Myocardial Infarction: Ventricular fibrillation was associated with high mortality rates

Clinical bottom line (level 2b)

  1. In patients with first myocardial infarction, advanced age and occurrence of ventricular fibrillation (especially after 48 hours) were associated with in hospital mortality (RR 2.5 to 20)
  2. In patients with first myocardial infarction, a non-Q wave infarct had a better in-hospital mortality rate (RR=0.08)
  3. In patients with first myocardial infarction, hypokalaemia, hypotension, smoking, extensive infarct, inferoposterior location, age <70y and bradycardia all independently predicted early onset ventricular fibrillation (RR 1.39 to 1.97)
  4. In patients who had survived ventricular fibrillation after first myocardial infarction, there was no residual effect on mortality rates
Volpi et al: American Journal of Cardiology 1998; 82: 265-271
Expires March 2003

The study

Prospective cohort study with objective outcomes, adjusted for confounding factors, not validated in an independent set of patients.

Setting: multi-centre European hospitals

7755 patients (aged mean 60y (SD 11y), 81% male) with first myocardial infarction (diagnosed by chest pain with ST elevation within 6h onset of pain) and Killip class I

Excluded if
  • failure to enter RCT (GISSI-2) of streptokinase vs. alteplase



    • Factors studied:
  • serum K + <3.6mmol/L
  • systolic BP <120 mmHg
  • current smoking
  • >3 leads with ST elevation
  • age <71y
  • inferoposterior site
  • admission bradycardia (<60 bpm)
  • VF within 4 hours of infarction
  • VF >4hrs but <48hrs after infarction
  • VF >48hrs after infarction
  • advanced age (>70y)
  • anterior infarct
  • non-Q wave infarct
  • increased number of leads with ST elevation (6-7)
  • heart failure
  • 2nd or 3rd degree heart block
  • reinfarction
  • in-hospital VF


    • All patients had aspirin, thrombolysis and a beta-blocker unless contraindicated. Patients who developed VF were usually put on a lidocaine infusion.

    multivariate analysis with Mantel-Haenszel estimate of relative risk

    98% total (11% via census office) followed for 6 months
    Outcomes studied:
  • early ventricular fibrillation within 4 hours, Killip class I
  • in-hospital mortality Data relates to patients without VF
  • 6 month mortality Data relates to patients without VF

    • The evidence

    outcome time to outcome number of patients/total number %
    (95% CI)
    early ventricular fibrillation 6 months 435/7755 5.6%
    (5.1% to 6.1%)
    in-hospital mortality 6 months 293/7320 4%
    (3.6% to 4.4%)
    6 month mortality 6 months 141/7044 2.0%
    (1.7% to 2.3%)

    prognostic factor for
    early ventricular fibrillation     		time to outcome control rate (%) adjusted OR
    (95% CI)
    serum K + <3.6mmol/L ? 1.97
    (1.51 to 2.56)
    systolic BP <120 mmHg ? 1.74
    (1.34 to 2.26)
    current smoking ? 1.66
    (1.15 to 2.41)
    >3 leads with ST elevation ? 1.66
    (1.24 to 2.23)
    age <71y ? 1.61
    (1.08 to 2.40)
    inferoposterior site ? 1.45
    (1.10 to 1.91)
    admission bradycardia (<60 bpm) ? 1.39
    (1.05 to 1.84)

    prognostic factor for
    in-hospital mortality     		time to outcome control rate (%) adjusted OR
    (95% CI)
    VF within 4 hours of infarction ? 2.47
    (1.48 to 4.13)
    VF >4hrs but <48hrs after infarction ? 3.97
    (1.51 to 10.5)
    VF >48hrs after infarction ? 20.2
    (11.5 to 35.5)
    advanced age (>70y) ? 7.0
    (4.8 to 10.2)
    anterior infarct ? 2.15
    (1.61 to 2.89)
    non-Q wave infarct ? 0.08
    (0.03 to 0.23)
    increased number of leads with ST elevation (6-7) ? 1.86
    (1.26 to 2.74)
    heart failure ? 2.47
    (1.84 to 3.33)
    2nd or 3rd degree heart block ? 2.89
    (2.03 to 4.09)
    reinfarction ? 2.07
    (1.17 to 3.67)

    prognostic factor for
    6 month mortality     		time to outcome control rate (%)
    in-hospital VF ? 7/327
    (2.1%)

    Comments

    1. Randomised controlled trial patients who are a selected population
    2. There was no clear difference in post-discharge mortality between patients with and without primary VF.

    Citation

    1. Volpi A, Cavalli A, Santoro L, et al: Incidence and prognosis of early primary ventricular fibrillation in acute myocardial infarction - results of the GISSI-2 database. American Journal of Cardiology 1998; 82: 265-271
    Contributor: Bob Phillips; Chris Ball; Lee Bailey, November 1999
    Reviewer: Dwight Peretz

    Clinical Question.
    Patient patients with first acute myocardial infarction
    Intervention or Exposure no heart failure (Killip I); ventricular fibrillation
    Outcome ventricular fibrillation; mortality