© 2014, Wiley Periodicals, Inc. DOI: 10.1111/echo.12687
Two-Dimensional Measurement of Tricuspid Annular Plane Systolic Excursion in Children: Can It Substitute for an M-Mode Assessment? Muhammad Yasir Qureshi, M.B.B.S.,* Benjamin W. Eidem, M.D.,*† Chelsea L. Reece, R.D.C.S.,† and Patrick W. O’Leary, M.D.*† *Division of Pediatric Cardiology, Mayo Clinic, Rochester, Minnesota; and †Division of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota
Background: Tricuspid annular plane systolic excursion measured by M-mode (MM-TAPSE) has been validated as a marker of right ventricular systolic performance. A similar measurement by 2D imaging (2D-TAPSE) can be obtained. We sought to determine the correlation and strength of agreement between MM-TAPSE and 2D-TAPSE in children. Methods: Echocardiographic studies performed for clinical indications were reviewed retrospectively. All consecutive subjects ≤18 years of age were included. The cohort was divided into those with normal echocardiographic ﬁndings and those with disorders affecting the right ventricle. Digitally recorded images were analyzed for both MM-TAPSE and 2DTAPSE. Measurements of 2D-TAPSE were made in an apical four-chamber view, from the tricuspid valve annulus to a consistent point at the apex of the imaging sector at end-diastole and end-systole, with the difference representing the 2D-TAPSE value. Results: A total of 329 subjects (mean age 9.0 6.1 years) met entry criteria. Correlation coefﬁcient between MM-TAPSE and 2D-TAPSE was 0.90. Bland–Altman analysis showed agreement between the two methods to be within 1.2 2.6 mm (mean percentage difference of 6.5%). About 1 mm difference between MM-TAPSE and 2D-TAPSE was consistently observed in all diagnostic subgroups, and across all age categories. Conclusion: MM-TAPSE and 2D-TAPSE correlate strongly, with 2D-TAPSE being consistently about 1 mm less than values obtained by the M-mode technique. We conclude that 2D-TAPSE can provide a reliable alternative to MM-TAPSE to quantitatively measure right ventricular systolic function and may be especially useful in situations where retrospective comparisons are sought. (Echocardiography 2015;32:528–534) Key words: TAPSE, pediatric echocardiography, right ventricular function Echocardiographic assessment of right ventricular function can be challenging due to the unique morphology of that chamber. The evaluation of right ventricular systolic function by tricuspid annular plane systolic excursion (TAPSE) has been validated in several pediatric1,2 and adult3,4 studies. It has been shown to correlate with other quantitative assessments of right ventricular systolic function measured by 2D echocardiography,5 radionuclide angiography6 and magnetic resonance imaging7,8 in adults. The prognostic value of TAPSE has also been established in pulmonary hypertension,1,9 congestive heart failure,10 pulmonary embolism,11 and early detection of right ventricular dysfunction.12 As a result, TAPSE is widely used for the assessment of longitudinal right ventricular systolic performance owing to its feasibility and reproducibilAddress for correspondence and reprint requests: Muhammad Yasir Qureshi, M.B.B.S., Mayo Clinic, 200 1st St. SW, Rochester, Minnesota 55905. Fax: (507) 284-0160; E-mail: [email protected]
ity.5,13,14 In fact, the American Society of Echocardiography identiﬁes TAPSE as one of the tools for functional assessment of the right ventricle in children.15,16 The standard measurement of TAPSE is made by using M-mode echocardiography, with the M-line passing through the lateral annulus of the tricuspid valve in the apical four-chamber orientation. The excursion of the lateral annulus is then measured from the M-mode tracing. It may also be useful to assess TAPSE when an annular M-mode tracing is not available. This situation can be encountered when one wishes to make a comparison of a current study with an older echocardiographic examination of the same patient or in the setting of retrospective research studies where right ventricular function needs to be assessed in a serial quantitative manner. In these instances, an M-mode tracing of the tricuspid valve lateral annulus may not have been originally acquired. As an alternative, a similar measurement of the excursion of tricuspid valve
2D Measurement of TAPSE
lateral annulus can be made by using 2D images (2D-TAPSE), because an apical four-chamber view has almost always been recorded during a standard echocardiographic examination. Some studies have used a similar measurement of TAPSE by magnetic resonance imaging17 and speckle tracking echocardiography.18 However, the strength of agreement between MM-TAPSE and 2D-TAPSE by echocardiography has not been established in the literature. The objective of this study was to evaluate the correlation and strength of agreement between MM-TAPSE and 2D-TAPSE in children with normal and abnormal cardiac anatomy and right ventricular systolic function. Methods: Study Participants: Approval from the Institutional Review Board of Mayo Clinic, Rochester was obtained. Subjects with either normal ﬁndings on transthoracic echocardiogram or diagnoses impacting the right ventricle were identiﬁed by review of our clinical echocardiographic database. All consecutive subjects of ≤18 years of age with echocardiograms that included assessments of MM-TAPSE were included. The diagnoses impacting the right ventricle included repaired tetralogy of Fallot, repaired pulmonary atresia with or without ventricular septal defect, repaired pulmonary stenosis, pulmonary hypertension, and unrepaired atrial septal defect. These diagnoses were chosen to create a variety of right ventricular pressureand volume-overloading conditions in comparison to those with normal echocardiograms. The indications for echocardiography in the subjects with normal ﬁndings were heart murmur, chest pain, presyncope, syncope or a screening examination due to a relative with congenital heart disease or channelopathy. The electronic medical record and digital echocardiographic studies were reviewed retrospectively. Only the studies which had an M-mode tracing at the tricuspid valve lateral annulus and adequate apical fourchamber image quality for analysis were included. Exclusion based on inadequate image quality occurred for one or more of the following reasons: (1) less than three beats acquired during the M-mode tracing or the apical four-chamber view; (2) too much patient movement during acquired images; or (3) inadequate visualization of right ventricle in apical four-chamber view. Echocardiographic Measurements: The echocardiographic measurements were performed off line on digitally recorded images by the same observer. TAPSE in every study was measured by M-mode and 2D techniques on separate occasions to maintain blinding. The
average of 3 cardiac cycles was used for all measurements. For MM-TAPSE, the M-mode tracing with the M-line at the lateral tricuspid valve annulus in apical four-chamber view was used. The MM-TAPSE was calculated by the difference between the distance from a consistent point as close to the apex of the image as possible (a surrogate for the skin-transducer interface or zero mark on the scale) to the lateral tricuspid valve annulus at end-systole and end-diastole (Fig. 1). For 2D-TAPSE, the measurements were made in an apical four-chamber view. Similar to MMTAPSE, the distance from the lateral tricuspid valve annulus to the zero mark on the scale (closest to the apex of the sector image, representing transducer-skin interface) was measured at enddiastole and end-systole, with the difference representing the 2D-TAPSE (Fig. 2). The timing of end-diastole and end-systole for both methods was determined by the maximum and minimum distance between the lateral annulus and the zero mark, respectively. The zero mark on the scale was used as the reference point instead of the right ventricular apex for better reproducibility. The right ventricular apex is challenging to be pinpointed consistently, whereas the zero mark on the scale is an easily deﬁned reference point. Also for MM-TAPSE, the TV excursion is calculated in reference to the zero mark on the scale and the distance from RV apex is not regarded. Inter-observer and intra-observer variabilities were assessed on 50 consecutive subjects. A different reviewer measured MM-TAPSE and 2DTAPSE on separate occasions, while maintaining
Figure 1. Measurement of tricuspid valve annular plane systolic excursion (TAPSE) by M-mode echocardiography (MMTAPSE). Distance was measured from tricuspid valve lateral annulus to the skin-transducer interface in systole and diastole and was averaged over three cardiac cycles. The MM-TAPSE was calculated by subtracting systolic distance from the diastolic distance which is mathematical equivalent of the clinically measured TAPSE demonstrated as the green arrow.
Qureshi, et al.
Figure 2. Measurement of tricuspid valve annular plane systolic excursion by 2D-echocardiography (2D-TAPSE). Distance was measured from tricuspid valve lateral annulus to the skin-transducer interface in A. diastole and B. systole and was averaged over three cardiac cycles. The MM-TAPSE was calculated by subtracting systolic distance from the diastolic distance.
blinding. The variability (mean percentage error) was calculated as the absolute difference between the 2 sets of measurements divided by the overall mean of the observed values by the 2 methods. These 50 subjects were also used to compare our technique of determining TAPSE to the standard method in which the calipers are placed directly on the closest and farthest points on the M-mode tracing of tricuspid valve annulus.16 Statistical Analysis: Statistical analyses were performed using MedCalc statistical software (version 18.104.22.168; MedCalc Software, Inc., Mariakerke, Belgium). Descriptive data are presented as mean standard deviation; median with range; or percentage with 95% conﬁdence interval. Correlation and regression analyses as well as Bland–Altman plot were used to assess correlation and agreement between the two methods for TAPSE measurement. Further subgroup analysis was performed based on age, degree of right ventricular dysfunction, and cardiac pathology. For age subgroups, subjects were grouped as