 For the next presentation, we'll be staying on the island. Our next speaker is Dr. Roberto Mosca, who's a staff anesthesiologist at the University Hospital of South Manchester. He received his training in Naples, Italy, and UK. He has a special interest in echo training and quality assurance. Dr. Mosca is a British Society of Echo Examiner, and is an active member of the ACTIAC subcommittee, and he regularly lectures and runs the Manchester Tohoby training program as well as the Quality Assurance Program in Manchester. Dr. Mosca will be talking to us about 3D LV assessment. Dr. Mosca? Hello everyone. I'm sharing my slides. OK, good afternoon from UK, at least. Thanks for the presentation, the introduction. Yeah, I'm working a busy Cadet Rassi unit in Manchester. And I thank the organizer to choose me for presenting this fascinating and controversial topic of 3D LV volume assessment. I have no financial disclosure to share. However, I'm going to show imaging from a particular vendor, which provides echo cards in my workplace. So the aims of these sessions are to describe the utility and the advantages of 3D transit of a geo-echo cardigraphy, and also consider limitations. I will go through an example to show how to use 3D echo to measure LV volume and function in the interoperative setting. For more than 20 years, the continuous advancement in computer increased technology resulted in a revolutionary transducer architecture, where a new sophisticated generation of over 3,000 crystals is arranged in rowing columns. Today, these innovative metrics are rated as user can acquire a pyramidal volumetric data set, displaying the three-dimensional echo cardigraphy imaging real time. These characteristics have made that real time 3D echo no anymore a research or a luxury tool. But an extraordinary diagnostic modality in clinical practice. One of its first and still current popular application is the LV volume assessment. Real time 3D echo generally has three main acquisition mode, light 3D, 3D zoom and full volume. So the light 3D is a button used to switch the system from 2D mode into real time 3D to watch and acquire real time 3D volumetric motion without any deconstructions. However, light 3D is limited to a narrow angle with a partial volume and may not provide much information. So it generally recommended for interventional guidance, rhythm disturbances, any situation in which the volume interest is within that limit angle. 3D zoom display a focus volume of interest, defined by a truncated slice of an arbitrary sector angle and the greater details, with or without ECG gating. So real time 3D zoom is indicated for the acquisition of valves and smaller structure as a trombost or masses. Full volume 3D instead combines a series of sub volumes acquired with ECG gating to create a final, larger, reconstructed full volume image. So the 3D full volume acquisition is the only image in modality that can capture the entire left and trigger volume at a sufficient frame rate, 20 yards or more, to allow dynamic assessment. So you can also use other modality like true view or glass view with color, but they are limited by narrow angle like 3D live and not usable for volume assessment. So the broad term or real time is typically applied to all current 3D echocardograph images to distinguish them from the earlier generation or complicated reconstructed 3D images, as we remember. However, we should consider the light 3D and 3D zoom modes as a true real time. It was in simultaneous ECG g-recording, whereas the full volume used ECG gating to synchronize image portion fused together over sequential cardiac cycles, and we could call them as a near real time, if we wish. So this is because the full image is unavailable until the final record the cycle is completed. So as clearly can be illustrated by the image on the right, showing stitching lines in the middle of a jail two chamber view, merging the four narrow wedges of the left ventricle volume over four consecutive art bits. The assessment of the left ventricle c-story function is a corner store on echocardograph examination, and the left ventricle ejection fraction is the most requested and used parameter. We also know that the LV volumes and ejection fraction are the key parameters to establish diagnosis and stratify prognosis. Moreover, important treatment decisions will be either surgery, device implantation, medical therapy, and evaluation of the therapeutic effects are based on these parameters. So as a result of these significant factors, we need to make any possible efforts to gain the most accurate and reproducible measurement. So for pure left ventricle function assessment, we can narrow down to three modality, full volume, biplane, and next plane. And I'm going to explain next how to achieve the measurements. For full volume, we need to start to get a decent, re-optimized middle of a jail zero degrees for chamber view. And after all the mechanical ventilation, electrocortory or manipulation, if you are in theater, we press the full volume button and acquire a four-beat cycle. We need to remember to check the frequency must be over 20 hertz to have any meaningful images. Once the acquisition is correctly done and of course restart the ventilation, we can start to do offline work. As you can see, the screen is now split in four images. The four chamber, its orthogonal view corresponding to the two chamber and a short axis view plus the 3D view with each individual bit, ECG trace at the bottom. To note, each box has the same color sector line to be positioned after confirming to be into the endostolic frame. After confirming the endostolic frame, the sector green and red lines need to be aligned in such a way to go through the left ventricle apex, the blue lines to cut through the middle left ventricle chamber and the yellow arrow to point the middle of the ventricle septum. Next, we will need to label the mitre annus reference points as a septal and lateral in the four chamber view and anterior inferior into the two chamber view. Finally, we will need to label the epical point. And after that, the soft analysis will show the endostolic volume. Then, we have to repeat the process for anti-stolic frame and label the reference points without moving the sector lines. So we have the anti-stolic volume. Although in both frames, it's very likely we will need to use manual board or editing along the endocardial line to ensure the entirety of the left ventricle chamber is included. Once the volumes are finalized, we can start the sequence analysis process, which takes a few seconds and presents the two view views and the LV cast playing along the cardiac cycle. And the bottom of the screen, it shows 17 different lines corresponding to each of the LV segments contribution to the change of LV volume along the cardiac cycle. So essentially, if lateral is the line, less contractile they are. We can also notice that each of the segmental line has a red arrow indicating the deepest point of the curve. So it could help to clearly identify not only the hypokinetic territory in the case of ischemia, but also related to acrony, dysynchrony, and indication for re-synchronization therapy. So you can also switch the cast view in color color segments to have visual confirmation of the anomaly itself. You can switch into other different pages for global volume summary. And a precise regional RR times so to help decision to treat our failure with CRT, for example. But again, this is not something related to the perioperative setting. However, I found some evidence useful in the perioperative setting as I show you in the next case. So in this case, after acquiring decent endocardial volume and I had to add endocardial border of the endocardial volume, which is, as I said, is not a rare happening, I'm afraid. And once we move into the analysis of the f-entrical segmental lines, it's easy to spot which territory has less discursion. We can easily identify it over in our course on the line or the left ventricle color square. And we can also highlight the excursion by each of the segmental layers, either basal, mid, or apical. And finally, we could select the parametric imaging page, which allows you to check the 17-segment discursion with the help of LV, Bell's Eye. So the report includes two color-coded polar maps, the top one for segmental timing, and the bottom one for segmental excursion. And it also showed the intraventricular disciclone indices. So the timing analysis of regional endocardial motion is the essential feature of the parametric display. The polar map is created showing the average timing when you get the motion in green. Early contracting segments, before the average, are displayed in shade of blue, while the lay contraction are in red color. For the visual interpretation of regional LV contracting or excursion, endocardial motion is depicted as shade of blue, representing normal inward motion segments, and red for outward moving this kinetic segment. The black show for achinetic segments. As I mentioned before, there are a few evidence like this paper showing that how the synchrony on more than 72 milliseconds lead to higher risk or complication for patients that are going conuretory by past graphs. This is something we can predict with our full volume assessment. A second modality to assess LV volume is the 3D biplane, which is the modality we can use a single bit acquisition, which reduce the frequency, but allow us to do a true real-time LV assessment. For this reason, this modality can find the scope in arts with the regular art rate or in situational gross and more dynamically instability. With the biplane, the four chamber, the two chamber images are acquired in the same cardiac cycle. This is important because it means that as opposed as 2D Simpsons method, we are assured that they displayed the same cycle volume and ejection fraction. And once acquired the image, we need to optimize the plane as per full volume before, with exception of having a yellow arrow to position. So for the advantage of this modality, is the ability to find the true apical point and avoid for shortening. Then after drawing a line across the metronomes, we can drag down the area to fill the LV cavity as four and two chamber, just like Simpsons, first and last and then in sisterly. And after the border editing, we have the volume and the ejection fraction reports. The third 3D LV assessment modality is explain. Explain function is able to show two orthogonal images simultaneously. Although in this we are unable to access the LV global volume, I think it can be extremely useful to analyze segmental wall motion or modality, especially in the operating room. In this example, we can appreciate the anterior wall like in Asia in transgastric short axis LV view. However, only with the help of explain image on the right, we can immediately appreciate the Akinesia limit only for the mid-territory of the anterior wall. So from Saska talk, we now know that 2D is the most widespread method using clinical practice yet, mainly due to its feasibility, wide distribution and rapid acquisition. But however, 2D echo methods are multiple limitation. They are operator dependent, they rely on visual interpretation and they are subjected to inter and intra-observer variability. In 2D echo to calculate a volume, geometrical modeling or chamber shape must be performed, and consequently, the ejection fraction estimation is subjected to error in presence of any pathology. So then the kind of visualization necessary to define chamber dimension is often difficult into the echo. And to counter this, the probe might be rotated, obtain better images. And this eventually produce an inherent problem, notably for shorting view and provide incorrect location on the apex. On the other hand, 2D echo has better spatial resolution, meaning the ability of a non-transound system to distinguish between two points at a particular depth in tissue. This is because we have in 3D reduced line density and therefore worse lateral resolution. 2D also better temporal resolution, which represent the extent of which an ultrasound system is able to distinguish changes between successive image frames over time. So in essence, the movements. Temporal resolution is determined by the image frame rate of the system, essentially nerds, which may vary depending on number of factors like propagation speed of the sound wave, the depth of the width of the field, the number of beam lines per field and the numbers of focal points. So however, meta analysis like this one where Doros and collaborator put 23 studies with more than 1600 echo algorithms, I found that compared with 2D echo, 3D was more accurate for left ventricular endostolic volume and systolic volume and ejection fraction, taking CMR as a golden standard. Like in this example from my clinical practice, 3D echo LB volume are found significantly larger compared to 2D echo. However, more clinical decision are based on ejection fraction. So with ejection fracture, there is no statistical difference in bias between 3D echo and 2D echo. So the major benefits of 3D echo really can be appreciated by significant 2D echo, intra and intraobserved variability, where two 3D demonstrate much lower variance on both. So if we compare instead the 3D echo against CMR, we found that different meta analysis did also identify the underestimation of 3D echo volumes compared to CMR, but they are much less than 2D echo. Hers disease like dilatation and hypertrophy leads to a greater distance between the ultrasound beam and the ventricle. So decreased image quality further. Irregular birders as a result of this pathologist impairing the accuracy of 3D echo of the border tracing and analysis and suggest a further contribute to the greater variation between 3D echo and CMR. So although every meta analysis showed that the accuracy of EF ejection fraction determined by 3D echo was excellent in normal art, the underestimation from a disease left ventricle fat and the stolic and systolic volume resulting in underestimation of ejection fraction in this patient population. For the comparable point is the ability to real time 3D echo to produce volume time curves. And these allow more detailed qualitative quantitative analysis of left ventricle performance like LB filling rates. As a reflection of the continuous LB volume change throughout the cardiac cycle. So these have been demonstrated correlate well with the CMR. CMR is not without its own limitations, of course, notably the cost, the increasing time, reduce the capacity and some devising compatibility your course of patient claustrophobia. So if we move into looking at transverse of agile trans thoracic, we need to remember that all the major supporting evidence of the accuracy and reproducibility of real time 3D echo or left ventricle volume are based on multi-center study and guidelines for the use of trans thoracic echo, TT echo. So we also know that although the visualization and many cardiac structures improve with transverse of agile echo, some difference in measurements have been found between transverse of agile echo and trans thoracic, particularly for chamber dimension and thickness. So these difference are primary due to the ability of tain from the transverse of agile approach to standardize imaging frames and views used when quantified chamber dimension with trans thoracic echo. Then we have so far evidence that real time 3D echo has really showed to be accurate tool for LB volume assessment. However, LB border detection in 3D echo remains at time consuming task. The job paradise application of modality routine practice really. So to overcome this, a 3D automated segmentation framework has been developed with the ability to capture the LB morphology in real time. So these automate algorithm allows for a fast and accurate quantification of the 3D cardiac volume and the global function with the minimal user input. So it may therefore contribute to the further integration of 3D echo in routine clinical practice. But unfortunately, these automatic algorithm is currently available only from trans thoracic modality, not in transverse of agile. And moreover, as I showed in semi-automatic border detection software as difficult tracking the endocardium. Even despite clear endocardial border definition, the naked eye. So this require manual adjustment of the border and could potentially introduce additional error in volume and ejection fracture measurements. So the transverse of a jail probe is relatively fixed in the sofagus with the apex in the far sector of field. Unlike the mobile probe in trans thoracic with its proximity of the apex and being able to include even the most dilated mitral annulus in the wider sector. So not always possible in transverse of a jail. Also calcification of mitral apparatus can cause image drop out of the ventricle of the endocardial borders with transverse of a gel echo. So the interoperative transverse of a gel echo involve additional challenges such as mechanical ventilation, data interference, direct cardiac manipulation and rapidly changing loading condition amongst the many others. And finally, the two modalities have very different implications. So the trans thoracic echidography 3D LV assessment is more indicated for risk stratification and planning new therapy or modify them. The trans thoracic 3D LV has a strong evidence in early identification of cardiotoxicity induced by chemotherapy in a trophysology therapeutic indication such as implant of devices or planning surgery. If we consider left ventricle and systolic volume a better pronostic value of left ventricle and systolic diameter. The trans thoracic transverse of a gel echo 3D volume instead has a smaller range of indication in perioperative setting showing superiority to 2D echo for identification of new segmental wall motion normality, left ventricle aneurysm or the synchrony after coronary artery bypass graphs. So when we talk about 3D LV volume meditation we need firstly reiterate the concept of poor 2D images makes a 3D image poor. We also clarify it's near real time 3D modality across the list for consecutive bits which results of being subjected to stitching artifacts and non-compatibility with the rimeus. It needs at least 20 years frequency to achieve meaningful images. It can be time consuming especially when there is a need of border editing. Certainly having the automated algorithm like the trans thoracic is a further limiting factor. And as just explained sometimes difficult to include the entirety of the late heart, the late left ventricle and the effects is not very visualized being as a far field. However, most important the 3D echo ejection fracture doesn't different from the 2D echo. Before conclude this controversial talk I'd like to leave you reflecting on this case and as it dies last week. As you can see from this 2D picture we can semi quantify quantitatively classify this LV function as poor but how poor is the ejection fraction? If you use conventional 2D measurements you find consistent results between LV and mode ejection fracture 26%. A fraction I assure to you 24% and Simpson biplane of 26%. But the 3D LV assure an ejection fracture as low as 18%. So we can also appreciate the basal inferior aneurysmal segment visible in the green square which protrudes the reference mesh of the bottom right of the video. So my personal take reflection at this was that I was glad to be able to use this additional modality for the patient's sake. So in the summary, I'll say the 3D LV assessment is the closest modality to CMR results is mainly validated with trans-toracic echo is strongly indicate a modality of choice in our patient's setting and provide similar ejection fraction but greater volume than 2D Simpsons. Procedure less inter and intra-observer variability than 2D and the interoperative 3D LV assessment carries more challenges. I got limited benefits but it's always a useful ability to have. Thank you. Thank you very much, Dr. Musker for that excellent presentation.