Sara Anne Hayes MBBCh, Presenter: Nothing to Disclose

Micheal Breen MBBCh, MRCPI, Abstract Co-Author: Nothing to Disclose

Patrick Mc Laughlin FFRRCSI, Abstract Co-Author: Nothing to Disclose

Michael Henry, Abstract Co-Author: Nothing to Disclose

Michael M. Maher MD, FRCR, Abstract Co-Author: Nothing to Disclose

Max Frederick Ryan MBBCh, Abstract Co-Author: Nothing to Disclose

To demonstrate the development and successful implementation of an automated PACS-based alert system, in a tertiary referral center, for the communication of unexpected and significant findings on chest radiographs.

Our radiology department utilizes a AGFA IMPAX PACS framework. As part of a quality assurance program, an inbuilt keyword entitled “Unexpected and Significant Clinical Findings” was enabled on PACS. The program represented an attempt to address problems seen in current communication work flows between radiologists and clinicians, by initiating a well-documented communication sequence. When unexpected findings were identified on a radiological study, the radiologist selected the keyword and a streamlined alert system highlighted the case to a Radiology Information System administrator, who contacted the referring clinician by email. We retrospectively reviewed all included chest radiographs and analyzed the referring sources, follow-up investigations and outcomes. An emphasis was placed on the timeline from the initial study to follow-up imaging and final outcome.  

In the 14 months since the system launch, 39,665 chest radiographs were performed at our institution. The keyword was applied to 158 chest radiographs (0.4%). 46.2% (n=73) of referrals were from the emergency room, 30.3% (n=48) were outpatients, 21.5% (n=34) were inpatients and 1.9% (n=3) were family doctor referrals. The mean time for clinician response was 3.1 days (S.D. 9.37). 77.8% of patients had a relevant follow-up study at our institution and the mean time interval to the next relevant radiological investigation was 26 days (S.D 33.98, 95% confidence interval 5.98). On follow-up, 13.2% of patients (n=21) subsequently received a malignant diagnosis. 9.5% (n=15) were new cases of primary lung cancer and 3.8% (n=6) had new metastatic disease from a known primary tumour. 49.4% of patients (n=78) received benign diagnoses and 20.2% (n=32) had no relevant abnormality on follow-up imaging.   

There is a growing need for the development of robust automated systems to facilitate the timely communication of unexpected radiological findings. We hope to share our early experiences of such a system and evaluate its effectiveness and safety. 

Failure to communicate radiological results is estimated to be the fourth most common cause of litigation against radiologists in the United States.

Hayes, S, Breen, M, Mc Laughlin, P, Henry, M, Maher, M, Ryan, M, Communication of Unexpected and Significant Findings on Chest Radiographs with an Automated PACS Alert System.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11012818.html