INTRODUCTION

Fat embolism syndrome (FES) is a rare, potentially life-threatening condition, resulting from the release of fat globules into the microcirculation causing end-organ damage [1]. It is most commonly encountered following skeletal trauma resulting in long-bone or pelvic fractures as well as high-risk orthopedic surgical procedures such as intramedullary nailing. It is less frequently described in other conditions, including pancreatitis, liposuction, sickle cell disease, and osteomyelitis [1,2]. FES is classically characterized by respiratory failure, fever, petechial rash, and neurologic symptoms manifesting 24 to 72 hours after a precipitating event; however, most patients do not exhibit the full spectrum of clinical findings [1]. While most patients have evidence of fat globules in their blood following long-bone fracture, the proportion who develop FES is much lower, with estimates of approximately 0.1% to 0.3% for isolated fractures, although the risk increases nearly 10-fold with multiple fractures [2,3].

One of the most worrisome complications of FES is cerebral fat embolism (CFE), which can be variable in presentation and manifest with acute altered mental status, seizures, and/or focal neurologic deficits [1]. The diagnosis of FES and CFE can be elusive and requires a high index of clinical suspicion, as there are no specific diagnostic tests with sufficiently high predictive value. Magnetic resonance imaging (MRI) is the most sensitive technique to establish a diagnosis of CFE and classically demonstrates a “starfield pattern” with widespread nonconfluent microinfarcts throughout the bilateral hemispheres [4]. While MRI is a useful modality for diagnosis of CFE, it is incapable of predicting the risk of recurrent embolism. In contrast, transcranial Doppler ultrasound (TCD) provides a portable and noninvasive diagnostic approach for assessing CFE in real time. Prior investigations using TCD for CFE detection in long-bone fractures have found a high frequency and intensity of emboli, particularly in patients with a patent foramen ovale (PFO), which are directly correlated with risk of developing neurologic symptoms [5,6]. Given these advantages, screening high-risk trauma patients with point of care TCD may allow for early detection, risk stratification, and timely interventions in patients with CFE.

We report the case of an otherwise healthy 16-year-old female patient who presented following a motor vehicle collision with femur and pelvic fractures leading to CFE/FES with large vessel occlusion (LVO). TCD was utilized to diagnose ongoing emboli and demonstrated resolution after definitive treatment with fracture fixation.

CASE REPORT

The patient was an otherwise healthy 16-year-old female patient who presented to the emergency department following a high-speed motor vehicle collision. The patient received a chest needle decompression in the field for presumed pneumothorax. Upon arrival to the emergency department, she was found to have a Glasgow Coma Scale of 3 and was intubated emergently for airway protection. She subsequently had bilateral chest tubes placed in the setting of shock and was resuscitated with an appropriate hemodynamic response.

A full trauma workup revealed a left femoral shaft comminuted fracture (Fig. 1A), complete zone 2 sacral fracture, lateral compression type 1 pelvic ring injury, multiple rib fractures, mandible fracture, and a C1 fracture with atlanto-occipital dislocation. Head computed tomography (CT) demonstrated subarachnoid hemorrhage at the basal cisterns, intraventricular hemorrhage, and diffuse cerebral edema. A CT angiogram of the head and neck was performed and showed an LVO of the basilar artery. The patient was then taken emergently to interventional neuroradiology for mechanical thrombectomy with successful recanalization of the basilar artery through residual right posterior cerebral artery occlusion (Fig. 1B). The clot appeared to be composed of adipose-like tissue. An external ventricular drain was placed given severe traumatic brain injury with initial low pressure. The patient was subsequently taken emergently to surgery for an occiput to C2 posterior spinal fusion for atlanto-occipital dislocation.

CFE was considered the likely mechanism of the basilar artery occlusion in the setting of long-bone and pelvic fractures, as well as evidence of fat material in the retrieved basilar thrombus. Prolonged TCD monitoring of the bilateral middle cerebral arteries (MCAs) was performed on hospital day 2 to evaluate the risk of ongoing embolization and demonstrated bilateral spontaneous microembolic signals (MES; 14 MES over 30 minutes; range, 7–14 dB) suggesting a proximal source (Fig. 2). Subsequent TCD evaluation on hospital day 3 demonstrated ongoing MES in the bilateral MCAs (10 MES over 30 minutes; range, 7–15 dB). A transthoracic echocardiogram (TTE) was performed with agitated saline to evaluate for PFO; however, it demonstrated no evidence of an intracardiac shunt.

A repeat CT demonstrated new evolving multifocal infarcts involving the right parietal, left frontoparietal and cerebellum consistent with embolic cerebral infarcts. Given the evidence of new multifocal embolic infarcts, the absence of other vascular injuries, and TCD demonstrating continuous MES in the bilateral cerebral hemispheres, CFE was felt to be most likely in the setting of her multiple orthopedic injuries. The patient was then taken for expedited closed reduction of the femur fracture, as well as closed reduction and percutaneous fixation of the pelvic ring injury on hospital day 5 to reduce the risk of further fat embolization. Repeat TCD obtained on hospital days 6 and 7 following fracture stabilization demonstrated a complete absence of MES in the bilateral MCAs, suggesting a successful treatment response (Fig. 3). A TCD study with agitated saline was also performed following surgery to evaluate for right-to-left shunt (RLS) and confirmed the presence of a grade II shunt despite a repeat TTE continuing to find no evidence of PFO.

Despite evidence for resolution of CFE, the patient’s examination remained poor and notable for absent eye opening and flexor posturing to noxious stimuli. A brain MRI was obtained on hospital day 6 demonstrating multifocal ischemic strokes involving the brainstem in addition to bilateral cerebellar, occipital, frontoparietal and thalamic structures (Fig. 1C, D). In the setting of substantial neurologic injury and after extensive family meetings with palliative care, the family decided to pursue comfort care measures and she died following compassionate extubation.

Ethics statement

Informed consent for publication of the research details and clinical images was obtained from the legally authorized representative of the patient.

DISCUSSION

Patients with CFE may present with a constellation of neurologic symptoms either in isolation or in combination with other end-organ damage related to FES. The precise pathophysiologic mechanisms underlying FES remain elusive, although two models predominate. The mechanical theory hypothesizes that the traumatic release of fat particles into the venous system enters the systemic circulation via intracardiac or pulmonary shunts, causing vasculature occlusion and tissue ischemia [7]. The biochemical theory, in contrast, proposes that increased lipase levels following trauma lead to catalyzation of fat globules into free fatty acids in the bloodstream, which pass through the pulmonary vasculature and cause endothelial damage with resulting microthrombi formation and a widespread inflammatory response [1,7]. The latter model likely better accounts for the typical delay in systemic manifestations, including acute respiratory distress syndrome and diffuse petechial rash, following an inciting event and provides a rationale for the development of neurologic symptoms even in the absence of an RLS.

While the full spectrum of clinical manifestations related to FES can likely be explained by an overlap of these two mechanisms, the occurrence of an acute presentation with LVO is almost certainly related to shunted macroembolic material. Compared to classic CFE neuroimaging findings of diffuse punctate infarcts throughout the subcortical white matter suggestive of microvascular cytotoxic injury, fat emboli–related LVO, such as in our patient’s case, has been rarely reported in the literature and results in blockage of a major cerebral artery with symptoms of acute stroke. In a recent meta-analysis, 18 cases of LVO related to CFE were identified, and in contrast to typical features of FES, 77% experienced neurologic symptoms within 6 hours of the injury and rarely exhibited other systemic manifestations of FES [8]. Furthermore, the majority of patients screened had evidence of PFO, further supporting the role of fat macrothrombi shunting into the systemic circulation, resulting in mechanical occlusion of the cerebral vasculature. Although the overall mortality of FES is approximately 10%, with a large proportion achieving functional recovery within a year even after presenting with coma or posturing, LVO cases appear to be associated with more severe neurologic injury, often necessitating aggressive intervention including decompressive hemicraniectomy for management of malignant cerebral edema, with a higher likelihood of mortality (Table 1) [8–17]. Though limited to case reports, urgent recognition and revascularization via endovascular thrombectomy is the favored approach for LVO due to fat emboli, with some reports of good functional outcomes [8,14,17].

For patients presenting with CFE outside of an LVO, the management is largely supportive and focused on optimizing hemodynamics and respiratory status. Given their potent anti-inflammatory properties, prophylactic corticosteroids have been widely studied and appear to decrease the incidence of FES, though notably without a mortality benefit, and their use remains controversial due to increased risk of infection, delayed wound healing, and metabolic derangements [18]. The most well-established intervention to decrease the incidence and severity of FES is early immobilization and fixation [19]. Although circulatory levels of fat emboli transiently increase during closed manipulation and surgical fixation, evidence suggests the risk of FES increases by over twofold when the time to operation is delayed longer than 24 hours [1,2,19]. However, certain trauma patients may not be stable enough for expedited orthopedic surgery, for example, those patients with significant polytrauma or severe traumatic brain injury, contributing to delays in the definitive management of skeletal injuries. In these cases, risk-stratifying based on CFE burden would be invaluable to identify patients at the highest risk for neurologic deterioration.

While several clinical criteria exist for FES, none are designed to predict the risk of CFE, and no diagnostic modalities are widely utilized to predict the occurrence of cerebral ischemia [20]. TCD monitoring is highly efficacious for detection of spontaneous microemboli entering the cerebral circulation and has been validated to predict ischemic stroke risk in a variety of conditions, including carotid stenosis, blunt cerebrovascular injury, infective endocarditis, and cardiac thrombus [21]. MES are defined as high-intensity transient signals at least 3 dB greater than the spectral background, short duration (<300 msec) and unidirectional, in keeping with the direction of blood flow in the parent vessel (Fig. 2) [21]. TCD is well suited for the critical care setting as it is portable and noninvasive; furthermore, with recent innovations in robotic headset technology, prolonged continuous monitoring is feasible even without experienced operators for real-time cerebral hemodynamic assessment and evaluation of embolic events. Furthermore, TCD possesses nearly three times higher sensitivity for PFO assessment compared to TTE, as exhibited in our patient, making it an ideal diagnostic modality for noninvasive detection of RLS when evaluating the risk of paradoxical emboli [22].

Prior investigations by Forteza et al. [5,6] demonstrated that spontaneous MES on TCD were common in patients with long-bone fractures and that a combination of RLS, high frequency of MES (>4 per hour), and high signal intensity (>13 dB) was highly predictive of developing neurologic symptoms (positive predictive value 86%, negative predictive value 97%). Furthermore, TCD characterizes the embolism size in FES, which correlates with risk of neurologic deficits, as patients with RLS on average have higher signal intensities, suggestive of a larger mass, than those without a shunt [5,23]. In our patient, the persistent elevation in MES frequency (10–14 over 30 minutes) with high intensity (7–15 dB) in combination with radiographic evidence of new cerebral infarcts justified surgical fixation to mitigate the risk of further cerebral injury. Ideally, the presence of persistent MES in this clinical setting could be used to proceed with surgical fixation before significant cerebral injury occurs. Another benefit of daily TCD assessment is the evaluation of response to intervention. Our case is the first to characterize the resolution of microemboli on TCD following surgical fixation of fractures in FES, thus confirming an efficacious treatment response. Other investigations have reported similar utility of TCD emboli detection to gauge response to initiation of anticoagulation in cerebrovascular dissection as well as confirming successful PFO closure [23,24].

In summary, we report a case of FES following skeletal trauma with CFE initially presenting with basilar LVO requiring thrombectomy. Although the patient ultimately died due to substantial neurologic injury related to their initial brainstem ischemic stroke, the case highlights the utility of TCD microemboli detection in critically ill patients at risk of FES to gauge risk of cerebral embolic events as well as to confirm the response to definitive therapy.

ARTICLE INFORMATION

Author contributions

Conceptualization: RM, IN, CC, JRV; Investigation: MJ, HN, KR, AA, JRV; Methodology: MJ, JRV; Visualization: HN, KR, AA; Writing–original draft: MJ, JRV; Writing–review & editing: RM, IN, CC, JRV. All authors read and approved the final manuscript.

Conflicts of interest

The authors have no conflicts of interest to declare.

Funding

The authors received no financial support for this study.

Data availability

Data sharing is not applicable as no new data were created or analyzed in this study.

Fig. 1.

Multimodal imaging findings. (A) Acute comminuted left femoral shaft fracture. (B) Large vessel occlusion (arrow) in the mid basilar artery on digital subtraction angiography. (C) Multiple areas of diffusion restriction in the supratentorial brain on magnetic resonance imaging, consistent with cytotoxic edema. (D) Diffusion-weighted imaging signal abnormalities in the paramedian pons and bilateral cerebellar hemispheres on magnetic resonance imaging.

Fig. 2.

Transcranial Doppler ultrasound (Viasonix Dolphin/XF, Imaging Monitoring USA) results. (A) Transcranial Doppler spectral waveform of the left middle cerebral artery (MCA) obtained via transtemporal approach with a high-intensity transient signal (HITS) noted (arrow). (B) Power M-mode with a red band extending from 35 to 65 mm denoting anterograde blood flow in the MCA and a blue band extending from 65 to 75 mm denoting retrograde flow in the anterior cerebral artery (ACA). Note the HITS in green box, which is brief (<0.3 seconds) and unidirectional (slightly negatively slanted orientation, reflecting movement towards probe over time as blood flows laterally through MCA). (C) High-resolution M-mode allows further characterization of HITS (green box) at 24 msec resolution to confirm that the signal is unidirectional along its entire length, increasing the likelihood that it reflects true microemboli and is not an artifact (D) Complex velocity analysis demonstrates matching Doppler shift frequencies for HITS (green waveform) and blood flow (black waveform) suggesting that both are in-phase, increasing the likelihood of a true embolic signature rather than an artifact. Energy measurement showing that the event is of sufficiently high intensity (15 dB) to be consistent with microembolic signal.

Fig. 3.

Chronological graph of transcranial Doppler ultrasound findings in the right and left middle cerebral artery (MCA) by hospital day. Recordings performed on hospital days 2, 3, 6, and 7. Surgical fixation occurred on hospital day 5. MES, microembolic signals.

REFERENCES

• 1. Luff D, Hewson DW. Fat embolism syndrome. BJA Educ 2021;21:322–8.ArticlePubMedPMC
• 2. Kainoh T, Iriyama H, Komori A, Saitoh D, Naito T, Abe T. Risk factors of fat embolism syndrome after trauma: a nested case-control study with the use of a nationwide trauma registry in Japan. Chest 2021;159:1064–71.ArticlePubMed
• 3. Stein PD, Yaekoub AY, Matta F, Kleerekoper M. Fat embolism syndrome. Am J Med Sci 2008;336:472–7.ArticlePubMed
• 4. Parizel PM, Demey HE, Veeckmans G, et al. Early diagnosis of cerebral fat embolism syndrome by diffusion-weighted MRI (starfield pattern). Stroke 2001;32:2942–4.ArticlePubMed
• 5. Forteza AM, Koch S, Campo-Bustillo I, et al. Transcranial Doppler detection of cerebral fat emboli and relation to paradoxical embolism: a pilot study. Circulation 2011;123:1947–52.ArticlePubMed
• 6. Forteza AM, Koch S, Romano JG, et al. Transcranial Doppler detection of fat emboli. Stroke 1999;30:2687–91.ArticlePubMed
• 7. Godoy DA, Di Napoli M, Rabinstein AA. Cerebral fat embolism: recognition, complications, and prognosis. Neurocrit Care 2018;29:358–65.ArticlePubMedPDF
• 8. Ooi S, Raviskanthan S, Campbell BC, Hutton EJ, Mitchell PJ, Cloud GC. Cerebral large vessel occlusion caused by fat embolism: a case series and review of the literature. Front Neurol 2021;12:746099.ArticlePubMedPMC
• 9. van Oostenbrugge RJ, Freling G, Lodder J, Lalisang R, Twijnstra A. Fatal stroke due to paradoxical fat embolism. Cerebrovasc Dis 1996;6:313–4.ArticlePDF
• 10. Wang YC, Fu JH, Lai PH. Teaching NeuroImages: hypodense artery sign in acute cerebral infarction by contrast-enhanced CT. Neurology 2009;73:e16.ArticlePubMed
• 11. Kellogg RG, Fontes RB, Lopes DK. Massive cerebral involvement in fat embolism syndrome and intracranial pressure management. J Neurosurg 2013;119:1263–70.ArticlePubMed
• 12. Piuzzi NS, Zanotti G, Comba FM, Buttaro MA, Piccaluga F. Paradoxical cerebral fat embolism in revision hip surgery. Case Rep Orthop 2014;2014:140757.ArticlePubMedPMCPDF
• 13. Avila JD. Hypodense artery sign in cerebral fat embolism. Pract Neurol 2017;17:304–5.ArticlePubMed
• 14. Fowler JB, Fiani B, Sarhadi K, Cortez V. Cerebral fat embolism in the absence of a long bone fracture: a rare case report. Surg Neurol Int 2021;12:78.ArticlePubMedPMC
• 15. Koh D, Nair G, Teoh E, Sannigrahi A. Hypodense MCA sign: large-vessel occlusion secondary to fat embolism post-right hip hemiarthroplasty. J Med Imaging Radiat Oncol 2021;65:222–3.ArticlePubMedPDF
• 16. Kamal RH, Ramadhan FR, Wibowo MD, Sasono B. A rare fatal cerebral fat emboli syndrome with large vessel occlusion post femur fracture plating in an older female: a case report. Int J Surg Case Rep 2024;120:109828.ArticlePubMedPMC
• 17. Maier B, Badat N, Boulouis G, Zuber M. Mechanical thrombectomy for a cerebral fat embolism. Intensive Care Med 2019;45:1151.ArticlePubMedPDF
• 18. Bederman SS, Bhandari M, McKee MD, Schemitsch EH. Do corticosteroids reduce the risk of fat embolism syndrome in patients with long-bone fractures? A meta-analysis. Can J Surg 2009;52:386–93.ArticlePubMedPMC
• 19. Talbot M, Schemitsch EH. Fat embolism syndrome: history, definition, epidemiology. Injury 2006;37 Suppl 4:S3–7.ArticlePubMed
• 20. Morales-Vidal SG. Neurologic complications of fat embolism syndrome. Curr Neurol Neurosci Rep 2019;19:14.ArticlePubMedPDF
• 21. Kargiotis O, Psychogios K, Safouris A, et al. The role of transcranial Doppler monitoring in patients with multi-territory acute embolic strokes: a review. J Neuroimaging 2019;29:309–22.ArticlePubMedPDF
• 22. Rubin MN, Shah R, Devlin T, et al. Robot-assisted transcranial Doppler versus transthoracic echocardiography for right to left shunt detection. Stroke 2023;54:2842–50.ArticlePubMedPMC
• 23. Forteza AM, Rabinstein A, Koch S, et al. Endovascular closure of a patent foramen ovale in the fat embolism syndrome: changes in the embolic patterns as detected by transcranial Doppler. Arch Neurol 2002;59:455–9.ArticlePubMed
• 24. Koch S, Romano JG, Bustillo IC, Concha M, Forteza AM. Anticoagulation and microembolus detection in a case of internal carotid artery dissection. J Neuroimaging 2001;11:63–6.ArticlePubMed

Figure & Data

References

Citations

Citations to this article as recorded by