CXR eponyms in pulmonary embolism

The classic chest x-ray eponymous terms of pulmonary embolism are often presented as isolated signs. In reality, they describe different vascular and parenchymal consequences of the same disease process. Proximal pulmonary arterial enlargement, distal vascular depletion, pulmonary infarction and secondary atelectasis. Their historical development also overlaps considerably, and several modern definitions have drifted from the original descriptions.

This review traces the development of the Westermark, Hampton, Fleischner, Chang and Palla eponyms, separates overlapping terminology, and examines what these signs mean in the modern CTPA era.

PathophysiologyCXR findingsEponym
Proximal pulmonary arterial enlargementProminent/dilated central or proximal PA related to obstruction and acute pulmonary hypertensionFleischner sign
Transient descending PA enlargementNew enlargement of a previously normal right or left descending PA, potentially reversibleChang sign
Right descending PA shape distortionProximal enlargement with abrupt distal tapering producing the characteristic “sausage” morphologyPalla sign
Distal vascular depletionRegional oligaemia with attenuated/absent peripheral vascular markingsWestermark sign
Peripheral pulmonary infarctionPleural-based haemorrhagic infarction with a characteristic convex inward marginHampton hump
Secondary plate atelectasisLinear/discoid atelectasis; an indirect and non-specific manifestationFleischner lines

Related non-eponymous finding: the knuckle sign describes abrupt pulmonary arterial termination and should not be conflated with Fleischner, Chang or Palla signs.

Historical timeline

1910 – Otto von Dehn published Über röntgenologische Lungenbefunde im Vergleich zu den Ergebnissen der Sektion. His study comparing chest radiographic appearances with postmortem findings was an early report of radiographic findings in pulmonary thrombosis. Dehn noted stark ausgesprochene Hilusschatten (marked hilar shadows) in a patient with thrombosis of both pulmonary arteries and almost no hilar lymph-node enlargement at autopsy.

1931 – Clarence H. Boswell and Harold D. Palmer published Progressive thrombosis of pulmonary artery. Their case provided a strong radiographic–pathological correlation. The chest radiograph showed an increased density of the shadow of the right pulmonary artery noting that its significance was not recognized before death. An abnormal hilar/pulmonary arterial shadow was retrospectively correlated with an obstructing thrombus.

1936 – Felix George Fleischner published three papers investigating pulmonary collapse, plate-like atelectasis and basal horizontal pulmonary shadows. These establish the basis of the radiographic findings later termed Fleischner lines.

1938 – Westermark sign: distal pulmonary oligaemia

1938 – Nils Westermark published On the roentgen diagnosis of lung embolism, based on 26 patients with pulmonary embolism subsequently confirmed at autopsy and a further 18 clinical cases followed radiographically during evolution and recovery.

What he described: embolism without infarction produced a relatively hyperlucent pulmonary territory with diminished or absent peripheral vascular markings. Central vessels remained visible but could terminate abruptly at the edge of the oligaemic territory.

Modern eponym: Westermark sign: regional pulmonary oligaemia distal to embolic arterial obstruction.

What has been distorted: the original observation was regional vascular depletion rather than hyperlucency alone. The autopsy series comprised 26 patients, not 28.

1940 – Hampton hump and pulmonary infarction

1940 – Aubrey Otis Hampton and Benjamin Castleman published Correlation of postmortem chest teleroentgenograms with autopsy findings with special reference to pulmonary embolism and infarction. Their study correlated postmortem radiographs with detailed gross and microscopic lung pathology and provided one of the foundational descriptions of the evolution and radiographic morphology of pulmonary infarction.

What they described: peripheral pleural-based infarcts with variable morphology and a characteristically convex medial/cardiac margin.

Modern eponym: Hampton hump

What has been distorted: “wedge-shaped opacity” is useful shorthand, but Hampton and Castleman actually described rounded, hemispherical and truncated-cone lesions; the hump came from the convex inward margin.

1941 – Fleischner, Aubrey Otis Hampton and Benjamin Castleman published Linear shadows in the lung (interlobar pleuritis, atelectasis and healed infarction). They distinguished plate-like atelectasis from linear shadows caused by pleural disease and healing/healed pulmonary infarction.

1951 – Hanelin and Eyler published Pulmonary Artery Thrombosis: Roentgen Manifestations reviewing five cases involving the main pulmonary artery or major branches. They described dilatation proximal to the obstruction, enlargement or altered contour at the thrombus, and diminished distal vessels producing regional hyperlucency.

1958–1962 — Fleischner sign and the enlarged pulmonary artery

1958 – Felix George Fleischner published Pulmonary embolism. Separates thrombus-related hilar enlargement from dynamic pressure-mediated arterial dilatation and demonstrates serial 11 → 14 → 22 → 15 mm change.

1959 – Fleischner described unilateral oligaemia with compensatory contralateral pleonemia. This was not the original description of the Fleischner sign.

1962 – Fleischner published Pulmonary embolism and provided his synthesis of the roentgen manifestations of PE.

Modern eponym: Fleischner sign = prominent, dilated proximal or central PA.

Separate eponym: Fleischner lines = plate, discoid atelectasis.

What has been distorted: Fleischner did not first discover PA enlargement in 1959. His contribution was largely the haemodynamic explanation and demonstration of dynamic dilatation.

1962–1965 — Chang sign and pulmonary artery measurement

1962 – C. H. Joseph Chang published The normal roentgenographic measurement of the right descending pulmonary artery in 1,085 cases, establishing reference measurements for the vessel on PA chest radiography.

April 1965 – Chang and W. Clayton Davis published A Roentgen sign of Pulmonary Infarction. They studied 23 consecutive patients with pulmonary infarction and found enlargement of a descending pulmonary artery in all cases : 17–22 mm in 19 right-sided infarcts and 17–26 mm in four left-sided infarcts.

December 1965 – Chang published Clinical application of the measurement of the right descending pulmonary artery in the radiological diagnosis of pulmonary hypertensions from various causes. He applied the measurement method to pulmonary hypertension from several causes.

Modern eponym: Chang sign = transient calibre enlargement, not vessel amputation.

What has been distorted: abrupt cutoff was not part of Chang’s original definition.

1972 – Murray G. Baron published Fleischner lines and pulmonary emboli stating that the linear shadows were “commonly referred to as Fleischner lines”. He defined their morphology and emphasised that they were an indirect, non-specific manifestation of PE caused by focal bronchial obstruction and atelectasis rather than by direct visualisation of vascular occlusion.

1983 – Palla sign and right descending pulmonary artery morphology

1983 – Antonio Palla et al. published Enlargement of the right descending pulmonary artery in pulmonary embolism. They reviewed 374 consecutive patients admitted with suspected pulmonary embolism and quantitatively compared right descending pulmonary artery dimensions in patients with and without confirmed PE.

What they described: during acute PE, the proximal right descending pulmonary artery could become enlarged, dense and conspicuous, with relatively abrupt distal tapering

Modern eponym: Palla sign = enlargement and characteristic sausage-shape distortion of the right descending pulmonary artery

What has been distorted: Palla was not the first to describe descending pulmonary arterial enlargement. Palla used Chang’s 1962 normal measurements as a reference standard but did not cite the 1965 Chang–Davis paper. Later authors use Palla sign for almost any enlarged right descending pulmonary artery, blurring its distinction from the earlier Chang sign.

From descriptive signs to diagnostic evidence: 1982–2000

Prospective evaluation has demonstrated that these chest x-ray findings are poor standalone diagnostic tests. In PIOPED, Westermark sign, prominent central PA, Fleischner sign and Hampton hump were all poor independent predictors of PE.

With the development of CTPA, chest radiography became primarily a tool for identifying alternative diagnoses and recognising occasional clues rather than directly diagnosing or excluding PE.

1982 – Greenspan et al published Accuracy of the chest radiograph in diagnosis of pulmonary embolism. They randomised chest radiographs from 152 patients suspected of PE to nine blinded readers. The study demonstrated poor overall diagnostic accuracy of chest radiography for PE and found no meaningful relationship between reader experience and performance.

1993 – Worsley et al. published Chest Radiographic Findings in Patients With Acute Pulmonary Embolism: Observations From the PIOPED Study.. They reviewed CXRs from 1,063 patients with suspected PE. 383 had angiographically confirmed PE and 680 had PE excluded. The paper defines:

  • oligaemia = Westermark sign
  • prominent central pulmonary artery = Fleischner sign
  • pleural-based opacity = Hampton hump

All were poor predictors of PE

2000 – C. Gregory Elliott et al published findings from the International Cooperative Pulmonary Embolism Registry (ICPOER). Among 2,322 patients with CXRs, pulmonary artery enlargement was reported in 19%, compared with cardiomegaly 27%, pleura

Eponymythology
How the names changed the observations

These eponyms did not arise as five discrete diagnostic signs. They were applied retrospectively to observations developed over several decades, and later teaching progressively narrowed or altered their meanings.

  • Westermark’s vascular depletion became “hyperlucency”
  • Hampton’s variable infarct morphology became a “wedge”
  • Fleischner’s haemodynamic observations became a single enlarged artery
  • Chang and Palla became near-synonyms despite describing different phenomena.

The names remain useful shorthand, but the original papers show a much more interconnected radiographic physiology of pulmonary embolism.

References

Original articles

Review articles

eponymythology

the myths behind the names

more eponyms

BA MA (Oxon) MBChB (Edin) FACEM FFSEM. Emergency physician, Sir Charles Gairdner Hospital. Passion for rugby, medical history, medical education, and asynchronous learning #FOAMed evangelist. Co-founder and CTO of Life in the Fast lane | On Call: Principles and Protocols 4e| Eponyms | Books | Horology

One comment

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.