Impact of selected individual dog traits on echocardiographic parameters obtained in 1-dimensional (M-mode) and 2-dimensional (2D) imaging

Abstract

The popularity and availability of echocardiography in veterinary practice for companion animals have substantially increased in recent years. The results obtained during the procedure are compared to reference values established for the general dog population or to standards developed for a specific dog breed. The aim of this study was to determine whether individual dog traits, such as body weight, chest structure, and level of physical activity and performance, affect the reference values for echocardiographic parameters. Published reference values for echocardiographic examination parameters for 32 dog breeds were analyzed and the relationship between individual echocardiographic parameters and body weight, chest structure, and level of physical activity and performance was then statistically analyzed. It was found that echocardiographic parameters are affected by the dog’s weight and physical activity. There was no significant relationship between heart size and chest structure. The great variety of dog breeds means that echocardiographic findings should be individually interpreted rather than establishing reference ranges for each breed in population studies. This will allow for a more accurate interpretation of the results obtained in the echocardiographic examination and consequently lead to earlier diagnosis of changes in myocardial morphology.

Introduction

Echocardiography is a non-invasive imaging method for diagnosing heart disease. It allows the morphology and function of the cardiac muscle and major blood vessels to be assessed. Reference values for echocardiographic examination for the general dog population are available in the form of tables (1–4), echocardiographic calculators (www.beta.vin.com), or as applications for iPhones (Cardio-Calculator by Melastudio Arti Grafiche, www.appadvice.com).

All echocardiographic reference values for dogs have been set based only on the body weight or body surface area (BSA) and do not include other characteristics, such as breed, chest structure, or level of physical activity or performance. Due to the growing popularity and availability of echocardiographic examination in veterinary medicine in recent years, a trend has emerged to set specific reference values for individual breeds. This would increase the accuracy of the examination and allow even small deviations from the normal value to be detected, as well as improve breeding programs for a given breed. To date, reference values for echocardiographic examination have been published for over 37 breeds (5).

The aim of this study was to determine the relationship between the individual dog’s characteristics, such as body weight, chest structure, and level of physical activity and performance, and reference values for echocardiographic parameters.

Materials and methods

Reference values for echocardiographic examination for 32 dog breeds published in the scientific literature were reviewed (Table I). In the available publications, the examined dogs were of different ages and sexes, except for beagles, whose standards were set on a homogeneous female population, aged 21 mo (6). All dogs for which the breed reference values were established were clinically healthy and in good condition, with no acquired or congenital heart disease detected by echocardiography. Chest structure and level of physical activity and performance of individual dog breeds were determined according to the American Kennel Club guidelines (www.akc.org) (Table II). Reference values for 4 breeds (German shepherd, whippet, Irish wolfhound, and Labrador retriever) were described in 2 independent publications, which were also included in the statistics (2,7,13,18,25,26,31,32). Reference values for echocardiographic parameters for Labrador retrievers from Saini’s study have been broken down by range of body weight (7).

Table I.

Sources of reference values for echocardiographic measurements for 32 dog breeds.

Breed	Age (months)	Number of dogs enrolled in study	Reference
Miniature poodle	24 to 84	20	(12)
Italian greyhound	18 to 108	20	(18)
Dachshund	9 to 192	41	(19)
Indian spitz	36 to 60	24	(20)
English cocker spaniel	—	12	(15)
Hungarian mudi	12 to 144	28	(21)
Whippet	10 to 169	105	(22)
Whippet	18 to 84	20	(18)
Cavalier King Charles spaniel	12 to 145.2	134	(11)
Corgi (Welsh corgi)	24 to 84	20	(12)
Beagle (females)	21	6	(6)
Border collie	24 to 144	20	(5)
English springer spaniel	30 to 153.6	39	(23)
English pointer	—	16	(16)
Hungarian greyhound	12 to 132	22	(21)
Hungarian vizsla	6 to 120	45	(21)
Afghan hound	24 to 84	20	(12)
Greyhound	18 to 84	20	(18)
Saluki	28 to 124	75	(24)
Labrador retriever	16 to 48	24	(25)
Labrador retriever	12 to > 60	24	(7)
Labrador retriever	12 to > 60	7	(7)
Golden retriever	24 to 84	20	(12)
German shepherd	12 to 96	50	(26)
German shepherd	12 to 60	60	(13)
Doberman pinscher	12 to 132	39	(27)
Estrela Mountain dog	18 to 123	74	(28)
English bull terrier	9 to 30	14	(14)
Alaskan malamute	24 to 108	77	(17)
English bulldog	—	18	(29)
Boxer	25.2 to 132	81	(30)
Irish wolfhound	12 to 108	20	(31)
Irish wolfhound	12 to 102	262	(32)
Great Dane	12 to 72	15	(31)
Spanish mastiff	24 to 48	12	(33)
Bordeaux dog	12 to 84	31	(34)
Newfoundland	12 to 132	27	(31)

Table II.

Chest structure and level of physical activity and performance of individual dog breeds.

Breed	Chest structure	Physical activity	Performance
Miniature poodle	Deep	Moderate	Hunting
Italian greyhound	Deep	Vigorous	Racing
Dachshund	Deep	Moderate	Hunting
Indian spitz	Deep	Light	Companion
English cocker spaniel	Deep	Moderate	Hunting
Hungarian mudi	Deep	Moderate	Shepherd
Whippet	Deep	Vigorous	Racing
Cavalier King Charles spaniel	Barrel	Light	Companion
Corgi (Welsh corgi)	Barrel	Moderate	Shepherd
Beagle	Barrel	Moderate	Hunting
Border collie	Deep		Shepherd
English springer spaniel	Deep	Moderate	Hunting
English pointer	Deep	Moderate	Hunting
Hungarian greyhound	Deep	Vigorous	Racing
Hungarian vizsla	Deep	Moderate	Hunting
Afghan hound	Deep	Moderate	Shepherd
Greyhound	Deep	Vigorous	Racing
Saluki	Deep	Vigorous	Racing
Labrador retriever	Deep	Moderate	Water
Golden retriever	Deep	Moderate	Water
German shepherd	Deep	Moderate	Shepherd
Doberman pinscher	Deep	Moderate	Defensive
Estrela Mountain dog	Deep	Moderate	Shepherd
English bull terrier	Barrel	Moderate	Defensive
Alaskan malamute	Barrel	Vigorous	Draft
English bulldog	Barrel	Light	Companion
Boxer	Barrel	Moderate	Defensive
Irish wolfhound	Deep	Moderate	Hunting
Great Dane	Deep	Moderate	Defensive
Spanish mastiff	Barrel	Moderate	Defensive
Bordeaux dog	Barrel	Moderate	Defensive
Newfoundland	Barrel	Moderate	Water

Echocardiographic parameters were divided into primary and secondary measurements. Primary measurements were left ventricular dimension in diastole (LVDd), left ventricular dimension in systole (LVDs), interventricular septum in diastole (IVSd), interventricular septum in systole (IVSs), left ventricular free wall in diastole (LVWd), left ventricular free wall in systole (LVWs), left atrial dimension (LA), and aortic root diameter (Ao).

Secondary echocardiographic measures were fractional shortening (FS%) and left atrium-to-aorta ratio (LA:Ao). Fractional shortening (FS%) is an echocardiographic parameter calculated using the following formula:

$$\text{FS}\%=\frac{{\text{LVD}}_{\text{d}}-{\text{LVD}}_{\text{s}}}{{\text{LVD}}_{\text{d}}}\times 100$$

LA:Ao was calculated using the following formula:

$$\text{LA:Ao}=\frac{\text{LA}}{\text{Ao}}$$

We were not able to determine the impact of level of performance on echocardiographic parameters as echocardiographic values for too few existing dog breeds have been examined and published. Thirty-two dog breeds included in the analysis were assigned to 7 types of performance, which meant that the groups obtained were too small for conducting the analysis (Table II).

Statistical analysis

The analysis was based on the mean or median values of echocardiographic measurements and body weight published in scientific articles. In several articles in which only ranges were presented, the average of the minimum and maximum was included in the analysis. Normality of distribution of numerical variables was verified using the Shapiro-Wilk W-test and body weight (BW), the only non-normally distributed variable, was logarithmically transformed (natural logarithm). The relationship between echocardiographic measurements and BW was analyzed using Pearson’s linear correlation coefficient (r) and reported with its 95% confidence interval (CI), and the coefficient of determination (R²), which indicates the part of variance of the echocardiographic measurement that could be explained by the correlation with BW. Based on the Pearson’s coefficient, correlations were classified as very high when absolute value of r ≥ 0.9 to 1.0, high when r ≥ 0.7 to 0.9, moderate when r ≥ 0.5 to 0.7, low when r ≥ 0.3 to 0.5, and negligible when r = 0.0 to 0.3 (8).

As BW proved significantly linked to almost all echocardiographic measurements, it was included as a covariate in the further analyses of the relationship between echocardiographic measurements and categorical characteristics of dog breeds. These analyses were carried out first with the use of the multivariate analysis of covariance (MANCOVA) and then, if it proved significant, with the univariate analysis of covariance (ANCOVA) with the Bonferroni post-hoc test. All tests were 2-sided. A significance level (α) was set at 0.05. Statistical analysis was conducted in TIBCO Statistica 13.3.0 (TIBCO Statistics, Palo Alto, California, USA).

Results

Impact of body weight on echocardiographic measurements

All primary measurements revealed a significant high positive linear correlation with the logarithm of body weight (Table III). Fractional shortening (FS%) showed a significant moderate negative linear correlation with the logarithm of body weight (Figure 1). There was no significant correlation between LA:Ao and the dog’s body weight (P = 0.759) (Figure 2; Table III).

Table III.

Relationship between echocardiographic measurements and dog’s body weight.

Echocardiographic measurement	n	Pearson’s linear correlation coefficient (r) with 95% CI	P-value	R2
LVDd	37	0.87 (0.76, 0.93)	< 0.001	75%
LVDs	33	0.79 (0.62, 0.89)	< 0.001	63%
IVSd	37	0.75 (0.56, 0.86)	< 0.001	57%
IVSs	33	0.72 (0.50, 0.85)	< 0.001	52%
LVWd	37	0.69 (0.47, 0.83)	< 0.001	47%
LVWs	33	0.76 (0.56, 0.88)	< 0.001	58%
LA	36	0.81 (0.66, 0.90)	< 0.001	65%
Ao	36	0.89 (0.79, 0.94)	< 0.001	79%
FS%	37	−0.52 (−0.72, −0.24)	0.001	27%
LA:Ao	36	−0.05 (−0.37, 0.28)	0.759	—

R² — coefficient of determination; LVDd — left ventricular dimension in diastole; LVDs — left ventricular dimension in systole; IVSd — interventricular septum in diastole; IVSs — interventricular septum in systole; LVWd — left ventricular free wall in diastole; LVWs — left ventricular free wall in systole; LA — left atrial dimension; Ao — aortic root diameter; FS% — fractional shortening; LA:Ao — left atrium-to-aorta ratio.

Figure 1.

A scatter plot of the correlation between the fractional shortening (FS%) and the natural logarithm of body weight.

Figure 2.

A scatter plot of the correlation between the left atrium-to-aorta ratio (LA:Ao) and the natural logarithm of body weight.

Impact of chest structure on echocardiographic measurements

No significant relationship was found between echocardiographic measurements and chest conformation (P = 0.727).

Impact of physical activity on echocardiographic measurements

Multivariate analysis, including body weight as a covariate, showed that there was a significant relationship between level of physical activity of the dog breed and one of the echocardiographic measurements (P = 0.041). One-dimensional analysis showed that such a relationship existed for fractional shortening (FS%). Controlling for body weight, FS% was significantly linked to physical activity (P = 0.002). In dogs with high activity, FS% was significantly lower than in dogs with low (P = 0.008) and moderate (P = 0.047) activity (Figure 3).

Figure 3.

Relationship between the mean fractional shortening (FS%) (with 95% CI) and the physical activity of the dog breed adjusted by the natural logarithm of body weight.

Discussion

It is important to establish echocardiographic values for individual dog breeds, not only for primary measurements such as left ventricular dimension or left atrial dimension, but also for secondary measurements. Even though FS and LA:Ao are secondary echocardiographic measurements, they play an important role in diagnosing numerous heart diseases in dogs. Fractional shortening (FS%) is a parameter based on which left ventricular systolic function is assessed. Its measurement is used to diagnose and monitor the course of dilated cardiomyopathy (DCM), which occurs, among other criteria, with left ventricular systolic dysfunction (1). The LA:Ao parameter is used, among other criteria, in patients diagnosed with mitral valve disease (MVD). It is one of the criteria used to assign patients to individual stages of the disease. According to the guidelines of the American College of Veterinary Internal Medicine, MVD has 4 stages: A, B, C, and D, and its treatment should be started when the heart cavities have become enlarged (stage B2). Thanks to the LA/Ao calculation, among others, it is possible to recognize whether the patient has an enlarged left atrium, classify them to stage B2 of the disease, and start treatment to delay its progression (9).

Many publications to date have clearly stated that the size of the heart cavity depends primarily on body weight (1–4,7). However, it seems that this parameter may not be the only significant indicator affecting echocardiographic reference values. The published reference values are based on healthy dogs, usually with normal body weight. Over the past years, obesity has become a major medical problem in pets. There is a growing tendency for pathological weight gain in dogs, which seems to reflect the scale of the problem in humans (10). Dogs whose body weight deviates significantly from the permissible weight standards are often brought to the cardiology practice. These situations may be associated with obesity, but can also be associated with cachexia, which can be cardiac or resulting from comorbidities, such as cancer. After echocardiographic examination and measurements, it is not clear whether obtained results should be compared to the reference values set according to the patient’s current, too high or low body weight, or to their ideal body weight. As this issue has not been clarified, more research is needed to answer this question and determine whether echocardiographic parameters in obese dogs can be compared to the norms for their breed.

In this study, the body condition was not analyzed because authors of the publications reviewed stated that the population of dogs studied was in normal condition.

Contrary to the conclusion in publications outlining echocardiographic standards for individual breeds that the traits of dogs affect echocardiographic parameters (11–14), the analysis conducted by the authors of the present study demonstrated that chest structure, one of these characteristics, does not affect heart size. In addition, dogs do not always participate in activities typical of their breed, e.g., it cannot be assumed that each of the tested sighthounds, which is a sporting breed, was subjected to intense training. The analysis should therefore be repeated in the future when most dog breeds will undergo cardiology tests.

There are many studies on the significant effects of physical activity on cardiac morphology (12,15–17). Intense physical effort has been shown to result in weight gain and so-called “athlete’s heart” in working and athletic dogs, including border collies, Alaskan malamutes, and whippets and other sighthounds. Vigorous physical activity leads to cardiac dilation, hypertrophy, or increased vagal nerve tension, which result in a slower heart rate (17). The use of reference values for the general population in dogs subjected to intense physical effort will therefore probably lead to misinterpretation of the obtained echocardiographic examination results and prevent a correct diagnosis.

The analysis in this study was based on various studies carried out on a specific breed of dogs of different sex and age (except for beagles), which is the limitation of this study. Another limitation was the large difference between the minimum and maximum values, which is a common problem in clinical trials conducted on animals due to the differences in size and weight in a given breed.

Due to the wide variety of dog breeds and inhomogeneous research groups, echocardiographic measurements should be established individually for each breed based on a population study. This will allow for a more accurate interpretation of the results obtained in the echocardiographic examination and lead to earlier diagnosis of heart disease.

The analysis in this study confirmed that all primary echocardiographic parameters are dependent on body weight, although other factors that may affect these values, such as chest structure and physical activity, were also noted. These particular correlations were not found in this study, however, and should be confirmed in another study with a research group selected on the basis of chest structure or properly defined physical activity, not just breed. While this study has several limitations, it has shown that many factors should be considered when determining echocardiographic values.