Massive cerebral involvement in fat embolism syndrome and intracranial pressure management

Case report

Full access

Fat embolism syndrome (FES) is a common clinical entity that can occasionally have significant neurological sequelae. The authors report a case of cerebral fat embolism and FES that required surgical management of intracranial pressure (ICP). They also discuss the literature as well as the potential need for neurosurgical management of this disease entity in select patients. A 58-year-old woman presented with a seizure episode and altered mental status after suffering a right femur fracture. Head CT studies demonstrated hypointense areas consistent with fat globules at the gray-white matter junction predominantly in the right hemisphere. This CT finding is unique in the literature, as other reports have not included imaging performed early enough to capture this finding. Brain MR images obtained 3 days later revealed T2-hyperintense areas with restricted diffusion within the same hemisphere, along with midline shift and subfalcine herniation. These findings steered the patient to the operating room for decompressive hemicraniectomy. A review of the literature from 1980 to 2012 disclosed 54 cases in 38 reports concerning cerebral fat embolism and FES. Analysis of all the cases revealed that 98% of the patients presented with mental status changes, whereas only 22% had focal signs and/or seizures. A good outcome was seen in 57.6% of patients with coma and/or abnormal posturing on presentation and in 90.5% of patients presenting with mild mental status changes, focal deficits, or seizure. In the majority of cases ICP was managed conservatively with no surgical intervention. One case featured the use of an ICP monitor, while none featured the use of hemicraniectomy.

Abbreviations used in this paper:DWI = diffusion-weighted imaging; FES = fat embolism syndrome; ICP = intracranial pressure; TBI = traumatic brain injury.

Abstract

Fat embolism syndrome (FES) is a common clinical entity that can occasionally have significant neurological sequelae. The authors report a case of cerebral fat embolism and FES that required surgical management of intracranial pressure (ICP). They also discuss the literature as well as the potential need for neurosurgical management of this disease entity in select patients. A 58-year-old woman presented with a seizure episode and altered mental status after suffering a right femur fracture. Head CT studies demonstrated hypointense areas consistent with fat globules at the gray-white matter junction predominantly in the right hemisphere. This CT finding is unique in the literature, as other reports have not included imaging performed early enough to capture this finding. Brain MR images obtained 3 days later revealed T2-hyperintense areas with restricted diffusion within the same hemisphere, along with midline shift and subfalcine herniation. These findings steered the patient to the operating room for decompressive hemicraniectomy. A review of the literature from 1980 to 2012 disclosed 54 cases in 38 reports concerning cerebral fat embolism and FES. Analysis of all the cases revealed that 98% of the patients presented with mental status changes, whereas only 22% had focal signs and/or seizures. A good outcome was seen in 57.6% of patients with coma and/or abnormal posturing on presentation and in 90.5% of patients presenting with mild mental status changes, focal deficits, or seizure. In the majority of cases ICP was managed conservatively with no surgical intervention. One case featured the use of an ICP monitor, while none featured the use of hemicraniectomy.

Fat embolism syndrome was first described by Zenker in 1861 and is characterized by the clinical triad of petechiae, neurological dysfunction, and pulmonary involvement.56 While fat embolism is defined as simply the presence of fat globules within the blood or any parenchymal structure, FES is a complex and catastrophic clinical syndrome seen in only a fraction of fat embolism cases.8 There is no universally accepted definition of the syndrome, although Gurd and Wilson proposed the first set of diagnostic criteria that are still widely in use today.26 Under their system, a diagnosis of FES could be made if 1 major feature and 4 minor features plus fat macroglobulinemia were present (Table 1). This clinical syndrome typically occurs shortly after long bone fractures but has also been described during different orthopedic procedures and cardiac surgery.23 Risk factors for the development of FES include multiple long bone fractures, femoral shaft fractures, acute pancreatitis, diabetes mellitus, cardiopulmonary bypass, and parenteral infusion of lipids.34 The incidence of FES is difficult to estimate but has been reported in the literature as up to 10% in patients with long bone and pelvic fractures.8

TABLE 1:

Gurd's criteria for FES*

Major FeatureMinor FeatureLaboratory Findings
petechial rashpyrexiaanemia
respiratory insufficiencytachycardiathrombocytopenia
neurologic involvementretinal changeselevated ESR
jaundicefat macroglobulinemia
renal insufficiency

* ESR = erythrocyte sedimentation rate.

Despite a relatively elevated incidence, FES usually presents in a mild form. Cases requiring specialized neurosurgical consultation and particularly ICP management are thought to be rare and to usually evolve to an ominous outcome. We present a case of posttraumatic cerebral fat embolism with catastrophic neurological consequences as well as a review of current neurosurgical implications and treatment strategies. The number of patients requiring neurosurgical consultation may not be as small as generally thought, and a good outcome may be possible despite an initial presentation suggesting otherwise.

Case Report

History

A 58-year-old woman with a medical history of mixed connective tissue disease, previous left occipital lobe ischemic stroke, hypertension, and end-stage renal disease on dialysis presented to a local community hospital after a mechanical fall at home. She suffered no head trauma and was able to provide a history of the incident to emergency room staff at the time of her presentation. An initial survey was significant only for an intertrochanteric fracture of the right femur; neurological examination was unremarkable on admission. Six hours after admission, she experienced a generalized tonic-clonic seizure. Initial management included administration of benzodiazepines; however, she failed to return to her neurological baseline, respiratory insufficiency developed, and she was subsequently intubated and maintained on mechanical ventilation. She was transferred to the intensive care unit, and maintenance levetiracetam was started. Despite these measures, she still did not return to her neurological baseline after the clinical seizures subsided. Head CT scanning was performed 12 hours after the initial injury, revealing multiple hypodense areas at the gray-white matter junction predominantly in the right hemisphere (Fig. 1A and B). Findings in the left hemisphere consisted of only scattered small hypodensities in the high left frontal lobe, which was interpreted as pneumocephalus and attributed to an occult skull base fracture. Aside from the well-documented absence of head trauma, a thin-cut skull base CT study failed to demonstrate any such fracture. Radiodensity measurement of the intracranial hypodensities was found to be about 90 HU. Supportive treatment was maintained, as were prophylactic broad-spectrum antibiotics.

Fig. 1.
Fig. 1.

A and B: Axial head CTs obtained on postinjury Day 1 demonstrating hypodensities at the gray-white matter junction in multiple areas. C and D: Axial T2-weighted and FLAIR MR images obtained on postinjury Day 3 showing edema and midline shift.

The patient's neurological status failed to improve, and on postinjury Day 3 she underwent noncontrast MRI of the brain to further delineate the etiology of her continued comatose state (Fig. 1C and D). Magnetic resonance imaging demonstrated increased T2 and FLAIR signal throughout the right centrum semiovale and temporal lobe with 10 mm of midline shift to the left and subfalcine herniation. Diffusion-weighted imaging revealed widespread areas of restricted diffusion predominantly within the right hemisphere; a few scattered areas in the distribution of the left anterior and middle cerebral arteries were also seen. This MRI finding was interpreted as possible cerebritis due to the occult skull base fracture, and the patient was transferred to our institution for further management.

Examination

Upon arrival, she exhibited spontaneous eye opening and localized painful stimuli, and was hemiplegic on the left. She could not follow commands and her Glasgow Coma Scale score was 10T. Brainstem reflexes were intact. Laboratory abnormalities included only a platelet count of 77,000/μl and hemoglobin of 8 g/dl. After thoroughly reviewing her history, clinical examination, and imaging, we hypothesized that FES was the problem. Platelet transfusion with a goal of 100,000/μl and initial neurological resuscitation were initiated. Echocardiography did not demonstrate a patent foramen ovale. Brain MR angiography and chest CT angiography demonstrated no abnormalities of the circle of Willis, aortic arch, or origin of the great vessels.

Operation

Given the patient's failure to improve after initial resuscitation and the significant cerebral edema, 10-mm midline shift, and subfalcine herniation, she was taken to the operating room for decompressive hemicraniectomy and duraplasty (Fig. 2A). Intraoperatively, the only significant finding was diffuse swelling and hyperemia of the right cerebral hemisphere.

Fig. 2.
Fig. 2.

A: Operative photograph demonstrating ample frontotemporoparietal decompression and edematous, hyperemic brain parenchyma with sulcal effacement. B and C: Postoperative axial CTs demonstrating improvement in midline shift and effective decompression.

Postoperative Course

Postoperative CT demonstrated effective decompression (Fig. 2B). On postinjury Day 9 the patient was opening her eyes spontaneously and following commands on her right side. Tracheostomy and gastrostomy were performed on postinjury Day 11. The orthopedic service elected nonoperative management of the intertrochanteric fracture given the expected long bed rest period, and she was transferred to a long-term rehabilitation facility. At 6 months postinjury, she was at a specialized rehabilitation facility being weaned from her tracheostomy, speaking through a Passy-Muir valve, communicating in written form, and ambulating with assistance (modified Rankin Scale Score 4).

Literature Review

A review of the literature was conducted with the use of PubMed (http://www.ncbi.nlm.nih.gov/pubmed/). The search terms “cerebral AND fat AND embolism” and “fat AND embolism AND syndrome” were used. Only reports from the post-CT or post-MRI era were included; this period was arbitrarily defined as post-1980. Searches utilizing those 2 strings yielded a combined 129 results. Abstracts were screened, and only articles containing reports of patients with neurological manifestations of FES were analyzed (85 results). Every article was reviewed by 2 of the authors (R.G.K. and R.B.V.F.). Articles written in languages other than English, French, German, Portuguese, Italian, or Spanish; containing nonhuman subjects; and having missing or vague clinical descriptions, management details, or imaging data were excluded. Cases describing subarachnoid dissemination of fat, simple fat embolism due to direct intravenous injection (that is, not FES), unclear etiology, and/or unknown connection to the neurological findings were also excluded. Cases meeting the aforementioned criteria are featured in Table 2.2,3,6,8,10–12,15,17,18,20,21,23,24,26–32,34–38,40,43–45,47–49,52–54

TABLE 2:

Literature summary of cases of FES, 2007 to 2012*

Authors & YearAge & SexInjury/Precipitating EventSymptomImaging ResultsICP ManagementOutcome
Ammon et al., 200738, Fhip arthroplastymental status changeMRI: multiple embolic infarctssupportivedeath
Anegawa et al., 199120, Mlong bone Fxsseizure, comaCT: normal; MRI: T2 starfieldsteroids, osmolar therapyintact
Beretta et al., 200822, MORIFcoma, anisocoriaCT: swelling; MRI: T2 starfieldosmolar therapy, ICP monitor (40 mm Hg)severe disability
Buskens et al., 200832, MORIFcomaCT: normal; MRI: T2 starfieldsupportiveintact
Butteriss et al., 200618, FORIFcomaMRI: T2 starfieldsupportivesevere disability
Chen et al., 200817, MORIFmental status changeCT: swelling; MRI: T2 starfieldsupportiveintact
18, MORIFcomaMRI: T2 starfieldsupportiveintact
25, MORIFmental status changeCT: normal; MRI: T2 starfieldsupportiveintact
42, MORIFcomaCT: embolic infarcts; MRI: T2 starfieldsupportivesevere disability
Chrysikopoulos et al., 199618, Mlimb lacerationcomaCT: normal; MRI: T2 starfieldsupportiveintact
18, Mlong bone FxscomaCT: normal; MRI: T2 starfieldsupportiveintact
Citerio et al., 199517, Mlong bone FxscomaCT: swelling; MRI: T2 starfieldosmolar therapymod disability
Dive et al., 200216, Mlimb lengtheningmental status changeMRI: T2 starfieldsupportiveintact
Eguia et al., 200757, MORIFmental status changeCT: normal; MRI: T2 starfieldsupportiveintact
Erdem et al., 199338, Mlong bone Fxsaphasia, comaMRI: T2 starfieldsupportiveintact
Findlay & DeMajo, 198434, FORIFmental status change, hemiparesisCT: hemispheric swellingosmolar therapy, steroids, hyperventilationintact
Gombar et al., 200518, Mlong bone Fxscoma, anisocoriaCT: swellingsupportiveintact
Gregorakos et al., 200018, Mlong bone Fx + external fixationcomaCT: swelling; MRI: T2 starfieldsteroids, hyperventilationintact
20, Mlong bone Fx + closed reductioncomaCT: swelling; MRI: T2 starfieldsupportiveintact
Guillevin et al., 200533, Mlong bone Fx + closed reductioncomaCT: subcortical hypodensities; MRI: DWI starfieldsupportivemild disability
Jacobson et al., 198624, Mlong bone Fxslt hemiparesis, comaCT: swellingsteroidsintact
21, Mlong bone Fxsrt hemiplegia, comaCT: normalsteroidsintact
20, Mlong bone Fxslt hemiparesis, comaCT: normalsteroidsintact
Kamano et al., 200124, Mlong bone Fxmental status changeCT: normal; MRI: T2 starfieldsteroids, anticoagulationmild disability
Kamenar & Burger, 198045, Mlong bone, rib Fxscoma, myoclonusCT: swellingsupportivedeath
Kawano et al., 199116, Mlong bone Fxsmental status changeCT: normal; MRI: T2 starfieldsupportiveintact
21, Mlong bone Fxsseizures, mental status changeCT: normal; MRI: T2 starfieldsupportiveintact
Kim et al., 200825, MORIFmental status changeMRI: T2, DWI starfieldsupportivemild disability
Kumar et al., 201227, MORIFmental status changeMRI: T2, FLAIR starfieldsupportiveintact
Lee et al., 201265, Fknee arthroplastycomaMRI: DWI starfieldsupportiveintact
71, Fknee arthroplastymental status changeMRI: DWI starfieldsteroidsintact
78, Fknee arthroplastymental status changeMRI: DWI starfieldsteroidsintact
60, Fknee arthroplastymental status changeMRI: DWI starfieldsteroidsintact
Marshall et al., 200423, MORIFcoma, pyramidal signsCT: normal; MRI: DWI starfieldsupportivemod disability
Metting et al., 200921, Mlong bone FxcomaCT: diffuse edema; MRI: FLAIR & DWI starfieldsupportivemild disability
Meyer et al., 200723, MORIFmental status change, hemiparesisMRI: fat emboli (delayed)supportiveintact
18, MORIFcomaCT: embolic infarcts; MRI: T2 starfieldsteroidsmod disability
31, FORIFmental status changeMRI: T2 starfieldsupportiveintact
Nastanski et al., 200517, Mlong bone FxscomaCT: normal; MRI: DWI, T2 starfieldsupportivemild disability
Parizel et al., 200118, Flong bone Fxmental status changeMRI: T2 starfieldsupportiveintact
Pfeffer & Heran, 201020, FORIFcomaMRI: T2 starfield, restricted DWIsupportivesevere disability
21, MORIFcomaMRI: T2 starfield, restricted DWIsupportivesevere disability
Rughani et al., 201121, Mlong bone FxscomaMRI: DWI starfieldsupportivemod disability
Sakamoto et al., 198318, Mlong bone FxscomaCT: cortical hypodensitysupportiveintact
Sasano et al., 200475, Mhip arthroplastycomaMRI: DWI starfieldsupportivesevere disability
Satoh et al., 199722, MORIFcomaMRI: T2 starfieldsupportiveintact
Scopa et al., 199424, Mlong bone, pelvic Fxscoma, seizuresCT: edema; MRI: T2 starfieldsupportiveintact
Simon et al., 200328, MORIFmental status changeMRI: FLAIR starfieldsupportiveintact
Stienen & Gautschi, 201223, MORIFcomaMRI: T2, DWI starfieldsupportivesevere disability
Walshe et al., 200719, FORIFmental status changeCT: diffuse edema, bilat watershed infarctssupportivedeath
Yeon et al., 200364, Fknee arthroplastyhemiparesisMRI: T2 starfieldsupportiveintact
Yoshida et al., 199618, Mlong bone Fxmental status changeCT: normal; MRI: T2 starfieldsupportiveintact
79, Flong bone Fxcoma, decorticationCT: normal; MRI: T2 starfieldsupportivemild disability
21, Flong bone Fxscoma, decorticationCT: normal; MRI: T2 starfieldsupportivedeath
present case58, Flong bone Fxcoma, hemiplegia, midline shiftCT: hypodensities at gray/white junction; MRI: diffuse rt hemisphere T2 changes/restricted diffusionhemicraniectomysevere disability

* Fx(s) = fracture(s); mod = moderate; ORIF = open reduction and fixation.

Thirty-eight reports describing 54 cases (40 males and 14 females) were found in the literature. Given the syndrome's well-known association with trauma and long bone fractures, it was not surprising to find that patients were predominantly young males (mean age 30.0 years, range 16–78 years). Most of them had been the victims of high-energy trauma; these patients presented with cerebral fat embolism immediately following either a fracture event (24 patients [44.4%]) or an early open reduction and fixation (21 patients [38.9%]). Cerebral fat embolism was also described following elective orthopedic surgery (8 patients [14.8%]) and was due to extensive soft tissue trauma in a single case (1.8%). Neurological symptoms ranged from mild (mental status changes) to severe encephalopathy (coma with or without abnormal posturing), including in a minority of patients with seizures or focal signs; no patients had only the latter symptom. We considered either head CT or brain MRI necessary for the exclusion of direct TBI and the diagnosis of cerebral FES. Computed tomography scans were frequently normal (16 [51.6%] of 31 scans) or demonstrated nonspecific findings, such as generalized swelling (11 [35.5%] of 31 scans). Only rarely did CT localize focal pathology (16.1% multiple infarcts), and only once (3.2%) was edema predominantly confined to one hemisphere, although the CT in this case was not available for review in the article. Magnetic resonance imaging most often revealed the typical T2- or DW-hyperintense lesions in the supratentorial white matter, the so-called starfield pattern (44 [95.6%] of 46 images) referenced in Table 2, whereas localized embolic infarcts were seen in only 2 cases (2 [4.3%] of 46 images).38

Therapy directed at the neurological manifestations of FES has largely been supportive. Airway protection, mechanical ventilation, and seizure control were universally adopted when appropriate. Steroids were used in 11 (20.4%) of 54 patients. We imagined this would reflect an older clinical trend but were surprised to find recent papers describing its use.33 Nevertheless, only 2 publications accounted for 6 of these patients.27,33 Despite the relatively elevated percentage of patients who demonstrated swelling on the imaging studies, osmolar therapy was used in only 4 (7.4%) and hyperventilation in 2 (3.7%) cases. Direct measurement of ICP was reported only once, with values up to 40 mm Hg.6

Outcome was relatively good, but for patients with coma and/or abnormal posturing at presentation, the outcome was considerably worse. Overall mortality was 4 (7.4%) of 54 patients, with a good outcome (intact or mild disability) in 39 (72.2%) of 54. If the patients with coma and/or abnormal posturing are considered separately, mortality is comparable (2 [6.1%] of 33 patients) but good outcome is decreased (19 [57.6%] of 33 patients). Inversely, the patient population presenting with mild mental status change, seizures, or focal deficits had a far better evolution: mortality was similar (2 [9.5%] 21 patients) but all survivors had a good outcome (17 [90.5%] of 21 patients).

Discussion

Fat embolism syndrome is a consequence of fat embolism that classically presents with cutaneous, neurological, and pulmonary manifestations. Asymptomatic fat embolism occurs in almost all patients with multisystem trauma, but FES itself is rare: it is estimated to occur in 0.5%–11% of all long bone fractures.1 It is also widely known to occur in elective orthopedic procedures: Barak et al. demonstrated that 41% of patients undergoing fixation of an intertrochanteric fracture had microemboli detected on transcranial Doppler monitoring.5 On the other hand, the amount of fat globules required to elicit neurological symptoms is massive; in an autopsy study, Kamenar and Burger documented as many as 100 fat globules per square millimeter of brain area.29 The full triad is not always present, but CNS manifestations occur in as many as 80% of patients with FES.27 Acute onset is the hallmark of FES encephalopathy, and such patients typically present with nonlocalizing symptoms, usually mental status changes that may quickly progress to coma. In our literature search, 98% of patients presented in this manner while only 22% of patients had focal signs and/or seizures.

The complete pathophysiology of FES is still not completely understood. The most widely accepted theory was initially proposed by Bergman et al. in 1873: mechanical occlusion of the microvascular bed by fatty emboli is the central event that precipitates all of the syndrome components in each individual organ.37 This finding of mechanical occlusion was originally described in the pulmonary vasculature, but we now know of several mechanisms that explain how the fat globules bypass the pulmonary circulation and reach other organs, triggering the systemic effects of FES. The filtration ability of the lungs is approximately 80%; smaller fat droplets 7–10 μm in diameter will pass through the capillaries.36 Furthermore, a patent foramen ovale is present in 20%–25% of the normal population.37 The mechanical theory was further complemented by an increased understanding of the following biochemical events: intravascular lipolysis releases free fatty acids systemically, thus precipitating a local and systemic inflammatory reaction. Intravascular lipolysis and release of free fatty acids directly affect pneumocytes in the lung and impair gas exchange; the exact correlate of this process at the cellular level in the brain remains unknown.20,40 An adjunctive role for catecholamines released in major trauma has also been suggested, releasing free fatty acids and contributing to the microvascular damage.37

Diagnosis of FES is classically made using Gurd's criteria—clinically, that is. However, the features of FES are nonspecific, especially in patients presenting predominantly with neurological symptoms. In the last major review of FES, Bulger et al. emphasized the “exclusion” nature of the FES diagnosis and the fact that a diagnosis should be made clinically.8 This may have been true 15 years ago; at the time, the supportive treatment provided to patients would have been adequate not only for FES but also for other differential diagnoses. Given that most patients with FES exhibit symptoms in the first 48 hours, it is imperative to exclude other treatable causes of neurological decline, such as increasing cerebral contusions or diffuse brain injury with swelling.37,53 In the largest study of cerebral FES to date, Takahashi et al.51 suggested, for the first time, incorporating neuroimaging characteristics into the diagnostic criteria, which was finally accomplished by Lee et al.33 in 2012 with the creation of Gurd's modified criteria, an approach that we also favor.

The first use of CT in a patient with FES was documented in 1983 by Sakamoto et al.44 In this pioneering report initial CT scans were considered normal, but on the 8th postinjury day multiple hypodense areas suggestive of embolic infarcts were seen. While this pattern of embolic infarcts was found in a few reports included in our literature review (5 of 31 papers that describe CT findings consistent with embolic infarct areas), the most common finding was nonspecific brain swelling (11 of 31 papers) or the examination was considered normal (16 of 31 papers). At this point, when evaluating an unconscious patient, further diagnostic workup with MRI is mandatory, given that the differential diagnosis of diffuse TBI may require specific treatment not applicable to FES. Magnetic resonance imaging findings in FES were first described almost simultaneously by Saito et al., Kawano et al., and Anegawa et al. in the early 1990s.3,30,43 The typical findings of scattered, increased T2 signal and restricted diffusion areas in the centrum semiovale were finally described by Parizel et al. with the term “starfield” pattern.39 This feature has been considered highly specific for cerebral FES: in our search, 44 (95.6%) of 46 MRI studies demonstrated these findings. Data in the present report stand apart from the typical imaging characteristics of FES in that CT demonstrated, for the first time to our knowledge, fat globules in the gray-white matter junction and predominantly in one hemisphere. Identification of the hypodensities seen on CT is corroborated by measuring their radiodensity at −100 HU. This measurement is particularly helpful, as the HU of air and fat are quite different (approximately −1000 and −85 HU, respectively), although their appearance on CT is comparable.7 Besides the presumably large size of the fat globules, the speed with which a CT scan was obtained after the initial symptoms was probably crucial for visualization of the fat globules. A similar pattern was demonstrated only once in the literature but inside the ventricular system after cerebrospinal embolization of fat from a ruptured Tarlov cyst.17 Guillevin et al. may have described this pattern before but unfortunately did not include the figure in their report.15,25 A hemispheric predominance was reported once by Thaunat et al., but this finding differed significantly from features in our case since the predominance was the result of a direct facial injection of fat.52

The fact that some, if not most, of the effects of FES are caused by the secondary inflammation and not the mechanical occlusion of the microvascular bed underlies the treatment principles that guided the management of our patient. A reliable estimate of the duration of the inflammatory process cannot be found in the literature; clinical resolution is generally thought to occur earlier (7–30 days) than complete resolution of the imaging alterations,4 which may still be present as late as 90 days after the precipitating event.9 Whether clinical improvement correlates with the severity and resolution of the starfield changes on MRI is not firmly established in the literature. It seems that early authors relying on T2 changes did not appreciate that relationship;52 since reports of DWI usage in FES became more common in the early 2000s, there is an overall impression that the extent and velocity of clinical recovery correlates with DWI changes.39,41 Assuming that the inflammatory component is potentially reversible and that the neurons in the affected area are therefore viable, we made every effort to salvage the affected cerebral territory in the present case. The morbidity and mortality data in the literature review in Table 2 demonstrate a better outcome than comparable cases of pure TBI or embolic infarcts of the same magnitude from a cardiogenic source. It should be remembered that the FES population corresponds very closely with the TBI population; therefore, other clinical comorbidities should not affect outcome. Even in the subgroup presenting with coma and/or abnormal posturing, 57.6% still achieved a satisfactory recovery. Patients in a TBI cohort with similar neurological findings on admission—for example, those participating in the DECRA (Decompressive Craniectomy in Patients with Severe Traumatic Brain Injury) study—fared comparatively worse; poor outcome rates ranged between 70% and 51% depending on the treatment group.14

It is evident from our literature search that neurosurgical consultation may be necessary for a significant fraction of cases. Neurosurgery was historically involved before the CT and MRI era to create exploratory bur holes to rule out an expanding mass lesion. Warthin was the first to create such a diagnostic bur hole in 1913, and Schneider reported a second case in 1952.47 The creation of such holes to rule out an expanding lesion is now a historical curiosity, and while a decompressive procedure such as that demonstrated in our case will rarely be needed, ICP monitoring may be considered in a comatose patient with multisystem trauma, especially if the initial imaging study reveals a nonspecific finding such as swelling. It was surprising to realize that of the 33 patients who presented with coma and/or abnormal posturing in our review, only one underwent ICP monitoring.6 And although ICP monitoring was seldom utilized in the literature, it is clear that increased ICP was indeed a concern, and other forms of therapy, such as osmolar therapy (4 [12.1%] of 33 patients) or hyperventilation (2 [6.1%] of 33 patients), were attempted in the comatose patients with brain swelling.42 Steroids were also sporadically used, as mentioned above; however, there are no clinical data to support or discourage their use. It is our firm belief that osmolar therapy should not be administered empirically without ICP measurement. The alternative would be to treat ICP in a definitive manner through a decompressive procedure. Asymmetrical embolization in the reported case enabled a decompressive hemicraniectomy, but in more typical cases with bilateral involvement, a decompressive procedure would have to be bilateral (bifrontal). We have no particular explanation for why our case had preferentially unilateral embolization. Magnetic resonance angiography performed after admission to our institution revealed no stenosis or other abnormalities of the extracranial carotid arteries or circle of Willis. Additionally, we obtained a contrast-enhanced CT of the chest and found no apparent abnormalities of the aortic arch or origins of the common carotid arteries. We suspect the unilateral embolization may be related to the causes of the prior contralateral ischemic stroke, but we do not have any direct evidence to support this statement.35

We are convinced by this case and the literature findings that neurosurgeons may be an integral part of care for the most critical portion of FES patients with cerebral involvement. Despite the fact that FES was described more than 150 years ago, diagnosis and adequate treatment of its cerebral implications still represent a significant clinical challenge. Understanding that it is primarily an inflammatory pathology and not simply massive neuronal death following mechanical occlusion is essential so that every member of a neurocritical care team can devote the extra effort that a reversible disease process in a young patient deserves, including neurosurgical intervention when appropriate.

Disclosure

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author contributions to the study and manuscript preparation include the following. Conception and design: Kellogg. Acquisition of data: Kellogg. Analysis and interpretation of data: Kellogg. Drafting the article: Kellogg. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Kellogg. Administrative/technical/material support: Kellogg, Lopes. Study supervision: Lopes.

Acknowledgements

The authors thank the neurocritical care team at Rush University Medical Center led by Richard Temes, M.D., for assistance in the management of the case described in the current report. They were integral to the medical management of this patient and her preparation for successful rehabilitation. The authors believe that it is only through this type of collaboration that the best outcomes for our patients can be achieved.

References

Article Information

Address correspondence to: Robert G. Kellogg, M.D., 1725 W. Harrison St., Ste. 855, Chicago, IL 60612. email: Robert_Kellogg@Rush.edu.

Please include this information when citing this paper: published online August 16, 2013; DOI: 10.3171/2013.7.JNS13363.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    A and B: Axial head CTs obtained on postinjury Day 1 demonstrating hypodensities at the gray-white matter junction in multiple areas. C and D: Axial T2-weighted and FLAIR MR images obtained on postinjury Day 3 showing edema and midline shift.

  • View in gallery

    A: Operative photograph demonstrating ample frontotemporoparietal decompression and edematous, hyperemic brain parenchyma with sulcal effacement. B and C: Postoperative axial CTs demonstrating improvement in midline shift and effective decompression.

References

1

Allardyce DBMeek RNWoodruff BCassim MMEllis D: Increasing our knowledge of the pathogenesis of fat embolism: a prospective study of 43 patients with fractured femoral shafts. J Trauma 14:9559621974

2

Ammon JTKhalily CLester DK: Fatal cerebral emboli in the absence of a cardiac arterial-venous shunt: case report. J Arthroplasty 22:4774792007

3

Anegawa SHayashi TTorigoe ROgasawara THashizume T: Magnetic resonance imaging of fat embolism syndrome— case report. Neurol Med Chir (Tokyo) 31:3593611991

4

Babikian VLCaplan LR: Brain embolism is a dynamic process with variable characteristics. Neurology 54:7978012000

5

Barak MKabha MNorman DSoudry MKats YMilo S: Cerebral microemboli during hip fracture fixation: a prospective study. Anesth Analg 107:2212252008

6

Beretta LCalvi MRFrascoli CAnzalone N: Cerebral fat embolism, brain swelling, and severe intracranial hypertension. J Trauma 65:E46E482008

7

Brooks RA: A quantitative theory of the Hounsfield unit and its application to dual energy scanning. J Comput Assist Tomogr 1:4874931977

8

Bulger EMSmith DGMaier RVJurkovich GJ: Fat embolism syndrome. A 10-year review. Arch Surg 132:4354391997

9

Buskens CJGratama JWCHogervorst Mvan Leeuwen RBRommes JHSpronk PE: Encephalopathy and MRI abnormalities in fat embolism syndrome: a case report. Med Sci Monit 14:CS125CS1292008

10

Butteriss DJAMahad DSoh CWalls TWeir DBirchall D: Reversible cytotoxic cerebral edema in cerebral fat embolism. AJNR Am J Neuroradiol 27:6206232006

11

Chen JJSHa JCMirvis SE: MR imaging of the brain in fat embolism syndrome. Emerg Radiol 15:1871922008

12

Chrysikopoulos HManiatis VPappas JFilalithis PGogali CSfyras D: Case report: post-traumatic cerebral fat embolism: CT and MR findings. Report of two cases and review of the literature. Clin Radiol 51:7287321996

13

Citerio GBianchini EBeretta L: Magnetic resonance imaging of cerebral fat embolism: a case report. Intensive Care Med 21:6796811995

14

Cooper DJRosenfeld JVMurray LArabi YMDavies ARD'Urso P: Decompressive craniectomy in diffuse traumatic brain injury. N Engl J Med 364:149315022011

15

Cox GTzioupis CCalori GMGreen JSeligson DGiannoudis PV: Cerebral fat emboli: a trigger of post-operative delirium. Injury 42:Suppl 4S6S102011

16

Dive AMDubois PEIde CBulpa PABroka SMInstallé E: Paradoxical cerebral fat embolism: an unusual cause of persistent unconsciousness after orthopedic surgery. Anesthesiology 96:102910312002

17

Duja CMBerna CKremer SGéronimus CKopferschmitt JBilbault P: Confusion after spine injury: cerebral fat embolism after traumatic rupture of a Tarlov cyst: case report. BMC Emerg Med 10:182010

18

Eguia PMedina AGarcia-Monco JCMartin VMonton FI: The value of diffusion-weighted MRI in the diagnosis of cerebral fat embolism. J Neuroimaging 17:78802007

19

Erdem ENamer IJSaribas OAras TTan EBekdik C: Cerebral fat embolism studied with MRI and SPECT. Neuroradiology 35:1992011993

20

Fabian TC: Unravelling the fat embolism syndrome. N Engl J Med 329:9619631993

21

Findlay JMDeMajo W: Cerebral fat embolism. Can Med Assoc J 131:7557571984

22

Gombar SDey NDeva C: Pupillary signs in fat embolism syndrome. Acta Anaesthesiol Scand 49:7232005. (Letter)

23

Gossling HRPellegrini VD Jr: Fat embolism syndrome: a review of the pathophysiology and physiological basis of treatment. Clin Orthop Relat Res 16568821982

24

Gregorakos LSakayianni KHroni DHarizopoulou VMarkou NGeorgiadou F: Prolonged coma due to cerebral fat embolism: report of two cases. J Accid Emerg Med 17:1441462000

25

Guillevin RVallée JNDemeret SSonneville RBolgert FMont'alverne F: Cerebral fat embolism: usefulness of magnetic resonance spectroscopy. Ann Neurol 57:4344392005

26

Gurd ARWilson RI: The fat embolism syndrome. J Bone Joint Surg Br 56B:4084161974

27

Jacobson DMTerrence CFReinmuth OM: The neurologic manifestations of fat embolism. Neurology 36:8478511986

28

Kamano MHonda YKitaguchi MKazuki K: Cerebral fat embolism after a nondisplaced tibial fracture: case report. Clin Orthop Relat Res 3892062092001

29

Kamenar EBurger PC: Cerebral fat embolism: a neuropathological study of a microembolic state. Stroke 11:4774841980

30

Kawano YOchi MHayashi KMorikawa MKimura S: Magnetic resonance imaging of cerebral fat embolism. Neuroradiology 33:72741991

31

Kim KCHwang DSShin HD: Post-traumatic cerebral fat embolism prior to operative repair of femoral and tibial fractures. Orthopedics 31:1702008

32

Kumar SGupta VAggarwal SSingh PKhandelwal N: Fat embolism syndrome mimicker of diffuse axonal injury on magnetic resonance imaging. Neurol India 60:1001022012

33

Lee SCYoon JYNam CHKim TKJung KALee DW: Cerebral fat embolism syndrome after simultaneous bilateral total knee arthroplasty: a case series. J Arthroplasty 27:4094142012

34

Levy D: The fat embolism syndrome. A review. Clin Orthop Relat Res 2612812861990

35

Marshall GBHeale VRHerx LAbdeen AMrkonjic LPowell J: Magnetic resonance diffusion weighted imaging in cerebral fat embolism. Can J Neurol Sci 31:4174212004

36

Metting ZRödiger LARegtien JGvan der Naalt J: Delayed coma in head injury: consider cerebral fat embolism. Clin Neurol Neurosurg 111:5976002009

37

Meyer NPennington WTDewitt DSchmeling GJ: Isolated cerebral fat emboli syndrome in multiply injured patients: a review of three cases and the literature. J Trauma 63:139514022007

38

Nastanski FGordon WILekawa ME: Posttraumatic paradoxical fat embolism to the brain: a case report. J Trauma 58:3723742005

39

Parizel PMDemey HEVeeckmans GVerstreken FCras PJorens PG: Early diagnosis of cerebral fat embolism syndrome by diffusion-weighted MRI (starfield pattern). Stroke 32:294229442001

40

Pell ACChristie JKeating JFSutherland GR: The detection of fat embolism by transoesophageal echocardiography during reamed intramedullary nailing. A study of 24 patients with femoral and tibial fractures. J Bone Joint Surg Br 75:9219251993

41

Pfeffer GHeran MKS: Restricted diffusion and poor clinical outcome in cerebral fat embolism syndrome. Can J Neurol Sci 37:1281302010

42

Rughani AIFlorman JESeder DB: Clinical and radiographic improvement following cerebral fat emboli. Neurocrit Care 15:1901932011

43

Saito AMeguro KMatsumura AKomatsu YOohashi N: Magnetic resonance imaging of a fat embolism of the brain: case report. Neurosurgery 26:8828851990

44

Sakamoto TSawada YYukioka TYoshioka TSugimoto TTaneda M: Computed tomography for diagnosis and assessment of cerebral fat embolism. Neuroradiology 24:2832851983

45

Sasano NIshida STetsu STakasu HIshikawa KSasano H: Cerebral fat embolism diagnosed by magnetic resonance imaging at one, eight, and 50 days after hip arthroplasty: a case report. Can J Anaesth 51:8758792004

46

Satoh HKurisu KOhtani MArita KOkabayashi SNakahara T: Cerebral fat embolism studied by magnetic resonance imaging, transcranial Doppler sonography, and single photon emission computed tomography: case report. J Trauma 43:3453481997

47

Schneider RC: Fat embolism; a problem in the differential diagnosis of craniocerebral trauma. J Neurosurg 9:1141952

48

Scopa MMagatti MRossitto P: Neurologic symptoms in fat embolism syndrome: case report. J Trauma 36:9069081994

49

Simon ADUlmer JLStrottmann JM: Contrast-enhanced MR imaging of cerebral fat embolism: case report and review of the literature. AJNR Am J Neuroradiol 24:971012003

50

Stienen MNGautschi OP: Cerebral fat embolism after severe road traffic accident. Am J Emerg Med 30:831.e1831.e22012

51

Takahashi MSuzuki ROsakabe YAsai JIMiyo TNagashima G: Magnetic resonance imaging findings in cerebral fat embolism: correlation with clinical manifestations. J Trauma 46:3243271999

52

Thaunat OThaler FLoirat PDecroix JPBoulin A: Cerebral fat embolism induced by facial fat injection. Plast Reconstr Surg 113:223522362004

53

Walshe CMCooper JDKossmann THayes IIles L: Cerebral fat embolism syndrome causing brain death after longbone fractures and acetazolamide therapy. Crit Care Resusc 9:1841862007

54

Yeon HBRamappa ALandzberg MJThornhill TS: Paradoxic cerebral embolism after cemented knee arthroplasty: a report of 2 cases and prophylactic option for subsequent arthroplasty. J Arthroplasty 18:1131202003

55

Yoshida AOkada YNagata YHanaguri KMorio M: Assessment of cerebral fat embolism by magnetic resonance imaging in the acute stage. J Trauma 40:4374401996

56

Zenker FA: Beiträige zur normalen und pathologischen Anatomie der Lunge DresdenG. Schonfeld1862

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 14 14 14
Full Text Views 138 138 99
PDF Downloads 77 77 55
EPUB Downloads 0 0 0

PubMed

Google Scholar