Drug-Eluting Stents in the Elderly: Long-Term (> One Year) Clinical Outcomes of Octogenarians in the DESIRE (Drug- Eluting Stent

ABSTRACT: Background. Despite the increasing number of elderly people, this high-complexity subset of patients is often excluded from randomized trials of percutaneous coronary intervention (PCI) and, therefore, limited data are available about their outcomes after drug-eluting stent (DES) implantation. We sought to compare the very long-term (> 1 year) clinical follow up of octogenarians treated with DES compared to younger individuals. Methods. The DESIRE registry is a prospective, nonrandomized, single-center registry with consecutive patients treated solely with DES between May 2002 and May 2007. The only exclusion criteria were the treatment of patients in the setting of acute myocardial infarction (MI) (< 72 hours) and lesions located in non-native coronary arteries. The primary endpoint was the occurrence of combined major adverse cardiac events (MACE) (cardiac death, non-fatal MI and target vessel revascularization) in-hospital and in very long term (> 1 year) follow up. Patients were clinically evaluated at 1, 3 and 6 months and then annually up to 5 years. Stent thrombosis was classified according to the ARC definition. Results. A total of 1,364 patients matched the inclusion/exclusion criteria and were sorted into 3 groups according to their ages: Group I = patients < 70 years of age (n = 914); Group II = patients ≥ 70 and < 80 years of age (n = 334); and Group III = patients ≥ 80 years old (n = 116). As expected, octogenarians had significantly more comorbid and complex anatomic lesion presentation. Nevertheless, in-hospital success was comparable among the 3 groups. Long-term (2.6 ± 1.2 years) follow up was obtained from 97% of the total population and showed equivalent cumulative MACE in all age ranges (7.6% for < 70 years old vs. 5.4% for septuagenarians and 6.0% for octogenarians, p = 0.7). However, octogenarians had markedly more cardiac death, reflecting the severity of their comorbidity and the natural evolution of coronary disease. Of note, very few cases of stent thrombosis were noticed in the overall population (20 patients, 1.5%), with no difference among the groups. Conclusions. PCI with DES represents an efficient and safe approach to treat coronary artery disease in the elderly, with low rates of combined MACE comparable to other age ranges.

J INVASIVE CARDIOL 2008;20:404–410

Despite the rapidly-increasing elderly population and the fact that coronary artery disease (CAD) is the most prevalent cause of death among them, this subset of patients is frequently excluded from major revascularization trials, thus a paucity of information is available about their outcomes, in particular long-term results after either coronary artery bypass graft surgery (CABG) or percutaneous coronary intervention (PCI).1–3

It is well recognized that compared to the general population, the elderly are more likely to present with more complex lesions, unstable angina and more severe comorbid conditions.4–8
The recent introduction of drug-eluting stents (DES) for the treatment of CAD has proved to considerably reduce the rates of restenosis and, subsequently, the need for repeat revascularization.9–11 This benefit has been extended to complex subsets of patients and lesions such as diabetics, those with small vessels, long lesions, etc.12–14 At the moment, little is known about the long-term safety and efficacy of these new devices for the treatment of elderly patients, especially those > 80 years of age.

We sought to evaluate the long-term clinical results of PCI with DES in octogenarians compared to septuagenarians and patients < 70 years of age.

Methods

Patient population and study design. Since May 2002 when the first DES was clinically and market-approved in our country, we began enrolling patients in the prospective, single-center, nonrandomized DESIRE (Drug-Eluting Stent In the REal world) registry.

To be included in this registry, patients had to present either with angina or objective evidence of ischemia in the territory supplied by the vessel to be treated and lesions with angiographic evidence of ≥ 50% diameter stenosis. Additionally, they had to be treated with at least 1 DES. Patients receiving both DES and bare-metal stents (BMS) were excluded from this analysis. The study was designed to assess long-term outcomes, thus patients with < 1-year  follow up were excluded (n = 369). However, if a patient had a major adverse cardiac event (MACE), including death or stent thrombosis in the first year of follow up that required < 1 year of clinical follow up, that patient was included in the analysis. Furthermore, patients with ST-elevation myocardial infarction (STEMI) and lesions located in saphenous vein grafts were allocated to specific databases and therefore excluded from the general registry.

Antithrombotic regimen. Patients were premedicated with aspirin (200 mg), which was recommended to be continued lifelong. Additionally, they received clopidogrel (loading dose of 300–600 mg) initiated 24 hours before the intervention or ticlopidine (250 mg twice daily) administered 72 hours prior to the percutaneous procedure. Until December 2005, our hospital recommended that patients stay on thienopyridines for a minimum of 3 and 6 months after Cypher™ (Cordis Corp., Miami Lakes, Florida) and Taxus® (Boston Scientific Corp., Natick, Massachusetts) deployment, respectively. Starting in 2006, our institutional board decided to prolong the dual antiplatelet regimen for a minimum of 1 year for both DES following international recommendations. During the procedure, heparin was administered as a bolus dose of 100 IU/kg, with an additional bolus to maintain an activated clotting time > 250 seconds. The use of other adjunctive medications, including glycoprotein IIb/IIIa inhibitors, was left to the operator’s discretion.

Intervention and follow up. Intracoronary stenting was performed using standard interventional techniques. The stent was to be deployed from a “normal-to-normal” reference segment and when more than 1 stent was necessary to treat the same lesion, overlapping was strongly recommended. An in-stent residual lesion of < 10% was always our goal. Two different DES were used: sirolimus-eluting (Cypher) and paclitaxel-eluting (Taxus) stents. These stents were available in lengths ranging from 8 mm to 33 mm and diameters from 2.25 mm to 3.5 mm for Cypher stents, and lengths from 8 mm to 32 mm and diameters from 2.25 mm to 3.5 mm for Taxus stents. The type of DES to be implanted was left to the operator’s discretion.

Electrocardiograms (ECG) and creatine-kinase (CK) and CK-MB values were routinely obtained preintervention and repeated 8–12 hours after the procedure.

Clinical follow up was conducted by office appointment or, when deemed appropriate, by phone call at 1, 6 and 12 months and then annually up to 5 years following the baseline procedure.
Angiographic follow up was not part of this study protocol, and whenever performed, was ischemia-driven.
Quantitative coronary analysis. Intracoronary nitroglycerin (0.1–0.2 mg) was given prior to and after each intervention to achieve maximal vasodilatation. Quantitative coronary analysis (QCA) was performed immediately before and after stent placement by an experienced technician. The analysis of the stented segment comprised the stent itself and the 5 mm proximal and distal to it. Angiographic measurements included proximal, distal and interpolated references, minimum lumen diameter (MLD), percentage of lesion stenosis and lesion length. Also, acute gain was measured and defined as the difference between the MLD after stent deployment and the baseline MLD.

Definitions and endpoints. The study’s primary endpoint was the occurrence of MACE in-hospital and long term (minimum 12 months). MACE was defined as cardiac death, non-fatal MI and target vessel revascularization (TVR). TVR was only based on the presence of symptoms and/or signs of ischemia. All deaths were considered to be cardiac-related unless a non-cardiac origin could be clearly established by clinical and/or pathological study. The diagnosis of MI was based on either the development of new pathological Q-waves in ≥ 2 contiguous electrocardiographic leads and/or an elevation of CK-MB isoenzyme > 3 times the upper limit of normal post procedure during the index hospitalization, or cardiac enzyme elevation > 2 times the upper limit of normal thereafter. The secondary endpoint was the incidence of stent thrombosis in the total population. Stent thrombosis was classified as definite, probable and possible according to definitions proposed by the Academic Research Consortium (ARC), and was stratified as acute (< 24 hours), subacute (24 hours to 30 days) and late (1 to 12 months).15

Left ventricular function was accessed by the ejection fraction (%EF), and patients were classified as: a) normal (%EF ≥ 55%); b) mild dysfunction (%EF ≥ 40% and < 55%); c) moderate dysfunction (%EF ≥ 30% and < 40%), and; d) severe dysfunction (%EF < 30%).

Chronic renal failure was defined by the presence of creatine ≥ 1.4 mg/dl or a calculated glomerular filtration rate of < 60 ml/minute/1.73 m2.

Statistical analyses. All analyses were performed on the intention-to-treat basis. Continuous variables are expressed as mean and standard deviation. Categorical data are presented as counts and percentages and compared using Fisher’s exact test. When three groups were compared, overall probability values were derived from one-way ANOVA. The cumulative incidence of adverse events was estimated according to the Kaplan-Meier method. Differences between the event-free survival curves for the three groups were compared using the log-rank test. All statistical tests were performed at a 0.05 level of significance. Statistical analyses were performed using a commercially available software (SPSS version 9.0, SPSS, Inc. Chicago, Illinois).

Results

Between May 2002 and May 2007, a total of 2,081 consecutive patients were included in our database. Among them, 348 individuals presented with the culprit lesion in a saphenous vein graft or were treated in the setting of acute MI and were excluded from the current investigation. Another 369 patients were later excluded because they did not complete a minimum of 1 year of clinical follow up. For the purpose of this analysis, the remaining 1,364 patients were divided into three groups according to their ages: Group I = patients < 70 years of age (n = 914); Group II = patients ≥ 70 and < 80 years of age (n = 334); and Group III = patients ≥ 80 years of age (n = 116).

The baseline clinical and procedural characteristics are indicated in Table 1. As expected, the proportion of women with CAD significantly increased over the decades, with greater prevalence among Group III patients (38.2% vs. 33.8% for Group II, and 18.5% for Group I; p = < 0.001). Also, octogenarians had a significantly greater history of stroke (3.5% vs. 2.9% in Group II, and 1.3% in Group I; p = 0.02) and moderate-to-severe left ventricular dysfunction (40% vs. 16.7% in Group II, and 22.2% in Group I; p < 0.001). Moreover, acute coronary syndrome as the initial clinical presentation was more prevalent among octogenarians (38.7% vs. 28.7% for Group II, and 28.6% for Group I; p = 0.02).

Notably, the occurrence of more complex lesions (Types B2 and C) was higher in Group III (74.4% vs. 64.4%, and 64.3% for Groups II and I, respectively; p = 0.02).

Overall, the Cypher was the predominant DES used in all three groups (84.6%), with similar distribution among the three groups.

In-hospital MACE rates were similar for the three groups (Table 3). Of note, very few complications related to puncture site (bleeding/pseudoaneurysm formation) were observed, and they were equally distributed among the groups (0.5% in patients < 70 years of age vs. 0.7% in patients ≥ 70 and < 80 years of age, and 0.65% in patients > 80 years of age; p = 0.7).

Clinical follow up was obtained from 97% of the patients (< 70 years of age = 97.2%; ≥ 70 < 80 years of age = 96.7%; and > 80 years of age = 97.1%; p = 0.8). The mean follow-up period was 2.6 ± 1.2 years. Remarkably, at the end of 5 years more than 90% of patients in all three groups were completely free of MACE. The primary endpoint of this study, cumulative long-term MACE, was comparable among the three groups. However, when the events were analyzed separately, patients > 80 years of age had significantly higher rates of cardiac (and non-cardiac) death than the remaining subjects (Table 3 and Figure 1). It is important to note that two-thirds of the cardiac deaths among the octogenarians were a consequence of progressive heart failure in the follow-up period and therefore not directly related to PCI. Except for death, the other major adverse events were equivalent among the three groups.

The overall DES thrombosis rate was very low, with only 20 (1.5%) cases detected which were equally distributed among the three groups (Table 4).

Discussion

The current study highlights the excellent immediate and long-term clinical results of PCI with DES in octogenarians. This therapeutic approach proved to be a safe and efficient alternative treatment to this complex group of patients.

Due to their clinical and anatomical complexity, very elderly patients, particularly those > 80 years of age have traditionally being denied invasive treatment for CAD. In 2003 Pfisterer et al published the results of the TIME trial comparing, in a random fashion, invasive strategy (PCI or CABG) to optimized medical therapy in 282 patients ≥ 75 years of age. Despite the equivalence in symptoms, quality of life and death or non-fatal infarction, patients allocated to medical treatment carried a 50% greater chance of later hospitalization and revascularization in the 1-year follow-up period.16
A potential criticism to the TIME trial was to consider CABG and PCI together as an invasive therapeutic approach. Numerous studies have demonstrated that CABG for elderly patients carries a higher risk of immediate adverse events. Alexander et al, analyzing 67,764 patients (4,743 octogenarians) from the National Cardiovascular Network, demonstrated that patients > 80 years of age who underwent CABG had significantly higher in-hospital mortality than younger patients (8.1% vs. 3.0%; p < 0.05). In addition, they had twice the rate of postoperative stroke and renal failure (3.9% and 6.9%, respectively vs. 1.8% and 2.9%; p < 0.005 for both comparisons).17 To corroborate with the National Cardiovascular Network findings, an analysis conducted by Peterson et al using the data on 24,461 patients aged 80 years or older (Medicare database), showed an in-hospital mortality rate of 11.5% after CABG alone.18 Overall, these studies had an event rate 10 times higher than those described in our study for the same age ranges (in-hospital mortality rate of 0.8% for octogenarians).
Prior to the introduction of stents, balloon angioplasty also presented a very high risk of acute adverse events, especially when performed in complex patients such as the elderly. Several reports from the 1980s described procedural mortality rates of 1.4% to 10% and an emergency CABG rate ranging from 0.7% to 7.0% in that particular subgroup of patients.7,8 However, PCI data from the beginning of the 1990s had already demonstrated a changing trend toward lower procedural complication rates following the advent of high-pressure balloons, intravascular ultrasound guidance and better dual antiplatelet regimens as a substitution for anticoagulation.19

The introduction of stents to the percutaneous interventional arsenal dramatically changed the immediate- and long-term results of this revascularization procedure. However, studies from that period still showed poorer outcomes following PCI in the elderly. Abizaid et al, reporting their preliminary results with Palmaz-Schatz™ stents for the treatment of octogenarians, showed an increased rate of acute complications, in-hospital mortality and late mortality among those patients when compared to individuals < 80 years of age.5 De Gregorio et al, analyzing data on 2,688 patients (137 of them ≥ 75 years of age) treated between 1993 and 1997 at Centro Cuore Columbus, also showed higher rates of procedure-related complications among the elderly, including emergency CABG (3.7% elderly vs. 1.4% for those < 75 years of age; p = 0.04) and death (2.2% vs. 0.12%; p < 0.0001). At the end of 1-year follow up, 91% of the patients ≥ 75 years of age were alive, however only 54% were event-free, mainly due to the need for repeat revascularization.20 Also, Cohen et al, reporting the findings of the National Heart, Lung and Blood Institute’s (NHLBI) Dynamic Registry, demonstrated an adjusted relative risk of in-hospital death and MI that increased by age group. One-year mortality rates among the elderly were higher than in younger groups, but similar to what was expected among persons of similar ages in the general population.21

Two contemporary real-world series of elderly patients treated with DES corroborated with our findings. First, Vijayakumar et al, analyzing the data of 46 octogenarians enrolled in the RESEARCH registry, showed repeat revascularization and mortality rates of 9.0% and 4.0%, respectively, at 1-year follow up in this particular subset of patients.22 Also, Hassani et al, reporting the Washington Hospital Center experience, demonstrated similar in-hospital complication rates among octogenarians and patients < 80 years of age. At 6 months, restenosis rates were equivalent between the groups.23 The major limitation of these two studies is the lack of long-term (> 1 year) data. The present analysis not only confirms the excellent immediate- and middle-term results of DES in the elderly, but also demonstrates the long-term safety and efficacy of these devices.

Although age has not been reported as an independent predictor of stent thrombosis, many clinical and angiographic findings common to the elderly (e.g., heart and kidney failure, acute coronary syndromes, long calcified lesions, etc.) have been associated with a higher incidence of these serious adverse events.24 Even though the septuagenarians and octogenarians in our series presented with a significantly greater number of the above-mentioned predictors of thrombosis, they were not associated with an increase in acute, subacute or late stent thrombosis rates when compared to patients < 70 years of age. It is worth mentioning that the majority of our patient population (72%) was on the dual antiplatelet therapy (AAS + thienopyridine) for 6 months or less.

Study limitations. This was not a randomized study. Our registry represents a single-center experience. Sirolimus-eluting stents were the predominant DES used in this series of patients, preventing any comparison between the two types of DES used in this registry. The exclusion of patients treated in the setting of acute MI and with saphenous vein graft lesions may have played a role in the low rate of MACE observed in all three groups of this study. In addition, the lack of angiographic follow up, although representing a real-world scenario, also precludes our ability to assess late lumen loss, loss index and binary restenosis rates.

Conclusions

PCI with drug-eluting stent implantation represents an efficient and safe approach to the treatment of CAD in elderly patients, with low rates of combined MACE that are comparable to other age groups. Octogenarians have a greater mortality rate in their long-term clinical evolution, probably as a direct consequence of their more severe comorbid conditions.