What causes the Silhouette Sign in the thorax on chest radiography?
Let me know in the comments.
Click to reveal the answer
Answer: The silhouette sign is the loss of a normal border on a chest X-ray – the heart, aorta or diaphragm – because an opacity of soft-tissue or fluid density now lies against it. A border is only visible where air-filled lung meets soft tissue. Replace that lung with pus, fluid, collapsed lung or tumour and the contrast disappears, so the edge disappears too.
Its real value is localisation. Each border is touched by a specific lobe or segment, so the border you lose tells you where the disease is, from a single frontal film. Lose the right heart border and the disease is anterior, in the right middle lobe. Keep the right heart border while the right lower zone is white, and the disease is posterior, in the right lower lobe.
Key facts for practice and radiology board exams
| At a glance | |
|---|---|
| What it is | Loss of a normal soft-tissue/air interface (heart, aorta, diaphragm, paraspinal line) on a radiograph |
| Why it happens | Adjacent lung becomes opaque, so there is no longer a density difference to create an edge |
| What it tells you | Where the opacity is – anterior vs posterior, and which lobe or segment |
| What it does not tell you | What the opacity is. Pneumonia, collapse, tumour, effusion and a mediastinal mass all do it |
| Classic pairs | Right heart border → right middle lobe. Left heart border → lingula. Hemidiaphragm → lower lobe. Aortic knob → left upper lobe (apicoposterior segment) |
| Commonest false positive | Pectus excavatum blurring the right heart border |
| Commonest false negative | Small, segmental or ground-glass opacities, and anything seen only on the lateral view |
| Named by | Benjamin Felson and Henry Felson, Radiology, 1950 |
Why is it called so?
A silhouette is an outline. On a normal chest X-ray, the heart, aorta and diaphragm are seen only as outlines – their shapes are drawn by the air in the lung around them. When the lung next to one of those outlines fills with material of the same density, that part of the outline vanishes. The sign is named for the lost outline, not for the lesion. That is why the more accurate name is the loss of silhouette sign.
The term was coined by Benjamin Felson and Henry Felson in their 1950 paper[1], which set out the rule: an intrathoracic lesion touching a border of the heart, aorta or diaphragm will obliterate that border, and a lesion not in contact with it will not. The underlying principle had been noted by earlier radiologists (it is often credited to H. Kennon Dunham), but the Felsons turned it into a working localisation tool.
Pathophysiology
You see an edge on a radiograph only when two things are true.
- There is a density difference. Air against soft tissue gives a sharp edge. Soft tissue against soft tissue, or fluid against soft tissue, gives none – they have almost the same X-ray attenuation.
- The X-ray beam hits the interface tangentially. The edge is drawn by the part of the surface that runs parallel to the beam. That is why the right heart border is drawn by the right atrium and the left heart border by the left ventricle.
Consolidation, collapse, a mass or pleural fluid replaces air with soft-tissue density. If that happens in the lung that sits against the tangential part of a border, the border goes.
A 1977 Radiology study by Longuet and colleagues[2] refined Felson’s rule in a way that matters in practice. They showed the sign works because of radiographic contrast and how we perceive it, not because of anatomical contact as such. Two consequences follow:
- A lesion can touch a structure and still leave its border visible if it does not touch the tangential part of the surface.
- A border can look blurred with nothing touching it, if the density difference is reduced for another reason – a depressed sternum, overlying soft tissue, low penetration or poor inspiration.
So treat the sign as strong evidence, not proof. Confirm it with the lateral view or with a second sign before you name a lobe.
Alternative names: Loss of silhouette sign, loss of outline sign, Felson’s silhouette sign.
Other associated named signs: Hilum overlay sign, cervicothoracic sign, thoracoabdominal sign, spine sign, incomplete border sign, Luftsichel sign.
Which border localises which lobe?
This is the table to learn. Each row is a border you check on every frontal film.
| Border lost on the frontal film | Structure that forms it | Lung (or space) against it | Anterior or posterior? |
|---|---|---|---|
| Right heart border | Right atrium | Right middle lobe, medial segment | Anterior |
| Right upper mediastinal border / ascending aorta | Superior vena cava, ascending aorta | Right upper lobe, anterior segment | Anterior |
| Right hemidiaphragm | Diaphragm dome | Right lower lobe, basal segments (or subpulmonic effusion) | Posterior / inferior |
| Left heart border | Left ventricle | Lingula, inferior segment | Anterior |
| Aortic knob | Aortic arch | Left upper lobe, apicoposterior segment | Posterior / superior |
| Descending aorta | Descending thoracic aorta | Left lower lobe (superior and posterior segments), or posterior mediastinal mass | Posterior |
| Left hemidiaphragm | Diaphragm dome | Left lower lobe, basal segments | Posterior / inferior |
| Paraspinal line | Pleura against the spine | Posterior lower lobe, posterior mediastinum, paravertebral mass | Posterior |
The four patterns examiners ask about most:
- Right lower zone opacity, right heart border lost → right middle lobe. Heart border and middle lobe are both anterior.
- Right lower zone opacity, right heart border sharp, hemidiaphragm lost → right lower lobe. The lower lobe is posterior and never touches the heart.
- Left heart border lost → lingula (or left upper lobe collapse, which pulls the lingula with it, as in the image above).
- Left hemidiaphragm or descending aorta lost behind the heart → left lower lobe. Look for a triangular retrocardiac density.
Both pairs can coexist. Right middle and lower lobe disease together will efface the right heart border and the right hemidiaphragm, and lingular plus left lower lobe disease will efface the left heart border and the left hemidiaphragm.
The silhouette sign on the lateral chest X-ray
The same principle works on the lateral film, and it rescues the pneumonias the frontal view misses.
| Lateral-view finding | Meaning |
|---|---|
| Anterior part of the left hemidiaphragm lost against the heart | Normal – the heart sits directly on it. Use this to tell the two hemidiaphragms apart |
| Whole length of a hemidiaphragm lost, including posteriorly | Lower lobe consolidation or collapse, or effusion, on that side |
| Inferior vena cava shadow lost | Right lower lobe disease, especially the medial basal segment; also retrocardiac mass or subpulmonic effusion (Foote and Meredith, 1979)[3] |
| Lower thoracic vertebrae do not get darker from top to bottom | Spine sign – posterior lower lobe opacity, often invisible on the frontal view |
| Anterior heart border blurred on the lateral | Right middle lobe or lingula; confirms an equivocal frontal finding |
This matters because a meaningful share of pneumonias are visible on only one projection. In a review of 100 emergency chest radiographs whose pneumonia reports were amended by attending radiologists (Ojutiku 2005), the pneumonia was conspicuous on only one view in 43% of cases, split evenly between frontal and lateral.[8]
How reliable is the silhouette sign?
Better at telling you where than whether – and it is less common than textbooks imply.
| Setting | What the evidence shows |
|---|---|
| Emergency department pneumonia (Ojutiku 2005, 60 confirmed pneumonias) | A silhouette sign was present in only 40%. Increased opacity was present in 100%, loss of vascular markings in 72%, air bronchograms in 37%. 75% of the pneumonias were segmental or smaller |
| Ventilated patients, autopsy-proven pneumonia (Wunderink 1992, n=69) | No radiographic sign exceeded 68% diagnostic efficiency. The silhouette sign was not an independent predictor – only air bronchograms were[7] |
| Intubated patients, portable film vs CT (Tierney 2020, n=67) | Portable radiograph localised disease to the correct lobe in only 62%, against 87% for 9-point lung ultrasound[11] |
| Infants with a poorly defined right heart border (Culham 1981)[4] | CT found pulmonary or pleural disease in 7 of 8; the thymus caused it in 1. A blurred right heart border in the first year of life is usually abnormal |
| Anterior vs posterior localisation (Longuet 1977) | Reliable for separating anterior lesions from posterior or lower lesions |
Practical reading: a clear-cut silhouette sign on a good PA film localises well. Its absence does not exclude pneumonia, and on a supine portable ICU film it localises poorly.
Causes of the silhouette sign
The sign tells you the opacity is against the border. It does not tell you what the opacity is.
| Cause | Clues to tell it apart |
|---|---|
| Consolidation (pneumonia) | Air bronchograms, no volume loss, clinical infection |
| Lobar collapse (atelectasis) | Volume loss: fissure displacement, hilar shift, elevated hemidiaphragm, mediastinal shift, crowded vessels |
| Central obstructing tumour | Collapse with a convex bulge at the hilum (Golden S sign), collapse that does not clear |
| Pleural effusion | Meniscus, blunted costophrenic angle, fluid tracking up the lateral wall; subpulmonic effusion effaces the diaphragm |
| Anterior mediastinal mass | Effaces the heart border but the hilar vessels stay visible through it (hilum overlay sign) |
| Pericardial or cardiophrenic lesion | Pericardial cyst, prominent epicardial fat pad, Morgagni hernia at the right cardiophrenic angle |
| Diaphragmatic or subdiaphragmatic lesion | Hydatid cyst of the diaphragm, eventration, hernia |
Pitfalls: false positives and false negatives
| Pitfall | Why it happens | How to avoid it |
|---|---|---|
| Pectus excavatum (false positive) | The depressed sternum and chest-wall soft tissue lie against the right heart border and reduce contrast. Ward 1989 (72 patients vs 72 controls): straight left heart border, leftward heart displacement and an indistinct right heart border were the most specific signs; three or more signs were 95.5% specific[6] | Look at the lateral view for the depressed sternum. Check for horizontal posterior ribs and steep anterior ribs |
| Epicardial fat pad, obesity | Fat at the cardiophrenic angle softens the heart border | Fat is less dense than consolidation; no air bronchograms, no volume loss |
| Poor inspiration, AP supine, rotation | Crowded bases and a magnified heart blur both lower borders | Count ribs; compare with a prior; do not name a lobe on one poor film |
| Thymus in infants | Normal thymus can efface the right heart border | Still presume disease first – 7 of 8 in Culham’s CT series were abnormal |
| Post-cardiac surgery left base | Left lower lobe atelectasis is very common after CABG (32.5% to 85% across groups in Benjamin 1982, linked to phrenic nerve cooling)[5] | Expected finding early after surgery; report it but do not over-call pneumonia |
| Small or ground-glass opacity (false negative) | Not dense enough to remove the air-soft tissue contrast | Look for loss of vascular markings, the lateral view, the spine sign |
| Lesion not at the tangent (false negative) | Touches the structure but not the part that draws its edge | Use the lateral view; if still unclear, CT |
| Chronic lobar collapse | A shrunken lobe can hide against the spine and leave only indirect signs, such as an apparently absent hilum (Saliba 2025)[12] | Look for secondary signs of volume loss, not just lost borders |
Felson’s other localisation signs
The silhouette sign is one of a family of rules that use the same logic – is the lesion surrounded by air or by soft tissue?
| Sign | Finding | Meaning |
|---|---|---|
| Cervicothoracic sign | Upper margin of a mediastinal mass stays sharp above the clavicles | Mass is posterior (surrounded by lung). A margin lost at the clavicles means an anterior mass entering the neck |
| Thoracoabdominal sign | Lower margin of a mass stays visible below the dome of the diaphragm | Mass is posterior, in the retrocrural space, outlined by posterior lung |
| Hilum overlay sign | Hilar vessels visible through a mass projected over the hilum | Mass is not hilar – it is anterior or posterior mediastinal |
| Incomplete border sign | Lesion with one sharp and one fading border | Extrapulmonary (pleural, chest wall or mediastinal) lesion |
Silhouette sign vs the cardiac silhouette
The phrases sound alike but mean different things. The cardiac silhouette (or cardiomediastinal silhouette) is the normal outline of the heart and mediastinum on a chest X-ray – the shape you assess for size and contour. The silhouette sign is the loss of part of that outline because something of soft-tissue density is lying against it. Reports that say “the cardiomediastinal silhouette is normal” mean the outline is normal in size and shape and every border is sharp.
What to do when you see it
Reporting checklist
- Name the border lost and the lobe or segment it implies (for example, “loss of the right heart border indicates right middle lobe disease”).
- Say whether there is volume loss. Consolidation and collapse have different implications.
- Check the lateral view for the matching opacity, the spine sign and the IVC.
- Look for a cause of obstruction: hilar mass, Golden S sign, endobronchial foreign body in a child.
- If the patient is a smoker or over 50, recommend a follow-up radiograph to confirm clearance.
Follow-up after pneumonia
The British Thoracic Society community-acquired pneumonia guideline advises a repeat chest radiograph at about 6 weeks for patients with persistent symptoms or signs, and for those at higher risk of an underlying malignancy – especially smokers and people over 50.[9] A silhouette sign that persists after treatment, especially with volume loss, should prompt CT to look for a central obstructing lesion.[10]
Frequently asked questions
What is the silhouette sign on a chest X-ray?
It is the loss of a normal border, such as the heart, aorta or diaphragm, because an opacity of soft-tissue or fluid density lies against it. Borders are only visible where air-filled lung meets soft tissue, so replacing that lung with pus, fluid, collapsed lung or tumour removes the edge.
What does loss of the right heart border mean?
Disease in the medial segment of the right middle lobe, which lies against the right atrium. If the right lower zone is opaque but the right heart border stays sharp, the disease is in the right lower lobe instead, which is posterior. Pectus excavatum is the commonest false positive.
Which lobe obscures the left heart border?
The lingula, specifically its inferior segment, which lies against the left ventricle. Left upper lobe collapse also effaces the left heart border because it carries the lingula with it.
What does loss of the hemidiaphragm outline mean?
Disease in the basal segments of the lower lobe on that side, or a subpulmonic effusion. On the lateral view the anterior part of the left hemidiaphragm is normally lost against the heart, which is how the two hemidiaphragms are told apart.
How reliable is the silhouette sign?
It localises well on a good PA film but is present less often than expected. In one emergency series only 40 percent of confirmed pneumonias showed a silhouette sign, and in ventilated patients it did not independently predict autopsy-proven pneumonia. Its absence does not exclude pneumonia.
What is the difference between the silhouette sign and the cardiac silhouette?
The cardiac or cardiomediastinal silhouette is the normal outline of the heart and mediastinum. The silhouette sign is the loss of part of that outline because something of soft-tissue density is lying against it.
Who described the silhouette sign?
Benjamin Felson and Henry Felson named it in Radiology in 1950. Longuet and colleagues showed in 1977 that it depends on radiographic contrast and the tangential X-ray beam rather than on anatomical contact alone.
Related radiology signs
- Air bronchogram sign – the best single radiographic predictor of pneumonia
- Spine sign – the lateral-view counterpart for lower lobe disease
- Luftsichel sign – left upper lobe collapse
- Golden S sign – right upper lobe collapse from a central mass
- Juxtaphrenic peak sign – upper lobe volume loss
- Hilum overlay sign and cervicothoracic sign – mediastinal mass localisation
References
- Felson B, Felson H. Localization of intrathoracic lesions by means of the postero-anterior roentgenogram; the silhouette sign. Radiology. 1950;55(3):363-74. PMID: 14781343.
- Longuet R, Phelan J, Tanous H, Bushong S. Criteria of the silhouette sign. Radiology. 1977;122(3):581-5. PMID: 841093.
- Foote GA, Meredith HC. The silhouette sign and the inferior vena cava. Radiology. 1979;133(3 Pt 1):583-5. PMID: 504636.
- Culham JA. The right heart border in infancy. Radiology. 1981;139(2):381-4. PMID: 7220884.
- Benjamin JJ, Cascade PN, Rubenfire M, Wajszczuk W, Kerin NZ. Left lower lobe atelectasis and consolidation following cardiac surgery: the effect of topical cooling on the phrenic nerve. Radiology. 1982;142(1):11-4. PMID: 6975951.
- Ward CS, Halpin SF, Wilson AG. The posteroanterior chest radiograph in depressed sternum. Clin Radiol. 1989;40(2):139-43. PMID: 2624616.
- Wunderink RG, Woldenberg LS, Zeiss J, Day CM, Ciemins J, Lacher DA. The radiologic diagnosis of autopsy-proven ventilator-associated pneumonia. Chest. 1992;101(2):458-63. PMID: 1735272.
- Ojutiku O, Haramati LB, Rakoff S, Sprayregen S. Radiology residents’ on-call interpretation of chest radiographs for pneumonia. Acad Radiol. 2005;12(5):658-64. PMID: 15866141.
- Lim WS, Baudouin SV, George RC, Hill AT, et al. BTS guidelines for the management of community acquired pneumonia in adults: update 2009. Thorax. 2009;64 Suppl 3:iii1-55. PMID: 19783532.
- Mullett R, Jain A, Kotugodella S, Curtis J. Lobar collapse demystified: the chest radiograph with CT correlation. Postgrad Med J. 2012;88(1040):335-47. PMID: 22282740.
- Tierney DM, Huelster JS, Overgaard JD, et al. Comparative performance of pulmonary ultrasound, chest radiograph, and CT among patients with acute respiratory failure. Crit Care Med. 2020;48(2):151-7. PMID: 31939782.
- Saliba T, Rotzinger D, Tack D. The missing hilum: chronic lobar collapse. Respir Med Case Rep. 2025;58:102333. PMID: 41438902.
Access all radiology signs posted so far: https://radiogyan.com/radiology-signs/
Improve Content And Help Your Colleagues!
Found an error in the post? Please let us know using our contact page and we will update it with due credits!
If you wish to contribute radiological images for this case (or any other article on the website), you can submit them here and you will be duly credited: Submit radiology case