ST changes during Holter monitoring predicted poor outcome after myocardial infarction
ACP J Club. 1992 July-Aug;117:25. doi:10.7326/ACPJC-1992-117-1-025
Petretta M, Bonaduce D, Bianchi V, et al. Characterization and prognostic significance of silent myocardial ischemia on predischarge electrocardiographic monitoring in unselected patients with myocardial infarction. Am J Cardiol. 1992 Mar 1;69:579-83.
To study the prevalence, characteristics, and prognostic significance of ischemic ST changes in predischarge Holter monitoring in patients with a first myocardial infarction.
Inception cohort followed for 2 years.
Coronary care unit of an Italian university medical center.
441 consecutive patients with a first myocardial infarction were evaluated and included if they could have 24-hour continuous electrocardiographic (Holter) monitoring, were ≤ 70 years old, and had no early postinfarction angina. Patients with conduction disturbances, valvular heart disease, pacemaker rhythm, left ventricular hypertrophy, resting ST-segment elevation ≥ 1.5 mm, dominant S or Q waves in recorded channel leads, postural ST changes, or severe ventricular arrhythmias were excluded. 2-year follow-up was 87.4%.
Assessment of prognostic factors
270 patients (214 men, mean age 57.9 y) had continuous dual-channel Holter monitoring (leads V2 and V6) before discharge, an average of 13 days after symptom onset. Patients charted symptoms, and a technician analyzed heart rate, and onset, duration, and end of ischemic events. Ischemic episodes were those with > 1 mm ST depression at the J point and at 80 ms thereafter in each complex in consecutive beats for ≥ 60 seconds. All patients also had serial echocardiographic evaluation for analysis of left ventricular wall motion.
Main outcome measures
Cardiac death and re-infarction, determined at clinic visits or by phone call.
64 (24%) of 270 patients had ischemic ST episodes. Patients with ST changes had a higher incidence of > 10 premature ventricular contractions per hour (P < 0.01) and a lower incidence of left ventricular wall motion abnormalities(P < 0.05) than patients without ST changes. After 2 years 53 patients with and 183 without ST changes were analyzed. Cardiac death and re-infarction were higher in patients with ST changes (26.4% vs 12.6%, odds ratio [OR] 2.5, 95% CI 1.1 to 5.6; P < 0.05). The groups did not differ for unstable angina or cardiac vascularization. Multivariate analysis showed Killip class was the best predictor of a first recurrent cardiac event (OR 3.0, CI 1.6 to 4.3; P < 0.01) followed by ST changes (OR 1.8, CI 1.1 to 2.4; P < 0.05).
After myocardial infarction, ST changes during Holter monitoring were associated with increased cardiac death and re-infarction.
Source of funding: Not stated.
Address for reprint: Dr. D. Bonaduce, Via Aniello Falcone, 394, 80127 Napoli, Italy.
Recurrent ischemia after myocardial infarction is associated with increased risk for cardiac death, recurrent myocardial infarction, and multi-vessel coronary artery disease. It is therefore standard practice to determine prognosis in postinfarction patients by screening for inducible ischemia with noninvasive methods. This study shows that ST changes detected by 24-hour Holter monitoring may also identify high-risk patients.
Petretta and colleagues studied an inception cohort of patients admitted to a coronary care unit with a first myocardial infarction. Unfortunately, the referral pattern description was incomplete, and physicians may have difficulty determining if the study patients are similar to their own. The end point of cardiac mortality was well defined, but the absence of cardiac enzyme analysis makes the definition of infarction or re-infarction unconventional. Although follow-up was 87%, the telephone survey to determine cardiac death and re-infarction may be inaccurate. It is also unclear whether the outcomes assessment was "blind."
The study confirms the findings of earlier investigations that a substantial proportion of postinfarction patients have silent ischemia and that asymptomatic ST changes during Holter monitoring can predict future myocardial events. However, the frequency of silent ischemia and the number of myocardial events in both the ST-change and no ST-change groups may have been higher than expected because anti-ischemic therapy was suboptimal as shown by the low proportion of patients on β-blocker ers.
If the physician has access to Holter monitoring equipment capable of detecting ischemic ST-segment shifts (1), this technique allows risk stratification of patients who cannot participate in exercise testing. I would prefer to use radionuclide imaging with pharmacologic vasodilation, however, because this method determines the location and severity of ischemia with greater precision.
Allan D. Kitching, MSc, MD
The Toronto HospitalToronto, Ontario, Canada