Computed tomography pulmonary angiography: an assessment of the radiology report

The aim of this study was to evaluate the uncertainty in computed tomographic pulmonary angiography (CTPA) radiology reports, manifested by descriptions of report limitations and image quality.

CTPA reports between 2004 and 2006 were reviewed for patient demographic data (age, gender, pregnancy state), radiologist data (years of experience, subspecialty, final dictation by an attending radiologist vs a resident being present and dictating the report), the presence of pulmonary embolism (PE), and key words describing examination quality and limitations.

There were 2151 CTPA reports. Patterns of reporting CTPA in the impression sections of radiology reports were as follows: (1) PE conclusively positive (10%), (2) PE conclusively negative (29%), (3) PE negative to segmental arteries (27%), (4) PE negative to central pulmonary arteries (21%), (5) PE negative but suboptimal examination (8%), and (6) nondiagnostic examination (5%). Among the last three categories, seven PEs were not initially diagnosed but were found on subsequent imaging examinations. Limitations in image quality, respiratory motion artifact, and contrast enhancement were most frequently mentioned as limitations in image quality (62% and 28% of all reports, respectively). Radiologists tended to report limitations in image quality if they were thoracic radiology subspecialists, had >10 years of experience, or worked independently (P < .001).

Different patterns of reporting CTPA exist and vary on the basis of individual radiologists' subspecialties, experience, and whether they work independently or with residents. Certain wording regarding the presence of PE may falsely imply negativity of PE in a limited examination.