Achilleas Gikas, MD

Diamantis Kofteridis, MD
Demosthenes Bouros, MD
Argiro Voloudaki, MD
Yiannis Tselentis, MD
Nikolaos Tsaparas, MD

Q Fever Pneumonia:
Appearance on Chest
Radiographs1
PURPOSE: To determine the radiographic features of Q fever pneumonia.

Index term:
Lung, infection, 60.2029, 60.211,
60.213
Q fever, 60.2029
Radiology 1999; 210:339343
Abbreviations:
IgG immunoglobulin G
IgM immunoglobulin M
1

From the Departments of Internal

Medicine (A.G., D.K., N.T.), Clinical Bacteriology, Parasitology, and Geographical Medicine (A.G., D.K., Y.T.), Pneumonology (D.B.), and Radiology (A.V.),
University Hospital of Heraklion, 71100
Heraklion, Crete, Greece. Received February 24, 1998; revision requested May
4; revision received June 4; accepted
August 10. Address reprint requests
to A.G.

MATERIALS AND METHODS: Chest radiographs in 85 patients admitted to the

hospital during a 7-year period with Q fever pneumonia were retrospectively
reviewed by two observers.
RESULTS: The most commonly recorded abnormalities were segmental (n 53
[62%]) and lobar (n 15 [18%]) areas of opacity. Segmental pneumonia was
observed as a unilateral single area of opacity in 38 (72%) patients. It was more
frequently located in the upper lobes. The left upper lobe was involved in 31% of
patients; the right upper lobe, in 23%; and the right lower lobe, in 27%. Lobar
pneumonia was less frequently observed as a single lesion in eight (53%) of 15
patients: It was located in the left upper lobe in 31% and in the right middle lobe in
27% of patients. There was no correlation between the extent of pulmonary
involvement and the course of the disease; the outcome was favorable in all patients.
Complete resolution of the radiographic findings occurred in a mean of 39 days.
CONCLUSION: The radiographic differentiation of Q fever pneumonia from the
other community-acquired pneumonias is not possible. Clinical, serologic, and
epidemiologic data provide the best basis for diagnosis.

r RSNA, 1999

Acute Q fever disease is a benign infection with a worldwide distribution. It is due to

infection with Coxiella burnetii, an obligate intracellular parasite living in the phagolysosomes of the host cells. The microorganism is found in the urine, feces, milk, and aborted
fetuses of infected animals (1). Humans most often acquire the disease from inhaled dust or
by ingesting unpasteurized milk or fresh cheese (2).
Derrick (3) described the disease in 1937. It is characterized by a wide diversity of clinical
manifestations: prolonged fever, pneumonia, granulomatous hepatitis, and meningoencephalitis. Reports from several countries indicate that the epidemiologic and clinical
features of the disease may vary from one area of the world to another. Pneumonia (47)
and hepatitis (911) were the predominant clinical manifestations in most published series.
We (12) have recently documented that on the Greek island of Crete, C burnetii is a not
uncommon cause of community-acquired pneumonia. As far as we know from the
international literature, the extent, variations, and evolution of the radiographic findings
of Q fever pneumonia have been rarely reported. The purpose of our study was to
determine the extent, variations, and evolution of the radiographic findings in patients
with Q fever pneumonia.

Author contributions:
Guarantor of integrity of entire study,
A.G.; study concepts and design, A.G.;
definition of intellectual content, A.G.;
literature research, A.G.; clinical studies, A.G., D.K., D.B., A.V., Y.T., N.T.;
data acquisition, D.K.; data analysis,
A.G., D.K., D.B., A.V., Y.T., N.T.; manuscript preparation and editing, A.G.;
manuscript review, A.G., D.K., A.V.

MATERIALS AND METHODS

During 7 years (19891996), serum samples from 3,600 patients suspected of having a C
burnetii infection were assayed for the presence of antibodies against antigen phase II of the
microorganism by using the indirect immunofluorescence antibody technique. Patients
were selected at a retrospective review of records of adults hospitalized for a febrile illness
associated with serologic evidence of acute C burnetii infection.
The determination of acute Q fever infection was based on both serologic and clinical
criteria, as previously described in detail elsewhere (12). These included immunoglobulin G
339

a.

b.

Figure 1. Chest radiographs obtained at the time of hospital admission in a 37-year-old man with a 9-day history of fever, nonproductive cough,
right-sided chest pain 3 days after the onset of symptoms, headache, malaise, and IgM titers against C burnetii of 1:3,200 measured with the indirect
immunofluorescence antibody technique. (a) Posteroanterior view shows lobar pneumonia in the right middle lobe. (b) Lateral view shows some loss
of volume in the consolidated middle lobe, evidenced by a slight upward displacement of the major fissure.

(IgG) antibody titers of 1:960 or greater or

immunoglobulin M (IgM) titers 1:400 or
greater and/or a fourfold increase in the
titer between two consecutive assays, as
well as one or more of the following: fever
(temperature 38C), respiratory disease
(dyspnea, expectoration, cough, chest
painaccompanied by appropriate findings on the chest radiograph), hepatitis (a
greater than twofold increase in serum
levels of aspartate aminotransferase, alanine aminotransferase, or both), and central nervous system involvement.
By using the aforementioned criteria,
138 cases of Q fever were identified. A
computerized questionnaire file was completed for each of these patients by using
the EPI INFO database (version 6.04b; Centers for Disease Control and Prevention,
Epidemiology Programs Office, Atlanta,
Ga). Findings from the initial and followup chest radiographs that were representative of the radiographic course of the
disease were included as a subfile within
this questionnaire.
Of this group of 138 patients, 85 had
abnormal chest radiographs available for
review. All chest radiographs in each patient obtained at admission, during the
hospital stay, and at discharge were reviewed by two of the authors (D.B., A.V.).
After an initial independent evaluation,
340 Radiology February 1999

discrepancies were discussed, and a final

consensus was reached about the predominant type of parenchymal abnormality, the distribution, the pleural involvement, and the mediastinal and hilar changes.
The mean age (SD) of the patients
was 36 years 15 (age range, 1892
years); 65 patients were men (76.5%), and
20 (23.5%) were women. All patients
were admitted to the hospital with a
clinical diagnosis of community-acquired
pneumonia. Fever was present at the time
of admission in 82 (96.5%) patients;
cough, in 30 (35.3%); expectoration, in
10 (11.8%); bloodstained sputum, in 13
(15.3%); wheezing, in three (3.5%); and
pleurodynia, in 16 (18.8%). The physical
examination of the chest revealed rales in
52 (61.2%) patients and rhonchi in five
(5.9%); in 25 (29.4%) patients, the examination results were negative.
The mean time (SD) between onset of
the disease and acquisition of the first
chest radiograph was 6.6 days 5.9. The
second radiograph was obtained 13.2
days 10.5 after the onset of disease; the
third, 24.2 days 16.7 after onset; and
the fourth, 26.4 days 16.5 after onset.
In four patients, five radiographs were
obtained, and the mean time between
admission and acquisition of the fifth
radiograph was 51.7 days 31.8.

TABLE 1
Initial Radiographic Findings in 85
Patients with Q Fever Pneumonia
Radiographic Pattern
Air space opacities
Lobar
Segmental
Patchy
Nodular findings
Micronodular (0.50.9 mm)
Macronodular (1.02.9 cm)
Masslike (3 cm)
Interstitial findings
Reticular
Reticulonodular
Associated findings
Pleural effusion
Volume loss
Increased volume
Basal linear opacities
Adenopathy

No. of
Patients
77 (90.5)
15 (17.6)
53 (62.3)
9 (10.6)
6 (7.1)
2 (2.4)
2 (2.4)
2 (2.4)
2 (2.4)
1 (1.2)
1 (1.2)
15 (17.6)
6 (7.1)
2 (2.4)
3 (3.5)
0 (0)

Note.Numbers in parentheses are percentages.

RESULTS
The chest radiographic patterns at the
time of admission are presented in Table
1. In the majority of patients (90.5%), air
space opacification predominated and was
Gikas et al

Figure 2. Anteroposterior chest radiograph obtained at the time of

hospital admission in a 19-year-old man with a 3-day history of fever
and nonproductive cough, a fourfold increase in IgM and IgG titers,
and C burnetii infection. A peripheral rounded area of soft-tissue
opacification (arrow) in the anterolateral aspect of the right lower lobe
mimics a tumor mass.

TABLE 2
Distribution of Lobar and Segmental
Opacities in 68 Patients with Q Fever
Pneumonia
Variable
Affected lobe
Right upper
Right middle
Right lower
Left upper
Left lower
Total no. of lobes
No. of affected zones
per patient
One
Two
Three
Four

Lobar Segmental
Opacity Opacity
4 (15)
7 (27)
4 (15)
8 (31)
3 (12)
26
8 (53)
4 (27)
2 (13)
1 (7)

Total no. of patients 15

17 (23)
5 (7)
20 (27)
23 (31)
9 (12)
74
38 (72)
11 (21)
2 (4)
2 (4)
53

Note.Numbers in parentheses are percentages.

expressed as segmental or lobar consolidation and patchy infiltrates. Some degree

of volume loss was noticed in six (40%) of
15 patients with lobar pneumonia, which
was seen as displacement of interlobar
fissures, (Fig 1), while an increase in lobe
volume was observed in two (13%) of 15
patients, which was seen as bulging of
interlobar fissures. A round pneumonia
pattern, mimicking a single pulmonary
mass, was seen in two patients (Fig 2).
Volume 210 Number 2

Figure 3. Anteroposterior chest radiograph obtained at the time of

hospital admission in a 36-year-old man with a 10-day history of
fever, cough, chest pain, myalgias, malaise, IgM titer of 1:1,600, and C
burnetii infection. Multiple bilateral nodular areas of consolidation
(arrows) are shown. These areas of consolidation are several centimeters in diameter.

Segmental and lobar opacities, observed in 15 (17.6%) and 53 (62.3%) of

the 85 patients, respectively, were the
most commonly recorded abnormalities.
A pattern of multiple disseminated nodules (micronodular, macronodular, and
masslike opacities) was observed in six
patients (7.1%). The diameter of the nodules ranged from a few millimeters (two
patients) to 4 cm (four patients) (Fig 3).
Only two of the 85 patients had interstitial pneumonia at the time of presentation, which was depicted as a reticular or
reticulonodular pattern (Fig 4). A small
pleural effusion and basal linear opacities
accompanying the parenchymal lesions
were observed in 15 (17.6%) and three
(3.5%) patients, respectively. None of the
patients had evidence of mediastinal or
hilar abnormalities.
In a total of 170 lungs examined in 85
patients, right lung involvement was
slightly predominant (59 [34.7%] right
lungs vs 41 [24%] left lungs). All patients
with nodular and interstitial patterns
showed bilateral lung involvement (Figs
3, 4). A bilateral distribution was also
observed in four (26.6%) of 15 patients
with lobar consolidation but in only one
(1.9%) of 53 patients with segmental
opacification and in only two (22.2%) of
nine patients with patchy infiltrates.
Areas of opacification of more than one
lobe were seen in almost half the patients
(seven [47%] of 15) with a lobar consolidation pattern. In contrast, opacification of

a single segment was seen in the majority

of patients (38 [72%] of 53) with a segmental consolidation pattern (Table 2, Fig 5).
In a total of 26 lobar opacities encountered in patients with lobar pneumonia,
the most frequently involved sites were
the left upper lobe (eight patients [31%])
and the right middle lobe (seven patients
[27%]).
In a total of 74 segmental opacities
encountered in patients with a segmental
pneumonia pattern, the predominant
sites of involvement were the upper lobes
(left upper lobe in 31%, right upper lobe
in 23%), while the right lower lobe was
involved in 27% of patients, and the left
lower lobe was involved in 12% (Table 2,
Fig 5).
In 70 (82%) of 85 patients, a second
chest radiograph, obtained between approximately the 2nd and 3rd week after
admission, was available for review. Only
18 (25%) of these 70 patients showed an
improvement in terms of radiographic
findings. Extension of pneumonia was
noted in 16 (23%) patients, and no change
was seen in 32 (45%) patients. Spread of
the disease occurred by means of direct
extension, not by means of migratory
lesions. Resolution was noted in four
(6%) patients (Table 3).
A third radiograph, obtained between
the 3rd and 4th week after admission, was
available in 40 patients, and the most
common findings were improvement (52%)
and resolution (18%). These were also the

Radiographic Appearance of Q Fever Pneumonia 341

most common findings in the small number of patients in whom a fourth and/or
fifth radiograph was available.
On the basis of the available radiographs, the mean time to complete resolution was estimated to be up to 39 days.
There was no relationship between the
extension of pulmonary involvement and
the course of the disease; the outcome
was favorable for all patients.

DISCUSSION
Q fever is a zoonosis caused by C burnetii
(1). Transmission to humans is by means
of inhalation of infected droplets or consumption of infected raw milk or fresh
cheese (2).
The clinical picture of acute infection
due to C burnetii consists of pneumonia,
prolonged fever of unknown origin,
granulomatous hepatitis, and meningoencephalitis (47,912). Endocarditis is
the best-described chronic entity of C
burnetii infection (18). The presence of
antibodies against antigen phase II of the
microorganism (by means of the indirect
immunofluorescence antibody technique),
associated with one or more of the cardinal manifestations of the disease (fever,
respiratory disease with radiographic findings, hepatitis, central nervous system
involvement, skin rash), define the diagnosis of Q fever (10,12).
Tetracycline is the drug of choice
against C burnetii. Chloramphenicol, cotrimoxazole, and rifampicine are also effective, whereas the use of erythromycin
in the treatment of Q fever is disputed
(10,18). Evidence of the clinical effectiveness of quinolones is still lacking (18).
Pulmonary involvement is a major
manifestation of acute Q fever infection.
In a number of clinical reports from Spain
(4), Canada (5), Switzerland (6), the United
States (7), and, recently, from Greece (12),
pneumonia was the predominant clinical
manifestation of the disease, but detailed
descriptions of the radiographic findings
are lacking.
In the 85 patients in the present study
with clinical and serologic evidence of Q
fever pneumonia, the radiographic findings were not specific for the disease. The
most common abnormalities we noted
were unilateral, single-segmental opacities, which were located mainly in the
upper lobes. As far as we know, this distribution of the segmental opacities has not
been described as a principle characteristic of Q fever pneumonia. Lobar opacities
were found less commonly. Similar findings have been reported in other pub342 Radiology February 1999

a.

b.

Figure 4. Chest radiographs obtained at the time of hospital admission in a 39-year-old woman
with a 10-day history of fever, productive cough, headache, myalgias, malaise, confusion, and a
fourfold increase in IgM and IgG titers. (a) Anteroposterior view shows reticulonodular infiltrations bilaterally, simulating interstitial disease. However, a peripheral rim of heavy consolidation
(arrows) is visible. (b) Lateral view depicts mild volume reduction of the left upper lobe, which is
seen as upward displacement of the major fissure.

TABLE 3
Radiographic Follow-up in 85 Patients with Q Fever Pneumonia
Chest
Radiography
Series
First study (n 85)
Second study
(n 70)
Third study (n 40)
Fourth study
(n 18)
Fifth study (n 4)

Patients with
Patients
Patients
Patients
Time between
with Stable Clearance
with
with
Symptom Onset
of Disease
and Radiography (d)* Extension Improvement Findings
6.6 5.9

NA

NA

NA

NA

13.1 10.5
24.2 16.7

16 (23)
4 (10)

18 (26)
21 (52)

32 (46)
8 (20)

4 (6)
7 (18)

26.4 16.5
51.7 31.7

1 (6)
0 (0)

10 (56)
1 (25)

4 (22)
1 (25)

3 (17)
2 (50)

Note.Unless otherwise noted, data are the number of patients, with percentages in parentheses.
NA not applicable.
* Data are the mean SD.

lished series (1316) about the radiographic

pulmonary manifestations of Q fever. Segmental or lobar consolidations were observed by Jacobson et al (13) in 59 (90.7%)
of 65 cases and by Smith et al (16) in 51
(82%) of 62 cases. In these studies, the
lower lobes were more frequently involved, whereas in our series, upper lobe
involvement was predominant.
We could not confirm the findings of
Millar (17), who reported linear atelectasis in 50% and multiple round opacities
in 68% of a series of 35 cases. In our series,
linear atelectasis was noted in only three
patients, and multiple nodular opacities
were seen in six patients.

The radiographic differentiation of Q

fever pneumonia from the other community-acquired pneumonias is not possible. Segmental or lobar opacities and an
occasional pleural effusion, which in our
series were noted in 15 (17.6%) patients,
are common in other community-acquired pneumonias as well. On the other
hand, Q fever pneumonia has a tendency
to resolve more slowly; in our series,
resolution of radiographic findings was
found to take up to 39 days.
When lobar or segmental pneumonia
involves the upper lobes, it resembles
primary tuberculosis or fungal disease.
However, cavitation and hilar or mediastiGikas et al

vascular engorgement and the migratory

extension of the disease seen in atypical
pneumonia are very rare in Q fever pneumonia. These characteristics were absent
in our series.
In summary, a single, unilateral, segmental, upper-lobar opacity was the main
finding observed in our series of patients
with Q fever pneumonia. Unilateral lobar
consolidation was less frequently observed. Resolution of these lesions is slow.
The definite diagnosis of Q fever can be
established only on the basis of laboratory results.

Figure 5. Anteroposterior chest radiograph

obtained at the time of admission in a 36-yearold woman with a 3-day history of fever,
cough, frontal headache, arthralgias, myalgias,
malaise, and a fourfold increase in IgM and IgG
titers. Airspace opacification (arrow) involves a
single segment of the right upper lobe.

nal involvement were characteristically

absent in our and other published series
(1317).
Although atypical pneumonia has a
somewhat similar clinical picture, it can
often be differentiated from Q fever pneumonia. Segmental and lobar involvement
are less frequent in atypical pneumonia
(13,14). On the other hand, the hilar and

Volume 210 Number 2

References
1. Marrie TJ. Epidemiology of Q fever. In:
Marrie TJ, ed. Q fever. Vol 1, The disease.
Boca Raton, Fla: CRC, 1990; 4970.
2. Marmion RA, Stoker MQP. Q fever in
Britain: epidemiology of an outbreak. Lancet 1959; 2:611616.
3. Derrick EH. Q fever, a new fever entity:
clinical features, diagnosis and laboratory
investigation. Med J Aust 1937; 2:281299.
4. Montejo Baranda M, Corral Carranceja J,
Aguirre Errasti C. Q fever in the Basque
country: 19811984. Rev Infect Dis 1985;
7:700701.
5. Marrie TJ. Q fever, 1979-1987, Nova Scotia. Can Dis Wkly Rep 1988; 14:6970.
6. Dupuis G, Peter O, Pedroni D, Petite J.
Aspects cliniques observes lors dune epidemie de 415 cas de fievre Q. Schweiz Med
Wochenschr 1985; 115:814818.
7. Reilly S, Northwood JL, Caul EO. Q fever
in Plymouth, 197288: a review with particular reference to neurological manifestations. Epidemiol Infect 1990; 105:391
408.

8.

9.
10.
11.

12.

13.
14.
15.

16.

17.
18.

Tissot Dupont H, Raoult D, Brouqui P, et

al. Epidemiologic features and clinical presentation of acute Q fever in hospitalized
patients: 323 French cases. Am J Med
1992; 93:427434.
Vellend H, Salit IE, Spence L, et al. Q fever,
Ontario. Can Dis Wkly Rep 1982; 8:171.
Spelman DW. Q fever: a study of 111
consecutive cases. Med J Aust 1982; 1:547
553.
Raoult D, Drancourt M, De Micco C, et al.
Les hepatites de la fievre Q: a propos de 14
cas et revue de la litterature. Semin Ho
pital 1986; 26:997999.
Tselentis Y, Gikas A, Kofteridis D, et al. Q
fever in the Greek island of Crete: epidemiology, clinical and therapeutic data from
98 cases. Clin Infect Dis 1995; 20:1311
1316.
Jacobson G, Denlinger RS, Carter RA.
Roentgen manifestation of Q fever. Radiology 1949; 53:739749.
Glucker LG, Munk J. Roentgenologic pulmonary manifestation of Queensland fever. J Faculty Radiol 1952; 3:186192.
Pickworth FE, El-Soussi M, Wells IP, McGavin CR, Burge PS. The radiological appearance of Q fever pneumonia. Clin
Radiol 1991; 44:150153.
Smith DL, Wellings R, Walker C, et al. The
chest x ray in Q fever: a report of 69 cases
from the 1989 West Midlands outbreak.
Br J Radiol 1991; 64:11011108.
Millar JK. The chest film findings in Q
fever: a series of 35 cases. Clin Radiol
1978; 29:371375.
Marrie TJ. Coxiella burnetii (Q fever). In:
Mandell GL, Douglas RG, Bennet JE, eds.
Principles and practice of infectious diseases. 3rd ed. New York, NY: Livingstone,
1990; 14721476.

Radiographic Appearance of Q Fever Pneumonia 343