Measurement of M-mode echocardiographic parameters in healthy adult Maine Coon cats

OBJECTIVE: To determine reference values for M-mode echocardiographic parameters in nonsedated healthy adult Maine Coon cats and compare those values with data reported for nonsedated healthy adult domestic cats. DESIGN: Prospective study. ANIMALS: 105 healthy adult Maine Coon cats. PROCEDURE: Over a 3-year period, M-mode echocardiographic examinations (involving a standard right parasternal transthoracic technique) were performed on Maine Coon cats as part of prebreeding evaluations; values of M-mode parameters in healthy individuals were collected, and mean values were calculated for comparison with those reported for healthy adult domestic cats. RESULTS: The mean +/- SD weight of Maine Coon cats was significantly greater than that of domestic cats. Mean values of left ventricular internal dimension at end diastole and end systole (LVIDd and LVIDs, respectively), interventricular septal thickness at end systole (IVSs), left ventricular posterior wall thickness at end systole (LVPWs), left atrial dimension at end systole (LADs), and aortic root dimension (Ao) in Maine Coon cats differed significantly from values in healthy domestic cats. The greatest differences detected between the 2 groups involved values of LVIDd, LADs, and Ao. Linear regression analysis revealed a weak but significant correlation between weight and each of LVIDd, LVPWs, IVSs, Ao, LADs, and left ventricular posterior wall thickness at end diastole. CONCLUSIONS AND CLINICAL RELEVANCE: Values of several M-mode echocardiographic parameters in Maine Coon cats differ from those reported for domestic cats; these differences should be considered during interpretation of echocardiographic findings to distinguish between cardiac health and disease in this breed.     

Evaluation of four 2-dimensional echocardiographic methods of assessing left atrial size in dogs

Two-dimensional (2D) echocardiography is the cornerstone of noninvasive evaluation of the cardiac patient, and often involves estimating left atrial (LA) size. However, 2D echocardiographic methods of estimating LA size have been inadequately described, and most reference intervals are based on M-mode echocardiographic measurements. We determined reference intervals for 4 different 2D echocardiographic methods of estimating LA size in adult (> or =9-month-old) dogs without cardiovascular disease. Thirty-six dogs, placed in right lateral recumbency, were examined by 2D echocardiography. The left atrium was measured at specific time points in the cardiac cycle. Measurement methods were LA diameter in short axis, LA diameter in long axis, LA circumference in short axis, and LA cross-sectional area in short axis. Comparisons of these LA dimensions to appropriate aortic dimensions provided body weight-independent estimates of LA size. We found strong associations of LA dimensions with body weight (r2 = .76-.88). Comparable body weight-independent 2D echocardiographic estimates of LA size in short axis exceeded historical M-mode reference intervals. These data provide echocardiographers with reference intervals for 2D echocardiographic estimates of LA size in adult dogs.     

Two-dimensional echocardiographic assessment of the feline left atrium

Atrial size determined by echocardiography provides a surrogate measure of the hemodynamic burden of cardiac disease. Linear atrial dimensions often are indexed to aortic diameter. Whereas quantitative variables obtained from healthy cats, using 2-dimensional echocardiography (2DE), have been reported, indices from 2DE, have not. Using 2DE and M-mode echocardiography, we calculated indices of left atrial size and a single index of atrial function, left atrial fractional shortening, in 17 healthy cats. Specifically, left atrial dimensions from short- and long-axis 2DE planes were indexed to aortic diameter and also to end-diastolic left ventricular dimension. Additionally, left atrial circumference and area were indexed to aortic circumference and area, respectively. The same variables were obtained from 20 cats with hypertrophic cardiomyopathy (HCM), so that agreement between 2DE indices and indices from M-mode echocardiography could be evaluated over a clinically relevant range of atrial sizes. Atrial dimensions and indices of atrial size from cats with HCM exceeded those of healthy cats. Left atrial dimension from 2D short-axis images indexed to aortic diameter generally was less than the analogous index obtained from M-mode (mean bias, [95% limits of agreement] -0.13, [-0.42, 0.17]). Left atrial dimension from 2D long-axis images indexed to aortic diameter generally was greater than the index obtained from M-mode (0.15, [-0.28, 0.58]). We conclude that ratios of left atrial size and aortic diameter, from 2DE and M-mode echocardiography, are not interchangeable. Normative data that may serve as reference intervals for 2DE assessment of atrial size are presented.     

M-MODE ECHOCARDIOGRAPHY AS A DIAGNOSTIC AID FOR FELINE CARDIOMYOPATHY

Normal cats and cats with congestive cardiomyopathy (CCM) and hypertrophic cardiomyopathy (HCM) were examined using M-mode echocardiography to determine its diagnostic capabilities. Sixteen normal cats were examined to verify previously reported data and to add further echocardiographic inforamtion (left atrial/aortic root ratio, left posterior wall thickness at end systole and end diastole, amplitude of mitral valve excursions, and velocity of valve opening and closure) to aid in differential diagnosis. Significant (p<0.05) changes were detected between the normal cats and those with cardiomyopathy. In each type of cardiomyopathy, alterations in left atrial dimension, left atrial/aortic root ratio, left ventricular dimension, left ventricular wall thickness and percentage of ventricular dimensional change were identified. Altered mitral valve motion was found with HCM. Echocardiography was found to be an accurate technique for definitive diagnosis of feline cardiomyopathy.     

M-mode echocardiographic measurements in nonanesthetized healthy cats: effects of body weight, heart rate, and other variables

Body weight, heart rate, and 19 M-mode echocardiographic variables were measured in 41 nonanesthetized healthy cats. Estimated limits were determined for the echocardiographic variables, and each variable was then correlated to body weight, heart rate, and the 18 other variables. A significant (P less than 0.05) positive correlation to body weight was found with aortic diameter, left atrial dimension, septal and left ventricular systolic and left ventricular diastolic wall thicknesses, and left and right ventricular diastolic and right ventricular systolic internal dimensions. Significant inverse correlation (P less than 0.05) to heart rate was found with body weight, left ventricular systolic and diastolic and right ventricular systolic internal dimensions, left atrial dimension, left atrial dimension to aortic ratio, mitral valve E point to ventricular septal separation, and left ventricular ejection time. Left ventricular shortening fraction in the short axis and velocity of circumferential fiber shortening were significantly correlated (P less than 0.05) to heart rate. Significant correlation (P less than 0.05) was also found between many echocardiographic variables.     

Two-dimensional and M-mode echocardiographic reference values in healthy adult Saanen goats

Echocardiography has become a routine non-invasive cardiac diagnostic tool in most species. Accurate measurement of cardiac dimensions requires reference values, which are poorly documented in goats. The aim of the present study was to test the inter-day repeatability and to establish the reference values of two-dimensional (2D-) and time-motion (M-) mode echocardiographic variables in healthy adult Saanen goats. Six goats were investigated three times by the same observer at one-day interval using a standardised 2D- and M-mode echocardiographic protocol. The intra-observer inter-day repeatability was tested using analysis of variance, calculation of the coefficient of variation and confidence intervals. A single echocardiographic examination was performed in six other goats, and values obtained in the 12 goats were used to establish the 2D- and M-mode echocardiographic reference values in healthy adult female Saanen goats. Statistical analysis revealed a good inter-day repeatability of the echocardiographic cardiac measurements. Echocardiographic reference values obtained in healthy adult Saanen goats seemed slightly higher than those reported in healthy Swedish domestic goats and were similar to those reported in healthy adult sheep.     

Assessment of left atrial appendage flow velocity and its relation to spontaneous echocardiographic contrast in 89 cats with myocardial disease

The hypotheses of this prospective study were that (1) left atrial appendage (LAA) blood flow velocities can be recorded in cats with myocardial disease by transthoracic Doppler echocardiography, (2) LA enlargement, LA mechanical dysfunction, and left ventricular (LV) diastolic abnormalities are associated with decreased LAA flow velocities, and (3) low LAA flow velocities predict the appearance of spontaneous echocardiographic contrast in cats with cardiomyopathy. Transthoracic 2-dimensional, M-mode, and Doppler echocardiographic studies were performed in 89 cats with hypertrophic, restrictive, dilated, or unclassified cardiomyopathy or with hyperthyroid heart disease. Maximal LAA flow velocity (LAAmax) was decreased (P < .001) in cats with cardiomyopathy (median, 0.28 m/s; range, 0.08-1.35) compared to normal cats. Associated with decreased LAA flow velocities were increased LA size, decreased LA function, increased severity of LV diastolic dysfunction, and the presence of congestive heart failure. Multivariate logistic regression analysis detected an LAAmax <0.20 m/s as the only independent variable to predict LA spontaneous echocardiographic contrast (odds ratio, 30.1; 95% confidence interval [CI], 4.1 222.3; P < .001). Receiver operating characteristic analysis performed to predict spontaneous echocardiographic contrast indicated an area under the curve of 0.88 (95% CI, 0.80-0.95; P < .001) with sensitivities of 100 and 74% and specificities of 69 and 83% for LAAmax <0.25 and <0.20 m/s, respectively. Thus, low LAA flow velocities identified a subgroup of patients at increased risk of spontaneous echocardiographic contrast and possible thromboembolism. These findings may have important clinical implications for anticoagulation therapy and prognostication in cats with cardiomyopathy.     

ASSESSMENT OF CARDIAC CHAMBER SIZE USING ANATOMIC M-MODE

Anatomic M-mode (AMM) is an echocardiographic technique that is capable of generating M-mode studies from two-dimensional (2D) cine loops. Unlike conventional M-mode (CMM) whose scan line must lie along the axis of the ultrasound signal, AMM produces M-mode studies independent of the orientation of the ultrasound beam. We sought to determine the ability of AMM to measure cardiac dimensions in normal dogs and to assess the accuracy and variability of AMM and CMM vs. 2D measurements. Thirty-eight healthy dogs underwent physical exam and 2D, CMM, and AMM echocardiographic studies. The end-diastolic and end-systolic dimension of the left ventricle and the diameter of the left atrium (LAD) and aortic root were evaluated from the right parasternal short- and long-axis views. Results of the AMM and CMM study were compared with the 2D study via linear regression and calculation of a coefficient of correlation. AMM increased the level of correlation with both the left ventricular dimensions and LAD. Bland-Altman analysis revealed that AMM increased the level of agreement with 2D measurements and CMM greatly underestimated LAD vs. AMM. In healthy dogs, cardiac AMM measurements are associated with greater accuracy and less variability than CMM. AMM has the potential to improve quantification of cardiac dimensions.     

Meta-analysis of normal canine echocardiographic dimensional data using ratio indices

ObjectivesTo investigate the dependence of echocardiographic ratio indices (ERIs) on age, body weight (BW) and breed/study group using individually contributed and published summarized data in dogs.BackgroundERIs allow for narrow prediction intervals of M-mode echocardiographic measurements in generic adult dogs. Breed and age-specific differences have not been examined systematically using ERI methods.Animals, materials and methodsIndividual M-mode measurements were contributed by 15 published investigators from 661 dogs, allowing direct calculation of ERIs and summary statistics for each of these breed/study groups. M-mode ERI summary statistics were estimated from published summaries of 22 additional groups that included 527 adult and 36 growing dogs. Individual two-dimensional (2DE) left atrial (LA) and aortic root (Ao) measurements were contributed from 36 dogs. ERIs were analyzed for dependence on BW, breed/study group and age.ResultsThe majority of variation among ERIs was due to differences in the breed or study technique with comparatively little dependence on BW. Age dependence of ERIs was seen in the early growth phases of young dogs, but expected values for each ERI became static long before maturity, roughly at 10–12 weeks of age. ERIs derived from individual 2DE LA and Ao measurements showed no significant dependence on BW.ConclusionsERIs are well normalized for body size and may be useful for clinical evaluation of individuals, prediction of expected M-mode and 2DE cardiac dimensions, and investigation of age or breed-specific cardiac shape changes.     

Comparison of the Repeatability of Echocardiographic Measurements from Different Modes and Views in Horses of Various Breeds and Sizes

The objective of this study was to compare the repeatability of echocardiographic measurements obtained from different echocardiographic modes and views in healthy adult equids of various sizes, breeds, and thorax shapes. Ten equids (body weight: 120–662 kg; age: 1–26 years) from various breeds, free of cardiac disease, were used in this study. Each animal was submitted to a standardized echocardiographic and Doppler protocol 3 times at 1 day interval. This protocol included the measurements of left and right ventricular, aortic, pulmonary, and left atrial parameters obtained from different views using the bidimensional (2D) or the motion (M) modes, and the measurement of several parameters of blood flow obtained from the pulsed wave Doppler mode. Repeatability of each measurement was estimated on the basis of the residual variance using a linear model and the coefficient of variation of repeated measurements. A two by two comparison of the repeatability of measurements performed in different views was performed using the residual variances in a variance ratio F test. Results showed that repeatability of echocardiographic or Doppler measurements in equids of various sizes, breeds, and thorax shapes are comparable to previously reported results in thoroughbred and standardbred horses. Left ventricular morphologic parameters showed a good repeatability in the classic M-mode right parasternal short axis view at the level of the chordae tendineae, but the 2D-mode right parasternal long axis four-chamber view appeared to offer an interesting alternative measurement. This latter view also allowed obtaining the most repeatable measurement of right ventricular internal diameter. The left atrial diameter was maximal and most repeatable in the 2D-mode left parasternal long axis four-chamber angled view, and the repeatability of the aortic diameter was best in the 2D-mode right parasternal long axis five-chamber view. Finally, aortic systolic time intervals were more repeatable when measured from the Doppler mode as compared with the M-mode. In conclusion, repeatability of echocardiographic measurements in horses could be optimized after the following protocol: (1) M-mode right parasternal short axis view at the level of chordae tendineae to measure left ventricular morphologic parameters, and 2D-mode right parasternal long axis four-chamber as an alternative view; (2) 2D-mode right parasternal long axis five-chambers to measure the aortic diameter; (3) 2D-mode left parasternal long axis four-chambers angled view to measure the left atrial diameter.     

Normal values of the canine M-mode echocardiogram

M-mode echocardiograms were recorded from 40 healthy dogs, awake and unsedated, in left lateral recumbent position. Fifteen echocardiographic measurements were taken and correlated with body weights using linear regression equations. The left ventricular internal dimension in systole and diastole, the left ventricular wall thickness, the aortic root dimension, and the left atrial dimension had high correlation coefficients (r2), ranging from 0.756 to 0.619. The fractional shortening of the left ventricle in systole (39% +/- 6%) and the left atrial to aortic root ratio (0.99 +/- 0.10) were not linearly related to body weights and had constant values.     

Echocardiographic reference values in healthy cats sedated with ketamine hydrochloride

An M-mode echocardiographic examination was performed in a consistent manner in 30 clinically healthy cats under light ketamine hydrochloride sedation. There was a significant linear relationship between increasing body size and increasing cardiac dimensions for several echocardiographic values. Positive correlation existed between body weight and body surface area with aortic root, left ventricular caudal wall thickness (LVCW), interventricular septal thickness (IVS), IVS/LVCW, and mean velocity of circumferential fiber shortening (Vcf); there was a negative correlation between body weight and body surface area with left ventricular ejection time (LVET). Body surface area also correlated positively with percentage of ventricular minor axis dimensional change (% delta D). Positive correlations were recorded between left ventricular end-diastolic dimension (LVEDD) and left ventricular endsystolic dimension (LVESD), LVESD and LVET, LVCW and IVS, LVET (calculated by LVCW motion) and LVET (calculated by aortic valve motion), % delta D and Vcf, heart rate and Vcf, and Vcf (calculated using aortic valve motion to compute LVET) and Vcf (using LVCW motion to compute LVET). There were negative correlations between LVEDD and % delta D, LVEDD and Vcf, LVESD and Vcf, LVET and Vcf, LVET and heart rate, LVET and % delta D. Significant differences were recorded between means of echocardiographic reference values generated in this and other studies, except for LVESD.     

Two-dimensional and M-mode echocardiographic measurements of cardiac dimensions in healthy standardbred trotters

The aim of the study was to establish normal echocardiographic values of healthy Standardbred trotters not published previously. Twenty-three clinically normal horses weighing between 350 and 490 kg were examined in the same manner: first a thorough physical and then detailed echocardiographic examination were performed. Standardised two-dimensional (2D) and guided M-mode echocardiographic imaging techniques were used to measure interventricular septal thickness (IVS), left ventricular internal diameter (LVID), left ventricular wall thickness (LVW), left atrial internal diameter (LAID) in end-systole (s) and end-diastole (d) and aortic diameter (AOD) in end-diastole. Mean, range and standard deviation of the different parameters were calculated. The mean values (in centimetres) were as follow (2D/M-mode): IVSs: 4.6/4.7; IVSd: 3.1/3.0; LVIDs: 7.0/7.0; LVIDd: 10.7/10.7; LVWs: 3.9/3.9; LVWd: 2.7/2.7; LAIDs: 10.4/-; LAIDd: 11.3/-; AODd: 7.2/-. Results of two-dimensional and M-mode measurements were compared to each other and to normal values obtained from other breeds.     

Echocardiography in the chinchilla

Heart disease has been described in the chinchilla and, with increasing popularity as a pet, the demand for diagnostic evaluation and treatment has increased. The goal of this study was to determine reference values for echocardiographic measurements in chinchillas and the effect of anesthesia on these measurements. Seventeen clinically healthy adult chinchillas were studied. All animals were anesthetized with isoflurane by mask. Standard echocardiographic views were used. A difference was seen in the echocardiographic measurements for left ventricular systolic dimension, fractional shortening, aortic (Ao) diameter, left atrial (LA) diameter, ratio of LA diameter to Ao diameter, and peak flow velocities and ejection times for Ao and pulmonary artery flows between awake and anesthetized chinchillas.     

The effect of hydration status on the echocardiographic measurements of normal cats

BACKGROUND: Diagnosis of cardiomyopathy of cats is based on 2-dimensional (2D) echocardiography. However, circulating fluid volume largely determines diastolic cardiac chamber dimensions, and reduced diastolic volume in other species results in what has been called "pseudohypertrophy of the ventricular myocardium." HYPOTHESIS: Altered hydration produces changes on 2D echocardiography that may confound the diagnosis or severity assessment of cardiomyopathy of cats. ANIMALS: Ten normal colony-sourced mixed breed cats were included. METHODS: Cats were examined by echocardiography at baseline and at completion of 3 protocols (volume depletion and maintenance-rate and anesthetic-rate IV fluid administration) applied in randomized crossover design with a 6-7 day washout period. RESULTS: Volume depletion increased diastolic left ventricular interventricular septal (IVSd) and free wall diameter (4.5 +/- 0.4 to 5.8 +/- 0.6 mm; P < .001) with wall thickness exceeding 6 mm in 4 cats. Diastolic left ventricular internal diameter (LVIDd) decreased, and reduction in systolic left ventricular internal diameter (LVIDs) produced end-systolic cavity obliteration in 7 cats. Left-atrial-to-aortic-root ratio (LA: Ao, 1.4 +/- 0.2 to 1.2 +/- 0.1, P < .05) and left atrial area in diastole (LAAd) decreased with volume depletion. Maintenance-rate IV fluid administration increased LAAd and fractional shortening (FS%). Anesthetic-rate IV fluid administration increased LVIDd, FS%, LAAd, and LA:Ao ratios (to 1.7 +/- 0.1, P < .01), producing an LA: Ao ratio above normal limits in 6 cats. A systolic heart murmur developed with administration of fluid at maintenance (n = 1) and anesthetic rates (n = 6). CONCLUSIONS: Altered hydration status produces changes in the echocardiographic examination of normal cats that may lead to an erroneous diagnosis of cardiomyopathy or mask its presence. Hydration status should be considered during echocardiographic examination in cats.     

Echocardiographic findings in 103 cats with hyperthyroidism

Using M-mode echocardiography, cardiac abnormalities were studied in 103 cats with untreated hyperthyroidism. In addition, follow-up echocardiography was performed on 24 of these cats to assess the long-term (4 to 21 months) effect of treatment (thyroidectomy or radioiodine) on thyrotoxic cardiac disease. The most common echocardiographic abnormality in the 103 untreated hyperthyroid cats was hypertrophy of the left ventricular caudal wall (71.9%). Hypertrophy of the interventricular septum also was documented in 39.8% of the 103 cats. Other abnormalities included high values for left atrial diameter (70.0%), aortic root diameter (18.5%), and left ventricular diameter at end diastole (45.6%). In some of these cats, indices of contractility were enhanced; in 21.4% and 14.6% of the cats, values for shortening fraction and velocity of circumferential fiber shortening, respectively, were greater than those values measured in clinically normal cats. After treatment of the hyperthyroidism, left ventricular hypertrophy resolved or improved in many of the cats, as indicated by decreases in left ventricular caudal wall and interventricular septum thicknesses. Hyperdynamic wall motion resolved in all cats after treatment, as evidenced by consistent decreases in shortening fraction and velocity of circumferential fiber shortening. Despite these improvements, some cats had one or more persistently abnormal echocardiographic values after treatment. These results suggested that in cats, hyperthyroidism commonly is associated with largely reversible cardiomyopathy. In those cats in which cardiomyopathy persists or worsens after treatment, underlying primary cardiomyopathy or thyroid hormone-induced cardiac structural damage may exist.     

Spectrum of M-mode echocardiographic abnormalities in 75 cats with systemic hypertension

A retrospective study was performed in 75 hypertensive cats to determine the spectrum and frequency of M-mode echocardiographic abnormalities. Results indicated that 21.3% of the cats had M-mode measurements within normal reference ranges. For cats with echocardiographic abnormalities, changes were variable. Thirty-nine percent of hypertensive cats had hypertrophy of the interventricular septum in diastole, and 41.3% had hypertrophy of the left ventricular (LV) posterior wall in diastole. One cat in five had a dilated left atrium, while fractional shortening and LV internal dimension in diastole were normal in 82.7% and 86.7% of the cats, respectively. The marked variability of echocardiographic findings in hypertensive cats made echocardiography an unreliable screening test for hypertension.     

Two-dimensional,long-axis echocardiographic ratios for assessment of left atrial and ventricular size in dogs

Introduction

Left ventricular (LV) and left atrial (LA) enlargement affect management and outcome of dogs with cardiac disease. Short-axis, two-dimensional echocardiographic (2DE) images, indexed to the aorta (Ao), are frequently used to identify cardiomegaly. Long-axis images offer complementary views of the left heart.

Prospective echocardiographic and tissue Doppler screening of a large Sphynx cat population: Reference ranges,heart disease prevalence and genetic aspects

Objectives(1) To investigate heart morphology and function using echocardiography and tissue Doppler imaging (TDI), (2) to determine heart disease prevalence and characteristics, and (3) to assess potential genetic features in a population of Sphynx cats presented for cardiovascular screening.AnimalsA total of 147 echocardiographic examinations, including 33 follow-ups, were performed by trained observers on 114 Sphynx cats of different ages (2.62 ± 1.93 years [0.5–10.0]) from 2004 to 2011.MethodsSphynx cats underwent a physical examination, conventional echocardiography, and, if possible, two-dimensional color TDI.ResultsConventional echocardiographic findings included 75/114 normal (65.8%) and 39/114 (34.2%) abnormal examinations with a diagnosis of either congenital heart diseases (n = 16) or hypertrophic cardiomyopathy (HCM, n = 23). In adult healthy cats, a significant body weight effect was observed for several echocardiographic variables, including end-diastolic left ventricular (LV) free wall (P < 0.01), interventricular septum (P < 0.001), and LV diameter (P < 0.001). Mitral valve dysplasia (MVD) was observed as a single or associated defect in 15/16 cats with congenital heart diseases. A significant increase in HCM prevalence (P < 0.001) was observed according to age. The pedigree analysis of a large family (n = 81) suggested an autosomal dominant mode of inheritance with incomplete penetrance for HCM.ConclusionsBody weight should be taken into account when interpreting values of diastolic myocardial wall thicknesses in Sphynx cats. Additionally, HCM and MVD are two relatively common heart diseases in this feline breed. More pedigree data are required to confirm the inheritance pattern of HCM at the breed level.     

RADIOGRAPHIC AND ECHOCARDIOGRAPHIC ASSESSMENT OF LEFT ATRIAL SIZE IN 100 CATS WITH ACUTE LEFT‐SIDED CONGESTIVE HEART FAILURE

The aims of this study were to evaluate left atrial size in cats with acute left‐sided congestive heart failure. We hypothesized that left atrial size as determined by thoracic radiography can be normal in cats with acute left‐sided congestive heart failure. One hundred cats with acute left‐sided congestive heart failure in which thoracic radiography and echocardiography were performed within 12 h were identified. Left atrial size was evaluated using right lateral and ventrodorsal radiographs. Measurements were compared to two‐dimensional echocardiographic variables of left atrial size and left ventricular size. On echocardiography, left atrial enlargement was observed in 96% cats (subjective assessment) whereas maximum left atrial dimension was increased (>15.7 mm) in 93% cats. On radiographs left atrial enlargement (subjective assessment) was found in 48% (lateral view), 53% (ventrodorsal view), and 64% (any view) of cats whereas left atrial enlargement was absent in 36% of cats in both views. Agreement between both methods of left atrial size estimation was poor (Cohen's kappa 0.17). Receiver operating characteristic curve analysis identified a maximum echocardiographic left atrial dimension of approximately 20 mm as the best compromise (Youden index) between sensitivity and specificity in the prediction of radiographic left atrial enlargement. Left atrial enlargement as assessed by thoracic radiography may be absent in a clinically relevant number of cats with congestive heart failure. Therefore, normal left atrial size on thoracic radiographs does not rule out presence of left‐sided congestive heart failure in cats with clinical signs of respiratory distress.