embolia gordurosa

1. Review article
Fat embolism syndrome: State-of-the-art review focused on
pulmonary imaging findings
Katrina Newbigin a, 1
, Carolina A. Souza b, 2
, Carlos Torres b, 2
, Edson Marchiori c, 3
,
Ashish Gupta b, 2
, Joao Inacio b, 2
, Mitchel Armstrong b, 4
, Elena Pe~na b, 2, *
a
Wesley Hospital, Brisbane, Australia
b
The Ottawa Hospital, Ottawa, Ontario, Canada
c
Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
a r t i c l e i n f o
Article history:
Received 30 March 2015
Received in revised form
22 October 2015
Accepted 28 January 2016
Available online 1 February 2016
Keywords:
Fat embolism
Non thrombotic pulmonary embolism
High resolution computed tomography
Pulmonary imaging
a b s t r a c t
Background: Fat embolism syndrome (FES) is a rare but potentially fatal complication of trauma or or-
thopedic surgery, which presents predominantly with pulmonary symptoms. Modern intensive care has
improved the mortality rates, however diagnosis remains difficult, relying predominantly on a combi-
nation of a classic triad of symptoms and non-specific, but characteristic radiological features. The aim of
this review is to describe the main clinical and imaging aspects of FES, ranging from pathophysiology to
treatment with emphasis on pulmonary involvement.
Methods: We reviewed the currently published literature on the main characteristics of FES.
Results: In a hypoxic patient with recent trauma or orthopedic surgery, the presence of diffuse, well-
demarcated ground glass opacities or ill-defined centrilobular nodules on computed tomography (CT)
of the chest are suggestive of FES.
Conclusions: Combination of the classic clinical syndrome in the appropriate clinical setting, together
with the characteristic imaging findings on chest CT, can help to achieve the correct diagnosis. Man-
agement remains predominantly supportive care, and the benefit of medical therapies such as cortico-
steroids and heparin remains unclear.
© 2016 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
2. Definition and classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
3. Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4. Pathophysiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.1. The mechanical theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.2. The biochemical theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5. Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6. Clinical manifestations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
7. Biochemical testing and bronchoalveolar lavage in FES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Abbreviations: FES, fat embolism syndrome; CT, computed tomography; ARDS, adult respiratory distress syndrome; HRCT, high-resolution computed tomography; MR,
magnetic resonance imaging; DWI, diffusion-weighted images.
* Corresponding author.
E-mail addresses: drnewbigin@gmail.com (K. Newbigin), csouza@toh.on.ca (C.A. Souza), catorres@toh.on.ca (C. Torres), edmarchiori@gmail.com (E. Marchiori), ashgupta@
toh.on.ca (A. Gupta), jinacio@toh.on.ca (J. Inacio), mitarmstrong@toh.on.ca (M. Armstrong), epena@toh.on.ca (E. Pe~na).
1
Department of Radiology, Wesley Hospital, 2 Chasley St, Auchenflower, Brisbane, QLD, 4066, Australia.
2
Department of Medical Imaging, Ottawa Hospital Research Institute, The Ottawa Hospital, University of Ottawa, 501 Smyth Road, Box 232, K1H8L6, Ottawa, ON, Canada.
3
Department of Radiology, Federal University of Rio de Janeiro, Rua Thomaz Cameron, 438, Valparaíso, CEP 25685-120, Petropolis, Rio de Janeiro, RJ, Brazil.
4
Department of Orthopedic Surgery, Ottawa Hospital Research Institute, The Ottawa Hospital, University of Ottawa, 501 Smyth Road, Box 232, K1H8L6, Ottawa, ON,
Canada.
Contents lists available at ScienceDirect
Respiratory Medicine
journal homepage: www.elsevier.com/locate/rmed
http://dx.doi.org/10.1016/j.rmed.2016.01.018
0954-6111/© 2016 Elsevier Ltd. All rights reserved.
Respiratory Medicine 113 (2016) 93e100

2. 8. Thoracic imaging manifestations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
8.1. Chest radiograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
8.2. Computed tomography of the chest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
8.2.1. Patchy ground-glass opacities and consolidations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
8.2.2. Small centrilobular nodules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
8.2.3. Other CT findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
9. Differential diagnosis of FES on chest CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
10. Imaging findings in cerebral fat embolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
11. Clinical course and prognosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
12. Management and treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
13. Prevention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
14. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
1. Introduction
Traumatic fat embolism syndrome (FES) is a rare entity, usually
occurring after long bone fracture or orthopedic surgery and is
classically described as triad of pulmonary, central nervous system
and skin manifestations. Although first described by Zenker in 1861
[1], the diagnosis of FES remains challenging due to non-specific
symptoms, with respiratory symptoms dominating the clinical
picture in 90% of cases. These are indistinguishable from other more
common causes of respiratory distress, in the post-traumatic and
post-operative period, due to the lack of reliable diagnostic labo-
ratory tests. Thoracic imaging, particularly thoracic computed to-
mography (CT), is the imaging modality of choice in the
investigation of pulmonary complications, playing an important
role in the diagnosis of FES and alternative differentials. The aim of
this article is to provide a comprehensive review of the clinical and
imaging manifestations of FES and the underlying pathological
process behind the imaging appearances.
2. Definition and classification
Long bone fractures and orthopedic surgery disrupt intra-
medullary fat in 95% of cases but only a minority of patients
develop clinical symptoms [2]. Therefore, a distinction needs to be
made between fat embolism and FES [3]. Fat embolism simply
implies the presence of fat droplets in the systemic circulation
which may be detected in the blood and urine of nearly all patients
with long bone fractures [4]. FES in contrast, is a clinical diagnosis
that requires the presence of at least one major and four minor
criteria, as proposed by Gurd and Wilson in 1974 (Table 1). These
criteria remain widely used today and are centred on a triad of
hypoxia (95%), confusion (60%) and petechial rash (33%) [5]. The
classic triad is not present in all patients, producing a clinical pic-
ture that can vary from an indolent process to fulminant respiratory
failure [3].
3. Epidemiology
FES is relatively rare with reported incidence in the setting of
trauma of 1e2% [6]. The incidence is higher in bilateral femoral
fractures (4.8%e7.5%), and following intra-medullary nail fixation
(up to 11%) [7]. There are reports suggesting that the risk of FES is
higher if the timing of intra-medullary nailing is delayed [8]. It is
unclear whether this is compounded by the complex nature of
orthopedic surgery in multi-trauma patients who have multiple
injuries and therefore are already at elevated risk of FES [9]. Non-
traumatic FES account for 5% of cases and causes include severe
pancreatitis, prolonged corticosteroid therapy and sickle cell dis-
ease, among others [10].
4. Pathophysiology
The complexity and unpredictability of FES has resulted in
numerous attempts to better understand the pathological process,
in particular why some individuals suffer severe pulmonary
distress while others are spared.
4.1. The mechanical theory
The mechanical theory, first described by Gauss in 1924 [11],
proposes that an increase in intramedullary pressure forces
marrow fat into the circulation via disrupted venous channels at
the fracture site. Embolization of fat particles cause direct
obstruction of the pulmonary capillaries, leading to a ventilation-
perfusion mismatch, low oxygen saturation and dyspnea, with
severity proportional to the quantity of fat present in the circulation
[12].
4.2. The biochemical theory
The biochemical theory described in 1927 by Lehmann and
Moore [13]postulated that fat emboli trapped in the small capil-
laries [14] release free fatty acids and glycerol, which are toxic to
the lung and initiate an inflammatory cascade resulting in localized
endothelial injury, permeability edema and hemorrhage. This may
result as diffuse alveolar wall damage and ultimately manifest as
adult respiratory distress syndrome (ARDS) [14]. The time delay to
Table 1
Gurd's criteria. Major and minor criteria of fat embolism syndrome [5].
Fat embolism syndrome (FES ¼ 1 major þ 4 minor)
Major criteria Minor criteria
Hypoxia (60 mmHg O2)
Confusion
Petechial rash
Pyrexia (39
C)
Tachycardia (120 beats per minute)
Retinal changes (petechiae)
Anuria or Oliguria
Anemia (hemoglobin drop 20%)
Thrombocytopenia (drop 50%)
High ESR (71 mm per hour)
Fat macroglobulinemia
Abbreviations: mmHg ¼ milimiters of mercury, O2 ¼ oxygen, ESR ¼ eythrocyte
sedimentation rate.
K. Newbigin et al. / Respiratory Medicine 113 (2016) 93e10094

3. trigger this biochemical cascade potentially explains the variable
delay between release of fat and clinical symptoms.
5. Diagnosis
There is no gold standard diagnostic test for the diagnosis of FES
[15]. The diagnosis can be achieved in most cases through the
combination of the classic clinical syndrome in the appropriate
setting together with typical imaging manifestations [16], [17].
6. Clinical manifestations
FES frequently presents in a subacute manner with symptoms
peaking at 48e72 h after injury [18]. Isolated transient respiratory
distress can be attributed to other causes of respiratory distress in
the post-traumatic setting including pulmonary contusion, aspi-
ration, pulmonary edema and acute pulmonary embolism. An
attempt to stratify the probability of FES has been made by
Schonfeld et al., with points given for each criteria met (Table 2)
[19]. This system allocates the highest number of points to the
classic feature of petechial rash, which typically occurs over the
upper anterior torso, axillary regions, and conjunctiva (Fig. 1a,b).
However, petechial rash does not appear until 3e5 days after the
onset of respiratory symptoms and is therefore not helpful in
establishing an early diagnosis [20].
In most cases, fulminant FES present with severe respiratory
distress symptoms. It often develops in the first 12 h after surgery, a
phenomenon likely related to high burden and larger size of emboli
resulting in acute pulmonary hypertension and right heart failure
[21,22]. The high pulmonary arterial pressure displaces fat droplets
across the pulmonary capillary bed, shunting them into the sys-
temic circulation which may lead to cerebral fat embolism mani-
festing as confusion and loss of consciousness [23]. The importance
of a patent foramen ovale in contributing to the development and
severity of cerebral fat embolism is controversial, with more recent
studies downplaying early reports of higher mortality rates in this
group [24e26]. Risk factors for severe FES have been described and
include high-velocity trauma, surgical delay of more than 10 h,
fixation of multiple fractures during the same procedure, fixation of
pathological fractures [9,27] as well as presence of contused lung.
The latter may occur because the non-contused lung may receive a
relative higher burden of fat emboli due to physiological shunting
[28].
7. Biochemical testing and bronchoalveolar lavage in FES
Laboratory tests are often abnormal in patients with FES, how-
ever there is no single reliable diagnostic test. There is a common
misconception that presence of fat globules in the urine or
increased serum lipase implies a diagnosis of FES. The demon-
stration of fat in the systemic circulation, either through urinalysis,
elevated serum lipase or serum phospholipase-A2, a breakdown
product of lipase, is however non-specific particularly in the setting
of multi-trauma [29,30]. Quantification of cells containing fat
droplets on bronchoalveolar lavage within 24 h after trauma has
been suggested as a potential tool to aid in the diagnosis of FES or to
help distinguish fulminant FES from other causes of ARDS [31].
However, the presence of fat globules within pulmonary macro-
phages is also non-specific and can be present in the setting of
multi-organ failure, sepsis or when lipid infusions are part of
nutritional regimes [32].
8. Thoracic imaging manifestations
8.1. Chest radiograph
Chest radiograph is usually the first imaging modality acquired
in patients with respiratory distress. In their sentinel paper Gurd
and Wilson described bilateral diffuse or patchy ill-defined opaci-
ties on the chest radiographs of 43 of the 52 patients with FES
(Fig. 2) [5], however these radiographic findings are non-specific
and may be indistinguishable from pulmonary edema, aspiration
or infection. Therefore the usefulness of chest radiographs in the
initial diagnosis of FES is limited, however it is a useful modality for
Table 2
Schonfeld's Criteria: points were allocated to symptoms known to be associated
with FES. Five or more points are required for diagnosis.
Schonfeld's criteria [19]
Symptoms Points
Diffuse petechiae 5
Alveolar infiltrates on chest radiograph 4
Hypoxemia (70 mmHg) 3
Confusion 1
Fever 38
C 1
Heat rate 120 beats per minute 1
Respiratory rate 30 breaths per minute 1
Fig. 1. Petechiae in a patient with FES. 38-year old man who presented with shortness
of breath, oxygen desaturation and petechiae involving the conjunctivae (a) as well as
the typical rash in the axillary regions, upper anterior and lateral chest wall (b).
K. Newbigin et al. / Respiratory Medicine 113 (2016) 93e100 95

4. monitoring disease progression. A study evaluating serial chest
radiographs in 20 patients with FES and abnormal chest radio-
graphs at the time of diagnosis showed resolution at one week in
50% of patients and complete resolution within two weeks in the
remainder of the cases [33,34].
8.2. Computed tomography of the chest
CT of the chest and in particular high-resolution CT (HRCT), is
the modality of choice to assess the lung parenchyma in suspected
cases of FES [35]. It may not only suggest the diagnosis but also
indicate alternate causes for respiratory distress.
8.2.1. Patchy ground-glass opacities and consolidations
The most commonly described pattern is patchy ground-glass
opacities, often associated with smooth interlobular septal thick-
ening, the so-called crazy-paving pattern (Fig. 3). Distinct demar-
cation between the affected and normal lung with areas of lobular
sparing has also been described (Fig. 4) [35]. This geographic dis-
tribution presumably reflects variations in lung perfusion at time of
embolization with under-perfused lobules spared from the embolic
event [36,37]. Airspace consolidation has also been reported,
apparently in more severe cases of FES, reflecting more extensive
pulmonary hemorrhage and edema [38]
A direct correlation between the extent of airspace opacities and
the severity of pulmonary symptoms has been suggested, likely
reflecting the extent of acute lung injury causing hemorrhage and
edema (Fig. 5) [39]. Malagari et al. described a predominant patchy
ground-glass pattern on HRCT in patients with mild pulmonary FES,
whereas Arakawa et al. demonstrated a direct correlation between
the extent of lung involvement and the severity of hypoxemia
[35,38]. The temporal evolution of imaging findings correlates with
the progression of respiratory symptoms, with both symptoms and
extent of the disease peaking approximately 48 h after the post-
traumatic release of fat [40].
8.2.2. Small centrilobular nodules
A less frequent imaging pattern includes small (10 mm), ill-
defined centrilobular nodules, typically described in a predomi-
nantly peripheral and upper lobe distribution (Fig. 6) [34,41] and
often located in the subpleural regions and along the interlobular
septa [42]. Nodules have been described in the earlier phases of the
syndrome and allegedly reflect the vasculogenic origin of me-
chanical obstruction by fat globules with subsequent resolution.
This hypothesis is supported by animal models that demonstrated
nodules at 2 h after embolism compared to ground-glass changes at
24 h [35,41].
8.2.3. Other CT findings
Small bilateral pleural effusions have also been reported.
Notably, although a few reports have described the direct visuali-
zation of macroscopic fat emboli within the pulmonary vasculature
this is an extremely uncommon finding [43,44].
Complete resolution of CT findings is expected to occur by
approximately two weeks [34,35]. The description of chronic
sequelae is limited to case reports and animal models [45]. There
has been one case report of diffuse micro-calcification with sub-
segmental intravascular distribution appearing after six weeks in a
patient with severe FES. These calcifications were associated with
adjacent thickening of the bronchovascular bundles, parenchymal
distortion and traction bronchiectasis on subsequent HRCT [46].
Fig. 2. Fat embolism syndrome in a 23-year-old-man with bilateral femoral fractures
presenting with hypoxia and confusion. Chest radiograph demonstrates patchy, ill-
defined opacities in the perihilar and lower lobe regions (arrows). These findings
were monitored with plain radiographs and resolved completely in one week.
Fig. 3. 66-year-old man with shortness of breath and unexplained drop in hemoglobin
after intramedullary nail fixation of a right femoral fracture. Axial HRCT image shows
diffuse ground-glass opacities and interlobular septal thickening causing crazy-paving
pattern in the anterior aspects of the lungs. Patchy consolidations are noted with a
dependent distribution.
Fig. 4. 18-year-old man presenting with acute respiratory distress, 48 h after intra-
medullary nail fixation of a femoral fracture. Axial HRCT image demonstrates bilateral
ground-glass opacities and interlobular septal thickening resulting in the crazy-paving
pattern. Note the sharp margins between the parenchymal abnormalities and areas of
normal lung (white arrows) with areas of lobular sparing (black arrows).
K. Newbigin et al. / Respiratory Medicine 113 (2016) 93e10096

5. Fig. 7 shows a case of severe FES which progressed to chronic
changes of fibrosis with traction bronchiectasis and architectural
distortion 2 months after the initial CT. Animal models have
demonstrated increased collagen deposition in the arterioles and
pulmonary septa six weeks after injection with triolein, a lipid-
based molecule used to reproduce FES in animal models [45].
Nonetheless, the frequency of chronic sequelae is thought to be
extremely rare although the exact incidence remains unknown.
9. Differential diagnosis of FES on chest CT
The differential diagnosis of parenchymal findings of traumatic
FES on CT includes several entities such as pulmonary contusion,
pulmonary edema, thromboembolic pulmonary embolism, aspira-
tion and pneumonia. Typical imaging findings and distinguishing
imaging features for each diagnosis are summarized in Table 3.
10. Imaging findings in cerebral fat embolism
Cerebral manifestations of FES occur to some degree in the
majority of fat embolism patients [48]. Clinical manifestations of
cerebral fat embolism are nonspecific and vary from confusion to
encephalopathy, but may include headache, seizures and coma
[49]. Magnetic resonance imaging (MR) has been reported to be the
most sensitive modality to diagnose cerebral fat embolism [48,49],
suggesting it may be wise to perform cerebral MR in patients with
acute alteration in mental status after orthopedic surgery or trauma
in the presence of a normal cerebral CT scan [50].
Typical MR findings include multiple small, scattered, non-
confluent, hyperintense lesions on T2-weighted images, that
involve both the grey and white matter [49] (Fig. 8a). The number
and size of the lesions is variable but correlates with the degree of
neurological disability as measured by the Glasgow Coma Scale
score [51]. These lesions may appear as early as 4 h after the onset
Fig. 5. 19-year-old man presenting with pronounced hypoxia following femoral nail
fixation three days after trauma, progressing to respiratory failure requiring respiratory
and circulatory support. Axial HRCT image demonstrates diffuse ground-glass opacities
with areas of lobular sparing (arrows).
Fig. 6. 23-year-old man with bilateral femoral fractures and spinal injury after falling
eight stories. The patient developed acute severe hypoxia and confusion requiring
intubation after intramedullary nail fixation of femoral fractures. Axial CT image per-
formed 2 h after surgery demonstrates bilateral centrilobular nodules, more numerous
in the right upper and superior segments of the right lower lobe. Note of atelectasis of
the left lower lobe.
Fig. 7. a. 67-year-old man with displaced right acetabular fracture and fracture-
dislocation of the right hip secondary to a motor vehicle accident, presenting with
acute respiratory failure two days after internal fixation of the fractures. (a) Axial CT
image demonstrates diffuse ground-glass opacities with areas of lobular sparing (white
arrows) and small peripheral consolidation in the right middle lobe (black arrow). b.
Two months after the initial presentation, same patient as in Fig. 7a, developed hyp-
oxic arrest and bedside echocardiogram demonstrated enlargement of the right
ventricle. Contrast-enhanced CT angiography performed to exclude pulmonary em-
bolism demonstrated coarse reticulation, traction bronchiectasis and architectural
distortion bilaterally, compatible with fibrotic changes.
K. Newbigin et al. / Respiratory Medicine 113 (2016) 93e100 97

6. of symptoms, and gradually disappear within a few weeks to a few
months [49,51]. However, lesions could be detected as early as 1 h
after symptom onset on diffusion-weighted images (DWI). On DWI,
multiple small, dot-like hyperintense foci are seen in a dark back-
ground, following a watershed distribution (Fig 8b) resulting in a
“starfield” appearance [49,52]. The DWI findings are thought to
represent foci of cytotoxic edema in acute cerebral infarcts arising
from fat emboli that occlude cerebral arterioles [48,52e56]. Free
fatty acids have been shown to be highly toxic to brain tissue and
may be the cause of vasogenic edema as well [55].
11. Clinical course and prognosis
FES is generally a self-resolving entity with good outcome. The
prognosis of FES has markedly improved with advances in critical
care medicine. The mortality rate reported by Bulger in a ten-year
review from 1985 was 7%, significantly improved from earlier
studies, which quoted 15e20% mortality rate [3,5,57]. Respiratory
failure is the most common cause of death while severe cerebral
edema is a contributing factor in many cases [58]. Severe cases may
progress to ARDS resulting in hypoxic arrest [59]. In cases of mild
FES, the prognosis is excellent with complete clinical and radio-
logical resolution of pulmonary manifestations within a few weeks
from the traumatic event [33,60]. Although neurological deficit
confers a poorer prognosis, neurological recovery is also usually
complete with residual lingering deficits reported only in severe
cases [61].
12. Management and treatment
There is no specific treatment for FES and supportive care is the
mainstay therapy for FES. In patients with hypoxemia and severe
respiratory distress mechanical ventilation is often required to
maintain arterial oxygen saturation aimed at minimizing further
lung damage [62]. In severely hypovolemic patients fluid resusci-
tation to limit or prevent shock states is required63 16
. Management
of right heart failure remains challenging, with inotropic support
and volume being the cornerstone of management [64]. In cases of
central nervous system dysfunction, frequent neurological exami-
nations and intracranial pressure monitoring may be necessary
[65].
Use of corticosteroids and heparin has been suggested as
possible treatments but remains controversial and has not shown
to reduce morbidity or mortality [63]. Corticosteroids may be
considered in cases of fulminant FES however; there is not enough
evidence to support its routine administration in the majority of
patients. Heparin increases lipase activity and may improve clear-
ance of lipids from the blood stream, however it also results in an
increase in free fatty acids, which could worsen the underlying
inflammatory cascade. Moreover, heparin may be considered un-
safe in the setting of trauma or surgery [63,66].
13. Prevention
Surgical timing and technique plays a role as preventive inter-
vention, with lower incidence of FES in early intramedullary nailing
with a five-fold decrease in the rate of FES if surgery is performed in
the first 24 h [8].
Corticosteroid administration has been proposed as a preventive
measure and has shown a decrease in the incidence of FES [67]. A
meta-analysis of 389 patients, demonstrated that prophylactic
corticosteroid administration reduced the risk of FES with a
number-needed eto-treat of eight to prevent one case of FES [67].
However, there is no published benefit in mortality and larger
randomized trials are required to obtain further confirmatory evi-
dence and to guide patient care [67].
14. Conclusion
Pulmonary FES is an uncommon but important cause of
morbidity following long bone trauma and subsequent orthopedic
surgery. The clinical diagnosis remains difficult and is often delayed
due to non-specific clinical manifestations overlapping with other
causes of post-traumatic or post-operative respiratory distress. The
absence of a specific diagnostic laboratory test further hinders a
timely diagnosis [15]. Although imaging findings are deemed non-
specific, chest CT plays an important role in the diagnostic work-up
of patients with suspected FES. Specifically, the presence of diffuse
well demarcated ground glass opacities or ill-defined centrilobular
nodules on chest CT are suggestive of the diagnosis in the appro-
priate surgical or post traumatic scenario. It is paramount that ra-
diologists, orthopedic surgeons and clinicians are familiar with
these clinical and imaging manifestations of FES to allow early
diagnosis and treatment.
Table 3
Differential Diagnosis for FES and distinguishing CT imaging features.
Differential diagnosis of FES Typical clinical and imaging feature Distinguishing imaging features of FES
Pulmonary contusion [37] - Develops at the time or shortly after trauma (within 6 h).
- Localized or multifocal GGO or consolidation with distribution related to
mechanism of trauma.
- Traumatic pneumatoceles/laceration often present.
- GGO and consolidation 24e72 h after trauma/
orthopedic surgery.
- Septal lines and lobular sparing
- Often upper lobes and dependent distribution.
Pulmonary edema [35] -Bilateral symmetrical septal lines, GGO and peribronchial cuffing, vascular
engorgement and pleural effusions.
- Patchy often asymmetrical.
- No peribronchial cuffing.
-Normal bronchus/artery diameter ratio.
Thromboembolic pulmonary
embolism [43,44]
-Intraluminal filling defects within pulmonary arteries with soft tissue
attenuation.
-No significant parenchymal abnormalities.
- Filling defects with fat (low) attenuation (extremely
uncommon finding).
- Predominance of parenchymal findings.
Aspiration Pneumonitis [34] -Small centrilobular nodules tree-in-bud opacities, peribronchial opacities in
the dependent lungs.
- Develop within a few hours after aspiration.
- Patients with reduced level of consciousness and head trauma.
- Tree-in-bud opacities uncommon in FES.
- Centrilobular nodules more common in the upper
lobes.
Pneumonia [47] -Fever and other infectious signs/symptoms.
- GGO or consolidation, localized or patchy, no zonal predominance. It may have
tree-in-bud opacities.
- Septal lines are uncommon.
- GGO and consolidation with areas of lobular sparing
(geographic pattern).
-Septal lines commonly seen.
Abbreviations; FES ¼ fat embolism syndrome; GGO ground-glass opacities.
K. Newbigin et al. / Respiratory Medicine 113 (2016) 93e10098

7. Conflict of interest statement
Dr. Souza discloses consultant fees for Pfizer and Boehringer
Ingelheim
Dr. Pe~na discloses a GE Healthcare lecture fee.
The remainder of the authors have no conflict of interests to
disclose.
Acknowledgments
None.
References
[1] Z. FA, Beitrage zur anatomie und physiologic der lunge, 1861.
[2] D.B. Allardyce, R.N. Meek, B. Woodruff, M.M. Cassim, D. Ellis, Increasing our
knowledge of the pathogenesis of fat embolism: a prospective study of 43
patients with fractured femoral shafts, J. Trauma 14 (1974) 955e962.
[3] E.M. Bulger, D.G. Smith, R.V. Maier, G.J. Jurkovich, Fat embolism syndrome. A
10-year review, Arch. Surg. 132 (1997) 435e439.
[4] E.J. Riseborough, J.H. Herndon, Alterations in pulmonary function, coagulation
and fat metabolism in patients with fractures of the lower limbs, Clin. Orthop.
Relat. Res. (1976) 248e267.
[5] A.R. Gurd, R.I. Wilson, The fat embolism syndrome, J. Bone Jt. Surg. Br. 56B
(1974) 408e416.
[6] C. Muller, B.A. Rahn, U. Pfister, R.P. Meinig, The incidence, pathogenesis,
diagnosis, and treatment of fat embolism, Orthop. Rev. 23 (1994) 107e117.
[7] G.M. Kontakis, T. Tossounidis, K. Weiss, H.C. Pape, P.V. Giannoudis, Fat em-
bolism: special situations bilateral femoral fractures and pathologic femoral
fractures, Injury 37 (Suppl. 4) (2006) S19eS24.
[8] L.B. Bone, K.D. Johnson, J. Weigelt, R. Scheinberg, Early versus delayed stabi-
lization of femoral fractures. A prospective randomized study, J. Bone Jt. Surg.
Am. Vol. 71 (1989) 336e340.
[9] R.C. Talucci, J. Manning, S. Lampard, A. Bach, C.J. Carrico, Early intramedullary
nailing of femoral shaft fractures: a cause of fat embolism syndrome, Am. J.
Surg. 146 (1983) 107e111.
[10] D. Levy, The fat embolism syndrome. A review, Clin. Orthop. Relat. Res. (1990)
281e286.
[11] H. Gauss, Pathology of fat embolism, Arch. Surg. 9 (1924) 593e603.
[12] A.C. Pell, J. Christie, J.F. Keating, G.R. Sutherland, The detection of fat embolism
by transoesophageal echocardiography during reamed intramedullary nailing.
A study of 24 patients with femoral and tibial fractures, J. Bone Jt. Surg. Br. 75
(1993) 921e925.
[13] P. Lehman EaM, Fat embolism including experimental production without
trauma, Arch. Surg. 14 (1927) 621e662.
[14] G. Hulman, The pathogenesis of fat embolism, J. Pathol. 176 (1995) 3e9.
[15] T.M. Dudney, C.G. Elliott, Pulmonary embolism from amniotic fluid, fat, and
air, Prog. Cardiovasc. Dis. 36 (1994) 447e474.
[16] E. Kosova, B. Bergmark, G. Piazza, Fat embolism syndrome, Circulation 131
(2015) 317e320.
[17] A.R. Gurd, Fat embolism: an aid to diagnosis, J. Bone Jt. Surg. Br. 52 (1970)
732e737.
[18] J.D. Prologo, V. Dogra, R. Farag, CT diagnosis of fat embolism, Am. J. Emerg.
Med. 22 (2004) 605e606.
[19] S.A. Schonfeld, Y. Ploysongsang, R. DiLisio, J.D. Crissman, E. Miller,
D.E. Hammerschmidt, H.S. Jacob, Fat embolism prophylaxis with corticoste-
roids. A prospective study in high-risk patients, Ann. Intern. Med. 99 (1983)
438e443.
[20] S.S. Tachakra, Distribution of skin petechiae in fat embolism rash, Lancet 1
(1976) 284e285.
[21] S.R. Hagley, The fulminant fat embolism syndrome, Anaesth. Intensiv. Care 11
(1983) 167e170.
[22] D.P. O'Toole, D. Gilroy, I.M. Bali, Fulminant fat embolism associated with
closed fracture reduction, Ulst. Med. J. 55 (1986) 86e88.
[23] V. Palmovic, J.R. McCarroll, Fat embolism in trauma, Arch. Pathol. 80 (1965)
630e635.
[24] H.E. Emson, Fat embolism studied in 100 patients dying after injury, J. Clin.
Pathol. 11 (1958) 28e35.
[25] M.W. Nijsten, J.P. Hamer, H.J. ten Duis, J.L. Posma, Fat embolism and patent
foramen ovale, Lancet 1 (1989) 1271.
[26] T.C. Fabian, Unravelling the fat embolism syndrome, N. Engl. J. Med. 329
(1993) 961e963.
[27] S.J. Pinney, J.F. Keating, R.N. Meek, Fat embolism syndrome in isolated femoral
fractures: does timing of nailing influence incidence? Injury 29 (1998)
131e133.
[28] H.C. Pape, M. Auf'm'Kolk, T. Paffrath, G. Regel, J.A. Sturm, H. Tscherne, Primary
intramedullary femur fixation in multiple trauma patients with associated
lung contusionea cause of posttraumatic ARDS? J. Trauma 34 (540-7; dis-
cussion) (1993) 547e548.
[29] A.R. Gurd, A method of demonstrating the change in serum fat in patients
with the fat embolism syndrome, Br. J. Surg. 56 (1969) 381.
[30] A.P. Ross, The value of serum lipase estimations in the fat embolism syn-
drome, Surgery 65 (1969) 271e273.
[31] J. Chastre, J.Y. Fagon, P. Soler, A. Fichelle, M.C. Dombret, D. Huten, A.J. Hance,
Fig. 8. a. Cerebral fat embolism in a patient with decreased level of consciousness after
a motor vehicle accident and femoral fracture. (a) Axial fluid-attenuated inversion
recovery (FLAIR) image of the brain demonstrates multiple tiny hyperintense foci
distributed in a linear fashion along the centrum semiovale bilaterally (arrows). b.
Diffusion-weighted imaging (DWI), reveals multiple lesions of high signal intensity
(arrows) in a watershed distribution, along the bilateral centrum semiovale, indicating
restricted diffusion due to cytotoxic edema.
K. Newbigin et al. / Respiratory Medicine 113 (2016) 93e100 99

8. C. Gibert, Bronchoalveolar lavage for rapid diagnosis of the fat embolism
syndrome in trauma patients, Ann. Intern. Med. 113 (1990) 583e588.
[32] J.M. Vedrinne, C. Guillaume, M.C. Gagnieu, P. Gratadour, C. Fleuret, J. Motin,
Bronchoalveolar lavage in trauma patients for diagnosis of fat embolism
syndrome, Chest 102 (1992) 1323e1327.
[33] N. Muangman, E.J. Stern, E.M. Bulger, G.J. Jurkovich, F.A. Mann, Chest radio-
graphic evolution in fat embolism syndrome, J. Med. Assoc. Thail. ¼ Chot-
maihet Thangphaet 88 (2005) 1854e1860.
[34] X. Gallardo, E. Castaner, J.M. Mata, J. Rimola, J. Branera, Nodular pattern at
lung computed tomography in fat embolism syndrome: a helpful finding,
J. Comput. Assist. Tomogr. 30 (2006) 254e257.
[35] K. Malagari, N. Economopoulos, C. Stoupis, Z. Daniil, S. Papiris, N.L. Muller,
D. Kelekis, High-resolution CT findings in mild pulmonary fat embolism, Chest
123 (2003) 1196e1201.
[36] A.K. Bartelink, E.S. Boon, L.M. Budel, Clinical picture: asymmetrical fat em-
bolism, Lancet 358 (2001) 795.
[37] P.R. Miller, M.A. Croce, T.K. Bee, W.G. Qaisi, C.P. Smith, G.L. Collins, T.C. Fabian,
ARDS after pulmonary contusion: accurate measurement of contusion volume
identifies high-risk patients, J. Trauma 51 (223-8; discussion) (2001)
229e230.
[38] H. Arakawa, Y. Kurihara, Y. Nakajima, Pulmonary fat embolism syndrome: CT
findings in six patients, J. Comput. Assist. Tomogr. 24 (2000) 24e29.
[39] S.E. Rossi, P.C. Goodman, T. Franquet, Nonthrombotic pulmonary emboli, AJR
Am. J. Roentgenol. 174 (2000) 1499e1508.
[40] A.P. Ross, The fat embolism syndrome: with special reference to the impor-
tance of hypoxia in the syndrome, Ann. R. Coll. Surg. Engl. 46 (1970) 159e171.
[41] L.E. Heyneman, N.L. Muller, Pulmonary nodules in early fat embolism syn-
drome: a case report, J. Thorac. Imaging 15 (2000) 71e74.
[42] F.G. Van den Brande, S. Hellemans, A. De Schepper, R. De Paep, B. Op De Beeck,
H.R. De Raeve, P.G. Jorens, Post-traumatic severe fat embolism syndrome with
uncommon CT findings, Anaesth. Intensiv. Care 34 (2006) 102e106.
[43] J.G. Ravenel, L.E. Heyneman, H.P. McAdams, Computed tomography diagnosis
of macroscopic pulmonary fat embolism, J. Thorac. Imaging 17 (2002)
154e156.
[44] G. Nucifora, F. Hysko, A. Vit, A. Vasciaveo, Pulmonary fat embolism: common
and unusual computed tomography findings, J. Comput. Assist. Tomogr. 31
(2007) 806e807.
[45] A.M. Poisner, F. Adler, B. Uhal, T.E. McIff, J.P. Schroeppel, A. Mehrer,
B.L. Herndon, K.M. Lankachandra, A. Molteni, Persistent and progressive
pulmonary fibrotic changes in a model of fat embolism, J. Trauma Acute Care
Surg. 72 (2012) 992e998.
[46] J. Hamrick-Turner, P.L. Abbitt, R.B. Harrison, P.E. Cranston, Diffuse lung cal-
cifications following fat emboli and adult respiratory distress syndromes: CT
findings, J. Thorac. Imaging 9 (1994) 47e50.
[47] T. Franquet, Imaging of pneumonia: trends and algorithms, Eur. Respir. J. 18
(2001) 196e208.
[48] Y. Zaitsu, S. Terae, K. Kudo, K.K. Tha, M. Hayakawa, N. Fujima, D. Yoshida,
A. Tsukahara, H. Shirato, Susceptibility-weighted imaging of cerebral fat
embolism, J. Comput. Assist. Tomogr. 34 (2010) 107e112.
[49] P.M. Parizel, H.E. Demey, G. Veeckmans, F. Verstreken, P. Cras, P.G. Jorens,
A.M. De Schepper, Early diagnosis of cerebral fat embolism syndrome by
diffusion-weighted MRI (starfield pattern), Stroke 32 (2001) 2942e2944.
[50] J.J. Chen, J.C. Ha, S.E. Mirvis, MR imaging of the brain in fat embolism syn-
drome, Emerg. Radiol. 15 (2008) 187e192.
[51] M. Takahashi, R. Suzuki, Y. Osakabe, J.I. Asai, T. Miyo, G. Nagashima,
T. Fujimoto, Y. Takahashi, Magnetic resonance imaging findings in cerebral fat
embolism: correlation with clinical manifestations, J. Trauma 46 (1999)
324e327.
[52] C.W. Ryu, D.H. Lee, T.K. Kim, S.J. Kim, H.S. Kim, J.H. Lee, C.G. Choi, D.C. Suh,
Cerebral fat embolism: diffusion-weighted magnetic resonance imaging
findings, Acta Radiol. 46 (2005) 528e533.
[53] G.B. Marshall, V.R. Heale, L. Herx, A. Abdeen, L. Mrkonjic, J. Powell, R.J. Sevick,
W. Morrish, Magnetic resonance diffusion weighted imaging in cerebral fat
embolism, Can. J. Neurol. Sci. 31 (2004) 417e421.
[54] D.J. Butteriss, D. Mahad, C. Soh, T. Walls, D. Weir, D. Birchall, Reversible
cytotoxic cerebral edema in cerebral fat embolism, AJNR Am. J. Neuroradiol.
27 (2006) 620e623.
[55] H.J. Kim, J.H. Lee, C.H. Lee, S.H. Lee, T.Y. Moon, B.M. Cho, H.K. Kim, B.R. Park,
K.H. Chang, Experimental cerebral fat embolism: embolic effects of triolein
and oleic acid depicted by MR imaging and electron microscopy, AJNR Am. J.
Neuroradiol. 23 (2002) 1516e1523.
[56] A.D. Simon, J.L. Ulmer, J.M. Strottmann, Contrast-enhanced MR imaging of
cerebral fat embolism: case report and review of the literature, AJNR Am. J.
Neuroradiol. 24 (2003) 97e101.
[57] J.P. Moreau, Fat embolism: A review and report of 100 cases, Can. J. Surg. J.
Can. Chir. 17 (1974) 196e199.
[58] E.E. Etchells, D.T. Wong, G. Davidson, P.L. Houston, Fatal cerebral fat embolism
associated with a patent foramen ovale, Chest 104 (1993) 962e963.
[59] E.E. Husebye, T. Lyberg, O. Roise, Bone marrow fat in the circulation: clinical
entities and pathophysiological mechanisms, Injury 37 (Suppl. 4) (2006)
S8eS18.
[60] J.W. O'Higgins, Fat embolism, Br. J. Anaesth. 42 (1970) 163e168.
[61] J.E. Thomas, D.R. Ayyar, Systemic fat embolism. A diagnostic profile in 24
patients, Arch. Neurol. 26 (1972) 517e523.
[62] P. Glover, L.I. Worthley, Fat embolism, Crit. Care Resusc. 1 (1999) 276e284.
[63] N.M. Habashi, P.L. Andrews, T.M. Scalea, Therapeutic aspects of fat embolism
syndrome, Injury 37 (Suppl. 4) (2006) S68eS73.
[64] S. Akhtar, Fat embolism, Anesthesiol. Clin. 27 (2009) 533e550 table of
contents.
[65] R.G. Kellogg, R.B. Fontes, D.K. Lopes, Massive cerebral involvement in fat
embolism syndrome and intracranial pressure management, J. Neurosurg. 119
(2013) 1263e1270.
[66] A. Mellor, N. Soni, Fat embolism, Anaesthesia 56 (2001) 145e154.
[67] S.S. Bederman, M. Bhandari, M.D. McKee, E.H. Schemitsch, Do corticosteroids
reduce the risk of fat embolism syndrome in patients with long-bone frac-
tures? A meta-analysis, Can. J. Surg. J. Can. Chir. 52 (2009) 386e393.
K. Newbigin et al. / Respiratory Medicine 113 (2016) 93e100100