Effect of Duration From Symptom Onset on the Negative Predictive Value of a Normal ECG for Exclusion of Acute Myocardial Infarction, ,

Study objective: We hypothesized that the negative predictive value of the ECG would improve with time and assessed the effect of time elapsed from symptom onset to ED presentation on the negative predictive value, sensitivity, specificity, and positive predictive value of the initial ECG in patients presenting with symptoms suggestive of acute myocardial infarction (AMI). Methods: We conducted a cross-sectional study in a university teaching hospital. Our structured data instrument, completed at the time of presentation included demographics, time of onset of symptoms, history, laboratory, and ECG findings. AMI was diagnosed with the use of international diagnostic criteria. Patients were stratified according to duration of time from symptom onset at 3-hour intervals. Results: We enrolled 526 patients in the study group. The mean age was 59 years; 40% were female. The mean time elapsed from symptom onset to presentation was 185 minutes. A diagnosis of AMI was made in 104 patients (20%). The negative predictive values of a normal ECG for exclusion of AMI, stratified by duration of time from symptom onset, were: 0 to 3 hours, 93.2% (95% confidence interval [CI], 87.4% to 96.1%); 3 to 6 hours, 93.0% (95% CI, 83.0% to 98.1%); 6 to 9 hours, 92.6% (95% CI, 75.8% to 99.1%); and 9 to 12 hours, 94.1% (95% CI, 71.3% to 99.9%) (P=1.0). The sensitivity, specificity, and positive predictive value of the ECG were similar in all groups. Conclusion: The negative predictive value of a normal ECG for exclusion of AMI does not improve as the duration of time from symptom onset to presentation increases. Normal ECG findings cannot be used to rule out an AMI, even those obtained up to 12 hours after symptom onset. [Singer AJ, Brogan GX, Valentine SM, McCuskey C, Khan S, Hollander JE: Effect of duration from symptom onset on the negative predictive value of a normal ECG for exclusion of acute myocardial infarction. Ann Emerg Med May 1997; 29:575-579.]