Unrecognized non-Q-wave myocardial infarction: prevalence and prognostic significance in patients with suspected coronary disease

Abstract

Background: Unrecognized myocardial infarction (UMI) is known to constitute a substantial portion of potentially lethal coronary heart disease. However, the diagnosis of UMI is based on the appearance of incidental Q-waves on 12-lead electrocardiography. Thus, the syndrome of non-Q-wave UMI has not been investigated. Delayed-enhancement cardiovascular magnetic resonance (DE-CMR) can identify MI, even when small, subendocardial, or without associated Q-waves. The aim of this study was to investigate the prevalence and prognosis associated with non-Q-wave UMI identified by DE-CMR.

Methods and findings: We conducted a prospective study of 185 patients with suspected coronary disease and without history of clinical myocardial infarction who were scheduled for invasive coronary angiography. Q-wave UMI was determined by electrocardiography (Minnesota Code). Non-Q-wave UMI was identified by DE-CMR in the absence of electrocardiographic Q-waves. Patients were followed to determine the prognostic significance of non-Q-wave UMI. The primary endpoint was all-cause mortality. The prevalence of non-Q-wave UMI was 27% (50/185), compared with 8% (15/185) for Q-wave UMI. Patients with non-Q-wave UMI were older, were more likely to have diabetes, and had higher Framingham risk than those without MI, but were similar to those with Q-wave UMI. Infarct size in non-Q-wave UMI was modest (8%+/-7% of left ventricular mass), and left ventricular ejection fraction (LVEF) by cine-CMR was usually preserved (52%+/-18%). The prevalence of non-Q-wave UMI increased with the extent and severity of coronary disease on angiography (p<0.0001 for both). Over 2.2 y (interquartile range 1.8-2.7), 16 deaths occurred: 13 in non-Q-wave UMI patients (26%), one in Q-wave UMI (7%), and two in patients without MI (2%). Multivariable analysis including New York Heart Association class and LVEF demonstrated that non-Q-wave UMI was an independent predictor of all-cause mortality (hazard ratio [HR] 11.4, 95% confidence interval [CI] 2.5-51.1) and cardiac mortality (HR 17.4, 95% CI 2.2-137.4).

Conclusions: In patients with suspected coronary disease, the prevalence of non-Q-wave UMI is more than 3-fold higher than Q-wave UMI. The presence of non-Q-wave UMI predicts subsequent mortality, and is incremental to LVEF.

Trial registration: Clinicaltrials.gov NCT00493168.

Figure 1. Typical DE-CMR images.

Short and long axis views of DE-CMR images from four patients are shown. Patients A–C demonstrate hyperenhancement (red arrows) consistent with prior myocardial infarction. None had Q-waves on electrocardiography, and all three were classified as having non-Q-wave UMI. Of note, Patient C has evidence of two distinct infarcts. Patient D has hyperenhancement (blue arrows) involving the midwall of the interventricular septum, sparing the subendocardium. This pattern is not consistent with prior myocardial infarction, and this patient was categorized in the “no MI” group. See text for further details.

Figure 2. Prevalence of UMI stratified by angiographic extent and severity of coronary artery disease.

The prevalence of non-Q-wave and Q-wave UMI increased with both the extent and severity of CAD. See text for further details. CAD = coronary artery disease; UMI = unrecognized myocardial infarction; 1V = single vessel; 2V = double vessel; 3V = triple vessel.

Figure 3. Kaplan–Meier estimates of survival (A) and cardiac survival (B) in patients with unrecognized non-Q-wave MI (blue line) and without MI (red line).

Overall and cardiac survival in patients with non-Q-wave UMI was significantly reduced in comparison to patients without MI (p<0.0001 for both). The annual mortality in patients with non-Q-wave MI was 15-fold higher than that in patients without MI (10.8% per year versus 0.8% per year, respectively).