Nonthrombotic Pulmonary Embolism *

Etiology, Prevalence, and Epidemiology

The term fat embolism refers to the presence of globules of free fat within the pulmonary vasculature. Fat embolism is common among trauma patients, especially those with long bone or pelvic fractures. In the context of trauma, fat embolism is also common after extensive injury to subcutaneous fat, such as occurs in severe beatings or liposuction. Fat embolism must be differentiated from fat embolism syndrome, which is defined by the presence of clinical signs and symptoms resulting from fat emboli. The clinical manifestations of fat embolism syndrome consist of the triad of hypoxia (95%), confusion (60%), and petechial rash (33%). Unfortunately, the classic triad is rarely present, and the clinical manifestations of fat embolism syndrome range from an indolent course to fulminant respiratory failure. The incidence of clinically significant disease in patients who have simple tibial or femoral fractures is generally believed to be about 1% to 3%. In individuals who have more severe trauma, the incidence of clinically evident embolism is probably in the range of 10% to 20%. Less common causes of fat embolism include severe beatings (embolism of subcutaneous fat), orthopedic procedures (e.g., intramedullary prosthesis insertion, arthroplasty), liposuction, severe pancreatitis, sickle cell disease, hematopoietic stem cell (bone marrow) transplantation and harvesting, and venous hyperalimentation.

Clinical Presentation

The symptoms of fat embolism syndrome usually appear gradually; dyspnea, neurologic symptoms, fever, and petechial rash typically develop 12 to 36 hours after injury. The delay in the development of clinical symptoms is presumably due to the time required for hydrolysis of neutral fat to toxic free fatty acids. Cough, hemoptysis, and pleuritic chest pain occur occasionally. Signs include pyrexia, tachypnea, and tachycardia. Acute cor pulmonale with cardiac failure, cyanosis, and circulatory shock may occur.

Symptoms of systemic fat embolism are seen in up to 85% of patients who have pulmonary disease. The symptoms are mainly related to the central nervous system and include confusion, restlessness, stupor, delirium, seizures, and coma. A petechial rash often develops 2 to 3 days after embolization, particularly along the anterior axillary folds and in the conjunctiva and retina.

Pathophysiology

Two pathogenetic mechanisms have been implicated in the development of pulmonary abnormalities after fat embolism. The first is mechanical vascular obstruction, predominantly by fat globules themselves and possibly enhanced in some cases by platelet or red blood cell aggregates. A second potential mechanism involves conversion of neutral triglycerides, the form in which fat appears to be transported to the lungs, into free fatty acids by endothelial lipases. On histologic examination, the presence of fat within pulmonary arterioles and capillaries can be suspected on hematoxylin and eosin–stained sections when there are round to oval spaces, 20 to 40 µm in diameter, apparently compressing red blood cells to one side. Definitive histologic diagnosis requires the use of fat-soluble dyes on unfixed (frozen) tissue or other special techniques.

Manifestations of the Disease

Radiography

The radiographic findings may be subtle and consist of bilateral hazy areas of increased opacity (ground-glass opacities) or patchy consolidation ( Fig. 52.1 ). In more severe cases, widespread consolidation resembles the radiographic appearance of acute respiratory distress syndrome from any cause. In one review of 22 patients with fat embolism syndrome, 2 had normal chest radiographs throughout hospitalization, and 20 developed pulmonary abnormalities. In all cases the radiographic findings consisted of bilateral areas of consolidation consistent with diffuse pulmonary edema. The time to appearance of evident radiographic lung injury was within 24 hours of initial trauma in 50% of cases, between 24 and 48 hours in 20%, and more than 48 hours in 30%. Of the patients with abnormal radiographs, 10 of 20 (50%) had complete resolution of the edema pattern within 1 week of development of opacities and 30% within 1 to 4 weeks; 20% died without resolution of the radiographic findings. Pleural effusions are typically absent.

Computed Tomography

The computed tomography (CT) findings of fat embolism syndrome include bilateral patchy or diffuse ground-glass opacities, patchy or confluent areas of consolidation, and poorly defined centrilobular nodules measuring less than 10 mm in diameter ( Fig. 52.2 ; see also Fig. 52.1 ). CT may demonstrate parenchymal abnormalities in patients with normal chest radiographs. In an investigation of 9 patients with fat embolism syndrome who had normal radiographs, all were found to have abnormalities on high-resolution CT; 7 had ground-glass opacities, and 2 had small nodular opacities. A CT review of 18 patients with fat embolism syndrome demonstrated consolidation with a dependent distribution (see Fig. 52.2 ) and patchy or peripheral predominant ground-glass opacities in the majority of patients. There was often lobular sparing with sharp margination between areas of involved and noninvolved lung that is thought to be secondary to variations in perfusion at time of embolization (see Fig. 52.1 ). Less common manifestations included lobular ground-glass opacities or consolidation, smooth septal thickening, bronchial wall thickening, and a crazy paving pattern. The degree and extent of consolidation and ground-glass opacities correlate well with disease severity. Centrilobular and subpleural nodules are usually seen in association with ground-glass opacities but may be the predominant or only abnormality evident on high-resolution CT in patients with fat embolism. The nodular opacities tend to be located in the centrilobular regions, along the interlobular septa, and along the interlobar fissures. Tree-in-bud opacities are uncommon but have been reported. Intraarterial fat is usually not seen, but the pulmonary trunk is often dilated (size >29 mm). Rarely, there are reports of patients developing pulmonary fibrosis months after fat embolization.

Differential Diagnosis

The time lapse between trauma and radiographic signs of fat embolism is usually 12 to 36 hours. This delay differentiates fat embolism from traumatic lung contusion, in which the radiographic opacity invariably appears immediately after injury. In addition, although the lung contusion usually clears rapidly (in about 24 hours), resolution of fat embolism generally takes 7 to 10 days and occasionally as long as 4 weeks. Further differentiation lies in the extent of lung involvement; contusion seldom affects both lungs diffusely and symmetrically.

Several tests have been evaluated to identify fat embolism, but none of these is specific for the diagnosis. Thrombocytopenia is frequently present and may be associated with disseminated intravascular coagulation (DIC). Hypocalcemia may develop because of the affinity of calcium ions for free fatty acids released by the hydrolysis of embolized fat. Lipiduria is relatively common, and hematuria and proteinuria are seen occasionally.

The diagnosis can be difficult to make, partly because of the relative nonspecificity of signs and symptoms and partly because clinical abnormalities may be related more directly to the cause of the emboli (e.g., trauma-associated shock). Some investigators have advocated the use of bronchoalveolar lavage and analysis of harvested macrophages for the presence of fat. However, patients who do not have fat embolism syndrome may also have lipid-laden macrophages in their bronchoalveolar lavage fluid, and the definitive diagnostic threshold is unclear; some investigators propose a value as low as 5% and others a value as high as 30%.

Synopsis of Treatment Options

There is no specific treatment available for fat embolism syndrome. The mainstay of treatment is therefore supportive. Prophylactic corticosteroids (methylprednisolone) and heparin may have beneficial effects, but their utility is controversial. The outcome in patients who receive supportive care is generally favorable; the mortality is less than 10%.

Etiology, Prevalence, and Epidemiology

Amniotic fluid embolism is a rare but severe complication of pregnancy. In a population-based cohort of 3 million hospital deliveries in Canada between 1991 and 2002, the total rate of amniotic fluid embolism was 14.8 per 100,000 multiple-birth deliveries and 6.0 per 100,000 singleton deliveries. Of the 180 cases of amniotic fluid embolism in women with singleton deliveries during the study period, 24 (13%) were fatal. Medical induction of labor nearly doubled the overall cases of amniotic fluid embolism, and the association was stronger for fatal cases. Maternal age of 35 years or older, cesarean or instrumented vaginal delivery, polyhydramnios, cervical laceration or uterine rupture, placenta previa or abruption, eclampsia, and fetal distress were also associated with an increased risk.

Clinical Presentation

The majority of patients are in the 35th to 42nd week of pregnancy at the time of embolization. The clinical manifestations are typically abrupt, with sudden onset of cardiovascular collapse, cyanosis, and hemorrhage or DIC. In less severe cases the initial manifestation is progressive dyspnea. Although these abnormalities begin during spontaneous labor in most patients, they occur after delivery in about 30% (10% spontaneous and 20% after cesarean section). The majority of patients experience seizures.

Pathophysiology

It is likely that little if any amniotic fluid enters the maternal circulation during normal labor and that significant embolism occurs only when there is disruption of the uterine wall in association with rupture of the placental membranes. Such disruption can occur at several sites, the most common probably being the endocervix or lower uterine segment. Traumatic tears in the small veins in these regions can occur during normal labor but are of no significance if they are covered by fetal membranes; however, if such veins have separated from the fetal membranes, uterine contractions can “pump” amniotic fluid into the maternal venous circulation. Amniotic fluid can also enter the maternal circulation at the placental site, usually in cases of uterine rupture, placenta previa, or cesarean section when the incision involves the site.

The pathophysiologic consequences of intravascular amniotic fluid are complex and probably related to several processes. These include pulmonary vascular obstruction by amniotic fluid particulates, such as meconium; left ventricular dysfunction, possibly as a result of ischemia secondary to pulmonary hypertension and right ventricular dysfunction; pulmonary edema, related to the left ventricular dysfunction just described or to damage to the vascular endothelium by some constituent of amniotic fluid; and an immunologic reaction pathogenetically similar to anaphylaxis. The main histologic abnormality is the presence of squames, mucin, and bile derived from meconium within small pulmonary vessels.

Manifestations of the Disease

Radiography and Computed Tomography

The principal radiographic finding is bilateral airspace consolidation with or without pleural effusions, which is indistinguishable from acute pulmonary edema of other causes ( Fig. 52.3 ). Whether cardiac enlargement accompanies the edema depends on the severity of pulmonary hypertension and consequent cor pulmonale with or without left ventricular failure. The consolidation may persist or resolve within a few days.

Differential Diagnosis

The diagnosis of amniotic fluid embolism is based on the rapid development of a complex constellation of findings with sudden cardiovascular collapse: acute left ventricular failure with pulmonary edema, DIC, and neurologic impairment. Because the predominant radiographic manifestation is widespread airspace consolidation, the main differential diagnoses are massive pulmonary hemorrhage and aspiration of liquid gastric contents. The diagnosis of amniotic fluid embolism should be considered, particularly in patients with risk factors such as medical induction of labor and placenta previa or abruption; it is supported by identification of squames, mucin, or hair fragments in samples of pulmonary capillary blood aspirated through a pulmonary arterial catheter.

Synopsis of Treatment Options

There are no specific pharmacologic or other therapies that prevent or treat the amniotic fluid embolism syndrome. The treatment is therefore supportive and involves aggressive management of multiple types of shock simultaneously. The mortality rate of clinically recognized cases remains high at 19% in a recent study and up to 61% in older cohort studies; 25% to 50% of patients die within the first hour of the disease and most of the rest within 12 hours. Serious neurologic sequelae are common in survivors.