Prediction of cardiovascular events using risk scores or electrocardiogram: A farewell to arms

‘The war is not won by victory’ Ernest Hemingway The discovery of the electrocardiogram (ECG) by the Dutch physiologist Willem Einthoven in 1902 gave physicians a powerful tool to understand the complexity of cardiac physiology and pathology.1 The analysis of cardiac electric activity was used for years to diagnose cardiac diseases and arrhythmias.2 In the field of coronary artery disease, the ST interval evaluation represents the milestone of acute coronary syndrome diagnosis,3 and the Q wave analysis continues to provide useful information regarding the extension of myocardial necrosis even in the troponin or cardiac magnetic resonance era.4 In the last few decades, the interest of cardiologist has progressively shifted from the diagnostic application of ECG to its prognostic utility.5 In fact, a large body of evidence has been produced on the predictive value of ECG intervals and waves. The P wave axis,6 duration7 and dispersion,8 PR interval,9 Q wave,10 QRS11 and QTc intervals,12 QT dispersion13 and QRS-T angle14 are all associated with major adverse cardiovascular events (CVEs) or cardiac arrhythmias. Taken together, these findings suggest that ECG should be used in clinical practice as a simple, reproducible and inexpensive tool only for diagnosis but also to predict prognosis to better stratify the patients’ cardiovascular risk. In this issue of EJPC, the prognostic value of electrocardiographic unrecognized myocardial infarction (MI) was investigated by Øhrn et al.15 in a general population of 5,686 patients derived from the Tromsø Study and prospectively followed for a mean of 5.5 years. The ECG criteria of the third universal definition of MI16 were used to identify a population without clinically recognized MI. The aim of the study was to investigate whether electrocardiographically detected unrecognized MI was independently associated with CVEs and death. An association of unrecognized myocardial MI with a 1.8-fold higher risk of both MI and all-cause mortality was reported by the authors; however, after adjustment for the traditional cardiovascular risk factors, these associations were not considered significant. In the era of widely available coronary revascularization and cardioprotective therapies, controversial results on the prognostic role of unrecognized MI have been found in previous studies in which several selected confounders, instead of the global cardiovascular risk (assessed with international scores), were used to identify the independent predictive value of ECG abnormalities.17,18 In the paper proposed by Øhrn et al., an unrecognized electrocardiographic MI did not improve risk prediction of future CV events when compared with the Framingham Risk Score (area under the curves 0.68 versus 0.68, p = 0.96) or the European HEART-SCORE (area under the curves 0.63 versus 0.63, p = 0.65). Nevertheless, these results provide new and useful information and suggest some interesting considerations. Even though the Framingham and HEART-SCORE risk scores were both intended for primary cardiovascular prevention, the current study reinforces the concept that these scores may also be able to predict CVEs in secondary cardiovascular prevention19; in fact, all proposed scores for secondary cardiovascular prevention (i.e. SMART risk score) include substantially similar items to calculate both Framingham and HEART-SCORE risk scores20 and the cornerstone of secondary prevention remains the rigorous management of all cardiovascular risk factors.21 Moreover, the similar predictivity between cardiovascular risk scores and ECG unrecognized MI does not deny the value of these ECG findings, rather suggesting that patients with these ECG abnormalities have to be considered at the same cardiovascular risk of both Framingham and HEART-SCORE high-risk patients. In this sense, paraphrasing the wisdom of the great writer Ernest Hemingway, this predictive war between ECG and cardiovascular risk scores ‘was not won by victory’. Finally, the results proposed by Øhrn et al. seem to suggest not to separate ECG data from the anamnestic evaluation, thereby reminding that, instead of a blind evaluation of instrumental data, a better way to predict CVE is represented by an integrated analysis of all cardiovascular risk factors. The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) received no financial support for the research, authorship, and/or publication of this article.