ABC | Volume 114, Nº2, February 2020

Update Update of the Brazilian Guideline on Nuclear Cardiology – 2020 Arq Bras Cardiol. 2020; 114(2):325-429 Figure 13 – Importance of amount of myocardium at risk (extension and intensity of ischemia) on myocardial perfusion imaging with radiopharmaceuticals ( 99m Tc-sestamibi or thallium-201) to post-test probability of coronary artery disease (CAD). For a given pretest probability (50% indicated in the graph), the post-test probabilities will be significantly higher according to imaging findings. With the condition of high-risk ischemia or > 20% extent of ischemic myocardium, the clinical implications for decision making become practically independent of pre-test probability of CAD. Source: Adapted from Udelson JE et al. 130 POST-TEST PROBABILITY OF CAD % PRETEST PROBABILITY OF CAD % SEVERE ABNORMALITY MODERATE ABNORMALITY MILD ABNORMALITY Figure 14 – Hypothetical examples of ROC curves, with: the area under the curve representing maximum or perfect diagnostic accuracy of the standard utilized (curve A; AUC = 1); “real” area under the curve representing good efficiency of the method used, often found in clinical practice (curve B; AUC = 0.85); the 45-degree diagonal line corresponding to random chance (curve C; AUC = 0.50), with the area under the ROC demonstrating the averages of diagnostic accuracy across a spectrum of cutoff values. On rare occasions, the estimatedAUC is less than 0.5, indicating that the test being evaluated performs worse than random chance.Adapted from Zou KH et al. 133 SENSITIVITY 1 − SPECIFICITY 352