Cerebral fat embolism without intracardiac shunt: A novel presentation

Abstract

Fat embolism syndrome (FES) is defined as an uncommon life-threatening disease process consisting of pulmonary, central nervous system (CNS), and cutaneous manifestations. The pathophysiology of this secondary injury is poorly understood. In the setting of the multiply injured patient, the diagnosis of FES is difficult to ascertain. A case report of a posttraumatic death caused by acute dissemination of diffuse fat emboli to the brain and lungs in the absence of a right-to-left heart defect after femur fracture is presented. The transesophageal echo cardiogram with bubble study failed to demonstrate an intracardiac defect or AV malformation in the lung further supporting a biochemical process. The acute decompensation of the patient within 2 h of the injury would favor mechanical emboli. Supportive care continues to be the mainstay of treatment for FES. Cerebral fat embolism should be considered in traumatically injured patients with unexplained decline in their neurologic examination. Cerebral fat embolism may occur without an intracardiac shunt.

INTRODUCTION

Fat embolism syndrome (FES) is defined as an uncommon life-threatening disease process consisting of pulmonary, central nervous system (CNS), and cutaneous manifestations.[1] This process has been recognized following orthopedic procedures and injuries. It is difficult to identify this disorder upon acute presentation of trauma patients secondary to the multiplicity of injuries and manifestation of symptoms.[2,3] A case report of a post-traumatic death caused by acute dissemination of fat emboli to the brain and lungs in the absence of a right-to-left heart defect after femur diaphyseal fracture is presented.

CASE REPORT

A healthy 54-year-old male was involved in a motorcycle accident at an approximated speed of 100 mph. Upon arrival to the emergency department, the patient was determined to be hemodynamically stable with a temperature of 36.7° C, pulse of 95, blood pressure of 111/64, respiratory rate of 17 and SaO₂ of 95%. The patient was alert, awake, talking with a Glascow Coma Score (GCS) of 15. CT scans documented a right-sided pneumothorax with associated first through eighth rib fractures, right pulmonary contusion, right closed diaphyseal midclavicular fracture, and left closed middiaphyseal femoral fracture. Within 2 h of the injury and 53 min of presentation, the patient had an abrupt neurologic deterioration to a GCS of six. Recorded vital signs were within normal limits except for tachypnea and oxygen desaturation which had dropped to 84% on a non-rebreather mask. Secondary to the alteration in GCS and decreased oxygen saturation, he was intubated and a repeat CT scan of the head was found to be normal. The patient was taken to the surgical intensive care unit for supportive care. Shortly after being transferred to the ICU, the patient developed tachycardia (heart rate=130). On hospital day one, he failed to show improvement in his neurological exam. A third head CT scan was obtained demonstrating subtle diminution of the ventricular system and visualization of the subarachnoid pathways compared to prior studies. An MRI of the head demonstrated dramatic bilateral periventricular white matter, subcortical white matter, bilateral basal ganglia, midbrain, and cerebellar foci in confluent area T2, hyperintensities suggestive of fat emboli [Figure 1]. At that time, neurosurgery inserted a ventriculostomy revealing intracranial pressures (ICP) of 18–24. Mannitol was administered to the patient who was sedated into a pentobarbital coma. The patient manifested several minor criteria of FES including: anemia (hemoglobin 7.0 g/dL), thrombocytopenia (platelet count = 25), acute renal insufficiency, ongoing tachycardia, and cooling techniques were required to maintain the patient normothermic. The ICPs continued to increase despite aggressive preventative treatment. The patient progressed to brain death on hospital day 5. During the patient's course, transthoracic echocardiogram (TTE) and transesophageal echocardiogram (TEE) were negative for evidence of patent foramen ovale (PFO), atrial septal defect (ASD), or ventricular septal defect (VSD).

Figure 1.

MRI showing Starfield findings of FES

An autopsy confirmed the fractures documented at admission. The patient's heart, liver, and kidneys were harvested for transplantation prior to autopsy. Sectioning of the brain revealed numerous petechial hemorrhages within the white matter and corpus callosum [Figure 2]. Microscopically, the cortex and white matter of the brain had perivascular hemorrhages corresponding to the petechiae. The osmium stained sections of the brain revealed numerous fat emboli within the arterioles and capillaries [Figure 3]. Osmium stained lung sections showed extensive fat embolization [Figures 4 and 5]. Many alveolar macrophages contained phagocytosed fat globules [Figure 5].

Figure 2.

Sectioning of the brain revealed numerous petechial hemorrhages within the white matter and corpus callosum

Figure 3.

Osmium-stained sections of the brain revealed numerous fat emboli within the arterioles and capillaries

Figure 4.

Osmium-stained lung sections showed extensive fat embolization

Figure 5.

Osmium-stained lung sections showed extensive fat embolization as well as many alveolar macrophages contained phagocytosed fat globules

DISCUSSION

FES is characterized by both major and minor findings following long-bone trauma and/or major orthopedic procedures as defined by Gurd.[1] Major criteria include hypoxia, deteriorating mental status, and petechiae. Minor criteria consist of tachycardia, fever, anemia, and thrombocytopenia. FES is thought to occur in approximately 0.9–2.2% of patients with long bone fractures.[4] The pathophysiology of this secondary injury is poorly understood.

A diagnosis of FES is difficult to determine in a patient with multiple injuries. Classically, an initial asymptomatic interval of 12–72 h is followed by pulmonary, neurologic, and/or dermatologic changes. Respiratory symptoms (e.g., hypoxia, tachypnea, and dyspnea) are noted in 75% of cases and pulmonary failure in 10%.[5] Altered mental status resulting in confusion, headache, stupor, coma, rigidity, or convulsions has been reported in 86% of patients.[6] Cutaneous manifestations, a reddish-brown nonpalpable petechial rash diffusely covering the upper torso, neck, and arms, appear within 12–36 h.[7] Retinal changes can also occur, resulting in exudates, cotton-wool spots, edema, hemorrhage, or intravascular fat globules.[8] Other nonspecific changes include lipuria, jaundice, thrombocytopenia, elevated Westergren Sedimentation Rate (WESR), anemia, fat macroglobulinemia, tachycardia, and pyrexia.[5] Our patient deteriorated quickly after his injury suggesting that fat embolism may begin at the time of injury.

The standard trauma patient evaluation often fails to document the occurrence of FES. Concentrations of neutral lipids, especially cholesterol and cholesterol esters obtained from a bronchoalveolar lavage, may assist with the diagnosis.[9] Chest radiographs, though non-specific, may show an evenly distributed increase in pulmonary markings and dilatation of the right heart.[10] Since the embolic load is distributed throughout both lungs, ventilation–perfusion scans are often nondiagnostic. Head CT scans are often normal.[11] MRI of the brain is the most sensitive test available, and correlates well with the clinical severity of brain injury.[12] T2-weighted images revealing nonconfluent regions of high signal intensity have been associated with FES. The addition of diffusion-weighted scans revealing corresponding bright spots on dark background (“starfield” pattern) are sensitive markers for FES.[13] Transcranial Doppler has been used to document cerebral microembolic signals up to 4 days after long bone fractures.[14]

Supportive care continues to be the mainstay of treatment for FES. Multiple treatment options have been evaluated in the past without significantly changing clinical outcomes including clofibrate, dextran-40, ethyl alcohol, heparin, aspirin, and steroids.[15,16] A recent meta-analysis supports the use of corticosteroids to prevent FES but note that further studies are required to evaluate this topic.[17] Early splinting and fixation of orthopedic fractures improve outcomes in trauma patients.[18] Decreased manipulation of intramedullary fat leads to a lower incidence of fat embolism.[19] Some have described venting the bone while performing intramedullary reaming in order to divert any subsequent fat emboli.[20]

Two reported theories attempt to explain the physiology of fat emboli. The first theory is based on mechanical emboli in the form of intramedullary fat entering venous circulation and traveling to the pulmonary vasculature.[21] The microcirculation of the lungs sequesters the embolic particles leading to an acute increase in oxygen requirement. Previous observations during intraoperative orthopedic manipulation support this embolic theory.[22] In order to cause CNS, renal, retinal, and cutaneous manifestations, these embolic particles must cross an intracardiac shunt (PFO, ASD, or VSD) or pulmonary arterio-venous malformation (PAVM), thus gaining access to the systemic circulation. The incidence of intracardiac shunts has been described to occur in 20–34% of the population.[3] Additionally, microfat droplets can theoretically traverse the pulmonary circulation without sequestration, resulting in systemic symptoms.[23]

The second theory suggests that FES is a biochemical process which is inflammatory in nature.[24] The authors have hypothesized that inflammatory reactants, including lipoprotein lipase, cause the release of fatty acids thus altering the fat transport mechanisms of the plasma. This change in homeostasis results in fat droplet aggregation with systemic sequestration in the microvasculature. These emboli exacerbate the development of organ dysfunction.[3]

Although neither theory has been well substantiated, the systemic manifestations seen in our patient could favor either process. The TEE with bubble study failed to demonstrate an intracardiac defect or AV malformation in the lung, further supporting a biochemical mechanism. However, the acute decompensation of the patient within 2 h of the injury would favor the mechanical emboli hypothesis.

CONCLUSION

Cerebral fat embolism should be considered in traumatically injured patients with unexplained decline in their neurologic examination. Early external fixation of the femur in the ICU might have reduced continued embolic dissemination. However, since our patient acutely had pulmonary decompensation without any warning signs, preventative measures could not be taken. Despite aggressive supportive treatment, the patient succumbed to the FES disease process.