Pulmonary Pseudoneoplasms: Data Sources.-Literature and Personal Review of Cases
Pulmonary Pseudoneoplasms: Data Sources.-Literature and Personal Review of Cases
Pulmonary Pseudoneoplasms: Data Sources.-Literature and Personal Review of Cases
● Context.—Not uncommonly, a surgical pathologist will Data Sources.—Literature and personal review of cases
be requested to review excised material, with a clinical with focus on inflammatory pseudotumors of the lung, or-
diagnosis of cancer, in which no malignancy can be iden- ganizing pneumonia, nodular lymphoid hyperplasia, apical
tified. Often, sampling may be the issue. However, differ- cap, round atelectasis, and sclerosing mediastinitis with its
ent nonneoplastic processes may mimic cancer clinically pulmonary counterpart, hyalinizing granuloma.
and not be recognized histologically. These are commonly Conclusions.—When reviewing specimens that appear
referred to as pseudoneoplasms and can involve the lung, nondiagnostic for malignancy, it is important to consider
pleura, and mediastinum. one of these pseudoneoplasms in the differential diagnosis
Objective.—To review the most commonly encountered as they may explain the clinical and radiologic information.
pseudoneoplasms of the thoracic cavity in surgical pathol- (Arch Pathol Lab Med. 2010;134:417–426)
ogy and discuss the main differential diagnosis.
418 Arch Pathol Lab Med—Vol 134, March 2010 Pulmonary Pseudoneoplasms—Yi & Aubry
IPT occurring in the pediatric age group is mostly a neo- et al19 reported ALK expression in 5 of 61 IPT cases in-
plasm with frequent ALK positivity and might properly volving various sites (8.2%). All of their positive cases in-
be referred to as IMT, IPT in adults may include more cluded patients younger than 40 years (range, 0.5 to 37
nonneoplastic processes with much less frequent ALK years), with the rate for ALK positivity becoming 21.7%
positivity.3 Zen et al4 have also suggested that there is a only when those patients in their cohort who were 40
subset of IPTs that represents an IgG4-related disease. years or younger were considered. Another study on ALK
Some IPT cases may be related to an infection or may expression in IMT in various sites20 reported that 44 of 73
represent postinfectious sequelae, as some studies,5,6 but cases (60%) were ALK positive, with 6 of 13 pulmonary
not all,7–9 have found an association with human herpes- IMT cases (46%). As mentioned earlier, it is not clear
virus or Epstein-Barr virus. Inflammatory pseudotumor in whether these series encompassed only neoplastic or both
the lower urogenital tract has been postulated to be a re- neoplastic and nonneoplastic-appearing subsets when the
parative process with a morphologic resemblance to nod- term IPT or IMT was used. However, it seems quite clear
ular fasciitis.10 However, several recent studies have also that not all neoplastic IMTs stain positive for ALK.
reported neoplastic-appearing, ALK-1–positive IMTs in- Sensitivity and specificity of ALK immunoreactivity are
volving the urinary bladder,11,12 again raising the possi- limited for IMT with the differential diagnosis of other
bility that some IPTs are reactive and some are neoplastic soft tissue tumors. A study on ALK expression in 135 cas-
IMTs. Meis et al13 documented a tumor, closely simulating es including 10 cases of IMT and 125 cases of its mesen-
IPT in the mesentery and retroperitoneum, under the chymal mimics reported the following results21: cytoplas-
name of inflammatory sarcoma on the basis of its behavior mic ALK positivity in IMT (4 of 10; 40%), malignant pe-
as a low-grade malignancy. Coffin et al14 later discussed ripheral nerve sheath tumor (4 of 10; 40%), rhabdomyo-
the relationship between IMT and inflammatory fibrosar- sarcoma (6 of 31; 19%), leiomyosarcoma (1 of 10; 10%),
coma with a historic review of these entities. Kutok et al6 and malignant fibrous histiocytoma (1 of 11; 9%). Alveolar
reported that IPT of the lymph node and spleen is a bio- rhabdomyosarcomas (4 of 16; 25%) displayed a distinctive
logically distinct entity from IMT and based their conclu- dot-like cytoplasmic positivity.21 No staining was ob-
sion on the absence of ALK expression in their 13 cases served in nodular fasciitis, desmoids, infantile myofibro-
(9 involving lymph nodes and 4 involving splenic lesions). matosis, infantile fibrosarcoma, synovial sarcoma, leio-
In a recent study on IMT of the central nervous system myoma, or myofibrosarcoma.21 Another study22 also re-
(CNS),9 5 of 6 IMTs in the CNS, characterized by neoplas- ported ALK expression in a wide variety of non-IMT soft
tic-appearing spindle cell proliferation, were positive for tissue tumors, although with low-level expression in most
ALK protein overexpression and/or 2p23 rearrangement, non-IMT mesenchymal tumors as compared to the diffuse
while 18 ‘‘nonneoplastic-appearing’’ IPTs were negative positivity in IMT. Therefore, ALK positivity is still useful
for both tests. The authors emphasized that the neoplastic in differentiating IMT from its mimickers, which are part
IMTs in CNS should be distinguished from nonneoplastic of the differential diagnosis in a practical setting. The pos-
IPTs for proper clinical management.9 sible prognostic significance of ALK expression in IMT
In 1995, Snyder et al2 first reported clonal changes in has been explored and concluded as being unclear at this
IPT of the lung involving 2p23, the encoding site for ALK time.23
gene, and a subsequent report by Griffin et al15 demon- While morphologic findings of neoplastic IMT are rel-
strated recurrent involvement of 2p23 in IMTs of various atively distinctive (Figure 1, a), a wide range of histopath-
body sites, suggesting the role of ALK outside of its pre- ologic features can be seen in the broad category of IPT
viously well recognized realm of lymphomas. Indeed, (Figure 1, b), reflecting its heterogeneous nature. The pul-
these observations were followed by the discovery of monary IMTs with neoplastic characteristics typically
TPM3-ALK and TPM4-ALK fusion oncogenes in IMTs that form a well-demarcated but nonencapsulated, usually sol-
transform, in vivo, both mesenchymal and lymphoid hu- itary, mass that replaces the underlying lung tissue. On
man cell lineages, as well as clathrin heavy chain gene, the other hand, IPTs with nonneoplastic features have ill-
CLTC. 16,17 Yousem et al18 reported that 3 of 9 pulmonary defined or irregular outlines without obliterating the un-
IPTs showed involvement of 2p23 by fluorescence in situ derlying alveolar parenchyma completely. Inflammatory
hybridization; in 1 case, null-cell type ALK-1–positive an- infiltrates in nonneoplastic IPTs are proportionally more
aplastic large cell lymphoma was present in the cervical prominent. The central portion of these lesions tends to
lymph node 2 years before the discovery of lung mass. show abundant plasma cells and lymphocytes, cavitation,
In IMT, ALK immunopositivity is a highly sensitive and and less conspicuous proliferation of fibroblasts and myo-
specific indicator of a 2p23 abnormality and is thus a use- fibroblasts.
ful surrogate for molecular or cytogenetic testing. Overall Matsubara et al24 and Gal et al25 have proposed mor-
sensitivity of ALK in IPT is approximately 50%, although phologic subtypes including organizing pneumonia vari-
it varies greatly depending on the age of patients. Chan ant, fibrohistiocytic variant, and plasma cell granuloma/
←
Figure 2. a, Organizing pneumonia. A well-delineated nodular lesion composed of alveolar-filling fibroblastic plugs characteristic of organizing
pneumonia (hematoxylin-eosin, original magnification ⫻40). b, Computerized tomography scan of the chest shows a spiculated opacity in the left
lower lobe, suggestive of malignancy. The wedge resection of this lesion showed only focal organizing pneumonia without any evidence of
malignancy.
Figure 3. a, Nodular lymphoid hyperplasia. An excised nodular lesion is composed of numerous germinal centers with fibrosis. b, Prominent
lymphoid follicles and plasma cells within the nodular hyperplasia. The reactive nature of these infiltrates should be proved by a careful evaluation
with immunophenotypic study as well as by histologic examination to rule out malignant lymphoma, which would more likely obliterate the
underlying alveolar parenchyma, interlobular septa, and pulmonary vessels (hematoxylin-eosin, original magnifications ⫻40 [a] and ⫻200 [b]).
Arch Pathol Lab Med—Vol 134, March 2010 Pulmonary Pseudoneoplasms—Yi & Aubry 419
Table 1. Reported Terminology for Clinicopathologic are usually negative for desmin, myogenin, myoglobin,
Spectrum of Inflammatory Pseudotumor (IPT) CD117, and S100 protein, in keeping with a myofibro-
blastic immunophenotype. ALK stains positively in ap-
Nonneoplastic IPT variants
Plasma cell granuloma type proximately 40% of IMTs, mostly from patients younger
Lymphoplasmacytic or plasma cell type than 40 years of age. ALK positivity is much lower in the
Organizing pneumonia type broader category of IPT. Protein p53 positivity is rare and
IgG4-related is reported in association with recurrence and malignant
Neoplastic IPT (IMT) transformation.27 Plasma cells are polyclonal.
Fibrous histiocytoma
Differential diagnoses for IMT or IPT include a variety
Histiocytoma
Myxoma of benign and malignant spindle cell neoplasms (espe-
Inflammatory fibrosarcoma cially malignant fibrous histiocytoma and inflammatory
Plasma cell granuloma sarcomatoid carcinoma), infection, and nonspecific fi-
Inflammatory fibromyxoid tumor broinflammatory changes. The most important radio-
Abbreviation: IMT, inflammatory myofibroblastic tumor. graphic differential diagnosis in adults with IPT is pri-
mary lung cancer. Inevitably, the distinction of IPT cases
with nonneoplastic features from a variety of nonspecific
inflammatory myofibroblastic variant for their IPT cases reactive conditions could be difficult and may even be ar-
that appeared to represent both neoplastic and nonneo- bitrary. Similarly, the distinction between IMT and inflam-
plastic conditions. Coffin et al14 have used similar but matory fibrosarcoma can be difficult, a fact highlighted by
slightly different terms to describe the microscopic pat- their designation as synonyms in the current World Health
terns of IPT (probably also encompassing both neoplastic Organization classification of the soft tissue tumor.
and nonneoplastic processes): fibrohistiocytic, plasma cell At present, surgery is the principal treatment, although
granuloma, largely sclerosed or fibrosed, compact spindle cell pat- nonsteroidal anti-inflammatory drugs, anti–TNF-␣ anti-
tern, hypocellular fibrous, and myxoid/vascular pattern. Their body, corticosteroids, and chemotherapy have been re-
morphologic descriptions seem to indicate that the term ported in a few cases.28 Inflammatory pseudotumors as-
fibrous histiocytoma type or histiocytoma type corresponds to sociated with IgG4-related lung disease respond to corti-
the neoplastic subtype. One could postulate that the mor- costeroid therapy.
phologic features described as ‘‘plasma cell granuloma
type,’’ ‘‘organizing pneumonia type,’’ or ‘‘lymphoplas- ORGANIZING PNEUMONIA
macytic type’’ are likely to represent a nonneoplastic pro- Organizing pneumonia (OP) occasionally forms a mass-
cess on the basis of the morphologic descriptions for these like lesion that can masquerade as a tumor and may result
lesions. However, it is difficult to determine the true na- in a surgical resection. This presentation commonly causes
ture of these morphologic types. It should be also noted concern for malignancy and leads to a surgical resection
that Dr. Liebow used the term plasma cell granuloma for of the lesion, especially when a needle biopsy is appar-
both neoplastic and nonneoplastic-appearing groups. A ently not satisfactory to explain the nodule found on ra-
recent study4 reported the presence of IgG4-positive plas- diologic studies. In this setting, OP may be cryptogenic or
ma cells in plasma cell granuloma type of IPTs of the lung. secondary to a number of other causes, including infec-
This study purported that the clinicopathologic similari- tion, radiation treatment, underlying malignancy, or drug
ties between IPT of the lung and sclerosing pancreatitis reaction.29–31 Histologically, OP consists of proliferating fi-
suggest that IgG4-related autoimmune processes might be broblasts forming organized plugs within the air spaces
involved in the pathogenesis of some pulmonary IPTs. It (Figure 2, a). This is often accompanied by varying de-
is especially important because IgG4-related disease is grees of interstitial chronic inflammation and type II
sensitive to corticosteroid therapy. These findings de- pneumocyte hyperplasia.
scribed in the literature are summarized in Table 1. Organizing pneumonia presenting as a solitary nodule
Previous studies have provided detailed morphologic by radiologic imaging (Figure 2, b) often is cause, first, for
descriptions of IPT or IMT. Mitotic rate in IMT or IPT is a computed tomography (CT)–guided needle biopsy. If
low and a myxoid stroma may be prominent.3 Some le- the needle biopsy shows OP on histologic examination, the
sions contain prominent fibroxanthomatous features with patient may still undergo surgical biopsy to rule out ma-
prominent endogenous lipid. Multinucleated giant cells lignancy. A recent study at Mayo Clinic32 reported the
including Touton-type cells are seen in some cases but clinical and radiologic manifestations of 26 cases diag-
granulomas and neutrophils are usually inconspicuous or nosed by surgical lung biopsy as OP from 1997 to 2004.
absent.24 Inflammatory pseudotumor can invade incorpo- All patients presented with a unifocal opacity detected on
rated or contiguous small vessels, particularly veins. In- chest radiography or CT scans. At the time of presentation,
vasion of pulmonary vessels distant from the main mass only 10 patients (38%) had symptoms, including cough,
also occurs, but this is rare. A purported case of IPT with shortness of breath, or chest pain. Contrast-enhancement
an extension into large pulmonary veins and pericardium CT scan or positron emission tomography scanning was
led to a patient’s death.26 Endobronchial involvement as a performed for 11 patients and the results were positive in
polypoid lesion occurs in about 10% of IPT cases with or all. Surgical procedures included wedge resection for 21
without extension to the bronchial wall.24 Mediastinal or patients (81%), segmentectomy for 3 patients (11%), and
hilar involvement, pleural penetration, and multifocal le- lobectomy for 2 patients (8%). Three cases of focal OP
sions have been described.24 In IPTs, intralesional calcifi- were related to infections, but the remaining cases were
cation occurs in about 15% of cases and focal necrosis or cryptogenic. Follow-up over a median interval of 11
hemorrhage may be present.24 months (range, 1 to 71 months) yielded no recurrence of
Spindle cells of IPT are uniformly positive for vimentin OP.
and also for smooth muscle actin in many cases, but they The radiologic features of focal OP are often indistin-
420 Arch Pathol Lab Med—Vol 134, March 2010 Pulmonary Pseudoneoplasms—Yi & Aubry
guishable from those of lung cancer and include positivity The most important differential diagnosis for NLH is
on contrast-enhancement CT scan and positron emission mucosa-associated lymphoid tissue (MALT) lymphoma
tomography scanning. Most cases of focal OP are cryp- but it also needs to be differentiated from lymphocytic
togenic and infection is identified in a minority of cases. interstitial pneumonia, lymphomatoid granulomatosis,
Occult aspiration may also contribute to some of these and IPT or IMT. Nodular lymphoid hyperplasia differs
cases. Surgical resection alone appears to suffice in the from lymphocytic interstitial pneumonia in that it is focal
management of focal cryptogenic OP. rather than diffuse. In contrast to IPT or IMT, spindle cells
in the background are absent in NLH. Unlike lympho-
NODULAR LYMPHOID HYPERPLASIA matoid granulomatosis, markedly atypical B cells are ab-
Pulmonary nodular lymphoid hyperplasia (NLH), a term sent in NLH, which will be negative for Epstein-Barr virus
first coined by Kradin and Mark,33 refers to 1 or more by either immunohistochemistry or in situ hybridization.
nodules or localized pulmonary infiltrates consisting of a The distinction from MALT lymphoma is particularly dif-
reactive lymphoid proliferation. The concept of ‘‘pseudo- ficult if not impossible without molecular studies. Any
lymphoma’’ for reactive localized masses of lymphoid tis- significant degree of lymphangitic pattern of infiltrative
sue in the lung was initially proposed by Saltzstein34 in activity would favor lymphoma. Helpful findings in favor
his review of pulmonary lymphoid lesions showing low of MALT lymphoma include diffuse infiltrative architec-
histologic grade, the presence of numerous lymphoid fol- ture with invasion of pleura and bronchial cartilage, prom-
licles, and a benign clinical course. Most recently, Abbo- inent monocytoid B cells, centrocyte-like (cleaved) atypical
danzo et al35 reported a clinicopathologic study on 14 cas- lymphocytes, intranuclear inclusions (Dutcher bodies),
es of NLH by applying more modern immunophenotypic prominent lymphoepithelial lesions, and light chain re-
and molecular methods to exclude a clonal or neoplastic striction (demonstrable in approximately 40% of cases). It
process. should be noted that some cases remain unclear even after
It has been now well recognized that most localized extensive immunophenotypic and molecular studies. It is
lymphoid lesions arising in the lung are marginal zone B- appropriate to label such cases as an ‘‘atypical lymphoid
cell lymphoma. However, there are some cases that have proliferation’’ if there are sufficient atypical histopatholog-
proved to be polyclonal and reactive in nature. For those ic features in favor of malignancy.
cases, the term nodular lymphoid hyperplasia is appropriate
and is preferred to pseudolymphoma, as it more accu- APICAL CAP
rately reflects its histogenetic characteristics. Abbodanzo Apical cap, also referred to as apical scar, is a form of
et al35 reported that cases collected at the Armed Forces localized pulmonary fibrosis. Although historically, apical
Institute of Pathology showed patients’ ages ranging from caps were thought to be caused by tuberculosis, most of-
19 to 80 years (mean, 65 years) with a slight female pre- ten the etiology is unknown.36–39 Other rare etiologies in-
dominance (ratio of men to women, 3:4). Ten of 14 cases clude radiation, infection other than tuberculosis, trauma,
(71%) were found incidentally on routine chest radio- and vascular abnormalities.36,37,40 In idiopathic cases, the
graphs but some patients were symptomatic with signs of apical cap is a common incidental radiologic finding, re-
cough, shortness of breath, and pleuritic pain.35 Most le- ported as unilateral in 11.2% and as bilateral in 12.2% of
sions were solitary and subpleural. When the disease was 258 adults undergoing routine chest radiographs.37 A sim-
multifocal (5 cases), it was usually unilateral. Five patients ilar frequency is reported in an old autopsy series.36 The
(36%) also had concomitant hilar, mediastinal, or para- prevalence of apical caps increases with age and is usually
esophageal lymphadenopathy. No recurrence of a pul- found in patients older than 45 years but may be recog-
monary lesion was documented after surgical resection nized in younger individuals.36 The prevalence is similar
and all 7 patients with follow-up information were alive, between men and women. Radiologically, these lesions are
with no evidence of disease in a period from 8 months to described as soft tissue densities, with sharply demarcat-
6 years after surgical excision.35 ed smooth or undulating margins, measuring on average
Grossly, NLH comprises a well-circumscribed, fleshy or of 5 to 6 mm in height. However, radiologic variations of
rubbery nodule (or nodules) usually measuring 2 to 4 cm, apical caps can result in diagnostic challenges, in partic-
but larger lesions have been reported. Histologically, lym- ular with apical carcinomas (Pancoast tumor).38 Therefore,
phoid follicles are typically well developed. Plasma cells apical caps are occasionally biopsied, thus the need, as a
are frequently prominent between the follicles (Figure 3, surgical pathologist, to be aware of their morphologic ap-
a and b). Most cases show at least mild interstitial fibrosis, pearance.
but in some cases it is dense and remarkable. Focal and Although usually present in the upper lobes, apices of
limited lymphangitic spread of lymphocytes and plasma lower lobes can also be involved. The gross and histologic
cells around the bronchovascular bundles and interlobular features of apical caps are very characteristic and descrip-
septa may be seen. Despite the frequent subpleural loca- tions from autopsy and surgical series are very similar.36,39
tion of NLH, plaque-like lymphoid infiltrates in the pleura Grossly, the apical cap is a slightly depressed, gray-white
are rare.35 Large airways are usually not particularly in- opaque plaque lesion, which on cross section appears tri-
volved. Regional lymph nodes biopsied at the same time angular (Figure 4, a). Pleural thickening is usually present,
show reactive follicular hyperplasia. opaque and white, with sharp lateral margins. The un-
Immunohistochemistry shows a mixture of B cells (with derlying lung parenchyma appears gray-white and usu-
polytypic and light chain expression) and T cells. Bcl- ally contains black anthracotic streaks. Histologically, at
2 stain is negative in germinal centers but positive in man- low power, the triangular shape is apparent with the
tle zone cells and T lymphocytes within and between ger- broad base extending along the pleural surface (Figure 4,
minal centers. Immunoglobulin heavy chain gene rear- b). In smokers, the peripheral lung parenchyma often dis-
rangement should be absent, by definition, when molec- plays emphysematous changes with respiratory bronchi-
ular genetic analysis is performed. olitis. The lesion comprises mostly areas of dense fibrosis,
Arch Pathol Lab Med—Vol 134, March 2010 Pulmonary Pseudoneoplasms—Yi & Aubry 421
which appear to have obliterated the air spaces with the
underlying elastic framework intact. However, the elastic
framework is collapsed and fibers are typically increased
and fragmented (Figure 4, c). This most distinctive feature
distinguishes apical caps from other types of pulmonary
scars. Metaplastic ossification or dystrophic calcification of
the area of fibrosis is frequently noted in surgically re-
sected fibrotic caps. Necrosis is not usually present.
Chronic inflammation is usually mild and acute inflam-
matory reactions absent. Granulomas are an uncommon
finding, seen in only 15% (7 of 48) of autopsy cases re-
ported in 1970 and in a single case (1 of 13) from a recent
surgical series.36,39 When present, they are located in the
adjacent lung parenchyma and the cap itself is not in-
volved. These findings further support causes other than
tuberculosis as the etiology. Moderate to severe vascular
abnormalities, characterized by mural thickening and old
fibrotic thrombosis with recanalization of arteries and mu-
ral sclerosis of pulmonary veins, located in the immediate
adjacent lung parenchyma, are another distinctive feature,
seen in more than half of cases. Based on this observation,
along with the apical location, the leading hypothesis for
the pathogenesis of apical caps is ischemic injury. Another
significant feature is the pneumocyte hyperplasia present
at the edge of the cap or within entrapped residual air
spaces, which can be misinterpreted as carcinoma asso-
ciated with a scar. Pneumocyte hyperplasia tends to be
pleomorphic and nonuniform in contrast to well-differ-
entiated adenocarcinoma, in particular bronchioloalveolar
carcinoma, in which the cell atypia is monotonous and
regular. Nuclear overlap with loss of polarity and in-
creased nuclear cytoplasmic ratio are other features favor-
ing malignancy.41 Other differential diagnoses include or-
ganizing pneumonia (discussed above), arterial infarct,
and silicotic nodule. Infarcts share some histologic fea-
tures with apical cap lesions, notably a pleural-based lo-
cation, triangular shape, and the presence of arterial vas-
cular abnormalities. However, infarcts are more common-
ly located in the lower lobes, may be of varying age, and
will show varying amounts of hemorrhage and organi-
zation with proliferating fibroblasts. Furthermore, the elas-
tic framework is often not well preserved in an infarct, in
contrast to the prominent elastosis of an apical cap. Sili-
cotic nodules are typically spherical and well-demarcated
and comprise concentric collagen fibers with interspersed
dust-laden macrophages.
ROUND ATELECTASIS
Round or rounded atelectasis is known by many other
terms including folded lung, shrinking pleuritis with atelec-
tasis, atelectatic pseudotumor, and pleuroma, a reflection of its
appearance, location, and presumed pathogenesis.42–47 Pa-
tients with round atelectasis are predominantly men
(⬎90%) and are often smokers, with a mean age of 60
years (range, 20 to 83 years).42–47 Patients are usually
Figure 4. Apical cap. a, Cross section of gross lung specimen showing asymptomatic, with incidental discovery of a mass on rou-
the typical triangular-shaped scar located beneath slightly thickened
pleura. The scar is gray-white with focal areas of anthracosis. b, Low-
tine chest radiograph. When symptoms occur, they most
power photomicrograph of an apical scar. The pleural thickening is often include cough, dyspnea, and fever with fatigue or
mild, dense, and eosinophilic, which contrasts with the basophilic ap- weight loss. Rarely, hemoptysis and chest pain have been
pearance of the triangular pulmonary scar. c, High-power photomicro- reported. Although most cases have been attributed to as-
graph highlights the composition of the apical scar with collapse of the bestos exposure, other potential etiologies comprise tu-
pulmonary elastic framework and increase in elastic fibers, which are berculosis, Dressler syndrome, cardiac failure, uremia,
usually fragmented (hematoxylin-eosin [b and c], original magnifica-
tions ⫻20 [b] and ⫻400 [c]).
and trauma.42–47 Occasionally, a specific etiology may not
be elicited from the clinical history.
Two main theories for the formation of round atelectasis
422 Arch Pathol Lab Med—Vol 134, March 2010 Pulmonary Pseudoneoplasms—Yi & Aubry
have been proposed.48 The original theory suggests that
round atelectasis is the result of the mechanical influence
of a pleural effusion on the lung, with compression and
folding of the lung and fixation in this position by a pleu-
ral fibrinous exudate; as the effusion resolves, the normal
adjacent lung hyperexpands to fill the space and thus en-
gulfs the atelectatic lung. The second theory proposes a
fibrosing mechanism by which a primary localized in-
flammation of the pleura becomes progressively fibrotic
and contracts, leading to folding of the pleura and col-
lapse of the underlying lung.
Both theories could explain the typical radiologic find-
ings. On plain chest radiograph, round atelectasis is de-
scribed as a rounded mass, 2.5 to 5 cm in dimension, al-
though lesions up to 10 cm have been reported.44 The most
common location is the posterior surface of the lower lobe
(approximately 75% of the lesions). It is pleural based and
associated with pleural thickening. The most characteristic
and specific feature (reported specificity of 92%) is the
curvilinear displacement of the vessels and bronchi to-
wards the center of the mass described as the ‘‘comet tail’’
sign. Chest CT with contrast usually confirms the diag-
nosis and precludes the need for further treatment. How-
ever, the comet tail sign is not always present (reported
sensitivity of 83%), and a number of variants have been
reported with CT; therefore, these lesions may appear
clinically suggestive of lung cancer or mesothelioma, par-
ticularly in older men who are smokers and with a history
of exposure to asbestos.49,50
Morphologic features are distinctive. Grossly, the visceral
pleura appears thickened and fibrotic, and sections of the
underlying lung reveal no apparent parenchymal abnor-
mality that would correlate with the radiologic findings.
Prominent folding of the pleura with invagination into the
underlying lung is seen underneath the area of fibrosis.42,46 Figure 5. Round atelectasis. a, Intermediate-power photomicrograph
Histologically, the pleural fibrosis is superficial to the elastic of round atelectasis showing the superficial pleura uniformly thickened
and interstitial layer of the pleura, which is of normal thick- by fibrosis, while the elastic layer of the pleura shows duplication with
ness but characterized by prominent wrinkling and infold- prominent wrinkling. b, At low power, the pleura displays deep infold-
ing (Figure 5, a).42,46 The infolding is usually deep, resulting ing with superimposition of the wrinkled pleura. The underlying lung
parenchyma is collapsed with no other abnormality (hematoxylin-eo-
in superimposition of the wrinkled pleura (Figure 5, b). sin, original magnifications ⫻100 [a] and ⫻40 [b]).
These distinctive features of round atelectasis separate it
from simple pleural fibrosis. The fibrosis is mostly acellular
with mild chronic inflammation. The appearance of the un- contiguous involvement of the mediastinum. Some pa-
derlying parenchyma is variable; it often appears collapsed tients may manifest both lesions concomitantly and may
but otherwise unremarkable. Focal nonspecific fibrosis and also have features of other fibrosing disorders, such as ret-
sclerosis of vessels can be seen. Asbestos bodies can occa- roperitoneal fibrosis.
sionally be identified. Some authors have suggested that
round atelectasis is part of a spectrum of pleural fibrotic Sclerosing Mediastinitis
diseases, from localized pleural fibrosis without associated Sclerosing mediastinitis is a rare condition thought to
round atelectasis to diffuse pleural fibrosis, both of which result from many etiologies, an abnormal immunologic re-
may be associated with pulmonary atelectasis, including in- sponse to histoplasmosis being the most prominent one in
volvement of an entire lobe.51,52 the United States (Table 2).53–59 It affects predominantly
When observed, round atelectasis remains stable al- young adults (mean age varies between 30 and 45 years)
though cases of spontaneous regression, as well as pro- and men slightly more than women.53–59 Distribution of the
gression, have been reported. In patients undergoing sur- fibrosis and varying degrees of compression of the me-
gical excision, no further lesions develop on follow-up.42–47 diastinal and hilar structures determine the type and se-
verity of the clinical presentation (Table 3). Venous infarcts
SCLEROSING (FIBROSING) MEDIASTINITIS AND have received much attention, as they can be the first man-
HYALINIZING GRANULOMA ifestation of sclerosing mediastinitis and present in a sur-
Sclerosing mediastinitis and hyalinizing granuloma gical lung biopsy of a patient with localized infiltrates.60–63
share similar histologic features and are both thought to Radiologically, sclerosing mediastinitis is divided be-
be immunologically derived. Sclerosing mediastinitis in- tween 2 types: focal, the most common, and diffuse.55,56
volves predominantly the mediastinum and hilar regions The focal type presents mainly as a localized and calcified
and can extend to the lung parenchyma, while hyalinizing mass in the paratracheal or subcarinal compartments of
granuloma occurs within the lung parenchyma without the mediastinum or in the pulmonary hilum. The diffuse
Arch Pathol Lab Med—Vol 134, March 2010 Pulmonary Pseudoneoplasms—Yi & Aubry 423
Table 2. Major Causes of Sclerosing
(Fibrosing) Mediastinitis
Infectious
Histoplasmosis
Other fungal infections (aspergillosis, cryptococcosis, blasto-
mycosis, mucormycosis)
Tuberculosis
Noninfectious
Autoimmune conditions (eg, IgG4-related immuno-
pathologic process)
Sarcoidosis
Radiation
Drugs (eg, methysergide)
Familial
Idiopathic
Arch Pathol Lab Med—Vol 134, March 2010 Pulmonary Pseudoneoplasms—Yi & Aubry 425
25. Gal AA, Koss MN, McCarthy WF, Hochholzer L. Prognostic factors in pul- 46. Menzies R, Fraser R. Round atelectasis: pathologic and pathogenetic fea-
monary fibrohistiocytic lesions. Cancer. 1994;73:1817–1824. tures. Am J Surg Pathol. 1987;11:674–681.
26. Warter A, Satge D, Roeslin N. Angioinvasive plasma cell granulomas of 47. Hillerdal G. Rounded atelectasis: clinical experience with 74 patients.
the lung. Cancer. 1987;59:435–443. Chest. 1989;95:836–841.
27. Ledet SC, Brown RW, Cagle PT. p53 immunostaining in the differentiation 48. Dernevik L, Gatzinsky P. Pathogenesis of shrinking pleuritis with atelec-
of inflammatory pseudotumor from sarcoma involving the lung. Mod Pathol. tasis—‘‘rounded atelectasis’’. Eur J Respir Dis. 1987;71:244–249.
1995;8:282–286. 49. O’Donovan PB, Schenk M, Lim K, Obuchowski N, Stoller JK. Evaluation
28. Applebaum H, Kieran MW, Cripe TP, et al. The rationale for nonsteroidal of the reliability of computed tomographic criteria used in the diagnosis of round
anti-inflammatory drug therapy for inflammatory myofibroblastic tumors: a Chil- atelectasis. J Thorac Imaging. 1997;12:54–58.
dren’s Oncology Group study [discusson in J Pediatr Surg. 2005;40:1003]. J Pe- 50. Batra P, Brown K, Hayashi K, Mori M. Rounded atelectasis. J Thorac Im-
diatr Surg. 2005;40:999–1003. aging. 1996;11:187–197.
29. Myers JL, Colby TV. Pathologic manifestations of bronchiolitis, constrictive 51. Cohen AM, Crass JR, Chung-Park M, Tomashefski JF Jr. Rounded atelectasis
bronchiolitis, cryptogenic organizing pneumonia, and diffuse panbronchiolitis. and fibrotic pleural disease: the pathologic continuum. J Thorac Imaging. 1993;
Clin Chest Med. 1993;14:611–622. 8:309–312.
30. Colby TV. Bronchiolitis: pathologic considerations. Am J Clin Pathol. 1998; 52. Chung-Park M, Tomashefski JF Jr, Cohen AM, el-Gazzar M, Cotes EE.
109:101–109. Shrinking pleuritis with lobar atelectasis, a morphologic variant of ‘‘round atel-
31. American Thoracic Society/European Respiratory Society International ectasis’’. Hum Pathol. 1989;20:382–387.
Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumo- 53. Mathisen DJ, Grillo HC. Clinical manifestation of mediastinal fibrosis and
histoplasmosis [discussion in Ann Thorac Surg. 1992;54:1057–1058]. Ann Thorac
nias. This joint statement of the American Thoracic Society (ATS), and the Euro-
Surg. 1992;54:1053–1057.
pean Respiratory Society (ERS) was adopted by the ATS board of directors, June
54. Mole TM, Glover J, Sheppard MN. Sclerosing mediastinitis: a report on 18
2001 and by the ERS Executive Committee, June 2001. Am J Respir Crit Care
cases. Thorax. 1995;50:280–283.
Med. 2002;165:277–304.
55. Devaraj A, Griffin N, Nicholson AG, Padley SP. Computed tomography
32. Maldonado F, Daniels CE, Hoffman EA, Yi ES, Ryu JH. Focal organizing findings in fibrosing mediastinitis. Clin Radiol. 2007;62:781–786.
pneumonia on surgical lung biopsy: causes, clinicoradiologic features, and out- 56. Rossi SE, McAdams HP, Rosado-de-Christenson ML, Franks TJ, Galvin JR.
comes. Chest. 2007;132:1579–1583. Fibrosing mediastinitis. Radiographics. 2001;21:737–757.
33. Kradin RL, Mark EJ. Benign lymphoid disorders of the lung, with a theory 57. Dunn EJ, Ulicny KS Jr, Wright CB, Gottesman L. Surgical implications of
regarding their development. Hum Pathol. 1983;14:857–867. sclerosing mediastinitis: a report of six cases and review of the literature. Chest.
34. Saltzstein SL. Pulmonary malignant lymphomas and pseudolymphomas: 1990;97:338–346.
classification, therapy, and prognosis. Cancer. 1963;16:928–955. 58. Dozois RR, Bernatz PE, Woolner LB, Andersen HA. Sclerosing mediastinitis
35. Abbondanzo SL, Rush W, Bijwaard KE, Koss MN. Nodular lymphoid hy- involving major bronchi. Mayo Clin Proc. 1968;43:557–569.
perplasia of the lung: a clinicopathologic study of 14 cases. Am J Surg Pathol. 59. Flieder DB, Suster S, Moran CA. Idiopathic fibroinflammatory (fibrosing/
2000;24:587–597. sclerosing) lesions of the mediastinum: a study of 30 cases with emphasis on
36. Butler C II, Kleinerman J. The pulmonary apical cap. Am J Pathol. 1970; morphologic heterogeneity. Mod Pathol. 1999;12:257–264.
60:205–216. 60. Katzenstein AL, Mazur MT. Pulmonary infarct: an unusual manifestation
37. McLoud TC, Isler RJ, Novelline RA, Putman CE, Simeone J, Stark P. The of fibrosing mediastinitis. Chest. 1980;77:521–524.
apical cap. AJR Am J Roentgenol. 1981;137:299–306. 61. Dye TE, Saab SB, Almond CH, Watson L. Sclerosing mediastinitis with
38. Im JG, Webb WR, Han MC, Park JH. Apical opacity associated with pul- occlusion of pulmonary veins: manifestations and management. J Thorac Car-
monary tuberculosis: high-resolution CT findings. Radiology. 1991;178:727–731. diovasc Surg. 1977;74:137–141.
39. Yousem SA. Pulmonary apical cap: a distinctive but poorly recognized 62. Berry DF, Buccigrossi D, Peabody J, Peterson KL, Moser KM. Pulmonary
lesion in pulmonary surgical pathology [comment in Am J Surg Pathol. 2001; vascular occlusion and fibrosing mediastinitis. Chest. 1986;89:296–301.
25(10):1344]. Am J Surg Pathol. 2001;25:679–683. 63. Williamson WA, Tronic BS, Levitan N, Webb-Johnson DC, Shahian DM,
40. Solovay J, Solovay HU. Apical pulmonary tumors—relation to apical scar- Ellis FH Jr. Pulmonary venous infarction. Chest. 1992;102:937–940.
64. Inoue M, Nose N, Nishikawa H, Takahashi M, Zen Y, Kawaguchi M. Suc-
ring. Dis Chest. 1965;48:20–28.
cessful treatment of sclerosing mediastinitis with a high serum IgG4 level. Gen
41. Colby TV. Malignancies in the lung and pleura mimicking benign pro-
Thorac Cardiovasc Surg. 2007;55:431–433.
cesses. Semin Diagn Pathol. 1995;12:30–44. 65. Crotty TB, Colby TV, Gay PC, Pisani RJ. Desmoplastic malignant mesothe-
42. Dernevik L, Gatzinsky P, Hultman E, Selin K, William-Olsson G, Zettergren lioma masquerading as sclerosing mediastinitis: a diagnostic dilemma. Hum
L. Shrinking pleuritis with atelectasis. Thorax. 1982;37:252–258. Pathol. 1992;23:79–82.
43. Stark P. Round atelectasis: another pulmonary pseudotumor. Am Rev Res- 66. Flannery MT, Espino M, Altus P, Messina J, Wallach PM. Hodgkin’s disease
pir Dis. 1982;125:248–250. masquerading as sclerosing mediastinitis. South Med J. 1994;87:921–923.
44. Schneider HJ, Felson B, Gonzalez LL. Rounded atelectasis. AJR Am J 67. Engleman P, Liebow AA, Gmelich J, Friedman PJ. Pulmonary hyalinizing
Roentgenol. 1980;134:225–232. granuloma. Am Rev Respir Dis. 1977;115:997–1008.
45. Mintzer RA, Cugell DW. The association of asbestos-induced pleural dis- 68. Yousem SA, Hochholzer L. Pulmonary hyalinizing granuloma. Am J Clin
ease and rounded atelectasis. Chest. 1982;81:457–460. Pathol. 1987;87:1–6.
426 Arch Pathol Lab Med—Vol 134, March 2010 Pulmonary Pseudoneoplasms—Yi & Aubry