Abstract: Poster Presentations |


Amy S. Guralnick, MD*; Renaud M. Gueret, MD; Aiman Tulaimat, MD
Author and Funding Information

Cook County Hospital, Chicago, IL


Chest. 2007;132(4_MeetingAbstracts):569c-570. doi:10.1378/chest.132.4_MeetingAbstracts.569c
Text Size: A A A
Published online


PURPOSE: At teaching hospitals, house-staff performs the initial assessment of patients considered for admission to a Medical Intensive Care Unit (MICU). Chest X-ray (CXR) interpretation is integral in evaluating these patients and can influence triage decisions. Disagreement between in-training physicians over CXR interpretations renders their risk stratification unreliable. We studied the inter-fellow agreement of CXR interpretation.

METHODS: Two Pulmonary fellows independently read 145 CXR of patients considered for admission to the MICU. Each CXR was divided into 4 quadrants by a standard technique. Each reader documented the percentage of each quadrant with infiltrate, amount of pleural effusion in each hemithorax, the presence of cardiomegaly and blood flow redistribution. Agreement was assessed by the Altman-Bland method, the coefficient of concordance, and Cohen's Kappa.

RESULTS: The reading of the right lower quadrant had the highest absolute bias (3.8%) and the right upper lobe had the lowest absolute bias (0.9%). The total lung bias was 1.1%. Both right and left effusion biases were equal (1%). The limits of agreement were widest for the left lower quadrant (2SD =30%) and narrowest for the left upper quadrant (2SD=20.6%). The total lung limit of agreement was 14.2%. The limits of agreement were equal for left and right side pleural effusions. The difference and the magnitude were correlated only for pleural effusions. The coefficient of concordance was highest for the left side pleural effusion (0.1) and lowest for the right upper quadrant (0.005). The Kappa was 0.66 for cardiomegaly and 0.52 for blood flow redistribution.

CONCLUSION: The limits of agreement between Pulmonary fellows are wide and their concordance is low.

CLINICAL IMPLICATIONS: The wide variations in CXR readings may lead to wide variation in triage. A standard for reading CXRs is needed to standardize risk stratification to optimize resource allocation for patient care.

DISCLOSURE: Amy Guralnick, No Financial Disclosure Information; No Product/Research Disclosure Information

Wednesday, October 24, 2007

12:30 PM - 2:00 PM




Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543