Two-Dimensional and M-Mode Echocardiographic Predictors of Disease Severity in Dogs With Congenital Subaortic Stenosis
Echocardiographic studies from 50 dogs with congenital subaortic stenosis were examined. The degree of concentric, left-ventricular hypertrophy as assessed by M-mode measurement demonstrated a positive relationship (P <0.05) to disease severity. However, the clinical utility of these measures is hindered by a large amount of individual variation (r2=0.243 to 0.473). Two-dimensional ultrasound was used to compare the cross-sectional area of the left-ventricular outflow tract to the cross-sectional area of the aortic root. The ratio of these two areas demonstrated a strong inverse relationship (P=0.001; r2=0.778) with disease severity. This ratio provides a method of estimating severity of disease by two-dimensional echocardiography.
Introduction
Subaortic stenosis (SAS) is caused by an abnormal ring or ridge of tissue that projects from the endocardial surface and encroaches into the lumen of the left-ventricular outflow tract (LVOT), thereby reducing the cross-sectional area through which blood can travel.1 The LVOT can be described as the cylindrical area directly below the aortic valve, bounded anteriorly by the interventricular septum and posteriorly by the anterior leaflet of the mitral valve.2 Ejection of left-ventricular blood volume is impeded by the stenosis, creating an environment of systolic pressure overload. The cardiac response to pressure overload is the development of concentric hypertrophy, which is recognized during echocardiography as an increase in myocardial wall thickness relative to the radius of the ventricular chamber. This adaptive response helps to normalize myocardial wall stress and preserve normal systolic function.
Presumptive diagnosis of SAS can be made based on physical examination, electrocardiography (ECG), and thoracic radiographs, although definitive diagnosis usually involves echocardiography. Three main findings form the basis of the echocardiographic diagnosis and include a narrowed LVOT, left-ventricular concentric hypertrophy (LVH), and increased blood-flow velocity within the stenotic LVOT.34 These features can be identified with the use of two-dimensional (2DE), M-mode, and Doppler echocardiography, respectively. Echocardiography is also invaluable for the assessment of disease severity. Previous studies have shown that the majority of dogs with severe disease experience sudden death at a young age, while dogs with mild disease remain clinically unaffected.5 Currently, the evaluation of blood-flow velocity by Doppler ultrasound is the most widely used index to grade severity of disease.5 The correlation between disease severity and M-mode and 2DE parameters (namely, the degree of LVH and the dimensions of the LVOT) has not been described. The purpose of this study was to examine the magnitude of LVH and cross-sectional area of the LVOT in dogs with congenital SAS.
Materials and Methods
Case records and echocardiograms from 50 cases of SAS diagnosed between October 1994 and January 1996 at the Veterinary Medical Teaching Hospital, University of California at Davis, were reviewed. Diagnosis of SAS was based on the following criteria: the presence of a left heart-base systolic murmur, blood-flow velocity across the LVOT >2.1 meters per second as measured by continuous-wave Doppler examination, and the exclusion of other congenital or acquired diseases that could elevate LVOT blood-flow velocities (e.g., patent ductus arteriosus).
Standard 2DE, M-mode, and Doppler echocardiograms were performed with dogs in right and left lateral recumbency using a 5.0- or 3.5-mHz imaging transducer. Additional continuous-wave Doppler studies were performed from a subcostal position with a dedicated 1.9-mHz transducer. One of two commercially available ultrasound machinesa,b was used to perform echocardiograms, and studies were recorded on standard half-inch S-VHS videotape. Videotaped studies were reviewed and reanalyzed. The following M-mode measurements were tabulated: left-ventricular internal dimension at end diastole (LVIDd) and end systole (LVIDs), diastolic thickness of the interventricular septum (IVSd), and diastolic thickness of the left-ventricular posterior wall (LVPWd). In cases where the quality of the M-mode study was deemed inadequate for obtaining these values (n=4), measurements were made using the 2DE right parasternal short-axis or long-axis views. In this technique, end-diastolic and end-systolic frames were frozen and electronic calipers were used to measure from leading edge to leading edge of the cardiac structure of interest.
The cross-sectional area of the LVOT was measured from the right parasternal short-axis view. The areas were determined planimetrically from the 2DE image by tracing the circumferential outline of each structure along the inner edge of the myocardium [Figure 1]. The cross-sectional area of the aortic root (Ao) was also measured by the same planimetric technique. The measurement was made at the level of the sinuses of Valsalva during diastole. The ratio between the two areas was expressed as LVOT:Ao. All 2DE and M-mode measurements were done in triplicate, and the resultant average was used for statistical analysis. In addition to the 50 affected dogs, 12 client-owned dogs that were deemed free of cardiac disease by physical examination and echocardiography were prospectively examined to establish normal values for the LVOT:Ao ratio by the technique previously described. None of the control dogs possessed a cardiac murmur, and echocardiographic examination indicated that the blood-flow velocity across the LVOT did not exceed 1.7 meters per second. This value falls within the normal reference range of LVOT velocities in dogs.6
Within the affected population, severity of disease was determined by examining the maximal velocity of blood flow across the LVOT from either the left parasternal or subcostal view. The single highest value obtained from either of these locations was then used to calculate the instantaneous pressure gradient (PG) between the left ventricle and aorta via the modified Bernoulli equation:2
\(PG\ (mm\ Hg)\ =\ 4\ {\times}\ (blood{-}flow\ velocity\ across\ LVOT\ [meters\ per\ second])^{2}\)Affected dogs were categorized into three levels of disease severity, based on their Doppler-derived PG. Dogs with a PG from 18 mm Hg to 50 mm Hg were classified as having mild disease. Dogs with a PG of 50 mm Hg to 80 mm Hg were classified as having moderate disease, and dogs with a PG >80 mm Hg were classified as having severe disease.
The left-ventricular mass (LVM) of affected dogs was calculated using the cube formula of Teichholz, where b = the specific density of myocardial tissue = 1.05 g/cm3.7
\(LVM(g)\ =\ \frac{7\ {\times}\ (LVIDd\ +\ IVSd\ +\ LVIDd)^{3}}{(2.4\ +\ LVIDd\ +\ IVSd\ +\ LVPWd)}{-}\ \frac{7(LVIDd)^{3}}{2.4\ +\ LVIDd}{\times}\ 1.05g/cm^{3}\)In order to standardize the degree of concentric LVH among dogs of different sizes, IVSd and LVPWd values from each dog were indexed to their LVIDd. For dogs whose weight was recorded at the time of the ultrasound (n=36), IVSD, LVPWd, and LVM measurements were also indexed to body weight (BW) in kg and body surface area (BSA) in m2.
Statistical Analysis
Differences in age, weight, and sex of the two populations were determined by the Mann-Whitney U or chi-square test. Two-dimensional and M-mode measures were adjusted for age and sex using multivariate analysis and were plotted against the Doppler-derived PG. From these linear regressions, coefficients of determination were calculated. Log values were used for the analysis of the LVOT:Ao ratio. Differences between the mean LVOT:Ao ratio of the two populations were determined by an unpaired t-test. Differences between the categories of disease severity were determined by analysis of variance (ANOVA) and the Tukey Studentized Range Test. Data is reported as mean values ± standard error (SE). Significance was set at a P value of ≤0.05.
Results
There were no significant differences in the age or sex of the affected versus the control population, as shown in Table 1. The Newfoundland, golden retriever, rottweiler, boxer, and German shepherd dog comprised 40 (80%) of the 50 affected dogs. A complete listing of breeds is presented in Table 2. Doppler examination of the affected dogs revealed a mean and median PG of 100.9±10.6 mm Hg and 76 mm Hg, respectively. The PG ranged from 18.5 to 303.5 mm Hg. In the control population, the PG did not exceed 11.5 mm Hg for any dog. Separation of the 50 affected dogs into the three classes of disease severity was performed based on the PG, resulting in 23 dogs with severe disease, 10 dogs with moderate disease, and 17 dogs with mild disease [Table 3].
Values of the eight indices of hypertrophy were plotted against the PG and demonstrated a significant and positive relationship to the severity of disease [Table 4]. The coefficients of determination were relatively low for all measures of LVH, ranging from 0.243 to 0.473. The LVOT:Ao ratio demonstrated a significant and inverse relationship to the PG [Figure 2]. Compared to measures of concentric LVH, the LVOT:Ao ratio possessed a greater correlation to disease severity with a r2 value of 0.778. There was no effect of gender or age on indices of hypertrophy or on the LVOT:Ao ratio.
The average LVOT:Ao ratio of the affected population (0.408±0.031; range, 0.105 to 0.977) was significantly smaller (P=0.0001) than that of the control population (0.637±0.023; range, 0.509 to 0.737). Within the affected population, there was a significant difference (P≤0.05) of the mean LVOT:Ao ratio between all three levels of disease severity. The LVOT:Ao ratios in dogs with mild, moderate, and severe disease were 0.656±0.034, 0.391±0.030, and 0.232±0.023, respectively. When the control population was compared to the three levels of disease, dogs with moderate and severe disease had significantly lower ratios than the control population [Figure 3]. There was no difference between dogs with mild disease and the controls.
Classification of Disease Severity Based on PG and LVOT:Ao Ratio
Affected dogs were divided according to disease severity, based on two different methods. Dogs had originally been classified based on their Doppler-derived PG [Table 3]. The results of this study allowed dogs to be classified based on their LVOT:Ao values. Dogs with mild disease were defined as having a ratio >0.5; dogs with moderate disease were defined as having a ratio of 0.3 to 0.5; and dogs with severe disease were defined as having a ratio <0.3. The level of agreement between the Doppler-derived PG and LVOT:Ao methods was compared, and agreement was found in 42 (84%) of the 50 dogs. This included 20 out of 23 dogs with severe disease, seven of 10 dogs with moderate disease, and 15 of 17 dogs with mild disease. In eight (16%) of 50 dogs, the two methods were in disagreement [Figure 4].
Discussion
Concentric Hypertrophy
The results of this study demonstrate a positive relationship between the degree of pressure overload and the magnitude of concentric LVH. This result reflects the fact that concentric LVH is a compensatory response, helping maintain normal systolic function in the face of pressure overload. If the response were perfectly controlled, the amount of hypertrophy for any given pressure load would be uniform and just enough to normalize the wall stress. The authors’ results reveal that the correlation between indices of hypertrophy and severity of disease is low, suggesting substantial individual variability in the magnitude of this compensatory response. This variability appears to limit the clinical usefulness of this index in predicting disease severity. Variability of concentric LVH has been previously reported in human patients with aortic stenosis.8–10 DePace found that measures of LVH correlated poorly with the severity of disease in adults with aortic stenosis (r=0.17).11 The etiology of the heterogeneous response may be related to gender and other inherited traits.12 Human males develop less hypertrophy than females and experience greater incidence of myocardial failure.1314 No differences in the magnitude of LVH due to gender were found in the present study. The translation from pressure overload to concentric LVH is complex and multifactorial. A fundamental cipher of this translation is the production of angiotensin II, a potent stimulator of myocardial growth.15–18 In humans, insertion (I)/ deletion (D) polymorphisms of the angiotensin-converting enzyme gene have been correlated with degree of hypertrophy.1920 Patients possessing the D/D polymorphism have greater degrees of LVH than patients with the I/D or I/I polymorphism for any given degree of disease severity.
Variability of LVH is an important consideration, given that patients who develop minimal LVH are more likely to experience loss of systolic function and congestive heart failure, while human patients with excessive LVH enjoy super-normal cardiac performance but are more predisposed to arrhythmias and sudden death.21 The detrimental effects of LVH reflect the evolution of diastolic dysfunction and decreased myocardial perfusion. Excessive LVH induces myocardial ischemia and creates a myocardial environment primed for the development of malignant arrhythmias. The presence of pathological LVH is a well-documented risk factor for adverse cardiac events, including sudden death.2223 Current therapy for SAS involves treatment with beta-adrenergic blockers that reduce heart rate and improve myocardial perfusion, thus shifting the balance of oxygen supply and demand back to a more favorable level. Although previous authors have shown that severity of canine disease, based on the Doppler PG, correlates with survival, the specific effect of LVH on patient outcome has yet to be determined.5
LVOT:Ao Ratio
Evaluation of the LVOT:Ao ratio may offer several advantages over the traditional Doppler-based PG. Pressure is the product of resistance and flow and is therefore dependent on both the anatomical dimensions of the stenosis (i.e., resistance) and cardiac output (i.e., flow). An examiner relying only on the Doppler PG disregards the effects of flow by assuming each patient has a normal cardiac output. In certain instances, this assumption may lead to erroneous results. In dogs with concurrent myocardial failure and hence low cardiac output, the PG will underestimate disease severity, whereas in dogs that are excited and sympathetically driven, cardiac output will be elevated and the PG will overestimate disease severity. The direct measurement of the stenotic region via the LVOT:Ao ratio is unaffected by flow and should be accurate across a wide range of cardiac outputs.
The results of this study indicate a significant difference in the LVOT:Ao ratios between the normal and affected populations. Wingfield described a similar finding in 11 dogs with SAS but did not attempt to correlate the ratio with disease severity.4 In the present study, the LVOT:Ao ratio was able to distinguish between three levels of disease severity. In dogs with Doppler-confirmed SAS, the LVOT:Ao ratio and the Doppler PG have a high level of agreement. In the study population, a ratio of >0.5 indicated mild disease, ratios between 0.3 and 0.5 indicated moderate disease, and a ratio <0.3 indicated severe disease. In dogs with physical examination findings consistent with SAS (i.e., left basilar heart murmur), LVOT:Ao ratios of 0.3 to 0.5 suggest moderate disease while values <0.3 suggest severe disease.
Study Limitations
The LVOT is a relatively small structure and can be difficult to image. Measurements of this region are often confounded by systolic motion of the heart, which tends to move the LVOT out of the imaging plane. Accurate measurements of this area are particularly difficult in animals that are panting or resistant to restraint. The LVOT:Ao ratio in normal and mildly affected dogs was not significantly different and cannot be used to differentiate between these two groups. This finding may reflect the inherent inaccuracy in measuring the LVOT, as well as the fact that in dogs with mild disease the anatomical narrowing of the LVOT is extremely subtle. The LVOT:Ao ratio is probably best suited to help confirm a diagnosis made in conjunction with conventional Doppler ultrasound. Further studies should be performed to determine the utility of the ratio in dogs with depressed cardiac function and poor cardiac output.
There is considerable debate involving the specific Doppler velocity that separates normal dogs from those with the mildest form of SAS. The subtle nature of the anatomical lesion and its expected effect on the LVOT velocity, as well as the impact of cardiac output on the Doppler study, strongly suggests that this “magic number” does not exist. The current diagnostic gold standard involves postmortem examination of the LVOT,1 a technique that is obviously impractical for studies involving client-owned animals. Despite the authors’ best efforts to define a population of mildly affected dogs, it is conceivable that normal dogs were inadvertently included and affected the results of this study. Until more sensitive and specific diagnostic tools are available (e.g., genetic testing), this difficulty will continue to hinder diagnosis of the mildest forms of disease.
Cardiac ultrasound machine, 128XP/10; Acuson Corporation, Mountain View, CA.
Cardiac ultrasound machine, ImagePoint; Hewlett-Packard Corporation, Palo Alto, CA
Acknowledgments
The authors thank Dr. Gary L. Wood for assistance with clinical evaluation of dogs and Dr. Randy Singer, Felicia Trachtenberg, and Dr. David Schaeffer for assistance with statistical analysis of data.
Citation: Journal of the American Animal Hospital Association 38, 3; 10.5326/0380209
Citation: Journal of the American Animal Hospital Association 38, 3; 10.5326/0380209
Citation: Journal of the American Animal Hospital Association 38, 3; 10.5326/0380209
Citation: Journal of the American Animal Hospital Association 38, 3; 10.5326/0380209
Examples of echocardiograms used to determine the areas of the left-ventricular outflow tract (LVOT) and aorta (AO) in dogs with mild and severe subaortic stenosis. All images are center-sided parasternal short-axis views at the level of the LVOT or AO. The area measurements have been determined planimetrically by tracing the outline of the LVOT and AO. LA indicates the left atrium and RV the center ventricle.
Relationship between the Doppler-derived pressure gradient (PG) and the ratio of left-ventricular outflow tract area to aortic root area (LVOT:Ao) in 50 dogs with congenital subaortic stenosis.
The left-ventricular outflow tract area to aortic root area ratio (LVOT:Ao) in the control population and dogs with mild, moderate, and severe subaortic stenosis. Values that are statistically different from each other (P<0.05) possess different numbers of asterisks. Error bars indicate the standard error of the values.
Fifty dogs with subaortic stenosis grouped into nine cells of disease severity based on their Doppler-derived pressure gradients and the ratio of left ventricular outflow tract area to aortic root area (LVOT:Ao). Agreement between the two methods occurred in 42 (84%) patients.
Contributor Notes
