Retrograde Recanalization of a Left Anterior Descending
Chronic Total Occlusion via an Ipsilateral Intraseptal Collateral

Satoru Otsuji, MD, Kunihiko Terasoma, MD, Shin Takiuchi, MD
Satoru Otsuji, MD, Kunihiko Terasoma, MD, Shin Takiuchi, MD

Successful recanalization and percutaneous revascularization of coronary arteries with chronic total occlusions (CTO) is one of the last frontiers in percutaneous coronary interventions (PCI). Successful CTO treatment is associated with reduced angina, improved left ventricular function and, ultimately, improved long-term survival.1 Despite increasing knowledge of CTO recanalization strategies, advances in equipment and operator expertise and published success rates with the standard antegrade techniques appear insufficient.2 A retrograde approach using various collateral vessels is likely to improve these suboptimal results.3 However, there are limitations in choosing the route for the retrograde approach such as the extreme tortuosity of the epicardial collateral, a narrow septal collateral that cannot be visualized throughout its entire course despite multiple selective angiograms, or a collateral arising from around the stenosed donor artery being compromised with stenting. We describe 2 cases of the successful use of a retrograde approach via an ipsilateral intraseptal (proximal septal-to-distal septal) collateral vessel in the treatment of a mid-left anterior descending (LAD) CTO.

Case 1. A 70-year-old diabetic male with a 1-year history of angina and positive exercise test underwent diagnostic coronary angiography which revealed 2-vessel coronary disease with a focal lesion in the proximal right coronary artery (RCA) and diffuse atherosclerosis throughout the LAD, with a 90% stenosis in the proximal portion and a CTO in the mid portion. The distal LAD beyond the CTO was supplied via 2 collateral vessels. The predominant one was an epicardial collateral arising from the stenosed segment of the RCA, and the other, though to a lesser extent, was an ipsilateral intraseptal collateral that was best visualized via a right anterior cranial view. This view displayed a hammock-like connection from the proximal septal branch to the distal one (Figures 1A and B). The patient was referred for PCI, which began with PCI of the LAD CTO first. However, the standard antegrade approach was not performed because multiple angiographic views were unable to delineate the correct entrance to the CTO. We therefore attempted a retrograde approach. The predominant epicardial collateral from the RCA was considered unsuitable for the access route, mainly because itarose from just around stenosed segment and because it was severely tortuous throughout its course. The ipsilateral intraseptal collateral was comparatively flexible, which was likely to allow a retrograde approach. We therefore proceeded with the procedure via this intraseptal connection. Bifemoral access was established, and a 7 Fr Launcher SL4 guiding catheter (Medtronic, Inc., Minneapolis, Minnesota) was used for the left main engagement and a 7 Fr Launcher JR4 guiding catheter was used to engage the RCA, allowing for better visualization of the distal LAD and to prepare for possible PCI in the event that the RCA stenosis worsened during the procedure. A Whisper LS guidewire (Abbott Vascular, Santa Clara, California) with a 150 cm long Excelsior microcatheter (Boston Scientific Corp., Natick, Massachusetts) was introduced in an antegrade fashion down the LAD into the targeted proximal septal branch. At that point in the procedure, the ipsilateral intra-septal collateral was not well visualized angiographically. The guidewire was, therefore, navigated referring the diagnostic angiogram and advanced in a retrograde fashion to the distal LAD via the intraseptal collateral circulation. After advancing the microcatheter into the distal LAD, the wire was exchanged for one with a stiffer tip. The correct position of the guidewire and microcatheter was confirmed via contrast media injection through the guiding catheter in the RCA. Subsequently, the wire was exchanged for a Fielder FC wire (Asahi Intecc, Japan) and then a PT2 LS wire (Boston Scientific), which crossed the lesion from a retrograde approach toward the proximal septal branch, and finally looped back into the ascending aorta (Figure 1C). Subsequently, the proximal LAD lesion was predilated toward the septal branch with a 2.0 mm x 20 mm Sprinter rapid exchange balloon (Medtronic) via the same wire. This was performed in order to relieve ischemia due to worsening of the stenosis at this lesion site during these wire manipulations and to facilitate the advancement of the balloon used to cross the CTO from a retrograde approach. Finally, a 1.3 mm x 10 mm, 155 cm long Lacrosse monorail balloon (Goodman, Japan) was advanced to the proximal tip of the CTO, and multiple dilatations were performed throughout the lesion via a retrograde approach (Figure 1D). Recovery of antegrade flow was barely achieved. The lesion was then crossed with a Fielder FC wire from an antegrade approach followed by removal of the retrograde wire. The CTO lesion was predilated using a 2.0 mm x 20 mm Sprinter balloon in an antegrade fashion. Finally, three (2.5 mm x 28 mm, 2.5 mm x 28 mm and 3.0 mm x 33 mm) Cypher™ stents (Cordis Corp., Miami Lakes, Florida) were deployed throughout the artery, yielding an excellent angiographic result (Figure 1E). Revascularization of the RCA was performed in a staged manner without any complications or ischemia during the procedure. The patient was discharged free of angina.

Case 2. A 55-year-old hypertensive male was referred for preoperative coronary angiography evaluation. He suffered a cerebral infarction 1 month prior to his referral. Cerebral angiograms revealed complete occlusion of the left-middle cerebral artery and a superficial temporal-to-middle cerebral arterial bypass was scheduled. His electrocardiogram was abnormal and an echocardiogram showed global reduction of his left ventricular function. Coronary angiography was performed and revealed left dominant coronary anatomy and a long CTO in the middle segment of the LAD. The distal segment of the LAD was supplied via an ipsilateral intraseptal collateral channel. The CTO was very long and the entrance to the occlusion had an unusual configuration. It was difficult to determine whether the artery had a bend or an ulcer at the site of the occlusion (Figure 2A). The anterior wall thickness was preserved according to the echocardiogram, thus we attempted PCI of the mid-LAD CTO using a standard antegrade approach. After a lengthy struggle, a Conquest Pro guidewire (Asahi Intecc) successfully penetrated the distal septal true lumen using a parallel-wire technique. To facilitate wire manipulation toward the distal LAD, balloon dilatation was performed toward the septal branch, however, a large false lumen was created and the distal true lumen could not be captured (Figure 2B). Intravascular ultrasound (IVUS) revealed a large false lumen created at the entry of the CTO, which gave a misleading idea of its entrance point. The procedure was eventually abandoned. A second attempt was made after healing the dissection. An ipsilateral intraseptal collateral channel remained and was comparatively straight for better accommodation of the retrograde wiring. This collateral channel was clearly visualized with contrast media injection through the microcatheter (Figure 2C), and a retrograde approach was thus made A 7 Fr short length (85 cm) Judkins left (JL 4) guide (Heartrail, Terumo, Japan) was engaged in the LCA. A Whisper LS guidewire (Abbott Vascular) with a Finecross MG microcatheter (Terumo, Japan) was introduced in the proximal septal branch and was navigated into the distal LAD through the distal septal branch. Because its angulation and tortuosity were insignificant, wire manipulation to the distal LAD was relatively smooth. Once the microcatheter was advanced to the distal septal branch, the wire was navigated into the CTO retrogradely while exchanging the wire for a stiffer one (PT-2 LS, Boston Scientific). However, because the previously-created false lumen remained in the CTO, the wire easily migrated into the subintimal space and could not capture the proximal true lumen (Figure 2D). We then introduced another wire (Miracle 3, Asahi Intecc) with a microcatheter antegradely. The antegrade wire was also easily advanced into the false lumen. Several wire manipulations to connect the antegrade wire to the retrograde one in the distal true lumen were unsuccessfully attempted. Balloon dilatation of the distal false lumen via the distal septal branch was therefore performed. A 1.3 mm Lacrosse monorail balloon (Goodman, Japan) was introduced into the distal false lumen via the septal collateral in a retrograde fashion and was dilated with nominal pressure. This maneuver facilitated advancement of the antegrade wire into the distal septal true lumen. Finally, a Conquest Pro guidewire was able to penetrate the partition between the false and true planes toward the distal septal true lumen (Figure 2E). A continuous connection was consequently created. The true lumen of the distal main LAD was then successfully negotiated with a Fielder FC guidewire in an antegrade fashion. Balloon dilatation was performed and 2 Cypher stents (2.5 mm; 28 mm) were deployed, with a sufficient angiographic result (Figure 2F), and echocardiography showed that the patient’s left ventricular function subsequently improved.

With increasing operator expertise and the development of dedicated devices, the success rate of percutaneous treatment for CTOs has improved significantly. Newly-developed antegrade strategies including the parallel-wire technique, the IVUS-guided technique4 and subintimal tracking5 favorably contribute to the technical success rates. At present, the primary reason for failure of CTO-PCI is the inability to successfully cross the lesion into the distal true lumen with guidewires. Determining the exact entrance of the CTO using multipleview angiograms is crucial before commencing wire manipulations. The inability to determine the true entrance poses possible limitations for the antegrade approach and is one of the contributing factors of procedural failure. These difficulties are experienced when treating an occlusion with several branches around its origin, as in the first patient discussed here, or an occlusion in which the vascular stump has not been precisely identified, as in the second patient. Although IVUS can be useful in assessing the correct entrance of a CTO in certain cases, it might have been less useful in the first patient here due to diffuse narrowing and calcification of the diagonal branch, rendering it difficult to introduce the IVUS catheter. Branch dilatation is useful for introducing an IVUS catheter, however, it has a potential risk of dissection and may ultimately compromise subsequent wire manipulations made to select the correct route. In the second patient, the first attempt failed owing to a miscalculation of the CTO’s entrance, which was then confirmed by IVUS during the procedure. A particular angiographic configuration around the entry of the CTO makes it difficult to navigate the guidewire along the correct route. The absence of an appropriate side branch to allow introduction of the IVUS catheter impinges on conjunctive interrogation to determine the correct entrance. Despite detailed angiographic and IVUS investigation, there remain limitations in determining the correct entrance of the CTO, particularly for the antegrade approach, as in these two cases. Recent advancements in multidetector computed tomography technology may provide useful information for the location of the correct CTO entrance and its entire course.

Conversely, even in these two cases, the distal edge of the CTO was clearly visualized angiographically, allowing for precise placement of the wire tip at the distal exit point if the retrograde approach was successfully performed. The retrograde approach has been recently applied in CTO PCI with encouraging results.6 It has been suggested that the histology of the distal fibrous cap may be thin and less calcified than the proximal fibrous cap.7 These factors make the distal fibrous cap more amenable to penetration with a guidewire than the proximal one, and may reduce the guidewire’s tendency to migrate into the false lumen. Indeed, by using less stiff guidewires, successful crossing of the entire occlusion was achieved without significant difficulty in the first patient. The guidewire was finally advanced into the ascending aorta via the proximal LAD and left main trunk (LMT). It is suggested that advancing the guidewire via a retrograde approach through the occlusion back into the proximal LAD and LMT runs the risk of dissection. If the retrograde approach is conducted via a contralateral artery, it could result in LMT dissection. In the two cases described here, however, even in such instances, it is possible to restore flow immediately because the guidewire has already crossed the lesion in an antegrade fashion.

The second PCI attempt in the second patient appears to have been performed in an unconventional situation, as the retrograde wire was smoothly advanced through the lesion. Despite careful wire manipulation, however, the guidewire easily migrated into the previously-created false lumen and expanded it. Balloon dilatation was therefore performed, as described in the case presentation. This was followed by the recentlydescribed “Controlled Antegrade and Retrograde subintimal Tracking (CART)” method.8 The concept of the CART method involves connecting the antegradely-created subintimal plane to the retrogradely-created one. Retrogradely-introduced balloon dilatation plays an important role in navigating an antegrade guidewire into the distal true lumen. Our case was slightly different from the initially-demonstrated cases. The exit of the CTO was located at the septal bifurcation, and there remained only a small opening in the LAD’s true lumen for insertion of a balloon completely proximal to the bifurcation. Thus, we adopted the method to facilitate antegrade wire entry into the true lumen of the septal branch. Retrograde balloon dilatation from the distal septal branch toward the subintimal space of the CTO lesion successfully steered the antegrade guidewire into the septal true lumen. Subsequent capturing of the true lumen of the distal LAD with an antegrade guidewire was relatively smooth. The antegrade guidewire was successfully advanced into the true lumen of the septal branch via different approaches; nevertheless, it was difficult to negotiate the distal true lumen of the LAD at the time of the antegrade approach. Although the precise distal true lumen of the LAD was not successfully identified at the time of the antegrade approach, it would be related to the direction and location of the wire penetration. The penetration point for the retrograde approach appears to be more central, and not marginal, to the distal fibrous cap. The superiority of retrograde balloon dilatation to antegrade dilatation is attributed to the fact that the septal true lumen is connected to the distal true lumen of the LAD.

To our knowledge, these are the first case reports of retrograde PCI of a CTO via the ipsilateral intraseptal collateral in a mid-LAD CTO. Three routes have been previously described as retrograde approaches to reach the distal cap of the occlusion:3,6 1) via a bypass graft in patients who have previously undergone coronary artery bypass graft surgery; 2) via the epicardial collateral vessel; and 3) interseptal collaterals (between septal perforators and anterior and posterior descending arteries) in cases of RCA CTO or proximal or mid-LAD CTO.
Aside from anatomical problems, one of the major concerns during this procedure is the risk of jeopardizing a functional graft or nonoccluded coronary arteries. Use of this technique must be carefully evaluated, as damaging a donor conduit may result in serious consequences. In the first patient, there was a prominent epicardial collateral from the RCA to the distal LAD. However, this collateral was not suitable for the retrograde approach because it arose from the stenosed segment. Dilatation or stenting of this type of lesion runs the risk of compromising the collateral channel, even if protected with a guidewire.
Ipsilateral intraseptal collaterals have been frequently reported in mid-LAD occlusions.9 However, the exact frequency with which these collaterals grow to allow for the introduction of guidewires and, subsequently, balloon catheters is not known. This report shows the feasibility of using this artery for the retrograde approach. The technique described here is not always appropriate, as it requires adequately-sized collateral vessels. The most remarkable difference between the intraseptal channel and the interseptal one is its anatomical course. Intraseptal collaterals have a relatively sharp angle compared to interseptal collaterals, thus it is imperative that the devices pass through and do a U-turn at the angled point of the end of its course. Meticulous care is required when advancing devices to this site. A recent report has shown that interseptal collaterals are flexible and dilatable using a small-sized balloon.10 These collateral channels lie intramuscularly, thus it is suggested that intraseptal ipsilateral collaterals may tolerate device passage and become a potential source of access for the retrograde approach. In selected cases, this technique can be employed as an alternative to a conventional antegrade approach when the latter is considered unsuitable or is unsuccessful.




1. Olivari Z, Rubartelli P, Piscione F, et al. Immediate results and one-year clinical outcome after percutaneous coronary interventions in chronic total occlusions: Data from a multicenter, prospective, observational study (TOAST-GISE). J Am Coll Cardiol 2003;41:1672–1678.
2. Hoye A, van Domberg RT, Sonnenschein K, et al. Percutaneous coronary intervention for chronic total occlusions: The Thoraxcenter experience 1992-2002. Eur Heart J 2005;26:2630–2636.
3. Rosenmann D, Meerkin D, Almagor Y. Retrograde dilation of chronic total occlusion via collateral vessel in three patients. Catheter Cardiovasc Interv 2006;67:250–253.
4. Ito S, Suzuki T, Ito T, et al. Novel technique using intravascular ultrasound-guided guidewire cross in coronary intervention for uncrossable chronic total occlusions. Circulation 2004:68:1088–1092.
5. Colombo A, Mikhail GW, Michev I, et al. Treating chronic total occlusion using subintimal tracking and reentry: The STAR technique. Catheter Cardiovasc Interv 2005;64:407–411.
6. Kukreja N, Serruys P, Sianos G. Retrograde recanalization of chronically occluded coronary arteries: Illustration and description of the technique. Catheter Cardiovasc Interv 2007;69:833–841.
7. Masayuki K, Fujiwara H, Miyake M, et al. Histologic studies in percutaneous transluminal coronary angioplasty for chronic total occlusion: Comparison of tapering and abrupt types of occlusion and short and long occluded segments: J Am Coll Cardiol 1993;21:604–611.
8. Surmely JF, Tsuchikane E, Katoh O, et al. New concept for CTO recanalization using controlled antegrade and retrograde subintimal tracking: The CART technique. J Invasive Cardiol 2006:18;334–338.
9. Levin DC. Pathways and functional significance of the coronary collateral circulation. Circulation 1974;50:831–837.
10. Surmely JF, Katoh O, Tsuchikane E, et al. Coronary septal collaterals as an access for the retrograde approach in the percutaneous treatment of coronary chronic total occlusions. Catheter Cardiovasc Interv 2007;69:826–832