CJEM Articles: prognosis

Displaying 1-2 of 2 results

  • January 2006 8 1
    Bernd Eber, Christian Punzengruber, Edwin Maurer, Elisabeth Lassnig, Gudrun Lamm, Herbert Mayr, Johann Auer, Michael Porodko, Robert Berent, Thomas Weber

    Background: Accurate prediction of survival to hospital discharge in patients who achieve return of spontaneous circulation after cardiopulmonary resuscitation (CPR) has significant ethical and socioeconomic implications. We investigated the prognostic performance of serum neuron-specific enolase (NSE), a biochemical marker of ischemic brain injury, after successful CPR.
    Methods: In-hospital or out-of-hospital patients with nontraumatic normothermic cardiac arrest who achieved return of spontaneous circulation (ROSC) following at least 5 minutes of CPR were eligible. Neuron-specific enolase levels were assessed immediately, 6 hours, 12 hours and 2 days after ROSC. Subjects were followed to death or hospital discharge.
    Results: Seventeen patients (7 men, 10 women) were enrolled during a 1-year period. Median (range) NSE levels in survivors and non-survivors respectively were as follows: immediately after ROSC: 14.0 μg/L (9.1-51.4 μg/L) versus 25.9 μg/L (10.2-57.5 μg/L); 6 hours after ROSC: 15.2 μg/L (9.7-30.8 μg/L) versus 25.6 μg/L (12.7-38.2 μg/L); 12 hours after ROSC: 14.0 μg/L (8.6-32.4 μg/L) versus 28.5 μg/L (11.0-50.7 μg/L); and 48 hours after ROSC: 13.1 μg/L (7.8-29.5 μg/L) versus 52.0 μg/L (29.1-254.0 μg/L). Non-survivors had significantly higher NSE levels 48 hours after ROSC than surivors (p = 0.04) and showed a trend toward higher values during the entire time course following ROSC. An NSE concentration of >30 μg/L 48 hours after ROSC predicted death with a high specificity (100%: 95% confidence interval [CI] 85%-100%), and a level of 29 μg/L or less at 48 hours predicted survival with a high specificity (100%: 95% CI 83%-100%).
    Conclusions: Serum NSE levels may have clinical utility for the prediction of survival to hospital discharge in patients after ROSC following CPR over 5 minutes in duration. This study is small, and our results are limited by wide confidence intervals. Further research on ability of NSE to facilitate prediction and clinical decision-making after cardiac arrest is warranted.

  • January 2004 6 1
    Akbar Panju, Eric Stanton, Gordon H. Guyatt, John Opie, Lauren Griffith, Matthew J. McQueen, P.J. Devereaux, Stephen A. Hill

    Objective: To determine the ability of troponin I (TnI) measurement to predict the likelihood of a serious cardiac outcome over the subsequent 72 hours in patients presenting to the emergency department (ED) with symptoms suggestive of an acute coronary syndrome.

    Methods: This prospective observational study enrolled consecutive patients presenting to 2 urban tertiary care hospital EDs over a 5-week period. Eligible patients included those for whom a TnI test was ordered within 24 hours of arrival and in whom no serious cardiac outcome occurred before the test result was available. Patients were followed for 72 hours and serious cardiac outcomes documented; these included cardiovascular death, myocardial infarction, congestive heart failure, serious arrhythmia and refractory pain. We calculated likelihood ratios (LRs) to describe the association of the TnI result with serious cardiac outcomes.

    Results: Of the 352 enrolled patients, 20 had a serious cardiac outcome within 72 hours of ED presentation. The derived LRs (and 95% confidence interval [CI]) were 0.5 (0.3-0.9) for TnI values <0.5 µg/L, 1.6 (0.4-6.5) for TnI values from 0.5 to 2.0 µg/L, 5.8 (1.7-19.5) for TnI values from >2.0 to 10.0 µg/L and 14.4 (4.8-42.9) for TnI values >10.0 µg/L.

    Conclusions: TnI values >2.0 µg/L are associated with an increased probability of serious cardiac outcomes within 72 hours. TnI values between 0.5 and 2.0 µg/L are weakly positive predictors. TnI values <0.5 µg/L have LRs in the range of 0.5 and thus are weakly negative predictors, not substantially decreasing the likelihood of serious cardiac outcomes, particularly in patients with a moderate or high pretest probability.