Background
Assessment of mitral regurgitation (MR) severity by echocardiography is important for clinical decision making, but MR severity can be challenging to quantitate accurately and reproducibly. The accuracy of effective regurgitant orifice area (EROA) and regurgitant volume (RVol) calculated using two-dimensional (2D) proximal isovelocity surface area is limited by the geometric assumptions of proximal isovelocity surface area shape, and both variables demonstrate interobserver variability. The aim of this study was to compare a novel automated three-dimensional (3D) echocardiographic method for calculating MR regurgitant flow using standard 2D techniques.
Methods
A sheep model of ischemic MR and patients with MR were prospectively examined. Patients with a range of severity of MR were examined. EROA and RVol were calculated from 3D color Doppler acquisitions using a novel computer-automated algorithm based on the field optimization method to measure EROA and RVol. For an independent comparison group, the 3D field optimization method was compared with 2D methods for grading MR in an experimental ovine model of MR.
Results
Fifteen 3D data sets from nine sheep (open-chest transthoracic echocardiographic data sets) and 33 transesophageal data sets from patients with MR were prospectively examined. For sheep data sets, mean 2D EROA was 0.16 ± 0.05 cm 2 , and mean 2D RVol was 21.84 ± 8.03 mL. Mean 3D EROA was 0.09 ± 0.04 cm 2 , and mean 3D RVol was 14.40 ± 5.79 cm 3 . There was good correlation between 2D and 3D EROA ( R = 0.70) and RVol ( R = 0.80). For patient data sets, mean 2D EROA was 0.35 ± 0.35 cm 2 , and mean 2D RVol was 58.9 ± 52.9 mL. Mean 3D EROA was 0.34 ± 0.29 cm 2 , and mean 3D RVol was 54.6 ± 36.5 mL. There was excellent correlation between 2D and 3D EROA ( R = 0.94) and RVol ( R = 0.84). Bland-Altman analysis revealed greater interobserver variability for 2D RVol measurements compared with 3D RVol using the 3D field optimization method measurements, but variability was statistically significant only for RVol.
Conclusions
Direct automated measurement of proximal isovelocity surface area region for EROA calculation using real-time 3D color Doppler echocardiography is feasible, with a high correlation to current 2D EROA methods but less variability. This novel automated method provides an accurate and highly reproducible method for calculating EROA.
Mitral regurgitation (MR) is one of the most prevalent forms of valvular malfunction, and moderate to severe MR, even if asymptomatic, carries a 5-year mortality rate of up to 14% if untreated, because of the high likelihood that left ventricular dysfunction will develop. Surgical treatment of MR has improved life expectancy, but the accurate quantification of MR severity is important for decisions regarding the timing of surgical intervention and prognostication.
Echocardiography is the primary clinical tool used for the evaluation of the mechanism and severity of MR. The traditional approach for the assessment of MR severity by two-dimensional (2D) echocardiography requires the integration of qualitative and quantitative measures. Existing quantitative measures such as the effective regurgitant orifice area (EROA) and regurgitant volume (RVol) using the 2D proximal isovelocity surface area (PISA) are time consuming to perform, based on geometric assumptions, and subject to interobserver variability. Hence, an automated measurement of the flow convergence zone could improve the quantification of MR severity.
Three-dimensional (3D) echocardiography has been shown to provide more accurate measurements for ventricular volumes and function compared with 2D echocardiography. Advances in the technology of 3D echocardiography now enable semiautomated MR quantification on the basis of the 3D PISA method, with recent studies demonstrating that real-time 3D full-volume color Doppler acquisition of the regurgitant jet and 3D PISA analysis are feasible and potentially more accurate method for quantifying MR. However technical limitations, such as multiple gated acquisition and time-consuming manual processing, prevent its becoming widely adopted into daily clinical practice.
The field optimization method (FOM) is a novel algorithm that adopts a computed approach of nonlinear curve fitting of Doppler signals within the flow convergence zone, allowing an automated estimation of MR RVol. Essentially, for each 3D color flow volume, the FOM iterates on a fluid dynamics model that, when processed by a model of ultrasound physics, attempts to agree with the observed velocities in a least squares sense. The output of this model is an estimate of the regurgitant flow and the location of its associated orifice. The FOM algorithm does not make any orifice shape assumption; the algorithm uses the nonlinear curve-fitting technique, iterating on the parameters of both the orifice position and regurgitant flow. In a previous study, we found that a setting with the orifice position locked offers the best performance and robustness of the current 3D FOM. This algorithm develops a 3D map of the flow convergence zone for the PISA measurement from 3D color Doppler velocities and hence is not confined by the geometric assumptions of 2D techniques. As this algorithm has been successfully validated in in vitro models, and ex vivo models, the aims of this study were to (1) compare and validate the 3D FOM against standard 2D MR quantitation methods in an experimental in vivo model of MR and (2) examine the assessment of MR severity from 3D color Doppler echocardiographic images using this novel automated algorithm with standard 2D measurements in a clinical setting.
Methods
Animal Studies
All animal experiments had approval from the Massachusetts General Hospital Institutional Subcommittee on Research Animal Care. A chronic ischemic MR model in sheep was used. Briefly, Polypay sheep were anesthetized with sodium thiopental (12.5 mg/kg intravenously), and the trachea was intubated and ventilated at 15 mL/kg with a mixture of 2% isoflurane and oxygen. All animals received glycopyrrolate (0.4 mg intravenously) and vancomycin (0.5 g intravenously) 1 hour before incision. The heart of each animal was exposed through a thoracotomy, and the second and third circumflex marginal branches were ligated to infarct the inferoposterior wall. The chest was closed and the animals cared for during a further 8 weeks after the induced infarct, before being subjected to a second thoracotomy for echocardiographic evaluation of chronic MR.
Patient Studies
The protocol for this study had approval from the Massachusetts General Hospital Institutional Review Board. Only patients with (1) the presence of at least mild MR in the standard 2D color Doppler evaluation of regurgitant jet size, (2) the absence of concomitant lesions (more than mild aortic stenosis, aortic insufficiency, or mitral stenosis), and (3) only one flow convergence region were included in this prospective study. Transesophageal echocardiograms were obtained because the higher resolution images of transesophageal echocardiography, multiplane capabilities, and proximity to the mitral valve makes PISA measurement easier and quantification likely more accurate. For patients with atrial fibrillation, an average of measurements from two cycles each containing two beats was used.
Echocardiography
All sheep 2D and 3D data sets were sequentially acquired using the iE33-xMATRIX ultrasound system (Philips Medical Systems, Andover, MA). All sheep 2D and 3D data sets were sequentially acquired using a 5-MHz X5-1 transthoracic probe applied directly on the exposed heart following thoracotomy (i.e., open-chest epicardial imaging; Philips Medical Systems). All patient 2D and 3D data sets were sequentially acquired by transesophageal echocardiography using the iE33-xMATRIX ultrasound system and a 7-MHz X7-3t transesophageal probe (Philips Medical Systems). Regurgitant jets were classified as eccentric if they were in close contact with the mitral leaflet behind the regurgitant orifice and impinged the medial or lateral wall of the left atrium, whereas central jets were initially directed into the center of the left atrium. The severity of MR was assessed offline from 2D transesophageal images using quantitative measures (EROA and RVol by PISA) and the methodology outlined in the guidelines of the American Society of Echocardiography by level 3 trained echocardiographers. Visualization of the PISA was optimized by baseline shifting the color Doppler aliasing velocity. The maximal velocity of the regurgitant jet was determined using the continuous-wave Doppler method. The EROA was calculated using the formula:
EROA = [ 2 π × ( PISA radius ) 2 × V aliasing ] / V max ,
For the animal studies, the regurgitant MR volume was also determined using a separate volumetric method. In this method, MR volume was obtained by comparing the difference between the measured stroke volume (3D ventricular end-diastolic volume − end-systolic volume) and the calculated stroke volume by the Doppler method.
Three-dimensional EROA and RVol were obtained from four- or six-beat full-volume 3D color Doppler acquisitions in real time and analyzed offline using custom propriety software (Philips Medical Systems), automated flow quantification, which uses the FOM algorithm. Briefly, the custom software measures the 3D PISA using FOM from 3D color data sets using the same simple hemispheric fluid dynamics model as in PISA. The 3D FOM algorithm first makes initial guesses as to the underlying instantaneous flow rate, orifice location, and orifice shape on the basis of the acquired 3D velocity vector field on the proximal side of the orifice. From these three parameters, the software produces an initial model of the 3D velocity vector field. Comparing the modeled field with the observed field, it then iterates on these three parameters until the two fields agree (on a least mean squares basis). Finally, the 3D FOM algorithm estimates RVol by integrating the instantaneous flow rate over all systolic volumes. One of the optional settings in the FOM program is also to enable “multiple-orifice” simulation. This setting models complex (e.g., slit) orifices as two or more discrete, simple “round orifices.” We disabled this setting because our trials thus far have revealed greater robustness using a single orifice. The orifice location was then locked and its position specified by the user (similar to 2D PISA), and the ability of the algorithm to adapt to the orifice shape was disabled, limiting the model to a round orifice. These settings were chosen on the basis of their superior performance in our previous study and to improve the robustness of the 3D FOM ( Figure 1 ). The overall work flow involved in the application of this algorithm is also very simple because of its semiautomated features. Essentially, planar reconstruction was used to identify the plane of the mitral regurgitant orifice, which is perpendicular to the mitral regurgitant jet as visualized by color Doppler on a QLAB station (Philips Medical Systems). A marker corresponding to a reference point of interest is used to mark the origin of the regurgitant orifice in the 3D data set, and when locked, the algorithm then activated to calculate EROA and RVol. Interobserver variability was also assessed. A second independent blinded observer analyzed the same full-volume acquisition of the PISA for 10 randomly selected cases. Readers of 3D data sets were blinded to 2D values.
