Aortic annular plane systolic excursion in cats with hypertrophic cardiomyopathy

Abstract Background Impairment of left ventricular (LV) longitudinal function is an early marker of systolic dysfunction in hypertrophic cardiomyopathy (HCM). Aortic annular plane systolic excursion (AAPSE) is a measure of LV longitudinal function in people that has not been evaluated in cats. Hypothesis Aortic annular plane systolic excursion is lower in cats with HCM compared to control cats, and cats in stage C have the lowest AAPSE. Animals One hundred seventy‐five cats: 60 normal, 61 HCM stage B and 54 HCM stage C cats. Materials Multicenter retrospective case‐control study. Electronic medical records from 4 referral hospitals were reviewed for cats diagnosed with HCM and normal cats. HCM was defined as LV wall thickness ≥6 mm and normal cats ≤5 mm. M‐mode bisecting the aorta in right parasternal short‐axis view was used to measure AAPSE. Results Aortic annular plane systolic excursion was lower in HCM cats compared to normal cats (3.9 ± 0.9 mm versus 4.6 ± 0.9 mm, P < .001) and was lowest in HCM stage C (2.4 ± 0.6 mm, P < .001). An AAPSE <2.9 mm gave a sensitivity of 83% (95% CI 71%‐91%) and specificity of 92% (95% CI 82%‐97%) to differentiate HCM stage C from stage B. AAPSE correlated with mitral annular plane systolic excursion (r = .6 [.4‐.7], P < .001), and atrial fractional shortening (r = .6 [.5‐.7], P < .001), but showed no correlation with LV fractional shortening. Conclusions and Clinical Importance Aortic annular plane systolic excursion is an easily acquired echocardiographic variable and might be a new measurement of LV systolic performance in cats with HCM.


| INTRODUCTION
[4][5][6][7] Systolic function of the LV is determined by a complex arrangement of myofibers which contribute to longitudinal LV shortening, circumferential and radial shortening and an axial twist. 8In people with HCM, reductions in LV longitudinal function, assessed by global longitudinal strain (GLS), occur before documentation of a reduced ejection fraction, 9,10 and this is documented in cats. 11Strain imaging requires high echocardiographic quality along with specialist software, both of which are frequently unavailable to the small animal practitioner.Some of these difficulties have been overcome by assessing mitral annular plane systolic excursion (MAPSE), which is a marker of LV longitudinal systolic function and correlates with GLS in people. 12,13ts with HCM with a reduced MAPSE are more likely to be in congestive heart failure (CHF) and have a worse prognosis. 2,4,6But this measurement requires a left apical view, which can be challenging to acquire especially in unstable cats.[16] Currently there are no published reference intervals for AAPSE in cats and no reports on the diagnostic use of AAPSE in cats with HCM.
We hypothesized that AAPSE would be lower in cats with HCM compared to control cats, and cats in CHF (stage C) would have the lowest AAPSE measurements.We aimed to provide AAPSE values for normal cats and determine if cats with HCM stage B have a lower AAPSE than normal control cats, and if AAPSE is associated with CHF.

| Animals
Multicenter retrospective case-control study.The electronic medical records of client-owned cats from 4 specialist referral hospitals from November 2019 to May 2022 were reviewed for cats diagnosed with HCM and normal cats.Cases were included in the study if they had a complete case record (signalment, history, physical examination and current medications), diagnostic echocardiographic examination including a right-parasternal short-axis view (RPSA) at the level of the aortic valve with M-mode of sufficient quality to measure AAPSE or a RPSA of sufficient quality to perform anatomical M-mode assessment.
Exclusion criteria for this study included systemic diseases known to affect the cardiovascular system or mimic HCM (phenocopies), such as systemic hypertension (systolic blood pressure > 180 mm Hg), hyperthyroidism (all cats >6 years of age had T4 measured), dehydration, diabetes mellitus, anemia (packed cell volume of <20%), neoplasia, and arrhythmias, congenital heart disease and cardiomyopathies other than HCM.Additionally, cases with incomplete clinical records or echocardiography examination were excluded.
Cats sedated for echocardiography with butorphanol alone or in combination with alfaxalone were not excluded, as these sedatives have been reported to have little to no effect on echocardiographic variables. 17 Cats with HCM were considered as stage B if they were not receiving diuretics and did not have a history of tachypnea, dyspnea, syncope or signs of arterial thromboembolism.Cats could be included as stage B if they were on any other medication except for diuretics.
Cats with HCM were in the stage C (CHF) group if they had increased respiratory rate and/or effort and either ultrasonographic or radiographic evidence of pulmonary edema or pleural effusion.Echocardiography could have been performed either before the administration of diuretics or when the cats had recently been diagnosed with CHF (within 24 hours of echocardiography).
All echocardiographic exams were performed and measured by board-certified veterinary cardiologists or a cardiology resident under direct supervision of a board-certified cardiologist.

| AAPSE measurement
Aortic annular plane systolic excursion (AAPSE) was measured as described in human literature. 12,22Both ventral and dorsal AAPSE (AAPSE-V and AAPSE-D respectively) were measured as shown in Figure 1 from a RPSA view at the level of the aortic valve.A minimum of 5 cardiac cycles were measured and averaged to calculate the mean values for each cat.Intra-and inter-observer reliability was determined by repeat measurements of a random 10% of the cohort.Intra-observer repeatability was determined for a single observer (Luke C. Dutton) on measurements obtained at least 1 week apart.Inter-observer reproducibility was determined by measurements performed between 2 independent observers (Luke C. Dutton and Jose Novo Matos).Strength of correlation was defined in our study as follows: 0 to .2, little or no correlation; .21 to .50, fair; .51-.75, moderate to good;

| Statistical analysis
.75-1.0 good to excellent. 23Receiver operating characteristic curve analysis was performed to obtain the optimal cut-off of AAPSE to differentiate stage C from stage B cats, and the area under the curve (AUC) was calculated.The latter was defined as follows: poor, 0-.20; slight, .21-.40; moderate, .41-.60; good, .61-.80; and very good, .81-1.0. 24AAPSE values for the control cohort were based on the 90% confidence intervals as previously recommended for the sample size. 25P values <.05 were considered statistically significant.Intraand inter-observer reliability and variability for AAPSE were quantified by intraclass correlation coefficient (ICC) and average percent coefficient of variation (CV), respectively.Intraclass correlation coefficient (ICC) estimates and their 95% confidence intervals were calculated based on a single-observer, absolute agreement, 2-way mixed-effects model.Percent CV was calculated as (SD of the measurements/ average of the measurements) Â 100.In our study, an ICC > .75 and CV < 10% were considered to indicate excellent measurement reliability and low measurement variability, respectively. 26

| Power calculation
Since no AAPSE measurements are reported for cats, the power calculation was based on reported measurements for MAPSE, which in F I G U R E 1 M-mode right parasternal short-axis view at the level of the aortic valve.To measure aortic annular plane systolic excursion (AAPSE) the cursor was placed to bisect the aorta and left atrium, with displacement of the ventral border (AAPSE-V) and dorsal border (AAPSE-D) measured in mm using the leading-edge to leading-edge method.
T A B L E 1 Cohort characteristics and echocardiographic values for the cats in this study.humans is similar to and correlated with AAPSE values. 14Based on this assumption a group size of n = 30 cats would provide a power of 93% to detect a >10% difference in AAPSE measurements between groups (assuming an AAPSE of 5.2 mm in normal cats), with an α value of .05.

| Study cohort
We

| Aortic annular plane systolic excursion values in control cats and reliability
We obtained AAPSE values for cats without evidence of HCM using our control cohort (

| Aortic annular plane systolic excursion in HCM compared to control cats
Compared to control cats, both AAPSE-V and AAPSE-D were lower in cats with HCM stage B (P < .001,Table 1 and Figure 2A,B), and cats with HCM stage C had both an AAPSE-V and AAPSE-D that was lower than cats with HCM stage B (P < .001,Table 2 and Figure 2A,B).In addition, receiver operating characteristic curves were obtained for both AAPSE-V and AAPSE-D to differentiate cats with HCM stage C from stage B. The AAPSE-V receiver operating characteristic curve had a similar AUC compared to AAPSE-D, and both were good at predicting stage C compared to stage B cats (0.93, 95% CI 0.89-0.98versus 0.90, 95% CI 0.84-0.96, Figure 3).An AAPSE-V cut-off of 2.9 mm gave a sensitivity of 83% (95% CI 71%-91%) and specificity of 92% (95% CI 82%-97%) to differentiate cats with HCM stage C from stage B. Similarly, an AAPSE-D cut-off of 2.9 mm gave a sensitivity of 80% (95% CI 67%-88%) and specificity of 82% (95% CI 71%-90%) to differentiate cats with HCM stage C from stage B. F I G U R E 3 Receiver operating characteristic (ROC) curve to assess the ability of aortic annular plane systolic excursion (AAPSE) from the ventral aortic wall (AAPSE-V) to differentiate cats with hypertrophic cardiomyopathy (HCM) stage B and stage C cats.The AUC was 0.93.An AAPSE-V cut-off of 2.9 mm gave a sensitivity of 83% and specificity of 92% to differentiate HCM stage C from stage B cats.

| DISCUSSION
1][32] Cardiac contraction depends on longitudinal shortening, circumferential shortening, radial shortening and an axial twist. 8,33[38] In people with HCM both GLS and MAPSE reduce as the disease progresses. 3,7,10,38Therefore, in humans GLS and MAPSE are used to document systolic dysfunction in patients with HCM and this correlates with disease severity and acts as a prognostic indicator.Similarly in cats, reduction in MAPSE is associated with a worse prognosis 6 and cats with HCM have a reduced GLS, suggesting reduced overall longitudinal systolic function. 11,28 humans, displacement of the aortic root is used to assess longitudinal systolic function.This is referred to as systolic aortic root motion (SARM), aortic root systolic excursion (ARSE) and anterior aortic plane systolic excursion (AAPSE). 12,14,16,39We opted to use the abbreviation AAPSE (and change anterior aortic to aortic annular) to complement MAPSE and tricuspid annular plane systolic excursion to aid ease of use and understanding.In humans AAPSE correlates with other measures of longitudinal systolic function such as MAPSE and GLS. 12,15Although GLS was not performed in our retrospective cases, AAPSE did correlate with MAPSE and tissue Doppler S 0 wave from the IVS.AAPSE has advantages over other methods to assess systolic function as it uses a standard echocardiographic view that is routinely obtained in feline echocardiography, even in unstable cats (unlike MAPSE), and requires no calculation to compare to reference values.To highlight this, AAPSE is used in people as a measure of LV systolic function in the emergency setting. 22Our study suggests that AAPSE could be used as a rapidly acquired and easy to learn measurement to assess LV systolic performance in the acute setting.Many emergency and critical care clinicals will be familiar with obtaining a view to assess the LA:Ao, and since AAPSE utilizes the same view, it might be feasible to a less experienced operator.However, we have not assessed this use in the present report.
Movement of the aortic root can be affected by several factors given the anatomical arrangement.The aortic root is connected to the cardiac skeleton, meaning movements of the heart base would subsequently move the aortic root.In people, the aortic root motion correlates with LV stroke volume, 16,40,41 and decreases when stroke volume is diminished by higher cardiac pacing rates. 40 emptying, but data in people suggests that aortic root motion is mainly attributed to LV systolic function. 16Regardless of the different potential factors affecting AAPSE, this could be a useful and easy to acquire measure in cats with HCM, however the prognostic significance of AAPSE is yet to be determined.
Our study has several limitations.This was a retrospective study, therefore no standard echocardiographic protocol was followed, and thus, not all variables were assessed in every cat (eg, TDI and MAPSE).
Most cats with HCM stage B had normal left atrial size (stage B1), and thus mild HCM.This might have been the cause for the overlap in the AAPSE values between control cats and HCM stage B cats.The difference between control (normal) and HCM stage B2 cats would be larger with less overlap (given the correlation between AAPSE and left atrial size).However, we do not believe this affects our ability to say that cats with HCM overall have a lower AAPSE compared to control cats.We used a referral cohort, therefore a selection bias was introduced.Additionally, we excluded cases with comorbidities that mimic HCM (HCM phenocopies) or could possibly affect AAPSE measurement.These limitations can potentially lead to an inflated performance of the echocardiographic variables used and to spectrum bias and limited challenge bias that might lead to an overestimation of the sensitivity and specificity of AAPSE. 44,45Moreover, our results might not be applicable to a general cat population. 45Tissue Doppler imaging was only available for the IVS and not the FW, limiting correlation with MAPSE FW.Although AAPSE was lowest in cats with HCM stage C and correlates with other prognostic markers, further longitudinal data is required to fully assess the prognostic significance of AAPSE, ideally using multivariable analysis to assess if AAPSE is an independent outcome predictor.We did not assess observer repeatability and reproducibility for all echocardiographic measurements, but these were considered to be good to excellent given the level of experience in feline cardiology of the observers and standardization of echocardiographic views.We also did not assess the impact of different echocardiographic machines used in this study on the measurements obtained.However, we used standard B-mode and M-mode linear measurements that are unlikely to vary between echocardiography systems.
[48] In conclusion, we found that AAPSE was lower in cats with HCM, and cats with HCM stage C have the lowest values of AAPSE.This might be a new measurement of LV systolic performance, most likely related to longitudinally systolic function.We provide values for AAPSE in normal cats, and show that the measurement is easy to acquire and has excellent intra-and inter-observer reliability.

, 18 2. 2 |
Echocardiographic assessment A complete echocardiographic examination was performed in all cats including B-mode, M-mode and Doppler studies from standard rightand left-sided views as outlined in published veterinary guidelines. 1 If a cat had >1 echocardiogram stored over the study period, only the first echocardiogram was used.The LV wall thickness was measured at the thickest part of both the interventricular septum (IVS) and free wall (FW), in 3 2-dimensional image planes (right parasternal long axis 4-chamber view, right parasternal long-axis 5-chamber view and RPSA view at the level of the papillary muscles) using the leading-edge to leading-edge method, over at least 3 cardiac cycles and averaged.A diagnosis of HCM was made if LV wall thickness in any segment measured at end-diastole on B-mode was ≥6 mm.Normal cats were defined as maximum LV wall thickness at end-diastole of ≤5 mm.Left atrial size was assessed by the left atrium-to-aortic root ratio (LA:Ao) in the RPSA view and left atrial diameter (LAD) in a right parasternal long-axis 4-chamber view measured at end-systole. 19-21Left atrial function was assessed by calculating left atrial fractional shortening (LAFS%) from a RPSA view using M-mode. 21The presence of systolic anterior motion of the mitral valve (SAM) was documented based on 2-dimensional and color Doppler assessment of a right parasternal long-axis 5-chamber view, with evidence of movement of the anterior mitral valve leaflet toward the IVS in systole with turbulent flow in the LV outflow tract.The LV systolic function was measured by LV fractional shortening on M-mode (LVFS%) from a RPSA view and where available the S 0 wave of the IVS on tissue-Doppler imaging (TDI) obtained from a left apical 4-chamber view.Mitral annular plane systolic excursion (MAPSE) was also measured if there was sufficient quality in the left-apical 4-chamber view (for anatomic M-mode) or if recorded at time of image acquisition.The LV internal dimension in end-diastole (LVIDd) was measured from a right parasternal long-axis 4-chamber view using the trailing edge to leading edge method on B-mode.The LV outflow tract velocities and velocity time integral were also recorded from a left apical 4-chamber view.
Statistical analysis was performed using commercially available software (GraphPad Prism version 9.3.1 for Mac OS X, GraphPad Software, San Diego, California, USA; IBM SPSS Statistics for Mac, Version 23.0.Armonk, NY: IBM Corp).Data was assessed for normality by both histogram analysis and Shapiro-Wilk test.Parametric data are presented as mean (SD) and nonparametric data presented as median (range).Categorical data is presented as frequency and percentage.Comparison between 2 groups were carried out using Student's t-test for continuous parametric data and Mann-Whitney U test for continuous nonparametric data.Comparisons between 3 or more groups was performed using 2-way ANOVA with Tukey posthoc analysis for continuous parametric data and Kruskal-Wallis test for continuous nonparametric data with Dunn's post-hoc analysis.Chi-square test was used to compare categorical variables between 2 or more groups.Pearson's correlation coefficient was performed to assess correlation between AAPSE and echocardiographic variables of LV systolic function, forward aortic flow, LAFS% and heart rate.

1 F I G U R E 2
Aortic annular plane systolic excursion of ventral aortic wall and AAPSE-D showed a moderate positive correlation with MAPSE IVS T A B L E 2 Values for aortic annular plane systolic excursion (AAPSE) based on measurements from the control group of cats (n = 60).Box and whisker plots showing distribution of aortic annular plane systolic excursion (AAPSE) measurements for each of control cats, cats with hypertrophic cardiomyopathy (HCM) stage B and stage C. AAPSE ventral was lower in cats with HCM stage B and lowest in stage C cats (A, P < .001).Distribution of AAPSE dorsal measurements for each group show a similar pattern (B, P < .001).****P < .001based on ANOVA with post-hoc Tukey analysis.

F I G U R E 4
Correlation of aortic annular plane systolic excursion (AAPSE) with other echocardiographic indices.Aortic annular plane systolic excursion from the ventral aortic wall (AAPSE-V) showed a positive correlation with mitral annular plane systolic excursion (MAPSE) from the interventricular septum (IVS; A), tissue Doppler systolic velocity of the interventricular septum (IVS TDI S 0 ; B) and left atrial fractional shortening (LAFS%; C).There was no correlation with left ventricular fractional shortening (LVFS%; D).There was also no correlation between AAPSE and aortic velocity time integral (VTI; E). compared to control cats and is lowest in cats with HCM stage C. We provided AAPSE values for normal cats.AAPSE demonstrated a high sensitivity and specificity to distinguish cats with HCM stage B from stage C. AAPSE showed positive correlation with other indices of LV longitudinal systolic function (MAPSE, TDI S 0 ), but seemed independent of indices of short axis function (ie, LVFS%) and had a weak association with aortic flow variables.Furthermore, AAPSE showed excellent intraand inter-observer repeatability and reproducibility, respectively.

Table 2
Normally distributed data are presented as mean (±SD), non-normally distributed data is presented as median (range).Categorical data are presented as frequency and percentage.Greek values indicate statistically significant result between groups based on post-hoc analysis.α = compared to control group, β = compared to hypertrophic cardiomyopathy (HCM) stage B group, γ = compared to HCM stage C group.Bold values indicate P < .05.Abbreviations: AAPSE, aortic annular plane systolic excursion; AAPSE-V, aortic annular plane systolic excursion from the ventral aortic wall; AAPSE-D, aortic annular plane systolic excursion from the dorsal aortic wall; LAAo, left atrium to aortic root ratio; LAD, left atrial diameter; LAFS%, left atrial fractional shortening; LVOT Vmax, left ventricular outflow tract maximum flow velocity; LVFS%, left ventricular fractional shortening; M/F, male or female; MAPSE, mitral annular plane systolic excursion; MAPSE-FW, mitral annular plane systolic excursion from the free wall; MAPSE-IVS, mitral annular plane systolic excursion from the interventricular septum; Other breeds: * 1 cat of each breed: Russian Blue, Persian, Exotic Shorthair, British Longhair, Norwegian Forest, Birman, Thai; ** 1 cat of each breed: British Longhair, Exotic Shorthair, Chartreux; *** 1 cat of each breed: Ragdoll, British shorthair, Himalayan, Norwegian Forrest, Chartreux, Sphynx, British Longhair, Cornish Rex.SAM, systolic anterior motion of the mitral valve; IVS TDI S 0 , systolic velocity of the interventricular septum obtained from tissue Doppler imaging; VTI, velocity time integral; Y/N, yes or no.
42 addition, the aorta shares its dorsal border with the left atrium, therefore changes in the emptying and filling of the left atrium can be reflected in the movement of the aortic root.42Displacement of the posterior wall of the aorta correlates with left atrial volume changes.
43In our study AAPSE had a moderate positive correlation with LAFS%, which could support this hypothesis.As indices of cardiac performance generally decrease with advancing disease and onset of CHF the fact that 2 measures are correlated does not necessarily imply they are assessing the same variable.It seems plausible that AAPSE is affected both by LV systolic function and left atrial filling and