Directional Atherectomy Research Articles

Randomized trials of myocardial revascularization

Over the past two decades there has been considerable refinement in randomized cardiovascular clinical trials. The common aim of randomized clinical trials of myocardial revascularization has been to understand the relative benefits of each technique on survival and nonfatal end points. The bypass surgery versus medicine trials that began in the 1970s provided evidence that the patients with advanced ischemic heart disease--three-vessel disease and/or substantially impaired LV function--have the most to gain from aggressive therapy (i.e., bypass surgery). In these cases, surgical revascularization provides survival benefit and has emerged as the reference standard for providing the most definitive revascularization. In long-term follow-up, however, surgery does not reduce the occurrence of myocardial infarction or angina compared with medical therapy. For patients with less extensive atherosclerosis and/or preserved ventricular function, trials comparing medical therapy, percutaneous coronary interventions, and bypass surgery have shown similar medium-term rates of death and infarction. In general, patients undergoing bypass surgery require the fewest subsequent antianginal medications, whereas patients undergoing PTCA require a moderate amount of antianginal medications, and patients treated solely with medical therapy require the most antianginal medications. Regardless, during long-term follow-up a similarly high number of patients are angina-free, although patients in the PTCA group require reintervention more often. Among patients treated percutaneously, techniques such as balloon angioplasty, directional atherectomy, stenting, rotablation, and laser may be considered. Compared with balloon angioplasty, greater acute gains in angiographic lumen have been obtained after directional atherectomy and stenting, but at the expense of increased periprocedural infarction after atherectomy, increased peripheral vascular complications after stenting, and increased late loss of lumen after both. Elective stenting has been associated with improved clinical outcome, whereas laser and rotational atherectomy have not, in comparison to balloon angioplasty. Restenosis remains the major limitation of all percutaneous approaches. Guidance for the individual patient is often less straightforward. Although general conclusions can be derived from patients cohorts in randomized trials, only a "gestalt" can be provided for the individual patient. For example, we have poor predictive capacity for restenosis, especially when this is recurrent despite repeated intervention. Only the demographic criteria of severe unstable angina and insulin-dependent diabetes mellitus are helpful in categorizing patients as "restenosis-prone." A substantial number of patients do not fit into the criteria adopted for entry into the revascularization trials--a point that is all too frequently forgotten.(ABSTRACT TRUNCATED AT 400 WORDS)

Design and initial testing of an ultrasound-guided directional atherectomy device

Directional atherectomy removes plaque in a targeted portion of a vessel wall. In practice, orienting the cutter toward maximal plaque accumulation and assessing the depth of vessel excision is difficult with angiographic guidance alone. Accoridngly, we designed and tested a prototype catheter that combines ultrasound imaging capability with directional atherectomy in a single device. Twenty-seven in vitro vessels (32 lesions) were treated with atherectomy alone or with atherectomy combined with ultrasound. Lesion characteristics before and after the procedure were similar in each group. A significant decrease in the incidence of subintimal tissue excision was observed in the atherectomy-ultrasound group (21.1%) compared to the group that had debulking with atherectomy alone (54.5%). In addition, among specimens with media and/or adventitia the relative amount of subintimal tissue was significantly less ( p < 0.001) with ultrasound guidance than with atherectomy alone (11.2% ± 10.1% vs 34.2% ± 8.5%). We conclude that ultrasound-guided directional atherectomy is technically feasible and may aid in achieving maximal plaque debulking and reduce the amount of subintimal injury.

Distal embolization is common after directional atherectomy in coronary arteries and saphenous vein grafts

Coronary embolization is a complication of coronary intervention procedures. The incidence, predictors, and clinical significance of this phenomenon during directional atherectomy were examined in 111 consecutive patients who underwent directional atherectomy to 120 lesions. Distal embolization occurred in 31 (28%) of the patients. It was noted mainly in the saphenous vein graft group of patients (12 [48%] of 25) versus the native coronary group (19 [22% of 86]; p = 0.01). Clinical predictors were age and de novo lesions. Morphologic predictors were larger artery size, larger postprocedure minimal luminal diameter, calcific lesions, and type C lesions. The only difference in clinical outcome was a longer hospitalization in the distal embolization group with 3.9 ± 3.7 days versus the rest of the patients 2.4 ± 2.4 days ( p = 0.01). In the majority of patients there was no significant adverse clinical outcome.

782-5 Balloon Angioplasty, Palmaz-Schatz Stent, and Directional Coronary Atherectomy for Restenotic Lesions: Retrospective Comparison in a Single Center

Both Palmaz-Schatz stents (PSS) and directional atherectomy (DCA) are reported to be suitable for treating restenotic lesions and claim to decrease subsequent restenosis. The purpose of this study was to compare in hospital outcome and follow-up (FU) of these two technologies with balloon angioplasty. Records of 1087 patients with restenosis that underwent elective (PSS) implantation (116 patients, 122 lesions). DCA (103 patients, 108 lesions), balloon angioplasty in 868 patients, 965 lesions) from January 1988 to March 1994 were reviewed. Mean follow-up was 1.22 ± 0.73 years. PSS DCA Balloon P Lesion success (%) 117 (95.9) 102 (944) 920 (95.3) 0.87 MLD post (mm) 2.87 ± 0.67 2.84 ± 0.77 2.20 ± 0.96 0.0001 In hospital death (%) 1 (0.9) 0 (0.0) 4 (0.5) 0.63 O-Wave MI (%) 2 (1.8) 1 (1.0) 15 (1.8) 0.85 CABG(%) 5 (4.3) 3 (2.9) 30 (3.5) 0.84 6 months follow-up PTCA(%) 31 (35) 30 (31) 247 (31) 0.74 CABG(%) 19 (22) 21 (22) 141 (19) 0.58 Recurrent angina (%) 34 (37) 34 (34) 340 (42) 0.25 Death (%) 9 (8.5) 4 (39) 37 (4.3) 0.14 Although a larger lumen was obtained with PSS and DCA, this was not associated with a significant reduction in revascularization, recurrent angina or death at six months. Event free survival curves reveal overlapping and low event free survival at 3 years.

935-31 Rebulking After Directional Atherectomy Compared with Balloon Angioplasty: Quantitative Lesion Morphology

We have previously shown that the proportional loss in minimum lumen diameter (MLD) after directional atherectomy (DCA) is greater than after balloon angioplasty (BA). To examine whether other parameters of lesion morphology show the same increase in loss with DCA. we analyzed the post-procedure and 6 month follow-up angiograms in 228 patients randomized to DCA or BA in the Canadian Coronary Atherectomy Trial (CCAT) in whom no crossover between procedures occurred. Initial gain and late loss in MLD, initial reduction (red) and late increase (Inc) in lesion length (LL) and initial “debulk” and late “rebulk” in plaque area (PA) were compared by procedure. Mean values ( ± sem), measured by automatic interpolative quantitative coronary analysis (CMS) are given: MLD(mm) LL(mm) PA(mm 2 ) (n) Gain * Loss * Red ** Inc * Debulk † Rebulk † DCA (111) 1.47 ± 0.04 0.81 ± 0.06 6.75 ± 0.47 5.58 ± 0.61 10.2 ± 0.53 6.8 ± 0.71 SA (117) 115 ± 0.04 0.41 ± 0.05 4.69 ± 0.47 2.80 ± 0.51 8.7 ± 0.59 3.9 ± 67 * P &lt; 0.001 ** P &lt; 0.01 † p &lt; 0.05 Compared to BA, DCA produced a significantly ( * P &lt; 0.001, ** P &lt; 0.01. † P &lt; 0.05) larger increase in MLD and decrease in both LL and PA (from pre- to post-procedure). However. this was offset by a greater late loss in MLD and increase in LL and PA (P values as above from post- to follow-up procedure). Resultant net changes (from pre- to follow up procedure) were not different. These results confirm that DCA more effectively “debulks” lesions. DCA. however. was associated with more “rebulking” in the follow-up period. so that the net results on quantitative lesion morphology of the two interventions were the same. These findings are consistent regardless of the quantitative angiographic parameter examined. DCA appears to have a device specific effect on the restenotic process which is probably related to the amount of vessel wall injury associated with this procedure

730-6 Histopathologic Findings in Ostial Coronary Lesions Treated with Directional Atherectomy

Conventional balloon angioplasty has had limited success in treating patients with ostial coronary stenoses. Directional coronary atherectomy (DCA) has been used in this setting with improved results. We hypothesized that this difference in device outcome is the result of differences in ostial lesion histopathology. Accordingly, we compared quantitative histopathology on tissue obtained from DCA in pts with ostial (n = 21) and non-ostial (n = 160) coronary stenoses. The % area occupied by adventitia, media, fibrocellular, sclerotic, and atheromatous tissue and new or old thrombus was calculated using computerized planimetry of appropriately stained specimens. Pts with ostial lesions were less likely to be hypertensive (29% vs 56%, p &lt; 0.05) and tended more often to be female (43% vs 26%, p = 0.09). In addition, pts with ostial lesions were less likely to have angiographic evidence of thrombus (0% vs 20%, p &lt; 0.05) or plaque ulceration (5% vs 29%, p &lt; 0.05). Reference segment diameter was similar in both groups. All pts with ostial lesions had successful DCA procedures. The histopathology (% total specimen area) is shown:Empty CellOstialNon-OstialP ValueAdventitia1.7 ± 2.80.3 ± 1.2&lt;0.001Media8.0 ± 15.62.3 ± 5.5&lt;0.01Fibrocellular50± 3334 ± 25&lt;0.01Sclerotic49 ± 3462 ± 25&lt;0.05Atheromatous15 ± 2429 ± 25&lt;0.05New thrombus8 ± 1613 ± 19NSOld thrombus11 ± 199 ± 18NS Thus, DCA of ostial lesions more commonly yields deep tissue elements compared to non-ostial lesions. In addition, ostial lesions are more cellular with a reduced content of sclerotic and atheromatous plaque. Given the high success rate of DCA in this group, it is possible that ostial plaque composition is ideally suited to DCA. These findings suggest that ostial lesions have a different histopathology which may be relevant to device selection.

730-2 Acute Angiographic, Intravascular Ultrasound and Clinical Results of Directional Atherectomy in the Optimal Atherectomy Restenosis Study

The Optimal Atherectomy Restenosis Study (OARS) is a 200 patient multicenter (4 sites) registry designed to assess an intravascular ultrasound (IVUS) guided “optimal” directional atherectomy (DCA) on acute angiographic results, in-hospital complications, and late outcomes. At present, preliminary acute results are available for 155 lesions in 146 consecutively treated patients. Baseline patient characteristics were age 58 ± 11 years, male gender 74%, LAD treatment vessel 54%, CCVS angina class III or IV 75%, diabetes 15%, and prior PTCA 23%. Pre-treatment target lesions were 7.9 ± 2.6 mm in length, 63% were eccentric, 13% were ulcerated, and 20% had either moderate or severe calcification. Overall, procedure success (&lt; 50% final stenosis without major complications) was achieved in 143 (98%) patients and there were in-hospital major complications in 3 (2.1%) patients including no deaths 10%). Q wave MI in 1 (0.7%), and emergency CABG in 2 (1.4%). Significant angiographic dissections ( ≥ grade C) after DCA were present in 9 (5.9%) patients and bailout stents were used to treat dissections in 4 (2.7%) without subsequent ischemic events. There was 1 perforation 10.7%) treated successfully with PTCA. Quantitative angiography was performed using an automated edge detection algorithm (CMS); reference diameter was 3.23 ± 0.48 mm and pre-treatment minimum lumen diameter (MLD) increased from 1.18 ± 0.44 mm to 2.74 ± 0.69 mm after DCA, and furtherto 3.14 ± 0.56 mm after adjunct PTCA (performed in 89% of treated lesions). Similarly, diameter stenosis was reduced from 63 ± 12% pre-treatment to 19 ± 19% after DCA to 7 ± 12% after adjunct PTCA. Sequential IVUS revealed an increase in lesion site cross-sectional area from 8.2 mm 2 following DCA to 9.0 mm 2 after adjunct PTCA (proximal reference area 10.5 mm 2 ). Despite the large final lumen dimensions (14% final cross-sectional area stenosis by IVUS). % cross-sectional narrowing (residual plaque burden) still averaged 57%. We conclude: (1) “optimal” DCA can be performed with high procedure success and few major in-hospital complications; (2) adjunct PTCA is usually required to achieve maximum lumen dimensions, and (3) despite these favorable angiographic results, IVUS reveals considerable residual plaque burden ( &gt; 50%) after “optimal” DCA.

1018-50 Procedural Results and Long Term Clinical Outcomes in Aorto-Ostial Saphenous Vein Graft Lesions After New Device Angioplasty

Balloon angioplasty of aorto-ostial (AO, ≤3 mm from aorta) saphenous vein graft (SVG) lesions has been associated with suboptimal results and high restenosis frequency. New angioplasty devices are being utilized to treat AOSVG lesions but comparative efficacy has not been established. Therefore, we compared the early procedural results and late clinical outcomes of 85 patients (90 lesions) treated with the Palmaz-Schatz stent (S) and 34 pts (41 lesions) treated with directional atherectomy (DCA). There was no difference in pt demographics between the 2 groups. TLR = target lesion revascularization, procedural success = final % stenosis &gt;50% and no major complications (death, MI, CABG), and event free survival = freedom from major complications or repeat PTCA at follow-up. stent DCA P value Angiographic Findings : Ulceration (%) 16.3 0.0 0.005 Lesion length/Reference size (mm) 8.9 ± 6.0/3.8 ± 0.7 5.3 ± 2.9/3.7 ± 0.7 0.0002/0.8 Acute gain (mm)/Final % stenosis 2.4 ± 0.9/4+7 2.1+0.7/13 ± 14 0.13/0.004 Clinical Findings : Major complications (%) 2.4 5.9 0.32 Procedural success (%) 96.5 94.1 0.62 %TLR @ 6/12 months 20.6/32.8 30.8/41.7 0.05 % Event free survival 6/12 months 73.6/52.8 57.8/42.4 0.15 1) Despite a higher frequency of adverse lesion characteristics including ulceration and longer lesions, S placement achieved improved angiographic results (lower final % stenosis) compared to DCA. 2) Similarly, late clinical outcomes favored S (vs DCA) with fewer TLR events and a higher event free survival. 3) Nevertheless, the overall results of these new device modalities (S and DCA) for AO-SVG lesions are discouraging (&lt;50% 12-month event free survival) and mandate further therapy refinements to improve late clinical outcomes.

782-4 Influence of Epicardial Lesion Location on Late Luminal Renarrowing and Angiographic Outcome After Successful Balloon Angioplasty, Directional Atherectomy, Stent Implantation and Excimer Laser Angioplasty

Left anterior descending (LAD) coronary artery location is reported to be associated with increased propensity to restenosis after intervention. This study, in 3736 patients, investigates whether this finding is equally applicable to intervention with balloon angioplasty (BA, n = 3797 lesions), directional atherectomy (DCA, n = 200 lesions), Palmaz-Schatz stent implantation (PS, n = 229 lesions) and excimer laser angioplasty (ELCA, n = 116 lesions). Automated core laboratory quantitative analysis (OCA) was performed in multiple matched angiographic projections pre and post intervention and at 6 month follow up (overall QCA follow up 93%). Multiple linear regression analysis was employed to determine whether LAD lesion location was associated with greater propensity to luminal renarrowing, independently of lesion severity pre, procedural luminal increase and reference vessel diameter. Because of the large variation in vessel size, lumen renarrowing was measured as relative luminal loss (minimal luminal diameter (MLD) post-follow up/reference diameter pre) and angiographic outcome as relative lumen at follow up (MLD at follow up/reference diameter pre). In multivariate analysis applied to the entire patient population, LAD location was independently associated with greater relative loss and smaller lumen at follow up (p = 0.007). However, application of the model to each patient group separately, revealed analogous findings for the DCA (p = 0.003) and BA (p = 0.03) groups, but no significant association between LAD location and late angiographic results was evident in the PS (p = 0.18) or ELCA (p = 0.55) groups. Retrospective exclusion of totally occluded lesions at baseline and/or at follow up from the analysis did not appreciable alter these findings. LAD lesions were independently associated with less favourable late angiographic results after balloon angioplasty or directional atherectomy, whereas late results after Palmaz-Schatz stent implantation or excimer laser angioplasty did not vary with lesion location. In contrast with previous reports, these findings suggest a differential influence of lesion location on late interventional results, depending on the device used. Accordinglyfurther investigation will be required for clarification of this association.

Multivariate predictors of intravascular ultrasound end points after directional coronary atherectomy

Objectives. This study attempted to identify the clinical, angiographic, procedural and intravascular ultrasound predictors of directional atherectomy results assessed by intravascular ultrasound.Background. Several angiographic and intravascular ultrasound variables have been associated with the outcome of directional coronary atherectomy. No study has incorporated both modalities into a predictive model.Methods. One hundred seventy patients were analyzed using preintervention and postintervention intravascular ultrasound and quantitative angiography. Clinical and procedural variables were collected by independent chart review. Quantitative and qualitative angiographic analysis was performed by a core laboratory in blinded manner. Intravascular ultrasound was performed using a transducer-tipped catheter, rotating within a stationary imaging sheath, and withdrawn automatically at 0.5 mm/s. Clinical, procedural, angiographic and ultrasound variables were tested iin a multivariate linear regression model. Dependent ultrasound variables included postatherectomy lumen cross-sectional area and percent cross-sectional narrowing (plaque plus media/external elastic membrane cross-sectional area) and, in a subgroup of 47 patients studied using volumetric analysis, percent plaque volume removal.Results. By multivariate stepwise linear regression analysis, predictors of residual lumen cross-sectional area (correcting for reference lumen area) included arc of calcium and preatherectomy plaque plus media cross-sectional area; predictors of residual cross-sectional narrowing were arc of calcium, preatherectomy plaque plus media cross-sectional area and lesion length; and predictors of percent plaque volume removal were arc of calcium and atherectomy device size.Conclusions. The preintervention lesion arc of calcium measured by intravascular ultrasound is the most consistent predictor of the effectiveness and results of directional coronary atherectomy.

760-1 The Precision Guided Directional Laser Atherectomy Catheter: A New Approach to Percutaneous Coronary Revascularisation

Directional coronary atherectomy has been shown to be a successful method of percutaneous coronary revascularisation, particularly in large vessels with eccentric stenoses. The application of laser technology to coronary revascularisation has not been as successful, mainly because of a relatively high rate of vessel wall perforation. We describe a combination endovascular imaging/laser atherectomy catheter (Precision Guided Directional Laser Atherectomy Catheter, PGDLAC) that may overcome these shortcomings by allowing precision guided intervention. The catheter consists of a laser cutting device with an integrated imaging system (fiberoptic or ultrasound). Fiberoptic imaging allows exact positioning and orientation of the laser tipped catheter within the vessel lumen, and visual control of plaque excision. In the second step, the vessel wall morphology is demonstrated using ultrasound, and the result of the intervention assessed without removing the catheter. Excised plaque fragments are extracted by way of the catheter, Preliminary results of in vitro experimental application of this device in human coronary arteries is promising and confirms the technical feasibility. Such a device potentially offers the advantages of effective, precision guided atherectomy in plaques of variable constitution, with a potential reduction in procedure time and radiation dose. By combining state of the art imaging and interventional technology, we have developed a precision guided directional laser atherectomy system that may provide more effective percutaneous coronary revascularisation than previously developed techniques.

760-6 Initial In-Vivo and In-Vitro Testing of an Ultrasound-Guided Directional Coronary Atherectomy Catheter

We report the results of initial in-vitro and in-vivo testing of a new, ultrasoundguided directional coronary atherectomy (GDCA) catheter (DVI, Redwood City, CA) which is being developed as a potential first-generation clinical product. A 30 MHz, focused transducer is incorporated in the body of the rotating cutter. Because of the relatively large aperture of the transducer (0.040”). resolution and penetration are superior to stand-alone IVUS catheters. In order to provide maximum rotational stability for imaging, a fixed-wire catheter configuration has been selected for initial production. The GTOTM shaft design has been adopted for torque control. 11 GDCA catheters, 8 with standard (3.25 mm) balloons and 3 with graft (4.5 mm) balloons, were tested in 4 normal farm pigs and 3 pigs with injury models of coronary stenosis. All catheters were able to be selectively directed into the target vesseil/lesion with good torque control. High resolution of the images allowed the clear demonstration of intima, venous structures, pericardium and myocardium. With balloon inflations the adequacy of engagement of the arterial wall by the cutter was directly visualized. In vitro testing using fresh porcine aorta compared the cutting ability of the combined device to the standard GTO. The weights of single cuts were not significantly different at 10 psi (5.8 mg vs 6.2 mg). and 30 psi (7.9 mg vs 8.1 mg). To fill the nose cone of the GDCA 6.4 cuts were required compared to 5.8 for the GTO (p =p ns). The weights of tissue from the filled nose cones were not significantly different (26.5 gm vs 20.5 gm). High resolution ultrasound imaging has been combined with a directional atherectomy catheter in a configuration suitable for clinical trials.

730-5 Why Do Aorto-Ostial Lesions Behave Differently than Non-ostial Lesions? Histologic Findings in Directional Atherectomy Specimens of Aorto-Ostial vs. Non-ostial Lesions

Conventional angioplasty of aorto-ostial lesions (≤3 mm from aorta) is frequently associated with suboptimal results, high procedural complications, and increased restenosis rate compared to non-ostial lesions. To gain insight into the differences between aorto-ostial and non-ostial lesions, directional atherectomy (DCA) samples of de novo native aorto-ostial lesions from 32 patients (64 ± 11 years; 22 men) were compared to randomly selected DCA samples of de novo native non-ostial lesions from 34 patients (61 ± 11 years; 23 men). (1) The amount of tissue removed was similar between the two groups (6.4 ± 3.6 μm 2 vs. 6.2 ± 3.8 μm 2 ). with predominantly fibrosclerotic plaque (59% vs. 68%): (2) The aorto-ostial group had significantly more areas of high cellularity vs. non-ostial group (0.73 ± 0.74 μm 2 vs. 0.41 ± 0.49 μm 2 ; p &lt; 0.05): (3) The aorto-ostial group tended to contain more thrombus (0.61 ± 0.79 μm 2 vs. 0.29 ± 0.35 μm 2 ; p &lt; 0.08); and (4) the aorto-ostial DCA samples more frequently had adventitial tissue (0.1 0 ± 0.14 μm 2 vs. 0.04 ± 0.06 μm 2 ; p = 0.07). We conclude that: (1) The aorto-ostiallesions seem to experience greater spontaneous trauma and stimulus for proliferation, as evidenced by the more abundant reactive cellular component. This histologic confirmation of the known rheologic effect on aorto-ostial regions may explain the increased restenosis rate observed with these lesions.: (2) The aorto-ostiallesions are more unstable, with a trend for more thrombotic component. This may explain the high procedural complications during angioplasty of the aorto-ostiallesions.: and (3) DCA specimens from the aortoostial lesions tended to have greater amount of adventitia, another potential source for procedural complication and increased restenosis. These findings suggest that there are biologic explanations for observed suboptimal acute and chronic results after angioplasty of aorto-ostial lesions, and further suggest a need for an improved strategy, such as minimization of the aorto-ostial turbulent flow possibly with stenting combined with local pharmacologic treatment.

990-49 Inadequate Pre-Hospital Medical Therapy in Patients Undergoing Percutaneous Coronary Revascularization: Results from the CAVEAT Trial

Patients with known coronary artery disease rnay be receiving inadequate medical therapy prior to referral for percutaneous coronary revascularization. To examine the medical treatment of patients referred for percutaneous coronary revascularization, we analyzed pre-hospital treatment data and concomitant medical conditions from the CAVEAT I trial. Of the 1,012 patients referred for revascularization 68.3% had unstable angina, 17.6% had stable angina, 9.8% had chest pain after myocardial infarction, and 4.3% were asymptomatic with a positive functional study. Prior to PTCA or directional atherectomy (DCA), patients had low rates of treatment with aspirin, nitrates, calcium channel blockers, beta-blockers, and anti-lipid agents (Table). Pre-hospital medications were associated with the following conditions: aspirin-none, nitrates-none, calcium channel blockershypertension (p = 0.004) and absence of unstable angina (p = 0.046), beta blockers-hypertension (p = 0.01), and anti-lipid agents-hyperlipidemia (p = 0.0001) and number of diseased vessels (p = 0035). These data strongly indicate that patients referred for percutaneous coronary revascularization have inadequate pre-hospital medical treatment. PTCA (N = 500) DCA (N = 512) Aspirin 60.4% 58.6% Nitrates 53.8% 51.8% Ca Blockers 49.0% 45.7% Beta Blockers 32.2% 34.8% Anti-lipid Agents 11.0% 11.1%

Differential Expression of Tissue Factor Protein in Directional Atherectomy Specimens From Patients With Stable and Unstable Coronary Syndromes

Tissue factor (TF) is a cell membrane-associated protein that catalyzes the rate-limiting step of the extrinsic coagulation pathway, which is the major source of thrombin production in vivo. To explore the potential role that TF may play in ischemic coronary syndromes, directional coronary atherectomy specimens were tested for the presence of TF protein using immunohistochemical techniques. Frozen sections from atherectomy specimens in 61 patients were examined for TF expression using an IgG murine monoclonal antibody against human TF. Patients were classified according to their admission diagnosis as having either an unstable or a stable coronary syndrome. An unstable coronary syndrome was defined as either angina pectoris occurring at rest or post-myocardial infarction (< 1 week) angina. Stable coronary syndromes included patients with stable, progressive, and new-onset (< 6 weeks) angina without rest pain. TF was detected in 15 (43%) of 35 patients with unstable coronary syndromes versus only 3 (12%) of 26 patients with stable coronary syndromes (odds ratio, 5.7; 95% confidence interval, 1.3 to 24.3; P = .018). Within the subgroup of patients with unstable coronary syndromes, TF was detected in 14 (60%) of 25 patients with de novo lesions versus only 1 (10%) of 10 patients with a restenosis lesion (P < .02). An additional 8 patients with stable coronary syndromes due to a restenosis lesion were also negative for TF. Therefore, the overall incidence of TF expression was only 6% (1 of 18) in restenosis lesions compared with 33% (14 of 43) in de novo lesions (P < .03). This study provides the first description of TF protein expression in human coronary artery lesions in vivo. Tissue factor was readily detected in de novo lesions in patients with unstable coronary syndromes, suggesting a role for TF in the pathogenesis of this disease process. Conversely, TF was rarely detected in patients with restenosis lesions even if the resulting clinical presentation was an unstable coronary syndrome. These results may have implications for the management of patients with unstable angina from de novo lesions and patients with ischemic symptoms from a restenosis lesion.

918-21 Intravascular Ultrasound Evidence of Significant Oversizing of Angioplasty Balloons in Female Patients

Although women may have an increased risk of complications following interventions, the etiology of this gender-related effect remains uncertain. While oversizing of angioplasty balloons is associated with increased complications, no angiographic data have demonstrated inappropriate balloon sizing in women. We performed intravascular ultrasound and quantitative angiography in 124 patients (101 males, 23 females) before and after coronary balloon angioplasty. In 51 patients, balloon angioplasty was the primary therapy, while in 73, balloon dilatation was performed as an adjunct to rotational or directional atherectomy. Separate core laboratories, blinded to outcome, measured angiographic and ultrasound dimensions in reference segments and target lesions. Results Men Women p value Angio Reference (mm) 3.50 3.29 NS Ultrasound Reference (mm) 3.37 2.85 0.001 Balloon Size (mm) 3.48 3.27 0.10 Balloon/Angio Ratio 1.00 1.02 NS Balloon/Ultrasound Ratio 1.06 1.17 0.05 These important differences in the ratio of balloon to ultrasound vessel diameter are partially explained by a greater plaque burden at reference sites in women (ultrasound area reduction of 43.7% vs. 37.6%). Although balloons were appropriately sized in women when compared to angiographic diameter, significant oversizing of balloon catheters in women was apparent when compared to ultrasound reference segment dimensions. These findings have important clinical implications for coronary interventions in women.

1018-51 Histomorphological Features in Primary and Restenotic Plaque Tissue Analyzed by a Relational Database System

A relational database was developed to enable relational analysis and correlation of clinical data with morphologic changes of the arterial vessel wall. A database system was designed which files all clinical, histological and immunohistochemical data acquired by serial analysis of 177 primary and restenotic tissue specimen obtained by directional atherectomy. The database has been developed in a 4 th Dimension (ACIUS, France) environment. Immunohistochemical analysis was performed of cellular and extracellular components with 35 antibodies and the results were added to the database to allow a relational analyzation of pathogenetic depencies. The percentage of positive stained cells in 3 μm thick serial cross-sections of biopsies in a standardized manner (magnification × 400, l0 × 10 frames). All values are given as percent of all present cells. Diabetic patients had a significant increase of foam-cells (31%) and collagen I (60%) compared with all other patients (18%, 32%). Smoking and high cholesterol together led to the highest incidence of restenosis (50%; total population: 27%) followed by smoking alone (41%) and was associated with enhanced cellular activation (i.e. high proliferation index and reduction of cytoskeletal filaments). Restenosis had the lowest occurrence of foam-cells (&lt;5%) and a significant reduction of contractile filaments (α-SMC-actin: 65%, desmin: 14%) versus 85% and 54% in all biopsies. Restenosis, diabetes and smoking together was associated with the highest occurrence of proliferation (restenoses 9%, diabetes and smoking 6.1%). Diabetes was associated with prolonged inflammation and enhanced extracellular matrix content. Determinants of clinical outcome after angioplasty were high cholesterol levels, smoking and cellular activation in primary lesions. (Supported by a grant from Friedrich Baur Stiftung No. 58/94).

Enhanced intracoronary thrombolysis with urokinase using a novel, local drug delivery system. In vitro, in vivo, and clinical studies.

Current pharmacological regimens for treating intracoronary thrombus in the cardiac catheterization laboratory generally involve the administration of thrombolytic agents that result in a systemic fibrinolytic state and/or require prolonged arterial drug infusion. The purpose of the present study was to assess a new technique for treating intracoronary thrombus consisting of the local infusion of limited quantities of urokinase with a novel drug delivery device. THe Dispatch coronary infusion catheter is a new local drug delivery system that allows for the prolonged infusion of therapeutic agents at an angioplasty site while distal coronary flow is maintained. Three experimental protocols were performed to determine the in vitro, in vivo, and clinical efficacy of this device. First, in vitro thrombolysis of fresh, porcine thrombus trapped in a 4-mm plastic tube with a 50% constriction and perfused with 20% porcine plasma was measured. Twenty-three thrombi were weighed before and after no treatment (n = 5), "systemic" urokinase administration (n = 4), local infusion of 150,000 U urokinase with a standard end-hole catheter (n = 4), local infusion of saline with the Dispatch catheter (n = 5), and local infusion of 150,000 U urokinase with the Dispatch catheter (n = 5). Second, 25 porcine coronary arteries in 23 pigs were dilated in vivo with conventional balloon angioplasty and then treated with 123I-labeled urokinase that was administered either by the Dispatch catheter (150,000 U; n = 16), intravenous systemic bolus (1,000,000 U; n = 3), guiding catheter infusion (500,000 U; n = 3), or local end-hole catheter infusion (150,000 U; n = 3). All vessels were subsequently harvested to quantify intramural deposition and subsequent washout of urokinase at the angioplasty site. Finally, 19 patients with angiographic evidence of intracoronary thrombus were treated with local urokinase infusion with the Dispatch catheter either before or after balloon angioplasty or directional atherectomy. In vitro studies demonstrated that infusion of urokinase with the Dispatch catheter decreased thrombus weight by 66% compared with no treatment (-25%), "systemic" urokinase administration (25%), end-hole catheter urokinase infusion (32%), or infusion of saline by the Dispatch catheter (32%) (P < or = .005). In vivo studies demonstrated immediate deposition of 0.12% of the urokinase delivered by the Dispatch catheter to the angioplasty site, compared with 0.0007% with systemic bolus, 0.003% with guiding catheter infusion, and 0.007% with local infusion with an end-hole catheter (P < .001). Urokinase deposited by the Dispatch catheter persisted intramurally for at least 5 hours. Patient studies demonstrated reduction of thrombus-containing stenoses and complete disappearance of intracoronary thrombus in all cases in which 150,000 U urokinase was locally infused over 30 minutes. There was no evidence of abrupt closure, distal embolization, or no reflow in any patient. Local urokinase delivery with the Dispatch catheter can result in rapid and complete intracoronary thrombolysis using substantially less drug than standard thrombolytic techniques. Intramural deposition of drug with this technique creates a local reservoir of urokinase that may provide prolonged thrombolytic activity at the infusion site.

918-7 Limitations of Percutaneous Interventions in the Treatment of Bifurcation Lesions Involving the Left Anterior Descending Coronary Artery

Serious complications may occur when intervention is unsuccessful in bifurcation lesions involving the left anterior descending (LAD) and first major diagonal (D), because of the large amount of involved myocardium. To determine this complication rate, we reviewed 82 consecutive cases, over a 3 year period, in which these lesions were attempted. Sixty-six percent of the subjects were male, and 37% had unstable angina. The mean age was 59 and the mean ejection fraction was 56%. Digital calipers were used to measure vessel minimum lumen (MLD) and reference diameters. For the LAD the final MLD was 1.81 mm and for the 0 1.32 mm. The final percent mean diameter stenoses for the LAD and D were 41% and 45%, respectively. There were no significant differences in the rates of success or complication between groups treated with angioplasty only (N = 68) or directional atherectomy (N = 14). The in-hospital event-free success rate was 55%. The in-hospital complication rates were: Recurrent Ischemia 16% Ventricular Tachycardia 2% Myocardial Infarction 14% Stroke 2% Bypass Surgery 12% Death 1% Repeat Procedure 4% Composite 34% LAD bifurcation lesion intervention is associated with a high in-hospital complication rate. Since these lesions are not amenable to stent placement or atherectomy with simultaneous protection of both vessels, these cases should be carefully evaluated before intervention, and bypass surgery should be considered as a treatment option.

950-94 Real-time Three-dimensional Intracoronary Ultrasonography: High Resolution Dynamic Images of Coronary Artery Lesions

Intravascular ultrasound displays coronary arteries in independent crosssectional images with no axial information. Three-dimensional reconstruction is mandatory, but cardiac cycle linked vessel motions and torsions of the arteries do not allow to apply simple devices that are used for peripheral vessels. Methods and results: We used a 2.9 F 30 MHz intravascular imaging system in combination with a motorized catheter pullback device. ECG and respiration triggering was performed and cubic datasets were acquired with a time resolution of 25 frames/s. Axial increments were set at 0.1 mm over a distance along the vessels of 25 to 35 mm. After processing the original registrations, three-dimensional real-time images could be created in any plane through the dataset. 25 patients with coronary artery disease were investigated before and after therapeutic interventions including Palmaz-Schatz stents (n = 6), laser angioplasty (n = 5), directional atherectomy (n = 41, and balloon angioplasty (n = 21). The system provided excellent images in all patients. Since the spatial geometry of the stents was known the reliability of the three-dimensional images could easily be proven. Dissection membranes, plaques and vessel takeoffs were clearly visualized in their spatial orientation, and the effect of interventions could be demonstrated in all cases. Major problems arise from tortuous vessels with regionally reduced original image quality and from surface definitions by user dependent settings of the noiselthreshold level. Real-time three-dimensional reconstruction of the coronary arteries is possible and extends future application of intravascular sonography.