Transcription
I want to show interesting inogram and tips, but I also want to remind you of some engagement tips specifically regarding the RCA. I have described earlier this year the five caveats in RCA engagement, plus two additional ones. I suggest you review this stock. I want to highlight the third one here, which I frequently see fellows struggle with. With while torquing from the right cusp and pulling to engage the RCA, your catheter frequently jumps to the left cusp, which is the higher cusp. What to do next when this happens? Certainly don't try to engage the RCA by pulling more and clocking from the left cusp. So what should you do next? And this will be the first step of this talk, and this is a case to illustrate that.
Here the fellow was torquing from the right CP. This is the catheter in the lower cusp, the right cusp, and you can see as he's torquing here and pulling, clocking and pulling, the catheter jumped up to the left cusp. You can see the jump here; it's jumping from right to left cusp. So what do you do next? The first step is to get it back back to the right CP. You counter-clock and push it back down to the right CP, as we did here: counter-clock and push down to the right CP, making sure you are low into that right CP.
Now once you're in the right CP, if you just clock and pull again, as most fellows do, you may jump right back onto the left CP. In such a patient with such an anatomy where you have a propensity to jump up either to the left cusp or to the hole in between and fall in the ventricle, so you cannot keep doing this pull and clock again. The way you do it is you have to clock and barely pull; clock while staying very low, barely above the valve. You clock without much pulling, keeping your catheter low.
Now you may ask me, "Normally when we clock in order to transmit the torque you have to move; you have to pull the catheter." So if you clock and barely pull, you worry you may not transmit your clock. So to transmit the clock torque while staying low without much pulling, you may use your left hand at the sheath level, and you may push and pull the catheter at the sheath level with your left hand to create movement and transmit your clock while staying low. So you clock at that low valve level with a slight push and pull until the torque transmits and the catheter starts looking at you. Once the catheter starts to transmit the torque and starts looking at you, especially at this level, then you can keep clocking, and at this point you can pull more; it's not going to jump up to the left cusp at this point. And here is an illustration of that. So once it looks at you, then you can pull more and clock more; it's not going to jump up to this side anymore. Once it has the momentum toward looking interiorly toward us and to the right CP, you can pull more; it's not going to jump anymore. And that's what we did here, and we successfully engaged it. So again, once it's starting to look at us or toward the right CP, then you can be more free, pull more, and clock more; it's not going to jump. So the key is to start the process by staying low, slight push and pull at the sheath level, start transmitting the torque while you're low, preventing it from jumping; then once it starts to transmit halfway, you can pull more and clock more.
Another tip regarding the RCA: this is an inogram I see frequently, especially from outside institutions. So here they are trying to engage the right with a JR4, and this is the image they got. How can you improve this image? How can you selectively engage the RCA in this case? What's the next step? You need to recognize this is an LAO view. The ostium of that right is anterior. You see the ostium; it's about here; it's anterior takeoff RCA from the right CP; it's the takeoff is looking at us. Therefore, in those cases, JR4 does not properly engage the coronary. This is the takeoff of that RCA; it's anterior. JR4 cannot reach an anterior takeoff RCA most often because JR4 only elongates when it's aiming toward the right surface of the aorta. When it aims anteriorly, it doesn't elongate and therefore it doesn't reach. So in those cases when you see an anterior takeoff RCA, what you need to do: you need to use AL1, typically Amplatz Left One, the catheter with big butt and large catheter overall, to reach that anterior takeoff RCA. Occasionally you may use A2 in cases of small aorta, such as a small woman. And in those cases you use AL1, and you make the catheter point at you in an LAO view. You make it point at you rather than point toward the right surface of the aorta; you make it point at you as you're engaging, or you may use an RAO view and make it point toward the anterior surface of the aorta. RAO will lay out the ostium better than LAO in this case. So that's what you need to do in this case, and that's what was done: AL1 catheter, this is it here, and it's pointing to the anterior surface of the aorta in an RAO view. So RAO view, and you point in this direction in RAO view. So this this is here the fourth caveat that I describ—one of the most important caveats when you have difficulty engaging the RCA—it's an anterior RCA takeoff.
This is another interesting case of difficulty engaging the RCA. So here we're trying to engage the RCA; we are in the lowermost cusp, and we're giving non-selective puffs, and we don't see the RCA. So we are presumably in the right CP, but we don't see the RCA on our non-selective puffs. So what's the next step here? What's the diagnosis? The first thought whenever something like this happens is that again this is an anterior takeoff RCA, and the JR4 catheter which we're using here is far from the ostium, and we're not able to engage it. So that's the first thought, and maybe one of the most common scenarios. In this case, what's the other possibility? Remember the five caveats and the additional two. So another possibility that we don't think of often is that the lowermost CP in an LAO view—I frequently assume it's the right CP—but it could also be the non-coronary cusp. This is the LAO view; it spreads apart the left cusp, which is the higher cusp, and the lower cusp, which are right and even lower than the right, the non-coronary cusp. But it's possible that the catheter is in that non-coronary cusp rather than the right cusp, and that's why you're not able to engage. So those are the big two possibilities in this case: either you have anterior RCA takeoff or your catheter is in the non-coronary cusp, which is something we don't often think about. So RAO view will help here. One, it will help you engage and aim toward an anterior takeoff RCA; it will also help tell you whether you're in the non-coronary cusp or right cusp. RAO will spread apart the posteriormost cusp, which is non-coronary versus the right and left cusp, whereas LAO spreads apart right and non-coronary on the one hand and left coronary cusp on the other hand. So RAO view spreads them like this: non-coronary, which is the posteriormost and low, and right and left, which is anterior. And we did an RAO view, and actually we realize in an RAO view there is a ridge here, so we're really in the posteriormost cusp; we're in the non-coronary cusp. So all we did, we jumped up from that non-coronary CP to the more anterior CPs, and you can see it here; we pulled; we jumped up. Now once we jumped up, then we go back to the LAO view; we verify: are we in the right CP now or in the left CP? Because now you have to distinguish: after I jumped up from the non-coronary, did I jump to this or to that? So after doing the RAO, knowing that we were in non-coronary, jumping to the right CP or left CP, we do LAO; we verify we're indeed in the right cusp because on the non-selective puffs we see that red here; then we know we're in the right CP; then we just pull and clock, and we successfully engaged it. So this is a case where the catheter falls in the non-coronary CP. This is the sixth caveat that I describe, which is not a common caveat. The anterior takeoff RCA is a more common caveat when you're struggling to engage the RCA, but that's something to think about at times.
By cuspid valve, the most common scenario is right and left are fused. If you have a bicuspid valve, those are typically fused between the right and left cusp. You still aim to put your catheter in the lowermost area of that common cusp. The only thing is that you won't see jumps when you're moving between cusps in a bicuspid valve, but the levels will be the same; it's just you won't see jumps as the catheter moves between cusps.
I want to move on to some interesting angiographic cases. This is a 75-year-old man; he had CABG uh in February 2023, including LIMA to LAD. His exertional chest pain resolved, but he's still having persistent exertional dyspnea. This is his LIMA to LAD inogram, and this is from my former great fellow, Wasawat; he shared this case with me. So this is his LIMA to LAD inogram; I want you to look at it carefully. So this is the LIMA to LAD, and there is a big side branch coming off that LIMA here, and beside the side branch that you see, there is another potential problem: the anastomosis has some stenosis. The question here: can you have coronary steal through that LIMA side branch? Is this a problem—that big side branch coming off the LIMA? This is a common scenario. LIMA gives anterior intercostal branches and may give lateral thoracic branches, such as here, which supply the pectoral muscles. Those are usually ligated by the surgeon during surgery, but some branches may not be ligated, and large unligated thoracic branches have been documented in 10 to 20% of LIMA grafts. So it's a really common scenario. The question becomes: can you have a steal diversion phenomenon through those branches? And the answer is overwhelmingly no. Why is that so? Side branches, unlike what a lot of doctors think, are not a problem; they do not create a steal, even something as big as this branch, which looks bigger than the LAD. And the reason is hemodynamic. In most tissues, blood flow peaks during systole because of the higher pressure in the aorta during systole and higher pressure in the aortic branches during systole. So arterial flow to the chest wall and to the pectoral muscle is predominantly systolic, whereas the coronary flow to the LV myocardium is predominantly diastolic—more than 80% diastolic. Therefore, that doesn't divert blood away from the myocardium during diastole. Multiple studies using intravascular Doppler techniques have mainly refuted this—this syndrome of steal phenomenon. Many studies failed to show any hemodynamic significance of those IMA side branches under adenosine and under exercise hyperemic conditions, with no change in coronary flow reserve with or without those side branches. And this is one of those studies using Doppler wire; they demonstrated that the CFR into the LAD with adenosine and with left arm exercise did not change with or without occlusion of the LIMA side branch. This is another study of 15 patients with and 15 patients without LIMA side branches. CFR with adenosine of the LIMA was the same proximally and distally in patients with or without side branches. So side branches did not create any effect on coronary myocardial flow. This is another study using EDO Doppler of the LIMA grafted to the LAD in 10 patients with and 10 patients without side branches at rest and with vasodilation; there was no difference in flow between the two groups, and there was proper rise of diastolic flow velocity in both groups. Also, thallium scan did not show any ischemia in those patients with side branches. So so again, multiple studies show that those branches are not hemodynamically significant because their flow is mainly systolic.
Now, can the side branch be significant? It can, and there are case reports of that, although a lot of those case reports use angina as their proof of significance and angina resolution, but you need more than that to consider a side branch as hemodynamically significant. One, it is uncommon, and before considering it, so the patient must have typical exertional angina, classic exertional angina that is otherwise unexplained by any other disease he has, and you must prove impairment of LAD flow objectively by CFR or FFR. And even if FFR through the LIMA into the LAD is significant, you need to do pull-back and prove that the drop is across the side side branch. You may also wire the side branch and preferably you would wire the side branch, occlude it with a balloon, and re-measure the FFR while the balloon is occluding the side branch to prove whether the side branch is truly the culprit of ischemia. And in this case, what they did, they used an FFR pressure wire across the LAD, and this is what they obtained. They did RFR, which is peak instantaneous pressure ratio, distal versus the guide pressure; it's close to IFR, and they had a 0.03 drop across the anastomosis, 0.03 drop across the band, and 0.04 drop across the side branch. You can see it; the summation of all those drops was 0.89, which is just borderline significant, with non-significant drop at each one of those three locations. So what should you do here? Should you stent those three locations since the RFR is 0.89? My answer is you should absolutely not stent here. The drop at every one of those focuses is mild, and the risk of stent restenosis in those high-stress areas, especially anastomosis, is much higher than the risk of progression of the rather moderate stable disease at that anastomosis. As I always say, stenting moderate CAD, even if it is significant or mildly significant by IFR, stenting may aggravate the progression of moderate CAD, meaning you may have have faster progression with stenting than with leaving it alone, particularly with stent restenosis, neointimal hyperplasia, and neoatherosclerosis induced by stenting. So absolutely I would not stent in this particular case. And this case also confirms that a huge side branch was not hemodynamically significant—only a 0.04 drop in RFR at this level.
Another idea: can someone tell me why did we have some drop in pressure across this area, across this loop? So note that the wire straightened that loop. You see the loop here; the wire is straightened that loop, and very likely that 0.03 drop across the loop is actually a false drop; it's a pseudo-lesion. Whenever you put a wire across a loop, but particular LIMA, which frequently has loops and a lot of tortuosity, you create pseudo-lesions, and with your wire you will have some hemodynamic effect, and you may even create a hemodynamically significant lesion—much more than a 0.03 drop—just from straightening a loop and creating a lesion. That's one of the rare cases of false-positive IFR or FFR outside technical issues that can cause false-positive FFR. This is a case where without a technical measurement issue you may have a false-positive IFR and FFR; it's the pseudo-lesion where you straighten a loop or tortuosity with your wire. So be careful; even that 0.03 here is probably a zero, not 0.03. So you need to recognize that, and and therefore the summation is probably not 0.89; it may be 0.91, 0.92.
I'll move on to the next case. This is an inogram here; it's an RAO view of the left coronary, and what do you see here? Something noticeable. I will start giving the answer. So you notice on this angiogram that we see some left ventricular filling. One is a reflex maybe that we have what we call a coronary arteriovenous fistula from the left coronary to the ventricle. Now, which ventricle? In order to define left or right ventricle, you cannot tell on this view; you have to do an LAO view. RAO view, which is this, tells you you're going to the ventricle rather than the atria, which are here on this side. This is the AV groove, the circumflex, and atria will be here. So this is potentially fistular to a ventricle. You do an LAO view, and an LAO view will tell you it's a potentially fistula to the left ventricle. This is the LAD here running over the septum, and this is on this side of the LAD. So LAO tells you right versus left; RAO tells you ventricle versus atria. However, is this really a fistula to the left ventricle, or is it something else? It is something else, in my opinion, and it's a fairly common inogram; you will see it one in every few hundred angiograms. The reason I don't think it's a fistula is the following: you don't see directly the left ventricle. What you see first, you see myocardial wall stain blush, and you see myocardial LV wall boundaries delineated before you see the cavity fill. So you see myocardial blush first—wall blush before you see cavity filling—okay? And the cavity filling is relatively minor, which tells you it's not a fistula. A fistula will go straight into the cavity, not so much stain the wall and the boundaries. So what we have here, it is what we call dilated myocardial veins; we most commonly call those dilated Thebesian veins. So again, you see the myocardial wall stain and blush and LV myocardial boundary delineated before you see the cavity, which tells you that it's not a fistula; it's dilated myocardial wall veins, the so-called dilated Thebesian veins. And you see that even on the RCA injection, and common when you have those dilated Thebesian veins, you would see them on injection from the right coronary and from the left coronary. So you have dilated Thebesian veins from the PDA and posterolateral branches into the left ventricle. And this is a case from Circulation where they describe the exact same angiographic findings, and they call them as dilated Thebesian veins—bizarre appearance of capillary blush draining into the left ventricular cavity.
Now let me explain to you those Thebesian veins. There is a lot of debate about it, but here is a histological description. Normally, blood drains from the myocardial capillary into the major coronary venous system, which ends into the coronary sinus. But there is a smaller venous system, so rather than draining into that major coronary venous coronary sinus system, there is a more minor venous system, the Thebesian vein system—this number one here—where those veins drain directly into a cardiac cavity, whether right or left cavity, atrial or ventricular cavity. So those are veins that drain directly into any cavity. Those Thebesian veins may drain directly into the cavity or may go through sinusoidal lakes before eventually draining into the cavity. When the vein here—that Thebesian vein or the Thebesian sinusoid—are dilated, you may have the appearance that we had on our inogram. And when they are dilated, that dilatation is more common in the left ventricle; that's why we see those dilated Thebesian veins more commonly on the left ventricle, even though Thebesian veins are present in all cavity walls, but the dilated veins are more commonly—much more commonly—seen toward the left ventricle. Unlike true coronary fistulas, which are much more common toward the right heart, it's rare to have a coronary arteriovenous fistula to the LV. A coronary arteriovenous fistula means you have a communication between the artery and the cavity; you have a track that connects that artery directly to the cavity, bypassing that muscle. Whereas with the dilated Thebesian vein, the artery is still feeding the myocardium, which therefore does not get ischemic; it's not getting bypassed; blood is not diverted away from it; it's just the venous blood that is dilated and that drains into the cavity. That's why those Thebesian veins are different from a fistula; they are not hemodynamically significant; the myocardium is not getting ischemic; it's just dilatation of normal anatomical venous structure. Whether this or that, it causes a little bit of venous deoxygenated blood to LV shunting. And a proof that it's not hemodynamically significant are twofold: if this was a true hemodynamically significant shunt, you'll have two problems: one, the myocardium itself will get ischemic, and therefore because of the steal phenomenon from myocardium, the proximal artery will become dilated, anormal as happens in coronary arteriovenous fistula, whereas here the coronary is not aneurysmal. So that's one indirect suggestion that those are not hemodynamically significant. Another way of proving that it's not hemodynamically significant is: if it is hemodynamically significant, you will have a left-to-left shunt; it's almost like an AI, and the LV will dilate, and the LV in those patients is not dilated. So that's another proof that it's not hemodynamically significant: no big left-to-left shunt recirculation. And this is a summary of what I just described: when the venoluminal Thebesian veins or the sinusoid are dilated, you may have the appearance of our inogram, and this dilatation is more often seen into the LV. This is not a traditional coronary arteriovenous fistula, as the artery does not communicate directly with the cavity; the artery feeds the myocardium, which therefore does not get ischemic. Then it is the venous blood that drains into the cavity; that's why it is not hemodynamically significant. And this is an anatomy paper that proved the presence of those Thebesian veins normally. Now I, like most interventionalists, like to call this angiographic appearance dilated Thebesian veins. However, the great Dr. Angelini believes that those communications are indeed coronary-to-LV microfistulas, not Thebesian veins, but regardless, even Dr. Angelini considers them overwhelmingly insignificant hemodynamically, and he writes, "The clinical relevance of these malformations is trivial, and the flow through those microfistulas are often not substantial enough to cause ischemia or angina at rest, making coronary steal unlikely. Blood flow in these microfistulas appears to account for less than 2% of total coronary flow." So it is fine to call them dilated Thebesian sinusoidal veins, and that's what I like to call them, like most interventionalists. It's okay to call them coronary-to-LV microfistulas, like Dr. Angelini says, but regardless, the most important practical idea is that they are not hemodynamically significant. Don't be alarmed by those, and don't alarm the patient by those. And again, the proof is that the upstream coronary is not usually significantly dilated, contrary to what happened in severe hemodynamic fistula, and also the LV does not dilate, unlike what would happen if you have significant left-to-left shunting.
I'll move on to another case. This is an LV recording. Looking at this LV recording, tell me what is the LV EDP and what is the mean wedge pressure in this patient? So this is the LVEDP point after the LV a-wave, after the p-wave. This is the LVEDP; it's close to 30 mm of mercury, about 28 mm of mercury. Now what is the mean wedge pressure? A lot of doctors will be inclined to say that it's the same as the LVEDP, and that's the point of this case. The mean wedge pressure is likely a lot lower than 28 in this case; it rather correlates with this point here, the pre-a LV pressure in this patient. So whenever you have discrepancy between a high LVEDP but a normal pre-a LV pressure—and the pre-a pressure here is about 12, 13 mm of mercury—whenever you have this big discrepancy, your mean wedge pressure rather correlates with the pre-a LV pressure rather than with the post-a LV EDP. If all LVic pressure is high—pre-a and post-a is high—then your mean wedge pressure correlates with LVP and may even exceed it. But when you have this particular pattern, you need to recognize this pattern; it's a very common pattern: normal LV pre-a pressure but high overshoot of LV pressure during a-wave and high LVEDP. Your wedge pressure—mean wedge pressure—correlates with the pre-a LV pressure. This pattern that you see here means means that the wedge pressure or the LA pressure is normal throughout most of diastole and most of the cardiac cycle; only the end-diastolic LA pressure, like the end-diastolic LV pressure, shoots up. But the mean wedge pressure averages the wedge pressure or the LA pressure throughout the whole cardiac cycle, and the wedge pressure—a pressure—is normal throughout most of the cardiac cycle; it only overshoots at end-diastole, like the LVEDP. This is a sign of compensated LV dysfunction. LA pressure, like LV diastolic pressure, is mostly normal, but LV compliance only gets overwhelmed with the extra volume in end-diastole. So those patients are compensated; they are not having active congestion or active pulmonary edema; they are prone to decompensating quickly if you give them fluids or if you put them in atrial fibrillation or if you make them do heavy exertion, but they are currently compensated. So this tracing is a marker of compensated LV dysfunction. There is LV dysfunction; it's an important—I'm not going to dismiss the fact that LVEDP is high; there is LV dysfunction—but the wedge pressure is normal. This patient is not in pulmonary edema; this patient likely lies flat without hypoxia from a cardiac condition. So this is LVEDP, and this is LV Pa which...
Approximate means wedge pressure in this patient, and this is an illustration. This is compensated LV dysfunction; you get an overshoot only during a wave of both LA and LV pressure. The blue is the LA, the black is the LV pressure, so we get an over-overshoot during a wave, but the pressure is normal throughout most of the cardiac cycle, and the mean wedge correlates with the number here.
Conversely, when you have decompensated LV dysfunction, the whole LV pressure is high in diastole before and after a wave. The whole wedge pressure is high, and actually, the wedge pressure may exceed the LVEDP because in those cases, you may have a big V-wave in systole that causes the mean wedge pressure to be higher than even the LVEDP. I have explained this phenomenon in a totally separate talk that I strongly suggest you review, and I have explained it in a paper in JACC Heart Failure, and those are other papers, old from '73 or recent from 2022, that further corroborate, corroborate my explanation.
Now, why is this important? This is important for two reasons. One, it's a high LVEDP, but it doesn't mean you should diurese that patient, certainly not aggressively diurese, because this is a case where you may create acute kidney injury with diuresis. So, often no need for diuresis if mean wedge pressure is normal, even if LVEDP is high—not currently congested. The second idea is decision about support device during PCI.
And here is the patient from whom we obtained that tracing. This is an 84-year-old man who presented with heart failure and non-STEMI with a troponin I in the 5000s. His EF is newly reduced at 30% from that MI. He got 12 diuretics, and this is the hemodynamic tracing we obtained after diuresis. Clinically, he appears euvolemic; he's not hypoxic. Now, this is his angiogram, and he has complex disease. He has severe and complex calcification proximal to mid-LAD disease, which you see here. Beside the diagonal disease, you see that complex calcified proximal mid-LAD disease, and he also has severe left circumflex and OM3 disease, which you also see in a cranial view. He also has an occluded RCA, and you see collateral filling of that RCA. So look at that—this is a patient with an EF of 30%, three-vessel CAD with occluded RCA, and we need to do a rotablation of the proximal LAD as well as stenting of the circumflex. We need to do multivessel PCI, somewhat complex, and his LVEDP is close to 30. 30%—should he get LV support? A lot of doctors will tell you yes. My answer to this was no, based on hemodynamic tracing. I decided to not do support. Beside that tracing, clinically he also looked well compensated; systolic blood pressure was 100s; he was not hypoxic; he was not tachycardic. You could have done a full-blown right heart cath to prove what I just explained to you as well. I did not do any support. I did rotablation, and I fixed everything, and the patient tolerated the procedure well, and he did not get any hemodynamic compromise or hypotension during the procedure. And the reason I didn't do support is you need to review my talk—an important talk of when to consider LV support in complex PCI—but there are top two factors predicting hemodynamic compromise during PCI: one is hemodynamics, and I have decided here that his hemodynamics were acceptable—that's not EF, that's hemodynamics—and two is the complexity of the procedure I'm going to do, but mainly it's the complexity of a left main multi-step, two-stent distal left main strategy. This patient did not have complex distal left main bifurcation with planned complex two-stent strategy, so he didn't have any of those two. For that reason, I felt comfortable doing it without support, and this is the result here. We, we got very good results across the LAD and very good results across the circumflex and OM3.
I will move on to a next case. This is shared with me by my friend Nestor Berardo. So this is a 35-year-old female, no prior cardiac history. She's a smoker. She presents with chest pain and a non-STEMI troponin elevation, and this is her angiogram. This is a shallow AR cranial. What's the diagnosis? So when you see this angiogram, this is the LAD here running to the apex, and you see a diagonal. So the LAD has a long, diffuse, smooth stenosis that's extending almost all the way distally. So when you see this, the top diagnosis is coronary epicardial vasospasm. So the first step is to give intracoronary nitroglycerin, at least 200, 200 to 400 micrograms, and consider even giving verapamil if it persists with nitroglycerin. So let's say we gave here in this case nitroglycerin, we gave even verapamil, and the same angiographic appearance persisted. What's the diagnosis then? The second diagnosis you need to consider in this middle-aged woman is SCAD, spontaneous coronary artery dissection, specifically SCAD type 2, okay, which gives you this angiographic appearance exactly, which is intramural hematoma. It's a bleeding in the media that causes compression of the lumen, and it tends to have the appearance of that long, smooth lesion over 30 mm, refractory to nitroglycerin, usually heavily tortuous, although not in this particular case, and with no obvious calcium; middle-aged woman. So those patients tend to have a lot of vessel overlap and a lot of tortuosity. This patient did not have that, but it had that long, smooth appearance refractory to nitroglycerin without significant calcium, so you should think SCAD in this patient. Another possibility is long, diffuse atherosclerosis, plus or minus plaque rupture, but this is the least likely diagnosis here. This is most likely SCAD type 2. It's type 2A because it doesn't extend all the way to the tip of the vessel. When the dissection extends all the way to the tip of the vessel, we call it type 2B. This is probably SCAD type 2A, and SCAD tends to involve distal vessels in 92% of the cases; it involves more proximal vessels in 8% of the cases. So this is most likely SCAD, and you really need to recognize it because this patient has TIMI 2 to 3 flow, and she's not having an ongoing ST elevation. In fact, therefore, you should not touch it; you should not try to do PCI because there is a PCI risk of failure and complication of over 50% in those patients, and that's what my friend did here, so did not intervene in this patient. One thing to do in those cases to prove SCAD beside the simple angiographic appearance is to try to do peripheral angiography, like aorto-iliac angiography, and see if you have FMD, fibromuscular dysplasia, which is present in up to 70% of those patients that can indirectly corroborate your diagnosis, and do also CT-coagulated cerebral and abdominal iliac femoral CT to look for FMD, fibromuscular dysplasia, and that's what my friend did, and indeed this patient had coagulated fibromuscular dysplasia, which in retrospect further corroborated the diagnosis. You can also do CTA 6 weeks later and prove that the patient has no longer significant obstruction of the LAD. Even if you cannot tell for sure it's a SCAD, you can tell whether there is obstruction or not at 6 weeks, and in most SCAD, over 80%, up to 97%, they will recover spontaneously. Okay.
This is another case we had yesterday. This is a 37-year-old man this time, no past medical history, had V-fib arrest after vomiting. He's now in cardiogenic shock, and this is his angiogram. You have an occluded ramus that you see here, but you also has a small, diffuse left main. Whenever you see a small left main like this—it's smaller than the branch vessels such as the LAD and circumflex—you should think significant left main disease, and this is what was suspected here. This is severe left main disease. Could it be left main atheroma or left main SCAD? In this particular patient, he's in shock, and he has left main SCAD, and he has a total occlusion of a ramus with V-fib. This patient definitely needs an intervention, even if it is a SCAD, one because you have an occlusion with an ongoing infarct—TIMI zero flow for the ramus. Number two, it's left main SCAD; you need to fix left main SCAD, so he needed intervention. So the diagnosis in this patient is less important than the prior patient because here with a SCAD we decide to leave it alone; here with or without SCAD we're going to intervene. However, it's nice to know what's the pathophysiological process in this patient. IVUS showed this before intervention. You need to recognize the IVUS imaging of SCAD. This is a classic image. This is a true lumen of the left main, and this is the intramural hematoma. It has always that shape of crescent or banana shape that is somewhat homogeneous, somewhat dark with white speckles. It gets darker usually as you go deeper, so it's a crescent or banana shape, somewhat homogeneous with white speckles. This was the distal left main, and it gets worse in the proximal left main. You have that crescent, banana shape with white speckles, dark with white speckles, but you have a bigger banana here that you see again, darker and somewhat homogeneous with white speckles compressing the lumen. This is a classic SCAD, and I showed in the past from the literature—this is a SCAD from the literature again—it could be circumferential, or it could be crescent, banana shape. You see the SCAD here as well from the literature. Here there are multiple pockets—this is one pocket, banana; this is another banana here; this is another banana again, darker with white speckles—and in this case, the left main was fixed as well as the ramus occlusion.
This is another case. This is a 52-year-old man who presents with anterolateral STEMI, and this is an AO-CA view. You can already see there is some proximal LAD disease at the level of the first septal, but it's not occlusive. There is a subtotally occluded diagonal that may be the culprit for that STEMI, but start thinking of the rest. Maybe this is a circumflex artery here, and whenever you cannot see well the distal circumflex in an AO-CA view, whenever you see a lot of overlap of branches, maybe you think this is a dominant left circumflex. Okay. This is a shallow AO-cranial view here. Now what do you see on this view? One, we see the LAD and diagonal disease that I showed. We have some LAD proximal disease, and it's subtotally occluded diagonal, which is the culprit for that STEMI, but what else do you see? How about that circumflex? Can anybody point to the circumflex here? So in this shallow AO-cranial, in cranial views, the circumflex tends to have a loop up, then go down. Even a dominant circumflex, you would see it very well on cranial view; you will see it very well, but it will start by looping up, then going down. So this is the circumflex, that vessel here that's going up, and it gives an obtuse marginal branch; it doesn't go down. You have another vessel here that's—it's not looping up; it's going down all the way, and it's giving a PDA that almost meets the LAD, but it's not a circumflex. And whenever you're thinking, well, maybe it's an anomalous RCA coming from the left, and whenever you think anomalous coronaries, always use AO and LAO views. LAO view is the view that will confirm to you that you have an anomalous coronary; AO view is the view that will tell you the course of the anomalous coronary. So LAO will tell you if there is an anomalous coronary, and we did LAO here, and you see all that LAO. Okay. This is the LAD here running over the septum; this is the LAD, and this is an artery going on the right side of the LAD. Again, LAO splits apart left versus right. This is an artery going into the right AV groove as opposed to the circumflex going to the left AV groove. So this is indeed an anomalous RCA coming from the left. You have to define: is that anomalous RCA coming posteriorly, looping like this and going to the right AV groove, or is it coming between the arteries, or is it coming anterior to both aorta and PA? Here I show the illustration. This is here anomalous left coronary from the right, but it's the same concept: is it going posteriorly, in between the vessels, or anterior to both? And AO view will help decide the, the course. LAO tells you it's anomalous; AO tells you what's the course: interarterial, posterior, or anterior to both, and this is in relation to the aorta. On an AO view, you see how that anomalous artery relates to the non-selective contrast spilling into the aorta, and here on the AO view, this is the anomalous right coronary, and you see you seek that contrast spilling into the aorta that you see here, and you relate that anomalous coronary to the aorta, and you can see that that RCA is anterior to that contrast spilling. Now it's immediately anterior to it, so it's not posterior; it's immediately anterior to it; therefore, it's an interarterial course. If it was anterior to the I and the PA, you will see it have a big loop anteriorly, big upward loop anteriorly, then it would go to the right AV groove, and this is how it would look if it was anterior to both. It will give a big anterior loop, looping around the PA, then it will go back to the AV groove. It could also have a big loop downward if it is—is what we call septal, subpulmonic, below the PA course, but this one has a sharp bend immediately in front of the aorta, so this is an interarterial course. So AO view here tells us very simple: AO view told us this is an interarterial RCA. Keep in mind that anomalous RCA are overwhelmingly interarterial anyway, more than 90% of the time, so it's no surprise that it is interarterial. You can review my talk on congenital coronary anomalies to further understand those ideas.
Now another important concept here: after PCI of the diagonal culprit and the LAD in this patient, should he eventually receive surgical repair of that anomalous RCA? So when do we do surgical repair? My answer in this particular case is no, and I will explain it. We do surgical repair one, if you have true, typical angina that is otherwise unexplained or truly abnormal exercise stress test, not vasodilator stress test, or VT otherwise unexplained, then yes. Now if you have an asymptomatic patient with no angina and no classic angina on an exercise stress test, but they have an interarterial course or other high-risk features, then if it is a left main anomalous coronary interarterial or other high-risk feature, yes, Class 2A. RCA, probably no; it's a Class 2B to fix it, especially if it is less than 3.5, but most often no for RCA if it doesn't have a classic angina clinically or by an exercise stress test. And by the way, when I, what I call high-risk feature, those are associated with an interarterial course, is basically slit-like orifice that further stretches out and narrows as the aorta distends during exercise, or an acute, sharp ostium that is kinked or an intramural course partially in that wall before it emerges out, and that can create narrowing of that ostium, and those are commonly associated with an interarterial course, and they are better defined by CTA. But anyway, even if you have those, if it is RCA, I would probably not refer for surgery if it is asymptomatic with no angina. Some groups and some papers have considered in those cases with high anatomical features but no symptoms or angina, they have considered doing resting IVUS and resting PDA ratio with a pressure wire to look for compressed ostium and impaired flow at rest, and they have considered doing dobutamine IVUS and dobutamine PDA ratio to see if there is ostial compression or flow impairment with high flow state, and at the very least, if you do want to refer that patient to surgery, you need to do those things: either IVUS or PDA at rest and with dobutamine to prove some hemodynamic significance before referring him for surgery. But even that, there is no large data supporting that strategy, but it's a reasonable strategy.
I will show another case. This is a 69-year-old man with type 2 diabetes, and he has exertional angina. This is his angiogram. Try to define what's going on in this patient. This is shallow AO-cranial of the left coronary. You see the left main, LAD, circumflex territory, and you see collateral to the RCA, but there is something unusual about that RCA. Can somebody tell me? So one can thing, there is a CTO of that RCA, but if you look further, this is another view, and you can see that there are collaterals to the RCA, but there is something unusual—that RCA is filling all the way backward, and it's filling all the way to the ostium, and it seems that this ostium is communicating with another artery. You see the ostium here; it's communicating, not with the aorta—this is the aorta—that ostium is communicating with a structure higher than the aorta and more anterior than the aorta. So this is a patient who has an anomalous right coronary artery that is arising from the PA, and it's getting collateral filling from the left coronary system. Another thing to notice on this patient: he has an occluded left circumflex, and that may be what caused him to have angina because that anomalous coronary, he's had it all his life, but the occluded LCx might be what triggered his angina. So let me explain this a little bit. This is anomalous coronary artery originating from the pulmonary artery. So this is not a fistula. This is a picture of anomalous left coronary originating from the PA. This is different from fistula. Fistula means the coronary artery arises normally from the aorta, and it gives a channel to a chamber, such, such as, for example, the PA, more commonly RV and RA, but it can give it to the PA, but it's arising normally, and it's giving a channel to the PA. It could terminate in the PA, or it can give a channel to the PA, but anomalous origin from the PA is much, much more serious. It means the coronary itself arises—the ostium arises from the PA—it's being fed from the PA, and that's a huge problem because unlike fistula, most often it is benign, and you get some coronary steal and some coronary shunting to the chamber, whereas with anomalous coronary origin, the main issue is the low pressure in the PA. You have a coronary that arises from the PA. Well, the blood of the PA is deoxygenated, but more importantly than the deoxygenation is the PA pressure is low, so you cannot feed the myocardium from that coronary. This is more dreadful than the low oxygenation of that anomalous coronary; the low PA pressure explains why the flow in that apparent coronary artery is directed retrogradely toward the pulmonary artery and is derived via anastomotic vessels from the controlateral normally arising coronary artery. So even the collateral flow is better, and the collateral pressure is better than the pressure arising from the PA. So those patients can get severe angina. Anomalous left coronary from the PA is a very serious condition; is one of the most serious coronary anomalies and often leads to death in infancy if untreated. This is what we call anomalous left coronary from the PA. The only patients like this one—he's 69 years old—the only patients you will see in adulthood with anomalous coronary from the PA are those with an RCA arising from the PA or rarely an anomalous left coronary from the PA with a relatively very small left coronary system and massive right-to-left collaterals. So again, look at this picture. This is a patient with anomalous right coronary from the PA, and it's being fed from collaterals from the left coronary system, and you see it filling all the way to that PA ostium. The PA being more anterior and higher than the aorta—this is the aorta, and this is the PA here—the reason he had angina again—the circumflex occlusion. It's also possible that that RCA was getting other collaterals from the left circumflex AV groove continuation. So he's getting angina because he's losing some of those collaterals to the RCA. PA angiogram may not show the anomalous coronary because you have reverse flow in that coronary. PA pressure may be slightly increased from some degree of left-to-right shunting, and in infantile type anomalous uh coronary artery from VPA, the treatment is direct reimplantation of the anomalous coronary into the aorta. This can be done in adults like this patient, or in adults you can ligate the anomalous coronary from the PA and do a bypass.
This is another quick and final case. 47-year-old female with no prior medical history, she presents with chest pain. She is found to have anterolateral ST elevation. You can see here lateral ST elevation and anterolateral ST elevation with a lot of PVCs. This is definitely STEMI. This is not pericarditis. This is the dome shape of ST-T that indicates a STEMI, and this is what they did. This is again from my former fellow, Wasawat. This is what they found: subtotally occluded left circumflex with TIMI 2 flow. What's the diagnosis here? Again, I want you to think: give nitroglycerin. This could be coronary vasospasm, so definitely give nitroglycerin. If this persists in this middle-aged woman without heavy calcium and with long disease, think SCAD again, think SCAD. Keep that diagnosis in your mind, and that's what they thought. They gave nitroglycerin; it didn't resolve. They thought SCAD. However, she was having ongoing STEMI, and what they thought is ongoing ischemic pain, so they did decide to intervene, but what, when you intervene, be minimalist in your intervention. Another thing to notice is that this patient also had a lesion of the third diagonal, so she had a SCAD of the third diagonal as well; it wasn't just the circumflex. Multivessel SCAD can be seen in 15% of patients. So what they did here: they wired and planned on doing low-pressure, small, undersized balloon. They did that, but even with small, undersized balloon, they got propagation of the SCAD more distally, so they had to keep ballooning more distally, and eventually they used the cutting balloon, which may be considered to decompress the intramural hematoma in those patients—small, undersized cutting balloon—and eventually they got good result just with balloon angioplasty, and they stopped there. That's the uh proper management of this patient.