Lipoplasty, or liposuction, the surgical process of removing excess fat, is an elective procedure with rising frequency in the United States. Fat embolism syndrome is a clinical diagnosis and is defined as fat in the circulation with an identifiable clinical pattern of signs and symptoms (eg, hypoxemia, respiratory insufficiency, neurologic impairment, and petechial rash) that occur in the appropriate clinical context. Fat embolism syndrome following liposuction is a life-threatening complication, although its incidence is low. Currently, there is no specific therapy for fat embolism syndrome, so prevention, early detection, and supportive therapy are critical. Many cases of fat embolism syndrome are undiagnosed or misdiagnosed; however, postmortem examination can provide the means for appropriate diagnosis. Therefore, a pathologist must keep a keen eye, as microscopic fat emboli are difficult to appreciate with routine tissue processing and staining.
Fat emboli are generated when fat globules enter the bloodstream from tissue (usually marrow or adipose) disrupted by trauma or, alternatively, via production of toxic intermediaries of plasma-derived fat (eg, chylomicrons). A significant burden of fat emboli may cause specific clinical manifestations, creating the phenomenon termed fat embolism syndrome (FES). Despite its having been initially described in 1861, there remains no diagnostic test that is sufficiently sensitive or specific for confirming or excluding FES.1
Fat embolism syndrome is most commonly associated with long bone and pelvic fractures (bone marrow has high fat content), but it can also arise from soft tissue trauma without fracture and from a variety of other nontraumatic and nonorthopedic traumatic causes, including, but not limited to, pancreatitis, sickle or thalassemia-related hemoglobinopathies, alcoholic (fatty) liver disease, renal angiomyolipoma invasion of the inferior vena cava, bone tumor lysis, steroid therapy, and notably, liposuction.2 The first case of liposuction-induced FES was reported by Christman3 in 1986. The number of liposuction procedures per year continues to steadily rise in the United States. It was the most common cosmetic surgical procedure in the United States in 2015 (305 856 procedures), and the number of procedures has nearly doubled since 1997 (176 863 procedures), when plastic surgery statistics started being recorded. Moreover, 53 554 more procedures occurred in 2015 compared with 2014.4 With rising incidence and a lack of diagnostic testing, antemortem and postmortem suspicion should be raised for FES in the setting of recent liposuction. To date, only 17 cases of liposuction-induced FES with or without fat grafting have been reported in the literature.3,5–20
The pathogenesis of FES is unknown. During liposuction and fat grafting, there is rupture of small blood vessels and damage to adipocytes, producing lipid microfragments that reach the venous circulation and consequently cause lung injury.16,21 Fat emboli can also reach the systemic circulation, affecting other organs, because of the patency of the foramen ovale in the interatrial septum, the existence of pulmonary arteriovenous microfistulas, and the deformation of the fat microglobules that cross the pulmonary capillaries.22,23 Liposuction-induced FES typically occurs 12 to 72 hours after surgery. In theory, 3 major models have been proposed to explain the pathogenesis of FES. These theories exist to help explain the timing and etiology of the embolic events.
The first theory is the mechanical theory. This theory postulates that fat droplets are released by disruption of fat cells in fractured bone or in adipose tissue. These fat droplets enter the torn veins near the site of injury, and are then transported to the pulmonary vascular bed, where large fat globules result in mechanical obstruction and are trapped as emboli in lung capillaries. The smaller droplets (7–10 μm) may pass through the lung and reach systemic circulation, causing embolization to the brain, kidney, skin, or retina. Large fat droplets may also enter systemic circulation through preexisting pulmonary precapillary shunts and pathologic venous-arterial communication such as patent foramen ovale.24 This theory is corroborated by studies involving intraoperative transesophageal echocardiography in various types of orthopedic surgeries in which hypoechogenic intracardiac material is seen passing through the right heart during the introduction of intramedullary prosthesis.23 However, this theory fails to explain cases with delayed symptom onset (>72 hours) after liposuction.
The second theory is the biochemical theory, which helps explain nontraumatic and delayed fat embolic events. The biochemical theory proposes that the clinical manifestations of FES are attributable to a proinflammatory setting. After reaching the pulmonary capillaries, fat globules are hydrolyzed by lipase produced by pneumocytes, producing high concentrations of glycerol and free fatty acids, which are toxic to alveoli and endothelial cells. At the onset of the local injury, vasoactive amines and prostaglandins are released. In addition, there is neutrophil recruitment, leading to hemorrhage, as well as interstitial and alveolar edema. Histopathologically, there is edema, transudate, and subsequent alveolar exudate, followed by type II pneumocyte apoptosis and formation of hyaline membranes. This hypothesis addresses delayed symptom onset following liposuction, because agglutination and degradation of the fat emboli, which are time-consuming biochemical processes, would be necessary to trigger the local inflammatory process.12,16,22,23 Further support of the biochemical theory is endorsed in nontraumatic settings, such as pancreatitis. Serum from acutely ill patients demonstrates the capacity to agglutinate chylomicrons, low-density lipoproteins, and liposomes of nutritional fat emulsions. C-reactive protein is elevated in these patients, and studies have demonstrated its ability to induce calcium-dependent lipid agglutination.25 Therefore, C-reactive protein likely participates in this proposed delayed pathogenesis of nontraumatic FES.24
The third theory, which is the most recent and appears least supported, is the coagulation theory. This theory articulates how tissue thromboplastin and marrow elements are released following long bone fractures, thus activating the complement system and extrinsic coagulation cascade. As a result, these events lead to intravascular coagulation via fibrin and fibrin degradation products. These products combine with leukocytes, fat globules, and platelets to increase pulmonary vascular permeability in 2 ways: direct actions on endothelial cells, and via release of vasoactive substances.24 The coagulation theory fails to explain the etiology of nontraumatic FES.
These 3 theories are not mutually exclusive, and have all been described after major traumas involving long bone fracture, as well as after intramedullary orthopedic procedures.22 It is likely they all play a contributory role in the pathogenesis of FES as it relates to the etiology (traumatic versus nontraumatic) and time course.
The clinical manifestations of and diagnostic criteria for FES were first described by Gurd in 1970,33 and later refined by Gurd and Wilson26 in 1974. The refined criteria are currently the gold standard for FES diagnosis, although alternative diagnostic criteria have been proposed.27 The refined criteria state that at least 2 major signs/symptoms, or 1 major and 4 minor signs/symptoms, must be present to diagnosis the syndrome (Table). Fat embolism syndrome is most common after long bone or pelvic fractures, and less common among all other etiologies. Although FES typically manifests 24 to 72 hours after initial insult, it may rarely occur as early as 12 hours or as late as 2 weeks after the inciting event.28 Many affected patients fail to develop the classic triad. A study by Bulger et al29 identified 27 patients with FES, of whom 26 (96%) exhibited respiratory manifestations, but only 16 (59%) and 9 (33%), respectively, demonstrated neurologic abnormalities and a petechial skin rash.
Respiratory changes are usually the first clinical manifestation, and most frequently includes dyspnea, tachypnea, and hypoxemia. Respiratory abnormalities tend to vary in severity, but the possibility of acute respiratory failure encourages early intervention. Neurologic symptoms typically present in the early stages after the development of respiratory distress. These changes vary widely, from mild confusion and drowsiness to obtundation or seizures. The petechial rash is least common, and usually the last sign to manifest. It most commonly arises in the head, neck, axillae, and anterior thorax.
MACROSCOPIC AND MICROSCOPIC FEATURES
Despite the well-known clinical characterization, debate exists within the literature regarding the most appropriate clinical diagnostic criteria for FES. Therefore, some forensic pathologists are cautious about making a postmortem diagnosis of FES, especially in cases where clinical information is limited.30 Nevertheless, FES does have unique autopsy features that aid in diagnosis, with a predilection for lung, brain, kidney, skin, and eye. Gross pulmonary findings may include increased lung weight and volume, as well as firm but not solid texture, with a marbled appearance of the visceral pleura due to alternating areas of hemorrhage. Tardieu spots (petechiae or ecchymoses) may be present beneath the visceral pleura.31 Although microscopic evaluation for fat emboli with routine hematoxylin-eosin is often unrevealing (Figure 1), if the pathologist is sensitive to the differential of FES and is assessing for it, oftentimes he or she may see intravascular round/ovoid negative staining suggestive of possible fat emboli. With enough suspicion, special staining may prove useful. Thin slices of lung tissue fixed in formalin can be postfixed and stained in osmium tetroxide solution and subsequently submitted for paraffin embedding. Fat emboli will appear as round, uniform, black-staining droplets within blood vessels (Figures 2 and 3). Alternatively, oil red O staining can be used, but only on frozen section tissue, as lipids are dissolved by xylene and alcohol solvents during standard tissue processing.
Central nervous system gross pathology in FES includes potentially innumerable, macroscopic, brown-red petechial hemorrhages that are produced by fat emboli to the brain, and are particularly restricted to the white matter and spinal cord, with possible limited extension into gray matter (Figure 4). The petechiae may vary from 4 mm to less than 1 mm. Hemorrhage may also be present, but is less common, and is found in the cerebral gray matter, brain stem, and cerebellum. Microscopic brain findings with routine hematoxylin-eosin will show the corresponding petechial cerebral white matter and/or spinal cord hemorrhages (Figures 5 and 6) and cerebrospinal vascular fat emboli (Figure 7).
Renal findings are generally unremarkable grossly. Furthermore, kidney function tests and renal parenchyma are not typically affected by fat emboli. Rather, renal glomeruli harbor the disease burden because of the large blood supply of the kidneys and the restriction of the capillary bed to the glomeruli. Hematoxylin-eosin staining shows fat droplet deposition in arterioles/capillaries and renal glomeruli. These observations can be further emphasized utilizing the aforementioned fat-specific staining methods.
Cutaneous manifestations are produced by fat embolization to small dermal capillaries that cause erythrocyte extravasation. Macroscopically, this correlates to a petechial rash of the anterior thorax, axillae, neck, oral mucosa, and conjunctiva.
LABORATORY AND IMAGING FINDINGS
Acute decrease in hematocrit, increase in serum lipase levels, hypofibrinogenemia, hypoxemia, thrombocytopenia (50% of patients), anemia (70% of patients), and hypocalcemia (due to free fatty acid binding to calcium) typically occur, if present, at the same time as clinical manifestations (24–72 hours after initial insult).24,32 Although these laboratory findings frequently occur in the settings of FES, they are not specific.
Chest radiographs typically appear normal early in the disease process, but gradual (1–3 days) progression to bilateral flocculent shadows, and possibly bilateral interstitial opacification, occurs in some cases. The radiologic signs may remain for up to 3 weeks.32 Computed tomography imaging of the chest generally shows focal areas of ground-glass opacification with interlobular septal thickening. However, ill-defined centrilobular and subpleural nodules representing alveolar edema, microhemorrhage, and inflammatory response secondary to ischemia and cytotoxic emboli may be observed.24
Computed tomography imaging of the brain, an initial modality used because of neurologic impairment, may show no abnormalities, or it may show diffuse white matter petechial hemorrhages, consistent with microvascular damage. Rarely, it can show generalized cerebral edema or atrophy in severe cases of FES. Magnetic resonance imaging may be useful in patients who present with neurological features of fat embolism but with a normal computed tomography brain image. T2-weighted images show scattered high–signal-intensity lesions early in the disease course, typically deep in the white matter, cerebellum, and brain stem.24
The differential diagnosis in a patient with signs and symptoms resembling FES is exhaustive because of the nonspecific and variable presentation. Primary considerations should be based upon the specific clinical presentation, which may or may not present with the typical triad. Respiratory insufficiency in isolation should prompt suspicion for postsurgical pulmonary thromboembolism, pneumonia, drug reaction, cardiogenic pulmonary edema, and FES. Imaging modalities are typically helpful to differentiate pulmonary thromboemboli from fat emboli. Computed tomography imaging, the gold-standard imaging modality of pulmonary thromboembolism, may be contraindicated in patients who cannot tolerate intravenous contrast. Ventilation perfusion scanning can detect areas of perfusion failures, but it does not reliably differentiate thromboembolism from FES.
Meningococcal septicemia may present with neurologic impairment and skin changes, but respiratory insufficiency would be unusual and would assist in excluding this disease. Patients who present primarily with neurologic manifestations (and limited respiratory insufficiency) should prompt evaluation for cranial hemorrhage. Noncontrast head computed tomography is a valuable tool to help rule such diagnoses in or out.
CURRENT TREATMENT AND PROGNOSIS
Treatment of FES is merely supportive, as there is no directed therapy. Thus, prevention, early detection, and prompt supportive therapy are critical. Early diagnosis not only limits morbidity and mortality, but also diminishes additional investigation cost burdens. Continuous positive airway pressure is usually the first-line treatment for respiratory insufficiency. This treatment generally fails quickly, and swift transition to intubation with mechanical ventilation and positive end expiratory pressure should be provided. Pharmacologic agents can be used to maintain hemodynamic stability. Heparin and ethyl alcohol were historically used for treatment, but they are ineffective and are rarely used today. Contrary to recent efforts, there are insufficient data to support the routine use of corticosteroids, aspirin, and prophylactic antibiotics.24
Fat embolism syndrome is a self-limiting disease. The overall mortality from FES after liposuction is approximately 10% to 15%, and mortality correlates with severity of respiratory insufficiency.32 Although the duration of FES is difficult to predict, survival beyond initial presentation generally leads to full recovery.
The authors have no relevant financial interest in the products or companies described in this article.