After 80 study

1

Invasive versus conservative strategy in patients aged 80 years or older with non-ST elevation myocardial infarction or unstable angina pectoris (After Eighty study): an open-label randomised controlled trial

Tegn. Lancet;2016;10023:1057-1065

Clinical Question

  • In patients aged 80 years or older, with unstable angina/non-ST elevation myocardial infaction (NSTEMI), does an early invasive strategy compared with a conservative strategy, reduce the composite outcome of myocardial infarction, need for urgent revascularisation, stroke and death?

Design

  • Randomised controlled trial
  • Permuted block randomisation
  • Stratified by hospital
  • Use of sealed opaque envelopes
  • Non-blinded
  • Consecutive patients
  • Intention to treat analysis
  • Median follow-up of 1.53 years
  • Power calculation: 412 patients required to give a 80% error to detect a 10% absolute risk reduction from a baseline risk of 21%, with a false positive rate of 5%

Setting

  • 16 academic and teaching hospital without percutaneous coronary intervention facilities in Norway
  • December 2010 – February 2014

Population

  • Inclusion criteria: aged 80 years or over, with NSTEMI/unstable angina
    • With or without ST depression
    • With normal or raised Troponin T/I
    • Assessed for inclusion within 2 days of hospital admission
  • Exclusion criteria:
    • Unstable with continuing chest pain, ischaemic symptoms/signs, cardiogenic shock, bleeding problems, life expectancy <12 months due to serious co-morbidity; substantial mental disorder, including severe dementia
  • 457 randomised out of 4187 patients assessed for inclusion:
    • 2214 met exclusion criteria
    • 1516 candidates who met inclusion not included:
      • 402 refused to participate
      • 1011 logistics

Intervention

  • Early invasive strategy
    • Patient transported to Oslo University Hospital 1 day after inclusion
    • Early coronary angiography with immediate assessment for percutaneous coronary intervention (PCI), coronary artery bypass graft of optimum medical treatment
    • Patients returned to their local community hospitals after:
      • 6-18 hours if underwent PCI
      • 4-6 hours if underwent angiography alone
    • Angiography not performed in 9 patients (4%)
    • PCI performed in 107 patients (47%)
    • CABG performed in 6 patients (3%)

Control

  • Optimal medical treatment
    • Patient remained in community hospital
    • If had reinfarction, refractory angina despite optimal medical treatment, malignant ventricular arrythmias, or increasing symptoms of heart failure, they were considered for urgent coronary angiography

All patients treated according to existing guidelines

Baseline characteristics and medical treatments at discharge were similar between intervention vs. control groups, apart from the use of warfarin and nitrates:

  • Age, mean (range): 84.7 (80-93) vs. 84.9 (80-94)
  • Previous MI: 47% vs. 39%
  • T2DM: 20% vs. 14%
  • Killip class 1: 74% vs 74%
  • Ejection fraction
    • <30%: 5% vs. 3%
    • 30-50%: 28% vs. 31%
  • ECG
    • Pathological Q waves: 15% vs. 18%
    • ST depression: 19% vs. 18%
    • Negative T wave: 15% vs. 21%
  • Troponin elevation: 94% vs. 92%
  • Creatinine (mg/dL): 1.15 vs. 1.19
  • GRACE score: 138 vs. 138
  • Medical treatment at inclusion
    • Aspirin: 99% vs. 97%
    • Clopidogrel: 85% vs. 82%
    • Ticagrelor: 5% vs. 5%
    • Warfarin: 17% vs. 9%
    • Nitrates: 46% vs. 55%
  • Medical treatment at discharge
    • Aspiring: 95% vs. 93%
    • Clopidogrel: 72% vs. 73%
    • Ticagrelor: 4% vs. 4%
    • Warfarin: 22% vs. 14%
    • Beta-blocker: 86% vs 85%
    • ACE inhibitor or ARB: 52% vs. 54%
    • Nitrates: 34% vs. 48%

Outcome

  • Primary outcome: composite of myocardial infarction, need for urgent revascularisation, stroke and death was significantly reduced in the invasive group
    • 40.6% vs. 61.4%, Hazard ratio (HR) 0.53 (95% C.I. 0.41-0.69), p=0.0001, NNT = 5
    • Fragility index: 26 patients
  • Secondary outcomes:
    • Myocardial infarction – significantly reduced in invasive group
      • 17% vs. 30%, HR 0.52 (95% C.I. 0.35-0.76), p=0.001
    • Need for urgent revascularisation – significantly reduced in invasive group
      • 2% vs. 11%, HR 0.19 (95% C.I. 0.07-0.52), p=0.001
    • Stroke – no significant difference
      • 3% vs. 6%, HR 0.60 (95% C.I. 0.25-1.46), p=0.27
    • Death from any cause – no significant difference
      • 25% vs. 27%, HR 0.89 (95% C.I. 0.62-1.28), p=0.53
    • Complications – no significant difference
      • Major bleeding: 2% vs. 2%
      • Minor bleeding: 10% vs. 7%
    • Length of hospital stay: 6 days vs. 5 days
  • Sub-group analysis for primary outcome:
    • Age <90 years: Significantly reduced in invasive group
      • Hazard ratio 0.47 (95% C.I. 0.35-0.62; p=0.0001)
    • Age >90 years (n=34): No significant difference
      • Hazard ratio 1.21 (95% C.I. 0.53-2.7; p=0.64)

Authors’ Conclusions

  • The study supports an invasive strategy in clinically stable patients who are aged 80 years or over with NSTEMI/unstable angina. In patients ages older than 90 years the evidence is unclear

Strengths

  • Randomised controlled trial
  • Intention to treat analysis
  • Listed on clinicaltrials.gov
  • Patients quickly returned to community hospital post intervention. This means that any differences found are likely to be due to the intervention rather than differences in medical management that patients received at the community vs. central hospital
  • Multi-centre

Weaknesses

  • Large numbers of patients excluded for logistical reasons
  • Use of composite outcome. The need for urgent revascularisation is often included as part of the composite outcome in these type of trials. However, I think what would be most important to me would be my risk of dying or having severe morbidity such as a stroke of myocardial infarction; rather than the risk that if I do not have an angiogram now I might have to have it in the future

The Bottom Line

  • In patients over the age of 80 years with NSTEMI/unstable angina, an invasive compared with a conservative approach, reduces the composite outcome of death, myocardial infarction, stroke or need for urgent revascularisation. This was predominately due to the reduction in myocardial infarction and the need for urgent revascularisation. The reduction in myocardial infarction means that I will recommend this treatment to my patients.
  • In subgroup analysis of patients aged over 90 years there was no benefit found. However with low numbers of patients included over the age of 90 years, caution is required with interpreting these results.

External Links

Metadata

Summary author: Dave Slessor
Summary date: 13th April 2016
Peer-review editor: Steve Mathieu

3 comments

  • Duncan Chambler

    Thanks for another great summary, this time with a focus on emergency and acute medical care.
    Along with the weaknesses you mention, I note how the observed outcome rates were very different to the 21% expected rate used in the power calculation. Did the authors comment on why?
    Regardless of this, the findings are highly clinically and statistically significant and provide compelling evidence to support early PCI even at the expense of an inter-hospital transfer.
    However, given all the PCI / CABG procedures were performed in a single centre and they experienced a high exclusion rate due to “logistics”, I hope this trial is repeated to validate the large effect size.

  • Paul Young

    This is not evidence that intervention makes patients feel better, have improved quality of life, or live longer. I may be missing something but largely this seems to be evidence that lots of intervention now prevents some heart attacks and a relatively small amount of intervention later.

    All the early intervention will have morbidity and cost and around eight patients need to have the aggressive strategy to prevent one MI. On the other hand, with a conservative strategy seven patients out of eight will be avoiding an unnecessary angiogram/PCI/CABG and can be reassured that they don’t have an increased risk of death as a result. I am probably missing something. @dogICUma

    • David Slessor David Slessor

      Thanks @dogICUma for your comments. It’s always good to have someone challenge what we have written as it makes us think, stops us becoming complacent and encourages debate!
      Unfortunately a lot of what we practice is not based on high quality evidence. I think this trial would have benefited from looking at the functional outcome of the patients. It would have been more useful if we could tell our patients with this treatment you are more/less likely to live at home independently as oppose to with this treatment you are less likely to have a myocardial infarction. However myocardial infarction are likely to cause morbidity to our patients and I think for a number of our patients this would outweigh the risks of an early angiogram with PCI if required. There were small numbers of patients included who went on to have a CABG. This operation has significant morbidity and therefore the risk/benefit may not be in its favour. Future studies will hopefully try to investigate this.

Leave a Reply