- Email: [email protected]
Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs
Journal of Veterinary Cardiology (2014)
-, -e-
www.elsevier.com/locate/jvc
Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs Faisal A. Torad, BVSc, MVSc, PhD , Elham A. Hassan, BVSc, MVSc, PhD* Department of Surgery, Anesthesiology and Radiology, Faculty of Veterinary Medicine, Cairo University, Giza Square P.O. 12211, Cairo, Egypt Received 14 November 2013; received in revised form 8 August 2014; accepted 18 August 2014
KEYWORDS Radiography; Heart; Cardiomegaly; Microcardia; Canine
Abstract Objectives: To evaluate cardiac size in normal German shepherd dogs (GSD) using the two-dimensional cardiothoracic ratio (CTR) and to use this measure for diagnosing GSD with altered cardiac size. Animals: One hundred clinically normal GSD as well as 46 GSD with altered cardiac size (microcardia or cardiomegaly). Methods: The CTR was computed as the percentage area of the cardiac silhouette relative to the area of the dog’s thorax. Measurements were performed using a digital software program on lateral and ventro-dorsal radiographs at the points of peak inspiration and expiration. Receiver operating characteristic (ROC) curve analysis was used to determine the diagnostic accuracy of the CTR for diagnosing cardiomegaly or microcardia. Results: The mean ( SD) CTR on lateral radiographs of normal dogs was 27.60% 1.10% and 30.13% 1.42% at the points of peak inspiration and expiration, respectively. For ventro-dorsal radiographs, mean CTR was 30.45% 1.39% at peak inspiration and 33.34% 1.46% at peak expiration. The cutoff value of the CTR for diagnosing microcardia on lateral radiographs was 22.98% (inspiration) and 25.06% (expiration), compared to 25.03% (inspiration) and 23.97% (expiration) on ventrodorsal radiographs. Cutoff values for diagnosing cardiomegaly were 30.28% (inspiration) and 33.44% (expiration) on lateral radiographs and 36.80% (inspiration) and 37.99% (expiration) on ventro-dorsal radiographs.
* Corresponding author. E-mail addresses: [email protected], elham_a. [email protected] (E.A. Hassan). http://dx.doi.org/10.1016/j.jvc.2014.08.001 1760-2734/ª 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
2
F.A. Torad, E.A. Hassan Conclusions: CTR may provide a clinically useful tool for evaluating cardiac size in dogs. ª 2014 Elsevier B.V. All rights reserved.
Abbreviations AoD AUC CI CTR FS GSD LADs LVIDd
aortic root dimension area under the curve confidence of interval two-dimensional cardiothoracic ratio fractional shortening German shepherd dogs left atrial dimension at end-systole left ventricular internal dimension at end-diastole LVIDs left ventricular internal dimension at end-systole ROC receiver operating characteristic RVIDd right ventricular internal dimension at end-diastole VHS vertebral heart scale
be detected by VHS. Considerable changes in the cardiac contour can exist without subsequent dilatation, as in diseases which cause concentric hypertrophy.1,10,11 The development of an objective method for evaluation of cardiac size based on the entire cardiac silhouette may provide a useful tool for diagnosing subtle changes in cardiac size as well as generalized cardiac enlargement. We hypothesize that cardiac size can be evaluated through evaluation of the cardiac area in comparison to the thoracic area in dogs by use of a two-dimensional cardiothoracic ratio (CTR). The purpose of the present study was to evaluate the cardiac size in normal German shepherd dogs (GSD) using CTR, and to use this measure for evaluating dogs with cardiomegaly or microcardia.
Materials and methods Dogs and radiography
Introduction Evaluation of cardiac size is an important clinical variable in dogs1 and changes in cardiac size may be used to monitor and grade disease severity.2 However, a normal cardiac size does not rule out the presence of disease.1 Various diagnostic tools have been used to evaluate cardiac size including thoracic radiography, echocardiography and electrocardiography.3 Thoracic radiography remains the most commonly used method.4,5 Subjective radiographic evaluation of cardiac size with a guide of 2.5e3.5 intercostal spaces on the lateral projection is commonly used in canine practice.6e8 The limitations of this subjective evaluation may include superimposition of the ribs as well as the effects of breed, thoracic conformation, and phase of respiration.1,8,9 An objective evaluation of cardiac size using the vertebral heart scale (VHS) is a widely accepted method first described by Buchanan and Bucheler in 1995.1 The VHS measurement is based on measuring the cardiac height (long axis) and width (short axis) compared to the animal’s vertebral body length. A possible limitation of the VHS is that it relies only on two linear measurements for evaluation of cardiac size and not to the entire cardiac circumference. Therefore, subtle changes in cardiac size may not
The present study was done in 146 GSD. All study procedures were approved by the animal care and use ethical committee of the Faculty of Veterinary Medicine, Cairo University. One hundred healthy client-owned GSD (67 males and 33 females) aged 2.8 0.6 years old and weighing 31.2 5.4 kg were admitted to the clinic of the Department of Surgery, Anesthesiology and Radiology, Faculty of Veterinary Medicine, Cairo University for routine examination. Dogs included in the study were free of clinical signs of cardiovascular or respiratory disease. All dogs were evaluated by physical examination, electrocardiography (standard 12-lead ECG) and echocardiography before enrollment in the study. In addition, 46 GSD (30 males and 16 females) aged 3.1 0.5 years and weighing 30.8 5.2 kg were also included in the study. These dogs were diagnosed with altered cardiac size (cardiomegaly or microcardia) based on subjective radiographic evaluation (hearts less than 2.5 or more than 3.5 intercostal spaces based on lateral radiographs),6e8 echocardiography (according to values obtained from previous studies on normal GSD)12,13 and electrocardiography.14 The 46 GSDs with altered cardiac size included 35 dogs with cardiomegaly [dilated cardiomyopathy (n ¼ 16), persistent arterial duct (6), mitral insufficiency (5), aortic insufficiency (3),
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
Cardiac size in German shepherd dogs tricuspid insufficiency (2), hypertrophic cardiomyopathy (2), pulmonic stenosis (1)] and 11 dogs with microcardia. All examinations were performed without sedation or anesthesia. Four sets of radiographs were taken of each dog consisting of right lateral and ventro-dorsal thoracic radiographs at both peak inspiration and peak expiration. On lateral radiographs, the point of full inspiration was defined as the time point when the lumbodiaphragmatic angle was located caudal to the 12th thoracic vertebra (T12), while the point of maximal expiration was defined as the time when the lumbo-diaphragmatic angle was located cranial to the 11th thoracic vertebra (T11). On the ventro-dorsal radiographs, the point of full inspiration was determined as the phase when the costo-diaphragmatic angles were caudal to the 10th thoracic vertebra (T10) and peak expiration defined as the time at which the diaphragmatic dome was positioned caudal to middle of the 8th thoracic vertebra (T8).2,15 Radiographic positioning was assured by the same radiologist using a standardized positioning technique.16 The radiographs were taken using a grid of medium (400) speed and (0.2 30 40 cm) dimensions, as well as settings of 70e80 kVp and 3 mAs with a focal film distance of 70e90 cm. Analog radiographs for all dogs were analyzed, scanned and transferred to a computer system for measurement of VHS and CTR.
3 program and the result was expressed as units of vertebral length (v). For measuring the cardiac area on lateral and ventro-dorsal radiographs, the “area” measuring tool of the software program was used to trace the contour of the cardiac silhouette along the cranial cardiac border and waist, cardiac apex, caudal cardiac border and waist, cardiac base and again at the cranial border. For measuring the thoracic area on lateral radiographs, the “area” measuring tool of the program was used to track the contour of the thoracic cavity in a closed path along the ventral border of the thoracic vertebrae, lumbodiaphragmatic angle, crura of the diaphragm, sterno-diaphragmatic angle, dorsal border of the sternum, thoracic inlet and back to the start point at the ventral border of the thoracic vertebrae (Fig. 1). On the ventro-dorsal radiographs, the thoracic area was traced from the thoracic inlet, lateral thoracic wall, costo-diaphragmatic angle, diaphragmatic dome, costo-diaphragmatic angle and lateral thoracic wall of the other side and again to the starting point at the thoracic inlet (Fig. 2). The cardiac area and thoracic area were expressed in mm2, all the obtained data were transferred to a spreadsheet, and the CTR was then calculated using the following equation: twodimensional cardiothoracic ratio (CTR) ¼ cardiac area/thoracic area * 100.
Radiographic measurements All radiographs were labeled and randomized for blind examination. Radiographic measurements of the VHS and CTR were done by the same examiner (E.A.H.) using a commercial software program.a The examiner was blinded to the results of other findings. Measurement of the VHS was done in accordance with the method previously described.1 On lateral radiographs, the long axis of the heart was measured starting from the ventral border of the largest main-stem bronchus (carina) to the most distant point of the cardiac apex. The short axis dimension was measured at the level of the caudal vena cava on a line perpendicular to the long axis. The measurements of the two axes were then compared to the vertebrae starting at the cranial edge of T4 using the “unit” measuring tool of the
a Digimizer 4.2.2.0 image analysis software, MedCalc Software bvba, Ostend, Belgium.
Figure 1 Right lateral thoracic radiograph of a healthy German shepherd dog during peak expiration demonstrating tracking of the cardiac and thoracic contour using Digimizer Software. The cardiac silhouette is tracked along the cranial cardiac border and waist, cardiac apex, caudal cardiac border and waist, cardiac base and again at the cranial border. The thoracic cavity is tracked in a closed path along the ventral border of the thoracic vertebrae, lumbo-diaphragmatic angle, crura of the diaphragm, sterno-diaphragmatic angle, dorsal border of the sternum, thoracic inlet and back to the start point at the ventral border of the thoracic vertebrae.
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
4
F.A. Torad, E.A. Hassan Receiver operating characteristic (ROC) curve analysis was used to determine the diagnostic accuracy and the cutoff values of the CTR of normal dogs and those with microcardia or cardiomegaly based on the measurement of left ventricular internal dimension at end-diastole (LVIDd) using graphing software.c
Results
Figure 2 Ventro-dorsal thoracic radiograph of a healthy German shepherd dog during peak expiration demonstrating tracking of the cardiac and thoracic contour using Digimizer Software. The cardiac silhouette is tracked along the cranial cardiac border and waist, cardiac apex, caudal cardiac border and waist, cardiac base and again at the cranial border. The thoracic cavity is tracked in a closed path starting from the thoracic inlet, lateral thoracic wall, costodiaphragmatic angle, diaphragmatic dome, costodiaphragmatic angle and lateral thoracic wall of the other side and again to the starting point at the thoracic inlet.
Statistical analysis Data were expressed as mean SD and were analyzed by a commercial software program.b Repeated measures analysis of variance (ANOVA) was used for the comparison within and between groups. When a significant difference was recorded, a post hoc test (least significance difference) was used to determine how the means differed. The results were considered statistically significant at P < 0.05.
b IBM SPSS Statistics for Windows, Version 21.0, Armonk, New York, USA.
The GSDs included in the study were of similar age and body weight and no statistically significant differences were found in age or body weight within or between the three groups. The mean echocardiographic measurements of normal, microcardia and cardiomegaly groups are tabulated in Table 1. Results of ROC curve analysis and the cutoff value of the echocardiographic measurements are presented in Table 2. The mean CTR in different groups at the different radiographic projections are demonstrated in Table 3. Data obtained from the ROC curve analysis and the cutoff values of the CTR in different radiographic positions are tabulated in Table 4. In dogs with normal cardiac size based on subjective radiographic, electrocardiographic and echocardiographic evaluations, the mean (SD, minimum to maximum) VHS was 9.8 0.5, 9.2ve10.3v. In dogs with reduced cardiac size (microcardia) based on echocardiography and subjective radiographic evaluation (less than 2.5 intercostal spaces on lateral radiographs), the mean VHS was 9.4 0.6, 9.2ve10.1v. The dogs with cardiomegaly had a mean VHS of 10.5 0.4, with a minimum to maximum of 10.2ve10.9v.
Discussion In the present study, an attempt was made using the CTR to evaluate cardiac size relative to thoracic size to provide an objective diagnostic tool for evaluation of cardiac size in the GSD breed. It has been suggested that a reliable method of cardiac evaluation should be based on the comparison of a dog’s radiographs with those of normal dogs of the same breed.16 In this study, measurements of the CTR were done on a large number of clinically normal GSD, with GSD selected to eliminate interbreed variations of cardiac size and shape.11,17e21
c SigmaPlot 12.5 Exact Graphs and Data Analysis Software, Systat Software Inc., San Jose, California, USA.
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
Cardiac size in German shepherd dogs
5
Table 1 Mean echocardiographic measurements (SD, minimumemaximum) and coefficients of variation (italic) in the three groups of German shepherd dogs studied. Echocardiographic measurement
Normal (n ¼ 100)
Microcardia (n ¼ 11)
Cardiomegaly (n ¼ 35)
LVIDd (mm)
45.2 2.2A,B (41.3e49.7) 4.8 33.0 1.0C,D (31.0e35.5) 3.0 24.4 0.2E,F (24.0e25.0) 0.9 11.6 0.9G,H (10.0e13.4) 8.1 25.9 0.9I (24.3e27.5) 3.5 27.0 4.3J,K (22.5e36.7) 15.8
36.3 1.6A (34.6e39.2) 4.3 27.6 1.0C (25.7e28.8) 3.7 22.2 1.6E (20.5e25.4) 7.0 9.4 0.6G (8.6e10.2) 6.2 21.8 1.1I (20.1e23.0) 5.1 24.0 4.6J (17.3e31.4) 19.3
53.1 1.3B (50.3e55.5) 2.5 36.9 0.8D (35.5e38.0) 2.1 27.0 0.9F (24.9e28.9) 3.3 14.5 0.9H (12.5e16.0) 6.3 26.2 1.0 (24.6e28.0) 3.7 19.9 2.3K (14.5e21.1) 9.9
LVIDs (mm)
LADs (mm)
RVIDd (mm)
AoD (mm)
FS (%)
LVIDd ¼ Left ventricular internal dimension at end-diastole; LVIDs ¼ Left ventricular internal dimension at end-systole; LADs ¼ Left atrial dimension at end-systole; RVIDd ¼ Right ventricular internal dimension at end-diastole; AoD ¼ Aortic root dimension; FS ¼ Fractional shortening. Within a row, identical superscript letters indicate significant differences between the normal group and the abnormal group (either microcardia or cardiomegaly) (P < 0.05).
Variations of the radiographic cardiac size that may arise from inconsistent positioning of the animal or phase of respiration19,20 were eliminated by the use of the same evaluator, standardization of radiographic positioning, and consideration of the phase of respiration. Although it has been reported that cardiac size evaluated by the VHS does not change with the dog’s age and developmental stage,22 the dogs used in the study were of similar age and weight range in each group. Radiographic evaluation of cardiac disease is based mainly on the evaluation of the size and shape of the cardiac silhouette.16,23 Increased
cardiac size is expected in dogs with cardiac dilatation, eccentric hypertrophy16,23e26 as well as pericardial effusion.27,28 During the cardiac cycle, variation of the cardiac size may be expected. However, use of the cardiac silhouette for evaluation of cardiac size assumes that the overall cardiac size will not change substantially as ventricular contraction will be associated with atrial relaxation and vice versa. Initial attempts at measuring the CTR for evaluation of cardiac size in the late 1960s and early 1970s measured the cardiac and thoracic dimensions using a simplified linear method on lateral
Table 2 Receiver operating characteristic (ROC) curve analysis of echocardiographic measurements in the prediction of microcardia and cardiomegaly in German shepherd dogs. Echocardiographic measurement Cutoff LVIDd (mm) LVIDs (mm) LADs (mm) RVIDd (mm) AoD (mm) FS (%)
Cardiomegaly (n ¼ 35)
Microcardia (n ¼ 11) 40.3 29.9 23.3 9.9 23.7 31.4
AUC SD
95% CI
Cutoff
1.00e1.00 1.00e1.00 1.00e1.00 0.99e1.00 1.00e1.00 0.50e0.85
50.0 35.2 25.3 13.5 27.6 23.1
1.00 1.00 1.00 1.00 1.00 0.68
0.00 0.00 0.00 0.01 0.00 0.09
AUC SD
95% CI
1.00e1.00 0.99e1.00 1.00e1.00 0.96e1.00 0.48e0.71 1.00e1.00
1.00 0.99 0.99 0.98 0.59 1.00
0.00 0.00 0.00 0.01 0.06 0.00
See Table 1 for key. AUC ¼ area under the curve; CI ¼ confidence of interval.
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
6
F.A. Torad, E.A. Hassan Table 3 Mean cardiothoracic ratio (SD, minimumemaximum) and coefficients of variation (italic) in the 3 groups of German shepherd dogs studied at different radiographic positions. CTR (%) Lateral
Inspiration
Expiration
Ventro-dorsal
Inspiration
Expiration
Normal (n ¼ 100)
Microcardia (n ¼ 11)
Cardiomegaly (n ¼ 35)
27.60 1.10a,c,A,B (24.04e30.45) 0.05 30.13 1.42a,d,C,D (25.05e32.99) 0.04 30.45 1.39b,c,E,F (27.08e33.74) 0.05 33.34 1.46b,d,G,H (24.45e35.87) 0.03
21.96 0.58e,g,A (20.29e21.93) 0.03 22.01 0.46e,h,C (21.05e22.59) 0.02 22.46 0.43f,g,E (21.69e22.97) 0.02 22.80 0.40f,h,G (22.04e23.48) 0.02
33.97 1.13i,k,B (30.11e35.84) 0.03 35.25 0.86i,j,D (32.65e36.98) 0.02 41.66 1.13j,k,F (39.87e43.78) 0.03 43.34 1.25j,l,H (40.11e44.87) 0.03
CTR ¼ cardiothoracic ratio. Within a row, identical superscript capital letters indicate significant differences between the normal group and the abnormal group (either microcardia or cardiomegaly) (P < 0.05). Within a column, identical superscript small letters indicate significant differences between radiographic positions within the same group (P < 0.05).
radiographs,29e31 but this linear measurement was not sufficient to differentiate between normal and enlarged hearts. In the present study, the use of a two-dimensional CTR for evaluation of cardiac size was done by tracking the overall cardiac and thoracic contours using a software program rather than linear measurements. In addition, the CTR determination of normal heart size in this study was compared against a gold standard (echocardiography). The cutoff value of the CTR at the points of peak inspiration (30.28%) and expiration (33.44%) could be considered an early indicator of subtle changes in cardiac size. In normal GSD, the mean VHS (9.8v 0.5v) was similar to findings of a previous study in GSD (9.7v 0.8v).17 While dogs diagnosed with microcardia based on subjective radiographic evaluation had a VHS similar to the group of control dogs without statistically significant differences, we assume that the decrease in the cardiac width
(short axis) was offset by an increase in the cardiac length. The upper limit of the reported VHS in normal dogs in the present study (10.3v) overlapped with the lower limit of the dogs having increased cardiac size (10.2v). Therefore the threshold value of the VHS in GSD is in agreement with the reported 10.2 v in a previous study.17 Many dogs with a change in cardiac size (microcardia or cardiomegaly) based on subjective and echocardiographic evaluation had a VHS within the reference range; this result was similar to findings of a previous study.1 On the other hand, some of the healthy GSD had VHS values above the reference range, as has also been shown in a previous study.11 It has been reported that the diagnostic accuracy of VHS for diagnosing cardiac disease in GSD is 75%.14 Prior reports found that VHS measurement is not affected by the experience of the examiner.32 However, some difficulties may exist in determination of the reference points of cardiac axes
Table 4 Receiver operating characteristic (ROC) curve analysis of cardiothoracic ratio measured at the different radiographic positions in the prediction of microcardia and cardiomegaly. CTR (%)
Lateral Ventro-dorsal
Microcardia (n ¼ 11) Inspiration Expiration Inspiration Expiration
Cardiomegaly (n ¼ 35)
Cutoff
AUC SD
95% CI
Cutoff
AUC SD
95% CI
22.98 25.06 25.03 23.97
1.0 1.0 1.0 1.0
1.00e1.00 1.00e1.00 1.00e1.00 1.00e1.00
30.28 33.44 36.80 37.99
1.0 1.0 1.0 1.0
1.00e1.00 0.99e1.00 1.00e1.00 1.00e1.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
See Tables 2 and 3 for key (P < 0.0001).
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
Cardiac size in German shepherd dogs during VHS measurement.32,33 One of the main limitations of using conventional measuring tools and graduated rulers in VHS measurement is the need for estimation of proportions when cardiac length and width do not span by a whole number of vertebrae. This was reduced in our study by using a software program incorporating a digital calculation of the exact heart length and width, which were then automatically expressed as units of vertebral length. The significant increase in the mean CTR during expiration compared to inspiration could be attributed to decreased thoracic area as the heart will be surrounded by less aerated lung. The difference in thoracic area between expiration and inspiration is expected to be equal to the volume of inspired or expired air in one breath (tidal volume). The CTR was more increased in dogs with cardiomegaly on the ventro-dorsal radiographs as compared to the lateral images, which may suggest that the increase of cardiac size in these dogs occurred in the latero-medial direction more than in the cranio-caudal direction. A previous study also reported that the mean VHS is significantly larger in ventro-dorsal or dorso-ventral radiographs as compared to lateral radiographs.10 The main limitation of using CTR is the difficulty in its applying this measurement to dogs with pericardial effusion, perihilar pulmonary edema, or moderate to severe pleural effusion where there is loss of architectural details of the thoracic cavity by the accumulated fluids. Also, the CTR is not recommended for dogs with pneumothorax as the thorax appears overinflated by the escaped air, which may lead to an inaccurate measurement. Limitations of the study include the use of analog radiographs that required scanning prior to transfer to the computer system. Also the study was done on only one breed of dogs (GSD); further studies are required for setting breed specific normal ranges and prior to applying the CTR to the evaluation of cardiac size on a large scale. In addition, data on the inter-observer variability of CTR measurement is needed.
Conclusions This study provides reference values for the evaluation of cardiac size in GSD based on the cardiac silhouette using the CTR. The CTR may be considered together with other clinical data for the evaluation of GSD with potential heart disease. The ease of tracking the radiopaque cardiac contour and radiolucent thoracic contour as well as
7 the simplicity of the calculation favor the use of the CTR in evaluating the cardiac size of GSD.
Conflict of interest statement The authors declare no conflict of interest related to this report.
References 1. Buchanan JW, Bucheler J. Vertebral scale system to measure canine heart size in radiographs. J Am Vet Med Assoc 1995;206:194e199. 2. Farrow CS. The thorax. In: Farrow CS, editor. Radiology of the cat. St Louis: Mosby Inc.; 1994. p. 50e52. 91e93. 3. Nakayama H, Nakayama T, Hamlin RL. Correlation of cardiac enlargement as assessed by vertebral heart size and echocardiographic and electrocardiographic findings in dogs with evolving cardiomegaly due to rapid ventricular pacing. J Vet Intern Med 2001;15:217e221. 4. Dark P, Bonagura JD, Kelly DF. Cardiac radiography. In: Dark P, Bonagura JD, Kelly DF, editors. Color atlas of veterinary cardiology. St Louis: Mosby; 1996. p. 32e37. 5. Thrall DE. Veterinary radiology. 5th ed. Philadelphia: Saunders Elsevier; 2007479e485. 6. Owens JM. Radiology of the heart. In: Tilley LP, Owens JM, editors. Manual of small animal cardiology. New York: Churchill Livingstone Inc.; 1985. p. 37. 7. Owens JM, Biery DN. Radiographic interpretation for the small animal clinician. 2nd ed. Williams and Wilkins Company; 1999. p. 190. 8. Lamb CR, Boswood A. Role of survey radiography in diagnosing canine cardiac disease. Comp Cont Ed Prac Vet 2002; 24:316e326. 9. Litster A, Buchanan JW. Vertebral scale system to measure heart size in radiographs of cats. J Am Vet Med Assoc 2000; 216:210e214. 10. Buchanan JW. Vertebral scale system to measure heart size in radiographs. Vet Clin North Am Small Anim Pract 2000;30: 379e393. 11. Lamb CR, Tyler M, Boswood A, Skelly BJ, Cain M. Assessment of the value of the vertebral heart scale in the radiographic diagnosis of cardiac disease in dogs. Vet Rec 2000;146: 687e690. 12. Kayar A, Gonul R, Erman M, Uysal A. M-mode echocardiographic parameters and indices in the normal German shepherd dog. Vet Radiol Ultrasound 2006;47:482e486. 13. Muzzi RA, Muzzi LA, de Arau ´jo RB, Cherem M. Echocardiographic indices in normal German shepherd dog. J Vet Sci 2006;7:193e198. 14. Tilley LP, Smith FWK. Electrocardiography. In: Tilley LP, Smith FWK, Oyama M, Sleeper MM, editors. Manual of canine and feline cardiology. Philadelphia: Saunders; 2007. p. 49e77. 15. Silverman S, Suter PF. Influence of inspiration and expiration on canine thoracic radiographs. J Am Vet Med Assoc 1975;166:502e510. 16. Lord PF, Suter PF. Radiology. In: Fox PR, Sisson D, Moise NS, editors. Textbook of canine and feline cardiology. Philadelphia: Saunders; 1999. p. 107e129. 17. Lamb CR, Wilkeley H, Boswood A, Pfeiffer DU. Use of breedspecific ranges for the vertebral heart scale as an aid to the
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
8
18.
19.
20.
21.
22.
23.
24.
25.
F.A. Torad, E.A. Hassan radiographic diagnosis of cardiac disease in dogs. Vet Rec 2001;148:707e711. Kraetschmer S, Ludwig K, Meneses F, Nolte I, Simon D. Vertebral heart scale in the beagle dog. J Small Anim Pract 2008;49:240e243. Suter PF, Lord PF. Thoracic radiography. In: Switzerland W, Suter PF, editors. A text atlas of thoracic diseases of the dog and cat 1984. p. 24e34. Bavegems V, Caelenberg AV, Duchateau L, Sys SU, Bree HV, DeRick A. Vertebral heart size ranges specific for whippets. Vet Radiol Ultrasound 2005;46:400e403. Jepsen-Grant K, Pollard RE, Johnson LR. Vertebral heart scores in eight dog breeds. Vet Radiol Ultrasound 2013;54: 3e8. Sleeper MM, Buchanan JW. Vertebral scale system to measure heart size in growing puppies. J Am Vet Med Assoc 2001;219:57e59. Suter PF, Lord PF. A critical evaluation of the radiographic findings in canine cardiovascular diseases. J Am Vet Med Assoc 1970;158:358e371. Hamlin RL. Analysis of the cardiac silhouette in dorsoventral radiographs from dogs with heart disease. J Am Vet Med Assoc 1968;153:1444e1460. Toombs JP, Ogburn PN. Evaluating canine cardiovascular silhouette: radiographic methods and normal radiographic anatomy. Compend Contin Educ Pract Vet 1985;7: 579e587.
26. Kittleson MD. Radiography of the cardiovascular system. In: Kittleson MD, Kienle RD, editors. Small animal cardiovascular medicine. St Louis: Mosby; 1998p. p. 47e71. 27. Guglielmini C, Diana A, Santarelli G, Torbidone A, Di Tommaso M, Baron Toaldo M, Cipone M. Accuracy of radiographic vertebral heart score and sphericity index in the detection of pericardial effusion in dogs. J Am Vet Med Assoc 2012;241:1048e1055. 28. Co ˆte ´ E, Schwarz LA, Sithole F. Thoracic radiographic findings for dogs with cardiac tamponade attributable to pericardial effusion. J Am Vet Med Assoc 2013;243:232e235. 29. Uhlig K, Werner JA. Ro ¨ntgenologische Verfahren zur Messung der Herzvergro ¨ßerung beim Hund. Berl Mu ¨nchner Wschr 1969;82:110e112. 30. Von Recum A, Poirson JP. Eine vereinfachtes ro ¨ntgenologisches Messverfahren zur Bestimmung von Kardiomegalie. Kleintierpraxis 1971;16:82e86. 31. Empel W. Beurteilung der Gro ¨sse und Gestalt des Herzschattens des Deutschen Scha ¨ferhundes. Kleintierpraxis 1974;19:7e13. 32. Hansson K, Ha ¨ggstro ¨m J, Kvart C, Lord P. Interobserver variability of vertebral heart size measurements in dogs with normal and enlarged hearts. Vet Radiol Ultrasound 2005;46:122e130. 33. Spasojevic-Kosic LJ, Krstic N, Trailovic RD. Comparison of three methods of measuring vertebral heart size in German shepherd dogs. Acta Veterinaria 2007;57:133e141.
Available online at www.sciencedirect.com
ScienceDirect
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
-, -e-
www.elsevier.com/locate/jvc
Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs Faisal A. Torad, BVSc, MVSc, PhD , Elham A. Hassan, BVSc, MVSc, PhD* Department of Surgery, Anesthesiology and Radiology, Faculty of Veterinary Medicine, Cairo University, Giza Square P.O. 12211, Cairo, Egypt Received 14 November 2013; received in revised form 8 August 2014; accepted 18 August 2014
KEYWORDS Radiography; Heart; Cardiomegaly; Microcardia; Canine
Abstract Objectives: To evaluate cardiac size in normal German shepherd dogs (GSD) using the two-dimensional cardiothoracic ratio (CTR) and to use this measure for diagnosing GSD with altered cardiac size. Animals: One hundred clinically normal GSD as well as 46 GSD with altered cardiac size (microcardia or cardiomegaly). Methods: The CTR was computed as the percentage area of the cardiac silhouette relative to the area of the dog’s thorax. Measurements were performed using a digital software program on lateral and ventro-dorsal radiographs at the points of peak inspiration and expiration. Receiver operating characteristic (ROC) curve analysis was used to determine the diagnostic accuracy of the CTR for diagnosing cardiomegaly or microcardia. Results: The mean ( SD) CTR on lateral radiographs of normal dogs was 27.60% 1.10% and 30.13% 1.42% at the points of peak inspiration and expiration, respectively. For ventro-dorsal radiographs, mean CTR was 30.45% 1.39% at peak inspiration and 33.34% 1.46% at peak expiration. The cutoff value of the CTR for diagnosing microcardia on lateral radiographs was 22.98% (inspiration) and 25.06% (expiration), compared to 25.03% (inspiration) and 23.97% (expiration) on ventrodorsal radiographs. Cutoff values for diagnosing cardiomegaly were 30.28% (inspiration) and 33.44% (expiration) on lateral radiographs and 36.80% (inspiration) and 37.99% (expiration) on ventro-dorsal radiographs.
* Corresponding author. E-mail addresses: [email protected], elham_a. [email protected] (E.A. Hassan). http://dx.doi.org/10.1016/j.jvc.2014.08.001 1760-2734/ª 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
2
F.A. Torad, E.A. Hassan Conclusions: CTR may provide a clinically useful tool for evaluating cardiac size in dogs. ª 2014 Elsevier B.V. All rights reserved.
Abbreviations AoD AUC CI CTR FS GSD LADs LVIDd
aortic root dimension area under the curve confidence of interval two-dimensional cardiothoracic ratio fractional shortening German shepherd dogs left atrial dimension at end-systole left ventricular internal dimension at end-diastole LVIDs left ventricular internal dimension at end-systole ROC receiver operating characteristic RVIDd right ventricular internal dimension at end-diastole VHS vertebral heart scale
be detected by VHS. Considerable changes in the cardiac contour can exist without subsequent dilatation, as in diseases which cause concentric hypertrophy.1,10,11 The development of an objective method for evaluation of cardiac size based on the entire cardiac silhouette may provide a useful tool for diagnosing subtle changes in cardiac size as well as generalized cardiac enlargement. We hypothesize that cardiac size can be evaluated through evaluation of the cardiac area in comparison to the thoracic area in dogs by use of a two-dimensional cardiothoracic ratio (CTR). The purpose of the present study was to evaluate the cardiac size in normal German shepherd dogs (GSD) using CTR, and to use this measure for evaluating dogs with cardiomegaly or microcardia.
Materials and methods Dogs and radiography
Introduction Evaluation of cardiac size is an important clinical variable in dogs1 and changes in cardiac size may be used to monitor and grade disease severity.2 However, a normal cardiac size does not rule out the presence of disease.1 Various diagnostic tools have been used to evaluate cardiac size including thoracic radiography, echocardiography and electrocardiography.3 Thoracic radiography remains the most commonly used method.4,5 Subjective radiographic evaluation of cardiac size with a guide of 2.5e3.5 intercostal spaces on the lateral projection is commonly used in canine practice.6e8 The limitations of this subjective evaluation may include superimposition of the ribs as well as the effects of breed, thoracic conformation, and phase of respiration.1,8,9 An objective evaluation of cardiac size using the vertebral heart scale (VHS) is a widely accepted method first described by Buchanan and Bucheler in 1995.1 The VHS measurement is based on measuring the cardiac height (long axis) and width (short axis) compared to the animal’s vertebral body length. A possible limitation of the VHS is that it relies only on two linear measurements for evaluation of cardiac size and not to the entire cardiac circumference. Therefore, subtle changes in cardiac size may not
The present study was done in 146 GSD. All study procedures were approved by the animal care and use ethical committee of the Faculty of Veterinary Medicine, Cairo University. One hundred healthy client-owned GSD (67 males and 33 females) aged 2.8 0.6 years old and weighing 31.2 5.4 kg were admitted to the clinic of the Department of Surgery, Anesthesiology and Radiology, Faculty of Veterinary Medicine, Cairo University for routine examination. Dogs included in the study were free of clinical signs of cardiovascular or respiratory disease. All dogs were evaluated by physical examination, electrocardiography (standard 12-lead ECG) and echocardiography before enrollment in the study. In addition, 46 GSD (30 males and 16 females) aged 3.1 0.5 years and weighing 30.8 5.2 kg were also included in the study. These dogs were diagnosed with altered cardiac size (cardiomegaly or microcardia) based on subjective radiographic evaluation (hearts less than 2.5 or more than 3.5 intercostal spaces based on lateral radiographs),6e8 echocardiography (according to values obtained from previous studies on normal GSD)12,13 and electrocardiography.14 The 46 GSDs with altered cardiac size included 35 dogs with cardiomegaly [dilated cardiomyopathy (n ¼ 16), persistent arterial duct (6), mitral insufficiency (5), aortic insufficiency (3),
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
Cardiac size in German shepherd dogs tricuspid insufficiency (2), hypertrophic cardiomyopathy (2), pulmonic stenosis (1)] and 11 dogs with microcardia. All examinations were performed without sedation or anesthesia. Four sets of radiographs were taken of each dog consisting of right lateral and ventro-dorsal thoracic radiographs at both peak inspiration and peak expiration. On lateral radiographs, the point of full inspiration was defined as the time point when the lumbodiaphragmatic angle was located caudal to the 12th thoracic vertebra (T12), while the point of maximal expiration was defined as the time when the lumbo-diaphragmatic angle was located cranial to the 11th thoracic vertebra (T11). On the ventro-dorsal radiographs, the point of full inspiration was determined as the phase when the costo-diaphragmatic angles were caudal to the 10th thoracic vertebra (T10) and peak expiration defined as the time at which the diaphragmatic dome was positioned caudal to middle of the 8th thoracic vertebra (T8).2,15 Radiographic positioning was assured by the same radiologist using a standardized positioning technique.16 The radiographs were taken using a grid of medium (400) speed and (0.2 30 40 cm) dimensions, as well as settings of 70e80 kVp and 3 mAs with a focal film distance of 70e90 cm. Analog radiographs for all dogs were analyzed, scanned and transferred to a computer system for measurement of VHS and CTR.
3 program and the result was expressed as units of vertebral length (v). For measuring the cardiac area on lateral and ventro-dorsal radiographs, the “area” measuring tool of the software program was used to trace the contour of the cardiac silhouette along the cranial cardiac border and waist, cardiac apex, caudal cardiac border and waist, cardiac base and again at the cranial border. For measuring the thoracic area on lateral radiographs, the “area” measuring tool of the program was used to track the contour of the thoracic cavity in a closed path along the ventral border of the thoracic vertebrae, lumbodiaphragmatic angle, crura of the diaphragm, sterno-diaphragmatic angle, dorsal border of the sternum, thoracic inlet and back to the start point at the ventral border of the thoracic vertebrae (Fig. 1). On the ventro-dorsal radiographs, the thoracic area was traced from the thoracic inlet, lateral thoracic wall, costo-diaphragmatic angle, diaphragmatic dome, costo-diaphragmatic angle and lateral thoracic wall of the other side and again to the starting point at the thoracic inlet (Fig. 2). The cardiac area and thoracic area were expressed in mm2, all the obtained data were transferred to a spreadsheet, and the CTR was then calculated using the following equation: twodimensional cardiothoracic ratio (CTR) ¼ cardiac area/thoracic area * 100.
Radiographic measurements All radiographs were labeled and randomized for blind examination. Radiographic measurements of the VHS and CTR were done by the same examiner (E.A.H.) using a commercial software program.a The examiner was blinded to the results of other findings. Measurement of the VHS was done in accordance with the method previously described.1 On lateral radiographs, the long axis of the heart was measured starting from the ventral border of the largest main-stem bronchus (carina) to the most distant point of the cardiac apex. The short axis dimension was measured at the level of the caudal vena cava on a line perpendicular to the long axis. The measurements of the two axes were then compared to the vertebrae starting at the cranial edge of T4 using the “unit” measuring tool of the
a Digimizer 4.2.2.0 image analysis software, MedCalc Software bvba, Ostend, Belgium.
Figure 1 Right lateral thoracic radiograph of a healthy German shepherd dog during peak expiration demonstrating tracking of the cardiac and thoracic contour using Digimizer Software. The cardiac silhouette is tracked along the cranial cardiac border and waist, cardiac apex, caudal cardiac border and waist, cardiac base and again at the cranial border. The thoracic cavity is tracked in a closed path along the ventral border of the thoracic vertebrae, lumbo-diaphragmatic angle, crura of the diaphragm, sterno-diaphragmatic angle, dorsal border of the sternum, thoracic inlet and back to the start point at the ventral border of the thoracic vertebrae.
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
4
F.A. Torad, E.A. Hassan Receiver operating characteristic (ROC) curve analysis was used to determine the diagnostic accuracy and the cutoff values of the CTR of normal dogs and those with microcardia or cardiomegaly based on the measurement of left ventricular internal dimension at end-diastole (LVIDd) using graphing software.c
Results
Figure 2 Ventro-dorsal thoracic radiograph of a healthy German shepherd dog during peak expiration demonstrating tracking of the cardiac and thoracic contour using Digimizer Software. The cardiac silhouette is tracked along the cranial cardiac border and waist, cardiac apex, caudal cardiac border and waist, cardiac base and again at the cranial border. The thoracic cavity is tracked in a closed path starting from the thoracic inlet, lateral thoracic wall, costodiaphragmatic angle, diaphragmatic dome, costodiaphragmatic angle and lateral thoracic wall of the other side and again to the starting point at the thoracic inlet.
Statistical analysis Data were expressed as mean SD and were analyzed by a commercial software program.b Repeated measures analysis of variance (ANOVA) was used for the comparison within and between groups. When a significant difference was recorded, a post hoc test (least significance difference) was used to determine how the means differed. The results were considered statistically significant at P < 0.05.
b IBM SPSS Statistics for Windows, Version 21.0, Armonk, New York, USA.
The GSDs included in the study were of similar age and body weight and no statistically significant differences were found in age or body weight within or between the three groups. The mean echocardiographic measurements of normal, microcardia and cardiomegaly groups are tabulated in Table 1. Results of ROC curve analysis and the cutoff value of the echocardiographic measurements are presented in Table 2. The mean CTR in different groups at the different radiographic projections are demonstrated in Table 3. Data obtained from the ROC curve analysis and the cutoff values of the CTR in different radiographic positions are tabulated in Table 4. In dogs with normal cardiac size based on subjective radiographic, electrocardiographic and echocardiographic evaluations, the mean (SD, minimum to maximum) VHS was 9.8 0.5, 9.2ve10.3v. In dogs with reduced cardiac size (microcardia) based on echocardiography and subjective radiographic evaluation (less than 2.5 intercostal spaces on lateral radiographs), the mean VHS was 9.4 0.6, 9.2ve10.1v. The dogs with cardiomegaly had a mean VHS of 10.5 0.4, with a minimum to maximum of 10.2ve10.9v.
Discussion In the present study, an attempt was made using the CTR to evaluate cardiac size relative to thoracic size to provide an objective diagnostic tool for evaluation of cardiac size in the GSD breed. It has been suggested that a reliable method of cardiac evaluation should be based on the comparison of a dog’s radiographs with those of normal dogs of the same breed.16 In this study, measurements of the CTR were done on a large number of clinically normal GSD, with GSD selected to eliminate interbreed variations of cardiac size and shape.11,17e21
c SigmaPlot 12.5 Exact Graphs and Data Analysis Software, Systat Software Inc., San Jose, California, USA.
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
Cardiac size in German shepherd dogs
5
Table 1 Mean echocardiographic measurements (SD, minimumemaximum) and coefficients of variation (italic) in the three groups of German shepherd dogs studied. Echocardiographic measurement
Normal (n ¼ 100)
Microcardia (n ¼ 11)
Cardiomegaly (n ¼ 35)
LVIDd (mm)
45.2 2.2A,B (41.3e49.7) 4.8 33.0 1.0C,D (31.0e35.5) 3.0 24.4 0.2E,F (24.0e25.0) 0.9 11.6 0.9G,H (10.0e13.4) 8.1 25.9 0.9I (24.3e27.5) 3.5 27.0 4.3J,K (22.5e36.7) 15.8
36.3 1.6A (34.6e39.2) 4.3 27.6 1.0C (25.7e28.8) 3.7 22.2 1.6E (20.5e25.4) 7.0 9.4 0.6G (8.6e10.2) 6.2 21.8 1.1I (20.1e23.0) 5.1 24.0 4.6J (17.3e31.4) 19.3
53.1 1.3B (50.3e55.5) 2.5 36.9 0.8D (35.5e38.0) 2.1 27.0 0.9F (24.9e28.9) 3.3 14.5 0.9H (12.5e16.0) 6.3 26.2 1.0 (24.6e28.0) 3.7 19.9 2.3K (14.5e21.1) 9.9
LVIDs (mm)
LADs (mm)
RVIDd (mm)
AoD (mm)
FS (%)
LVIDd ¼ Left ventricular internal dimension at end-diastole; LVIDs ¼ Left ventricular internal dimension at end-systole; LADs ¼ Left atrial dimension at end-systole; RVIDd ¼ Right ventricular internal dimension at end-diastole; AoD ¼ Aortic root dimension; FS ¼ Fractional shortening. Within a row, identical superscript letters indicate significant differences between the normal group and the abnormal group (either microcardia or cardiomegaly) (P < 0.05).
Variations of the radiographic cardiac size that may arise from inconsistent positioning of the animal or phase of respiration19,20 were eliminated by the use of the same evaluator, standardization of radiographic positioning, and consideration of the phase of respiration. Although it has been reported that cardiac size evaluated by the VHS does not change with the dog’s age and developmental stage,22 the dogs used in the study were of similar age and weight range in each group. Radiographic evaluation of cardiac disease is based mainly on the evaluation of the size and shape of the cardiac silhouette.16,23 Increased
cardiac size is expected in dogs with cardiac dilatation, eccentric hypertrophy16,23e26 as well as pericardial effusion.27,28 During the cardiac cycle, variation of the cardiac size may be expected. However, use of the cardiac silhouette for evaluation of cardiac size assumes that the overall cardiac size will not change substantially as ventricular contraction will be associated with atrial relaxation and vice versa. Initial attempts at measuring the CTR for evaluation of cardiac size in the late 1960s and early 1970s measured the cardiac and thoracic dimensions using a simplified linear method on lateral
Table 2 Receiver operating characteristic (ROC) curve analysis of echocardiographic measurements in the prediction of microcardia and cardiomegaly in German shepherd dogs. Echocardiographic measurement Cutoff LVIDd (mm) LVIDs (mm) LADs (mm) RVIDd (mm) AoD (mm) FS (%)
Cardiomegaly (n ¼ 35)
Microcardia (n ¼ 11) 40.3 29.9 23.3 9.9 23.7 31.4
AUC SD
95% CI
Cutoff
1.00e1.00 1.00e1.00 1.00e1.00 0.99e1.00 1.00e1.00 0.50e0.85
50.0 35.2 25.3 13.5 27.6 23.1
1.00 1.00 1.00 1.00 1.00 0.68
0.00 0.00 0.00 0.01 0.00 0.09
AUC SD
95% CI
1.00e1.00 0.99e1.00 1.00e1.00 0.96e1.00 0.48e0.71 1.00e1.00
1.00 0.99 0.99 0.98 0.59 1.00
0.00 0.00 0.00 0.01 0.06 0.00
See Table 1 for key. AUC ¼ area under the curve; CI ¼ confidence of interval.
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
6
F.A. Torad, E.A. Hassan Table 3 Mean cardiothoracic ratio (SD, minimumemaximum) and coefficients of variation (italic) in the 3 groups of German shepherd dogs studied at different radiographic positions. CTR (%) Lateral
Inspiration
Expiration
Ventro-dorsal
Inspiration
Expiration
Normal (n ¼ 100)
Microcardia (n ¼ 11)
Cardiomegaly (n ¼ 35)
27.60 1.10a,c,A,B (24.04e30.45) 0.05 30.13 1.42a,d,C,D (25.05e32.99) 0.04 30.45 1.39b,c,E,F (27.08e33.74) 0.05 33.34 1.46b,d,G,H (24.45e35.87) 0.03
21.96 0.58e,g,A (20.29e21.93) 0.03 22.01 0.46e,h,C (21.05e22.59) 0.02 22.46 0.43f,g,E (21.69e22.97) 0.02 22.80 0.40f,h,G (22.04e23.48) 0.02
33.97 1.13i,k,B (30.11e35.84) 0.03 35.25 0.86i,j,D (32.65e36.98) 0.02 41.66 1.13j,k,F (39.87e43.78) 0.03 43.34 1.25j,l,H (40.11e44.87) 0.03
CTR ¼ cardiothoracic ratio. Within a row, identical superscript capital letters indicate significant differences between the normal group and the abnormal group (either microcardia or cardiomegaly) (P < 0.05). Within a column, identical superscript small letters indicate significant differences between radiographic positions within the same group (P < 0.05).
radiographs,29e31 but this linear measurement was not sufficient to differentiate between normal and enlarged hearts. In the present study, the use of a two-dimensional CTR for evaluation of cardiac size was done by tracking the overall cardiac and thoracic contours using a software program rather than linear measurements. In addition, the CTR determination of normal heart size in this study was compared against a gold standard (echocardiography). The cutoff value of the CTR at the points of peak inspiration (30.28%) and expiration (33.44%) could be considered an early indicator of subtle changes in cardiac size. In normal GSD, the mean VHS (9.8v 0.5v) was similar to findings of a previous study in GSD (9.7v 0.8v).17 While dogs diagnosed with microcardia based on subjective radiographic evaluation had a VHS similar to the group of control dogs without statistically significant differences, we assume that the decrease in the cardiac width
(short axis) was offset by an increase in the cardiac length. The upper limit of the reported VHS in normal dogs in the present study (10.3v) overlapped with the lower limit of the dogs having increased cardiac size (10.2v). Therefore the threshold value of the VHS in GSD is in agreement with the reported 10.2 v in a previous study.17 Many dogs with a change in cardiac size (microcardia or cardiomegaly) based on subjective and echocardiographic evaluation had a VHS within the reference range; this result was similar to findings of a previous study.1 On the other hand, some of the healthy GSD had VHS values above the reference range, as has also been shown in a previous study.11 It has been reported that the diagnostic accuracy of VHS for diagnosing cardiac disease in GSD is 75%.14 Prior reports found that VHS measurement is not affected by the experience of the examiner.32 However, some difficulties may exist in determination of the reference points of cardiac axes
Table 4 Receiver operating characteristic (ROC) curve analysis of cardiothoracic ratio measured at the different radiographic positions in the prediction of microcardia and cardiomegaly. CTR (%)
Lateral Ventro-dorsal
Microcardia (n ¼ 11) Inspiration Expiration Inspiration Expiration
Cardiomegaly (n ¼ 35)
Cutoff
AUC SD
95% CI
Cutoff
AUC SD
95% CI
22.98 25.06 25.03 23.97
1.0 1.0 1.0 1.0
1.00e1.00 1.00e1.00 1.00e1.00 1.00e1.00
30.28 33.44 36.80 37.99
1.0 1.0 1.0 1.0
1.00e1.00 0.99e1.00 1.00e1.00 1.00e1.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
See Tables 2 and 3 for key (P < 0.0001).
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
Cardiac size in German shepherd dogs during VHS measurement.32,33 One of the main limitations of using conventional measuring tools and graduated rulers in VHS measurement is the need for estimation of proportions when cardiac length and width do not span by a whole number of vertebrae. This was reduced in our study by using a software program incorporating a digital calculation of the exact heart length and width, which were then automatically expressed as units of vertebral length. The significant increase in the mean CTR during expiration compared to inspiration could be attributed to decreased thoracic area as the heart will be surrounded by less aerated lung. The difference in thoracic area between expiration and inspiration is expected to be equal to the volume of inspired or expired air in one breath (tidal volume). The CTR was more increased in dogs with cardiomegaly on the ventro-dorsal radiographs as compared to the lateral images, which may suggest that the increase of cardiac size in these dogs occurred in the latero-medial direction more than in the cranio-caudal direction. A previous study also reported that the mean VHS is significantly larger in ventro-dorsal or dorso-ventral radiographs as compared to lateral radiographs.10 The main limitation of using CTR is the difficulty in its applying this measurement to dogs with pericardial effusion, perihilar pulmonary edema, or moderate to severe pleural effusion where there is loss of architectural details of the thoracic cavity by the accumulated fluids. Also, the CTR is not recommended for dogs with pneumothorax as the thorax appears overinflated by the escaped air, which may lead to an inaccurate measurement. Limitations of the study include the use of analog radiographs that required scanning prior to transfer to the computer system. Also the study was done on only one breed of dogs (GSD); further studies are required for setting breed specific normal ranges and prior to applying the CTR to the evaluation of cardiac size on a large scale. In addition, data on the inter-observer variability of CTR measurement is needed.
Conclusions This study provides reference values for the evaluation of cardiac size in GSD based on the cardiac silhouette using the CTR. The CTR may be considered together with other clinical data for the evaluation of GSD with potential heart disease. The ease of tracking the radiopaque cardiac contour and radiolucent thoracic contour as well as
7 the simplicity of the calculation favor the use of the CTR in evaluating the cardiac size of GSD.
Conflict of interest statement The authors declare no conflict of interest related to this report.
References 1. Buchanan JW, Bucheler J. Vertebral scale system to measure canine heart size in radiographs. J Am Vet Med Assoc 1995;206:194e199. 2. Farrow CS. The thorax. In: Farrow CS, editor. Radiology of the cat. St Louis: Mosby Inc.; 1994. p. 50e52. 91e93. 3. Nakayama H, Nakayama T, Hamlin RL. Correlation of cardiac enlargement as assessed by vertebral heart size and echocardiographic and electrocardiographic findings in dogs with evolving cardiomegaly due to rapid ventricular pacing. J Vet Intern Med 2001;15:217e221. 4. Dark P, Bonagura JD, Kelly DF. Cardiac radiography. In: Dark P, Bonagura JD, Kelly DF, editors. Color atlas of veterinary cardiology. St Louis: Mosby; 1996. p. 32e37. 5. Thrall DE. Veterinary radiology. 5th ed. Philadelphia: Saunders Elsevier; 2007479e485. 6. Owens JM. Radiology of the heart. In: Tilley LP, Owens JM, editors. Manual of small animal cardiology. New York: Churchill Livingstone Inc.; 1985. p. 37. 7. Owens JM, Biery DN. Radiographic interpretation for the small animal clinician. 2nd ed. Williams and Wilkins Company; 1999. p. 190. 8. Lamb CR, Boswood A. Role of survey radiography in diagnosing canine cardiac disease. Comp Cont Ed Prac Vet 2002; 24:316e326. 9. Litster A, Buchanan JW. Vertebral scale system to measure heart size in radiographs of cats. J Am Vet Med Assoc 2000; 216:210e214. 10. Buchanan JW. Vertebral scale system to measure heart size in radiographs. Vet Clin North Am Small Anim Pract 2000;30: 379e393. 11. Lamb CR, Tyler M, Boswood A, Skelly BJ, Cain M. Assessment of the value of the vertebral heart scale in the radiographic diagnosis of cardiac disease in dogs. Vet Rec 2000;146: 687e690. 12. Kayar A, Gonul R, Erman M, Uysal A. M-mode echocardiographic parameters and indices in the normal German shepherd dog. Vet Radiol Ultrasound 2006;47:482e486. 13. Muzzi RA, Muzzi LA, de Arau ´jo RB, Cherem M. Echocardiographic indices in normal German shepherd dog. J Vet Sci 2006;7:193e198. 14. Tilley LP, Smith FWK. Electrocardiography. In: Tilley LP, Smith FWK, Oyama M, Sleeper MM, editors. Manual of canine and feline cardiology. Philadelphia: Saunders; 2007. p. 49e77. 15. Silverman S, Suter PF. Influence of inspiration and expiration on canine thoracic radiographs. J Am Vet Med Assoc 1975;166:502e510. 16. Lord PF, Suter PF. Radiology. In: Fox PR, Sisson D, Moise NS, editors. Textbook of canine and feline cardiology. Philadelphia: Saunders; 1999. p. 107e129. 17. Lamb CR, Wilkeley H, Boswood A, Pfeiffer DU. Use of breedspecific ranges for the vertebral heart scale as an aid to the
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001
8
18.
19.
20.
21.
22.
23.
24.
25.
F.A. Torad, E.A. Hassan radiographic diagnosis of cardiac disease in dogs. Vet Rec 2001;148:707e711. Kraetschmer S, Ludwig K, Meneses F, Nolte I, Simon D. Vertebral heart scale in the beagle dog. J Small Anim Pract 2008;49:240e243. Suter PF, Lord PF. Thoracic radiography. In: Switzerland W, Suter PF, editors. A text atlas of thoracic diseases of the dog and cat 1984. p. 24e34. Bavegems V, Caelenberg AV, Duchateau L, Sys SU, Bree HV, DeRick A. Vertebral heart size ranges specific for whippets. Vet Radiol Ultrasound 2005;46:400e403. Jepsen-Grant K, Pollard RE, Johnson LR. Vertebral heart scores in eight dog breeds. Vet Radiol Ultrasound 2013;54: 3e8. Sleeper MM, Buchanan JW. Vertebral scale system to measure heart size in growing puppies. J Am Vet Med Assoc 2001;219:57e59. Suter PF, Lord PF. A critical evaluation of the radiographic findings in canine cardiovascular diseases. J Am Vet Med Assoc 1970;158:358e371. Hamlin RL. Analysis of the cardiac silhouette in dorsoventral radiographs from dogs with heart disease. J Am Vet Med Assoc 1968;153:1444e1460. Toombs JP, Ogburn PN. Evaluating canine cardiovascular silhouette: radiographic methods and normal radiographic anatomy. Compend Contin Educ Pract Vet 1985;7: 579e587.
26. Kittleson MD. Radiography of the cardiovascular system. In: Kittleson MD, Kienle RD, editors. Small animal cardiovascular medicine. St Louis: Mosby; 1998p. p. 47e71. 27. Guglielmini C, Diana A, Santarelli G, Torbidone A, Di Tommaso M, Baron Toaldo M, Cipone M. Accuracy of radiographic vertebral heart score and sphericity index in the detection of pericardial effusion in dogs. J Am Vet Med Assoc 2012;241:1048e1055. 28. Co ˆte ´ E, Schwarz LA, Sithole F. Thoracic radiographic findings for dogs with cardiac tamponade attributable to pericardial effusion. J Am Vet Med Assoc 2013;243:232e235. 29. Uhlig K, Werner JA. Ro ¨ntgenologische Verfahren zur Messung der Herzvergro ¨ßerung beim Hund. Berl Mu ¨nchner Wschr 1969;82:110e112. 30. Von Recum A, Poirson JP. Eine vereinfachtes ro ¨ntgenologisches Messverfahren zur Bestimmung von Kardiomegalie. Kleintierpraxis 1971;16:82e86. 31. Empel W. Beurteilung der Gro ¨sse und Gestalt des Herzschattens des Deutschen Scha ¨ferhundes. Kleintierpraxis 1974;19:7e13. 32. Hansson K, Ha ¨ggstro ¨m J, Kvart C, Lord P. Interobserver variability of vertebral heart size measurements in dogs with normal and enlarged hearts. Vet Radiol Ultrasound 2005;46:122e130. 33. Spasojevic-Kosic LJ, Krstic N, Trailovic RD. Comparison of three methods of measuring vertebral heart size in German shepherd dogs. Acta Veterinaria 2007;57:133e141.
Available online at www.sciencedirect.com
ScienceDirect
Please cite this article in press as: Torad FA, Hassan EA, Two-dimensional cardiothoracic ratio for evaluation of cardiac size in German shepherd dogs, Journal of Veterinary Cardiology (2014), http://dx.doi.org/10.1016/j.jvc.2014.08.001