What is the value of chest x-ray for patients with acute chest pain?

MASIA T, GRANDPIERRE RG, BOBBIA X, POMMET S, MOREAU A, BODIN Y, PERRIN-BAYARD R, TREILLE J, CLARET PG, DE LA COUSSAYE JE. What is the value of chest x-ray for patients with acute chest pain? Med Emergency, MJEM 2016; 25:25-9.

Key words: Acute chest pain, chest x-ray, emergency service, relevance

  • Authors’ affiliation
  • Article history / info
  • Conflict of interest statement
Correspondent author: Romain Genre GRANDPIERRE, MD

Pôle Anesthésie Réanimation Douleur Urgences, Nîmes University Hospital,

4 Rue du Professeur Robert Debré 30029 Nîmes, France

romainsbs@hotmail.com

Masia T, MD, Grandpierre RG, MD, Bobbia X, MD, Pommet S, MD, Moreau A, MD, Bodin Y, MD, Perrin-Bayard R, MD, Treille J, MD, Claret PG, MD, de La Coussaye JE, MD

Pôle Anesthésie Réanimation Douleur Urgences, Nîmes University Hospital, Nîmes, France.

Category: Original article

Received: April 13, 2016

Revised: May 11, 2016

Accepted: June 01, 2016

There is no conflict of interest to declare

ABSTRACT

Background: Chest pain is a frequent reason of consultation in Emergency Department (ED). Despite the lack of recommendations, chest X-Ray (CXR) is often performed in this context.

Methods: This is a single-center retrospective study. From September to November 2012, adult patients admitted to ED with chest pain were included in the study. CXR and patients characteristics were analyzed. An expert committee was in charge to determine if the realization of the CXR was useful or not.

Results: Of the 300 patients who consulted in our ED for chest pain, CXR were performed for 71% (N=205) of them. Twenty percent (N=40) of CXRs were interpreted as pathological. Radiological confirmation rate for the initial medical hypothesis was 12 % (N=24). Our expert committee judged 92% (N=188) of the CXR was relevant. Discordances between experts were observed for 20 medical records (kappa test = .109). In multivariate analysis, variables associated with abnormal CXR were: presence of one or more cardio-vascular risk factor (p=.05), clinical exam abnormalities (dyspnea, caught, pulmonary exam abnormalities) (p=.006) and laboratory abnormalities (troponin, D-dimers, CRP, and white blood cells) (p=.016).

Conclusions: One CXR of five find abnormalities. Experts judged relevant for nine cases of ten with a low level of concordance. The absence of dyspnea, fever, cough, auscultatory abnormalities, and biological inflammatory reaction are independent criteria of normal CXR. Pulmonary ultrasound and guidelines could help to optimize practices. 

Full article