is an important prognostic marker in patients with pulmonary hypertension. Therefore, reliable, highly feasible, and reproducible methods of right ventricular (RV) function assessment are of interest. Echocardiography is the most widely available imaging test to evaluate patients with pulmonary hypertension. However, the assessment of RV function using two-dimensional echocardiography relies on geometric assumptions that may reduce the accuracy of this method. In addition, some subgroups of patients with pulmonary hypertension, such as patients with chronic obstructive pulmonary disease, may not have appropriate apical acoustic windows to visualize the right ventricle. In this group of patients, RV systolic function is often evaluated in the subcostal view, which may provide better image quality. However, two-dimensional measurements in this view may not be comparable with measurements obtained from the apical view.
Speckle-tracking echocardiography permits the angle-independent evaluation of myocardial strain. This methodology does not rely on geometric assumptions and has been shown to provide reliable information on ventricular performance. In patients with pulmonary hypertension, the assessment of longitudinal strain of the right ventricle with speckle-tracking strain imaging may be a valid estimate of RV function. RV longitudinal peak systolic strain (LPSS) measurements can be obtained from the apical and subcostal views. However, it remains unclear whether these two approaches provide comparable results. The purpose of the present study was to assess the feasibility of RV LPSS measurement in the subcostal view in a broad spectrum of patients who were referred for the evaluation of pulmonary hypertension.
Methods
Patient and Data Collection
Patients who were referred to the outpatient clinic for diagnosis and treatment of suspected pulmonary hypertension between January 2005 and August 2011 were included. All patients were screened extensively according to the international guidelines for pulmonary hypertension. On the basis of the results of the screening protocol, patients were categorized into one of the five different groups according to the Dana Point classification for pulmonary hypertension etiology : group 1 consisted of patients with pulmonary arterial hypertension due to a primary disease process in the pulmonary arteries, group 2 of patients with pulmonary hypertension due to underlying left-heart disease, group 3 of patients with pulmonary hypertension secondary to pulmonary and/or hypoxic disease, group 4 of patients with pulmonary hypertension due to chronic thromboembolic disease, and group 5 of patients with pulmonary hypertension due to multifactorial mechanisms. For the purpose of this study, only patients in group 1 (pulmonary arterial hypertension), group 2 (left-heart disease with established RV dilatation and dysfunction), and group 3 (pulmonary disease) were included. Patients with complex congenital heart disease or who had permanent atrial fibrillation were excluded. Additionally, patients were divided into two groups on the basis of a cutoff value of tricuspid annular plane systolic excursion (TAPSE) of ≤18 mm, as a parameter for RV dysfunction.
Furthermore, 30 controls without evidence of structural heart disease were selected from an echocardiographic database. Control individuals who were referred for echocardiographic evaluation of known valvular heart disease, murmur, congestive heart failure, or cardiac transplantation were excluded. Therefore, the control group included subjects who were referred for atypical chest pain, palpitations, or syncope without murmurs and who showed normal echocardiographic results.
As part of the departmental protocol, complete transthoracic echocardiography was performed, including assessment of RV function with two-dimensional speckle-tracking strain analysis. RV LPSS was measured in the apical four-chamber and subcostal views. The feasibility of RV LPSS measurements in the apical four-chamber and subcostal views was evaluated in the group of patients with pulmonary disease, in the group of patients with systemic disorder, in the group of patients with RV dilatation and dysfunction, and in normal controls. Finally, the agreement between RV LPSS measured in the apical four-chamber view and in the subcostal view was evaluated in the overall population and for each group of patients.
Echocardiography
Echocardiography was performed using a commercially available system (Vivid 7 and E9; GE-Vingmed Ultrasound AS, Horten, Norway). Images were obtained in the parasternal and apical views with the patient in the left lateral decubitus position and in the subcostal view with the patient in the supine position using a 3.5-MHz transducer. Standard two-dimensional, color, pulsed, and continuous-wave Doppler data were acquired and saved in regular cine loop format. Using dedicated software, the images were analyzed offline (EchoPAC version 111.0.00; GE-Vingmed Ultrasound AS). First, left ventricular end-systolic volume and end-diastolic volume were measured in the apical two-chamber and four-chamber views, and left ventricular ejection fraction was derived according to biplane Simpson’s method. Thereafter, RV dimensions were assessed by measuring RV end-diastolic area (RVEDA) and RV end-systolic area (RVESA) in the apical four-chamber view. RV function was evaluated by calculating the fractional area change (FAC) and measuring TAPSE. TAPSE was measured in the apical four-chamber views by aligning the M-mode cursor along the movement of the lateral tricuspid annulus. Valve disease was evaluated according to the current guidelines. Systolic pulmonary arterial pressure was estimated by calculating the systolic pressure gradient between the right ventricle and the right atrium by the maximum velocity of the tricuspid regurgitant jet using the modified Bernoulli equation and adding right atrial pressure. Right atrial pressure was estimated according to the diameter and inspiratory collapse of the inferior vena cava.
Strain Analysis by Speckle-Tracking Echocardiography
RV LPSS was measured in the apical four-chamber and subcostal views using speckle-tracking analysis. As previously described, speckle-tracking echocardiography enables the angle-independent evaluation of myocardial strain by tracking frame to frame the movement of the speckles, or natural acoustic markers, within the myocardium on two-dimensional grayscale images. Longitudinal strain is defined as the percentage of shortening of the region of interest relative to the original length and is conventionally presented as a negative value. Only images with frame rates > 40 frames/sec were selected for reliable analysis. The endocardial border of the RV free wall was manually traced at an end-systolic frame, and the software displayed automatically a region of interest including the myocardial wall. This region of interest can be manually adjusted to the thickness of the myocardium to ensure adequate tracking. Peak systolic strain was measured in the basal, midventricular, and apical segments of the RV free wall in both the apical four-chamber and subcostal views and averaged to obtain RV LPSS ( Figure 1 ). Segments of poor quality were noted as not feasible in either the apical four-chamber or the subcostal view or in both views. Patients with nonfeasible analyses were excluded from further analysis.
Statistical Analysis
All continuous data are presented as mean ± SD. Categorical data are presented as frequencies or percentages. The feasibility of RV LPSS measurement was compared between the two echocardiographic views (apical four-chamber and subcostal) in all four patient groups using χ 2 tests. Differences in baseline clinical and echocardiographic characteristics between the four groups were compared using analysis of one-way variance. Post hoc analyses were performed using Bonferroni’s correction for multiple comparisons. Correlation between RV LPSS measurements in the apical four-chamber and subcostal views was calculated using intraclass correlation coefficients. Bland-Altman analysis was used to evaluate the agreement between RV LPSS measurements in the apical four-chamber and subcostal views for all patient groups. Conventional parameters to assess RV performance, including TAPSE and RV dimensions (RVEDA and RVESA), were correlated with RV LPSS, measured in the apical four-chamber and subcostal views, using Pearson’s correlation. Multivariate logistic regression analysis was performed to detect the accuracy of RV LPSS to diagnose RV dysfunction, defined as TAPSE ≤ 18 mm. In a series of nested models, parameters of RV function (RVEDA and FAC) were entered individually in a stepwise fashion. Furthermore, the incremental value of RV LPSS over RV FAC to detect RV dysfunction was established by calculating the change in global χ 2 values. Finally, 15 patients were randomly selected to test the intraobserver and interobserver reproducibility of RV LPSS measurements in the apical four-chamber and subcostal views. Two independent observers measured RV LPSS in the apical four-chamber and in the subcostal views in a blinded manner. For intraobserver reproducibility, the measurements were performed by the same observer with a 1-month interval. Bland-Altman analyses were performed. In addition, the coefficient of variation and the intraclass correlation coefficient were calculated. All analyses were performed using SPSS version 17.0 for Windows (SPSS, Inc., Chicago, IL).
Results
Clinical and Echocardiographic Characteristics
A total of 179 patients were included in this evaluation: 85 patients had underlying systemic disorder, 64 had underlying pulmonary disease associated with pulmonary hypertension, and 30 showed left-sided heart disease and RV dilatation and dysfunction. In addition, 30 normal controls were included. Table 1 summarizes the clinical and echocardiographic characteristics of the four subgroups. As expected, left ventricular function was significantly reduced in the group of patients with RV dilatation and dysfunction compared with the other patient groups. By definition, RV function assessed by TAPSE and RV FAC was depressed in the group of patients with RV dilatation and dysfunction. The estimated systolic pulmonary artery pressure was >50 mm Hg in 16 (19%), 20 (31%), and 18 (60%) patients with systemic disorder, pulmonary disease, and RV dilatation and dysfunction, respectively. Control individuals did not show pulmonary hypertension. The systolic pressure gradient between the right ventricle and the right atrium was measurable in 90.9% of the total population. In 9.1%, no significant tricuspid regurgitation was measurable. Additionally, in 82.8% of the total population, mild to moderate tricuspid regurgitation was present, while 8.1% of the population had severe tricuspid regurgitation.
| Clinical characteristic | Systemic disorder ( n = 85) | Pulmonary disease ( n = 64) | LV disease with RV dysfunction ( n = 30) | Normal controls ( n = 30) | P |
|---|---|---|---|---|---|
| Mean age (y) | 53 ± 14 | 63 ± 11 | 64 ± 10 | 53 ± 12 | <.001 |
| Men | 23 (27%) | 35 (55%) | 22 (73%) | 10 (33%) | <.001 |
| BSA (m 2 ) | 1.82 ± 0.2 | 1.86 ± 0.2 | 1.92 ± 0.2 | 1.86 ± 0.2 | .2 |
| NYHA functional class | 1.93 ± 0.8 | 2.44 ± 0.6 | 2.73 ± 0.58 | 1.00 ± 0.19 | <.001 |
| Echocardiography | |||||
| Left ventricle | |||||
| LVEDV (mL) | 94 ± 25 | 95 ± 24 | 144 ± 87 | 106 ± 28 | <.001 |
| LVESV (mL) | 42 ± 15 | 43 ± 13 | 99 ± 75 | 41 ± 14 | <.001 |
| LVEF (%) | 56 ± 8 | 55 ± 7 | 38 ± 14 | 62 ± 7 | <.001 |
| Right ventricle | |||||
| RVEDA (cm 2 ) | 19 ± 6 | 20 ± 6 | 29 ± 5 | 16 ± 3 | <.001 |
| RVESA (cm 2 ) | 12 ± 5 | 13 ± 4 | 21 ± 5 | 10 ± 2 | <.001 |
| RVD 1 (cm) | 3.8 ± 0.7 | 3.8 ± 0.7 | 5.1 ± 0.6 | 3.0 ± 0.3 | <.001 |
| RVD 2 (cm) | 2.6 ± 0.8 | 2.6 ± 0.6 | 3.4 ± 0.7 | 2.1 ± 0.1 | <.001 |
| RVD 3 (cm) | 7.1 ± 0.9 | 7.3 ± 0.9 | 8.1 ± 1.1 | 7.2 ± 0.8 | <.001 |
| TAPSE (mm) | 19 ± 4 | 19 ± 4 | 12 ± 2 | 23 ± 3 | <.001 |
| RV FAC (%) | 38 ± 10 | 37 ± 8 | 29 ± 10 | 37 ± 9 | <.001 |
| Estimated SPAP (mm Hg) | 39 ± 19 | 46 ± 16 | 59 ± 20 | 21 ± 5 | <.001 |
| Estimated SPAP > 50 mm Hg | 16 (19%) | 20 (31%) | 18 (60%) | — | <.001 |
Feasibility of RV Longitudinal Strain Measurements in the Apical Four-Chamber and Subcostal Views
In the total study population of 209 individuals, the measurement of RV LPSS was feasible in the apical four-chamber view in 187 individuals (89.5%) and in the subcostal view in 196 individuals (93.8%). The percentages of feasibility of RV LPSS measurement in the apical four-chamber view and subcostal view are presented in Figure 2 for all patient groups. There were no significant differences in the feasibility of RV LPSS measurement between the apical four-chamber view and the subcostal view in the systemic disorder group ( P = .26), the pulmonary disease group ( P = .58), the RV dysfunction group ( P = 1.00), or the normal control group ( P = 1.00) ( Figure 2 ).
Correlation and Agreement between RV Longitudinal Strain Measured in the Apical Four-Chamber View and Subcostal View
In the patient group with underlying systemic disorder, RV LPSS in the apical four-chamber view and in the subcostal view was −22.56 ± 7.07% and −22.42 ± 6.97% ( P = .304), respectively. In the group of patients with pulmonary disease, RV LPSS was −19.72 ± 6.08% in the apical four-chamber view and −19.99 ± 6.06% in the subcostal view ( P = .087). Among patients with RV dilatation and dysfunction, RV LPSS was −14.74 ± 4.87% in the apical four-chamber view and −15.04 ± 4.96% in the subcostal view ( P = .145). In the normal control group, RV LPSS was −25.86 ± 4.17% and −25.72 ± 4.16 for the apical four-chamber and subcostal views ( P = .431), respectively ( Table 2 ).
| Group | RV LPSS | P | |
|---|---|---|---|
| Apical four-chamber view (%) | Subcostal view (%) | ||
| Systemic disorder | −22.56 ± 7.07 | −22.42 ± 6.97 | .304 |
| Pulmonary disease | −19.72 ± 6.08 | −19.99 ± 6.06 | .087 |
| RV dilatation and dysfunction | −14.74 ± 4.87 | −15.04 ± 4.96 | .145 |
| Normal control | −25.86 ± 4.17 | −25.72 ± 4.16 | .431 |
Figure 3 A shows the correlation between RV LPSS measurements obtained from the apical four-chamber and subcostal views observed in the group of patients with systemic disorder ( r = 0.987, P < .001). In addition, Bland-Altman analysis showed good agreement between measurements in both echocardiographic views, with no significant underestimation or overestimation and tight limits of agreement in this group of patients (mean bias, −0.14; 95% limits of agreement, −2.4% to 2.2%; Figure 3 B). In the patient group with underlying pulmonary disease, the correlation between RV LPSS measurements in the apical four-chamber view and in the subcostal view was 0.987 ( P < .001), with good measurement agreement between both echocardiographic views ( Figures 3 C and 3 D). There was no significant underestimation or overestimation, and the mean bias was 0.28 (95% limits of agreement, −1.93% to 2.46%). The correlation between RV LPSS measurements in the apical four-chamber and subcostal views in the group of patients with RV dilatation and dysfunction was 0.977 ( P < .001; Figure 3 E). Figure 3 F shows good agreement between both echocardiographic views, with a mean bias of 0.30 (95% limits of agreement, −1.70% to 2.3%). There was no significant underestimation or overestimation observed in this patient group. In the normal control group, the correlation between RV LPSS measured in the apical four-chamber view and in the subcostal view was also good ( r = 0.973, P < .001; Figure 3 G). Additionally, Bland-Altman analysis demonstrated mean bias of −0.14 with good agreement (95% limits of agreement, −1.96% to 1.68%; Figure 3 H).
