Evaluating the utility of chest x-rays for non-traumatic chest pain in Australia- a retrospective cohort study.

Abstract

Purpose: The aim of this study was to quantify the proportion of chest x-rays (CXRs) for non-traumatic chest pain (NTCP) in the emergency department (ED) that were abnormal and assess the clinical significance of these abnormalities. We also aimed to explore the variables associated with abnormal and clinically significant abnormal CXRs, to predict a population where CXRs can be safely avoided.

Methods: A single center retrospective cohort study was conducted including all adult patients presenting to a single ED with NTCP between 01 Jan 2022 and 31 Dec 2022. We categorized the CXRs into abnormal, or normal as reported by a radiologist. Abnormalities were categorized to be clinically significant based on potential or actual changes in patient management. The association of patient demographics, presenting vital signs, and clinical characteristics with clinically significant abnormalities were explored using multivariable logistic regression analysis.

Results: There were 3,419 eligible patient encounters included for analysis. Of these, 746 (21.8%; 95%CI: 20.4-23.2%) CXRs had at least one abnormality detected. There were 218 (6.4%; 95%CI: 6.1-7.9%) CXRs deemed to have clinically significant abnormalities. Age categories of 50-64 years (aOR 1.64; 95%CI 1.04-2.60), and age > 64 years (aOR 2.32; 95%CI: 1.51-3.57), history of congestive heart failure (CHF) (aOR 1.86; 95%CI: 1.08-3.21), smoking (aOR 1.27; 95%CI: 1.04-1.57), hemoptysis (aOR 6.69; 95%CI: 1.92-23.33), diminished lung sounds (aOR 4.87; 95%CI:2.95-8.05), rales (aOR 4.49; 95%CI: 2.82-7.15), and abnormal oxygen saturations (aOR 1.98; 95%CI: 1.40-2.79) were associated with clinically significant abnormalities on CXRs. In the absence of these variables, 1.4% (95%CI: 0.6-2.6%) of CXRs were abnormal with clinical significance.

Conclusions: CXRs have a relatively high yield of abnormalities among patients with NTCP. However, some CXRs could be safely avoided in the absence of variables associated with clinically significant abnormalities. Further validation of these clinical characteristics is required before translation to clinical practice.