Influence of Age on Long-Term Outcome After Emergent Percutaneous Coronary Intervention for ST-Elevation Myocardial Infarction

273 - 277

Chin Hiew, FRACP*,
Trent Williams, BN*,
Rachael Hatton, FRACP*,
Seshasayee Narasimhan, FRACP*,
Simon O’Connor, FRACP*,
Fran Baker, PhD§,
Patrick McElduff, PhD§,
John Attia, PhD§,
Nicholas Collins, FRACP*

   ABSTRACT: Background. Percutaneous coronary intervention (PCI) provides advantages compared to thrombolytic therapy in the treatment of ST-elevation myocardial infarction (STEMI). Elderly patients have increased in-hospital mortality; the predictors of adverse outcomes are not well established, with limited data available regarding late follow up of these patients. We evaluated early and late outcomes of patients undergoing emergent PCI for STEMI to identify the predictors of subsequent late adverse events and, in particular, determine if age alone was a predictor of an adverse outcome. Materials and Methods. A retrospective review of all patients referred for emergent PCI for STEMI to a single tertiary referral center was performed. All patients undergoing primary PCI for STEMI or rescue PCI for failed thrombolysis between December 2003 and December 2007 were included for assessment. Results. During the study period, 269 patients underwent primary or rescue PCI for STEMI. Patients ≥ 70 years of age were more likely to have established cardiovascular risk factors and documented coronary artery disease. Thrombolysis in myocardial infarction (TIMI) 3 flow predicted a good outcome, whereas recurrent ischemia in hospital, prior aspirin therapy and discharge creatinine predicted a poor outcome; age alone was not an adverse prognostic factor. Conclusions. The increased mortality in elderly patients undergoing PCI for STEMI reflects comorbidity and suboptimal procedural out- comes rather than any age effect per se. Future approaches to optimize the management of STEMI in the elderly should include identification of those patients most likely to benefit from an invasive approach.

J INVASIVE CARDIOL 2010;22:273–277

   Primary percutaneous coronary intervention (PCI) remains the standard of care for patients presenting with ST-elevation myocardial infarction (STEMI) at centers capable for performing emergent PCI.1 PCI offers advantages in terms of recurrent ischemia and the risk of hemorrhagic stroke compared to thrombolytic therapy. There is also strong evidence supporting the use of PCI as rescue therapy in cases of failed thrombolytic therapy for STEMI,2 as well as in patients with cardiogenic shock complicating MI, including the elderly.3 However, elderly patients (> 70 years of age) are a unique population, previously underrepresented in clinical trials.4 While in-hospital mortality is noted to be higher in the elderly, the predictors of adverse outcomes are not well established. In addition, there are limited data regarding late follow up of these patients, particularly in the contemporary setting.

   We sought to evaluate the early and late outcomes of elderly patients undergoing emergent PCI for STEMI and to identify the predictors of subsequent adverse events in a single tertiary referral center with primary PCI facilities.


   The study was a retrospective review of all patients referred for emergent PCI for STEMI to a single tertiary referral center. All patients undergoing primary PCI for STEMI or rescue PCI for failed thrombolysis between December 2003 and December 2007 were included for assessment. Procedural data were obtained by the prospectively maintained primary PCI database at the John Hunter Hospital and based upon operator reports completed at the conclusion of each case; additional clinical data were retrieved from review of medical records as required. Femoral artery access was preferred by the operators performing primary PCI at the center, with bare-metal stents favored in accordance with institutional policy. Outcome data were obtained by review of admission codes collated through the Hunter New England Area Health Service Heart and Stroke Register, which prospectively records data on readmission and mortality from hospitals within the referral area. In those cases where data were incomplete, attempts to contact patients were made by telephone follow up.

   Clinical definitions. Patients aged ≥ 70 years of age were prospectively considered elderly for the purposes of the study.5 Procedural success was angiographically defined as a residual stenosis < 50% by visual estimation after intervention with thrombolysis in myocardial infarction (TIMI) flow grade ≥ 2. Recurrent MI was defined as recurrence of clinical symptoms with new elevation of troponin-I levels. Periprocedural complications were defined as follows: major bleeding complications were defined as a decrease in hemoglobin level > 4.0g/L, need for transfusion of > 2 units of blood or need for corrective vascular surgery for arterial injury; acute renal failure was defined as a deterioration in renal function with an absolute increase in the serum creatinine level of > 44 μmol/l or relative increase of serum creatinine by > 25% after the administration of an intravascular contract agent. Estimated glomerular filtration rate (eGFR) was calculated using the revised Modification of Diet in Renal Disease equation.6

   Statistical analysis. Statistical analysis was performed using STATA Version 10 (STATA, Inc., College Station, Texas). Comparison between patients less than or greater than 70 years of age was performed using the Student’s t-test for continuous, normally distributed variables, or the chi-square test for dichotomous variables. Mortality and recurrent MI during the follow-up period was graphed for the > 70-year-old group versus the < 70-year-old group using Kaplan-Meier curves; groups were compared using the Wilcoxon rank sum test. Long term follow-up mortality rates were adjusted for age. A
Cox-proportional hazards regression model was used to identify the determinants of mortality and recurrent MI; variables identified in the univariate analysis with p-values < 0.2 were included in the multivariate logistic regression model, and back- wards stepwise elimination was used to arrive at the most parsimonious model. A p-value < 0.05 was considered to be significant.


   Between December 2003 and December 2007, a total of 269 patients underwent primary or rescue PCI for STEMI at our institution; of these, 87 patients (32.3%) were ≥ 70 years of age.

   All surviving patients (251/269) were then followed up for a mean 1,108 days ± 437 (Range 357– 1,843) with complete clinical and procedural data available for analysis in 231 patients (85.8%).

   The elderly group was more likely to be female and hypertensive, with a background of dyslipidemia, previous MI and coronary artery bypass graft (CABG) surgery when compared to the younger age group. Younger patients were more often current smokers than their older counterparts (Table 1). Fur- thermore, the elderly group were more likely to be on medical therapy (aspirin, beta-blockade, angiotensin- converting enzyme inhibition and statin agents) prior to their index presentation. There were no significant differences in terms of in-hospital use of aspirin, clopidogrel, beta-blockade, angiotensin converting enzyme inhibitors and statin agents between the two groups (Table 2).

   In terms of angiographic and procedural characteristics, the infarct-related artery, stent usage, use of glycoprotein IIb/IIIa inhibition, intra-aortic balloon pump use and procedural success rates were similar between both groups, as was the frequency of post-procedural TIMI 3 flow (Table 3). The elderly were more likely to require post-procedure intensive care unit admission compared to younger patients (p = 0.1).

   In-hospital mortality was lower in the younger group compared to the elderly group (3.9% vs 6.9%, p = 0.28). There were 4 strokes during hospitalization, 2 in each group (1.1% vs 2.3%, p = 0.6). Of those patients with multivessel coronary disease, 5 underwent planned revascularization during the index admission with PCI; a single patient was referred for CABG. One additional patient in the younger group underwent urgent CABG for periprocedural complications. Major bleeding complications occurred in 5 patients in the younger group and 4 patients in the elderly group (2.3% vs 4.6%, p = 0.5). Acute renal failure after PCI occurred in 8 patients in the younger age group compared to 6 patients in the elderly group, one of whom required hemodialysis (4.4% vs 6.9%, p = 0.4).

   The composite outcome of recurrent MI and mortality within 30 days was higher in the elderly group (22.9% vs. 17.0%, p =0.24); most of this significance was due to the higher 30-day mortality in the elderly group (11.5% vs. 5.5%, p = 0.08). This trend continued over the long term with Kaplan Meier curves for mortality (Figure 1) differing significantly between the two groups.

   In order to determine whether the differences in outcomes between the groups were due mainly to age or to other con- founders, a Cox proportional hazards regression was performed testing the effect of other variables including medications on admission and on discharge, creatinine, hemoglobin, recurrent ischemia in hospital, TIMI 3 flow post PCI, risk factors, peak creatine kinase, intensive care unit admission, door-to-balloon time, past medical history, vessel(s) involved and body mass index. Stepwise forward and backward regression results are shown in Table 4. Achieving TIMI 3 flow predicted a good outcome, whereas recurrent MI and discharge creatinine were both predictors of a poor outcome; aspirin was a marker for increased MI and mortality, probably due to bias in prescribing to higher-risk patients. Even when age was forced into the par- simonious model, the p-value was 0.325, indicating that age per se was not a risk factor for poor outcome.


   In this contemporary cohort of patients, despite the use of high-dose clopidogrel loading, routine use of stents and utilization of glycoprotein IIb/IIIa inhibitors, the elderly remain at higher risk of both recurrent MI and mor- tality in both the short term (30 days) and long term (3 years) following presentation with STEMI. This reflects increased comorbidity and suboptimal pro- cedural outcomes, rather than age per se. Independent risk factors for MI and mortality included failure to achieve TIMI 3 flow, recurrent ischemia and high creatinine; age did not remain significant after adjustment for these factors.

Primary PCI remains the preferred strategy for reperfusion in patients with STEMI in whom PCI may be performed in a timely fashion by experienced operators.1 Similarly, there is strong evidence to support rescue PCI in cases of failed reperfusion after thrombolytic therapy.2 Primary PCI has been demonstrated to be superior to thrombolytic therapy in the treatment of the elderly with STEMI,7 with a previous study demonstrating a lack of survival benefit of thrombolytic therapy in those > 75 years of age.8 Furthermore, PCI offers a number of additional theoretical advantages over thrombolytic therapy in the elderly; increased age is a documented risk factor for hemorrhagic stroke after thrombolysis,9 with PCI clearly beneficial in terms of hemorrhagic stroke prevention. Advanced age is also recognized as a risk factor for cardiac rupture after thrombolytic therapy, which may reflect a reduced ability to achieve reperfusion (TIMI 3 flow) in these patients.10 As PCI is more likely to achieve TIMI 3 flow in STEMI compared to thrombolysis, it may therefore confer additional benefit in the elderly population.

Post new comment

  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

More information about formatting options

Enter the characters shown in the image.