The very elderly (those ≥ 85 years of age) are the most rapidly growing population in Western countries. The mortality rate of this group increases markedly following ST-segment elevation myocardial infarction (STEMI) [1,2]. However, limited data are available regarding the outcomes of primary percutaneous coronary intervention (PCI) in this age group, especially in nonagenarians, who are generally excluded from clinical trials [3]. In the present study, we evaluated the temporal trends and in-hospital outcomes of primary PCI in nonagenarian STEMI patients using the Korea Acute Myocardial Infarction Registry (KAMIR) and Korea Working Group on Myocardial Infarction (KorMI) databases.

From November 2005 to May 2010, 84 patients aged ≥ 90 years who presented within 12 hours of symptom onset and underwent primary PCI for STEMI were enrolled in this study (Fig. 1). Although the total number of participants was small, the proportion of nonagenarians among STEMI patients more than doubled during the study period (0.59% in KAMIR vs. 1.35% in KorMI). Therefore, the growth rate was the highest in this age group (Fig. 2A). According to this trend, the rate of use of primary PCI in nonagenarians also increased to the level of the average rate of use for each different age group (from 62.5% in KAMIR to 81.0% in KorMI) (Fig. 2B).

Baseline characteristics are presented in Table 1. The mean age was 92.3 ± 3.4 years and 63.1% of patients were women. Hypertension, diabetes mellitus, hyperlipidemia, and current smoking were present in 51.2%, 14.3%, 2.4%, and 8.3% of patients, respectively. Only a few patients suffered from coronary artery disease and 3.6% had received PCI. Comorbidity index score was generally low and 69.0% of patients had no comorbid conditions. On admission, 28.4% patients were in Killip class ≥ 3 and 17.9% patients were in cardiogenic shock. Mean pre-hospital delay time was 232 ± 132 minutes and 84.5% patients arrived at the hospital within 6 hours of symptom onset. Mean left ventricular ejection fraction was 50.8% ± 12.8%. Overall, the baseline characteristics of patients did not change over time.

Angiographic findings and procedural results are presented in Table 2. The mean time from symptom onset to PCI was 311 ± 162 minutes, and was shortened compared to the KAMIR data (377 ± 146 minutes in KAMIR vs. 290 ± 161 minutes in KorMI, p = 0.030). The mean door-to-balloon time was 100 ± 79 minutes, and was also shortened compared to the KAMIR data (140 ± 131 minutes in KAMIR vs. 87 ± 45 minutes in KorMI, p = 0.087). Multi-vessel disease was present in 73.8% of patients. The left anterior descending artery and right coronary artery accounted for the majority of the infarct-related arteries, which were observed in 51.2% and 41.6% of patients, respectively. Coronary stents were implanted in 90.5% of patients, 67.9% of which were drug-eluting stents. A final TIMI grade 3 flow was obtained in 84.5% of patients (75.0% in KAMIR vs. 87.5% in KorMI, p = 0.320). The overall rate of use of glycoprotein IIb/IIIa receptor inhibitors was low; it was administered to only 7.1% of patients.

In-hospital outcomes are presented in Table 3. Cardiogenic shock was present in 25% of patients. Incidences of recurrent myocardial infarction, stroke and acute renal failure were low, and occurred in 1.2%, 1.2%, and 8.3% of patients, respectively. Remarkably, there were no cases of major bleeding during hospitalization.

The overall in-hospital mortality rate was 21.4%, which is four times higher than that of STEMI patients who underwent primary PCI within 12 hours of symptom onset (5.04%). The in-hospital mortality rate did not improve significantly during the study period (25.0% in KAMIR vs. 20.0% in KorMI, p = 0.919), and this trend was the same for other age groups (Fig. 3).

Stepwise logistic regression analysis identified two independent predictors of in-hospital mortality, namely a final TIMI flow grade < 3 (odds ratio [OR], 13.7; 95% confidence interval [CI], 3.2 to 59.0; p < 0.001) and cardiogenic shock during hospitalization (OR, 6.7; 95% CI, 1.5 to 30.3; p = 0.013) (Table 4).

Acute myocardial infarction (AMI) is the leading cause of death in elderly patients [6], and age is an important predictor of death after myocardial infarction. Thus, very elderly patients such as nonagenarians represent a high-risk population. Although nonagenarians constitute a small proportion of AMI patients, more nonagenarian STEMI patients may be encountered in daily practice in the near future as the population ages. In our study, the total number of STEMI patients remained unchanged, but the proportion of nonagenarian STEMI patients more than doubled during the study period, and their growth rate increased most rapidly compared to other age groups. This increase contributes to the reason why evidence-based data are necessary to guide proper management in this population.

There is evidence supporting the use of reperfusion therapy in elderly patients up to the age of 85 years [7]. However, in the current guidelines for the treatment of STEMI patients [8,9], there is no age limit for reperfusion therapy, and the recent TRIANA (TRatamiento del Infarto Agudo de miocardio eN Ancianos) trial proposed primary PCI as the optimal reperfusion strategy for old patients [10]. Thus, there is a general acceptance in daily practice for nonagenarian STEMI patients to undergo primary PCI. Several studies have reported the outcomes of primary PCI in nonagenarian STEMI patients [11-14]; these showed that achievement of a TIMI grade 3 flow rate after the procedure ranges from 74.1% to 82%, and in-hospital mortality rate ranges from 18.5% to 34.2%. As a result, the authors concluded that primary PCI can be performed with a high success rate and an acceptable in-hospital mortality rate. Our results also support these previous studies, as achievement of a TIMI grade 3 flow rate after the procedure was 84.5%, while in-hospital mortality was 21.4%.

In the present study, we enrolled nonagenarian STEMI patients who arrived at the hospital within 12 hours of symptom onset. The overall rate of use of primary PCI in these patients was 75.7%, and this rate eventually reached the average of other age groups (from 62.5% in KAMIR to 81.0% in KorMI). A previous single-center study in nonagenarian STEMI patients reported a primary PCI rate of use of 54% and an associated in-hospital mortality rate of 23% [15]. Considering results from the GRACE registry, in which 30% of STEMI patients who presented within 12 hours of symptom onset did not receive reperfusion therapy [16], the primary PCI rate of use was considered high in our study. This result is due to the fact that the included patients were relatively healthy nonagenarians and many patients presented with a short pre-hospital delay. Thus, the responsible physicians may have attempted to perform primary PCI actively. Contrary to the previous study in which PCI use rate increased over time and resulted in reduced in-hospital mortality rate after STEMI in nonagenarians [17], our study did not result in a reduction in mortality despite the increased primary PCI use rate over time. This could be due to three reasons. The first is the different inclusion criteria used for patient selection. Secondly, as mentioned earlier, primary PCI use rate was sufficiently high in our study to determine an acceptable in-hospital mortality rate. Lastly, the use rate of glycoprotein IIb/IIIa receptor inhibitors, which have favorable effects on mortality in this population, was low [11].

In our study, a final TIMI grade < 3 and cardiogenic shock were identified as independent predictors of in-hospital mortality. Final TIMI flow grade is an important determinant of in-hospital and long-term adverse events following primary PCI, but elderly patients are less likely to achieve a final TIMI grade of 3 after PCI [18]. Although high bleeding risk and complex coronary anatomy are common in elderly patients, judicious use of glycoprotein IIb/IIIa receptor inhibitors and thrombosuction devices can represent meaningful approaches to improve myocardial perfusion in elderly STEMI patients. Cardiogenic shock was present in 17.9% of patients at the time of admission and eventually developed in 7.1% of patients after admission, for a total of 25% of patients. Similar to other studies [11,19,20], our in-hospital mortality rate was very high in patients with cardiogenic shock (42.9%), indicating that the presence of cardiogenic shock in nonagenarian STEMI patients requires more careful consideration when determining treatment strategy. In one notable study [19], the in-hospital mortality rate was 90% in patients with cardiogenic shock, and the authors concluded that primary PCI is unable to affect the poor prognosis of very old patients with cardiogenic shock.

We are aware of several limitations in the present study. This is a retrospective multicenter analysis and is subject to the inherent limitations of such studies. Although two registries were combined, the number of enrolled patients was small, as nonagenarian STEMI patients account for only 0.9% of total STEMI patients. Therefore, it can be assumed that many nonagenarian STEMI patients were not referred to primary PCI. The total mortality rate was too small to perform multivariate analysis, which should be taken into account when interpreting our results. Lastly, we were unable to assess long-term mortality rate and quality of life.