Imaging - chest CT
Choose the protocol before you order
| Protocol | Use | Key point |
|---|---|---|
| Non-contrast | Nodules, ILD, follow-up, screening | Lowest dose, no renal or allergy issue |
| HRCT | Interstitial lung disease | Thin (1 mm) slices, inspiratory + expiratory + prone; expiratory phase detects air trapping = small airways disease; prone excludes dependent atelectasis mimicking fibrosis |
| CTPA | Pulmonary embolism | Timed to the pulmonary arteries |
| CT with contrast (portal venous) | Staging, mediastinum, pleura, infection | Assesses nodes, pleura, chest wall |
| CT bronchial/aortic angiogram | Haemoptysis, aortic disease | Maps bronchial arteries before embolisation |
| Low-dose CT | Lung cancer screening | ~1-2 mSv |
- Ordering "CT chest" without specifying the question gets you the wrong protocol and a repeat scan
Radiation and screening
- Typical doses: CXR ~0.02 mSv; low-dose CT ~1-2 mSv; standard chest CT ~5-7 mSv; CTPA ~3-10 mSv
- Annual Australian background ~1.5-2 mSv
- Australia's National Lung Cancer Screening Program commenced 1 July 2025 - biennial low-dose CT for people aged 50-70 with a >=30 pack-year history who currently smoke or quit within 10 years
- Pulmonary nodules are found in >30% of CTs performed for any reason - the great majority are benign
- CTPA positivity rates in unselected practice are often <10%, indicating over-investigation
Density and why it matters
- Hounsfield units: air -1000, fat ~-50 to -100, water 0, soft tissue 20-60, blood 40-60 (acute), calcium/bone >150
- Windowing determines what you can see
- Lung window - parenchyma, nodules, emphysema
- Mediastinal/soft tissue window - nodes, masses, pleura, vessels
- Bone window - metastases, fractures
- A nodule invisible on mediastinal windows is not "absent" - always look at both
Patterns and what generates them
| Pattern | Mechanism | Typical causes |
|---|---|---|
| Consolidation | Alveolar filling (air bronchograms) | Pneumonia, haemorrhage, oedema, adenocarcinoma, organising pneumonia |
| Ground glass | Partial alveolar filling or interstitial thickening | Infection (PJP), oedema, haemorrhage, early fibrosis, NSIP, hypersensitivity pneumonitis |
| Reticulation + traction bronchiectasis + honeycombing | Established fibrosis | UIP/IPF (basal, subpleural) |
| Tree-in-bud | Bronchiolar filling | Endobronchial spread of TB, MAC, aspiration, bronchiolitis |
| Mosaic attenuation with air trapping | Small airways or vascular disease | Bronchiolitis obliterans, hypersensitivity pneumonitis, CTEPH |
| Crazy paving | Ground glass + septal thickening | PAP, PJP, oedema, haemorrhage |
| Nodules - perilymphatic | Along lymphatics | Sarcoidosis, silicosis, lymphangitis carcinomatosa |
| Nodules - centrilobular | Bronchiolocentric | Hypersensitivity pneumonitis, infection, smoking-related |
| Nodules - random | Haematogenous | Miliary TB, metastases |
Lung collapse - the worked example
- CXR: opacification of the affected field, tracheal and mediastinal shift TOWARDS the collapse, elevated hemidiaphragm, crowded ribs, displaced fissure
- LLL: "sail sign" behind the heart with a double left heart border and loss of the medial left hemidiaphragm
- Contrast a large effusion: opacification with shift AWAY
- CT chest: characterises the causative lesion - endobronchial tumour, mucus plug, extrinsic compression, node - and stages it
- Bronchoscopy: direct assessment of an endobronchial lesion, washings +/- biopsy, and therapeutic clearance of plugs
Nodule management - Fleischner Society
Nodule management - Fleischner Society (incidental nodules, age >=35, no known cancer)
| Solid nodule | Low risk | High risk |
|---|---|---|
| <6 mm | No routine follow-up | Optional CT at 12 months |
| 6-8 mm | CT at 6-12 months, then 18-24 months | CT at 6-12, then 18-24 months |
| >8 mm | CT at 3 months, PET-CT or biopsy | Same |
- Subsolid nodules need longer follow-up (up to 5 years) - part-solid nodules with a solid component >6 mm carry the highest malignancy risk
- *Fleischner criteria do not apply to lung cancer screening (use Lung-RADS), to patients with known malignancy, or to the immunosuppressed*
Pulmonary embolism
- CTPA: filling defect, RV/LV ratio >1 signals RV strain
- V/Q or perfusion scan preferred in: pregnancy, contrast allergy, renal impairment, young women (breast dose)
- Subsegmental PE on modern scanners is of uncertain significance - not every filling defect requires anticoagulation
Limitations and harms
- Contrast: AKI risk (lower than historically believed; hydrate and assess eGFR), allergy, thyrotoxicosis after an iodine load, extravasation
- Cumulative radiation, especially in young patients and with repeated scans
- Overdiagnosis: incidental nodules, adrenal and thyroid incidentalomas, coronary calcium
- CT does not replace the clinical question - a scan ordered without a question generates findings without answers
What a CT changes
- Lung cancer: diagnosis, TNM staging, target for biopsy (CT-guided vs EBUS vs bronchoscopy), response assessment (RECIST)
- ILD: a definite UIP pattern on HRCT can make the diagnosis of IPF without biopsy in the right clinical context - subpleural, basal-predominant reticulation with honeycombing and traction bronchiectasis, no features suggesting an alternative
- Infection: cavitation (TB, S. aureus, anaerobes), halo/reverse halo (invasive aspergillosis), empyema (split pleura sign, lenticular shape, septations) -> drain
- Haemoptysis: locates the bleeding lobe and maps hypertrophied bronchial arteries before embolisation
- Pre-operative: segmental anatomy, fissure completeness, nodal stations for EBUS targeting
- Emphysema: distribution and fissure integrity for endobronchial valve selection
- Pleural disease: distinguishes empyema from lung abscess, identifies nodular or circumferential pleural thickening and diaphragmatic involvement suggesting mesothelioma or malignant effusion
Practical rules
- Give the radiologist the clinical question and the relevant history - the report quality tracks the request quality
- Compare with prior imaging - 2 years of stability in a solid nodule is strong evidence of benignity
- Check renal function and contrast allergy before contrast; metformin need not be withheld at normal eGFR
- In women of childbearing age, exclude pregnancy or choose an alternative
Common indications
- Lung cancer; National Lung Cancer Screening Program eligibility
- Interstitial lung disease - IPF, NSIP, hypersensitivity pneumonitis, CTD-ILD, drug-induced (amiodarone, methotrexate, nitrofurantoin, bleomycin, checkpoint inhibitors)
- Occupational lung disease - asbestosis, silicosis (including accelerated artificial-stone silicosis - a major current Australian problem), coal workers' pneumoconiosis
- Pulmonary embolism and CTEPH
- Bronchiectasis, cystic fibrosis, allergic bronchopulmonary aspergillosis
- Immunocompromised host - PJP, invasive aspergillosis, CMV, nocardia
- Sarcoidosis; lymphangioleiomyomatosis; pulmonary alveolar proteinosis
- Incidental coronary artery calcification, thyroid and adrenal nodules, vertebral fractures
Screening and nodule stability
- Low-dose CT screening reduces lung cancer mortality ~20% (NLST) and ~24% in men at 10 years (NELSON) - the evidence behind the Australian program
- Costs: false positives, invasive procedures for benign disease, overdiagnosis, incidental findings
- A solid nodule stable for 2 years is almost certainly benign; subsolid nodules can grow slowly over 5+ years and need longer surveillance
- Radiological progression in ILD often precedes symptoms and physiology - serial HRCT guides antifibrotic decisions
- Post-infective changes (organising pneumonia, scarring) may persist for months - do not re-image too early and do not over-interpret residual opacity
- Cumulative radiation risk is small per scan but real across a lifetime of repeated imaging - justify each scan and use the lowest appropriate protocol
🔒
11 more sections, plus exam facts
Premium unlocks every note across every specialty, and the full exam fact library behind it.
Get premium access