The emerging prominence of computer electrocardiography in large population-based surveys

The year 1987 marks the centennial of electrocardiography. As Beamish reminds us, it was the first and foremost technical aid to the cardiologist. During the history of cardiovascular epidemiology, which dates back roughly forty years, electrocardiography has played a central role as a relatively specific marker for coronary and hypertensive heart disease. The lack of sensitivity of electrocardiography and the advent of new technologies such as ultrasonography for assessment of such heart disease in cardiology and cardiovascular epidemiology has raised questions as to whether electrocardiography might be replaced in the next century. Recent developments in computer electrocardiography have rendered suggestions of the decreased importance of electrocardiography premature. Use of computer analysis of electrocardiograms in epidemiologic studies, which dates back less than three decades, has renewed interest in and expanded the role of electrocardiography in cardiology in general and cardiovascular epidemiology in particular. The computer analysis of electrocardiograms has allowed the processing and analysis of large numbers of electrocardiograms in a rapid, efficient, systematic and consistent manner. This has led to new, more sensitive criteria for detection of left ventricular hypertrophy as a discrete variable. In addition, the numerous waveform amplitude and duration measurements by computer have allowed the development of mathematical models to express the likelihood of coronary heart disease (myocardial infarction) and degree of left ventricular hypertrophy (estimates of left ventricular mass) on continuous scales. These developments have contributed substantially to the potential of electrocardiography (and particularly computer electrocardiography) in future studies and clinical applications.