| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2008;7:871-877. doi:10.1510/icvts.2008.178632 © 2008 European Association of Cardio-Thoracic Surgery
Acute and long-term outcome of unprotected left main coronary angioplasty compared to the anticipated surgical risk
a Department of Cardiology, University Hospital of Jena, Erlanger Allee 101, 07747 Jena, Germany Received 27 February 2008; received in revised form 26 May 2008; accepted 17 June 2008
*Corresponding author. Tel.: +49 3641 9324101; fax: +49 3641 9324102.
Objectives: Percutaneous coronary intervention of unprotected left main (ULM) stenosis is increasing despite surgical revascularisation being the recommended treatment by the current guidelines. We compared the 30-day and 9-month mortality after ULM stenting with the predicted surgical outcome as determined by the EuroSCORE. Methods and results: We included 81 consecutive patients who underwent ULM stenting. The patient cohort was divided into a normal risk group (EuroSCORE  5, predicted 30-day mortality <3%), and a high-risk group (EuroSCORE >5, predicted 30-day mortality 11%). Follow-up examinations were scheduled for one and nine months after the initial PCI. The average EuroSCOREs in the normal- and high-risk group were 3.0 (29 patients, 36%) and 10.0 (52 patients, 64%), respectively. Mortality rates at 30 days were 3% in the normal-risk group and 8% in the high-risk group. Subacute stent thrombosis occurred in 3.4% (three patients) undergoing elective PCI. Major adverse cardiac events during the nine months follow-up were registered in 24% of patients in the normal-risk group, and in 27% of the patients in the high-risk group. Conclusion: Short-term outcome of ULM stenting in our high-risk patients is comparable to surgical outcome predicted by the EuroSCORE. Long-term outcome was less favourable due to a high mortality rate. ULM stenting in patients with low surgical risk could be associated with higher mortality rates compared to CABG because of the unpredictable risk of a fatal stent thrombosis.
Key Words: Left main coronary artery; Coronary artery bypass grafting; Angioplasty; Mortality; Risk stratification
Surgical revascularisation of unprotected left main stem (ULM) stenosis improved survival and symptoms compared to medical therapy [1]. In contrast, balloon angioplasty was initially associated with a high acute mortality rate [2]. However, due to technical advances, especially the availability of drug-eluting stents, ULM stenting offers an attractive alternative approach for inoperable patients. To date, it is not known whether patients with an acceptable surgical risk would also benefit from ULM stenting compared to coronary artery bypass graft surgery (CABG). Current guidelines, therefore, continue to recommend CABG as the standard treatment in ULM stenosis and indicate that ULM stenting should only be considered in inoperable patients [3]. However, an SCAI survey (Society for Cardiovascular Angiography and Interventions) of 515 interventionists [4] reveals that many of the interventionists perform ULM stenting individually even in patients with an acceptable surgical risk. Therefore, the purpose of this study was to compare the acute and long-term outcome after ULM stenting with the anticipated surgical risk as determined by the EuroSCORE. We chose the well established EuroSCORE [5] because risk stratification by this score yielded the highest predictive value in a comparison of six different risk scores in German heart surgery [6].
2.1. Study population A total of 81 consecutive patients with ULM stenosis who underwent PCI in two cardiac centres between 2003 and 2006 were included in the study. The left main coronary artery was defined as unprotected if there was no open bypass supplying the downstream vascular bed. All patients who underwent elective or emergency ULM stenting during the time period were included in this study. Stenting technique was performed using the standard percutaneous transfemoral approach, except in one patient who underwent PCI via the transradial approach due to extensive peripheral vascular disease. Lesions of the ostium or in the mid-segment of the ULM stem were treated with a single stent (n=29). Specific techniques, which have previously been described [7], were used for the treatment of distal bifurcational stenosis. These techniques were as follows: Single Stenting (n=34), Culotte Technique (n=3), T-stenting (n=7), Simultaneous Kissing Stent technique (n=2) and Crush Technique (n=6). The choice between drug-eluting stents (DES, n=65) and bare metal stents (BMS, n=16) was made by the interventional cardiologist. Additional dilatations using non-compliant balloons were performed for optimal stent positioning. Although intravascular ultrasound scan (IVUS) was sometimes necessary to confirm the optimal PCI result, it was not generally applied. Moreover, the intra-aortic balloon pump (IABP; n=40) or micro axial flow pumps (Impella 13F, AMED 16F; n=5) were employed in cases of complex stenosis for haemodynamic stabilisation. Percutaneous coronary intervention procedural success was defined as a Thrombolysis In Myocardial Infarction (TIMI) flow grade 3 with a final residual stenosis of <30%. All patients received Aspirin (100 mg/day) indefinitely, and a 300-mg loading dose of Clopidogrel followed by a daily dosage of 75 mg for one year after ULM-PCI. Cardiac markers (Troponin I, Creatine kinase) were measured routinely before and 24 h after ULM stenting. 2.3. Calculation and interpretation of EuroSCORE The standard EuroSCORE was applied to calculate the anticipated surgical risk of death within 30 days [5]. Patients were stratified as high-risk candidates in the presence of EuroSCORE >5. According to the EuroSCORE database, these patients have an anticipated mortality risk of 11% whereas patients at normal risk should have a mortality risk below 3%. The cumulative rate of major cardiac adverse events (MACE) and mortality was analysed after 30 days and at 9 months. The following complications are defined as MACE: cardiovascular death, non-fatal myocardial infarction, CABG, target lesion revascularisation (TLR), and subacute stent thrombosis. Death was classified as either cardiac or non-cardiac. Deaths that could not be classified were considered as cardiac. Diagnosis of procedure related non-ST-segment elevation MI (NSTEMI) was based on the presence of typical symptoms and ECG findings and a post-PCI elevation in cardiac markers (troponin I to >3-fold the upper limit of normal). The diagnosis of a periprocedural STEMI was based on ST-elevation in at least two contiguous ECG leads together with an elevation of CK to more than three-fold the upper limit of normal. In case of pre-existing elevated cardiac markers, a PCI-related MI was assumed when cardiac markers increase >2-fold from prior levels if elevated at baseline or when angina CCS-IV persisted or dyspnoea (as angina equivalent) was present. Telephone interviews of patients and/or their general practitioners were scheduled after 30 days and after nine months. Angiographic follow-up was performed 6–9 months after ULM-PCI, unless the patient refused the procedure. Target lesion revascularisation (TLR) was defined as repeated-PCI (Re-PCI) of the left coronary artery for dilatation of a luminal stenosis ( 50% lumen diameter reduction) in the pre-treated segment site. Surgical revascularisation was considered separately.
Continuous variables are presented as means with the lower and upper range and were compared by the Student t-test or Mann–Whitney U-test. The  2-test or Fisher's exact tests were used to determine the significance of differences in the categorical variables. Survival curves were generated by Kaplan–Meier's method. P-values <0.05 were considered statistically significant. Statistical analyses were performed using SPSS version 10.0 (SPSS Inc., Chicago, Illinois). Only two-sided tests were used.
3.1. Baseline characteristics Patients undergoing ULM stenting included those:
Demographic data and co-morbidity used for calculating the EuroSCORE are summarised in Table 1. A normal surgical risk was calculated for 29 patients (36%), and a high surgical risk in 52 patients (64%). The median EuroSCORE was 7 (range 0–20) in all patients. The median EuroSCORE of the normal risk group was 3 (range 0–5), and that of the high-risk group 10 [6–20].
Procedural characteristics of the whole cohort and the two subgroups are summarised in Table 2. Technical success was achieved in all patients.
3.2. Thirty-day follow-up Thirty-day follow-up is summarised in Table 3. The overall mortality at 30 days was 6.2% (five patients). Among the five patients, two high-risk patients who initially presented in cardiogenic shock died despite successfully performed emergency PCIs. One high-risk and one normal-risk patient died due to subacute stent thrombosis and one high-risk patient died from pneumonia. Mortality in the high-risk subgroup at 7.7% was below the predicted surgical mortality. In contrast, the normal-risk subgroup had a slightly higher mortality of 3.4% compared to surgery (Fig. 1).
The rate of short-term MACE is shown in Table 3. There was no difference in MACE rates between the risk subgroups. The characteristics of the patients with subacute stent thrombosis are illustrated in Table 4.
3.3. Nine-month follow-up The long-term follow-up is shown in Table 5. Whereas no difference in the long-term MACE rate was observed, high-risk patients had a three times higher all-cause-mortality during the 9-month follow-up compared to normal-risk patients. However, this difference between the subgroups diminished when the rate of cardiac mortality was analysed (Fig. 2). Control angiography performed in 65 patients approximately nine months after ULM stenting confirmed restenosis in six patients. This complication was independent of the surgical risk. All but one patient had clinically asymptomatic restenosis. The characteristics of patients with restenosis are listed in Table 6. The rate of restenosis in the BMS group was consistently higher than in the DES group (19% vs. 5%, respectively). In addition, two patients underwent CABG four months after ULM stenting, one patient because of TLR, and one patient due to progressive coronary artery disease, despite good long-term results after ULM stenting.
Surgical revascularisation is currently the standard treatment for patients with ULM stenosis. Although not recommended by the guidelines, ULM stenting is performed frequently in elective patients who are eligible for CABG. However, it is unknown whether ULM stenting is equal or even superior to CABG in all risk subgroups. To date, results of ULM stenting are heterogeneous depending on the risk subgroups tested, the anatomical characteristics of the ULM lesion, and especially the type of stent used [8–10]. In the present study, the EuroSCORE, which was used to anticipate the surgical risk was compared with the observed mortality after ULM stenting. The 30-day mortality following ULM stenting of 6.2% in our study corresponds to the published acute mortality of 0–14% in other ULM-studies depending on the included risk groups [8–12]. In normal-risk patients, mortality after PCI was slightly higher than that predicted by the EuroSCORE. The 30-day mortality for high-risk patients, after ULM stenting, was lower than the predicted EuroSCORE value. This favourable short-term outcome is most likely due to the reduced invasiveness of the PCI-procedure compared to open heart surgery involving a cardiopulmonary bypass. However, when using the EuroSCORE as a comparative measure, improvements in cardiac surgery during the last decade must be considered. The EuroSCORE database was established 10 years ago. Since then innovative techniques such as OPCAB (off-pump coronary artery bypass) and MIDCAB (minimally invasive direct coronary artery bypass) have been implemented and postoperative management has also been continuously improving [13]. Recent studies involving CABG in patients with ULM stenosis performed between 2000–2005 showed an overall in-hospital mortality after on-pump surgery of 3.4% and of 1.1% after OPCAB, respectively [14]. It would seem that the EuroSCORE does not reflect this current progress in cardiac surgery, rather overestimating mortality rate and is thus currently undergoing new calculation and that the mortality rate is rather over-estimated. Therefore, ULM-stenting would have to prove its superiority over surgery for it to be considered as an alternative approach. In our study, subacute stent thrombosis occurred in three patients (3.6%) after elective ULM stenting. Other studies report this life-threatening complication as not normally exceeding 1–1.5% of all PCIs performed [15]. Moreover, all our study patients with subacute stent thrombosis had a distal bifurcational stenosis, which is a known risk factor for post-PCI complications [16]. Similar rates of thrombosis after distal ULM stenting were reported by other authors [17, 18]. A second known risk factor of stent thrombosis involves the actual technique of stent delivery in bifurcational stenosis. Two of the three patients in the current study were treated using the Crush-technique. The advantage of this technique is that it provides complete coverage of the bifurcational ostium, thereby preventing restenosis. Otherwise, the threefold stent layer associated with crush stenting might possibly promote stent thrombosis. Hoye et al. have, however, reported a 4.3% incidence of stent thrombosis with this technique [19]. Currently, it is still a matter of debate as to which of the techniques is superior [7]. Following publication of the data from the BASKET-LATE-study [20], the use of DES anywhere in the coronary system is under discussion. Although DES are generally used ‘off-label’ in the left main stem, they are preferable since BMS are associated with higher rates of restenosis [11, 12]. Thus, for ULM stenosis, bare metal stents should only be considered when drug-eluting stents are suboptimal due to the prolonged platelet inhibition associated with these stents. IVUS, as a diagnostic adjunct in coronary angiography, allows optimal stenting in ULM-PCI, thereby preventing both early and late-stage complications [21]. In the current study, an IVUS control has been performed, though infrequently and mostly after complex PCI. Since stent thrombosis occurred in patients who did not undergo an IVUS, we cannot exclude that this could have prevented this complication. On this basis, it is recommended that IVUS should be a routine procedure, especially after distal ULM stenting. The long-term outcome with a MACE rate of 26% is comparable to other ULM studies reporting complication rates of between 17% and 30% [11, 12, 18]. A direct comparison with CABG is currently not yet possible since the SYNTAX study (first multicentre randomised trial comparing CABG and PCI in multi-vessel CAD and ULM stenosis), is still in progress [22] and the EuroSCORE only evaluates short-term outcomes. To date, only a few studies have investigated the long-term outcome between ULM stenting and CABG in a non-randomised manner [23–25]. In these studies, PCI was correlated with a higher incidence of TVR of between 6% and 38% compared to CABG which showed TVR rates of 3–5%. These are similar to the present study where 12% TVR rates were achieved. In addition, in these studies similar rates of mortality and myocardial infarction between surgery and PCI were gained after a one-year follow-up. In contrast, we had an increased long-term mortality of 22% compared to 13% in these studies. On analysing the cause of death in the present study, we found a considerable proportion of non-cardiac deaths. The reason for this high proportion of non-cardiac deaths can be found in the pronounced high-risk profile in our patient population. In our study, 40% of all patients were older than 80 years. The EuroSCORE median of all patients in our study was 10 points compared to 6 points in the study by Palmerini et al. Moreover, we included 64% of high-risk patients in our study in contrast to the 32% (EuroSCORE >5) recruited in the study by Chieffo et al.
In our normal-risk patients, ULM-PCI did not prove superior to surgical revascularisation due to the high incidence of subacute stent thrombosis. This complication occurred independently of the surgical risk and was associated with high mortality. Here, patients who are eligible for coronary surgery, but prefer PCI should be informed about this complication in detail. In surgical high-risk patients, the positive short-term outcome is qualified by the suboptimal long-term outcome associated with high mortality. The advantage of CABG is its known low MACE rate during long-term follow-up, whereas data after ULM-PCI are still limited. As long as large randomised trials comparing PCI and CABG in ULM disease, like the SYNTAX study, are still pending, interventionists should adhere to current guidelines and perform ULM stenting only in the absence of surgical revascularisation options.
The authors would like to thank Mrs. Nasim Krögel for proof-reading the manuscript.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
