Establishing National Clinical Diagnostic Reference Levels and Achievable Doses for CT examinations in Brazil: A Prospective Study

Saturday, May 6, 2023

First Author(s)

LK

Lina Karout, MD

Research Fellow
Massachusetts General Hsopital
CHELSEA, Massachusetts, United States

Co-Author(s)

MB

Mônica Oliveira Bernardo, MD

Radiologist
Hospital Miguel Soeiro, Pontificia University Catholic of São Paulo, United States
SE

Shadi Ebrahimian, MD

Resident
Icahn School of Medicine at Mount Sinai, Elmhurst Hospital Center, Elmhurst, NY, USA, United States
FM

Flávio Morgado, PhD

Physicist
Pontificia University Catholic of São Paulo, United States
AS

Alair Sarmet Santos, MD

Radiologist
Complexo Hospitalar de Niterói, Niterói, Rio de Janeiro, United States
HS

Henrique Schroeder, MD

Radiologist
Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto da Universidade de São Paulo, Ribeirão Preto, São Paulo, United States
GP

Graciano Paulo, PhD

Professor
Escola Superior de Tecnologia da Saúde do Politécnico de Coimbra, Coimbra, Portugal, United States
FK

Felipe Moura Kiipper, MSc

Medical Physicist
Hospital Sírio Libanes, São Paulo, São Paulo
Sao Paulo, Sao Paulo, Brazil
PC

Paulo Costa, PhD

Professor
IFUSP- Instituto de Física da Universidade de São Paulo, United States
MK

Mannudeep Kalra, MD

Professor
Massachusetts General Hospital, United States

Purpose: We conducted a prospective study on establishing multi-parametric, multi-site national (Brazil) DRL for CT exams based on patient gender, size, body regions, and clinical indications.

Methods/Materials: We prospectively included 4,787 adult patients (18-99 years; male: female 2041:2746) who underwent CT at 13 sites in Brazil between January and October 2021 including routine head (n=1475 patients), chest (n=1210), and abdomen-pelvis [AP] (n=857), head CTA (n=34), CT PE (n=102), cervical spine CT (n=204), paranasal sinuses CT (n=565) and chest abdomen pelvis CT (n=340). We recorded patients’ demographic and anthropometrics, clinical indications, scanner information (vendor, scanner name, slice profile), scan parameters (number of scan phases, kV, mA, pitch) and dose indices (phase-specific CT dose index volume [CTDIvol] and phase-specific and total dose length products [DLP]). Patients were divided into under-, over-, normal- and obese-BMI groups per the WHO classification. Median and 75th percentile CTDIvol and DLP values were estimated for patient groups (gender and size), body region, clinical indications.

Results: The estimated 75th percentile CTDIvol for head CT (45mGy), head CTA (26mGy), chest CT (11mGy), CT-PE (15 mGy), Chest abdomen pelvis CT (13mGy), Abdomen CT (12mGy), Kidney stone CT (13mGy), paranasal sinuses CT (19mGy) and cervical spine CT (23mGy). When compared with USA and UK DRL, Brazil had a lower DRL for head, chest, abdomen CT and chest-abdomen pelvis CT protocols and higher levels in CT-PE and Kidney stone CT protocols. For identical body part, gender, BMI, and clinical indications, there were significant and several-fold differences in both CTDIvol and DLP values across the 13 Brazilian hospitals (p < 0.001). In addition, a major difference in CT protocols and parameters employed for the same clinical indication was noted (p < 0.001).

Conclusions: In this study most Brazilian DRL (CTDIvol and DLP) were lower or similar to the US and European reported DRLs for corresponding CT exams and protocols with variations based on clinical indications and patient size. However, major improvements are necessary in CT-PE and kidney stone CT protocols.