Exploring real-life clinical decision pathways following non-diagnostic coronary computed tomography angiography

Abstract Introduction Coronary computed tomography angiography (CCTA) plays a crucial role in non-invasive stratification of patients (pts) with suspected symptomatic coronary artery disease (CAD). However, due to some pitfalls, some tests may be non-diagnostic leading to management challenges, especially whether to proceed with additional investigation. Objectives To identify and characterize pts with non-diagnostic CCTA and evaluate the subsequent clinical management and outcomes. Methods Single-center retrospective cohort study of 354 pts with suspected obstructive CAD who underwent CCTA between June 2022 and September 2024. Non-diagnostic test (CAD-RADs N) was defined according to the impossibility of excluding obstructive CAD (≥50% stenosis) in ≥1 segment and absence of ≥50% stenosis in interpretable segments. Demographic and baseline clinical characteristics, prior and subsequent diagnostic tests, follow-up information on clinical discharge from cardiology consultation (CC), cardiovascular (CV) outcomes or hospital admission due to chest pain were assessed. Statistical analyses were conducted using SPSS v29. Results Overall, 42 pts (11.9%) had a non-diagnostic test with a mean age of 56.7 years and 52.0% being male. Medical history and chronic medication are described in table 1. Nineteen pts (45.2%) had CCTA as first-line non-invasive stratification exam while 23 pts (54.8%) underwent previous testing, predominantly exercise electrocardiogram (ECG) test (14/23; 60.9%). The main reasons for non-diagnostic CCTA were poor apnea (40.5%) and ECG-related artifacts (31.0%). Over a mean follow-up of 15.4 months, most pts were discharged from CC (54.2%) and only 4 (9.5%) underwent additional testing which did not identify obstructive/significant CAD. During this period, 5 pts (11.9%) had unplanned hospital admission due to chest pain, with none of them with subsequent diagnostic tests or CC discharge, and only 1 patient (2.4%) presented with acute myocardial infarction (AMI). Conclusions In our study, the rate of non-diagnostic CCTA was higher than the registered in the PROMISE trial (11.9 vs 6.0%), which may reflect suboptimal patient selection in clinical practice. During follow-up, most pts were discharged from the CC and less than 10% were submitted to additional testing. However, only one patient experienced non-fatal AMI. A larger sample would be useful to assess whether further testing impacts clinical outcomes or not in this population.