Clinical And Echocardiographic Findings
Upcoming SlideShare
Loading in...5
×
 

Clinical And Echocardiographic Findings

on

  • 767 views

 

Statistics

Views

Total Views
767
Views on SlideShare
767
Embed Views
0

Actions

Likes
0
Downloads
13
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Clinical And Echocardiographic Findings Clinical And Echocardiographic Findings Document Transcript

  • Journal of Veterinary Cardiology (2007) 9, 83e89 www.elsevier.com/locate/jvc Clinical and echocardiographic findings of pulmonary artery stenosis in seven cats Donald P. Schrope, DVM, Dipl ACVIM a,*, b William J. Kelch, DVM, PhD, Dipl ACVPM a Oradell Animal Hospital, 580 Winters Avenue, Paramus, NJ 07652, USA b Department of Comparative Medicine, College of Veterinary Medicine, University of Tennessee, Box 1071, Knoxville, TN 37901-1071, USA Received 19 November 2006; received in revised form 19 November 2006; accepted 10 September 2007 KEYWORDS Abstract Objectives: Describe the clinical, electrocardiographic (ECG), radio- Pulmonary artery; graphic and echocardiographic findings in cats with isolated pulmonary artery ste- Stenosis; nosis. Assess the usefulness of systolic and diastolic Doppler measurements at Feline; predicting stenosis severity. Congenital Background: Pulmonary artery stenosis is an infrequent congenital cardiac defect in humans that has not been reported in cats. In humans, pulmonary artery stenosis is usually seen in conjunction with other cardiac defects and may lead to clinical signs if severe. Animals, materials and methods: Seven cats with pulmonary artery stenosis were retrospectively evaluated. Medical records, radiographs, ECGs, echocardiograms and angiocardiograms were reviewed. Severity of stenosis was assessed by two- dimensional and color Doppler echocardiographic evaluation and clinical findings. Peak systolic and diastolic gradients across the stenosis, and systolic and diastolic pressure decay half-times were graded using echocardiography. In addition, the du- ration of antegrade flow during diastole was subjectively assessed. Univariate anal- yses were performed to assess the best variable to predict stenosis severity. Results: Concurrent congenital defects were not identified. Only cats with severe obstruction showed clinical signs including exertional dyspnea and lethargy. Dia- stolic Doppler measurements were superior to systolic measurements at predicting severity of stenosis. Antegrade flow throughout diastole and/or a diastolic pressure half-time of >100 ms indicated severe obstruction. The prognosis for pulmonary artery stenosis appears to be good regardless of severity. * Corresponding author. E-mail address: dpsdvm@optonline.net (D.P. Schrope). 1760-2734/$ - see front matter ª 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jvc.2007.09.001
  • 84 D.P. Schrope, W.J. Kelch Conclusion: Among cats with pulmonary artery stenosis, clinical signs are uncom- mon and prognosis is good. Doppler assessment of diastolic flow appears to be superior to systolic flow at predicting severity. ª 2007 Elsevier B.V. All rights reserved. Congenital stenosis of the main and/or The P-wave and QRS complex amplitude and branched pulmonary arteries has been identified duration, and the PR interval were measured on in humans and is commonly identified with con- three consecutive cardiac cycles and averaged on current lesions such as ventricular septal defect, the electrocardiograms that were available for pulmonic stenosis, patent ductus arteriosus, or review. The mean QRS axis was also assessed. tetralogy of Fallot.1e3 Congenital obstruction of The vertebral heart scale (VHS) was measured on the right ventricular outflow tract is uncommon all lateral and ventrodorsal radiographs that were in cats with an incidence of about 2e3% of congen- available for review.12 ital cardiac defects.4,5 To the authors’ knowledge, Complete echocardiograms were performed us- stenosis of the main or branched pulmonary ar- ing standard echocardiographic views. Maximum teries has not been reported in a cat. pulmonary valve annulus diameter (PVD) and Stenosis of the pulmonary artery is hemody- pulmonary artery stenosis diameter (PSD) were namically similar to coarctation of the aorta and obtained from right or left parasternal short-axis was referred to as coarctation of the pulmonary two-dimensional (2D) images during systole. Pul- artery in early human studies.1 As with other types monary artery stenosis diameter was evaluated of stenotic lesions, Doppler echocardiography has with the assistance of color-flow Doppler (CFD) to been used to assess the severity of aortic coarcta- identify the vena contracta in all cases. In cats tion using the modified Bernoulli’s equation to es- with severe disease the stenotic orifice was often timate the systolic pressure gradient across the so small that it was not easily identified without obstruction, however, this does not consistently CFD. The ratio of stenosis diameter to pulmonary reflect severity of the stenosis when compared to valve annulus diameter (PSD/PVD) was calculated. gradients measured at cardiac catheterization.6,7 Maximum right atrial diameter (RAD) and left atrial Further studies of aortic coarctation in humans diameter (LAD) were obtained from right para- have identified variable degrees of diastolic flow sternal long-axis 2D images during diastole. Mea- across the stenosis. Both subjective and objective surements of the RAD were performed using assessment of this diastolic flow has been found to a method similar to published methods of measur- be more accurate than systolic gradients at assess- ing the LAD in long-axis.13 The ratio of RAD to LAD ing severity of coarctation in many patients.6,8,9 In was calculated and considered normal if <1.0.14 humans with pulmonary artery stenosis and in ex- Right ventricular end-diastolic diameter (RVDd) perimental studies of pulmonary artery banding and left ventricular end-diastolic diameter (LVDd) in dogs, there are varying conclusions about how were obtained from M-mode tracings. The RVDd closely the Doppler systolic and cardiac catheteri- to LVDd ratio was generated and considered nor- zation-derived gradients correlated.10,11 In cats mal if <0.33.14 The remainder of the left heart with pulmonary artery stenosis, it was hypothe- measurements was made from standard 2D and sized that diastolic Doppler evaluation would re- M-mode echocardiographic views. All 2D and M- late better to severity than systolic evaluation. mode measurements were made on three consec- utive cardiac cycles and averaged. Peak systolic flow velocity through the stenosis Animals, materials and methods was measured in all cats after normal pulmonary valve flows were confirmed and a velocity step-up Cats with an echocardiographic diagnosis of pul- was identified at the level of the stenosis. Estima- monary artery stenosis were retrospectively re- tion of the peak systolic pressure gradient (PaS) viewed. The majority of these cats were identified and peak diastolic pressure gradient (PaD) across through an animal shelter that receives animals the stenosis was calculated using the modified from multiple sites along the east coast of the Bernoulli’s equation. Peak diastolic flow velocity United States. All of the cats were initially evalu- was measured at the onset of diastole identified by ated because a murmur had been ausculted on the end of the T-wave on the electrocardiogram physical exam. (Fig. 1).8 The systolic pressure decay half-time
  • Clinical and echocardiographic findings 85 multiple determination which estimates the pro- portion of the response variation that can be explained by the continuous independent variable; and r-squared for an ordinal variable refers to the uncertainty coefficient which, analogous to the co- efficient of multiple determination for continuous variables, also explains the proportion of the re- sponse variation explained by the ordinal indepen- dent variable. Results Figure 1 Spectral Doppler findings in cats 3 (A) and 7 The signalment, clinical signs, and findings of (B) and technique used to identify diastolic Doppler cardiac auscultation are summarized in Table 1. measurements. The onset of diastole was identified at None of the cats had evidence of jugular pulses, the end of the T-wave and the peak diastolic gradient cyanosis, alterations in femoral pulse strength or was measured at this point. The systolic pressure decay quality, abnormalities in S1 or S2, or a gallop half-time (Spht) was measured from the peak systolic rhythm during a resting examination. flow to the onset of diastole. The diastolic pressure Six of the cats were asymptomatic at the time decay half-time (Dpht) was measured from the onset of presentation. One cat (diagnosed at 12 years of of diastole to the end of diastolic flow. age) was described as less active than expected all of her life by one of two owners. One cat, that had (Spht) was measured from the peak systolic flow to initially been asymptomatic, developed moderate the onset of diastole (the end of the T-wave). The exertional dyspnea soon after diagnosis. An arte- diastolic pressure decay half-time (Dpht) was mea- rial blood gas was performed after dyspnea was sured from the onset of diastole to baseline or to induced by exercise. The arterial blood gas was the onset of systole if antegrade flow was present assessed as a mixed metabolic and respiratory throughout diastole. All measurements were per- acidosis with hypoxia (pH 7.1, pCO2 45.6, pO2 formed on three consecutive waveforms and aver- 84.0, HCO3 14.0, BEe15, sO2 90%). The calculated aged. The presence or absence of diastolic flow alveolar to arterial gradient was nine. The findings (DFA) was also subjectively assessed from the were consistent with hypoperfusion from pulmo- spectral Doppler tracing and graded as; (0þ) no di- nary artery stenosis. The clinical signs in cat 7 astolic flow, (1þ) flow through 50% or less of dias- did not progress over time, and no treatment was tole, (2þ) flow through >50% but <100% of initiated. diastole, and (3þ) diastolic flow throughout dias- Three cats had an ECG available for review. No tole resulting in continuous flow across the stenosis abnormalities were identified in two of the cats (#2 (Fig. 1). and 6). The ECG of cat 5 revealed deep S-waves in Angiocardiography was performed to better leads I, II, III, aVF, a right mean electrical axis shift characterize the lesion in the two initial cases. (170 e180 ), and widening of the QRS complex The ratio of minimal stenosis diameter to PV (50 ms) suggestive of right ventricular enlargement annulus diameter was obtained and averaged or a partial right bundle branch block. Five cats from at least two cardiac cycles during systole. (#3e7) had chest radiographs available for review. For statistical analysis, severity of stenosis was All five cats had normal VHS on lateral and ventro- assessed based on the echocardiographic PSD/PVD dorsal views and one cat (#7) had subjective evi- ratio. The diastolic and systolic Doppler data were dence of right ventricular enlargement and compared to the PSD/PVD ratio. Univariate analy- a small pulmonary artery bulge. ses were performed to assess the correlation with Echocardiography was performed in all seven stenosis severity.c A p-value <0.05 was considered cats and identified pulmonary artery stenosis with significant; r refers to the coefficient of linear cor- variable degrees of main pulmonary artery dilation relation between two variables; r-squared for con- proximal to the stenotic lesion (Figs. 2 and 3). tinuous variables refers to the coefficient of Color-flow Doppler confirmed turbulence during systole at the suspected stenosis in all cats and di- astolic flow at the level of the stenosis in cats 6 c JMP, Version 5.1 statistical software (SAS Institute Inc., SAS and 7. No cat showed evidence of pulmonary valve Campus Dr, Cary, NC 27513, USA). disease or left heart abnormalities. A summary of
  • 86 D.P. Schrope, W.J. Kelch Table 1 Signalment, initial auscultation findings, and clinical signs in seven cats with pulmonary artery stenosis Cat Breed Sex Age at Age as of Auscultation at initial Clinical signs diagnosis publication (years) examination 1 DSH F 2 months 6.8 2e3/6 left and right basilar SM None 2 DSH F 8 months 7.4 2e3/6 right basilar SM None 3 DSH F 9 months 9 3/6 left basilar SM None 4 DSH M 3 years 9a 4/6 left basilar SM None 5 Persian M 5 months 4.6 3/6 right basilar SM None 6 DSH F 12.7 years 13.9b 4/6 left basilar SM Lethargy 7 DSH M 15 months 9.6 2e3/6 left basilar SM and 1/6 DM Exertional dyspnea DSH ¼ domestic short hair, M ¼ male, F ¼ female, SM ¼ systolic murmur, and DM ¼ diastolic murmur. a Lost to follow-up. b Euthanized. the 2D echocardiographic findings of the right just proximal to the bifurcation without involve- heart and spectral Doppler data is presented in Ta- ment of the branched or peripheral segmental ble 2. Visualization of the pulmonary artery bifur- pulmonary artery (Fig. 4). The average ratio of cation and branched pulmonary arteries from the PSD/PVD on the angiocardiograms was 0.50 in cat right parasternal short-axis view was accomplished 4 and 0.32 in cat 7. in the majority of cats. In cat 6 the branched pul- Based on the univariate analyses, the Dpht monary arteries could only be visualized from (p ¼ 0.002; r ¼ À0.94; r2 ¼ 0.88) was most closely a left parasternal short-axis view. In cats 4 and 7 associated with stenosis severity. The DFA the branched pulmonary arteries could not be visu- (p ¼ 0.025; r2 ¼ 0.94), PaD (p ¼ 0.003; r ¼ À0.92; alized well from the right or left side with 2D. The r2 ¼ 0.85) and PaS (p ¼ 0.009; r ¼ À0.88; r2 ¼ 0.78) CFD in both cats, though, did show flow conver- were also significantly associated with stenosis se- gence and turbulence at the bifurcation with split- verity. Cats 6 and 7 had evidence of right atrial ting of the color-flow pattern into the origins of and ventricular dilation as well as clinical signs. It the branched pulmonary arteries. was felt that these two cats had severe disease. Angiocardiography confirmed in cats 4 and 7 an Both cats with severe stenosis had a DFA of 3þ dia- isolated lesion of the distal main pulmonary artery stolic flow and Dpht >100 ms (Table 2). Figure 2 Two-dimensional (A) and color-flow Doppler (B) images of cat 1 with mild pulmonary artery stenosis. Im- ages were obtained during systole. Note the mild narrowing of the pulmonary artery just proximal to the bifurcation (A) and associated flow convergence using color-flow Doppler (B). (Ao ¼ aorta, MPa ¼ main pulmonary artery, RPa ¼ right-branched pulmonary artery, LPa ¼ left-branched pulmonary artery).
  • Clinical and echocardiographic findings 87 Figure 3 Two-dimensional (A) and color-flow Doppler (B) images of cat 6 with severe pulmonary artery stenosis. Im- ages were obtained during systole. Note the severe stenosis in the body of the pulmonary artery and severe dilation of the main pulmonary artery (A). The stenotic orifice was very difficult to identify using only 2D imaging in this cat. Color-flow Doppler identified the orifice of the stenotic lesion and confirmed flow convergence (B). (Ao ¼ aorta, MPa ¼ main pulmonary artery, RPa ¼ right-branched pulmonary artery, LPa ¼ left-branched pulmonary artery). All of the cats were alive well into maturity and conjunction with other congenital cardiac anoma- only cat 6 had died at the time of publication. Cat 6 lies. The cause of pulmonary artery stenosis is had been euthanized approximately 15 months unclear. Postmortem studies in adults reveal intimal after diagnosis due to repeated signs of anorexia, invasion by smooth muscle, medial hyperplasia, and depression, and sneezing, as well as a suspicious increased and disorganized elastin fibers at the site pulmonary nodule on radiographs. A necropsy was of stenosis. In contrast, postmortem studies in not allowed. Cat 4 was lost to follow-up at approx- infants reveal fibrous intimal proliferation, medial imately 8 years of age. At the time of publication, hypoplasia, and loss of elastin fibers.3 Postmortem serial echocardiograms have been performed on the samples were not available from cats in this study remaining five cats. No progression in the degree of for comparison with the findings in humans. stenosis has been identified. A classification scheme for pulmonary artery stenosis based on the stenosis location and number of lesions has been developed in humans.1 The Discussion classification divides the pulmonary artery into four segments; (1) main pulmonary artery, (2) bi- Pulmonary artery stenosis is a well-known congen- furcation, (3) left- and right-branched pulmonary ital anomaly in humans that most often occurs in arteries, and (4) peripheral segmental pulmonary Table 2 Echocardiographic findings at initial diagnosis Cat Wt RVDd/LVDd RAD/LAD PVD PSD PSD/PVD PaS Spht PaD Dpht DFA (kg) (cm) (cm) (mmHg) (ms) (mmHg) (ms) (þ) 1 0.8 0.19 0.92 0.50 0.30 0.60 13.0 37 0.00 0 0 2 2.5 0.31 0.85 0.90 0.45 0.50 16.2 47 5.86 30 1 3 2.8 0.20 1.10 0.90 0.45 0.50 21.5 33 4.67 43 2 4 2.3 0.13 0.98 0.85 0.36 0.42 37.7 43 13.69 30 2 5 2.3 0.27 0.83 0.74 0.31 0.41 21.2 58 4.75 50 2 6 3.7 0.40 1.15 0.75 0.22 0.29 59.9 61 20.61 132 3 7 4.3 0.46 1.18 1.04 0.27 0.26 45.2 50 28.94 130 3 The cats were ranked from least to most severe based on severity of the echocardiographic PSD/PVD ratio. RVDd/LVDd ¼ ratio of right ventricular to left ventricular end-diastolic dimensions, RAD/LAD ¼ ratio of maximal right atrial diameter in long-axis to max- imal left atrial diameter in long-axis (diastole), PVD ¼ pulmonary valve annulus diameter (systole), PSD ¼ pulmonary artery steno- sis diameter (systole), PaS ¼ pulmonary artery stenosis peak systolic gradient, Spht ¼ pulmonary artery stenosis systolic pressure half-time, PaD ¼ pulmonary artery stenosis peak diastolic gradient, Dpht ¼ pulmonary artery stenosis diastolic pressure half-time, and DFA ¼ subjective grading of diastolic flow (0þ to 3þ). Weight (Wt) listed is that recorded at the time of initial diagnosis.
  • 88 D.P. Schrope, W.J. Kelch Of the seven cats, six cats were mixed breed, and no sex predilection was suggested (Table 1). The age at the time of diagnosis for most of these cats suggests a congenital lesion although a herita- ble basis could not be assessed since the majority were stray animals. All seven cats presented with basilar systolic murmurs although the point of maximal intensity varied (Table 1). Only one cat with severe stenosis and visible Doppler flow throughout diastole had a diastolic murmur. In humans with pulmonary ar- tery stenosis the presence of a diastolic murmur is variable.1,2 Exertional dyspnea was seen in one of the cats with severe stenosis. Exertional dyspnea is one of the most common clinical signs identified in humans with pulmonary artery stenosis, and is likely related to hypoperfusion of the pulmonary arterial tree.3 Based on the blood gas analysis, this was also the likely cause for exertional dyspnea in this cat. It is also possible that certain lesions could result in asymmetric blood flow to branched Figure 4 Selective right ventricular angiocardiogram pulmonary arteries further contributing to ventila- from cat 4 with distal pulmonary artery stenosis. Opaci- tioneperfusion mismatch. fication of the right ventricle (RV), main pulmonary ar- The available radiographs and ECGs did not tery (MPa), and right- (RPa) and left (LPa)-branched pulmonary arteries is evident. Mild opacification of the reveal specific findings that would differentiate right atrium and the caudal vena cava (CVC) is also pres- pulmonary artery stenosis from other causes of ent resulting from retrograde movement of the catheter basilar systolic murmurs. In fact, radiographic or back into the right atrium during injection. Note the ste- electrocardiographic abnormalities were limited nosis of the main pulmonary artery just proximal to the to cats with moderate to severe disease. bifurcation (black arrow) with normal branched pulmo- Although PaS and PaD were correlated to dis- nary arteries. ease severity, a ‘‘natural’’ cut-off between mod- erate and severe disease was not obvious to the arteries. Single lesions of the main pulmonary ar- authors (Table 2). Studies in human aortic coarcta- tery or branched pulmonary arteries are classified tion support the value of evaluating other diastolic as Type I pulmonary artery stenosis. Type I lesions flow parameters.6,8,9 In this study, Dpht and DFA are further subclassified into Type Ia (isolated ste- were statistically significant and there appeared nosis of the main pulmonary artery), Type Ib (iso- to be natural cut-offs between moderate and se- lated stenosis of the right-branched pulmonary vere disease using Dpht and DFA. One human paper artery), and Type Ic (isolated stenosis of the left- evaluating patients with aortic coarctation sug- branched pulmonary artery). Isolated lesions lo- gested that a Dpht of >100 ms was consistent cated at the bifurcation with extension into the with severe stenosis.8 Findings in this study suggest proximal branched pulmonary arteries are classi- that a similar value may be appropriate in cats fied as Type II. Lesions involving the peripheral with pulmonary artery stenosis. Both cats with segmental pulmonary arteries with no abnormali- a Dpht >100 ms and a DFA of 3þ had the most se- ties of the main pulmonary artery, bifurcation, or vere stenosis when graded by 2D echocardiography left- or right-branched pulmonary arteries are and clinical signs. classified as Type III. Lesions involving the periph- The discrepancy between systolic and diastolic eral segmental pulmonary arteries with additional Doppler findings and obstruction severity in hu- lesions in the main, left-branched, and/or right- mans with aortic coarctation is not clear. Causes branched pulmonary arteries are classified as may be related to changes in cardiac output, Type IV. Type Ia is one of the least common classes collateral blood flow, ductal flow, and the shape of pulmonary artery stenosis seen in humans.1 In and length of the stenosis.6e8 Except for the mor- contrast, the cats in this study showed no evidence phology of the stenosis, none of these would likely of other concurrent cardiac defects and the lesion be a cause for similar findings in the pulmonary ar- was classified as Type Ia in all cats. tery. It is possible that, with more severe stenosis,
  • Clinical and echocardiographic findings 89 the flexible walls of the main pulmonary artery 5. Liu SK. Pathology of feline heart disease. Vet Clin North Am proximal to the obstruction absorb kinetic energy 1977;7:323e39. 6. Houston AB, Simpson IA, Pollock JC, Jamieson MP, Doig WB, as the right ventricle contracts against the steno- Coleman EN. Doppler ultrasound in the assessment of sever- sis. This could result in a lower peak systolic driv- ity of coarctation of the aorta and interruption of the aortic ing force across the stenosis resulting in a lower arch. Br Heart J 1987;57:38e43. relative gradient. 7. Marx GR, Allen HD. Accuracy and pitfalls of Doppler evalu- The greatest limitations to this study are those ation of the pressure gradient in aortic coarctation. J Am Coll Cardiol 1986;7:1379e85. inherent to a retrospective study and the small 8. Carvalho JS, Redington AN, Shinebourne EA, Rigby ML, number of cats available. Furthermore, grading of Gibson D. Continuous wave Doppler echocardiography and severity was based on 2D echocardiographic find- coarctation of the aorta: gradients and flow patterns in ings and clinical signs. Ideally, disease severity the assessment of severity. Br Heart J 1990;64:133e7. would have been graded by angiocardiographic 9. Valdez-Cruz LM, Cayre RO. Coarctation of the aorta. In: Valdez-Cruz LM, Cayre RO, editors. Echocardiographic and intra-cardiac pressure data but these invasive diagnosis of congenital heart disease: an embryologic and diagnostics were believed to not be in the best anatomic approach. Philadelphia: Lippincott-Raven; 1999. interest for the majority of these cats. p. 475e82. 10. Houston AB, Sheldon CD, Simpson IA, Doig WB, Coleman EN. The severity of pulmonary valve or artery obstruction in children estimated by Doppler ultrasound. Eur Heart J References 1985;6:786e90. 11. Valdes-Cruz LM, Horowitz S, Sahn DJ, Larson D, Oliveria 1. Gay BB, French RH, Shuford WH, Rogers Jr JV. The roent- Lima C, Mesel E. Validation of a Doppler echocardiographic genologic features of single and multiple coarctations of method for calculating severity of discrete stenotic obstruc- the pulmonary artery and branches. Am J Roentgenol Ra- tions in a canine preparation with a pulmonary arterial dium Ther Nucl Med 1963;90:599e613. band. Circulation 1984;69:1177e81. 2. D’Cruz IA, Agustsson MH, Bicoff JP, Weinberg M, Arcilla RA. 12. Litster AL, Buchanan JW. Vertebral scale system to measure Stenotic lesions of the pulmonary artery. Clinical and hemo- heart size in radiographs of cats. J Am Vet Med Assoc 2004; dynamic findings in 84 cases. Am J Cardiol 1964 April:441e50. 216:210e4. 3. Kreutzer J, Landzberg MJ, Preminger TJ, Mandell VS, 13. Rishniw M, Erb HN. Evaluation of four 2-dimensional echo- Treves ST, Reid LM, Lock JE. Isolated peripheral pulmonary cardiographic methods of assessing left atrial size in dogs. artery stenosis in the adult. Circulation 1996;93:1417e23. J Vet Intern Med 2000;14:429e35. 4. Buchanan JW. Causes and prevalence of cardiovascular dis- 14. Boon JA. Acquired heart diseases. In: Boon JA, editor. Man- ease. In: Kirk RW, Bonagura JD, editors. Kirk’s current veteri- ual of veterinary echocardiography. Baltimore: Williams & nary therapy XI. Philadelphia: WB Saunders; 1992. p. 647e55. Wilkins; 1998. p. 261e382. Available online at www.sciencedirect.com