ProtocolPulse

Paediatric Imaging Guidance

CT/MRI
NICE NG232 (May 2023)RCR/SCoR/RCPCH2017, revised 2018RCR iRefer2017, continuously updatedIR(ME)R 2017SI 2017/1322v2.0 · March 2026
Routine
EDInpatientOutpatient

Radiation Sensitivity in Children

— ALL of the following must be met

  • ALARA (As Low As Reasonably Achievable) principle mandatory for all paediatric exposures — children have up to 10-fold higher lifetime attributable cancer risk compared to adults for the same effective doseIR(ME)R 2017 Reg 11
  • Radiosensitivity is highest in rapidly dividing tissues — thyroid, breast buds, gonads, and bone marrow are particularly vulnerable in children
  • Every paediatric CT must be individually justified with a higher threshold of benefit than for adults, given the increased lifetime riskIR(ME)R 2017
  • Non-ionising alternatives (ultrasound, MRI) must be actively considered before CT in all paediatric casesRCR iRefer

Age-Specific CT Protocols

— ALL of the following must be met

  • IV contrast dose is weight-based: typically 1–2 mL/kg of iodinated contrast (300 mgI/mL), maximum dose should not exceed adult dose
  • Field of view (FOV) must be adapted to body habitus — use the smallest FOV that covers the clinical area of interest to maximise spatial resolution and reduce scatter
  • Reduce kVp for smaller patients: 80 kVp for neonates/infants, 100 kVp for children <40 kg, 120 kVp only for adolescents approaching adult size
  • Use automatic tube current modulation (ATCM) with weight/age-appropriate reference mAs — typical paediatric protocols use 30–70% lower mAs than adult protocols
  • Single-phase acquisition is standard — multi-phase CT is rarely justified in children and must have explicit senior radiologist approval

Non-Accidental Injury (NAI)

— ALL of the following must be met

  • Skeletal survey is the first-line imaging investigation for suspected NAI in children under 2 years — CT should not replace the skeletal surveyRCR/SCoR/RCPCH
  • CT head is mandatory for all children under 1 year with suspected NAI, even without neurological signsRCR/SCoR/RCPCH
  • Follow-up skeletal survey at 11–14 days recommended to identify healing fractures not visible on initial imagingRCR/SCoR/RCPCH
  • Refer to RCR/SCoR/RCPCH 'Radiological Investigation of Suspected Physical Abuse in Children' for full imaging algorithmRCR/SCoR/RCPCH
  • Safeguarding team must be notified before imaging — all NAI imaging requests should be discussed with a consultant paediatric radiologist where available

CT Head — Differences from Adult Criteria

— ALL of the following must be met

  • Lower GCS threshold for immediate CT: GCS <14 in children (vs ≤12 in adults), or GCS <15 in infants under 1 yearNICE NG232 (May 2023)
  • Dangerous mechanism thresholds differ: fall >3 m in children (vs >1 m in adults); high-speed RTA; high-speed projectile impactNICE NG232 (May 2023)
  • Vomiting threshold: ≥3 discrete episodes in children (vs >1 episode in adults)NICE NG232 (May 2023)
  • Tense fontanelle in infants is an additional indication for immediate CT not applicable to adultsNICE NG232 (May 2023)
  • Bruise, swelling, or laceration >5 cm on the head in infants under 1 year — immediate CT indication unique to paediatricsNICE NG232 (May 2023)

MRI as Preferred Modality

— ALL of the following must be met

  • MRI delivers no ionising radiation and is the preferred cross-sectional modality in children where clinically appropriate and timely access is availableRCR iRefer
  • MRI is preferred over CT for non-urgent brain imaging in children (e.g., epilepsy, developmental delay, headache) — CT only where MRI is unavailable or clinically urgent
  • MRI is the modality of choice for paediatric spinal pathology including cord compression, tethered cord, and spinal dysraphism
  • MRI is preferred for musculoskeletal soft tissue assessment, bone marrow pathology, and joint internal derangement in children

Sedation and Anaesthesia for MRI

— ALL of the following must be met

  • Children under 5–6 years typically require sedation or general anaesthesia (GA) for MRI due to inability to remain still for scan duration (20–60 minutes)
  • Feed-and-sleep technique may be adequate for neonates and young infants (under 3–6 months) — avoids sedation/GA risks
  • GA carries small but non-negligible risk — weigh the clinical benefit of MRI against anaesthetic risk, particularly for repeat imaging
  • Sedation/GA for paediatric MRI requires appropriately trained anaesthetic staff, MRI-conditional monitoring equipment, and recovery facilities
  • Standard fasting guidelines apply: 6 hours solids, 4 hours breast milk, 1 hour clear fluids — confirm local policy

Diagnostic Reference Levels (DRLs) for Paediatric CT

— ALL of the following must be met

  • National DRLs for paediatric CT are published by PHE/UKHSA and must be used as benchmarks — departments should audit against these regularly
  • Paediatric DRLs are stratified by weight bands (typically 5 kg, 10 kg, 20 kg, 30 kg, 50 kg) rather than age alone
  • Example paediatric DRLs for CT head: CTDIvol ~30 mGy (neonate), ~40 mGy (1 year), ~50 mGy (5 years) — significantly lower than adult DRL of ~60 mGy
  • If DRL is consistently exceeded, a formal investigation and protocol optimisation is required under IR(ME)R employer's proceduresIR(ME)R 2017

Notes

Warning

Children have up to 10-fold higher lifetime cancer risk from radiation than adults — always consider non-ionising alternatives first

Alert

Do not apply adult CT head criteria to children — paediatric GCS thresholds and mechanism criteria differ significantly

Information

Weight-based protocols are essential — never use adult default CT parameters for paediatric patients

Warning

For suspected NAI, skeletal survey must precede CT — follow RCR/SCoR/RCPCH guidance

Information

MRI is preferred over CT for non-urgent paediatric imaging wherever clinically appropriate

Change Log

v2.02026-03-08New cross-cutting paediatric guidance

Not clinical advice. This protocol is a reference tool only. All imaging justifications remain the clinical and legal responsibility of the authorising practitioner under IR(ME)R 2017 (as amended 2024). Protocol content should be verified against current NICE, RCR, and specialty guidelines before use in practice.

AI-assisted content. Clinical criteria were developed with AI assistance and cross-referenced against cited source guidelines. Verify against original sources. Guidelines referenced are current at the stated version date and may have been updated since.