Senior Scholars Program

UMMS Affiliation

Department of Internal Medicine

Faculty Advisor

Pape, Linda



Document Type


Medical Subject Headings

Mitral Valve; Aorta; Heart Defects, Congenital; Echocardiography; Reference Standards


Cardiology | Cardiovascular Diseases | Health Services Research


Background:Bicuspid aortic valve (BAV) is the most common congenital heart malformation with a reported prevalence of 0.5% to 2% in the general population (Edwards, 1961; Roberts 1970; Hoffman 2002). BAV results in considerable morbidity and mortality due to its well known correlation with many pathological cardiovascular conditions including aortic stenosis, aortic regurgitation, infective endocarditis, and aortic dissection (Ward, 2000). Recently, it has been appreciated that even normally functioning BAVs are predisposed to dilation of the aortic root and ascending aorta, leading to potentially fatal aneurysm formation and rupture when progressive (Hahn, 1992; Nistri, 1999; Keane, 2000; Fedak, 2002; Alegret, 2003; Novaro, 2003; Pape, 2003). Many have sought to further characterize aortic dilation in BAV patients using echocardiographic measurement (Nkomo, 2003; Dore, 2003; Yasuda, 2003; Andrus, 2003; Berkoukhim, 2006; Schaefer, 2007; Cecconi, 2005; Warren, 2006; Borger, 2004). However, these studies often include only limited views of the ascending aorta and can potentially miss dilation. Published methods vary but include measurement of the aorta at 1cm beyond the sinotubular junction (STJ), 2-3cm beyond the STJ, some at “maximum size” of the ascending aorta, and others at the “level of the pulmonary artery.” Some reports even neglect to specify a location. In addition, measurements are often made at various stages of the cardiac cycle, with end-diastole and end-systole being the most commonly employed.

Objective:This study sought to understand the implications of echocardiographic measurement of the ascending aorta in BAV patients at various locations and during different phases of the cardiac cycle. There is a lack of data investigating the significance of this diagnostic variability. We are optimistic that our findings will encourage adoption of a standardized method for aortic measurement in BAV patients.

Methods:We selected 78 BAV patients undergoing echocardiography as part of a large prospective trial studying the association of BAV and aortic dilation. We subsequently screened 204 age and sex matched controls with tricuspid aortic valves undergoing routine clinical echocardiograms performed from November 2005 – December 2006 at the University of Massachusetts Medical School. Patients with stage I/II hypertension, elevated cholesterol, diabetes, cardiac disease, Marfan’s syndrome, history of prior surgery on the aorta or aortic valve, severe valvular dysfunction, aortic coarctation, or associated congenital cardiac defects were excluded.

Echocardiographic analysis:Echocardiographic studies were performed using standard commercially available equipment (Hewlett Packard Sonos 1000, 2500, 4500 and 5500; Hewlett Packard, Andover, Massachusetts). All echocardiograms were read by experienced cardiologists who assessed aortic regurgitation, aortic stenosis, and ejection fraction. Stenosis was defined as mild if the peak velocity was between 2.5-2.9 m/s, moderate if it was between 3.0-4.0 m/s, and severe over 4.0 m/s. American Society of Echocardiography criteria were used in assessing aortic insufficiency and ejection fractions. Aortic diameters were measured on parasternal 2D images from inner edge to inner edge of the aortic lumen perpendicular to the long axis at three sites: 1cm, 2cm, and 3cm from the STJ (Figure 1). A fourth measurement was made at 4cm when possible. Patients with poor quality images or studies where less than 3cm of ascending aorta could be visualized were excluded. Each set of measurements was made at three points in the cardiac cycle: end-diastole (defined as the onset of the QRS), mid-systole (defined as the time just after aortic valve opening), and end-systole (defined as the time just before aortic valve closure. One author (AA) reviewed the images and measured aortic diameters for all subjects. These measurements were repeated in a subset of 18 patients in end-diastole and end-systole (108 data points). A second observer (LP) repeated measurements in a subset of 10 patients in end-diastole (30 data points). This data was used to assess both intra- and interobserver variabilities.

Statistical analysis: Data are expressed as mean ± SD. The upper limit of normal for aortic diameter was defined as control +2SD at each location and was used to determine the percentage of BAV patients with aortic dilation. Differences between groups and across times were evaluated by fitting a general linear mixed model (repeated measures ANOVA). In the presence of significant effects, pairwise comparisons were performed using Tukey’s HSD test with the covariance matrix estimated from the ANOVA. The distributional characteristics of the outcomes was evaluated both graphically by visual inspection of model residuals and by the Kolmogorov-Smirnov Goodness of Fit test for normality. A p value of Results: Of 282 subjects, 192 were excluded based on the above criteria to arrive at a final study population of 45 BAV patients and 45 matched controls. Median age of participants was 47.5 years (range 22 to 69) and 60% were men. There were no significant differences between the BAV and control populations with respect to height, weight, body surface area, heart rate, or blood pressure. Demographic data are summarized in Table 1. All patients in the study had normal systolic function. The median ejection fraction of the BAV patients was 63% (range 45 to 70), similar to that of the control population (range 50 to 75). Ten BAV patients (22%) and forty control subjects (89%) had completely normal valvular function. Nineteen BAV patients (42%) had mild aortic stenosis and two (4%) had moderate stenosis. Sixteen BAV patients (36%) had mild aortic regurgitation and twelve (27%) had moderate regurgitation. There were five control subjects (11%) with mild aortic regurgitation. None of the patients included in this study had severe valve abnormalities. There were minimal intra- and interobserver variabilities in the measurement of the aortic dimensions at any site. Squared correlation coefficients (R2) for intraobserver variability were 0.94 for 54 measurements (18 patients) conducted in end-diastole and 0.90 for 54 measurements (18 patients) conducted in end-systole. The squared correlation coefficient (R2) for interobserver variability was 0.92 for 30 measurements (10 patients) conducted in end-diastole (Figure 2). We were able to measure the diameter of the aorta 4cm beyond the STJ in 28 BAV patients. The echo studies of the remaining 17 BAV patients did not have adequate visualization of the ascending aorta to permit measurement at this distance. A graphical representation of these results demonstrates continued dilation of the ascending aorta up to 3cm beyond the STJ in some patients (Figure 3). We qualitatively noted minimal increases in aortic diameter beyond 3cm in this subset of our study population. The mean aortic diameters for 45 BAV patients and 45 control subjects measured at 1cm, 2cm, and 3cm beyond the STJ in end-diastole, end-systole, and mid-systole are listed in Table 2. In all three phases of the cardiac cycle we observed larger diameters at 3cm than at 1cm. In addition, average diameters were consistently largest in mid-systole and smallest in end-diastole. The upper limits of normal for ascending aorta dimensions as determined by our control population are listed in Table 3. The number and percentage of the 45 BAV patients who exceeded these criteria is also displayed using various measurement methodologies. In all phases of the cardiac cycle, significantly more dilated aortas were detected by measuring the aorta at least 2cm beyond the STJ. A marginal increase in the detection of dilation was achieved by measuring to 3cm. The mean increases in aortic diameter observed between three measurement intervals in various phases of the cardiac cycle are displayed in Table 4. In our BAV population, we observed a considerable increase in diameter between 1 and 2cm with a smaller increase between 2 and 3cm. These results are displayed graphically in Figures 4 & 5 for measurements made in end-systole. Importantly, a large number of patients demonstrated ≥ 5mm increase in diameter at 3cm compared with 1cm: 15, 10, and 12 patients measured in end-diastole, end-systole, and mid-systole respectively.

Discussion:In this study we demonstrate that echocardiographic measurement of the ascending aorta in BAV patients is significantly influenced by both location and phase of the cardiac cycle. We found that BAV patients often continue to have dilation of their ascending aorta up to 3cm beyond the STJ and measurement below this level could potentially miss dilation. In addition, assessment of aortic size at inconsistent phases of the cardiac cycle may also compromise diagnostic accuracy. The results of this study underscore the need to adopt a standardized method for echocardiographic measurement of the ascending aorta in BAV patients. We believe that a standardized method will increase detection of aortic dilation, improve reliability of serial measurements, and promote uniformity across laboratories. A dilated aorta is often the first indication of a bicuspid valve and usually prompts further clinical investigation for underlying cardiovascular pathology. Failure to recognize aortic dilation would represent a significant detriment to the patient. We recommend that all BAV patients undergoing echocardiography have aortic measurements made at least up to 3cm beyond the STJ in end-systole. This method takes advantage of the increased visualization of the ascending aorta afforded in end-systole without risking possible overestimation of aortic size that can occur with measurement in mid-systole. No literature has been published on the ability of sonographers to reliably obtain 3cm of ascending aorta on routine echocardiographic studies, and this would be an area of future investigation.





To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.