How is the mediastinum measured on a chest X-ray (Chest Radiograph)?

How to Measure the Mediastinum on Chest X-Ray

The most widely accepted measurement is the short-axis diameter of mediastinal lymph nodes, with >1 cm on transverse imaging considered abnormal, though chest radiography itself is an insensitive and unreliable method for mediastinal assessment. 1

Standard Measurement Technique

For Mediastinal Lymph Nodes (CT-based, not plain radiograph)

• Measure the short-axis diameter on transverse (axial) CT images, with >1 cm considered the threshold for abnormal nodes 1
• Alternative criteria that have been used include:
  • Long-axis diameter >1 cm 1
  • Short-axis diameter >1.5 cm 1
  • Short-axis diameter >1 cm plus central necrosis or capsular disruption 1
  • Short-axis diameter >2 cm regardless of morphology 1
• Normal-sized mediastinal lymph nodes are defined as short-axis diameter <1 cm on transverse CT 1

For Mediastinal Width on Plain Chest Radiograph

• The traditional 8 cm upper limit for mediastinal width on supine chest radiographs is outdated and does not apply in modern trauma settings 2
• Mean mediastinal width is 6.31 cm on CT, but magnification with standard radiographic techniques produces measurements of 8.93-10.07 cm 2
• For posteroanterior (PA) films, left mediastinal width (LMW) >4.95 cm has 90% sensitivity and 90% specificity for pathology 3
• For anteroposterior (AP) films, maximal mediastinal width (MW) >8.65 cm has 72% sensitivity and 80% specificity 3

Critical Limitations of Chest Radiography

Poor Diagnostic Performance

• Chest radiography is insensitive for detecting mediastinal lymph node involvement and mediastinal vascular injury 1, 4
• The positive predictive value of widened mediastinum for aortic injury is <1%, with sensitivity and specificity that are statistically and clinically unreliable 5, 4
• Mediastinal width measurements have unacceptable sensitivity for predicting aortic injury, with substantial inter-reader variability (kappa = 0.49 for "widening") 6
• Individual radiologists' sensitivity for detecting aortic injury varies from 0.50-0.97, with specificity varying from 0.62-0.94 6

When Plain Radiography May Be Sufficient

• Bulky lymphadenopathy in superior or contralateral mediastinal areas may be adequate evidence of metastatic disease on plain film 1
• This applies particularly when the patient is too ill or unwilling to undergo further treatment 1
• Tissue confirmation should still be obtained by the least invasive method available 1

Recommended Imaging Algorithm

First-Line Imaging

• Contrast-enhanced CT chest with thin-section imaging (≤5 mm slices) is the initial study of choice for suspected mediastinal masses or pathology 7, 8
• Pre- and post-contrast imaging is essential to distinguish vascular structures from lymph nodes 7, 8
• IV contrast helps delineate mediastinal invasion by centrally located tumors 1

Advanced Imaging for Indeterminate Findings

• MRI should be performed when CT findings are indeterminate, as it provides superior tissue characterization and can prevent unnecessary biopsies 1, 7, 8
• MRI is superior to CT for detecting invasion across tissue planes and involvement of neurovascular structures 1
• Diffusion-weighted imaging (DWI) assists in lesion characterization and can direct biopsy toward areas of higher cellularity 1, 7

Common Pitfalls to Avoid

• Do not rely on the 8 cm mediastinal width threshold from older literature—it does not apply to modern supine radiography 2
• Do not use size criteria alone for lymph node assessment, as nodes >1 cm have limited sensitivity (median 55%) and specificity (median 81%) 1
• Do not assume a widened mediastinum indicates aortic injury—most patients with widened mediastinum have either no findings or nontraumatic findings 5
• PA chest radiographs are significantly more accurate than AP radiographs for mediastinal assessment 3
• Chest radiography usage in trauma has declined from 75% in 2005 to 25% in 2019, replaced by whole-body CT (increased from 35% to 80%) 4