EDUCATION EXHIBIT 121

Spectrum of Clinical
and Radiographic Find-
ings in Pediatric Myco-
plasma Pneumonia1
ONLINE-ONLY Susan D. John, MD • Janaki Ramanathan, MD • Leonard E. Swischuk, MD
.......CME .......
See www.rsna.org
/education Clinical symptoms in mycoplasma infection are nonspecific. Pulmonary
/rg_cme.html involvement may be widespread or focal and segmental and is accom-
panied by signs including rales, rhonchi, and decreased breath sounds.
LEARNING Although manifestations of mycoplasma infection are usually confined
OBJECTIVES to the respiratory tract, a wide variety of extrarespiratory manifestations
After reading this
article and taking can also occur, including more severe associated diseases such as myo-
the test, the reader carditis, acute disseminated encephalomyelitis, and cerebral arterio-
will be able to:
■ Develop a frame-
venous occlusion. The radiographic findings in mycoplasma pneumo-
work for approach- nia are also nonspecific and in some cases closely resemble those seen
ing the diagnosis of
mycoplasma infec-
in children with viral infections of the lower respiratory tract. Focal re-
tion in children. ticulonodular opacification confined to a single lobe is a radiographic
■ Recognize the fo- pattern that seems to be more closely associated with mycoplasma in-
cal reticulonodular
pattern seen in some fection than with other types of pediatric respiratory illnesses, and the
patients, as well as diagnosis of mycoplasma pneumonia should be considered whenever
other common find-
ings.
focal or bilateral reticulonodular opacification is seen. Hazy or ground-
■ Identify the impor- glass consolidations frequently occur, but dense homogeneous consoli-
tant clinical features dations like those seen with bacterial pneumonias are uncommon. At-
of myoplasma infec-
tion and correlate electasis or transient pseudoconsolidations due to confluent interstitial
them with radio- shadows are often seen. Radiographic findings alone are not sufficient
graphic findings.
for the definitive diagnosis of mycoplasma pneumonia, but in combi-
nation with clinical findings they can significantly improve the accu-
racy of diagnosis in this disease.

Index terms: Lung, infection, 60.2061 • Pneumonia, 60.2061 • Radiography, in infants and children, 60.11

RadioGraphics 2001; 21:121–131

1From the Department of Radiology, University of Texas–Houston Medical School, 6431 Fannin–MSB2.100, Houston, TX 77030 (S.D.J.); and

the Department of Radiology, University of Texas Medical Branch, Galveston (J.R., L.E.S.). Recipient of a Certificate of Merit for a scientific ex-
hibit at the 1999 RSNA scientific assembly. Received March 27, 2000; revision requested April 25 and received June 9; accepted June 9. Address
correspondence to S.D.J. (e-mail: susan.d.john@uth.tmc.edu).
©RSNA, 2001
122 January-February 2001
Introduction
Mycoplasma pneumoniae, a small, cell wall–deficient
bacterium, is an important and treatable cause of
community-acquired, atypical pneumonia in chil-
dren (1–3). The M pneumoniae organism is pleo-
morphic and demonstrates variable clinical and
radiographic findings. At microscopic analysis,
these organisms assume variable forms that can
resemble fungal filaments. The lack of a cell wall
makes them insensitive to the usual antibiotics
that are used to treat other common pneumonias;
therefore, differentiation from bacterial pulmonary
infections is vital to successful treatment. Myco-
plasma pneumonia can mimic viral respiratory
tract infections both clinically and radiographi-
cally; unlike viruses, however, mycoplasma infec-
tion is treatable. Appropriate antibiotic therapy
for mycoplasma infections (usually with erythro-
mycin, azithromycin, or tetracycline) can help
prevent the spread of the infection to contacts
and limit the course of the disease in the patient.
Prompt therapy may also help avert extrarespir-
atory manifestations of mycoplasma pneumonia,
which can occasionally be severe.
A wide variety of radiographic findings have
been attributed to mycoplasma respiratory infec-
tions over the years (4–10). Several authors have
described the tendency of mycoplasma infection
to produce focal interstitial changes in the lungs
(4,6,7). We, too, made this observation, and so
began to catalog cases of children with clinical or
radiographic findings suggestive of mycoplasma
pneumonia.
In this article, we discuss, illustrate, and corre-
late the clinical and radiographic findings in chil-
dren and adolescents with serologic evidence of
M pneumoniae pulmonary infection.
Clinical Experience
We prospectively identified 120 children and
adolescents over a 10-year period who were sus-
pected of having mycoplasma pneumonia, based
on clinical findings or on the presence of a reticu-
lar or nodular pattern confined to one pulmonary
lobe at chest radiography. Clinical findings that
raised suspicion for mycoplasma infection were
variable, but the most common reasons for suspi-
cion were persistent dry cough, lack of fever or a
low-grade fever less than 38.5°C, prominent head-
RG ■ Volume 21 • Number 1
ache or myalgias, illness lasting more than 1 week,
and lack of response to antibiotics. At least one
posteroanterior and one lateral chest radiograph
were obtained in each patient. Two adolescents
were also evaluated with computed tomography
(CT). Radiographs were independently evaluated
by two pediatric radiologists (S.D.J., L.E.S.) for
the presence of the following findings: (a) a
focal reticular, nodular, or reticulonodular pat-
tern; (b) focal consolidation (subcategorized as
dense or hazy); (c) a diffuse interstitial pattern;
(d) parahilar peribronchial opacification; (e) hilar
lymphadenopathy; (f) pleural effusion; and (g) at-
electasis or air trapping. The findings were later
compared, and discrepancies were resolved by
consensus. Referring physicians were encouraged
to verify the diagnosis of mycoplasma pneumonia
with serologic titers whenever possible. A positive
diagnosis was made on the basis of a single ele-
vated titer or a fourfold increase in titer. Of the
57 patients who were tested for serologic evi-
dence of infection with M pneumoniae, 42 (74%)
had seropositive findings and were included in
the study. The study population included chil-
dren and adolescents of all ages except for infants
less than 1 year old (age range, 1–16 years; mean,
8.2 years). Although the majority of patients with
mycoplasma pneumonia in our study were more
than 5 years of age, some children were less than
5 years old, including two children between 12
and 24 months of age. Some were hospital pa-
tients, whereas others were outpatients.
Results
Radiographic findings included focal pulmonary
areas of increased opacity in 35 of 42 patients
(83%), diffuse or bilateral perihilar areas of in-
creased opacity in five (12%), and normal find-
ings in two (5%). Focal pulmonary areas of in-
creased opacity occurred on the right side in 17
patients (40%) and on the left side in 13 (31%)
and were seen bilaterally in five (12%). Focal lo-
bar involvement was much more common in the
lower lobes (28 of 35 patients [80%]) than in the
upper or middle lobes (seven of 35 [20%]). Pleu-
ral effusions were seen in seven patients (17%),
and all effusions occurred on the left side except
for bilateral effusions in one patient. Hilar lymph-
adenopathy occurred in three patients (7%); in all
three cases, the lymph nodes were unilateral and
right-sided.
RG ■ Volume 21 • Number 1 John et al 123

Table 1 rect contact or by aerosol. The incubation period

Clinical Signs and Symptoms in 42 Children is usually 1–2 weeks. Mycoplasma infection is
with Mycoplasma Pneumonia uncommon in the 1st year of life but is an impor-
tant cause of community-acquired respiratory
Manifestation No. of Patients*
infections in school-age children and in even
Cough† 38 (90) younger children who attend child care centers.
Rales 26 (62) The symptoms of M pneumoniae infections are
Rhonchi 7 (17) nonspecific, and the onset of illness is usually
Wheezing 8 (19) gradual. The illness may begin in the upper respi-
Decreased breath sounds 11 (26) ratory tract as a mild pharyngitis, often accompa-
Fever‡ 13 (31)
nied by low-grade fever, headache, and myalgias;
Elevated white blood cell count 7 (29)§
rhinorrhea and nasal congestion are uncommon.
Left shift 18 (75)§
Within a few days, symptoms of tracheobronchi-
*Numbers in parentheses are percentages. tis develop, and the predominant symptom is
†
Mean duration = 8.5 da. usually a nagging, dry cough (2). Wheezing is not
‡
>38.0°C. a common symptom in children with no history
§
Findings in 24 patients. of asthma, but mycoplasma infection can precipi-
tate an acute asthma attack. Indeed, mycoplasma
infection is an important cause of asthma exacer-
bation in children (11). Fever and sputum pro-
Table 2 duction can occur, but they are not as severe as
Extrarespiratory Manifestations of Mycoplas- in pneumococcal and other bacterial pneumonias
ma Pneumonia (1). Typically, the interval between initial symp-
Systemic manifestations toms and fully developed pneumonitis is 6–10
Headache days. Postinfectious bronchitis may continue for
Fever, chills weeks or months.
Myalgias Clinical signs of mycoplasma infection of the
Nausea, vomiting, diarrhea lungs include rales, rhonchi, and decreased breath
Arthropathy sounds (Table 1), but signs of major consolida-
Immune hemolytic anemia tion are rare. Pulmonary involvement may be
Cardiac manifestations widespread or focal and segmental. White blood
Pericarditis
cell count is usually within the normal range.
Myocarditis
Myringitis
When the white blood cell count is elevated, a
Cervical lymphadenopathy superimposed bacterial infection is often the
Neurologic manifestations cause. Sputum that exhibits polymorphonuclear
Meningoencephalitis and mononuclear cells but few bacteria should
Transverse myelitis raise suspicion for mycoplasma infection.
Cerebral infarction Although illness caused by M pneumoniae is
Striatal necrosis usually confined to the respiratory tract, a wide
Cranial neuropathy (Guillian-Barre syndrome) variety of extrarespiratory manifestations can oc-
Dermatologic manifestations cur (Table 2). The cause of these manifestations
Erythema is still unknown, but autoimmune responses to
Urticaria
the infection are suspected. Some of the more
Vesicles
Stevens-Johnson syndrome
severe associated diseases include myocarditis,
acute disseminated encephalomyelitis, and cere-
bral arteriovenous occlusion (12–14). In most
cases, nonrespiratory manifestations are preceded
by respiratory symptoms, but treatment of the
Clinical Features of
pulmonary disease seems to have little effect on
Mycoplasma Infection subsequent neurologic disease. The interval be-
Humans are the only known reservoir for M tween respiratory symptoms and neurologic
pneumoniae. This tiny organism measures less manifestations is approximately 10 days (15).
than 350 m in length and is not visible at light mi-
croscopy. Mycoplasma infection is spread by di-
124 January-February 2001 RG ■ Volume 21 • Number 1

Figure 1. Mycoplasma pneumonia. Pho-

tomicrograph (hematoxylin-eosin stain)
shows a bronchiole surrounded by mono-
nuclear cells. Note the polymorphonuclear
cells and desquamated epithelium in the lu-
men.

Histopathologic Table 3
Features of Mycoplasma Infection Radiographic Findings in 42 Children with
Mycoplasma Pneumonia
The histopathologic features of mycoplasma in-
fection are limited primarily to ciliated respira- Finding No. of Patients*
tory epithelium from the trachea to the respira- Focal reticular pattern
tory bronchioles (16). Airways are surrounded by Unilobar 22 (52)
mononuclear cell infiltrates (Fig 1). Peribron- Bilobar 4 (10)
chial infiltrate can extend into the interstitium Parahilar peribronchial opacification 5 (12)
along blood vessels and lymphatic vessels (17). Consolidation or pseudoconsolidation 14 (33)
Both polymorphonuclear and mononuclear cells Diffuse interstitial pattern 3 (7)
may be found in the lumen of the airways. This Atelectasis 12 (29)
pattern is similar to that seen with viral lower res- Hilar lymphadenopathy 3 (7)
piratory tract infections, which explains why many Pleural effusion 7 (17)
of the radiographic features of these infections Normal findings 2 (5)
are also similar. Hematogenous spread of myco- *Numbers in parentheses are percentages.
plasma infection is rare.

Radiographic involvement to be more common, but we were

Features of Mycoplasma Infection
The radiographic patterns observed in patients
with M pneumoniae infections of the lower respi-
ratory tract are variable (Table 3). No clear pat-
tern of lobar involvement was identified in our
study, but the lower lobes were more commonly
affected than the upper lobes. Previous studies
have also noted a lower lobe predominance (4,7).
Involvement of one or two lobes was more com-
mon in our study than diffuse pulmonary disease.
Previous studies have shown diffuse or multilobar
more likely to identify children with unilobar dis-
ease because of our interest in this pattern. Se-
vere pulmonary complications of mycoplasma
pneumonia have been reported, including Swyer-
James syndrome, pulmonary fibrosis, bronchiol-
itis obliterans, and adult respiratory distress syn-
drome. However, none of these complications
was encountered in our series of patients.
The most common radiographic pattern en-
countered in our study was a reticulonodular pat-
tern confined to one lobe (Figs 2, 3). This pat-
tern correlates with the histopathologic findings
of peribronchitis. In some cases, the nodules
RG ■ Volume 21 • Number 1 John et al 125

a. b.

c. d.
Figure 2. Focal reticulonodular patterns in four different patients. (a) Posteroanterior radiograph demonstrates a
localized nodular pattern in the right upper lobe. (b) Posteroanterior radiograph shows a reticulonodular pattern
confined to the left lower lobe. (c) Posteroanterior radiograph shows a reticular pattern in the right lower lobe ac-
companied by small, patchy areas of increased opacity (arrows). (d) Radiograph demonstrates a reticular pattern
with mild, hazy opacification in the right lower lobe.

a.

Figure 3. Focal reticulonodular pattern. Frontal (a) and

lateral (b) radiographs obtained in an adolescent girl show
opacification in the right lower lobe, which appears hazy be-
cause of overlying breast tissue. The reticular nature of this
infiltrate is seen in b (arrows).

b.
126 January-February 2001 RG ■ Volume 21 • Number 1

a. b.

c. d.
Figure 4. Diffuse interstitial and parahilar peribronchial patterns in four different patients. (a) Posteroanterior ra-
diograph shows a mild, diffuse nodular pattern throughout both lungs with bronchial wall thickening. (b) Posteroan-
terior radiograph demonstrates more severe interstitial disease bilaterally, with both patchy nodular areas and cen-
tral alveolar areas of increased opacity on the right side. (c) Posteroanterior radiograph obtained in a child shows a
pattern resembling that of a viral infection, with bilateral parahilar peribronchial opacification and multiple areas of
subsegmental atelectasis (arrows). (d) Radiograph demonstrates a slightly nodular pattern in the lung bases with
atelectasis bilaterally.

were less well defined and resembled small, focal
patchy areas of increased opacity (Fig 2c), but
the pattern of opacification was distinct from the
homogeneous consolidations seen with typical
bacterial pneumonias. In several patients, the fo-
cal reticulonodular pattern was the initial radio-
graphic finding but later progressed to other pat-
terns. In other cases, reticulonodular changes
were noted adjacent to more homogeneous lobar
areas of increased opacity. Other authors have
described similar reticulonodular interstitial infil-
trates but have not observed a propensity for lo-
bar involvement (4,6,7). The common occur-
rence of the lobar reticulonodular pattern in our
study may be partially due to selection bias. The
nodular appearance seen on radiographs obtained
in our patients with mycoplasma pneumonia
most likely correlates with the centilobular or aci-
nar shadows detected by Tanaka et al (9) in a re-
cent CT study of community-acquired pneumo-
nias. Although pathophysiologically mycoplasma
respiratory infections are most commonly diffuse
and bronchial or peribronchial in nature, the ra-
diographic manifestations often reflect the pres-
ence of lobar interstitial disease.
Diffuse interstitial and bilateral parahilar peri-
bronchial patterns are common in mycoplasma
respiratory infections (Fig 4). The radiographic
findings are indistinguishable from those seen in
children with various viral infections of the lower
respiratory tract or other pathogens that cause
primarily bronchial and peribronchial disease
RG ■ Volume 21 • Number 1 John et al 127

a. b.
Figure 5. Diffuse interstitial and parahilar peribronchial pattern. (a) Initial posteroanterior radiograph obtained in
a child shows a streaky, hazy area in the right lower lobe. (b) Posteroanterior radiograph obtained 2 days later shows
resolution of the atelectasis, leaving a focal reticular pattern.

a. b.
Figure 6. Unilateral (right-sided) hilar lymphadenopathy. (a) Posteroanterior radiograph obtained in a child dem-
onstrates a reticulonodular area of increased opacity in the right upper lobe (arrows) accompanied by a nodular pat-
tern in the right hilum due to lymphadenopathy. (b) Posteroanterior radiograph obtained in a different child demon-
strates right hilar prominence due to lymphadenopathy (arrows). The adjacent area of increased opacity in the lung
was predominantly due to atelectasis.

such as pertussis and chlamydia. Atelectasis is a
common associated finding in affected patients
(Figs 4d, 5), providing a clue to the bronchial na-
ture of the disease.
Hilar lymphadenopathy is uncommon in my-
coplasma pneumonia, but unilateral hilar lymph
node enlargement was seen in a few patients in
our study (Fig 6). The radiographic findings in
these patients may be indistinguishable from those
seen in children with primary tuberculosis. In a
recent high-resolution CT study of patients with
M pneumoniae infections, lymphadenopathy was
seen in only one of 14 patients (7%) (18). In other
studies, it has been reported in up to 22% of pa-
tients (4,8).
Segmental and lobar consolidations have been
reported with variable frequency (from rare up to
57% of cases) in previous series of patients with
mycoplasma pneumonia (4–8,10). The differ-
ences in the reported prevalence of consolida-
tions in mycoplasma pneumonia may be due to
considerable variability in the appearance of pul-
monary areas of increased opacity that are viewed
128 January-February 2001 RG ■ Volume 21 • Number 1

7a. 7b.

8. 9.
Figures 7–9. Consolidation and “pseudoconsolidation.” (7a) Radiograph shows bilateral areas of hazy consolida-
tion in the lower lobes with air bronchograms. (7b) CT scan demonstrates bilateral lower lobe consolidation that ap-
pears more solid than that seen at radiography. (8) Posteroanterior radiograph obtained in a child demonstrates a
round, hazy area of increased opacity (arrows) that could represent an early consolidation or pseudoconsolidation.
(9) Posteroanterior radiograph demonstrates bilateral consolidations with a hazy appearance (arrows), findings that
suggest pseudoconsolidation.

as consolidations by different observers. In our
study, consolidations were seen in 33% of pa-
tients (Figs 7–10). However, the majority (11 of
13) of these consolidations consisted of homoge-
neous areas of hazy or ground-glass opacification
(Figs 8–10) or patchy inhomogeneous areas of
increased opacity rather than the homogeneous
dense opacification characteristic of bacterial
pneumonias. In some patients, the consolidation
was preceded by focal reticulonodular opacifica-
tion (Fig 11) or accompanied by a reticulonodu-
lar pattern in another portion of the lungs. In
other patients, a slight reticulonodular appear-
ance was noted at the margins of more homoge-
neous hazy opacification. Occasionally, a tran-
sient consolidation was seen on the initial radio-
graphs but cleared rapidly, leaving behind a focal
reticulonodular pattern (Fig 5). In such cases,
these pseudoconsolidations were probably due to
atelectasis or transient edema complicating the
focal bronchial and interstitial process. Patients
in our study with consolidations were only
slightly more likely to have more severe clinical
signs and symptoms than were patients with dif-
fuse peribronchial or localized reticulonodular
changes. A well-known feature of mycoplasma
pneumonia is the discrepancy between the sever-
ity of the radiographic changes and the relatively
mild clinical findings in some patients.
RG ■ Volume 21 • Number 1 John et al 129

a. b.
Figure 10. Consolidation and pseudoconsolidation
in an 11-year-old patient. (a) Initial posteroanterior ra-
diograph demonstrates a mild reticular pattern on the
right side (arrows). (b) Posteroanterior radiograph ob-
tained 4 days later reveals that a focal, hazy area of in-
creased opacity has developed. (c) CT scan shows areas
of mild ground-glass attenuation and acinar attenuation
confined to the right lower lobe.

c.

a. b.
Figure 11. Progression of mycoplasma infection in a
child. (a) Initial posteroanterior radiograph shows only
a small reticulonodular area in the right lower lobe (ar-
rows). (b) On a posteroanterior radiograph obtained 2
weeks later, the reticulonodular area has become more
prominent (arrows). (c) Posteroanterior radiograph ob-
tained 1 week after b reveals atelectasis in the right mid-
dle and lower lobes and a hazy pseudoconsolidation in
the left upper lobe.

c.
130 January-February 2001 RG ■ Volume 21 • Number 1

a. b.
Figure 12. Pleural effusions. (a, b) Posteroanterior
(a) and lateral (b) radiographs obtained in a child dem-
onstrate a reticular and slightly nodular focal area of in-
creased opacity in the left lower lobe, with an associated
small pleural effusion (arrow). Note the slight bowing
of the oblique fissure (arrowheads in b), a finding that
indicates volume loss. (c) Posteroanterior radiograph
obtained in a different child demonstrates subsegmental
atelectasis and a vague reticular pattern in the left lower
lobe with a small pleural effusion (arrow).

Pleural effusions were uncommon in our study

and were uniformly small (Fig 12). Previous
studies have reported a prevalence of 5%–20%
for pleural effusions (4–6,8,18). These effusions
are usually transient and of no clinical signifi-
cance; occasionally, however, the pleural fluid
may persist after resolution of the pulmonary pa-
renchymal findings.
Conclusions
The radiographic findings in mycoplasma pneu-
monia are variable, and in some cases they closely
resemble those seen in children with viral infec-
tions of the lower respiratory tract. Although none
of the findings are specific, focal reticulonodular
opacification confined to a single lobe seems to be
more closely associated with mycoplasma infec-
c.
tion than with other types of respiratory illnesses
in children. The diagnosis of mycoplasma pneu-
monia should be considered whenever focal or bi-
lateral reticulonodular opacification is seen. Hazy
or ground-glass consolidations occur with M
pneumoniae infections, but dense homogeneous
consolidations like those seen with bacterial pneu-
monias are uncommon. Transient pseudocon-
solidations due to confluent interstitial shadows
or atelectasis are common. Definitive diagnosis of
mycoplasma pneumonia cannot be made on the
basis of imaging features alone, but a combination
of clinical and radiographic findings can signifi-
cantly improve the accuracy of diagnosis in this
disease.
RG ■ Volume 21 • Number 1
References
1. Cimolai N. Mycoplasma pneumoniae respiratory
infection. Pediatr Rev 1998; 19:327–332.
2. Mansel JK, Rosenow EC, Smith TF, et al. Myco-
plasma pneumoniae pneumonia. Chest 1989; 93:
639–646.
3. Clyde WA Jr. Mycoplasma pneumoniae respira-
tory disease symposium: summation and signifi-
cance. Yale J Biol Med 1983; 56:523–527.
4. Guckel C, Benz-Bohm C, Widemann, B. Myco-
plasmal pneumonias in childhood: roentgen fea-
tures, differential diagnosis, and review of the lit-
erature. Pediatr Radiol 1989; 19:499–503.
5. Putman OE, Curtis AM, Simeone JF, et al. Myco-
plasma pneumonia, clinical and roentgenographic
patterns. Am J Roentgenol 1975; 124:417–422.
6. Janower ML, Weiss EB. Mycoplasmal, viral and
rickettsial pneumonias. Semin Roentgenol 1980;
15:25–34.
7. Cameron DC, Borthwick RN, Philp T. The radio-
graphic patterns of acute mycoplasma pneumoni-
tis. Clin Radiol 1977; 28:173–180.
8. MacFarlane JT, Miller AC, Roderick Smith WH,
et al. Comparative radiographic features of com-
munity acquired Legionnaires’ disease, pneumo-
coccal pneumonia, mycoplasma pneumonia, and
psittacosis. Thorax 1984; 39:28–33.
9. Tanaka N, Matsumoto T, Kuramitsu T, et al.
High resolution CT findings in community-ac-
quired pneumonia. J Comput Assist Tomogr
1996; 204:600–608.
10. Cockcroft DW, Stilwell GA. Lobar pneumonia
caused by Mycoplasma pneumoniae. Can Med
Assoc J 1981; 124:1463–1468.
This article meets the criteria for 1.0 credit hour in category 1 of the AMA Physician’s Recognition Award. To obtain
credit, see accompanying test at http://www.rsna.org/education/rg_cme.html.
John et al 131
11. Kondo S, Ito M, Saito M, et al. Progressive bron-
chial obstruction during the acute stage of respira-
tory tract infection in asthmatic children. Chest
1994; 106:100–104.
12. Ali NJ, Sillis M, Andrews BE, et al. The clinical
spectrum and diagnosis of Mycoplasma pneu-
moniae infection. Q J Med 1986; 58:241–251.
13. Ponka A. The occurrence and clinical picture of
serologically verified Mycoplasma pneumoniae in-
fections with emphasis on central nervous system,
cardiac and joint manifestations. Ann Clin Res
1979; 24:1–60.
14. Decaux G, Szyper M, Ectors M, Cornil A, Franken
L. Central nervous system complications of Myco-
plasma pneumoniae. J Neurol Neurosurg Psychia-
try 1980; 43:883–887.
15. Levine DP, Lerner AM. The clinical spectrum of
Mycoplasma pneumoniae infections. Med Clin
North Am 1978; 62:961–978.
16. Spencer H. The bacterial pneumonias due to rick-
ettsiae, chlamydiae, viruses, and mycoplasma. In:
Spencer H, ed. Pathology of the lung. 4th ed. Ox-
ford, England; Pergamon, 1985; 213–259.
17. Paster F, Joliffe LS, Finland M. Primary atypical
pneumonia: report of eight cases with autopsies.
Arch Pathol (Chicago) 1947; 44:581–608.
18. Fine NL, Smith LR, Sheedy PF. Frequency of
pleural effusions in mycoplasma and viral pneu-
monias. N Engl J Med 1970; 283:790–793.