Author + information
- Received June 4, 1999
- Revision received January 20, 2000
- Accepted March 29, 2000
- Published online August 1, 2000.
- Remi Choussat, MD∗,
- Alexander J.R Black, MB, BS†,
- Irene Bossi, MD∗,
- Thierry Joseph, MD∗,
- Jean Fajadet, MD∗,* ( and )
- Jean Marco, MD∗
- ↵*Reprint requests and correspondence: Dr. Jean Fajadet, Unité de Cardiologie Interventionelle, Clinique Pasteur, 45 avenue de Lombez, Toulouse 31076, France
This study was designed to evaluate the immediate and long-term clinical results of patients undergoing endoluminal reconstruction in diffusely degenerated saphenous vein grafts (SVGs) with elective implantation of one or more less-shortening Wallstents.
The optimal treatment strategy for patients with diffusely degenerated SVGs is controversial. Endoluminal reconstruction by stent implantation is one proposed strategy; however, there are few data regarding long-term clinical outcome.
Between May 1995 and September 1998, 6,534 consecutive patients underwent angioplasty in our institution, including 440 who were treated for SVG lesions. Of these, 126 (115 men, 11 women, median age 69.5 years, range: 33–86 years) with old SVGs (mean age: 13 ± 5 years) diffusely degenerated stenosed or occluded (mean lesion length: 27 ± 12 mm) were treated electively with implantation of one or multiple (total 197) less-shortening Wallstents.
Before discharge, 13 patients (10.3%) sustained at least one major cardiovascular event, including 4 deaths (3.2%), 11 myocardial infarctions (MI) (8.7%), and 3 repeat revascularizations (target vessel = 1, nontarget vessel = 2, 2.4%). Surviving patients were followed for 22 ± 11 months: 13 patients (11.1%) died, 11 (9.4%) sustained an MI, 37 underwent angioplasty (31.6%), and 4 (3.4%) underwent bypass surgery. The estimated three-year event-free survival rates (freedom from death, and freedom from death/MI/target vessel revascularization) were (mean ± SE) 81.1 ± 7.8% and 43.2 ± 18.5%, respectively.
The long-term clinical outcome of patients undergoing endoluminal reconstruction in diffusely degenerated SVG is relatively poor, mainly because of a high incidence of death or MI and the frequent need for repeat angioplasty. It is unlikely that percutaneous intervention alone will provide a satisfactory or definitive solution for these patients.
Coronary artery bypass graft (CABG) surgery effectively relieves angina in patients with obstructive coronary artery disease and may prolong life in selected patients (1). Recurrence of angina, however, occurs in 5% to 10% of patients each year, mainly because of graft failure or a combination of graft failure and progression of coronary atherosclerosis (2,3). Beyond one year and particularly >3 years after surgery, atherosclerosis of vein grafts becomes increasingly prevalent (4,5). Atherosclerosis may coexist with intimal fibromuscular proliferation and, in patients with acute ischemic syndromes, thrombus of varying ages. Serial angiographic studies have revealed that 15% to 30% of grafts are stenosed at one year after surgery and that nearly 50% are occluded at 10 years after surgery (4,6). The increasing number of patients who have undergone bypass surgery in which vein grafts were used has made vein bypass graft disease a frequent therapeutic challenge. Repeat CABG results in increased perioperative mortality (3%–7%) and myocardial infarction (MI) incidence (3%–11%), and less effective symptom relief compared with the initial operation (7,8). Balloon angioplasty has been proposed as a reasonable alternative but is associated with a higher rate of acute complications in this situation than is observed during interventions on native coronary arteries. Distal embolization, which is particularly frequent (4%–13%), is associated with a greater incidence of immediate and long-term adverse events (9–11). Moreover, mid- and long-term outcomes include high rates of restenosis (up to 46%) with major adverse events (24%–43% at one year and 66%–74% at five years) (9,12,13). Nonrandomized studies strongly support the use of intracoronary stenting for focal lesions in stenotic grafts (14–17), and there has been one randomized clinical trial that compared stent implantation with percutaneous transluminal coronary angioplasty (PTCA) of highly selected lesions that were short, focal and nonocclusive (18). In that study, the incidence of non-Q wave MI was only 2% in patients receiving stents, compared with 7% in the PTCA group. There was a trend in favor of stent placement for reduced angiographic restenosis. However, in the case of diffusely degenerated saphenous vein grafts (SVGs), the optimal treatment strategy, in particular the strategy of performing “endoluminal reconstruction” by stent implantation, has not been systematically studied.
The aim of this study was therefore to evaluate the immediate and long-term clinical results of patients with diffuse SVG lesions who underwent SVG reconstruction with elective implantation of one or multiple less-shortening Wallstents.
Between May 1995 and September 1998, out of 440 patients who underwent SVG angioplasty in our institution, 126 consecutive patients (29%) had diffusely degenerated stenosed (i.e., isolated lesions of >20 mm in length, or diffusely diseased segments with multiple/sequential stenoses) or occluded SVG. These patients underwent endoluminal reconstruction using less-shortening Wallstents. These patients constituted 1.9% of the 6,534 patients who underwent coronary angioplasty in our institution during the same period. These 126 patients were considered to be poor candidates for repeat surgery because of advanced age, unfavorable coronary vessel anatomy, poor left ventricular function or severe concomitant noncardiac disease and were electively treated by implantation of one or multiple Wallstents.
Angioplasty was performed via the transradial (61%), transfemoral (37%) or transbrachial (2%) approach, using 6F (17%), 7F (48%) or 8F (35%) guiding catheters. The lesion was carefully crossed using a 0.014 in. guidewire and predilated with an undersized balloon inflated generally to 3–4 atm for ≤30 s in an attempt to minimize the risk of distal embolization. In all of the study patients, Wallstents were used for SVG reconstruction. This stent is a self-expanding device whose deployment does not require high-pressure balloon inflation, and which is available in a large range of lengths (up to 50 mm) and sizes (up to 6 mm). Stents were generally chosen with a diameter 1.5 mm larger than the reference diameter of the graft and with a length 15 mm longer than the stenotic lesion. Intravascular ultrasound was not used during any implantation procedures. The arterial access sheaths were removed immediately after transradial procedures and immediately or 4 to 6 h after transfemoral procedure. Hemostasis was achieved by using the Perclose device or mechanical compression (Femostop device) for up to 2 h in patients treated via the transfemoral approach, and by radial compression with a tourniquet for 1 h in patients treated via the transradial approach.
After intracoronary injection of isosorbide dinitrate (3 mg) and lisindomine (1 mg), reference vessel diameter, minimum lumen diameter and lesion length were calculated using an online quantitative coronary analysis (DCI Philips system, The Netherlands). The following definitions were used for angiographic assessment: diffuse degenerated graft disease = isolated lesions of >20 mm in length, or diffusely diseased segments with multiple/sequential stenoses; thrombus = circumscribed intraluminal filling defect; angiographic success = smooth lumen surface at the stent site with a final diameter stenosis <10%; distal embolization = new appearance of filling defects and/or abrupt cutoff of the vessel distal to the target lesion and/or decreased antegrade flow in the distal vessel previously patent, in the absence of an occlusion of the target lesion; no reflow = poor antegrade flow (Thrombolysis in MI [TIMI] grade ≤1) not explained by dissection or high-grade residual stenosis at or adjacent to the target lesion; abrupt closure = poor antegrade flow (TIMI grade ≤1) due to acute occlusion of the target lesion.
Anticoagulant and antiplatelet regimen
All patients except those presenting with acute MI were pretreated for at least three days with aspirin (100–250 mg o.d.), ticlopidine (250 mg b.d.) and enoxaparin (100 UI/kg b.d). An intravenous bolus of heparin (70 UI/kg) was administered after arterial sheath placement and additional heparin boluses were repeated as necessary to maintain the activated clotting time between 250 and 300 s. Thirty-seven (29.4%) patients were treated with abciximab therapy, including a 0.25 mg/kg bolus initiated immediately before the procedure followed by a 0.125 μg/kg/min infusion for 12 h. After the procedure, all patients received 1 month ticlopidine (250 o.d. or b.d.) and long-term aspirin (100–250 mg o.d.). In the beginning of our experience, 14 patients with intracoronary thrombus were treated by enoxaparin for one to three days after the procedure, together with aspirin and ticlopidine. After the occurrence of a major intracranial hemorrhage in one patient treated after the procedure by enoxaparin, we decided to discontinue the use of anticoagulation after PTCA.
Data collection and follow-up
Clinical, angiographic and procedural data as well as postprocedural complications were prospectively entered into a computerized database (AS400, Showcase Strategy). Clinical follow-up was performed by an experienced physician who made telephone contact with the patients or their referring physicians. Information obtained included occurrence of recurrent angina, MI (defined by the development of new Q waves or increase in the serum cardiac enzymes to more than twice the upper limit of normal), subsequent cardiac catheterizations, need for repeat PTCA or additional CABG of target and nontarget vessel and death (cardiac and noncardiac). Follow-up data were entered into the computerized database. Initial and follow-up data of the studied patients were retrospectively analyzed.
Categorical variables were expressed as percent frequencies and continuous variables as mean value ± SD, or median (range and interquartile range). Chi-square analysis or the Fisher exact test were used for comparison between groups for categorical variables where appropriate, and the Student t test was used for normally distributed continuous variables. Predictors of cardiovascular events were determined by univariate analysis. A multivariate model of predictors of adverse outcome did not have sufficient power to discriminate the events, and the data are not presented. Out-of-hospital event-free survival analysis was performed using the Kaplan-Meier method, and four hierarchical event-free survival curves were generated: (1) freedom from death; (2) freedom from death and nonfatal MI; (3) freedom from death, nonfatal MI and repeat target vessel revascularization; (4) freedom from death, nonfatal MI and any repeat coronary vessel revascularization. A p value <0.05 was considered significant.
The baseline clinical characteristics of the 126 study patients are shown in Table 1. There was a high incidence of unstable angina (48.4%). A previous MI (>1 month) had occurred in 51 (40.4 %) patients. Recent MI (<30 days) occurred in 20 patients, of whom 10 were admitted in our institution within 72 h. Four of these with acute MI (<12 h) were treated by immediate PTCA. These patients were given aspirin and heparin on admission but did not receive our usual prolonged antiplatelet pretreatment. The remaining six patients underwent diagnostic catheterization and then received pre-treatment for at least three days with aspirin, ticlopidine and enoxaparin before undergoing PTCA. Previous PTCA had been performed in 54 patients (42.8%), of whom 31 (24.6%) had restenosis of the target vessel. The average vein graft age was 12.8 years (range 1–23.7 years), with 110 (87.3%) > 4 years old.
Single and multivessel coronary artery disease was present in 3.9% and 96.1% of the patients, respectively. The mean left ventricular ejection fraction was 55 ± 12% (range 20%–80%). The treated SVGs were anastamosed to the left anterior descending artery and/or diagonal branch (36.2%), left circumflex artery and/or marginal branch (39.8%) and right coronary artery (24%), respectively. The PTCA was performed for de novo lesion in 95 patients and for restenosis in 31 patients (Table 1). There were 138 treated SVGs (1.1 grafts/patient), with stenoses located proximally in 28.6%, in the graft body in 93.6% and distally in 15.9%. Mean lesion length was 27.1 ± 12.1 mm (range 8–80 mm) and preprocedural percentage stenosis was 79.2 ± 10.4%. Angiographic evidence of significant thrombus at the target site was found in 48 patients (38%). Successful implantation of a total of 197 Wallstents was achieved, representing 1.43 Wallstents per graft and 1.56 Wallstents per patient, with a mean final diameter of 3.9 ± 0.5 mm. Length of the Wallstent per graft was 44.5 ± 16.7 mm (range 23–126 mm) (Table 2). Seventeen additional stents were implanted in the grafts in addition to the Wallstents, including five AVE stents, three Nir stents, three Wiktor stents, two Palmaz-Schatz stents, two Be-stents, one Enforcer stent, and one GFX stent. Total stented length per graft, including the 17 additional stents, was 46.4 ± 17.4 mm (range 23–126 mm). No-reflow was observed in one patient and abrupt closure in none. Distal embolizations occurred in nine patients (7.1%) of whom three had received abciximab during the procedure. Five of these cases occurred after the predilation with an undersized balloon inflation and four after Wallstent deployment. Among the nine patients with distal embolization, eight had ECG changes during the invasive procedure and four developed chest pain. Other coronary lesions were simultaneously treated in 38 patients (30.1%), located in a native coronary artery in 27 patients, in other SVG in 10 patients and in an internal mammary artery graft in 2 patients.
In-hospital complications are reported in Table 3. Four patients (3.1%) died, two from cardiac causes. The first developed cardiogenic shock following distal embolization. The second died from ventricular fibrillation complicating anterior MI three days after the procedure. Noncardiac deaths included two strokes, both confirmed by computed tomographic scanning: one intracranial hemorrhage occurring three days after the procedure and one subclavian dissection caused by a transradial guiding catheter complicated by carotid hematoma. One of these patients had received abciximab therapy during PTCA, and both were treated with enoxaparin after the procedure.
Nine other patients (7.1%), including three who had received abciximab therapy, developed MI that was frequently associated with distal embolization. Among these patients, one was electively treated with CABG three days after the procedure and one who developed recurrent angina 2 h after stent implantation underwent PTCA of the left main coronary and the marginal branch. As a result, the overall clinical success rate was 89.7%.
In addition, one patient had transient ischemic stroke without sequelae. Eight patients (6.3%) had vascular access site complications, of which two received blood transfusion and two required surgical repair. One patient who had received abciximab therapy during PTCA had major thrombocytopenia (800 platelets/mm3), necessitating blood transfusion 24 h after the procedure, and one patient developed gastrointestinal bleeding. One hundred twenty-two patients were discharged alive (96.8%). The median hospital stay for the total cohort was 5.0 days (range 2–16 days, interquartile range 4–8 days).
Clinical events after hospital discharge
Follow-up (mean duration 21.6 ± 11.3 months) was obtained in 117 (95.9%) of the 122 discharged alive patients. After the PTCA, 57 patients had no further angina and 60 patients developed angina during follow-up, of which 60% had at least one episode of unstable angina. Table 3 lists the occurrence of major events after hospital discharge. There were 13 deaths (11%), 9 of which were cardiac and 4 noncardiac. Univariate analysis indicated that increasing age of the patients (68.0 years ± 8.1 vs. 74.9 years ± 5.3, p = 0.001) and lower ejection fraction (56.9% ± 13.9% vs. 47.0% ± 17.9%, p = 0.02) were predictors of deaths during follow-up. Six patients (5.1%) sustained a nonfatal MI. Four patients (3.4%) underwent repeat bypass surgery. Two of these four patients had unstable angina and two had stable angina before surgery. The indication for repeat surgery was severe and diffuse restenosis in two cases at 2 and 42 months, and total occlusion of the target vessel associated with progressive coronary disease in nontarget vessels at 6 months follow-up. Diabetes mellitus, hypercholesterolemia, vein graft age, previous coronary angioplasty at the stent site, abciximab treatment during the procedure, lesion length, Wallstent length per graft, and final percent diameter stenosis were not predictive of the composite of death, MI and target vessel revascularization (Table 4).
Follow-up angiograms were performed in 58 patients (48%) at a mean follow-up duration of 8.8 ± 6.2 months (range 1–43 months). The indications for angiography were unstable angina in 27 patients, stable angina and/or positive exercise test in 26 and MI in 5. The follow-up angiograms showed 34 restenoses, of which 14 were totally occluded, giving a 58.6% “angiographic” restenosis rate in this group of symptomatic patients. Initial lesion length was found to be associated with further vessel occlusion in univariate analysis (33.0 ± 4.9 vs. 23.8 ± 1.8 mm: p = 0.03). Vein graft age, diabetes mellitus, previous coronary angioplasty at the stent site, reference vessel diameter, post-stent minimal luminal diameter and final percent diameter stenosis were not associated with restenosis. Furthermore, 27 patients had stenosis of a nontarget vessel: 17 in native coronary arteries and 10 in nontarget SVG.
Repeat PTCA was performed in 38 patients at a median interval of 7.6 months (range 1–32 months): 15 underwent repeat target vessel revascularization, 13 underwent nontarget vessel revascularization and 10 underwent both target and nontarget repeat PTCA. The indication to perform repeat target lesion angioplasty was restenosis in or adjacent to the stented graft segment in all cases, with total occlusion in three cases. Nontarget vessel revascularization was due to progressive coronary disease in nontarget vessels.
Survival and event-free survival
Survival at six months and at three years were 94.2 ± 2.2% and 81.1 ± 7.8%, respectively. Freedom from death, MI and target vessel revascularization at six months and at three years, including events ccurring during the initial hospitalization, was 79.1 ± 4.2% and 43.2 ± 18.5%, respectively (Fig. 1). At three years, only 36.2 ± 17.0% of the patients were free from death, MI, and any coronary revascularization.
Management of recurrent ischemia in patients who have diffuse SVG lesions remains challenging and controversial (19). This report documents a single center’s experience of 126 consecutive patients undergoing endoluminal reconstruction of diffusely degenerated stenosed or occluded SVG with less-shortening Wallstents. The most important findings of our study are that despite initially successful results, this practice was associated with a poor short-term and long-term clinical outcome. Increasing operator expertise and technical improvements in angioplasty equipment and adjunctive medical therapy such as aspirin, ticlopidine, heparin and abciximab have made it possible to perform endoluminal reconstruction of diffusely degenerated SVGs with a high technical success rate. Nevertheless, the long-term results remain disappointing, first because of frequent in-hospital complications and second because of graft restenosis and/or progression of coronary atherosclerosis. In all, 55% of patients in this study died, had an MI, or required repeat revascularization during the mean 21.6-months follow-up. Nevertheless, these results should be seen in the context that the majority of these patients were considered to be poor surgical candidates and at increased risk for a second or third bypass operation because of advanced age, unfavorable coronary vessel anatomy, poor left ventricular function or general status.
In our study, successful stent implantation was achieved in all cases, as in previous studies reporting the use of Wallstents for treating SVG disease (20–24). The frequency of in-hospital major events including death, MI, emergency CABG or PTCA was relatively high (see Table 3) compared with previous reports (0 to 11%) (15–18,20,23–28). Distal embolization, observed in nine patients (7.1%), occurred relatively frequently. As serum level measurements of creatine kinase with MB fraction were not systematically performed after the procedure, some non-Q wave MIs were undoubtedly missed, resulting in an underestimate of the cardiac complication rate. In a pooled analysis of 16 studies (9), SVG balloon angioplasty was associated with a high procedural success rate (about 90%) and a lower incidence of death (<1%), MI (<3%), distal embolization (<3%) and emergency CABG (<2%). However, these findings are primarily concerned with the treatment of focal lesions, with 70% to 80% of the patients having a discrete lesion <10 to 13 mm long. In addition, in our study, patients were generally older, with advanced graft age, and our results may well be influenced accordingly. It has been well established that procedural results depend in part on the “diffuseness” of SVG disease, the age of the graft, and presence of thrombus (9).
Implantation of Palmaz-Schatz stents has been shown to decrease the incidence of adverse events (20%–30% at one year) and restenosis (13%–37%) compared with balloon angioplasty (15,16). However, the majority of studies of stent implantation have primarily concerned the treatment of focal lesions. Moreover, the results of the sole randomized trial published are disappointing, with a 37% restenosis rate in the stent group (18).
Our treatment strategy was initially based on an angioscopic study that found evidence of thrombus in 71% of SVGs treated by angioplasty (29), and the proved beneficial effect of the use of ticlopidine and aspirin in reducing platelet aggregation and coagulation activation during angioplasty procedures (30). Furthermore, one-third of our patients received abciximab therapy according to the assumption that patients with complex or thrombus-associated morphologies might derive additional benefit from platelet IIb/IIIa inhibition. Nevertheless, in our study we found no reduction of major coronary events in patients treated by abciximab (Table 4). This result is consistent with data from Ellis et al. (31) concerning the absence of beneficial effect of abciximab compared with placebo for patients with degenerated SVG. One might postulate that the stimulus for thrombosis in this situation was more dependent on the coagulation cascade than on platelet activation and that pretreatment with enoxaparin, aspirin and ticlopidine may be more likely to be of benefit than acute administration of IIb/IIIa inhibitors. Moreover, we used Wallstents, whose self-expanding property allows avoidance of balloon inflation for implantation and whose elastic wire mesh design would theoretically aid the entrapment of macroscopic friable SVG material.
Major bleeding and vascular complications resulted in two deaths in our study. This may have resulted in part from the aggressive antiplatelet and anticoagulant regimen that was used. However, such complications were also frequent in the other series despite different antiplatelet and anticoagulant treatments. Bleeding/vascular complications have been observed in up to 33% of patients and have been frequently related to in-hospital mortality (15–18,20–26,28). De Scheerder et al. (23), who performed intragraft administration of thrombolytic agents during the procedure, observed bleeding complications in 33% of patients, including two fatal hemorrhagic strokes. On the other hand, there was no clinically apparent subacute stent thrombosis in our study, contrasting with an incidence of 1% to 10% reported in the literature (15–18,20–26,28).
During the mean follow-up period of 21.6 ± 11.3 months, one-third of our patients experienced major events (death or MI) and more than one-third underwent repeat revascularization. These results are quite similar to those previously reported after SVG stent implantation, even though our series included only patients with diffuse lesions. One-year event-free survival has been reported at from 46% to 80% (15–17,20,21,26–28). Importantly, the most disappointing results were observed in the series of De Jaegere et al. (20), in which 52% of stent implantations (mainly Wallstents) was performed for long SVG lesions. In that study, 23% of patients developed MI and 50% underwent repeat revascularization at a median of six months after the procedure. Adverse outcome after SVG stenting has been thought to be partly related to restenosis. Angiographic follow-up revealed a restenosis rate of 17% to 37% in the series utilizing Palmaz-Schatz stents (15–18,25,26,28) and 20% to 54% in those using Wallstents (20–24). In our study, only symptomatic patients underwent repeat angiography, and the 59% angiographic restenosis rate was therefore likely overestimated.
Our choice to perform elective SVG “reconstruction” with long stents in order to cover segments of the grafts that were not severely narrowed was based on studies that reported a 27% to 34% rate of revascularization procedures in relation to SVG disease progression (23,25). Indeed, Ellis et al. (32) have found that 45% of recurrent ischemic events after SVG interventions result from initially untreated 40%–50% stenosed sites, as compared with the 19% of events related to initially treated 40%–50% stenosed segments. However, our strategy to perform SVG reconstruction with long stents might well have been partly responsible for the high incidence of restenosis we observed.
The present study clearly demonstrates that reconstruction of diffusely degenerated SVGs by implantation of self-expandable less-shortening Wallstents with preventive antiplatelet and anticoagulant treatment can be achieved with high deployment and procedural success, but was associated with a significant incidence of early complications. Long-term clinical outcome was marked by a relatively high cardiovascular morbidity and mortality. However, the clinical status of our population, at high risk for repeat CABG, should be taken into account when interpreting these results. Further studies, including the use of new antiplatelet and anticoagulant therapies and new device technology (33), will be needed to clarify the optimal strategies for treating these difficult patients.
- coronary artery bypass graft
- myocardial infarction
- percutaneous transluminal coronary angioplasty
- saphenous vein graft
- Thrombolysis in Myocardial Infarction
- Received June 4, 1999.
- Revision received January 20, 2000.
- Accepted March 29, 2000.
- American College of Cardiology
- CASS principal investigators and their associates
- Campeau L.,
- Lesperance J.,
- Hermann J.,
- et al.
- Guthaner M.G.,
- Robert E.W.,
- Alderman E.L.,
- Wexler L.
- Weintraub W.S.,
- Jones E.L.,
- Morris D.C.,
- et al.
- De Feyter P.J.,
- Van Suylen R.J.,
- De Jaegere P.P.,
- et al.
- Dooris M.,
- Hoffmann M.,
- Glazier S.,
- et al.
- CAVEAT-II Investigators,
- Lefkovits J.,
- Holmes D.R.,
- Califf R.M.,
- et al.
- Douglas J.S.,
- Gruentzig A.R.,
- King S.B. III.,
- et al.
- Ozaki Y.,
- Violaris A.G.,
- Hamburger J.,
- et al.
- Palmaz-Schatz Stent Study Group,
- Wong S.C.,
- Baim D.S.,
- Schatz R.A.,
- et al.
- Piana R.N.,
- Moscucci M.,
- Cohen D.J.,
- et al.
- Holmes D.R.,
- Berger P.B.
- De Jaegere P.P.,
- Van Domburg R.T.,
- De Feyter P.J.,
- et al.
- White C.J.,
- Ramee S.R.,
- Collins T.J.,
- et al.
- Gregorini L.,
- Marco J.,
- Fajadet J.,
- et al.
- Ellis S.G.,
- Lincoff A.M.,
- Miller D.,
- et al.
- Bartorelli A.,
- Pieper M.,
- Kyriakides Z.,
- et al.