Setting

Hospital Italiano de Buenos Aires (HIBA) is a non-profit healthcare academic tertiary center with 785 beds, of which 200 are dedicated to critical care. The hospital has a busy clinical practice with 3 millions outpatient consultations, 45000 inpatient admissions and 52000 surgeries performed in 41 operating rooms yearly. During 2018, the Imaging Department received 583,455 patients to perform different studies (table 1).

Since 1998, the HIBA has run an in-house-developed health information system, which includes clinical and administrative data. It has been recently certified by the HIMSS as level 7 in the Electronic Medical Record Adoption Model. The HIBA health information Department, is in charge of the design, development, implementation and maintenance of almost all systems, including the Electronic health record.

A.- Task force

A1.- Under the leadership of a medical physicist, a team composed by a CT technologist and a board certificate radiologist specialized in CT imaging conducted the following actions:

B.- Data mining:

B1.- The 20 most-used CT protocols (10 patents each) were selected based on the clinical indications for adults. In the pediatric population we take the 4 main clinical indications in 5 different age groups: 0-1 year old, 1-4 years old, 5-9 years old, 10-14 years old and over 15 years old.

B2.- A table was elaborated with the following information of each protocol: scanned region, clinical indication, kVp, mA min, mA max, mAs, tscan, pitch, FOV, scanned length, noise index or SD, and cut thickness with its corresponding Dose Length Product (DLP) [mGy] and Computed Tomography Dose Index volume (CTDI vol) [mGy].

B3.- Statistics over DLPs and CTDI vol was done in order to determine DRLs for each protocol with its clinical indication.

C.- Data analysis:

C1.- Data analysis was blinded to technologists and radiologists and only performed by medical physicians.

C2.- The data collected for each protocol (DLPs and CTDT vol) related to clinical indication and patient’s age was compared with the European Guidelines on DRLs.

C3.- Those protocols whose DRL exceeded the upper limit of the recommended DRLs following the European Guidelines on DRLs were submitted to a technical evaluation for their optimization.

D.- Dose Optimization:

D1.- Data analysis was blinded to technologists and radiologists and only performed by medical physicists.

D2.- The doses and the technical CT parameters originally settled that resulted beyond the limits of European Guidelines on DRLs were subject to modification. As examples:

-Decrease in mAs

-Pitch increase

-Modification of automatic exposition control noise parameters

D3.- The radiologist checked the images obtained with the new optimized CT protocols for validation. If approved, the new local DRLs were then established and the corresponding protocols definitively modified; if unapproved, up-going modifications continued till optimization and agreement. This ongoing process allowed the development of the Hospital DRLs.

E.- Monitoring:

E1.- Quarterly, a random sample of patients among the 20 most-used CT protocols (10 patents each) are checked to determine the DRLs used during that time frame (manual dose management system).

E2.- Those protocols whose DRL exceeded the upper limit of the local developed CT Hospital DRLs are submitted to a technical evaluation to find out the reasons of radiation over-exposure and act accordingly for optimization.