The Draw-Back Stent Deployment Technique: A Strategy for the Treatment of Coronary Branch Ostial Lesions

Treatment of isolated ostial lesions of sidebranches continues to represent a challenge for the interventional cardiologist. In some ways similar to aorto-ostial lesions, these lesions are highly elastic and subject to recoil and so are unsuitable for conventional balloon angioplasty. Mathias et al.1 reported a primary success rate with balloon angioplasty of only 74%. Procedural success with rotational atherectomy is better at 94%, but the angiographic restenosis rate is still high at 44%.2 Because they prevent elastic recoil, stents represent an attractive approach to the recanalization of these lesions. However, stent positioning is problematic. The lesion must be properly covered, but without encroaching on the lumen of the main vessel. We have developed an approach to these lesions entitled the draw-back stent deployment technique. We have found it to be user friendly and successful. This report describes our experience in 14 patients. METHODS Technique. A 7 French guiding catheter was used in all cases. In spite of this, difficulty was encountered in three cases in moving two full systems freely within the guide. In these cases, the balloon and guidewire system for the main vessel were exchanged for a fixed wire system. Figure 1 is an angiogram of one of the cases showing a severe localized lesion at the ostium of a first obtuse marginal (OM1) branch. Both the ongoing main vessel and the branch are wired. A slightly undersized balloon is chosen for the main vessel. This balloon is advanced distally into the main vessel beyond the origin of the branch. The sidebranch stent is selected to match the diameter of the branch and is usually a short stent because the lesions are localized. The stent is advanced distally into the branch well beyond the lesion. The balloon in the main vessel is then withdrawn to straddle the branch and inflated usually at nominal or slightly subnominal pressure (Figure 2). The stent in the OM1 branch is withdrawn until it contacts the inflated balloon in the main vessel (Figure 3) and is then deployed while the inflation pressure in the main vessel balloon is maintained (Figure 4). After deflation of both balloons and angiographic confirmation of patency of both the main vessel and branch, both systems are removed and a final angiogram is performed (Figure 5) to establish a good target lesion result without aggravation of the main vessel. Figure 6 is an intravascular ultrasound (IVUS) of the withdrawal from the OM branch into the circumflex artery. Figure 6A is an IVUS at the very ostium of the sidebranch showing satisfactory stent deployment and lesion treatment. Figure 6B is the very next frame less than one millimeter proximal showing the circumflex proper with no evidence of stent protrusion, i.e., perfect stent position. Figure 7 is a schematic illustrating the technique. Patient population. The procedure was attempted in 14 consecutive patients fulfilling the following criteria: 1) a significant (>= 50% diameter stenosis) ostial lesion of a large sidebranch (>= 2.5 mm visually); 2) the sidebranch arises at close to a 90° angle from the main vessel; 3) no significant disease in the main vessel segment straddling the sidebranch; and 4) this main vessel segment had not been previously stented. Procedural success was attained in 13 patients. In the single unsuccessful case, the angulation into the sidebranch became less perpendicular after the wires were in place; on withdrawing the stent, in rode along side the main vessel inflated balloon rather than being contained by it. The technique was abandoned in this case and stent deployment was performed with a single system. In this case, the stent was deployed across the main vessel with some aggravation of the ongoing main vessel; this was of no clinical consequence since there was a distal patent graft to the main vessel. Table 1 summarizes the clinical information on the successful patients, the quantitative coronary angiographic (QCA) measurements, and the procedural data. The mean age was 66 ± 9 years (range, 49–76 years). There were 9 men and 4 women. Three had diabetes. Ten had stable Canadian Cardiovascular Society (CCS) class 3 angina before the angioplasty, one had class 2 angina and 2 had unstable angina. The sidebranches in 6 cases were first (OM1) or second obtuse marginal branches (OM2) of the circumflex (CX) artery in 6 cases, a first diagonal branch of the left anterior descending artery (LAD) in 2 cases, and a branch of the first diagonal in 1 case. The “ostial” lesions were in the ongoing CX just beyond the take-off large OM1 branches in 2 cases; similarly, the “ostial” lesions were in the LAD just beyond the origin of large diagonal branches in 2 cases. In these cases, the branches served as the main vessels and the CX and LAD were the branches. As dictated by the inclusion criteria, there were no significant lesions (>= 50% diameter stenosis) visually seen in the main vessel in the vicinity of the origin of the branch; however, on QCA there was invariably some disease in the main vessel ranging from 15–41% diameter stenosis. There were additional lesions treated by stent deployment in 3 cases, but they were at a distance from the ostial branch lesions. Quantitative coronary angiography. QCA was performed independently and off-line by an experienced QCA technician. The QUANTOR.QCA system (Siemens AG, Forchheim, Germany) was used. Reference diameters for both the sidebranches and the main vessels were the interpolated measurements at the lesions and these were used for the absolute minimum lumen diameters and the percent diameter stenosis measurements. RESULTS The results are summarized in Table 1. The stents used were the Multi-Link DUET, Tristar or Tetra stents (Guidant Corporation, Temecula, California) in 8 cases; the GFX2 or S670 stent (Medtronic Inc., Minneapolis, Minnesota) in 3 cases; a NIR stent (Boston Scientific/Scimed, Inc., Maple Grove, Minnesota) in 1 case; and a Biodivisio (Biocompatibles Ltd, Surrey, United Kingdom) in 1 case. In the majority of patients, angina occurred shortly after the main vessel balloon was inflated and continued until both balloons were deflated. However, there were no peri-procedural complications or myocardial infarctions [defined as a serum creatine kinase (CK) level to more than double the upper limit of normal]. The mean CK before the procedure was 171 IU/L and 132 IU/L after the procedure (upper limit of normal Study limitations. The number of cases is small and a larger series will be required to establish the efficacy and safety of this technique. It has not been compared in a randomized study to the standard single-system stent deployment approach. The follow-up in some instances is quite short and there was no routine follow-up angiography. Silent restenosis may have occurred either in the stent or main vessel. IVUS was not routinely performed post-stent deployment to further verify the final position of the stent relative to the lesion and the main vessel and that subclinical trauma had not occurred in the main vessel segment straddling the sidebranch. It is evident even from this brief experience that it is suitable for only a special anatomy. Both the main vessel and branch must be of reasonable size and the branch in particular must be of sufficient diameter to accept a stent (>= 2 mm). The angle between the branch and main vessel should be close to 90 degrees. When it approaches 60 degrees, the inferior stent edge will by necessity protrude into the lumen of the main vessel at the apex of the branch origin. Customized beveled stents would represent an innovation for these anatomical variations. CONCLUSION Because of high sheer stress, atherosclerotic lesions have a propensity for bifurcations and branch ostia.4 In the former, more common situation, strategies involving “Y” stenting, “T” stenting or specifically designed bifurcation stents are required. However, in the latter situation, when the main vessel is undiseased or even mildly diseased, it should not be stented and attention should be focused on the branch ostial lesion. The technique described in this report represents a safe and effective approach to this anatomical subset. In our experience, having a strategy for accurately positioning and deploying a stent for such lesions, although somewhat time-consuming, reduces operator anxiety and leads to a more satisfying result. Acknowledgments. The authors wish to thank Dr. Alnoor Abdulla and Dr. John Ross for their encouragement and Mr. Sanh Bui for his technical expertise in the performance of the quantitative coronary angiographic measurements.

1. Mathias DW, Mooney JF, Lange HW, et al. Frequency of success and complications of coronary angioplasty of a stenosis at the ostium of a branch vessel. Am J Cardiol 1991;67:491–495. 2. Commeau P, Zimarino M, Lancelin B, et al. Rotational coronary atherectomy for aorto-ostial and branch ostial lesions. Cathet Cardiovasc Diagn 1994;32:97A. 3. Suwaidi JA, Berger PB, Rihal CS, et al. Immediate and long-term outcome of intracoronary stent implantation for true bifurcation lesions. J Am Coll Cardiol 2000;35:929–936. 4. Saltissi S, Webb-Peploe MM, Coltart DJ. Effect of variation in coronary artery anatomy on distribution of stenotic lesions. Br Heart J 1979;42:186–191.