Assessment of diastolic function using mitral flow propagation velocity in cats

Abstract

The objectives of this study were to investigate the usefulness of mitral flow propagation velocity (Vp) in cats by evaluating the effect of the flow pattern summation and evaluation of Vp variables in cats with hypertrophic cardiomyopathy (HCM). Healthy cats were categorized into summation (Sum) and separation (Sepa) groups to evaluate the effects of the flow pattern summation on Vp. Cats with HCM were categorized into HCM left atrial (LA) (−), LA (+), and LA (++) groups according to the degree of LA enlargement to investigate the feasibility of Vp. There were no significant differences noted in Vp between the Sum and Sepa groups and no significant correlation between Vp and heart rate. Decline of Vp was associated with the degree of LA enlargement. Mitral flow propagation velocity appeared to be clinically feasible in cats and could possibly be useful in the detection of diastolic dysfunctions in cats with HCM.

Introduction

Hypertrophic cardiomyopathy (HCM) is the most common heart disease in cats (1). It is characterized by the presence of concentric left ventricular (LV) hypertrophy. This is diagnosed when the LV wall thickness is 6 mm or more on the echocardiographic examination without pressure overload and systemic diseases, which are known to cause LV hypertrophy. As LV wall thickness is reportedly influenced by volume depression, heart rate, and body weight (2–4) the measurement of LV thickness is insufficient to diagnose HCM in cats.

Diastolic dysfunction is also a primary abnormality of HCM (5). It leads to left atrial (LA) enlargement followed by arterial thromboembolism. Assessment of diastolic dysfunction is also considered key in diagnosing and managing HCM in cats.

Echocardiography allows for the non-invasive assessment of myocardial function and is used for detecting dysfunction in cats with HCM. Pulsed wave Doppler (PWD) is a traditional technique used for the assessment of the LV filling pattern (6). Tissue Doppler imaging (TDI) is a representative method. Peak early diastolic myocardial velocities are reportedly reduced in cats with HCM (7,8). Speckle tracking echocardiography (STE) is a recent ultrasound technique that assesses ventricular function and is useful for assessing cardiac function in cats with HCM (5,9).

Numerous studies on diastolic function in cats with HCM have reported an occasional summation of early diastolic (E) and atrial systolic (A) waves in PWD, TDI, and STE. Little is known about the causes of these summations, but they may be due to elevated heart rate (HR) or indicate diastolic impairment (7,9,10). Vagal maneuvers and ocular administration of timolol have been used to decrease HR (11,12), but E and A waves were not completely separated and cardiac function may have been influenced by vagotonia or the administration of beta-blockers. Ultimately, the assessment of cardiac function using the aforementioned methods has certain limitations.

Mitral flow propagation velocity (Vp) is determined from color M-mode Doppler echocardiography and is a measure of the rapid filling phase (13). It refers to the propagation of the maximum velocity across specific locations during early diastole (14). It has a strong and negative correlation with the time constant of LV relaxation; a slow Vp reflects abnormal LV relaxation (15).

The usefulness of Vp has been reported in humans with HCM and dogs with dilated cardiomyopathy (16,17). In cats, Vp is an index of LV relaxation and functions independently of preload and HR (18). The cats in these reports, however, were anesthetized or received nasal planum or ocular pressure vagal maneuvers to separate the E and A waves. Presently, there are no published reports regarding Vp in healthy cats or the potential diagnostic utility of Vp in HCM.

We hypothesized that Vp is an index that is not influenced by wave summation. We sought to evaluate Vp as an assessment of diastolic function, with summation of E and A waves in healthy cats, and to evaluate Vp availability in cats with HCM during various stages.

Materials and methods

Animals and study design

Study 1: Comparing the separation and summation of E and A waves

The study groups comprised 11 cats with the separation of E and A waves (Sepa group) and 17 cats with the summation of E and A waves (Sum group) (Figure 1). Cats in both groups were either client-owned (n = 18) or experimental cats (n = 10), and all were healthy on the basis of clinical history, physical examination, blood pressure measurement, conventional echocardiographic examination, thoracic radiographs, electrocardiogram, complete blood cell counts, and serum chemistry panels. None of the cats were receiving treatment at the time of enrollment into the study. We also assessed interobserver and intraobserver variability for Vp in both the Sepa and Sum groups.

Figure 1.

The mitral flow propagation velocity (Vp). (A) Vp in Sepa group. E and A waves were separated. (B) Vp in Sum group. E and A waves were summated.

Study 2: Comparing healthy cats and cats with HCM

We compared Vp between the Sepa group and cats with HCM and between the Sum group and cats with HCM. The cats with HCM were prospectively recruited and classified according to left atrium to aorta ratio (LA/Ao) by echocardiography examination into one of the following groups: HCM LA (−) (LA/Ao < 1.5), HCM LA (+) (LA/Ao = 1.5 to 1.8), and HCM LA (++) (LA/Ao > 1.8) (19). All cats in the HCM group belonged to owners and were patients of the Azabu University Veterinary Teaching Hospital or the cardiology service of the small animal medicine clinic from January 2017 to May 2018.

The assessment of mitral propagation velocity

Mitral propagation velocity was measured as previously described (18). It was determined from a color M-mode recording of LV inflow. The M-mode scan line is placed through the center of the LV inflow blood column. Doppler velocity curves were assessed using a sweep speed of 200 mm/s. The color velocity scale was adjusted by shifting the baseline to optimize color aliasing of the mitral inflow signal (between 35 and 55 cm/s). Vp of early diastolic flow into the LV was obtained by manually tracing the Vp line over a distance of 1 to 3 cm, starting at the mitral valve tips, and along the transition zone where velocity aliasing occurs (Figure 2).

Figure 2.

Vp can be measured from the slope of the line of the first aliasing velocity during early filling from the mitral valve. The sweep speed was 200 mm/s, the color velocity scale was between 35 and 55 cm/s, and Vp line was over a distance of 1 to 3 cm. MV — mitral valve; LA — left atrial; LV — left ventricular.

Conventional echocardiography

All echocardiographic images were acquired using an ultrasound unit equipped with 7 MHz and 10 MHz transducers (Vivid E9, Vivid 7, or Vivid S5; GE Healthcare, Tokyo, Japan). Echocardiographic examinations were performed by 2 experienced sonographers (KS and TA). All echocardiographic measurements were directly performed on screen freeze-frame images, with 5 consecutive measurements averaged to determine each value.

Cats were gently restrained in a lateral recumbent position. The aortic and left atrial diameters at end-systole (AoD and LAD, respectively) and LA/Ao were measured from the right-sided parasternal short-axis view at the level of the aortic valve using a 2-dimensional method, as previously described and validated (20,21). Left ventricular internal diameter (LVIDd) and interventricular septum and LV free wall thickness at end-diastole (IVSd and LVFWd, respectively) were measured from the right parasternal short-axis view at the level of the chordae tendineae, and the LV shortening fraction (FS) was calculated. Heart rate was calculated using M-mode images.

The HCM was determined when the IVSd and/or LVFWd was > 6 mm on the M-mode echocardiographic examination or in at least 4 left ventricular sites from the short-axis view, in the absence of pressure overload or systemic diseases known to cause LV hypertrophy (22). The mean value of the thickest LV segment among the measured ventricular sites (LVTSd) was calculated.

Statistical analyses

All measurements were expressed as mean ± standard deviation (SD). All data were visually inspected and tested for normality using the Kolmogorov-Smirnov test.

In study 1, we compared age, body weight (BW), Vp, and echocardiographic data using a Student’s t-test for normally distributed data. When the data were non-normally distributed, we used a Mann-Whitney U-test. The correlation between Vp and HR was examined using Spearman’s rank correlation coefficient. Gender differences were assessed using the χ² test.

In study 2, we compared age, BW, Vp, and echocardiographic data among the Sepa, Sum, and all HCM groups using a 1-way analysis of variance (ANOVA) for normally distributed data. When a significant difference was detected, multiple comparisons were evaluated using a Bonferroni correction. When the normality test failed, we applied the Kruskal-Wallis test, and when a significant difference was detected, multiple comparisons were evaluated using the Steel-Dwass test. Gender differences were assessed using the χ² test. In all cases, statistical significance was set at P < 0.05. Effect sizes for Vp were calculated as η2 and were defined as small (η2 > 0.01), medium (η2 > 0.06), and large (η2 > 0.14).

Interobserver and intraobserver variability for Vp was assessed in 6 cats in each group by calculation of coefficients of variation (CV) using the formula:

$$\text{CV}=(\text{SD}/\text{arithmetic\hspace{0.17em}mean\hspace{0.17em}of\hspace{0.17em}measurements})\times 100$$( 23)

For interobserver and intraobserver variabilities, one scan was randomly selected from each cat, and was remeasured 4 weeks after initial measurements. The degree of variability was arbitrarily defined as follows: CV < 15% classified as good and CV > 15% as poor (9,24,25).

Results

Age, gender, body weight, and breed

The results for age, gender, BW, and breed in studies 1 and 2 are shown in Table I. Age was significantly higher in the HCM LA (+) and HCM LA (++) groups (P < 0.001). There were no significant differences noted in gender and BW among any of the groups. In the HCM LA (+) group, 2 cats received acetaminophen (Aspirin; Bayer, Tokyo, Japan) and 1 cat received clopidogrel. In the HCM (++) group, 1 cat received Aspirin (Bayer) and 1 cat received antibiotics. All cats in the HCM groups had no clinical symptoms.

Table I.

Characteristics of separation (Sepa), summation (Sum), and hypertrophic cardiomyopathy left atrial (HCM LA) groups.

	Sepa	Sum	HCM LA (−)	HCM LA (+)	HCM LA (++)
Age (mo)	52.3 ± 40.5	45.9 ± 30.9	42.8 ± 32.5	90.5 ± 26.4*	97.7 ± 17.2*
Body weight (kg)	3.96 ± 0.71	3.85 ± 0.55	4.31 ± 1.19	3.8 ± 0.86	4.10 ± 1.69
Sex (F/M)	8/9	10/11	6/12	5/5	3/4
Breed	Sphynx 1	Scottish fold 2	Norwegian forest cat 1	Maine coon 1	Singapura 1
	Maine coon 2	Norwegian forest cat 2	American shorthair 2	Scottish fold 3	Bengal 1
	Norwegian forest cat 2	American shorthair 2	Maine coon 2	American shorthair 3	Siamese 1
	Scottish fold 3	Maine coon 3	Scottish fold 4	Domestic shorthair 3	Domestic shorthair 4
	Domestic shorthair 9	Domestic shorthair 12	Domestic shorthair 9

^*Significant difference compared with Sepa, Sum, and HCM LA (−) groups by multiple comparison.

Conventional echocardiography

The results of conventional echocardiography are shown in Table II and intraobserver variability of Vp indices is shown in Table III. The IVSd in all HCM groups was significantly higher than in the Sepa and Sum groups (all P < 0.001). The LVIDd was significantly lower in the HCM LA (++) group than in the Sepa and Sum groups (P = 0.034 and P = 0.012, respectively). Compared to the Sepa and Sum groups, LVFWd was significantly higher in the HCM LA (−) (both P < 0.001), HCM LA (+) (P = 0.0017 and P = 0.002, respectively), and HCM LA (++) (both P = 0.006) groups. Heart rate was significantly higher in the Sum group compared to the Sepa, HCM LA (−), and HCM LA (+) groups (P < 0.001, P < 0.001, and P = 0.022, respectively). The LAD and LA/Ao were significantly higher in HCM LA (+) and HCM LA (++) groups compared with other groups (all P < 0.001), and in the HCM LA (++) group compared with the HCM LA (+) group (P = 0.0018). LVSTd was significantly higher in the HCM groups compared to the Sepa and Sum groups (all P < 0.01).

Table II.

Results of conventional echocardiography in separation (Sepa), summation (Sum), and hypertrophic cardiomyopathy left atrial (HCM LA) groups.

	Sepa	Sum	HCM LA (−)	HCM LA (+)	HCM LA (++)
IVSd (cm)	0.40 ± 0.07	0.38 ± 0.01	0.47 ± 0.08a	0.48 ± 0.05a	0.51 ± 0.11a
LVIDd (cm)	1.68 ± 0.16	1.64 ± 0.13	1.53 ± 0.18	1.49 ± 0.20	1.43 ± 0.15a
LVFWd (cm)	0.35 ± 0.10	0.34 ± 0.06	0.51 ± 0.08a	0.54 ± 0.09a	0.59 ± 0.13a
FS (%)	45.3 ± 3.7	44.9 ± 1.9	47.4 ± 9.4	51.6 ± 10.2	56.5 ± 7.9
HR (bpm)	174.0 ± 15.5	209.6 ± 17.6c	178 ± 21.6	175.8 ± 26.8	196.4 ± 29.7
AoD (cm)	0.85 ± 0.09	0.86 ± 0.04	0.93 ± 0.12	0.91 ± 0.11	0.95 ± 0.11
LAD (cm)	1.13 ± 0.15	1.17 ± 0.04	1.14 ± 0.18	1.47 ± 0.22b	1.93 ± 0.31b,d
LA/Ao	1.32 ± 0.16	1.37 ± 0.09	1.23 ± 0.08	1.62 ± 0.08b	2.03 ± 0.12b,d
LVSTd (mm)	3.9 ± 0.5	3.8 ± 0.3	6.2 ± 0.2a	6.4 ± 0.3a	6.8 ± 0.6a

^aSignificant difference compared with Sepa and Sum groups.

^bSignificant difference compared with Sepa, Sum, and HCM LA (−) groups.

^cSignificant difference compared with Sepa, HCM LA (−), and HCM LA (+) groups.

^dSignificant difference compared with HCM LA (+) group.

IVS — interventricular septum; LVID — left ventricular internal diameter; LVFW — LV free wall; FS — LV shortening fraction; HR — heart rate; AoD — aortic atrial diameter; LAD — left atrial diameter; LA/Ao — left atrium to aorta ratio; LVSTd — left ventricular thickest segment diameters at end-distole.

Table III.

Intra- and interobserver variability in separation (Sepa), summation (Sum), and hypertrophic cardiomyopathy left atrial (HCM LA) groups.

	Intraobserver (%)	Interobserver (%)
Sum	7.82	13.95
Sepa	4.79	5.23
HCM LA (−)	5.21	6.88
HCM LA (+)	7.88	7.11
HCM LA (++)	6.25	10.45

Propagation velocity (Vp)

In study 1, there were no significant differences observed in Vp between the Sepa and Sum groups (P = 0.23) (Figure 3) and there was no significant correlation noted between Vp and HR in either the Sepa or Sum groups (P = 0.88 and P = 0.93, respectively) (Figure 4). In study 2, the results for the ANOVA indicated a statistically significant difference among the groups (P < 0.01 and η2 = 0.59). Nine cats in HCM LA (−) group, 4 cats in HCM LA (+) group, and 5 cats in HCM LA (++) group had summation of E and A waves. Vp was significantly lower in HCM LA (+) compared with Sepa, Sum, and HCM LA (−) groups (P < 0.001, P < 0.001, and P < 0.001, respectively) and in the HCM LA (++) group compared with the Sepa, Sum, HCM LA (−), and HCM LA (+) groups (P = 0.006, P = 0.002, P = 0.004, and P = 0.011, respectively) (Figures 5 and 6).

Figure 3.

Box and whisker plot for mitral flow propagation velocity (Vp) in separation (Sepa) and summation (Sum) groups. There were no significant differences observed in Vp between the Sepa and Sum groups.

Figure 4.

Correlation of mitral flow propagation velocity (Vp) and heart rate (HR) in separation (Sepa) and summation (Sum) groups. There was no significant correlation noted in both groups.

Figure 5.

Box and whisker plot for mitral flow propagation velocity (Vp) in separation (Sepa), summation (Sum), and all hypertrophic cardiomyopathy left atrial (HCM LA) groups. Vp was decreased in HCM LA (+) compared with Sepa, Sum, and HCM LA (−) groups and in the HCM LA (++) group compared with Sepa, Sum, HCM LA (−), and HCM LA (+) groups.

* P < 0.05. ** P < 0.01.

Figure 6.

Mitral flow propagation velocity (Vp) in summation (Sum), hypertrophic cardiomyopathy left atrial (HCM LA) (+), and HCM LA (++) groups. Vp was lower in cats with HCM with progressive LA enlargement. The sweep speeds were the same (200 mm/s) in all images.

Vp with summation of E and A waves was 86.3 ± 11.3 cm/sec in HCM LA (−) group, 63.1 ± 15.8 cm/sec in HCM LA (+) group, and 47.1 ± 19.6 cm/sec in HCM LA (++) group. Vp with separation of E and A waves was 82.3 ± 8.2 cm/sec in HCM LA (−) group, 58.2 ± 5.5 cm/sec in HCM LA (+) group, and 46.3 ± 4.3 cm/sec in HCM LA (++) group.

Discussion

Mitral flow propagation velocity is a significant measure of cardiac diastolic function and is an accurate index of global LV diastolic function in various human cardiac diseases, including HCM (26). Even during atrial fibrillation, Vp can be used for assessing LV relaxation (27). This method is basically a means of determining how rapidly blood travels from the base of the ventricle towards the apex. In the presence of diastolic dysfunction, the ventricle is stiffer and applies less suction during early diastole. Thus, blood flow is slower.

The results of the present study indicated that Vp was useful, even if E and A waves were summated and Vp functioned independently of HR in healthy and awake cats. Furthermore, in cats with HCM, the decline of Vp was associated with the degree of LA enlargement, which is significantly correlated with the progression of diastolic dysfunction in HCM (28,29).

The American Society of Echocardiography has issued comments on Vp, because of its lower feasibility and reproducibility, and on issues with angulation between M-mode cursor and flow because they may lead to erroneous measurements. However, the interobserver and intraobserver variabilities of Vp in cats appeared to be satisfactory in the present study. The results also indicated that Vp might be useful in cats with HCM to assess the diastolic function.

Reports that assessed Vp in cats determined a normal Vp range of 64 ± 20 cm/s in healthy cats under anesthetized conditions (18) and 37 to 74 cm/s in healthy cats which received nasal planum or ocular pressure vagal maneuvers (24). The results of this study in healthy cats were higher than those in previous reports, presumably because of decreased cardiac function secondary to anesthesia or vagotonia in previous reports.

In this study, 1 cat in the HCM LA (+) group and 2 cats in the HCM LA (++) group exhibited Vp outliers. Sasaki et al (30) and Rajagopalan et al (31) reported that Vp was higher in patients with restrictive cardiomyopathy and constrictive pericarditis. They noted that Vp may be increased in patients with restrictive filling patterns. In HCM, diastolic dysfunction is developed and ultimately reaches a restrictive filling pattern (32). Some cats, particular in LA enlargement groups, had rounded slope of the aliasing. Interobserver variability tended to be slightly high. This may have caused the measurement error, and Vp may be outliers.

There were several limitations to this study. First, the sample size (i.e., number of cats) in each group was small. This could have affected the ability of statistical tests to determine differences between them. Furthermore, all groups included several feline breeds, which could have interacted with echocardiographic parameters (7,33); different breeds may carry different HCM-associated genes, which could produce different or unknown functional manifestations. Second, some cats received treatment. To what extent their status and treatment influenced Vp remains unclear, primarily because of their small number. Third, it was not possible to blind the observer to the presence or absence of LA enlargement while performing the Vp analysis. Forth, Third, the age was higher in the HCM LA (+) and HCM LA (++) groups, which is a potential concern, as previous work has shown that Vp may be influenced by age (34). However, Chetboul et al (35) reported that age was not a major determinant of cardiac function as assessed by echocardiography. On the other hand, some TDI studies have demonstrated that diastolic function may decrease with advanced age in cats (36). Future studies are warranted to assess the effect of age on Vp.

In conclusion, the findings of the present study suggest that Vp is a feasible method for assessing diastolic function in cats with summation of E and A waves. It is also useful for assessing diastolic dysfunction in cats with HCM. A recent study showed that LV diastolic dysfunction is associated with a poor prognosis (37). Thus, Vp measurements in cats might be useful to detect progressive diastolic dysfunction. Further studies are warranted for demonstrating the clinical availability of Vp in cats.