Tricuspid Annular Pulmonary Systolic Excursion (TAPSE)

TAPSE is a quantitative measurement of the movement of the tricuspid annulus during the cardiac cycle. Use the phase array cardiac probe and change the exam to cardiac mode. You use M-mode (M-mode is used to measure the movement of a structure along one plane over time) in the cardiac apical four-chamber view to measure the distance the tricuspid annulus moves from end-diastole to end-systole. If you are having difficulty obtaining an optimal apical four-chamber view, ask the patient to lie in the left lateral decubitus position. Having the patient lie on their left side will bring the heart closer to the chest wall and the apex closer to the probe. The recording will be in the shape of a sinusoidal wave simulating the movement of the tricuspid annulus. You measure from peak to trough of the curves using the same colored line to obtain a TAPSE value. 

How to obtain TAPSE. image courtesy of thepocusatlas.com

There are multiple measurements that assess right ventricular strain, such as RV dilatation, pulmonary artery systolic pressure and RV ejection fraction, however given the complex structure and contraction pattern of the uniquely-shaped right ventricle, it may be difficult to reliably obtain these measurements.To understand right heart strain, please see this post on EM docs or this post on the ACEP ultrasound blog. As emergency physicians we are not trained to assess for the above advanced measurements, however the literature has shown that we can measure a TAPSE as a reliable surrogate. 

Lobo et al. conducted an a priori study based on the prospective, multicenter observational PROTECT study with 782 patients, who were diagnosed with acute PE on CT pulmonary angiogram (CTPA).[4] Every patient underwent a transthoracic echocardiogram evaluating for signs of right heart dysfunction. They showed that TAPSE may be a reliable surrogate for other signs of RV strain. For example, when TAPSE < 1.6 cm, there was a direct correlation with other signs of RV dysfunction such as increases in RV free wall hypokinesis (McConnell’s Sign), RV end-diastolic diameter, RV/LV end-diastolic diameter ratio, and elevated pulmonary artery systolic pressures. Even in the setting of normal BNP and troponin, an abnormal TAPSE in acute PE is an echo sign of RV strain. Therefore, an abnormal TAPSE may place the patient in the submassive PE classification.[4] However, a normal TAPSE alone does not mean there is no RV strain and does not mean the patient is safe for discharge. You have to take the entire clinical picture into consideration, including validated prognostication scores. 

Bottom Line: TAPSE < 1.6 cm in acute PE is a surrogate for RV strain and may be a sign of submassive pulmonary embolism despite normal ECG and laboratory findings 

Pitfall: Keep in mind that abnormal TAPSE measurement may be seen in other disease processes that cause RV strain, such as chronic pulmonary hypertension, RV ischemia, and congestive heart failure.

Lobo et al. also evaluated all-cause mortality as their primary outcome and death secondary to PE as their secondary outcome. [4] Among the 782 patients with acute PE, 35 patients died. 8% of patients with TAPSE < 1.6 cm died, compared to 3% in patients with TAPSE > 1.6 cm. After multivariate regression that controlled for history of cancer, TAPSE value, troponin level, and systolic blood pressure < 120 mm Hg, only TAPSE < 1.6 cm was a significant independent predictor of all-cause 30-day mortality and PE-related death. 

Pulmonary Embolism Severity Index (PESI) and sPESI are the validated prognostic tools in acute PEs to determine which PE patients are low-risk. The PESI or sPESI scores should be used to determine low-risk PEs, however, TAPSE may be used as an adjunct. For example, according to the 2019 European Society of Cardiology guidelines, evidence of echo RV dysfunction should be used in your disposition decision.[6] Based on moderate level of evidence and expert opinion, no evidence of RV dysfunction on TTE or CTPA, in addition to low risk on prognostication scores and good access to medical care, is needed for discharge. 

Disposition Algorithm for PE patients. Source: ESC Guidelines [6]

Pitfalls: TAPSE is only a one plane measurement in longitudinal view. Therefore, keep in mind RV dysfunction may still be present with normal TAPSE.

Daley et al. conducted a prospective, observational study (n=150) in a single, urban center assessing the accuracy of TAPSE in diagnosing PE. This study included patients greater than 18 years old and undergoing CTA to evaluate for PE.[7] Out of 150 patients, 32 patients (21.3%) had PE and 2 patients (1.4%) had massive PE. Every patient underwent a focused cardiac exam, evaluating TAPSE, septal flattening, tricuspid regurgitation, RV enlargements, and McConnel’s sign. 

In this study, TAPSE less than 1.7 cm was 56% sensitive and 79% specific for PE (LR+ 2.6, LR- 0.56) in normotensive patients using CTA PE as the reference standard. On further assessment, TAPSE less than 2.0 cm was 72% sensitive and only 66% specific (LR+ 2.1 , LR- 0.43). 

Of note, TAPSE was the most sensitive for diagnosis of PE when compared to other signs of right heart strain. However, McConnell’s sign (apical bowing) was the most specific (97%), followed by septal flattening (D-sign) at 94%. On post-hoc analysis, TAPSE was 94% sensitive in hypotensive and tachycardic patients with PE.[7] 

TAPSE’s high sensitivity in tachycardia is also suggested by another study assessing reliability in a focused cardiac ultrasound in patients with abnormal vital signs.[8] TAPSE less than 2.0 cm was the most sensitive (88%) component of focused cardiac exam in patients with acute PE and heart rate (HR) > 100/min. TAPSE sensitivity (93%) increased in patients with acute PE and HR > 110/min. 

See table below showing how TAPSE compared to other signs of RV dysfunction in the setting of tachycardia. 

Note: This study decided to use a TAPSE cut-off value of 2.0 cm to increase their sensitivity, however most studies use a cut-off value of 1.6 cm.

TAPSE may neither be sensitive nor specific to rule in/rule out a PE prior to CTA PE in normotensive patients, however, it may be used to determine your pretest probability. This is especially true in patients with tachycardia as an abnormal TAPSE value becomes more sensitive. TAPSE may be used to increase your pretest probability and aid in your decision to anticoagulate empirically in the setting of delayed or unavailable definitive imaging. 

Of note, although the above two studies are some of the best TAPSE studies to date, they have many limitations and flaws. For instance, they combined the sensitivities of HR > 100 and/or systolic blood pressure < 90 mm Hg, but did not elaborate on the statistical difference between the two vital signs. In addition, according to the papers, the principal investigators did not always agree with the measurements obtained by the ultrasonographer, but still included the bedside interpretations since TAPSE was measured at the bedside. There were no video recordings of how the measurements were obtained. Lastly, they do not elaborate on the 30 scans that they did not review. A significant number of these exams may have been performed by medical students. Given the low quality of the current evidence, further high-quality studies are needed to assess the use of TAPSE. 

Bottom line: TAPSE is neither sensitive nor specific enough to rule in/rule out a PE in normotensive patients; however sensitivity increases in patients with abnormal vital signs. 

Pitfall: TAPSE value can vary based on the angle of M-mode at the tricuspid annulus. One should aim to have the M-Mode line as perpendicular to the annulus as possible. 

Pitfall: Similar to any ultrasound measurement, TAPSE measurement is operator dependent. It is important to take the average of three separate measurements.