SPC365

ClinicAl verSus algorIthmic predictioN of Obstructive Coronary Artery Disease – the CASINO study

Session: Comunicações Orais - Sessão 09 - Síndrome coronária aguda 1
Speaker: Mariana Sousa Paiva

Congress: CPC 2024
Topic: E. Coronary Artery Disease, Acute Coronary Syndromes, Acute Cardiac Care
Theme: 13. Acute Coronary Syndromes
Subtheme: 13.4 Acute Coronary Syndromes – Treatment
Session Type: Comunicações Orais
FP Number: ---
Authors: Mariana Sousa Paiva; Rita A. Carvalho; João Presume; Sérgio Maltês; Catarina Brízido; Liliana Marta; Sérgio Madeira; Daniel Matos; Francisco Moscoso Costa; Marisa Trabulo; Jorge Ferreira; António M. Ferreira

<p style="text-align:justify"><strong><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">[Background]:</span></span></strong></p>

<p style="text-align:justify"><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">The estimation of pretest probability (PTP) is a key step when evaluating patients with suspected coronary artery disease (CAD). Current European guidelines recommend using an algorithm based on age, sex, and symptom typicality. However, many cardiologists do not regularly use this method, relying instead on overall clinical impression. </span></span></p>

<p style="text-align:justify"><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">The aim of this study was to compare clinical and algorithmic prediction of obstructive CAD in a set of symptomatic patients with suspected chronic coronary syndrome (CCS).</span></span></p>

<p style="text-align:justify">&nbsp;</p>

<p style="text-align:justify"><strong><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">[Methods]:</span></span></strong></p>

<p style="text-align:justify"><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">In this survey study, 10 cardiologists from a tertiary center were asked to estimate the probability of obstructive CAD (on a scale of 1-99% and without the aid of any score or algorithm) for 100 anonymized clinical vignettes of outpatients who underwent diagnostic workup for suspected CCS. </span></span></p>

<p style="text-align:justify"><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">The provided information included age, sex, symptom description, cardiovascular risk factors, known comorbidities, and history of peripheral and/or cerebrovascular disease. </span></span></p>

<p style="text-align:justify"><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">Whenever non-invasive tests had been performed, physicians were asked to estimate the probability of obstructive CAD before and after knowing the test results. Obstructive CAD was defined as diameter stenosis &ge;50% confirmed by invasive coronary angiography.</span></span></p>

<p>&nbsp;</p>

<p><strong><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">[Results]:</span></span></strong></p>

<p><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">Among the 100 included patients (52 women, mean age 60&plusmn;10 years), symptoms consisted of chest pain in 88 and dyspnea in the remainder. After diagnostic workup, the observed prevalence of obstructive CAD was 12% (n=12).</span></span></p>

<p><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">The median PTP of obstructive CAD was 13% (IQR 6-24) according to the ESC algorithm, yielding good calibration (predicted vs. observed p=0.766). </span></span></p>

<p><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">Clinical estimates of PTP varied widely between cardiologists, were significantly higher than algorithmic likelihoods (median 33%, IQR 23-48%, p&lt;0.001 for the comparison), and significantly overestimated the presence of obstructive CAD (predicted vs. observed p&lt;0.001) &ndash; Figure 1A.</span></span></p>

<p><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">The ESC algorithm showed good discriminative power to identify patients with obstructive CAD (C-statistic 0.74, 95% CI 0.63-0.86, p=0.006), whereas the C-statistic for clinicians ranged from 0.50 to 0.72 (average 0.65&plusmn;0.06) &ndash; Fig 1B&nbsp;(p values for individual comparisons 0.045&ndash; 0.700).</span></span></p>

<p><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">In the subset of 58 patients with prior non-invasive tests, clinicians significantly changed their predictions after knowing test results (mean absolute difference 21%&plusmn;5%, p&lt;0.001), but that change did not improve their discriminative power significantly (mean change in C-statistic of 0.01&plusmn;0.11, p=0.810). &nbsp;</span></span></p>

<p>&nbsp;</p>

<p><strong><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">[Conclusion]:</span></span></strong></p>

<p><span style="font-size:12pt"><span style="font-family:Calibri,sans-serif">Clinicians tend to overestimate the likelihood of obstructive CAD and, despite using more clinical data, they fail to outperform a simple algorithm using only age, sex, and symptom typicality. The systematic use of this tool should be promoted in order to improve disease prediction and guide non-invasive testing.</span></span></p>