Author + information
- Received June 11, 1999
- Revision received October 26, 1999
- Accepted December 2, 1999
- Published online March 15, 2000.
- S.H Wilson, MBBS, FRACPa,
- P.B Berger, MD, FACCa,
- V Mathew, MDa,
- M.R Bell, MBBS, FRACP, FACCa,
- K.N Garratt, MD, FACCa,
- C.S Rihal, MD, FACCa,
- J.F Bresnahan, MD, FACCa,
- D.E Grill, MSa,
- S Melby, RNa and
- D.R Holmes Jr., MD, FACCa,* ()
- ↵*Reprint requests and correspondence: Dr. David R. Holmes, Jr., Division of Cardiovascular Diseases, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905
The aim of our study was to compare the in-hospital and long-term clinical outcomes of direct coronary stenting with balloon predilation followed by stent placement.
With improvement in stent designs, the practice of direct stenting without balloon predilation has become more widespread.
We analyzed the Mayo Clinic Coronary Intervention data base between January 1, 1995 and March 5, 1999 and identified 777 patients who were treated with direct stenting (DS) and 3,176 patients treated with balloon angioplasty plus stenting (BA+S).
The procedural success rates between the DS and BA+S groups were not significantly different (96.3% vs. 96.4%). The ability to deliver the stent in a subgroup of patients who had DS was 95%, with 5% requiring crossover to predilation. Multivariate analysis showed no significant differences with respect to in-hospital death (odds ratio [OR] 0.9, 95% confidence interval [CI] 0.5 to 1.8), in-hospital myocardial infarction (OR 0.9, 95% CI 0.6 to 1.2) or revascularization (OR 0.7, 95% CI 0.4 to 1.5) in the DS compared with the BA+S group. Long-term outcomes were not significantly different between the DS and BA+S groups. The procedural duration was significantly shorter in the DS group, and there was a decreased utilization of contrast agent, balloons and wires.
The in-hospital and long-term clinical outcomes in patients undergoing a coronary intervention are equivalent when comparing stenting without balloon predilation with balloon angioplasty followed by stenting. Direct stenting is associated with decreased utilization of contrast agent and equipment and shorter procedure times. A randomized study should be performed to better determine the impact of this technique on short- and long-term procedural outcomes.
The use of coronary stents in interventional cardiology has increased exponentially over the last decade, both in bailout and elective settings (1–3). The standard stent implantation technique involves predilation of the target lesion with a balloon catheter to allow easy passage of the stent and to assess the likelihood of complete stent expansion after deployment (4). With improvement in stent designs, including the lower profile and greater flexibility of second-generation stents, placement of stents without balloon predilation has become more widespread. This technique is potentially less traumatic to the vessel wall and may cause less disruption to distal flow (4,5). This may be particularly relevant in the presence of thrombus or when treating degenerated vein grafts (6,7). Despite the increasing frequency of stent deployment without predilation, the immediate procedural and long-term results of this strategy are not clear. The aim of this study was to compare the in-hospital and long-term clinical outcomes of direct coronary stenting with the more conventional approach of balloon predilation followed by stent placement.
We performed a retrospective analysis of the Mayo Clinic Coronary Intervention data base and identified 777 patients who were treated with direct stenting (DS) and 3,176 patients treated with conventional balloon predilation followed by stenting (BA+S) between January 1, 1995 and March 5, 1999. All patients undergoing percutaneous coronary revascularization at the Mayo Clinic were followed according to a protocol approved by the Institutional Review Board of the Mayo Clinic and Foundation. This registry (8) includes baseline and in-hospital demographic, clinical and angiographic data.
Successful stent placement was defined as a ≥20% reduction in the lumen diameter stenosis, resulting in a final residual stenosis within the stent of <50% by visual estimation, with achievement of Thrombolysis in Myocardial Infarction (TIMI) flow grade 3, without the in-hospital occurrence of death, Q wave myocardial infarction (MI) or a complication requiring immediate coronary artery bypass graft surgery (CABG). Acute MI was considered to have occurred when at least two of the following three criteria were met: 1) chest pain >30 min in duration; 2) persistent electrocardiographic changes suggestive of ischemia; or 3) characteristic elevations in serum creatine kinase levels, with a corresponding rise in the MB isoform (9). Q wave MI was defined as two of these criteria with new pathologic Q waves in the coronary distribution of the stented artery. Thrombus was considered to be present when one or more lumen-filling defects in the coronary artery were visualized with contrast agent present on three sides (10). Dissection was considered to have occurred when contrast staining was apparent within the vessel wall (10,11). Dissections were classified according to the classification system of the National Heart, Lung, and Blood Institute of the National Institutes of Health, Bethesda, Maryland. Single-vessel disease was defined as the presence of a ≥70% lesion in one of the three major coronary arteries or their major branches. Multivessel disease was defined as the presence of a ≥70% lesion in a major coronary artery or its major branches and a ≥50% lesion in a second coronary artery or its major branches. The angiographic severity of coronary artery disease was assessed visually using two orthogonal views. Lesions were classified according to the modified American Heart Association/American College of Cardiology lesion classification scheme by the interventional cardiologist performing the angioplasty procedure, immediately before the procedure was performed (12). Left ventricular ejection fractions were calculated using a modified Simpson’s rule from ventriculography performed in a 30° right anterior oblique projection.
Intracoronary stent implantation technique
Coronary angioplasty and intracoronary stent implantation were performed using standard percutaneous techniques (13) via the femoral, brachial or radial artery. Techniques for implantation of different stent designs have been described previously (14). In earlier time periods, 8F guide catheters and over-the-wire systems were used. In the more recent study period, rapid-exchange systems and 6F and 7F guide catheters were more commonly used. The decision whether to perform predilation was made by the operator. Lesions that were extremely tortuous or calcified were usually predilated based on operator judgment regarding the feasibility of stent deliverability. If a decision was made to proceed without predilation, the stent was usually deployed at the nominal pressure of the stent balloon, and a high pressure balloon was subsequently used for dilation within the stent to 14 to 16 atm. With some of the newer generation stents, the stent delivery balloon could be inflated to 14 atm. With these stents, postdilation with another balloon was not routinely performed. The first-generation stents used in the study included the Palmaz-Schatz (SDS, Johnson & Johnson, Interventional Systems, Warren, New Jersey), Gianturco-Roubin I (Cook Inc., Miami, Florida), Wiktor (Medtronic Inc., Minneapolis, Minnesota) and biliary stents (Johnson & Johnson, Interventional Systems). The second-generation stents used included the Multilink (Advanced Cardiovascular Systems, Temecula, California), Microstent II (Arterial Vascular Engineering, Santa Rosa, California), Crown (Cordis, Miami, Florida) and Gianturco-Roubin II (Cook Inc.) stents. Intravascular ultrasound was used at the discretion of the operator.
All patients received preprocedural oral aspirin (325 mg) and intravenous heparin to achieve an activated clotting time of ∼300 s. In patients receiving a combination of Coumadin (warfarin sodium) and aspirin, Coumadin was commenced on the evening of the procedure, and intravenous heparin was continued until the International Normalized Ratio was ≥2.0. In patients receiving ticlopidine (commenced in early 1995), the first dose was administered either immediately before (within 1 h) or during the procedure, with an initial loading dose of 500 mg and 250 mg later that evening and 250 mg twice a day for two to four weeks after the procedure (15), in addition to 81 to 325 mg of aspirin daily. After March 1998, clopidogrel was substituted for ticlopidine and was given as a loading dose of 300 mg either immediately before (within 1 h) or during the procedure, followed by 75 mg daily for two weeks. In early 1995, a number of patients received Coumadin for suboptimal stent deployment, as assessed by coronary angiography or intravascular ultrasound. Abciximab (a bolus of 0.25 mg/kg body weight followed by a 12-h infusion of 10 μg/min) was administered immediately before or during the procedure at the discretion of the operator. In patients receiving abciximab, heparin was given, aiming for an activated clotting time between 250 and 300 s.
In-hospital and follow-up events, including death, Q wave MI, all acute MIs, CABG and repeat angioplasty of the target vessel, were analyzed.
All patients were interviewed in person or by telephone six and 12 months after the initial procedure. Data from visits and hospital periods at the Mayo Clinic and other institutions were obtained for review. After January 1998, the stent deliverability rate was recorded in all those patients in whom placement of a stent without predilation was attempted.
Differences between groups in terms of clinical and angiographic variables were tested for statistical significance using the Pearson chi-square test for discrete data or the Student t test for continuous data. Multivariate logistic regression models were constructed using the forward selection process for each of the in-hospital end points. Once a model was constructed, a dichotomous variable indicating primary stenting was added to the model. Results are presented as adjusted odds ratios (ORs) with 95% confidence intervals (CIs). Multivariate Cox proportional hazards models were developed for each of the end points recorded during follow-up, again using forward selection. The adjusted relative risk was obtained by adding the variable for primary stenting to the model. Univariate survival curves were calculated using the Kaplan-Meier method; the log-rank test was used to test for differences between survival. Based on the sample size and follow-up of our study, we estimate that for the combined end point of death, MI and repeat revascularization, we had statistical power of ∼80% to detect a 3% change in the proportional event-free survival between the two groups at one year.
The clinical characteristics of the patients in the two groups are described in Table 1. There was no difference in the mean age, gender, history of hypercholesterolemia, hypertension, smoking or diabetes between the two groups. However, more patients had undergone CABG in the DS group than in the BA+S group (26.9% vs. 20.7%, p < 0.0001).
Angiographic and procedural characteristics (Tables 2 and 3)
⇑⇓There was no difference in the number of diseased vessels between the two groups, but in the DS group, more patients with saphenous vein graft (SVG) lesions were treated (19% vs. 9%, p < 0.001). In the DS group, more lesions contained thrombus (22% vs. 18%, p < 0.05), and there was a lower frequency of lesion calcification (34% vs. 47%, p < 0.001), more ostial lesions (29% vs. 21%, p < 0.001) and a lower incidence of bifurcation lesions treated (4% vs. 7%, p < 0.01). In addition, there was a less severe preprocedural lumen diameter stenosis (80.0 ± 16.8% vs. 87.3 ± 11.0%, p < 0.001) in the DS group.
There was greater utilization of second-generation stents in the DS group than in the BA+S group (p < 0.001). The maximal stent deployment pressure was slightly higher in the DS group (Table 3), but the ultimate inflation pressure used within the stent after deployment was similar between the two groups (15.6 ± 4.2 vs. 15.9 ± 4.6 atm, p = NS). The DS group had fewer dissections after stent implantation than did the BA+S group (24% vs. 41%, p < 0.001). However, there was an increase in the percentage of patients with reduced TIMI flow after stent deployment in the DS group, perhaps reflecting a higher number of patients with thrombus-containing lesions and degenerated vein grafts in this group. The final residual diameter stenosis was similar between the two groups.
In those patients who underwent stent placement within a bypass graft, there were trends similar to those of the overall cohort, with a higher incidence of thrombus-containing lesions (48% vs. 31%, p < 0.001) and lower preprocedural stenosis (83.4 ± 13.4% vs. 88.4 ± 10.8, p < 0.001) in the DS group. The graft age was significantly older in the DS group than in the BA+S group (9.6 ± 4.6 vs. 8.6 ± 4.8 years, p < 0.05). Again, as in the entire cohort, there were similar balloon inflation pressures within the stent between the two groups.
In-hospital events (Table 4)
The procedural success rates between the DS and BA+S groups were similar (96.3% vs. 96.4%). The frequency of in-laboratory coronary occlusion (2.8% vs. 3.0%) and branch occlusion (7.1% vs. 5.8%) was also similar. Direct stenting was associated with a similar risk of in-hospital MI in a multivariable analysis (OR 0.9, 95% CI 0.6 to 1.2, p = 0.47) after adjusting for vein graft interventions, the number of segments treated, anginal class and differences in the frequency of use of second-generation stents. In addition, after adjusting for significant covariates (see Methods), DS was associated with a similar frequency of in-hospital death (OR 0.9, 95% CI 0.5 to 1.8, p = 0.84) and repeat revascularization (OR 0.7, 95% CI 0.3 to 1.5, p = 0.4).
In the subgroup of patients in both groups who underwent bypass graft interventions, after adjusting for significant covariates, DS was associated with a similar in-hospital mortality (OR 1.1, 95% CI 0.4 to 3.3, p = 0.77) and risk of in-hospital MI (OR 0.8, 95% CI 0.7 to 3.0, p = 0.22).
Follow-up events (Fig. 1)
The median follow-up times available for the patients who successfully survived the hospital period were 177 days for the DS group and 379 days for the BA+S group. The six-month mortality rates were 2.2% for the DS group and 2.8% for the BA+S group (p = NS). The frequency of MI during this period was 2.4% for the DS group and 1.9% for the BA+S group (p = NS). After adjusting for significant covariates, the long-term adjusted risk ratios for DS were 0.7 for death (95% CI 0.4 to 1.3, p = 0.28), 0.8 (95% CI 0.6 to 1.2, p = 0.29) for death and MI and 0.9 (95% CI 0.7 to 1.2, p = 0.55) for the combined end point of death, MI and repeat revascularization. In the subgroup of patients with bypass graft interventions, there was no significant difference in long-term outcomes for the end points of death and MI or death, MI and repeat revascularization.
Procedural time, contrast agent used and equipment utilization
The procedural time in the DS group was significantly lower than that in the BA+S group (median 1.3 vs. 1.4 h, p < 0.001), as was the fluoroscopy time (median 25 vs. 20 min, p < 0.0001). The contrast load was significantly less in the DS group than in the BA+S group (median 270 vs. 285 ml, p < 0.01). The number of balloons used (1.7 ± 1.6 vs. 2.2 ± 1.4, p < 0.001), as well as guide wires used (2.0 ± 1.5 vs. 2.3 ± 1.8, p < 0.001), was also significantly less in the DS group than in the BA+S group.
This study demonstrates, for the first time, that the in-hospital and long-term clinical outcomes in patients undergoing a coronary intervention are equivalent using the novel technique of DS compared with the more conventional technique of balloon angioplasty followed by stenting. This similar outcome was achieved despite more high risk characteristics in DS group, including a higher frequency of previous CABGs, vein graft interventions and thrombus at the treatment site. These outcomes were achieved with shorter procedure times and decreased utilization of contrast agent and equipment in the DS group.
Direct stenting has become increasingly popular with the availability of more deliverable stents. This study, however, includes our early experience using the first-generation Palmaz-Schatz and Gianturco-Roubin stents, although the majority of DS procedures employed second-generation stents such as the ACS Multilink stent. This technique has been advocated previously in a select group of patients with a lesion length <12 mm, in the absence of calcification or tortuosity (4). In this study, the safety of DS in a larger group of patients is demonstrated, including those with vein grafts and thrombus-containing lesions. Patients with calcified lesions were also treated with DS in our study, although those with very heavy calcification were not. Direct stenting in this study was performed at the discretion of the operator; lesions within vein grafts and those containing thrombus were commonly selected for the technique. The DS group also had a lower preprocedural stenosis as compared with the BA+S group, reflecting the concern that tighter stenoses may be more difficult to cross without balloon predilation, and thus may lead to a greater risk of stent dislodgment from the delivery balloon. Other potential concerns associated with the technique of DS include the potential for stent loss and embolization, the possibility of incomplete stent expansion, compromised lesion and distal vessel visualization and subsequent difficulty with exact stent positioning. Optimal guiding support is also necessary if DS is attempted.
It may be hypothesized that DS may improve clinical outcomes, partly by decreasing the number of balloon inflations and deflations and decreasing ischemia time (4). Direct stenting is potentially less traumatic to the vessel wall and may cause less ischemia, dissection and disruption to distal flow. This may be particularly the case when thrombus is present or when treating aging vein grafts (6). Residual dissections after stenting have been associated with an increased risk of major adverse cardiac events (16), and the frequency of dissections after stent implantation in this study was lower in the DS group. Moreover, the use of stents in a bailout situation after severe dissection has been associated with a worse outcome (16), supporting the possibility that predilation may be harmful in a small number of patients. This potential risk is diminished with the technique of DS. In contrast to these potential benefits, the frequency of reduced TIMI flow after the procedure in this study was increased in the DS group, perhaps because of the treatment of more vein grafts and lesions containing thrombus. This decreased TIMI flow in the DS group did not lead to excess in-hospital complications, but is concerning nonetheless and warrants further investigation.
In this nonrandomized study, patients selected for DS were significantly more likely to require treatment of a bypass graft, and ∼20% of the DS group underwent a bypass graft intervention. Previous reports have suggested that SVG stenting without predilation may decrease procedural complications such as distal embolization, no-reflow phenomenon and periprocedural MI (6). These complications are more common in vein grafts as compared with native vessel interventions (17), occurring in up to 15% of procedures (18). In our study, the group treated with DS had significantly more thrombus present and older vein grafts, representing a group at higher risk for distal embolization and periprocedural MI. Despite this, treatment of SVG lesions with DS instead of balloon predilation had a similar rate of in-hospital events and similar long-term clinical outcomes. A randomized study is needed to further assess the impact of DS in this high risk subgroup of patients.
The technique of DS has potential advantages in terms of shortened procedural time, decreased radiation and contrast load received by the patient and decreased cost of the procedure. In our study, both the contrast load used and the procedural time were decreased by DS. These decreases may be clinically important, especially in the high risk group of patients. In addition, we found a decrease in the number of balloons used in the DS group, potentially leading to a lower overall procedural cost.
We are not able to comment on the overall success rate of stent delivery in our cohort, as this study was retrospective, and for the initial part of the study, we were unable to ascertain how frequently patients in the DS group crossed over and required predilation. This information was recorded after January 1998, and subsequent to this, the stent deliverability rate in the 406 patients in whom DS was attempted was 95%. The success rate of DS would be expected to improve as stent technology is further refined. However, the patients treated with DS in this study were selected because of an operator’s decision regarding their “suitability” for DS, and the general applicability of this technique needs further evaluation. Variability of the success of DS between different stent designs may occur, and evaluation in each stent type is important. It should also be noted that the DS group had a lower preprocedural stenosis as compared with the BA+S group. Another limitation is that routine angiographic follow-up was not performed, and the frequency of angiographic restenosis in the two groups is unknown. However, the clinical event rate at six months was similar between the two groups.
This study demonstrates that the in-hospital and long-term clinical outcomes in patients undergoing a coronary intervention are equivalent using the novel technique of DS as compared with the more conventional technique of BA+S. The outcome was similar, even though the DS group included a higher risk group of patients. There was decreased utilization of contrast agent and equipment and shorter procedure time in the DS group. A randomized study is needed to further assess the ability of this technique to reduce the risk of short- and long-term procedural complications and resource utilization.
- balloon angioplasty plus stenting
- coronary artery bypass graft surgery
- confidence interval
- direct stenting
- myocardial infarction
- odds ratio
- saphenous vein graft
- Thrombolysis in Myocardial Infarction
- Received June 11, 1999.
- Revision received October 26, 1999.
- Accepted December 2, 1999.
- American College of Cardiology
- Palmaz-Schatz Stent Study Group,
- Wong S.C,
- Baim D.S,
- Schatz R.A,
- et al.
- de Feyter P.J,
- van den Brand M.M,
- Laarman G.J,
- et al.
- TAUSA Investigators,
- Ambrose J.A,
- Almeida O.D,
- Sharma S.K,
- et al.
- Nath F.C,
- Muller D.W,
- Ellis S.G,
- et al.
- ↵Ryan TJ, Bauman WB, Kennedy JW, et al. Guidelines for percutaneous transluminal coronary angioplasty. Circulation 88:2987–3007.
- ↵Holmes DR Jr. Technical aspects. In: Vliestra RE, Holmes DR Jr., editors. Percutaneous Transluminal Coronary Angioplasty. Philedelphia: F.A. Davis, 1987:35–48.
- Hasdai D,
- Garratt K.N,
- Holmes D.R Jr.,
- Berger P.B,
- Schwartz R.S,
- Bell M.R
- Schuhlen H,
- Kastrati A,
- Dirschinger J,
- et al.
- Frimerman A,
- Rechavia E,
- Eigler N,
- Payton M.R,
- Makkar R,
- Litvack F