Echocardiography is the golden standard in cardiology diagnostics and provides information about cardiac morphology and function. Echocardiographic measurements play an important role in detection of cardiac enlargement and assessment of cardiac function. In veterinary cardiology, echocardiographic assessment is confounded by the fact that there are numerous breeds of different body shapes and wide range of body weights. To deal with this, attempts have been made to create general population reference values. Unfortunately, studies on echocardiographic reference values have shown that dogs having the same weight, but different body shape have different heart size. To deal with the shortcomings of the general canine population reference values, a more accurate method was created defining the formula for indexing of cardiac linear measurements to body length called allometric scaling. Although this is currently the most used method in evaluating heart size and function in echocardiography, it is unfortunately not suitable for all dogs since there are more and more studies describing breed-specific echocardiographic reference values. In human cardiology, linear measurements have been widely replaced by volumetric measurements since they are more accurate and sensitive method of evaluating heart size and function. So far, in veterinary medicine, reference intervals for left ventricular volumes have been determined for a small number of breeds, and one recent study determined LV volumes indexed to body weight for a general dog population which provides reference intervals for two groups of dogs, sighthound and non-sighthound dogs. Breed-specific reference intervals are still the best and most accurate way for evaluation of heart size and function because certain breeds differ from general population. Cardiac biomarkers, cardiac troponin I (cTnI) and N-terminl Pro-B-type natriuretic peptide (NT-proBNP), have also been studied as additional diagnostic markers of cardiac health and structural heart damage. Although general cut-off values and reference intervals have been determined for dogs, studies have also confirmed that NT-proBNP reference intervals differ among different dog breeds. It is a well-known fact that there are numerous factors that influence heart morphology and function. Athlete's heart is a phenomenon established in human medicine that describes complex remodelling of heart anatomy and physiology to accommodate a state of enhanced cardiovascular performance during intense or chronic physical activity. Hallmarks of athlete’s heart are cardiac chamber enlargement, increased left ventricular wall thickness and enhanced diastolic and global longitudinal systolic function. Similar changes have also been observed in working and racing dogs with elevated level of average daily activity. Belgian Malinois is a working dog breed and echocardiographic reference intervals have not been established for this breed. Primary goal of this study was to determine echocardiographic reference intervals of cardiac size and function in Belgian Malinois. Secondary goals were to establish breed-specific cTnI and NT-proBNP reference intervals, to determine if there are differences in echocardiographic parameters between working dogs and pets, between males and females, to test is there a correlation of echocardiographic parameters with age and weight, to compare Teicholz method of assessing left ventricular volumes to Simpson's method of discs, and to compare the newly established reference intervals with published reference intervals for similar dogs, namely German Shepherd dogs. Seventy-seven Belgian Malinois dogs (37 female and 28 male dogs) aged one to 10 years were examined in this study. Dogs were admitted for general health screening. Based on the level of physical activity they were divided in two groups. Dogs that had an average of more than three hours of intense physical activity per day were classified as working dogs and dogs with less than an average of three hours of intense physical activity per day were classified as pets. All dogs underwent complete physical examination, and complete blood count and basic biochemistry profile was made to assess general health and rule out systemic diseases that could have an impact on heart size and function. Cardiopulmonary dirofilariasis was ruled out using Idexx Snap® 4Dx® Plus test and modified Knott test. In all dogs complete cardiologic examination was done that consisted of two-dimensional and motion-mode, color, spectral and tissue Doppler echocardiography and electrocardiogram. Echocardiography examination was performed in non-sedated animals in left and right lateral recumbency according to the recommendations for standards in transthoracic two-dimensional echocardiography in the dog and cat generated by the echocardiography committee of the specialty of cardiology with simultaneous ECG recording. Linear (thickness of the interventricular septum in diastole and systole, internal diameter of left ventricle in diastole and systole, thickness of the left ventricular free wall in diastole and systole, EPSS, left atrial diameter, left atrial to aortic root ratio) and volumetric measurements (end-diastolic and end-systolic volumes of the left ventricle, and end-systolic volume of the left atrium) of the left side of the heart were done. Volumetric measurements were determined using Simpson's method of discs. Diastolic function was assessed measuring the mitral inflow, whilst diastolic and systolic function of the left ventricle were assessed by tissue Doppler. Furthermore, peak aortic and pulmonic flow velocities were determined. All data were graphically inspected and tested for normality of distribution. Outliers were inspected and excluded if necessary. Statistical analyses were performed according to reference interval guidelines for determination of de novo reference intervals in veterinary species using SPSS® computer software. Because of a small sample size correlation was tested using Spearman correlation test, and differences between pets and working dogs were established using t-test or Mann-Whitney test depending on the normality of distribution. A p-value <0.05 was considered statistically significant. Sixty-five healthy Belgian Malinois were included in this study. There was 28 male and 37 female dogs, of which 31 were classified as pets and 34 as working dogs. Working dogs group consisted of 18 males and 16 females, while pets group consisted of 10 males and 21 females. Overrepresentation of female dogs in pets group could be explained with an easier temperament of female dogs. The average weight of examined dogs was 27,9 kg, with female dogs in general weighing significantly less (p < 0,001) than male dogs, a finding that is seen in numerous other breeds. The average age of examined dogs was 4,2 years and working dogs were significantly younger than pets, but it should be pointed that pets group consisted of many retired working dogs. Defined linear echocardiographic reference values for the Belgians Malinois are breed specific. There was statistically significant difference between measurements of left ventricular diameters and wall thickness between males and females, with male dogs having bigger hearts than females. These findings have also been reported in other breeds. Left ventricular volumes determined using the Simpson’s method of discs from right parasternal long axis and left apical view were similar, and indexed volumes were similar to those reported in other studies. Left ventricular volumes were also statistically significantly larger in males than in females. Furthermore, reference intervals for the assessment of left atrial size have also been established. There is a statistically significant difference in left atrial diameter and volume with no difference in LA/Ao. Although LA/Ao is the only parameter among these that is not affected by weight but by the width of the aorta, studies so far have pointed that since left atrium is a three-dimensional structure LA/Ao alone is not an accurate way to assess left atrial size. Reference intervals for mitral inflow E-wave velocity are higher than previously reported. This could be explained with overall greater amount of physical activity in Belgian Malinois breed and enhanced diastolic function. Tissue Doppler imaging measurements in Belgian Malinois show supranormal S- and E'-wave velocities when compared with those published in other studies. The supranormal velocities could reflect enhanced global longitudinal systolic and early diastolic function in this exceptionally active dog breed. There is even a statistically significant difference in tissue Doppler E'-wave between males and females, with males having greater values than females. The aforementioned difference in E’ wave velocity could be explained with a higher percentage of working dogs among males than females. EPSS measurement is in concordance with the one established in Doberman breed and lower than that established for the general dog population. Peak aortic and pulmonic velocities are similar to reference intervals for general population of dogs. Reference intervals for cardiac biomarkers, cTnI and NT-proBNP have been established with mean NT-proBNP values that are two times higher and upper reference limit almost five times higher than the ones proposed for the general dog population. Studies so far have pointed that NT-proBNP values differ among breeds and that breed-specific reference intervals should be established. Significant differences have been reported in this study between echocardiographic parameters of heart size and function between working and pet dogs. Although there was no statistically significant difference between linear measurements of the left ventricle, left ventricular volumes are significantly higher in working dogs than in pets. This is in agreement with changes described for athlete's heart. It is also an additional proof showing that volumetric method is more accurate and should be used instead of linear measurements since it is a more sensitive method for the assessment of left ventricular size and function. Interestingly, EPSS was also statistically significantly greater in working dogs than pets. This is also in agreement with the results of a study in which left ventricular volumes and EPSS changes were present before changes in linear measurements in dogs with dilatative cardiomyopathy. Left atrial volume and diameter were also statistically significantly larger in working dogs than pets, which is in concordance with changes reported in professional athletes and studies done in dogs with degenerative mitral valve disease where volumetric measurements of left atrial size have been superior to LA/Ao in identification of mild left atrial enlargement. Since LA volumes are within published reference limits in both groups, left atrial enlargement in working dogs is also considered mild. When comparing the Teicholz method and Simpson’s method of discs for left ventricular volume calculation it has been shown that in Belgian Malinois there is a statistically significant difference between these two methods. Namely, volumes calculated by Teicholz method were larger than the ones calculated using Simpson’s method of discs. It can be stated that the Teicholz method is not recommendable for the volume calculation in Belgian Malinois breed. Finally, comparison of generated reference intervals for the Belgian Malinois with those published for the German Shepherd dog has been done. Since there are no published volumetric reference values for the German Shepherd dog, only linear measurements have been compared. The comparison showed that most of the linear measurements (excluding the left ventricular internal diameter in diastole, and thickness of left ventricular free wall in diastole) were larger in German Shepherds when compared to those in Belgian Malinois, probably because German Shepherd are by 10% heavier than Belgian Malinois dogs. It can be concluded that in this study echocardiographic reference intervals of cardiac size and function and cardiac biomarkers (cTnI and NT-proBNP) for the Belgian Malinois breed have been established. Furthermore, statistically significant difference has been established in several echocardiographic parameters regarding the degree of physical activity, sex, as well as correlation in regard to age and body weight.