Clinical indication-based CT scan range reduction: A scoping review of radiation dose savings and diagnostic accuracy

Computed Tomography (CT) is central to modern imaging practice and accounts for over 60% of the collective radiation dose from medical imaging, underscoring the need for optimisation. Adjusting the scan range based on clinical indication is an evidence-based method for dose optimisation. However, the evidence remains dispersed across indications without a concise cross‑domain synthesis; this review consolidates evidence on indication-based scan range optimisation, evaluating its impact on radiation dose, diagnostic accuracy, and implications for missed findings across multiple CT applications.

This scoping review followed Joanna Briggs Institute methodology and PRISMA-ScR guidelines. PubMed, Embase, and Scopus were searched for studies (2010-2025) on indication-driven CT scan range optimisation. Eligible studies evaluated reduced z‑axis coverage based on clinical indications and reported at least one outcome: scan length (mm), radiation dose metrics (mSv), or diagnostic accuracy (%). Data were synthesised narratively with descriptive reporting of key outcomes.

Twenty-one studies across six clinical domains were included. Indication-driven protocols reduced radiation exposure proportionally to scan length, achieving reductions of 33-50% in scan length and 10-50% in dose. Organ-specific reductions reached 97% for breast and 81% for testes, with foetal dose reductions exceeding 80% in pregnancy-adapted protocols. Diagnostic sensitivity for the primary indication remained 98-100%. Aggressive truncation increased missed alternative diagnoses in up to 17% of patients, particularly in older or multimorbid cohorts.

Clinical indication-driven optimisation of CT scan range achieves substantial radiation-dose reduction without compromising diagnostic accuracy. These findings support evidence-based adoption of landmark-based protocols for focused clinical questions, with selective application in patients with broader differential diagnoses. Future directions include AI-assisted landmark detection and automated planning tools to enhance reproducibility and workflow efficiency.