Radiographic quantification of left atrial size in dogs with myxomatous mitral valve disease

Abstract Background In the absence of echocardiography, identification of cardiomegaly via thoracic radiography is a necessary criterion for classification of disease severity in dogs with myxomatous mitral valve disease (MMVD). Objective Modified‐vertebral left atrial size (M‐VLAS) facilitates objective radiographic assessment of the left atrium (LA) in 2 dimensions and identifies LA enlargement more accurately than existing methods. Animals Sixty‐four dogs with various stages of MMVD and 6 control healthy dogs. Methods Retrospective case–control study. Medical records were searched for dogs with varying severity of MMVD. Modified‐vertebral left atrial size, vertebral left atrial size (VLAS), vertebral heart size (VHS), and radiographic left atrial dimension (RLAD) were measured from thoracic radiographs and compared with echocardiographically derived measurements. Results Positive correlation to LA/Ao was identified for M‐VLAS (r = 0.77, P < .001), VLAS (r = 0.76, P < .001), RLAD (r = 0.75, P < .001), and VHS (r = 0.67, P < .001). Receiver operating characteristic analyzes provided an area under the curve of 0.97 (95% CI, 0.94‐1.00) for M‐VLAS, which was superior to VHS (0.90, 95% CI, 0.94‐1.00, P = .03) in identifying dogs with LA/Ao ≥1.6. A cut‐off value of ≥3.4 vertebrae using M‐VLAS provided 92.7% sensitivity and 93.1% specificity in predicting LA enlargement. Conclusions and clinical importance M‐VLAS, which is superior to VHS, offers an accurate and repeatable way to radiographically identify LA enlargement in dogs with MMVD.

disease progresses, [3][4][5][6] and when eccentric capacity is reached, the subsequent rise in left atrial (LA) pressure creates an increase in pulmonary capillary pressure to precipitate pulmonary edema and congestion. [7][8][9] As such, LA enlargement is well-established as a precursor to congestive failure and a strong prognostic indicator for dogs with MMVD. 2,[9][10][11][12] Pimobendan therapy delays the onset of CHF and reduces cardiac deaths in dogs with stage B2 MMVD. 13 Although echocardiography is recommended for definitive diagnosis and disease staging of MMVD in dogs, it is not necessarily available nor affordable to every dog, especially in the general practice setting. As such, the recently updated American College of Veterinary Internal Medicine (ACVIM) consensus statement for MMVD recommends the use of thoracic radiography, namely the vertebral heart size (VHS) and vertebral left atrial size (VLAS), to identify dogs with stage B2 MMVD when echocardiography is unavailable to identify the cardiac enlargement by which stage B2 is defined. 14 Various limitations including breed-associated variations, [15][16][17][18][19][20] interobserver variability 21,22 , and the influence of respiratory and cardiac cycle 23,24 are associated with the objective radiographic assessment of global cardiac size with VHS. Vertebral left atrial size 25 and radiographic left atrial dimension (RLAD) 26 are described as methods to specifically assess LA size via thoracic radiography in dogs with MMVD using a single-dimensional measurement. Both methods are positively correlated with echocardiographically-derived left atrial-to-aortic root ratio (LA/Ao). A VLAS of ≥2.5 vertebrae is 87% specific and 67% sensitive in predicting LA/Ao ≥1.6, 25 whereas an RLAD of ≥1.8 vertebrae is 93.5% specific and 96.8% sensitive in identifying the same LA/Ao. 26 The simplicity of the VLAS method is countered by its relatively poor specificity and sensitivity, whereas the superior accuracy of the RLAD measure is offset by the complexity of its acquisition. We propose a new and simplistic method for radiographic LA measurement will be superior to previous methods by assessment of LA size in 2 dimensions.
The objectives of this study are therefore (a) to compare the accuracy of VLAS, VHS, and RLAD at identifying LA enlargement (LA/Ao ≥1.6) in dogs with various stages of MMVD and (b) to introduce and compare a modified-VLAS (M-VLAS) measurement to existing radiographic methods for identification of LA enlargement. We hypothesize that M-VLAS is superior to published methods in radiographic identification of LA enlargement given its additional dimension of LA measurement.

| MATERIALS AND METHODS
Electronic hospital records of dogs diagnosed with MMVD between 3 hospital sites of Veterinary Specialist Services (Underwood, Jindalee and Carrara) by the cardiology service from October 2017 to March 2020 were retrospectively reviewed. Dogs that were < 20 kg, with physical examination, 2 or 3 view thoracic radiographs (dorsoventral and right +/− left lateral), and echocardiogram performed within a 24-hour period were included in the study. Echocardiograms were performed by either a board-certified cardiologist (FEM/BJG) or a resident (CL) under their direct supervision. Echocardiographic measurements were taken from established 2-dimensional (2D) views, 27 and the LA/Ao measured from a standard right parasternal short-axis view. 28 Left ventricular internal diameter in diastole (LVIDdN) was normalized to individual body weight. 29 The diagnosis of MMVD was made based on clinical findings of a left apical systolic heart murmur combined with characteristic morphological mitral valve changes and valvular regurgitation evident on echocardiogram. 4 Classification of MMVD disease stages was made according to the ACVIM consensus guidelines. 14 Dogs that were < 1-year old, dogs with acute chordae tendinae rupture, or concurrent cardiac disease were excluded from the study.
Dogs with tricuspid regurgitation with color Doppler regurgitant jet area < 50% of the atrial area and without right heart remodeling were not excluded from the study. Dogs with thoracic radiographs that were malpositioned or where exposure factors precluded clear assessment of thoracic vertebrae were also excluded. Dogs were included as controls if they were < 20 kg, with no heart murmur on physical exam, no cardiac abnormalities detected on echocardiogram, and had at least 2-view thoracic radiographs (dorsoventral and right lateral) taken within the same 24-hour period as echocardiographic assessment. Baseline data retrieved from medical records included sex, age, breed, bodyweight, and medications administered at the time of examination.
A right lateral inspiratory thoracic radiograph from each dog was used for all radiographic measurements. Radiographic images were reviewed and measured using a commercial digital radiography view-   ( Figure 2).

| DISCUSSION
The objectives of this study were to examine and compare different radiographic methods for identification of echocardiographically mea- The original VLAS method aimed to quantify LA size radiographically based on a single-dimensional measurement that roughly represented the transverse diameter of the LA body. Yet, LA functional anatomy and geometry is complex, with the left atrium described as a cylinder with an almost fixed head and distensible walls attached to a piston (ie, mitral annulus). 32 The process of LA remodeling is equally complex and non-uniform. 33 Recognizing the limitations of only using a single 2D lateral radiograph to quantify the complex LA anatomy,  One of the most important findings of the current study is the superiority of M-VLAS to identify LA enlargement compared to VHS. provided high specificity (97%) at the expense of low sensitivity (78%).
Multiple publications have evaluated the radiographic method to objectively assess heart size by VHS since its introduction in 1995. 37 The limitations of VHS include multiple breed-associated variations, 15-20 interobserver variability, 21,22 and the influence of respiratory and cardiac cycles. 23,24 While respiratory cycle can be controlled when radiographs are taken, cardiac cycle cannot. The mean VHS ± SD ranges from 9.9 ± 0.8 to 10.4 ± 0.8 vertebrae between end-diastolic and end-systolic measurements at peak inspiration for dogs positioned in right lateral recumbency. 24  The current study has several limitations. Firstly, because of the retrospective design sample sizes were small especially in the healthy group (n = 6). However, the aim of this study was not to differentiate between healthy and diseased dogs nor to establish normal reference ranges, but rather to assess the efficacy of objective radiographic measurements to identify LA enlargement (LA/Ao ≥1.6), which together with other cardiac criteria, is an indication for pimobendan therapy. 14 Being a retrospective study, dog recruitment also could not be strictly controlled, and echocardiographic measurements might be subjected to interobserver variability. To reduce the possibility of misclassification, all dogs were assessed and classified by board-certified cardiologists or a resident under direct supervision by board-certified cardiologists. In addition, all dogs with heart failure were classified by documentation of clinical signs of dyspnea, radiographic evidence of pulmonary edema, and positive response to diuretic therapy. Secondly, the presence of concurrent systemic disease and medical treatment were not controlled and therefore any secondary change in intravascular volume between when thoracic radiographs and echocardiogram were performed could contribute to discrepancies in cardiac size. However, our inclusion criteria were limited to dogs with echocardiography and thoracic radiographs performed within a 24-hour period to minimize treatment effects and changes in intravascular volume. Only 1 dog with stage B1 MMVD from the cohort received fluid therapy intravenously during assessment for treatment of pancreatitis. Thirdly, we recognize the limitations of 2D echocardiography in accurately quantifying LA size and volume. 38,39 Finally, because of its retrospective nature, cases were recruited across 3 Veterinary Specialist Services hospital locations. Individual radiology unit and exposure factors used for each case were consequently different which has the potential to reduce consistency. The current study also included a variety of breeds. The effect of breed variability to the outcome of the various radiographic measurements (especially VLAS, RLAD, and M-VLAS) is unknown and will likely be the target for future studies.

ACKNOWLEDGMENT
No funding was received for this study. The authors thank the staff from Veterinary Specialist Services for assistance and support during this study.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.