In dogs, pulmonary hypertension (PH) is associated with various conditions, such as idiopathic pulmonary arterial hypertension, pulmonary diseases, heartworm disease, thromboembolic disease, and cardiac disease.[1-10] Whether it is precapillary (pulmonary arterial hypertension) or postcapillary (pulmonary venous hypertension), PH may impair right ventricular performance as it progresses in severity.[11, 12]
Echocardiography is currently the best noninvasive tool to evaluate cardiac structure and function in the veterinary clinical setting. Although methods are well established for the left cardiac chambers, assessing the function of the right ventricle (RV) is technically challenging because of its complex geometry and contractile properties. The RV is triangular in shape and its cavity is divided into inflow and outflow tracts that cannot be imaged simultaneously. Moreover, experimental studies in dogs demonstrated that longitudinal displacement of the base of the ventricle towards the apex is the major contributor to RV contraction. Therefore, echocardiographic measurements and algorithms that are extensively used to quantify left ventricular performance in clinical practice cannot be applied to the RV.[14, 15]
Tricuspid annular plane systolic excursion (TAPSE) is an echocardiographic parameter that has been developed in people to evaluate right ventricular systolic function.[15, 16] It measures the apical displacement of the lateral portion of the tricuspid valve annulus during systole from an M-mode recording. In human adults, TAPSE has been validated against invasive methods as a marker of RV systolic function, with normal values reported between 1.5 and 2 cm, and it has prognostic value regardless of heart rhythm or rate.[15-21] In patients with PH, TAPSE values below the reference range identify those with decreased right ventricular systolic function and an associated higher mortality risk.
The use of TAPSE has not been evaluated in normal dogs or dogs with PH, and a reference range has not been established. Because few methods designed to assess right ventricular performance are applicable to clinical practice, evidence of right ventricular systolic dysfunction typically is inferred from the presence of right-sided congestive heart failure (eg, ascites). Thus, an important application of TAPSE measurement would be the early identification of right-sided cardiac systolic dysfunction in the presence of PH.
The 1st aim of the present study was to generate empirically derived reference limits for TAPSE from a sample population of healthy dogs and determine the intra and interobserver variability of this echocardiographic parameter. Our 2nd objective was to evaluate the effects of PH on TAPSE, based on the hypothesis that right ventricular systolic function and TAPSE values decrease as the severity of PH progresses.
- Top of page
- Materials and Methods
- Supporting Information
This study showed that TAPSE measurement is easily obtainable with a standard echocardiography system, and has adequate intra- and interobserver agreement. When applied to dogs with PH, a disease that can alter right ventricular function, TAPSE values were significantly decreased compared with the reference range established from a sample of healthy dogs. Moreover, TAPSE was typically below the lower limit of this interval in dogs included in the severe PH group.
Alteration in right ventricular function has been identified as an independent negative prognostic indicator in numerous studies evaluating the outcome of human patients with heart failure secondary to dilated or ischemic cardiomyopathy. Conversely, little information is available on dogs because the assessment of RV systolic function has been challenging. Among the noninvasive methods that have been validated in dogs, gated radionuclide ventriculography has been shown to accurately assess RV ejection fraction and RV filling and emptying rates when compared with cardiac magnetic resonance imaging, but this technique is expensive and its availability limited. As for echocardiography, the challenges for RV performance assessment are linked to the inherently complex RV geometry, a heavily trabeculated inner surface with poor endocardial definition, anatomically separate inflow and outflow which can only be seen from separate views, and the load dependency, in combination with the lower accuracy of most conventional echocardiographic parameters in comparison with invasive measures. Tissue Doppler imaging (TDI)-derived right ventricular myocardial motion has been shown to be a satisfactory method to assess the systolic function of the basal segments of the RV, but these measurements are more difficult to obtain than TAPSE, and the technology is only available on a limited number of ultrasound systems.[29, 30] Right ventricular Tei index, an index of myocardial performance, has been evaluated in healthy dogs and in dogs with right heart disease, but the need for 2 different echocardiographic views and therefore 2 different cardiac cycles increases the variability of the measurement.[31, 32] In addition, the use of Tei index is limited by the absence of the isovolumic periods in the normal RV as well as the pseudonormalization of the index when RA pressure is increased. Indeed, increased RA pressure determines a shortening of the isovolumic relaxation time, which will result in a decreased value of the Tei index.
Conversely, TAPSE is easily obtainable. In the present study, there was adequate agreement among observers (Table 2). This finding is comparable to the results of human studies, in which TAPSE was highly repeatable and reproducible.[16, 34, 35] In human adults, most studies report that a TAPSE value >1.5 cm is suggestive of normal right ventricular systolic function.[15, 16, 36] This use of a single threshold value of TAPSE to distinguish between normal and decreased right ventricular function is in contrast with current recommendations that favor the use of standard deviation or percentile-derived reference ranges. In addition, TAPSE is a linear measurement, the value of which increases with body dimensions. Therefore, the use of a single reference range applicable to all dogs is not possible. In this study, the use of quantile regression to determine the limits of the reference range allowed modeling the relation of specific percentiles (2.5th and 97.5th percentiles) of the response variable TAPSE with body weight, rather than simply estimating the variations of the mean TAPSE value. This method takes into account that some percentiles of TAPSE may be more affected by changes in body weight than other percentiles. Therefore, it defines with more precision the lower and upper limits of the reference interval and also is more robust against outliers, in particular considering the small size of the reference group.
TAPSE measures the displacement of the valve annulus from its end-diastolic position toward the cardiac apex in systole. Similar to longitudinal right ventricular myocardial velocities measured by TDI in dogs and people, the amplitude of tricuspid annulus motion along the long axis of the heart has been shown to be an acceptable index of systolic function.[30, 39, 40] It can indeed be correlated with the orientation of the deep muscle layer of the RV wall that is predominantly parallel to the longitudinal axis of the heart. More importantly, TAPSE, as a marker of systolic function, has been identified as a relevant tool in the prognostic stratification of human patients with left and right heart failure.[17, 20, 22, 26]
Compared with the left ventricle, the adaptation of the RV to increased afterload, as seen with PH, is poor. In the event of an abrupt increase in pulmonary vascular resistance, a linear decrease in stroke volume is observed as pulmonary pressure increases. Right ventricular hypertrophy only occurs with a more progressive increase in afterload, but usually is rapidly followed by a dilatation of the ventricle and ultimately failure. In people with adult-onset PH, TAPSE was a highly sensitive and specific indicator of decreased right ventricular function, as measured by the stroke volume index. In the same study, lower TAPSE values were associated with higher pulmonary vascular resistance, increased right atrial size and smaller dimensions of the left cardiac chambers. Moreover, nonsurvivors had the lowest TAPSE values. These results support the hypothesis of the current study that the decrease in TAPSE values as PH severity increases, and more importantly when signs of right-sided congestive heart failure are present is a marker of decreased RV systolic function. Based on the dogs evaluated in this study, it is noteworthy that only animals included in the severe PH group had TAPSE values below the reference limits, whereas, based on TAPSE only, the mild and moderate PH groups were indistinguishable from the healthy group. TAPSE values below the lower limit of the reference range in the severe PH group suggest a decrease in right ventricular systolic performance. However, systolic function was not directly measured because of the invasiveness, cost, and lack of availability of these techniques. Therefore, it cannot be ascertained that right ventricular systolic dysfunction was the only contributor to low TAPSE values in animals with high peak tricuspid regurgitation jet velocity. Indeed, there are examples of normal RV systolic function in the presence of severe PH, such as occurs in some human patients with Eisenmenger's physiology. However, this situation does not seem to be the case when PH is acquired.
A 2nd limitation of this study is that peak systolic tricuspid regurgitation jet velocity measured by Doppler echocardiography was used to estimate RV-to-RA PG using the simplified Bernoulli equation and subsequently define 3 PH groups. Although echocardiography is recognized as a reliable method to estimate pulmonary artery pressure, cardiac catheterization is preferable but rarely used in veterinary clinical practice.[42, 43] Underestimation of Doppler-derived RV-to-RA PG may result from a weak tricuspid jet signal or an excessive angle between the ultrasound beam and the direction of the flow. In addition, the simplified Bernoulli equation does not take into account the inertial component that is included in the complete equation. Estimated RA pressure was not used in this study to predict systolic pulmonary pressure, because methods to approximate RA pressures from echocardiographic data have not been validated in dogs.
A 3rd limitation is the heterogeneity of the population studied. Most dogs with PH had atrioventricular valve endocardiosis or heartworm disease, but other diseases represented included dilated cardiomyopathy in 1 dog and respiratory disease. The variety of diseases and their chronicity might have impacted the systolic function of the right heart differently. For example, whereas atrioventricular valve endocardiosis and dilated cardiomyopathy are mostly considered left-sided heart diseases, some animals could have had a reduction in right ventricular myocardial contractility. Moreover, because of the physiologic mechanism of ventricular interdependence, TAPSE values may be further decreased when left and right ventricular dysfunction is present concomitantly. In people, TAPSE values may be decreased in the setting of isolated left ventricular systolic dysfunction. Therefore, additional studies are needed in dogs to determine the exact contribution of left heart and right heart systolic dysfunction to TAPSE.
The importance of M-mode cursor alignment with the longitudinal motion of the tricuspid valve annulus also should be emphasized when TAPSE is measured. Excessive angulation of the cursor with respect to the displacement of the ventricular base may result in substantial underestimation of TAPSE.
Finally, the reference interval for TAPSE was established from a small population of healthy dogs. However, this number is comparable to most studies of dogs previously designed to establish normal echocardiographic dimensional data, which typically included an average of 30 dogs. In addition, the reference range obtained from this group of healthy dogs may not be suitable for accurate quantification of TAPSE in animals with a body weight <3 kg and >42 kg because dogs within these ranges of weight were not included in the reference group. Extrapolation of the lines to quantify TAPSE in smaller and larger dogs may be possible, but is hypothetical. Lastly, it cannot be ruled out that some dogs in the reference group had mild PH if measurable tricuspid regurgitation was absent.
In conclusion, this study provides information on the normal values of TAPSE in dogs for quantification in addition to qualitative evaluation of right ventricular systolic dysfunction in clinical practice (a supplemental online TAPSE reference range chart is provided to plot individual TAPSE measurements). It also describes the effects of PH, as estimated by peak systolic tricuspid regurgitation velocity, on TAPSE values. Additional studies should be performed to confirm the use of TAPSE as a marker of right ventricular systolic function, to test this parameter against already established methods, and to determine its role as a prognostic indicator in dogs with PH and right-sided heart disease.