Diagnostic

The Right Paratracheal Stripe 1 Radiology

Maj. Charles J. Savoca, M.C., U.S.A., 2 John H. M. Austin, M.D.,

and Henry I. Goldberg, M.D.

The right paratracheal stripe (RPS) is seen on postero-anterior chest radiographs as a thin,
water-density stripe between the air column of the trachea and the adjacent right lung. The
range of width of the RPS in 1,259 normal subjects was 1 to 4 mm. An RPS width of 5 mm
or more is reliable evidence of disease. The differential diagnosis of this finding is lengthy,
and best divided into those diseases affecting the (a) trachea; (b) mediastinum; and (c)
pleura.

INDEX TERMS: Mediastinum. Pleura. Pneumonitis, desquamative interstitial. (Thoracic

structures, anatomical detail, 6[8].920). (Thoracic structures, other normal variant, 6[8].139)
• Trachea

Radiology 122:295-301, February 1977

THIN STRIPE of water density is usually visible ration with the patient in the erect position at a focus-film
A between the tracheal air column and the adjacent
right lung on postero-anterior chest radiographs (Fig. 1).
distance of 183 cm (72 in.). The width of the RPS was
measured at a level 2 cm above the superior extent of the
The term "paratrachealline" has been ambiguously used
to describe either the entire density (3) or the pleural re-
flection which forms its lateral margin (1). The term' 'right
paratracheal stripe" (RPS) therefore appears to be more
descriptively accurate and less ambiguous.
Observation of widening of the RPS can be useful di-
agnostically. In our experience it has occasionally been
the earliest or only radiographic evidence of disease. Figley
(3) described the RPS as "uniformly thin {2-3 mm)," but
to our knowledge no previous study of its width has been
reported. This study was therefore undertaken to establish
the normal width of the RPS and the significance of its
widening.
A clear understanding of the anatomy of the right par-
atracheal region is needed to prevent misidentification of
the RPS. The mediastinum is deeply indented by the lung
(6, 7). The medial edge of the right lung at the level of the
fifth thoracic vertebra forms a tangent to the x-ray beam
in five planes, thus forming five lines on the radiograph
(Figs. 2 and 3). From posterior to anterior, these lines are
(a) the paraspinous line; (b) the posterior mediastinal
junction line; (c) the lateral margin of the RPS; (d) the lateral
margin of the superior vena cava; and (e) the anterior
mediastinal junction line. The lateral margin of the RPS
must be distinguished from the paraspinous line and the
superior vena cava in order to prevent a false interpretation
of widening of the RPS.

PART I: NORMAL VALUES

MA TERIALS AND METHOD Fig. 1. The open arrows outline the RPS at a level 2 cm above
the superior extent of the azygos arch. Arrows 1 cm above and 1
Postero-anterior chest radiographs of 1,259 normal cm below indicate the range within which the width of the RPS was
subjects were evaluated; all were obtained at end-inspi- measured.

1 From the Departments of Radiology, 97th U.S. Army Hospital, Frankfurt, Germany; University of California School of Medicine, San Francisco,
Calif.; and Columbia-Presbyterian Medical Center, New York, N. Y. Presented at the Sixty-First Scientific Assembly and Annual Meeting of the
Radiological Society of North America, Chicago, 111., Nov. 30-Dec. 5, 1975.
2 Currently NIH Trainee in Diagnostic Radiology, supported in part by National Institutes of Health Training Grant GM 01272 from the National
Institute of General Medical Sciences. ss

295
296 CHARLES J. SAVOCA AND OTHERS February 1977

ANTERIOR
Fig. 2. A cross-section at the level of the fifth thoracic vertebra. Arrows indicate where the edge of the right lung is tangent
to the postero-anterior x-ray beam, and thus forms lines on the frontal radiograph: (1) the paraspinous line; (2) the posterior
mediastinal junction line; (3) the lateral margin of the RPS; (4) the lateral margin of the superior vena cava; and (5) the anterior
mediastinal junction line.
Fig. 3. The numbered arrows indicate the lines on the radiograph which are formed by the edge of the right lung, as shown
in Figure 2: (1) the paraspinous line; (2) the posterior mediastinal junction line; (3) the lateral margin of the RPS; (4) the lateral
margin of the superior vena cava; and (5) the anterior mediastinal junction line.

azygos arch. If no azygos arch was seen, the width was unmeasurable. Measurements were determined to the
measured at a level 2 cm above the superior margin of the nearest 0.5 mm.
origin of the right main bronchus. The width was measured Five groups of subjects were studied: Group I: Male
as closely as possible to these levels, and always within soldiers (1,000/1,029 were measurable), ages 18-29,
1 cm above or below them (Fig. 1). If the width could not having annual or discharge medical examinations at the
be measured within this range, the case was considered 97th U.S. Army General Hospital, Frankfurt, Germany,
between December 1971 and March 1972. All had no
evidence of chest disease as determined by medical
histories, physical examinations, and postero-anterior
chest radiographs; Group 1/: 50 female soldiers (all
measurable), ages 18-29, with no evidence of intratho-
racic disease on postero-anterior chest radiographs; Group
11/: 62 older men (50 measurable), ages 61-83, with no
evidence of intrathoracic disease on postero-anterior chest
radiographs; Group IV: 68 older women (50 measurable),
ages 60-80, with no evidence of intrathoracic disease on
postero-anterior chest radiographs; and Group V: 50
pregnant women (all measurable), ages 17-36, with no
ANTERIOR
Fig. 4. Cross-sectional diagram of
evidence of intrathoracic disease on posteroanterior chest
the tissues of the RPS (arrows). The tissue radiographs.
planes measured include: all of the layers
of the trachea, mediastinal connective RESULTS
tissue, the parietal pleura. the pleural
space, and the visceral pleura. The widths of the RPS in all measurable subjects are
 Diagnostic
Vol. 122 RIGHT PARATRACHEAL STRIPE 297 Radiology

given in TABLE I. The range was 1 to 4 mm. The width of than that for the younger subjects. We have no explanation
the RPS was measurable in 94% (1,200/1,259) of all for this finding other than to speculate that perhaps the RPS
subjects. It was more readily measurable in the young adult increases in width with increasing age from accumulation
subjects (97 % in Groups I, II, and V) than in the older of mediastinal adipose tissue or slightly enlarged para-
subjects (77 % in Groups III and IV). The mean width of the tracheal lymph nodes due to old inflammation.
RPS was slightly greater in the older subjects (Groups III Mild degrees of patient rotation do not appear to sig-
and IV) than in the younger subjects (Group I) (p < nificantly affect the measurement; the data were tightly
0.0005). grouped despite the fact that many of the radiographs were
Pregnancy had no effect on the width of the RPS. This obtained with the subjects in mild degrees of rotation. All
finding differs from measurement of the width of the azygos the data were obtained from postero-anterior and upright
arch, which is affected by the cardiovascular changes of radiographs exposed with a six-foot focus-film distance;
pregnancy (4, 5). care must be taken in extrapolating these results to ante-
roposterior examinations.
DISCUSSION
This study of normal subjects indicates that the width PART II: WIDENING OF THE RPS
of the RPSfulfills one of the criteria for a useful quantitative
MA TERIALS AND METHOD
test: that the normal range be tightly grouped. The range
was 1 to 4 mm for all 1,200 measurable subjects. The cases which form the basis of the second part of
Although the width of the RPS was never greater than our study were collected prospectively from 1972 to 1974.
4 mm in any of these normal subjects, it is of interest that Every case seen by the authors in which the width of the
the mean width for the older subjects was slightly greater RPS was 5 mm or more was evaluated to determine the

Fig. 5. Radiograph of a 45-year-old man complaining of dyspnea. A. The upper portion of the RPS is 5 mm
wide (arrows). Because the normal range was established at a lower level, the significance of this possible
increase in width was uncertain.
B. Two months later, this width had increased to 9 mm. There were no other abnormal findings. A tracheal
carcinoma was found at this level.
298 CHARLES J. SAVOCA AND OTHERS February 1977

Fig. 6. A 49-year-old woman with dyspnea had painful lesions in the cartilage of her nose and ears,
arthralgias, sensory-neural hearing loss, and a history of anterior uveitis. A. The plain radiograph shows
a narrow trachea and a 6 mm-wide RPS.
B. A tomogram better delineates the tracheal narrowing. The diagnosis was relapsing polychondri-
tis.

Fig. 7. Radiograph of an asymptomatic 26-year-old male soldier. Sarcoidosis was suspected because of an RPS width of 7 mm and possible
left paratracheal adenopathy. The lungs appear normal. Scalene node biopsy showed noncaseating granulomas of the sarcoidosis type.
Fig. 8. Radiograph of an asymptomatic 23-year-old male soldier. A reticulonodular pattern is present throughout the lungs. The wide RPS
indicates adenopathy, making sarcoidosis likely. Scalene node biopsy showed noncaseating granulomas of the sarcoidosis type.
 Diagnostic
Vol. 122 RIGHT PARATRACHEAL STRIPE 299 Radiology

Fig. 9. A. Radiograph of a 51-year-old man with cavitating right upper lung bacteroides pneumonia. The RPS is thickened.
B. A decubitus radiograph shows diffuse thickening of the pleural shadow surrounding the RUL pneumonia. No free fluid is present. Because
the pleural shadows are thick lateral, superior, and inferior to the affected lobe, inflammatory pleural thickening medially is the most likely cause
of the widening of the RPS, although lymphadenopathy is also a tenable explanation.

cause of the widening. In the majority the cause for the authors over this period. Measurements were made ac-
widening was obvious, such as recent surgery, diffuse cording to the method described in Part I.
metastatic disease, or massive pleural effusion. Such
cases were numerous, and no attempt was made to record RESULTS
their exact number. Whenever the cause of the widening
of the RPS was uncertain, the case was followed until a The results were twofold. First, a great many diseases
distinct cause of the widening was established by means were found to widen the RPS. Second, there were no false
of either radiological, laboratory, or histologic findings. positive diagnoses: an RPS width of 5 mm or more was
Over 200 cases of widening of the RPS were seen by the always associated with disease.

Fig. 10. Thickening of the RPS from pleural effusion in a 53-year-old woman with polycystic kidneys and renal failure. A. The RPS measures
5 mm. There is no other evidence of effusion on this upright radiograph.
B. A right lateral decubitus examination shows a right pleural effusion. The effusion causes thickening of the paratracheal stripe, which narrows
when the patient is in the decubitus position.
300 CHARLES J. SAVQCA AND OTHERS February 1977

Table I: Frequency of measured widths of the right paratracheal stripe in selected groups of normal subjects
Number of Subjects with Measured Widths of:
Unmea- 3.5 4.0 Mean Widths
Group surable 1.0 mm 1.5mm 2.0 mm 2.5 mm 3.0 mm mm mm ±SD
I
1,029 men 29 76 295 421 145 53 7 3 1.9 ± 0.5 mm
age 18-29 yrs.
II
50 women 0 4 14 29 2 1 0 0 1.8 ± 0.4 mm
age 18-29 yrs.
III
62 men 12 0 3 9 19 11 5 3 2.7 ± 0.6 mm
age 61-83 yrs.
IV
68 women 18 2 6 9 16 8 6 3 2.5 ± 0.8 mm
age 60-80 yrs.
V
50 pregnant women 0 11 16 16 6 1 0 0 1.7 ± 0.5 mm
age 17-36 yrs.

DIFFERENTIAL DIAGNOSIS abnormalities which we have observed to cause Widening

of the RPSare amyloidosis, post-traumatic stenosis of the
The RPS is composed of 3 tissue planes: (a) the tra- trachea, and tracheopathia osteoplastica.
chea, including all of its layers; (b) the mediastinal con- Mediastinal Connective Tissue and Contents: The
nective tissue and its contents, especially lymph nodes; mediastinal connective tissue and its contents form an
and (c) the pleura and the pleural space (Fig. 4). Each of important component of the RPS. This segment of medi-
these tissues must be considered in the evaluation of astinal connective tissue is normally quite thin as viewed
possible causes of RPS widening. on a postero-anterior radiograph, so even small increases
Trachea: Abnormalities of any of the layers of the in the width of this tissue are easily appreciated as wid-
trachea, l.e., mucosa, submucosa, or cartilage rings, can ening of the RPS. Mediastinal abnormalities which we have
widen the RPS. Thickening of mucosal origin from a tra- observed to cause widening of the RPS are mediastinitis,
cheal carcinoma is shown in Figure 5. Thickening of the intrathoracic goiter, hemorrhage, and postsurgical changes
cartilage wall of the trachea due to relapsing polychondritis from mediastinoscopy, cardiac surgery, and right radical
is shown in Figure 6. In this rare disease, tracheal and neck dissection.
bronchial cartilage dissolves and is replaced by an over- The contents of the mediastinal connective tissue which
proliferation of fibrous connective tissue, which often most often cause widening of the RPS are lymph nodes.
narrows the lumens of the major airways (2). Other tracheal Widening of the RPS due to lymphadenopathy is a partie-

Fig. 11. Radiograph of a 28-year-old woman with recurrent pleuritic chest pain and dyspnea on exertion. A. The linear basilar densities
were thought to represent discoid atelectasis from a viral pleuritis.
B. Five months later, the basilar densities have increased, and the RPS has widened to 5 mm. Because of the RPS widening the radiologist
suggested a thorough medical workup, which eventually led to the diagnosis by open lung biopsy of desquamative interstitial pneumonitis.
 Diagnostic
Vol. 122 RIGHT PARATRACHEAL STRIPE 301 Radiology

ularly useful sign in sarcoidosis, where it may be the only is important. In the patient eventually proved to have
radiographic abnormality (Fig. 7). In addition, in cases of desquamative interstitial pneumonitis, for example, the
sarcoidosis with diffuse lung involvement, widening of the finding of widening of the RPS was considered evidence
RPS may be the only sign of lymphadenopathy, and can of progressive disease, and led to an extensive medical
thus serve to narrow the radiological differential diagnosis evaluation which eventually yielded the diagnosis.
(Fig. 8). In a patient with known sarcoidosis of the para-
tracheal nodes, changes in the width of the RPS can be CONCLUSIONS
followed serially as a sign of activity. In a patient with
The range of the width of the RPS in normal subjects is
possible carcinoma of the lung, the width of the RPSshould
1 to 4 mm. An RPS of 5 mm or more is reliable evidence
be carefully evaluated as a sign of metastatic disease in
of disease. The differential diagnosis is lengthy. The value
the mediastinal lymph nodes.
of the finding is therefore in early diagnosis, or in following
Other causes of lymphadenopathy which we have ob-
change in a patient with known disease. A wide RPS can
served to widen the RPS are pneumonia, cystic fibrosis,
be the first or only radiographic sign of disease, and in such
primary tuberculosis (8), coccidioidomycosis leukemia,
cases it should be considered an indication for further
lymphoma, metastases from carcinoma of the lung, and
evaluation. A check for widening of the RPS should be
metastases from a distant source. Disease involving other
included in the routine evaluation of all chest radio-
contents of the mediastinal connective tissue, e.g., neu-
graphs.
rofibromatosis of the vagus nerve, can also cause widening
Department of Radiology
of the RPS. University of California
Pleura: Diseases which cause thickening of the parietal San Francisco, Calif. 94143
or visceral pleura, or which cause an increase in pleural
REFERENCES
fluid, can widen the RPS. Right upper lobe infections can
cause widening from adenopathy and also from localized 1. Berne AS, Gerle RD, Mitchell GE: The mediastinum: normal
pleural thickening or effusion (Fig. 9). In addition, free roentgen anatomy and radiologic technics. Semin Roentgenol 4:3-21,
pleural effusion often widens the RPS. In a case of poly- Jan 1969
cystic kidneys with renal failure, widening of the RPS was 2. Dolan DL, Lemmon GB, Teitelbaum SL: Relapsing polychon-
dritis. Am J Med 41:285-299, Aug 1966
the only abnormality evident on the routine chest radio- 3. Figley MM: Mediastinal minutiae. Semin Roentgenol 4:22-32,
graphs (Fig. 10). Decubitus views demonstrated that pleural Jan 1969
effusion was present. In our hospitalized patients, pleural 4. Fleischner FG, Udis SW: Dilatation of the azygos vein: a
effusion is by far the commonest association with widening roentgen sign of venous engorgement. Am J Roentgenol 67:569-575,
Apr 1952
of the RPS. It is presumably due to localized effusion, al- 5. Keats TE, Lipscomb GE, Betts CS, III: Mensuration of the arch
though pleural or mediastinal edema may contribute. The of the azygos vein and its application to the study of cardiopulmonary
widening invariably disappears with successful treatment disease. Radiology 9:990-994, May 1968
6. Knutsson F: The mediastinal pleura. Acta Radiol 43 (Fasc 4):
of the effusion. 265-275, Apr 1955
Unknown Location: The tissue component causing 7. Lachman E: A comparison of the posterior boundaries of the
widening of the RPS cannot always be inferred, as in right lungs and pleura as demonstrated on the cadaver and on the roent-
upper lobe atelectasis, radiation fibrosis, polyarteritis genogram of the living. Anat Rec 83:521-542,25 Aug 1942
8. Morgan H, Ellis K: Superior mediastinal masses: secondary
nodosa, Wegener's granulomatosis, or desquamative in- to tuberculous lymphadenitis in the adult. Am J Roentgenol 120:
terstitial pneumonitis (Fig. 11). Nevertheless, the finding 893-897, Apr 1974