Acute Myocardial Infarction: Individual pre-discharge tests were insensitive predictors of poor outcome

Clinical bottom line (level 2a)

  1. About a tenth of patients with myocardial infarction will die or have an MI.
  2. In patients with myocardial infarction, certain characteristics on non-invasive testing techniques may be helpful in assessing predischarge risk, but they are not very accurate.
Shaw et al: Am J Cardiology 1996; 78: 1327-1337
Expires March 2003

The study

Systematic review of cohort studies published after 1980 of
  • Patients: myocardial infarction
  • Intervention: predischarge testing <6wk from MI
  • Outcome: cardiac death


    • Articles found in english alone using MEDLINE, 1980-1995 (search terms: myocardial infarction; exercise test; use of 'radionucleotide imaging' subheading; echocardiography; dipyramidole ) and searching review bibliographies

    Selection criteria: published studies with consecutive patients, >80% follow-up, adequate data on testing and outcome
    Appraisal criteria: used an instrument developed by Baker (JAMA 94;272:1528)
    Articles excluded if: excluded if insufficient data (14), mixed population of patients (8), pre-1980 study (11), follow-up short or <80% (5), duplicate reports (18)

    28 articles on 15613 patients with ECG data, 8 studies on 1247 patients with ventricular perfusion imaging, 11 studies of ventricular function on 1464 patients, and 8 studies of pharmacological stress testing on 1550 patients
    chi-squared and eyeball tests for heterogeneity were negative

    The evidence

    outcome time to outcome number of patients/total number %
    (95% CI)
    cardiac death ? / 4%
    (% to %)

    prognostic factor for
    cardiac death     		time to outcome control rate (%) adjusted OR
    (95% CI)
    ECG: ST depression >1mm ? 1.7
    (1.2 to 2.5)
    ECG: impaired systolic BP ? 4.0
    (2.5 to 6.3)
    ECG: limited exercise duration ? 4.0
    (1.9 to 8.4)
    Myocardial perfusion imaging; exercise testing: reversible perfusion defect ? 3.1
    (1.6 to 4.6)
    Myocardial perfusion imaging; pharmacological stress: reversible perfusion defect ? 1.2
    (0.4 to 3.7)
    Ventricular function; exercise stress: EF<40% ? 3.2
    ( to )
    Ventricular function; exercise stress: EF change <5% ? 4.2
    ( to )
    Ventricular function; exercise stress: new dyssynergy ? 1.2
    ( to )
    Ventricular function; pharmacological stress: new dyssynergy ? 2.7
    (1.4 to 5.2)

    Comments

    1. Exclusion of pre-1980 patients intended to reflect patients exposed to current therapies and technologies
    2. No combination of tests or direct comparison of tests performed

    Citation

    1. Shaw LJ, Peterson ED, Kesler K, et al: A metaanalysis of predischarge risk stratification after acute myocardial infarction with stress echocardiography, myocardial perfusion and ventricular function imaging. Am J Cardiology 1996; 78: 1327-1337
    Contributor: Bob Phillips and Clare Wotton, November 1999
    Reviewer:

    Clinical Question.
    Patient patients who had a myocardial infarction
    Intervention or Exposure non-invasive pre-discharge testing
    Outcome cardiac death