Article Access Statistics   Viewed  377   Printed  6   Emailed  0   PDF Downloaded  56   Comments  [Add]  

Click on image for details.



Year : 2021
 Volume
: 24  Issue : 2  Page
: 163171 

Intraoperative comparison of 2D versus 3D transesophageal echocardiography for quantitative assessment of mitral regurgitation 

Pravin S Lovhale^{1}, Shrinivas Gadhinglajkar^{2}, Rupa Sreedhar^{2}, Subin Sukesan^{2}, Vivek Pillai^{3}
^{1} Consultant Cardiac Anaesthesia, Raheja Hospital, Mumbai, Maharashtra, India ^{2} Department of Anesthesia, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum, Kerala, India ^{3} Department of CVTS, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum, Kerala, India
Click here for correspondence address and
email
Date of Submission  22Jan2020 
Date of Decision  30May2020 
Date of Acceptance  26Jun2020 
Date of Web Publication  19Apr2021 




Abstract   
Background: Effective regurgitant orifice area (EROA) can be represented by 3D echocardiographic vena contracta crosssectional area (3DVCA) as a reference method for the quantification of mitral regurgitation (MR) without making any geometrical assumptions. EROA can also be derived from 3D PISA technique with a hemispherical (HS) or hemielliptical (HE) assumption of the proximal flow convergence. However, it is not clear whether HSPISA and HEPISA has better agreement with 3DVCA. Aims: This study was conducted to compare the EROA and Rvol obtained from 3DVCA with those obtained from 2DVC, 2DHSPISA, 3DHSPISA, and 3DHEPISA. Setting: Tertiary care hospital. Design: Prospective observational study. Materials and Methods: After anesthesia induction, 43 consecutive patients were evaluated with RT3DTEE after acquiring images from midesophegeal views and performing the offline analysis of volume dataset. 3DVCA was measured using multiplanar reconstruction mode and EROA and regurgitant volume were estimated using HSPISA and HEPISA methods. The HEPISA was calculated by using the Knud Thomsen formula. Statistical Analysis: Agreement between methods to estimate EROA and regurgitant volumes were tested using Bland–Altman analysis. The interobserver variability and intraobserver variability were assessed using an intraclass correlation coefficient. Results: The EROA estimated by 3DVCA was larger than EROA obtained by 2DHSPISA and 3DHSPISA, which were significantly greater than 3DHEPISA. 3DHSPISAEROA showed the best agreement with 3DVCA (bias: 0.21; limits of agreement: −0.01 to 0.41; SD: 0.1). Correlation between various methods as compared to 3DVCA was better in the organic MR group than functional MR group. Conclusion: 3DHSPISA showed the best agreement with 3DVCA compared to other PISA methods. Better correlation between PISAEROA and 3DVCA was observed in patients with organic MR than functional MR.
Keywords: 3D echocardiography, mitral regurgitation, proximal isovelocity surface area, transesophageal echocardiography
How to cite this article: Lovhale PS, Gadhinglajkar S, Sreedhar R, Sukesan S, Pillai V. Intraoperative comparison of 2D versus 3D transesophageal echocardiography for quantitative assessment of mitral regurgitation. Ann Card Anaesth 2021;24:16371 
How to cite this URL: Lovhale PS, Gadhinglajkar S, Sreedhar R, Sukesan S, Pillai V. Intraoperative comparison of 2D versus 3D transesophageal echocardiography for quantitative assessment of mitral regurgitation. Ann Card Anaesth [serial online] 2021 [cited 2021 Aug 4];24:16371. Available from: https://www.annals.in/text.asp?2021/24/2/163/314139 
Introduction   
The flow convergence or proximal isovelocity surface area (PISA) method has been recommended for grading the severity of mitral regurgitation (MR).^{[1],[2]} Vena contracta width (VCW), which is the narrowest portion of the regurgitant color Doppler jet is considered as one of the quantitative parameters for the assessment of severity of MR.^{[3],[4]} However, the assumption of circular geometry of the regurgitant orifice is required for the 2DVCW measurement. The EROA can be represented by the 3D echocardiographic vena contracta crosssectional area (3DVCA) without making any geometrical assumptions. The utility of VCA and MR regurgitant volume (Rvol) derived from 3DE in predicting the severity of MR has been validated against left ventricular angiography,^{[5]} 2D volumetric Doppler methods,^{[6]} and cardiac MRI.^{[7]} The American society of Echocardiography (ASE) recommends integration of qualitative, semiquantitative, and quantitative methods for the estimation of MR severity.^{[8]} Although no parameter has been accredited as the true gold standard for the quantification of MR, some authors have recommended 3DVCA as the reference parameter for the same purpose.^{[5]} It is presumed that the hemispherical (HS) contour of the proximal flow convergence gives the best estimate of the EROA. However, Yosefy et al.^{[9]} using realtime 3D echocardiography (RT3DE) showed that a significant number of patients with MR predominantly have a hemielliptical (HE) contour of proximal flow convergence rather than the HS contour and suggested that the HEPISA assumption is more appropriate for the quantification of MR. Therefore, we conducted this study to compare the EROA and Rvol obtained from 3DVCA with those obtained from 2DVC, 2DHSPISA, 3DHSPISA, and 3DHEPISA.
Materials and Methods   
This prospective, observational study was conducted in a tertiary referral center and a universitylevel hospital annually performing more than 1500 adult cardiac surgeries from January to October 2017. After obtaining the approval from the Institutional Ethics Committee and informed consent from patients, a total of 43 consecutive patients meeting the inclusion criteria were recruited as study subjects. We hypothesized that the EROA obtained from transesophageal (TEE) RT3D color Doppler methods would be more accurate than the 2DTEE color Doppler methods. The primary study objectives were to compare the EROA and Rvol obtained by 3DVC with those obtained by 2DVCW, 2DHSPISA, 3DHSPISA, and 3DHEPISA methods. The secondary objectives were to assess the shape of the regurgitant orifice in different pathologies associated with MR and to evaluate the intraobserver and interobserver variability in MR quantification. To achieve 80% study power, to detect a small difference with an effect size of 0.5 and alpha error of 0.025, the minimum sample size estimated was 38 patients. A total of 61 consecutive adult patients who underwent cardiac surgery and had at least mild MR and underlying sinus rhythm were recruited for the study. Exclusion criteria were those with eccentric MR, multiple jets, previous mitral valve surgery, cleft mitral valve, infective endocarditis, significant mitral stenosis (area <1.5 cm^{2}), more than mild aortic valve stenosis or regurgitation, arrhythmias, and myocardial infarction 6 weeks prior to the surgery. Emergency or redo surgeries, esophagogastric pathologies, poor quality 2DE, or 3DE images and surgeries wherein TEE probe placement was contraindicated were also treated as exclusion criteria. A total of 43 patients were finally included as study subjects after the exclusion of 18 patients due to various reasons.
After anesthesia induction and establishment of standard monitoring, the trachea was intubated and artificial ventilation was instituted. An adultsize RT3DTEE probe was inserted and heart was examined using an ultrasound system (iE 33, Philips Ultrasound, Bothell, USA). All echocardiographic examinations were performed before establishment of cardiopulmonary bypass (CPB). The hemodynamic parameters were maintained close to the preoperative levels at the time of acquisition of images. The images necessary for assessment of MR were acquired from midesophegeal views and 3DE analysis was performed later using offline Qlab 3DQ software.
All echocardiographic examinations and data acquisition were performed by echocardiographers trained in RT3DTEE. The 2D and 3D vena contracta were acquired from a zoom mode in the midesophageal long axis (MELAX) view with the central beam passing through the leaflet tips at a Nyquist limit of 50 to 60 cm/sec. The color flow sector was made as small as possible to maximize lateral and temporal resolution. The VCW was defined as the narrowest width of the proximal jet measured at or in the immediate vicinity of the MR orifice at the leaflet tips. The severity of MR using VCW was graded as mild (<0.3 cm), moderate (0.3 to 0.69 cm), or severe (>0.7 cm). The EROA was calculated by the formula 2πr^{2} and the mitral regurgitant volume (MRRvol) by multiplying EROA with MRVTI. Using MRRvol, the severity of MR was graded as mild (<30 ml), mildtomoderate (30 to 44 ml), moderatetosevere (45 to 59 ml), and severe (≥60 ml). The proximal flow convergence was acquired in the zoom mode at MELAX view with the baseline shift of the Nyquist limit (30–40 cm/s) in the direction of MR. The sector width and depth were initially reduced to increase the resolution. Also the color Doppler box size was reduced that would include all the three components of the regurgitant jet [Figure 1]a.  Figure 1: Four methods of measurement of EROA are shown in the figure. (a) Shows 2DHSPISA radius measured in midesophageal long axis view using zoom mode (b) Method to measure 3DVCA from 3D color full volume dataset. The blue window displays enface view of vena contracta which is directly traced to measure its area. (c) Methods of 3DHSPISA and 3DHEPISA measurements: The diameters D2 and D3 of the PISA shell were measured in the blue window. Abbreviations:  PISA: Proximal isovelocity surface area. MR: Mitral regurgitation; 3DVCA: 3D vena contracta area; HS: Hemispherical; HE: Hemielliptical
Click here to view 
The EROA was calculated using the formula EROA = 2π × r^{2} × Va/Vmax, where r represents the maximal PISA radius; Va, the aliasing velocity of the proximal flow convergence (same as Nyquist limit) and Vmax, the maximal velocity of continuous wave Doppler MR signal. The maximum radius was measured from the tip of the leaflets to the point of first color aliasing. MRRvol was calculated as (EROA × MRVTI). A 3D color fullvolume of MR jet was obtained from the MELAX view over 7 beats using ECGgating by adjusting the Nyquist limit at 50 to 60 cm/s. Two orthogonal image planes parallel to the regurgitant jet direction were manually aligned across the regurgitant jet; a third cropping plane was placed perpendicular to the jet direction and then moved along the jet direction until the crosssectional area at the level of the vena contracta was visualized. The frame with the largest VCA in systole was magnified and VCA was measured by direct planimetry of the color Doppler flow signal [Figure 1]b. The shape of EROA was also observed and labelled as circular, elliptical, or crescentic. In the fullvolume 3D color dataset, the proximal flow convergence was acquired with the baseline shift of the Nyquist limit (30–40 cm/s) to optimize the visualization of flow convergence. The 3D volume data was presented in 4 quadrants using MPR mode, which included three 2DE orthogonal anatomic planes. The imaging planes were adjusted to get the best possible PISA shell. The 3DHSPISA was acquired from the displayed view by measuring the radius of the first aliasing velocity with the Nyquist limit adjusted offline to 30 to 40 cm/s and using the formula: PISA = 2πr^{2}. For HE3DPISA radius, r was measured as in the method of HS PISA. The blue cropping plane was now aligned perpendicular to regurgitant jet and advanced towards the mitral valve till it cut the PISA shell at its maximum diameter in a plane perpendicular to radius. The two diameters D1 (PISA width) in MELAX plane and D2 (PISA diameter) in the corresponding orthogonal mid commissural view were measured [Figure 1]c. The 3DHEPISA was then calculated from these three orthogonal parameters according to Knud Thomsen's formula.^{[10],[11]}
HE PISA = 2π ([rp (d1/2)p + rp (d2/2)p + (d1/2) p (d2/2) p]/3)1/p
Where p = 1.6075. The formula was incorporated in Microsoft excel sheet for calculations and values for r, D1 and D2 were entered manually.
During statistical analysis, categorical data was expressed as percentage (number of observations). The statistical mean and standard deviation were calculated for quantitative data and expressed as mean ± SD. Differences between the groups were analysed with the paired ttest. Pearson's correlation coefficient was used for all correlation evaluations. The r value of 0 to 0.35, 0.36 to 0.55, and more than 0.55 were considered as poor/weak correlation, good correlation, and significant correlation, respectively. A positive rvalue denoted direct correlation whereas a negative value signified inverse correlation. The agreement between the methods to calculate EROA and MR Rvol were tested using Bland–Altman analysis and plotted with lines representing mean ± 1.96SD. The interobserver variability and intraobserver variability was performed using intraclass correlation coefficient (ICC) and expressed as ICC value, 95% limits of agreement. The percentage of underestimation of EROA by different methods compared to 3DVCEROA was calculated by ratio of difference between the two methods divided by 3DVCEROA ×100. The P value of ≤0.05 was considered to be significant for all analyses. Receiver operating curve was used to analyze the value of echocardiographic parameters for grading severe MR. The value having optimal sensitivity and specificity was taken as the cutoff value. Statistical analysis was performed using the SPSS software version 22 and Graph pad prism 5.2.
Results   
Sixty one consecutive patients posted for elective cardiac surgery having at least mild MR were enrolled in the study. During the intraoperative period, patients with eccentric MR jets (n = 11), atrial fibrillation (n = 3), infective endocarditis (n = 2), and poor quality echocardiographic images (n = 2) were excluded from the study as per the protocol. About 43 patients satisfied the inclusion criteria. The mean age of the study population was 55.27 ± 15.08 years. More number of patients had functional MR (n = 34) compared to organic etiology (n = 9) as only those with central regurgitant jet were considered. The majority of the patients had elliptical geometry of regurgitant orifice, whereas a crescentic shape was more common in the functional group than the organic group. The demographic features, echocardiographic parameters, and shape of the 3DVCA in all patients are summarized in [Table 1].  Table 1: Demographic prof i les and preoperative echocardiographic parameters of patients. Majority of patients had functional MR
Click here to view 
The mean 2DVCW of the MR jet was 0.335 cm ± 0.0416 cm. A comparison of 2DVCW and 3DVCA showed a significant correlation (r = 0.69, P < 0.0001). In the subgroup analysis, the correlation value was better in organic etiology (r = 0.81) compared to functional etiology (r = 0.66). The EROA was calculated by three methods of PISA, namely 2DHSPISA, 3DHSPISA, and 3DHEPISA. The 3DHSPISA was larger than 2DHSPISA, but the difference was not significant. Both 2DHSPISA and 3DHSPISA were significantly larger than 3DHEPISA (mean ± SD: 1.88 ± 1.1 cm^{2} vs 1.32±0.79 cm^{2} and 1.97 ± 1.23 cm^{2} vs 1.32 ± 0.79 cm^{2}; both P < 0.001), respectively. All the three PISA methods showed significant positive correlation with each other. In the subgroup analysis, although 2DHSPISA correlated well with 3DHEPISA in both groups, the correlation in organic MR group (r = 0.97) was better than in functional MR group (r = 0.80).
The EROA estimated by 3DVCA (0.40 ± 0.13 cm^{2}) was larger than the EROA obtained by 2DVC (0.09 ± 0.02 cm^{2}), 2DHSPISA (0.18 ± 0.10 cm^{2}), 3DHSPISA (0.19 ± 0.12 cm^{2}), and 3DHEPISA (0.12 ± 0.07 cm^{2}). The paired difference between 3DVCEROA and EROA obtained by all other methods was statistically significant [Table 2]. The paired differences among the 3DVC and 2DVC, 2DHSPISA, 3DHSPISA, and 3DHEPISA were 0.31 ± 0.12, 0.22 ± 0.10, 0.21 ± 0.10, and 0.27 ± 0.09, respectively. There was a significant positive correlation among the paired comparison of different methods. The EROA obtained by 3DVC method was compared with all other methods and the limits of agreement were plotted. Considering the 3DVCA as the reference method, all other methods underestimated EROA; however, 3DHSPISAEROA showed the best agreement (bias: 0.21; limits of agreement: −0.01 to 0.41; SD: 0.1). In the subgroup analysis, the correlation between various methods as compared to 3DVCA was better in the organic MR group than functional MR. Maximum Rvol was seen with 3DVCRvol (mean ± SD 55.86 ± 20.80) and paired difference with all other methods was statistically significant (P < 0.001). All methods showed good correlation with 3DVCRvol [Table 2]. The grade of MR differed when estimated using EROA and Rvol by 3DVC, 2DHSPISA, and 3DHSPISA methods. The severity of MR was high when estimated with 3DVCEROA and 3DVCRvol parameters. The other methods underestimated the severity grading of MR in comparison with the 3DVCEROA and 3DVCRvol [Table 3]. The EROA and Rvol derived from 2DVC and 3DHEPISA showed maximum underestimation of severity grading compared to that from the 3DVC.  Table 2: Paired comparison of EROA and Rvol obtained by four methods with that of 3DVCA: The paired difference between 3DVCEROA and EROA obtained by other methods and 3DVCARvol and Rvol obtained by other methods is significant
Click here to view 
 Table 3: Regurgitant volume derived by five different methods: On comparing EROA obtained by four methods with 3DVC, the 2DVC and 3DHEPISA methods show maximum underestimation
Click here to view 
The Bland–Altman plot analysis was performed for the EROA and Rvol obtained by 3DVC and other methods. All methods underestimated EROA compared to 3DVC EROA and 3DHSPISA EROA which showed the best agreement among the methods used (bias: 0.21; limits of agreement: −0.01 to 0.41; SD: 0.1). Similarly, all methods underestimated Rvol compared to 3DVC Rvol. The 3DHSRvol showed the best agreement with the 3DVCRvol among the methods used (bias: 29.36; limits of agreement: 0.70 to 58.30) [Figure 2]. As the 2DVC and 2DHSPISA severely underestimated MR, the ROC analysis was performed to find the best cutoff value for severe MR by these methods, which were obtained after selecting the optimal values for sensitivity and specificity. At a sensitivity of 87% and specificity of 59.3%, a value of 2DVCW of >0.325 cm predicts severe MR. A 2DHSPISAEROA value of >0.149 cm^{2} is associated with a sensitivity of 75.0% and specificity of 74.1% that predicts severe MR. A 2DHSPISA Rvol value of >21.32 ml predicts severe MR when the sensitivity and specificity are 66.7% and 64.0%, respectively [Figure 3].  Figure 2: Bland–Altman plot analysis of EROA (a) and Rvol (b) obtained by 3DVC and other methods: (a) Difference in the paired EROA is plotted on the ordinate against mean EROA on abscissa. (b) Difference in the pairedRvol is plotted on the ordinate against meanRvol on abscissa. The observed mean difference is displayed as a continuous black line and the limits of agreement are displayed as dashed black lines. All methods underestimated EROA and Rvol compared to 3DVC EROA. 3DHS EROA showed the best agreement among the methods used
Click here to view 
 Figure 3: The cutoff values for severe MR using 2D echocardiographic parameters from ROC analysis. Considering 3D VCA as gold standard cut off values for 2DVCW, 2DPISAEROA and 2DPISARvol were significantly low
Click here to view 
The intraobserver and interobserver variability values for different methods used to calculate EROA are mentioned in [Table 4]. Excellent reliability was found for all methods between two analyses of a singleobserver and also between the observers.  Table 4: Table shows intraobserver and interobserver variability values for different methods used to calculate EROA. Excellent reliability was found for all methods between two analyses of a singleobserver and also between the observers
Click here to view 
Discussion   
Quantitative parameters are considered as important criteria to determine the severity of regurgitant lesions. Although, the published literature is inadequate to validate any parameter as a goldstandard for the echocardiographic estimation of EROA, the 3DVCA may be considered a reference method for the same purpose.^{[12]} Khanna et al.^{[5]} showed that 3DVCA correlates well with ventricular angiographic grading (coefficient r = 0.88) and is a feasible method for MR assessment. Excellent correlation between the 3DVCA and quantitative 2D Doppler parameters was demonstrated by others.^{[6],[13]} Grading the MR is reliable with 3DVCA, which distinguishes moderate from severe MR for all etiologies. It remains accurate to represent the EROA in the presence of central as well as eccentric jets and also in the presence of multiple jets,^{[14],[15]} and agrees with the parameters obtained by 3D ventricular volumes and thermodilution data. MR volume estimated using 3DTEE was found to have excellent correlation with CMR volume^{[7]} with a marginal underestimation by 1.2%. In a similar study, Marsan et al.^{[16]} reported the correlation between the CMRMR volume and 3DVCAderived MR volume with an insignificant difference of 0.08 ml/beat.
The regurgitant orifice is often inadequately visualized on 2DE and planimetric measurement of orifice area is rarely possible. The 2DVCW, which is regarded a simple and quick surrogate parameter for the EROA, has precision only for the circular regurgitant orifice. As the vena contracta is always a three dimensional structure having a variable length and width, the MR quantification using VCW is erroneous in the presence of elliptical or crescentic shape of regurgitant orifice. In our study, we found that VCW significantly underestimated the EROA. The circular shape of the regurgitant orifice was observed only in 4.65% of our patients. The elliptical shape of the regurgitant orifice was common in patients with organic etiology, whereas the crescentic shape was seen more commonly in the functional subgroup. In subgroup analysis, the correlation between 3DVCA and 2DVCW was found better in the MR of organic etiology compared to the functional etiology which is consistent with other studies.^{[17]} Since our study was confined to the central jets, we observed a strong correlation between the 2DVCW and 3DVCA in the organic group.
Grading the severity of MR using PISA method shows excellent correlation with angiographic methods.^{[18]} The concept of flow convergence is based on the assumption that the base of the hemisphere is planar and there is no constraining to the flow. This method may not be accurate for eccentric jets where the flow field is restricted.^{[19]} The EROA and Rvol estimated using PISA technique were shown to have a significant agreement with the data derived from the thermodilution technique only for central jets. Significant overestimation was seen in eccentric jets which were mostly due to flail leaflets.^{[20]} All patients in our study had central jets.
The hemispheric assumption holds true only if the ERO geometry remains circular. However, the ERO is rarely circular and may vary in shape from elliptical (most common) to crescentic or irregular. Therefore, the hemispheric PISA assumption especially in cases of functional MR, where the regurgitant orifice elongates along the mitral valve leaflet coaptation line, may lead to a discrepancy between estimated EROA and the actual area. Previous studies have shown that the EROA estimated by both HSPISA and HEPISA methods correlate well with 3DVCA; however, which method has a better agreement with the 3DVCA is still not established. The eccentricity of ellipse is an indicator of the deviation of the ellipse from a circle. Gorodisky et al.,^{[21]} using CMR found that in all cases of MR, including organic MR, PISA is eccentric in shape. It suggests that the 3D PISA shape resembles a hemiellipse rather than the hemisphere. The correlation between 3D PISAEROA and 3DVCA was better in the organic subgroup than functional in our study subjects. Using the Bland–Altman analysis, we found that both 3DPISA methods underestimated the 3DVCEROA. Since in our study, a large number of patients had the elliptical shape of regurgitant orifice, we expected that the 3DHEPISA would offer a more accurate value for EROA than the 3DHSPISA. However, the 3DHSEROA showed the best agreement among the methods used. Contrary to our observation, the hemispherical assumption of PISA was found to underestimate the regurgitant orifice area more than the hemieliptical assumption of PISA in some of the published studies. This underestimation was more evident in the functional subgroup.^{[10],[17]} Ashikhmina et al.^{[10]} reported larger values for HEPISA than HSPISA, although both were less than 3DVCA in their patients. Contrary to the observation of Ashikhmina et al., we found that the 3DHEPISA significantly underestimated the EROA compared to other PISA methods. A majority of our patients had functional MR due to coronary artery disease. The characteristics of regurgitant orifice in functional MR under anesthesia do not remain constant throughout the systole, but vary under the influence of multiple factors such as the fixed versus compliant nature of regurgitant orifice, left ventricular remodelling and dysynchrony, changing loading conditions, and severity of MR. Due to these irregular dynamic conditions, the shape of the regurgitant orifice may vary from circular to elliptical or crescentic at any given point of time, which directly influences the geometry of the PISA as well.
The PISA is a threedimensional structure which has three radii (r1, r2, and r3). The heightradius r1 is considered as a radius for both HS and HEPISA. The other two baseradii r2 and r3 are presumed to be equal to r1 in hemispherical geometry, whereas in hemielliptical geometry the r2 and r3 may be different from r1. For HEPISA to be larger than HSPISA, the sum of the two orthogonal baseradii (r2 and r3) must be more than twice the heightradius (r1) of PISA. We observed that similar to the 3DVCA, the 3DPISA shells also elongate along the commissural plane but remain constricted in the anteroposterior plane. In most of the cases, the anteroposterior radius (r2) was less than the radius (r1) of PISA. Therefore, for HEPISA value to become more than HSPISA, the radius along the commissural plane (r3) must be significantly large to compensate for the reduction in the r2 [Figure 4].  Figure 4: Base of the 3DHSPISA (a) and 3DHEPISA (b) are shown in the diagrammatic representation. The radius r1, which represents the height of both HSPISA and HEPISA, is same and not shown in the diagram. The r2 and r3 represent radius of the minor axis and major axis of base of the PISA, respectively. In HSPISA, r1, r2, and r3 are equal, whereas in HEPISA, r2 is less than r3. If the value of addition of r2 and r3 in HEPISA is less than the addition of r2 and r3 in HSPISA, the volume of HEPISA would be less than that of HSPISA
Click here to view 
In most of our patients, the regurgitant orifice was elliptical (79%) and the HEPISA was smaller than HSPISA because the summation of r2 and r3 was less than twice the r1. Therefore, the assumption that HEPISA would be more accurate than HSPISA for EROA estimations in asymmetric PISA shells may not be true for all the subsets of patients. Similar findings were reported in another study,^{[16]} wherein the authors noted that patients with functional MR had the smallest VCA. The mean EROA calculated by the HSPISA method had larger values than that of HE PISA. Schmidt et al.^{[22]} studied the feasibility and application of semiautomated PISA detection software. They found a better diagnostic performance in circular rather than elongated PISA shells not only for 2DPISA but also for 3DPISA. It suggests that the 3D PISA measurements work better for circular orifices. As with every technique, 3DE also has its limitations. It has poor temporal resolution, which can be overcome by multiple beat full volume acquisitions. But cardiac arrhythmias, respiratory motion or probe motion may lead to stitching artefacts. It also has limited spatial resolution. Currently, VCA measurement requires manual alignment of plane perpendicular to VC. Small mistakes in alignment can lead to under or overestimation of the EROA.
PISA is generally measured at only a single frame in systole, resulting in the overestimation of true flow, since the regurgitant flow varies over time, especially in cases of midsystolic and latesystolic MR.^{[21]} Automated 3D PISA measurements using dedicated softwares would probably overcome this problem. Recent studies^{[23]} comparing the 3D PISA using automated softwares have shown increased accuracy and less underestimation of EROA with these techniques compared to 2DPISA or geometrically assumed 3DPISA. However, the EROA derived by automated 3DPISA measurement was found to be less than the 3DVCA.
We do acknowledge the limitations to our study. EROA is a dynamic concept and changes throughout systole. Our reference standard of 3DVCA derived EROA from a single largest frame may not represent the EROA for the entire systole. Calculating the mean value by averaging EROA from all systolic frames would have better represented the EROA; however, this timeconsuming and cumbersome method may not be suitable for the intraoperative period. An automated EROA detection software taking the temporal variation into account would mitigate this problem, although, it is not yet validated. We excluded patients with eccentric jets from our study population. Since eccentric jets are common in patients with organic pathology, we could include a very small number of patients with this pathology in our study. With the current limitations to technology, no geometric assumptions would be ideal for the 3D PISA measurement. The development of automated software for PISA measurements in future may have an edge over the geometric PISA assumptions. The cardiac MRIbased 4DPISA may be able to assess MR severity quantitatively without any geometric assumptions.^{[21]}
In summary, the quantitative assessment of MR can be successfully performed using intraoperative TEE 2D color Doppler and 3D color full volume dataset. All the three methods of PISA and 2DVCW were found to have significant positive correlations with 3DVCA. The 3DHEPISA, however, significantly underestimated the EROA compared to other PISA methods. The correlation between PISAEROA and 3DVCA was better in the organic subgroup than functional. All methods to derive EROA underestimated EROA compared to 3DVCEROA. The 3D HS EROA showed the best agreement among the methods used. Similarly, the MR Rvol obtained by 2DHSPISA, 3DHSPISA, and 3DHEPISA techniques had significant positive correlations with the Rvol of 3DVCA. The 3DHEPISA significantly underestimated the Rvol. The elliptical shape of vena contracta was most commonly seen in the MR of both functional and organic etiologies. The crescentic shape was more commonly seen in the functional group than the organic group.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References   
1.  Zoghbi WA, EnriquezSarano M, Foster E, Grayburn PA, Kraft CD, Levine RA, et al. American Society of Echocardiography. Recommendations for evaluation of the severity of native valvular regurgitation with twodimensional and Doppler echocardiography. J Am Soc Echocardiogr 2003;16:777802. 
2.  Lancellotti P, Moura L, Pierard LA, Agricola E, Popescu BA, Tribouilloy C, et al. European Association of Echocardiography. European Association of Echocardiography recommendations for the assessment of valvular regurgitation. Part 2: Mitral and tricuspid regurgitation (native valve disease). Eur J Echocardiogr 2010;11:30732. 
3.  Hall SA, Brickner ME, Willett DL, Irani WN, Afridi I, Grayburn PA. Assessment of mitral regurgitation severity by Doppler color flow mapping of the vena contracta. Circulation 1997;95:63642. 
4.  Fehske W, Omran H, Manz M, Kohler J, Hagendorff A, Luderitz B. Colorcoded Doppler imaging of the vena contracta as a basis for quantification of pure mitral regurgitation. Am J Cardiol 1994;73:26874. 
5.  Khanna D, Vengala S, Miller AP, Nanda NC, Lloyd SG, Ahmed S, et al. Quantification of mitral regurgitation by live threedimensional transthoracic echocardiographic measurements of vena contracta area. Echocardiography 2004;21:73743. 
6.  Iwakura K, Ito H, Kawano S, Okamura A, Kurotobi T, Date M, et al. Comparison of orifice area by transthoracic threedimensional Doppler echocardiography versus proximal isovelocity surface area (PISA) method for assessment of mitral regurgitation. Am J Cardiol 2006;97:16307. 
7.  Shanks M, Siebelink HM, Delgado V, van de Veire NR, Ng AC, Sieders A, et al. Quantitative assessment of mitral regurgitation: Comparison between three dimensional transesophageal echocardiography and magnetic resonance imaging. Circ Cardiovasc Imaging 2010;3:694700. 
8.  Zoghbi WA, Adams D, Bonow RO, EnriquezSarano M, Foster E, Grayburn PA, et al. Recommendations for noninvasive evaluation of native valvular regurgitation: A report from the American Society of Echocardiography Developed in Collaboration with the Society for Cardiovascular Magnetic Resonance. J Am Soc Echocardiogr 2017;30:30371. 
9.  Yosefy C, Levine RA, Solis J, Vaturi M, Handschumacher MD, Hung J. Proximal flow convergence region as assessed by real time 3dimensional echocardiography: Challenging the hemispheric assumption. J Am Soc Echocardiogr 2007;20:38996. 
10.  Ashikhmina E, Shook D, Cobey F, Bollen B, Fox J, Liu X, et al. Threedimensional versus twodimensional echocardiographic assessment of functional mitral regurgitation proximal isovelocity surface area. Anesth Analg 2015;120:53442. 
11.  
12.  Lombardero M, Henquin R, Perea G, Corneli M, Izurieta C. True morphology of mitral regurgitant flow assessed by threedimensional transesophageal echocardiography. Echocardiography 2017;34:8793. 
13.  Yosefy C, Hung J, Chua S, Vaturi M, TonNu TT, Handschumacher MD, et al. Direct measurement of vena contracta area by realtime 3dimensional echocardiography for assessing severity of mitral regurgitation. Am J Cardiol 2009;104:97883. 
14.  Zeng X, Levine RA, Hua L, Morris EL, Kang Y, Flaherty M, et al. Diagnostic value of vena contracta area in the quantification of mitral regurgitation severity by color Doppler 3D echocardiography. Circ Cardiovasc Imaging 2011;4:50613. 
15.  Hyodo E, Iwata S, Tugcu A, Arai K, Shimada K, Muro T, et al. Direct measurement of multiple vena contracta areas for assessing the severity of mitral regurgitation using 3D TEE. JACC Cardiovasc Imaging 2012;5:66976. 
16.  Marsan NA, Westenberg JJ, Ypenburg C, Delgado V, van Bommel RJ, Roes SD, et al. Quantification of functional mitral regurgitation by realtime 3D echocardiography: Comparison with 3D velocity encoded cardiac magnetic resonance. JACC Cardiovasc Imaging 2009;2:124552. 
17.  Kahlert P, Plicht B, Schenk IM, Janosi RA, Erbel R, Buck T. Direct assessment of size and shape of noncircular vena contracta area in functional versus organic mitral regurgitation using realtime three dimensional echocardiography. J Am Soc Echocardiogr 2008;21:91221. 
18.  Bargiggia GS, Tronconi L, Sahn DJ, Recusani F, Raisaro A, De Servi S, et al. A new method for quantitation of mitral regurgitation based on color flow Doppler imaging of flow convergence proximal to regurgitant orifice. Circulation 1991;84:14819. 
19.  Simpson IA, Shiota T, Gharib M, Sahn DJ. Current status of flow convergence for clinical applications: Is it a leaning tower of “PISA”? J Am Coll Cardiol 1996;27:5049. 
20.  Pu M, Vandervoort PM, Griffin BP, Leung DY, Stewart WJ, Cosgrove DM 3 ^{rd}, et al. Quantification of mitral regurgitation by the proximal convergence method using transesophageal echocardiography: Clinical validation of a geometric correction for proximal flow constraint. Circulation 1995;92:216977. 
21.  Gorodisky L, Agmon Y, Porat M, Abadi S, Lessick J. Assessment of mitral regurgitation by 3dimensional proximal flow convergence using magnetic resonance imaging: Comparison with echoDoppler. Int J Cardiovasc Imaging 2018;34:793802. 
22.  Schmidt FP, Gniewosz T, Jabs A, Münzel T, Hink U, Lancellotti P, et al. Usefulness of 3DPISA as compared to guideline endorsed parameters for mitral regurgitation quantification. Int J Cardiovasc Imaging 2014;30:15018. 
23.  Hien MD, Weymann A, Rauch H, Grossgasteiger MC, Motsch J, Rosendal C. Comparison of intraoperative threedimensional Doppler color flow mapping to assess mitral regurgitation. Echocardiography 2012;29:84957. 
Correspondence Address: Shrinivas Gadhinglajkar Department of Anesthesia, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum  695 011, Kerala India
Source of Support: None, Conflict of Interest: None  Check 
DOI: 10.4103/aca.ACA_28_20
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4] 







