Venous thromboembolism (VTE) includes the spectrum of deep vein thrombosis (DVT) and pulmonary embolism (PE). Embolization of material into the pulmonary venous circulation may lead to marked cardiopulmonary dysfunction and is of particular interest to the critical care practitioner. Although VTE is extraordinarily common in hospitalized patients, estimating the frequency of VTE and PE is problematic. The nonspecific clinical findings and high rate of undiagnosed events likely underestimate the true incidence of disease, whereas autopsy data may overestimate meaningful events by detecting asymptomatic cases. That being considered, the incidence of VTE is thought to be in excess of 600,000 cases per year in the United States [1]. In one population-based study with autopsy data, the annual incidences (age and sex adjusted) of DVT and PE were 48 and 69 per 100,000, respectively [2]. In a study of critically ill patients, 10% to 30% of medical/surgical intensive care unit (ICU) patients experienced DVT within the first week of admission, and approximately 60% of trauma patients had DVT within the first 2 weeks, most of which were clinically silent [3]. The prevalence of VTE as a cause of critical illness is also uncertain, but approximately 15% to 20% of patients with diagnosed PE have significant hemodynamic and/or respiratory compromise [4,5]. Untreated symptomatic PE has a mortality rate of approximately 30%, but treatment reduces this risk considerably [6,7,8,9,10,11,12,13]. Adjusting for patients with concomitant terminal illnesses, acute PE likely account for more than 100,000 deaths per year in the United States in patients with an otherwise good prognosis [14,15].

Recognizing the presence of risk factors for VTE is essential because more than 90% of deaths due to PE occur in patients who are not treated because the diagnosis was unsuspected and thus undetected [14,15]. Virtually every risk factor for VTE can be derived from Virchow’s triad of stasis, venous injury, and hypercoagulability described nearly 150 years ago (Tables 52.1 and 52.2) [16]. Common major risk factors for VTE include increased age, malignancy, surgery, hospitalization with acute medical illness, and a history of prior VTE. Critically ill patients may be at especially high risk for VTE due to severe underlying disease, immobility, and venoinvasive catheters, and the incidence of VTE increases correspondingly with the number of risk factors present.

Inherited or acquired hypercoagulable states are now recognized in more than 20% of patients with VTE and occur at an even higher rate in patients with idiopathic or recurrent VTE (Table 52.3) [17,18]. Resistance to activated protein C due to the factor V Leiden mutation is one of the most common hypercoagulable states, with a prevalence of 5% in white individuals of European ancestry, and may be present in perhaps 20% of patients with PE [19,20,21,22]. Ethnicity is a risk factor, and whites and African Americans have a significantly higher incidence of VTE as compared with Hispanics and Asians [23].

Indwelling central venous catheters (CVCs), particularly common in critically ill patients, provide a constant nidus for clot formation. Thrombosis may occur in up to 67% of patients with invasive catheters [24]. Specific risk factors for thrombotic complications of CVCs may be related to technical or host factors and include femoral vein site, duration of cannulation (> days), nighttime placement, extremes of age, and multiple lines [25,26,27,28]. The frequency of clinically meaningful complications due to CVC-related thrombosis remains unclear.

The sequence of events leading to venous thrombosis is not fully understood and likely varies based on the dynamic interactions between genetic and acquired risk factors. In one proposed scheme, endothelial stimulation results from either blood
stasis-induced hypoxia or direct vein wall injury, after which point tissue factor is transferred to the endothelial cell and initiates the enzymatic cascade of coagulation reactions, leading to thrombin generation and fibrin deposition [29]. Thus, it is not surprising that many venous thrombi arise in valve pockets, where blood flow tends to stagnate, or at a specific area of vascular disruption, such as an indwelling catheter site.

Lower extremity DVT is the most frequent source of PE and, in untreated patients with proximal DVT, approximately half will develop PE [30,31]. Other sources of PE include pelvic, renal, or upper extremity veins, as well as the right heart. After traveling to the lungs, a large thrombus may occlude a major pulmonary artery and cause significant cardiovascular symptoms, or it may break up into smaller clots and travel distally, where it is more likely to produce pleuritic chest pain. Thrombi are most frequently carried to the lower lobes due to the higher blood flow. Pulmonary infarction is relatively uncommon because of incomplete vessel occlusion by emboli and bronchial artery anastomoses.

Ventilation/perfusion ([V with dot above]/[Q with dot above]) mismatch is the principal physiologic effect of PE and leads to hypoxemia in most patients. [V with dot above]/[Q with dot above] mismatch results from increased physiologic dead space and intrapulmonary shunting, which frequently produces an elevated alveolar-to-arteriolar oxygen gradient. Concomitant physiologic responses to PE include increased minute ventilation and airways resistance, as well as decreased vital and diffusion capacities [32,33,34,35,36]. In patients with a potentially patent foramen ovale, progressive pulmonary hypertension may lead to a right-to-left intra-atrial shunt, resulting in worsening hypoxemia and, rarely, paradoxical embolization.

The hemodynamic response to PE may vary depending on the degree of occlusion of the pulmonary arterial circulation and on the presence of underlying cardiopulmonary disease. Physiologically, PE causes a decrease in the cross-sectional area of the pulmonary arterial bed and leads to an increase in the pulmonary vascular resistance. This impedes right ventricular outflow, leading to reduced left ventricular preload and, ultimately, a diminished cardiac output. Progressive vascular obstruction and hypoxemia stimulates vasoconstriction and a further rise in pulmonary artery pressure. The normal right ventricle fails acutely when it cannot generate a systolic pressure to overcome a mean pulmonary artery pressure greater than 40 mm Hg needed to maintain pulmonary perfusion. The normal pulmonary circulation has a large reserve capacity and more than 50% obstruction is generally required for a substantial increase in the mean pulmonary artery pressure [37,38].
Patients with underlying cardiopulmonary disease have less physiologic reserve as compared with healthy individuals, and it follows that they may suffer right heart failure with a lesser degree of pulmonary vascular occlusion [39,40].

Recognizing the presence of VTE may be challenging as neither the signs nor symptoms associated with DVT and PE are sensitive or specific for the diagnoses [13,14,41,42]. Because DVT is usually asymptomatic and most cases of fatal PE are unsuspected prior to death, the most critical step in the diagnosis of VTE is the development of a clinical suspicion of the disease [6,43,44]. This suspicion is based on the constellation of risk factors, symptoms, signs, electrocardiography, blood tests, and chest radiographic findings. Although clinical assessment alone is inadequate in diagnosing and excluding DVT and PE, both clinical gestalt and clinical prediction rules are useful in establishing a pretest probability in which patients are typically classified into three groups based on the estimated prevalence of disease (Table 52.4) [45,46,47,48,49]. This clinical pretest probability serves as the root of algorithms for the diagnosis of DVT and PE [50,51,52,53].

The diagnosis of VTE in the critically ill patient may be particularly challenging. Underlying systemic illnesses or other superimposed acute illness may mimic or mask the common signs and symptoms of VTE. Also, common clinical likelihood models for predicting VTE may not be valid in the ICU setting [54]. Furthermore, definitive testing for VTE may be precluded by relative contraindications, such as mechanical ventilation, shock, and renal failure.