Abstract
Rationale: Right ventricular (RV) function is an important determinant of prognosis in pulmonary hypertension. However, noninvasive assessment of the RV function is often limited by complex geometry and poor endocardial definition.
Objectives: To test whether the degree of tricuspid annular displacement (tricuspid annular plane systolic excursion [TAPSE]) is a useful echo-derived measure of RV function with prognostic significance in pulmonary hypertension.
Methods: We prospectively studied 63 consecutive patients with pulmonary hypertension who were referred for a clinically indicated right heart catheterization. Patients underwent right heart catheterization immediately followed by transthoracic echocardiogram and TAPSE measurement.
Results: In the overall cohort, a TAPSE of less than 1.8 cm was associated with greater RV systolic dysfunction (cardiac index, 1.9 vs. 2.7 L/min/m2; RV % area change, 24 vs. 33%), right heart remodeling (right atrial area index, 17.0 vs. 12.1 cm2/m), and RV–left ventricular (LV) disproportion (RV/LV diastolic area, 1.7 vs. 1.2; all p < 0.001), versus a TAPSE of 1.8 cm or greater. In patients with pulmonary arterial hypertension (PAH; n = 47), survival estimates at 1 and 2 yr were 94 and 88%, respectively, in those with a TAPSE of 1.8 cm or greater versus 60 and 50%, respectively, in subjects with a TAPSE less than 1.8 cm. The unadjusted risk of death (hazard ratio) in patients with a TAPSE less than 1.8 versus 1.8 cm or greater was 5.7 (95% confidence interval, 1.3–24.9; p = 0.02) for the PAH cohort. For every 1-mm decrease in TAPSE, the unadjusted risk of death increased by 17% (hazard ratio, 1.17; 95% confidence interval, 1.05–1.30; p = 0.006), which persisted after adjusting for other echocardiographic and hemodynamic variables and baseline treatment status.
Conclusions: TAPSE powerfully reflects RV function and prognosis in PAH.
Pulmonary arterial hypertension (PAH) is a clinical syndrome characterized by progressive increases in pulmonary vascular load, leading to marked increases in pulmonary artery pressure, exercise intolerance, and ultimately death (1). However, the natural history of PAH is heterogeneous, with more rapid clinical deterioration seen in patients with the greatest degree of right ventricular (RV) dysfunction (2, 3). This underscores the importance of RV function assessment in patients with PAH, and places a premium on a modality that is noninvasive and comprehensive, yet practical. However, complex RV chamber geometry and suboptimal RV endocardial definition have limited the application of noninvasive RV function assessment in clinical practice (4).
Unlike the left ventricle, which shortens relatively symmetrically in the transverse and longitudinal planes, muscle fiber orientation of the right ventricle dictates that contraction occurs predominantly along the longitudinal plane (5). As a result, systolic displacement of the tricuspid annulus toward the RV apex (longitudinal plane), referred to as tricuspid annular plane systolic excursion (TAPSE), closely correlates with RV ejection fraction (6). Importantly, TAPSE does not require geometric assumptions or RV endocardial definition, and thus has been found to be highly reproducible and practical (6, 7). Subsequent studies have confirmed that TAPSE is an excellent measure of RV systolic function, and that a depressed TAPSE portends a poor prognosis in patients with dilated cardiomyopathy and after acute inferior wall myocardial infarction (7–9). However, the clinical and prognostic significance of TAPSE has not been tested in adult patients with PAH in whom the prevalence and prognostic importance of RV dysfunction would suggest that such an approach may be of great value. Thus, we conducted a prospective observational study to determine whether TAPSE represents a simple noninvasive measure of RV function and has prognostic value in patients with pulmonary hypertension, including PAH. Some of the results of the current study have been previously reported in abstract form (10).
This study was approved by the Johns Hopkins Institutional Review Board. Informed consent was obtained from all subjects prior to enrollment.
From March to September of 2004, 72 consecutive patients with known or suspected pulmonary hypertension were screened for study enrollment at the time of referral for a clinically indicated right heart catheterization (RHC). Within 1 h after the RHC, patients underwent a two-dimensional (2-D) echocardiographic/Doppler examination in the cardiac catheterization recovery suite. Inclusion criteria were a mean pulmonary artery pressure (mPAP) of 25 mm Hg or greater, mean pulmonary capillary wedge pressure (PCWP) of 15 mm Hg or less, and adequate echocardiographic windows. A total of nine patients were excluded: seven for normal pulmonary hemodynamics and two for technically inadequate echocardiographic windows. Thus, 63 patients were monitored in this prospective observational study.
All patients underwent a standard RHC for hemodynamic data collection.
Echocardiographic imaging was performed using a Philips Sonos 5500 and a 3.2-MHz transducer (Philips Medical Systems, Andover, MA). A 2-D Doppler examination was performed using a prespecified echocardiographic protocol using views specifically designed to optimize RV imaging. To obtain TAPSE, the apical four-chamber view was used, and an M-mode cursor was placed through the lateral tricuspid annulus in real time. Off-line, the brightness was adjusted to maximize the contrast between the M-mode signal arising from the tricuspid annulus and the background. TAPSE was measured as the total displacement of the tricuspid annulus (centimeters) from end-diastole to end-systole, with values representing the average TAPSE of three to five beats (9).
Additional hemodynamic and echocardiographic method details are reported in the online supplement.
Median follow-up was 19.3 mo. At study termination, the vital status of each patient was confirmed by review of medical records, phone contact, and the Social Security Death Index. The primary endpoint was all-cause mortality.
Continuous variables were summarized by mean ± SD or median (interquartile range), as appropriate. Differences were detected by two-tailed, unpaired Student's t test. p < 0.05 was considered significant.
A receiver operating characteristic (ROC) curve was used initially to test the ability of TAPSE to detect an RV stroke volume index (SVI) less than the median SVI value in the dataset (< 29 ml/m2). This value is an accepted lower limit of normal for SVI, corresponding to a cardiac index of 2.2 L/min, assuming a heart rate of 75 beats/min (11). The optimal cut point was based on the greatest proportion of correctly classified patients identified by ROC curve analysis (point highest and to the left on the ROC curve), and determined to be 1.8 cm.
Thereafter, we compared hemodynamic and echocardiographic variables in patients with a TAPSE less than 1.8 cm or 1.8 cm or greater to examine the association between TAPSE and various aspects of RV function. Variables were grouped as measures of RV systolic function, right heart remodeling, and RV–left ventricular (LV) disproportion to illustrate differences in these three important aspects of RV function in patients with a TAPSE value less than 1.8 cm or 1.8 cm or greater.
Using this dichotomized TAPSE (</⩾ 1.8 cm), a time-to-event analysis was performed. The Kaplan-Meier product limit estimator was used to compare the time to death between patients with a TAPSE less than 1.8 cm versus those with a TAPSE 1.8 cm or greater. Multivariable Cox proportional hazards models (12) were constructed using TAPSE as a continuous or a dichotomous variable as detailed in the online Methods.
Table 1 summarizes the clinical and hemodynamic characteristics of the overall study population. The majority of patients (47/63 patients; 75%) were in group 1 of the World Health Organization (WHO) classification of pulmonary hypertension, with the remaining patients having either pulmonary hypertension associated with respiratory disease and/or hypoxemia (WHO group 3, n = 13) or chronic thromboembolic pulmonary hypertension (WHO group 4, n = 3). At enrollment, 70% of the patients were WHO functional class III or IV. Approximately 50% of the patients were receiving therapy for pulmonary hypertension at the time of enrollment, with the remainder being newly diagnosed. During the study period, 95% of patients were receiving various medical therapies for pulmonary hypertension (Table 1).
Clinical characteristics | |
|---|---|
| Age, yr | 55 ± 15 |
| Female, n (%) | 52 (83) |
| White, n (%) | 48 (76) |
| African American, n (%) | 13 (21) |
| Other, n (%) | 2 (3) |
| WHO class III–IV, n (%) | 44 (70) |
| WHO class I–II, n (%) | 19 (30) |
| Diagnoses, n (%) | |
| PAH | |
| IPAH | 23 (37) |
| SSc-PAH | 20 (32) |
| Connective tissue disease PAH (other than SSc) | 4 (6) |
| Pulmonary hypertension associated with respiratory disorders and/or hypoxemia | 13 (21) |
| Chronic thromboembolic disease | 3 (5) |
| Treatment, pre/post* | |
| Epoprostenol (IV) | 16%/23% |
| Iloprost (INH) or treprostenil (SC) | 5%/8% |
| Bosentan | 22%/41% |
| Sildenafil | 6%/23% |
| Bosentan + sildenafil | 3%/13% |
| Epoprostenol + bosentan or sildenafil | 3%/13% |
| Hemodynamics | |
| Heart rate, beats/min | 82 ± 15 |
| Systolic BP, mm Hg | 119 ± 17 |
| Mean right atrial pressure, mm Hg | 10 ± 5 |
| Systolic pulmonary artery pressure, mm Hg | 75 ± 21 |
| Mean pulmonary artery pressure, mm Hg | 45 ± 12 |
| Pulmonary capillary wedge pressure, mm Hg | 10 ± 3 |
| Pulmonary vascular resistance, mm Hg/L/min | 10 ± 5 |
| Cardiac index, L/min/m2 | 2.3 ± 0.7 |
| Stroke volume index, ml/m2 | 29 ± 10 |
| Mixed venous O2 saturation, % | 64 ± 9 |
The baseline hemodynamics of the study population were notable for a markedly reduced SVI, cardiac index, Sv̄O2, and elevated right atrial pressure (RAP) as compared with normal reference values (11). Pulmonary vascular resistance (PVR) was also markedly elevated, whereas PCWP was normal, reflecting the precapillary etiology of pulmonary hypertension in our cohort.
Baseline echocardiographic values are listed in Table 2. Patients had TAPSE and RV fractional area change (RVFAC) values well below those previously reported in normal subjects (i.e., normal TAPSE ⩾ 2.5 cm; normal RVFAC ⩾ 40% [6, 13, 14]). In addition, patients also had evidence of marked right atrial and RV dilatation, both in absolute terms and in relative proportion to left heart dimensions (13, 14). LV size and systolic function were within normal range, implicating RV dysfunction as the source of the hemodynamic (and echocardiographic) derangements reported herein.
TAPSE, cm | 1.9 ± 0.6 |
| RV fractional area change, % | 30 ± 13 |
| RA, cm | 4.8 ± 0.8 |
| LA, cm | 3.4 ± 0.6 |
| RVIDd, cm | 4.6 ± 0.8 |
| LVIDd, cm | 4.1 ± 0.8 |
| RAA index, cm2/m | 14.2 ± 4.9 |
| RVAd index, cm2/m | 15.7 ± 3.7 |
| RA:LA area ratio | 1.5 ± 0.4 |
| RV:LV end-diastolic area ratio | 1.4 ± 0.5 |
| Diastolic eccentricity index | 1.4 ± 0.3 |
| Systolic eccentricity index | 1.5 ± 0.6 |
| Max TR velocity, m/s | 3.8 ± 0.7 |
| TR severity (grade 0–3) | 1.8 ± 1.0 |
| Pericardial effusion, n (%) | 16 (25) |
| LV ejection fraction, % | 57 ± 9 |
ROC curve analysis revealed that TAPSE was a highly sensitive and specific indicator of a depressed RV SVI (area under the ROC curve [AUC] 0.87, p < 0.0001; Figure 1). The optimal ROC-derived cutoff was 1.8 cm. Likewise, TAPSE also detected a depressed cardiac index (AUC, 0.76; p = 0.0006), also at a cutoff of 1.8 cm (Figure E1 of the online supplement). Figure 2 shows representative M-mode tracings and hemodynamics from two subjects: one with relatively preserved TAPSE (A) and the other with depressed TAPSE (B). The subjects had similar mPAP; however, the patient with a TAPSE of 1.5 cm had a much lower SVI and higher PVR than the subject with a TAPSE of 2.3 cm.
Figure 1. Receiver operating characteristic curve (ROC) demonstrating the ability of tricuspid annular plane systolic excursion (TAPSE) to predict a right ventricular stroke volume index < 29 ml/m2. The arrow indicates the optimal discriminatory cut point, corresponding to a TAPSE value of 1.8 cm. AUC = area under the ROC curve; CI = confidence interval.
[More] [Minimize]
Figure 2. Representative M-mode recordings through the lateral tricuspid valve annulus for the purpose of measuring the TAPSE. Tracings were taken from two separate patients: (A) one with relative preservation of right ventricular (RV) function and the other (B) from a patient with significant RV dysfunction. Below each panel are respective hemodynamics for each patient. CI = cardiac index; mPAP = mean pulmonary arterial pressure; PVR = pulmonary vascular resistance; RAP = right atrial pressure; SVI = stroke volume index.
[More] [Minimize]Figure 3 summarizes the striking differences in RV systolic function (A), right heart remodeling (B), and RV–LV disproportion (C) observed between patients with a TAPSE of 1.8 cm or greater and those with a TAPSE less than 1.8 cm. Patients with a TAPSE less than 1.8 cm also had significantly higher RAP (12 ± 5 mm Hg) than those with a TAPSE of 1.8 cm or greater (8 ± 5 mm Hg; p = 0.01).
Figure 3. (A–C) Bar graphs compare indices of RV systolic function, right heart remodeling, and RV–left ventricular (LV) disproportion in patients with a TAPSE ⩾ 1.8 cm (open bars) and < 1.8 cm (solid black bars). (A) Measures of RV systolic function. RVFAC = right ventricular fractional area change; RVSWI = right ventricular stroke work index. (B) Measures of right heart remodeling. RA and RV area index denote respective chamber area indexed to patient height. (C) Measures of RV–LV disproportion.
[More] [Minimize]Figure 4 highlights the differences in PVR and pulmonary arterial compliance (stroke volume to pulmonary artery pulse pressure ratio) between the TAPSE subgroups despite similar mPAP, implicating the significance of RV afterload on this measure of RV function. Similarly, we observed a linear inverse relationship between TAPSE and PVR (r = −0.52, p < 0.0001; y = −0.055x + 2.45).
Figure 4. Bar graphs illustrating mPAP, PVR, and pulmonary artery compliance (stroke volume to pulmonary artery pulse pressure ratio [SV/PP]) in patients with a TAPSE ⩾ 1.8 cm (open bars) and < 1.8 cm (solid black bars).
[More] [Minimize]Patients with a TAPSE less than 1.8 cm had significantly smaller left atrial (3.2 ± 0.6 vs. 3.6 ± 0.7 cm; p = 0.03) and LV diastolic dimensions (3.6 ± 0.7 vs. 4.5 ± 0.6 cm; p < 0.0001) than subjects with a TAPSE of 1.8 cm or greater. Furthermore, patients with a TAPSE less than 1.8 cm had comparatively worse LV diastolic function, because the maximal transmitral E-wave velocity (63 ± 33 vs. 80 ± 27 cm/s; p = 0.03) and E′ septum (6.4 ± 2.0 vs. 8.1 ± 2.6 cm/s; p = 0.009) were significantly lower than in subjects with a TAPSE of 1.8 cm or greater. Interestingly, there was also a linear inverse correlation between the diastolic eccentricity index and E′ septum (r = −0.45, p = 0.0004; y = −2.97x + 11.7), suggesting that LV relaxation was adversely affected by leftward interventricular septal displacement. Subjects with a TAPSE less than 1.8 cm also trended toward lower systolic blood pressure (115 ± 13 vs. 122 ± 16 mm Hg; p = 0.05), with higher heart rates (87 ± 17 vs. 78 ± 13 beats/min; p = 0.04) and systemic vascular resistance (1,980 ± 670 vs. 1,302 ± 373 dyne · s · cm−5; p < 0.0001) than those with a TAPSE of 1.8 cm or greater, indicating greater systemic hemodynamic compensation (and thus less hemodynamic reserve) in those with more depressed TAPSE.
Patients with a TAPSE less than 1.8 cm also had more tricuspid regurgitation (TR; grade 2.2 ± 0.9 vs. 1.4 ± 0.9; p = 0.002) than subjects with relative preservation of TAPSE. Likewise, patients with moderate or severe TR (grade ⩾ 2) had a significantly lower TAPSE (1.7 ± 0.1 vs. 2.2 ± 0.1 cm; p = 0.01) than patients with less TR (grade ⩽ 1). In addition, 83% of patients with a pericardial effusion greater than grade 1 had a TAPSE less than 1.8 cm (p = 0.001).
In the overall cohort, after a median follow-up of 19.3 mo (interquartile range, 9.5–22.1 mo), 23 of the 63 subjects had died (37% mortality). Sixteen of the 23 deaths (70%) occurred in subjects with a TAPSE less than 1.8 cm. The average TAPSE among survivors was 2.1 ± 0.6 cm, versus 1.6 ± 0.5 cm among nonsurvivors (p = 0.01).
As illustrated, survival of the patients with a TAPSE less than 1.8 was significantly shorter than patients with a TAPSE of 1.8 cm or greater (log-rank test, χ2 =10.00, p = 0.001) in the overall cohort (Figure 5A). The 1- and 2-yr survival estimates were 57% (95% confidence interval [CI], 37–73%) and 42% (95% CI, 24–60%), respectively, for patients with a TAPSE less than 1.8 compared with 85% (95% CI, 69–94%) and 79% (95% CI, 61–90%), respectively, for patients with a TAPSE of 1.8 cm or greater. For the PAH cohort, patients with a TAPSE less than 1.8 had a significantly shorter survival versus patients with a TAPSE of 1.8 or greater (log-rank test, χ2 = 6.81, p = 0.009; Figure 5B). Survival estimates for patients with PAH with a TAPSE less than 1.8 were 60% (95% CI, 40–75%) and 50% (95% CI, 31–66%) at 1 and 2 yr compared with 94% (95% CI, 65–99%) and 88% (95% CI, 61–97%) for patients with a TAPSE of 1.8 or greater. It was striking to note that 15 of the 16 patients with a TAPSE less than 1.8 cm who died in the overall cohort had PAH. In contrast, of the seven patients who died with a TAPSE of 1.8 cm or greater during follow-up, only two had PAH, whereas the remaining five had pulmonary hypertension associated with respiratory disease (four patients with interstitial lung disease, one patient with obstructive lung disease). Furthermore, mortality also differed across TAPSE tertiles in the PAH population, with the largest differences in outcome observed between patients with a TAPSE less than 1.5 cm and those with a TAPSE of 2.0 cm or greater (Figure 6).
Figure 5. Kaplan and Meier estimates of survival (primary endpoint, all-cause mortality) in patients stratified by TAPSE values. A represents the overall cohort; B represents the pulmonary arterial hypertension (PAH) population only. Numbers below each figure represent the number of patients at risk for death at each time point.
[More] [Minimize]
Figure 6. Bar graphs of crude mortality rate in patients with PAH across TAPSE tertiles. Numbers above bars indicate the absolute number of subjects dead/alive in each tertile.
[More] [Minimize]The unadjusted risk of death during the study period for patients in the overall cohort with a TAPSE less than 1.8 compared with those with a TAPSE of 1.8 cm or greater was 3.8 (95% CI, 1.6–9.3; p = 0.003). The risk of death was even greater for patients with PAH with a TAPSE less than 1.8 compared with those with a TAPSE of 1.8 or greater (unadjusted hazard ratio [HR], 5.7; 95% CI, 1.3–24.9; p = 0.02). When analyzed as a continuous variable in patients with PAH, for every 1-mm decrease in TAPSE, the unadjusted risk of death increased by 17% (HR, 1.17; 95% CI, 1.04–1.32; p = 0.006). Multivariable models using Cox proportional hazards were constructed that consisted of TAPSE as a continuous or dichotomous variable (</⩾ 1.8 cm), variables found to be significant in bivariable analyses (p < 0.20), and variables previously shown to have prognostic significance (Table 3). Although RAP was highly significant in bivariable analysis, and partially confounded the association between TAPSE and outcome on multivariable analyses, inclusion of RAP in the model did not add significantly to the predictive value of the overall model using global likelihood ratio testing, and thus was excluded.
Variable | Unadjusted HR (95% CI, p value) | Adjusted HR (95% CI, p value)* | |
|---|---|---|---|
| Age | 1.01 (0.98–1.05, p = 0.48) | N/A | |
| Baseline WHO | 6.8 (2.50–18.55, p < 0.001) | (C) | 11.60 (2.62–51.57, p < 0.01) |
| (D) | 9.29 (1.91–45.11, p < 0.01) | ||
| Baseline therapy (yes or no) | 0.53 (0.20–1.39, p = 0.20) | N/A | |
| Eccentricity index, diastole | 6.14 (1.03–36.48, p = 0.05) | (C) | 2.60 (0.20–34.01, p = 0.47) |
| (D) | 3.70 (0.40–34.33, p = 0.25) | ||
| Right atrial area index | 1.24 (1.12–1.38, p < 0.001) | † | |
| RV fractional area change | 0.92 (0.86–0.96, p = 0.02) | (C) | 0.94 (0.82–1.08, p = 0.39) |
| (D) | 0.94 (0.88–1.00, p = 0.05) | ||
| Tricuspid regurgitation | 2.59 (1.39–4.81, p = 0.003) | (C) | 1.68 (0.84–3.40, p = 0.15) |
| (D) | 1.24 (0.60–2.53, p = 0.56) | ||
| Pericardial effusion | 1.69 (1.12–2.54, p = 0.01) | (C) | 1.78 (1.13–2.82, p = 0.01) |
| (D) | 1.76 (1.11–2.78, p = 0.02) | ||
| TAPSE (continuous) | 1.17 (1.04–1.32, p < 0.01) | 1.16 | (1.03–1.28, p = 0.01) |
| TAPSE (dichotomous) | 5.69 (1.30–24.95, p = 0.01) | 7.03 | (1.49–33.18, p = 0.01) |
| Right atrial pressure | 1.22 (1.12–1.35, p < 0.001) | (C) | 1.12 (1.00–1.24, p = 0.04) |
| (D) | 1.10 (0.98–1.24, p = 0.09) | ||
| Mean PAP | 1.00 (0.96–1.04, p = 0.88) | † | |
| Cardiac index | 0.43 (0.18–1.02, p = 0.05) | † | |
| PVR | 1.07 (0.99–1.16, p = 0.10) | † | |
| PA oxygen saturation | 0.88 (0.83–0.94, p < 0.001) | † |
The most parsimonious model included TAPSE as a continuous variable, pericardial effusion, and baseline WHO functional class, yielding an adjusted risk of death increase of 16% for every millimeter decrease in TAPSE (HR, 1.16; 95% CI, 1.03–1.28; p = 0.01). Moreover, when stratified for treatment status at baseline, the HR was unchanged, suggesting that TAPSE predicted outcome regardless of whether patients were being treated for pulmonary hypertension at the time of enrollment.
Overall, the intraobserver (r = 0.96, p < 0.0001; mean bias, −0.03; 95% CI of agreement, −0.4 to 0.3) and interobserver reproducibility (r = 0.94, p < 0.0001; mean bias, 0.03 cm; 95% CI of agreement, −0.3 to 0.4) of TAPSE was excellent. This is in contrast with the intraobserver (r = 0.77, p = 0.008; mean bias, 2.4; 95% CI of agreement, −14 to 19%) and interobserver variability (r = 0.71, p = 0.003; mean bias, −4.5; 95% CI, −19 to 10) obtained for RVFAC.
In the present study, a TAPSE less than 1.8 cm identified the patients with pulmonary hypertension who had more advanced RV dysfunction, as compared with subjects with a TAPSE of 1.8 cm or greater. Patients with a TAPSE less than 1.8 cm had dramatically reduced survival over a median follow-up of 19 mo, a finding that was driven almost exclusively by the patients with a diagnosis of PAH. Patients with a TAPSE less than 1.5 cm had an especially poor outcome. The prognostic significance of TAPSE persisted after adjustment for several previously recognized echocardiographic and invasive predictors of outcome, as well as baseline treatment status. Thus, our results suggest that TAPSE is a robust measure of RV function and a powerful predictor of patient survival in pulmonary hypertension.
Rushmer and colleagues first demonstrated that RV shortening occurs primarily along its longitudinal axis or from base to apex, whereas the left ventricle shortens relatively symmetrically in the longitudinal and transverse axes (5). Investigators subsequently showed that echo-derived systolic displacement of the tricuspid annulus closely correlated with RV ejection fraction (r = 0.92) in a variety of non-PAH populations (6, 8, 15, 16). Moreover, investigators have shown that longitudinal RV shortening is depressed in children with idiopathic pulmonary hypertension as compared with age-matched control subjects and children with atrial septal defects (17).
In the current study, TAPSE was shown to be a highly sensitive and specific predictor of a depressed RV SVI, which parallels a recent study showing that longitudinal RV displacement similarly predicted a reduced RV SVI in patients with PAH (18). These findings support studies showing that TAPSE closely correlates with RV ejection fraction, which is the proportion of stroke volume to end-diastolic volume (6). We also observed striking differences between a variety of other invasive hemodynamic and echocardiographic measures of RV systolic function in patients with a TAPSE less than 1.8 cm or in those with a TAPSE of 1.8 cm or greater. Moreover, subjects with a TAPSE less than 1.8 cm had comparatively greater degrees of right heart remodeling and RV–LV disproportion than subjects with a TAPSE of 1.8 cm or greater, with a low TAPSE capturing a patient phenotype with more globally decompensated RV function manifesting as a triad of poor RV systolic function, right heart remodeling, and RV–LV disproportion. These same patients also displayed relative underfilling of the LV, both in terms of smaller LV dimensions and greater LV diastolic impairment as compared with those with preserved TAPSE, likely reflecting the in-series and interdependent effects of the failing right ventricle on the LV filling (19). It is also worth noting that TAPSE was inversely associated with the degree of tricuspid regurgitation, which parallels the association between poor RV function and greater TR, and argues against the theoretic possibility that decreased RV afterload in the setting of significant TR would lead to increases in tricuspid annular motion. Taken together, these results suggest that systolic (base to apex) displacement of the tricuspid annulus is an accurate reflection of RV function in patients with pulmonary hypertension.
Recognizing the strong association between measures of RV systolic function and afterload, it was logical that TAPSE correlated inversely with PVR, suggesting that TAPSE is afterload dependent. In keeping, recent data have shown a strong direct correlation (r = 0.75, p < 0.01) between an acute change in RV SVI and longitudinal RV shortening immediately after epoprostenol infusion (18). Likewise, we demonstrated that patients with a depressed TAPSE had significantly lower pulmonary artery compliance, which reflects afterload at the level of the proximal pulmonary arteries, and which has been shown to predict poor outcome in patients with PAH (20, 21). Future studies will be needed to determine if TAPSE can be followed serially to track RV responsiveness to chronic afterload-reducing therapies in PAH. In addition, it would be of interest to determine whether TAPSE is useful in detecting pulmonary hypertension, as others have shown that non-Doppler indices such as an increased ratio of RV diameter to tissue Doppler velocity of tricuspid annular motion are useful predictors of pulmonary hypertension (22).
In the current study, a TAPSE less than 1.8 cm was associated with a nearly fourfold increased risk of death in the overall population, and a nearly sixfold increased risk of death in the subgroup of patients with PAH. Importantly, although a TAPSE less than 1.8 cm was associated with an adverse prognosis in both the overall population and in patients with PAH, 15 of the 16 patients who died with a TAPSE less than 1.8 cm had PAH, indicating that the prognostic significance of a TAPSE less than 1.8 cm in the overall cohort was driven almost entirely by the patients with PAH. Conversely, only two patients with PAH with a TAPSE of 1.8 cm or greater died as compared with five with pulmonary hypertension associated with respiratory disease. Because patients with pulmonary hypertension associated with advanced lung disease are also subject to progressive respiratory failure, their poor outcome can be explained despite relatively normal RV function.
Importantly, the prognostic significance of TAPSE persisted across TAPSE tertiles, and when examined as a continuous variable, every 1-mm decrease in TAPSE conferred a 17% increase in risk of death in patients with PAH. This risk of death persisted after adjusting for echocardiographic and invasive hemodynamic variables previously established to be of prognostic importance in patients with PAH (2, 3). In light of these findings, future studies need to address the prognostic relevance of serial changes in TAPSE in the same patient, as well as determine if such changes can be used to alter treatment strategies. Likewise, it will be important to determine the precision with which changes in TAPSE can be identified, because very small changes in TAPSE may be difficult to appreciate given inherent limitations in echocardiographic resolution.
Overall, our results align with prior investigations showing that worse RV function, reflected as a depressed TAPSE, is an adverse prognostic indicator in patients with dilated cardiomyopathy, and in subjects after acute inferior myocardial infarction (7–9). However, the current results represent the first demonstration of the prognostic significance of TAPSE in patients with pulmonary hypertension. In contrast to the above studies, which reported cutoffs ranging from 1.3 to 1.5 cm, our study showed that the optimal TAPSE cutoff for predicting outcome was 1.8 cm. Although the tertile in the current study with a TAPSE less than 1.5 cm clearly had the worst outcome, we also observed a relatively high mortality rate in patients with more intermediate depression of TAPSE (1.5–1.9 cm). However, prior reports focused on patients with severe LV systolic dysfunction. This is an important caveat, as LV systolic function plays an important role in determining RV systolic function (so-called systolic ventricular interdependence), which may dictate the level at which TAPSE (and other measures of RV function) reflects RV dysfunction and thus becomes prognostically important (23).
Numerous studies have demonstrated the importance of RV function in predicting outcome in patients with PAH. However, most studies have relied on invasive hemodynamic indices of RV function; indirect measures of RV function, such as right atrial size, tricuspid regurgitation, and pericardial effusion; or more complex Doppler-derived methods, which, despite their validity, are not widely used in clinical practice (2, 3, 24–27). Thus, the finding from the current study that TAPSE imparts important prognostic information in patients with pulmonary hypertension is particularly relevant, because TAPSE is a direct measure of RV function that is easily obtained from a routine transthoracic echocardiographic examination, and does not require sophisticated post hoc analysis or expertise to measure. This combination of prognostic power and simplicity makes TAPSE especially applicable to clinical practice and facilitates its use equally in both community-based and tertiary care practice.
Consistent with prior reports, the measurement of TAPSE was highly reproducible in our hands, with excellent intraobserver and interobserver agreement (6, 7). Also consistent with prior reports, however, was our observation that RV FAC was far less reproducible, with intraobserver and interobserver reproducibility that was, arguably, unacceptably low (8). The variability of this measure likely reflects both its dependence on endocardial visualization to accurately obtain RV end-diastolic and end-systolic area, and the fact that this index is mathematically derived, which may further amplify variability.
In the current study, we chose RV SVI as the reference standard for ROC analysis with the specific intent of determining a rational cut point from which to further examine TAPSE in relation to numerous hemodynamic and echocardiographic measures of RV performance that have been validated in prior studies, and that are routinely used in the clinical assessment of patients with PAH. Therefore, the results from the current study allow the functional and prognostic significance of a depressed TAPSE to be more readily appreciated and therefore applied into clinical practice. We did not compare TAPSE against a volumetric standard of RV function, such as angiographic or magnetic resonance imaging–derived RV ejection fraction. However, the relationship of TAPSE to RV ejection fraction is well established in other patient populations, and perhaps more importantly, such a comparison is far less practical than our approach and not absolved of the limitations of RV chamber geometry and endocardial definition (28).
Although the vast majority (75%) of patients in our cohort had PAH (WHO classification group 1), patients with other forms of pulmonary hypertension, such as pulmonary hypertension associated with respiratory disease or hypoxemia, were included. This simply reflected the makeup of the population referred to our pulmonary hypertension clinic. As such, our population may better represent the population referred for pulmonary hypertension evaluation in clinical practice. Within the PAH cohort, there was heterogeneity that could have affected the survival and hazard analyses. About half of the PAH cohort were patients with established PAH currently receiving therapy; the others were newly diagnosed with PAH and were not receiving therapy at the time of RHC and echocardiography. Thus, it is possible that survival analyses were skewed by lead-time bias, because patients with established disease may have been more likely to die in the follow-up period than newly diagnosed patients. However, when stratified by whether or not patients were receiving therapy for PAH at baseline evaluation, survival estimates did not vary. In addition, the risk of death adjusted for treatment status at baseline remained significantly greater in patients with a lower TAPSE (adjusted HR, 1.16; 95% CI, 1.04–1.29; p = 0.01). Moreover, the current study was not designed to examine the effects of therapy on TAPSE and how such effects relate to outcome, which, as discussed above, will be an important future application of TAPSE. Our results may have also been subject to referral bias, which may have enriched our population with a group of patients at higher risk for poor outcomes. This raises the question as to whether the prognostic importance of TAPSE would hold among a broader, and perhaps less ill, population. Conversely, it is encouraging that TAPSE did prognosticate in a population such as ours.
Factors not included in these analyses (e.g., exercise capacity) or those factors that we have failed to identify may confound this relationship between TAPSE and mortality. Larger scale prospective studies will be needed to more directly address the prognostic ability of TAPSE within (e.g., idiopathic PAH vs. scleroderma-associated PAH) and across (e.g., idiopathic PAH vs. chronic thromboembolic pulmonary hypertension) the various categories of pulmonary hypertension, because the size of the current study population was not conducive to more definitive subgroup analyses. Finally, although the reproducibility of TAPSE was excellent, the experience and familiarity of the echocardiographic reviewers with the technique may have introduced bias.
The present study demonstrates that TAPSE is a simple, reproducible, yet robust measure of RV function in patients with pulmonary hypertension, which also has important prognostic implications for patients with PAH. Thus, TAPSE should be incorporated into the echocardiographic assessment of patients with PAH.
The authors thank Ellen G. Reather for expert manuscript preparation.
| 1. | Farber HW, Loscalzo J. Pulmonary arterial hypertension. N Engl J Med 2004;351:1655–1665. |
| 2. | D'Alonzo GE, Barst RJ, Ayres SM, Bergofsky EH, Brundage BH, Detre KM, Fishman AP, Goldring RM, Groves BM, Kernis JT. Survival in patients with primary pulmonary hypertension: results from a national prospective registry. Ann Intern Med 1991;115:343–349. |
| 3. | McLaughlin VV, Presberg KW, Doyle RL, Abman SH, McCrory DC, Fortin T, Ahearn G. Prognosis of pulmonary arterial hypertension: ACCP evidence-based clinical practice guidelines. Chest 2004;126:78S–92S. |
| 4. | Chin KM, Kim NH, Rubin LJ. The right ventricle in pulmonary hypertension. Coron Artery Dis 2005;16:13–18. |
| 5. | Rushmer RF, Crystal DK, Wagner C. The functional anatomy of ventricular contraction. Circ Res 1953;1:162–170. |
| 6. | Kaul S, Tei C, Hopkins JM, Shah PM. Assessment of right ventricular function using two-dimensional echocardiography. Am Heart J 1984; 107:526–531. |
| 7. | Karatasakis GT, Karagounis LA, Kalyvas PA, Manginas A, Athanassopoulos GD, Aggelakas SA, Cokkinos DV. Prognostic significance of echocardiographically estimated right ventricular shortening in advanced heart failure. Am J Cardiol 1998;82:329–334. |
| 8. | Ghio S, Recusani F, Klersy C, Sebastiani R, Laudisa ML, Campana C, Gavazzi A, Tavazzi L. Prognostic usefulness of the tricuspid annular plane systolic excursion in patients with congestive heart failure secondary to idiopathic or ischemic dilated cardiomyopathy. Am J Cardiol 2000;85:837–842. |
| 9. | Samad BA, Alam M, Jensen-Urstad K. Prognostic impact of right ventricular involvement as assessed by tricuspid annular motion in patients with acute myocardial infarction. Am J Cardiol 2002;90:778–781. |
| 10. | Forfia PR, Fisher M, Champion HC, Girgis RE, Hassoun PM. Tricuspid annular plane systolic excursion (TAPSE) is a simple noninvasive index of RV function and a powerful predictor of survival in pulmonary arterial hypertension [abstract]. Proc Am Thorac Soc 2006;3:A60. |
| 11. | Lange RA, Hillis LD. Cardiac catheterization and hemodynamic assessment. In: Topol EJ, editor. Textbook of cardiovascular medicine. Philadelphia: Lippincott-Raven; 1998. p. 1957–1976. |
| 12. | Cox DR. Regression models and life-tables. J R Stat Soc Ser B Stat Methodol 1972;34:187–220. |
| 13. | Raymond RJ, Hinderliter AL, Willis PW, Ralph D, Caldwell EJ, Williams W, Ettinger NA, Hill NS, Summer WR, de Boisblanc B, et al. Echocardiographic predictors of adverse outcomes in primary pulmonary hypertension. J Am Coll Cardiol 2002;39:1214–1219. |
| 14. | Galie N, Hinderliter AL, Torbicki A, Fourme T, Simonneau G, Pulido T, Espinola-Zavaleta N, Rocchi G, Manes A, Frantz R, et al. Effects of the oral endothelin-receptor antagonist bosentan on echocardiographic and doppler measures in patients with pulmonary arterial hypertension. J Am Coll Cardiol 2003;41:1380–1386. |
| 15. | Shah AR, Grodman R, Salazar MF, Rehman NU, Coppola J, Braff R. Assessment of acute right ventricular dysfunction induced by right coronary artery occlusion using echocardiographic atrioventricular plane displacement. Echocardiography 2000;17:513–519. |
| 16. | Hebert JL, Chemla D, Gerard O, Zamani K, Quillard J, Azarine A, Frank R, Lecarpentier Y, Fontaine G. Angiographic right and left ventricular function in arrhythmogenic right ventricular dysplasia. Am J Cardiol 2004;93:728–733. |
| 17. | Arce OX, Knudson OA, Ellison MC, Baselga P, Ivy DD, Degroff C, Valdes-Cruz L. Longitudinal motion of the atrioventricular annuli in children: reference values, growth related changes, and effects of right ventricular volume and pressure overload. J Am Soc Echocardiogr 2002;15:906–916. |
| 18. | Urheim S, Cauduro S, Frantz R, McGoon M, Belohlavek M, Green T, Miller F, Bailey K, Seward J, Tajik J, et al. Relation of tissue displacement and strain to invasively determined right ventricular stroke volume. Am J Cardiol 2005;96:1173–1178. |
| 19. | Gan TJ, Lankhaar JW, Marcus TJ, Westerhof N, Marques KM, Bronzwaer JG, Boonstra A, Postmus PE, Vonk-Noordegraaf A. Impaired left ventricular filling due to right to left ventricular interaction in patients with pulmonary arterial hypertension. Am J Physiol Heart Circ Physiol 2006;290:H1528–H1533. |
| 20. | Dyer K, Lanning C, Das B, Lee PF, Ivy DD, Valdes-Cruz L, Shandas R. Noninvasive Doppler tissue measurement of pulmonary artery compliance in children with pulmonary hypertension. J Am Soc Echocardiogr 2006;19:403–412. |
| 21. | Mahapatra S, Nishimura RA, Sorajja P, Cha S, McGoon MD. Relationship of pulmonary arterial capacitance and mortality in idiopathic pulmonary arterial hypertension. J Am Coll Cardiol 2006;47:799–803. |
| 22. | McLean AS, Ting I, Huang SJ, Wesley S. The use of the right ventricular diameter and tricuspid annular tissue Doppler velocity parameter to predict the presence of pulmonary hypertension. Eur J Echocardiog [E-pub ahead of print] Apr 29, 2006. |
| 23. | Santamore WP, Dell'Italia LJ. Ventricular interdependence: significant left ventricular contributions to right ventricular systolic function. Prog Cardiovasc Dis 1998;40:289–308. |
| 24. | McLaughlin VV, Shillington A, Rich S. Survival in primary pulmonary hypertension: the impact of epoprostenol therapy. Circulation 2002; 106:1477–1482. |
| 25. | Tei C, Dujardin KS, Hodge DO, Bailey KR, McGoon MD, Tajik AJ, Seward SB. Doppler echocardiographic index for assessment of global right ventricular function. J Am Soc Echocardiogr 1996;9:838–847. |
| 26. | Yeo TC, Dujardin KS, Tei C, Mahoney DW, McGoon MD, Seward JB. Value of a Doppler-derived index combining systolic and diastolic time intervals in predicting outcome in primary pulmonary hypertension. Am J Cardiol 1998;81:1157–1161. |
| 27. | Dyer KL, Pauliks LB, Das B, Shandas R, Ivy D, Shaffer EM, Valdes-Cruz LM. Use of myocardial performance index in pediatric patients with idiopathic pulmonary arterial hypertension. J Am Soc Echocardiogr 2006;19:21–27. |
| 28. | Pattynama PM, Lamb HJ, Van der Velde EA, Van der Geest RJ, Van der Wall EE, de Roos A. Reproducibility of MRI-derived measurements of right ventricular volumes and myocardial mass. Magn Reson Imaging 1995;13:53–63. |
