Chapter 10:

Thrombosis and Embolism

Pulmonary embolism (PE) is probably the leading cause of death associated with liposuction. It occurs most frequently with general anesthesia or heavy intravenous (IV) sedation; unnecessary perioperative IV fluid infusion, causing hemodilution; and excessive liposuction.

Thrombosis is a manifestation of a complex series of events leading to vascular inflammation. Recent advances in understanding the pathogenesis of thrombosis have identified common triggering events, including surgery (especially with systemic anesthesia), infection, pregnancy, and malignancy. A relationship may exist between the risk of thromboembolism and surgical or anesthetic techniques. New information may permit the liposuction surgeon to identify more easily and thus avoid patients at increased risk (Box 10-1).

Pulmonary thromboembolism is a major cause of death in surgery patients under general anesthesia for more than 30 minutes.1 A large prospective multicenter study found a 1% incidence of fatal PE among all general surgery patients.2 Another study of postoperative surgery patients found an 18% incidence of abnormal lung scans, consistent with PE.3

The major risk factors for postoperative PE include a history of previous deep venous thrombosis (DVT), trauma, obesity, age greater than 40 years, prolonged immobility, varicose veins, and inherited molecular defects in several hemostatic components. These factors are often found in liposuction patients.

Deep Venous Thrombosis

Since fatal pulmonary emboli are usually the direct result of a DVT, researchers have focused on assessing its incidence. The accuracy of detecting a DVT depends on the diagnostic methodology. Ascending venography is the gold standard, but it is time consuming, invasive, and expensive. Radiolabeled fibrinogen uptake testing and duplex ultrasonography are less accurate.

Excessive liposuction appears to be associated with an increased risk of DVT and fatal pulmonary emboli. A pulmonary embolus is a plug of material, such as a thrombus or fat embolus, that is transported by the bloodstream from a distant site to the lungs.

An acute massive pulmonary embolus or embolon (large embolus) can cause sudden death in the postoperative period. The typical scenario involves an asymptomatic postoperative patient who, on arising from bed, suddenly collapses and dies. The actual pathophysiology is complex.4 The sudden dislodgment of a large, distal venous thrombus and its transportation through the right ventricle into the pulmonary artery can cause a complex pulmonary artery occlusion, tremendous right ventricular strain, autonomic reflex-mediated dysrhythmia, and cardiac arrest. A massive shower of multiple, small pulmonary emboli may have the same effect by initiating widespread mechanical endovascular irritation and precipitating reflex pulmonary arteriolar vasospasm.

Without perioperative antithrombotic prophylaxis, at least 30 cases of DVT diagnosed by phlebography can be expected among 100 patients who have had general surgical procedures of moderate severity.5 Intensive care unit patients have a high rate of DVT.6 Also, patients with DVT have such a high incidence (40%) of asymptomatic PE that some authors believe DVT and PE should be considered a single disease.7,8 DVT of the lower extremity is responsible for almost 90% of all pulmonary emboli.9 Other sources of PE are the deep pelvic vein, renal veins, and inferior vena cava.

Among hospitalized patients with various causes of death, 64% of consecutive autopsies found evidence of subclinical PE.10 Another autopsy study found DVT in 65% of fatally injured trauma patients, and PE was the cause of death in 20%.11 Venous thromboembolism is a common complication in patients with major trauma. DVT in the lower extremities was found in 201 (58%) of 349 trauma patients with adequate venographic studies; before venography, only three patients with DVT had clinical symptoms suggesting DVT.12

When properly diagnosed and treated, PE is an uncommon cause of death,13,14 but the diagnosis of DVT and PE is often missed. An untreated DVT in a thigh has at least a 50% chance of leading to PE and 10% for a fatal PE. The long-term morbidity of the postphlebitic syndrome is valvular imcompetence, collateral reflux, chronic venous hypertension, stasis dermatitis, and ulcerations. Without prophylaxis, the frequency of fatal PE has been estimated at 0.1% to 0.8%.2,15,16 The safest approach in cosmetic surgical patients is to minimize the risks (Box 10-2).

Systemic Anesthesia

In this discussion, systemic anesthesia includes inhalational as well as IV sedation-analgesia that precludes a patient from ambulating to the bathroom to urinate. When all major surgeries required systemic anesthesia, the question of whether or not it was a predisposing factor for DVT and fatal PE was merely of academic interest. With the advent of the tumescent technique for liposuction totally by local anesthesia, however, the question has become more important.

Within the past 15 years, clinicians have begun considering the possible association of systemic anesthesia with DVT, PE, and fatal PE. Because of the relatively high incidence of DVT and PE associated with surgery for hip fractures and elective hip replacement surgery, some of the most revealing studies appear in the orthopedic literature.

The mechanism by which systemic anesthesia predisposes to DVT is not precisely known. Venostasis and impaired venous flow rates have been suggested as mechanisms for precipitating DVT. General anesthesia decreases lower extremity blood flow rates by approximately 50%.17,18 Blood hypercoagulability is significantly greater with general anesthesia than with epidural anesthesia in orthopedic surgery that uses tourniquets.19 General anesthesia for cesarean section is associated with accelerated coagulation compared with spinal anesthesia.20 Blood hypercoagulability is thought to predispose to thromboembolism.21,22 Moderate surgical trauma with blood loss greater than 300 ml can activate thrombin generation with hypercoagulability and fibrinolysis.23

The incidence of pulmonary thromboembolism after total hip replacement with general anesthesia or regional anesthesia (epidural block) was assessed prospectively with random allocation of patients in two studies (n = 60 and n = 94 patients, respectively).24,25 In both studies, using perfusion lung scans, the incidence of PE was 33% with general and 10% with regional anesthesia.

The incidence of DVT after total hip replacement, diagnosed by both venography and fibrinogen uptake, was 53% (n = 47) under general anesthesia and 29% (n = 38) under subarachnoid block.26 Another study of DVT using venography found an incidence of 76% with systemic anesthesia (n = 21) and 40% with subarachnoid block (n = 20).27

A revealing study of total knee replacement found that the incidence of DVT diagnosed by venography was 59% for general anesthesia and 19% for regional anesthesia.28 Both groups used tourniquets for hemostasis, and both groups had prolonged exposure to local anesthesia through low-dose bupivacaine for analgesia.

Among liposuction patients, systemic anesthesia is the greatest risk factor for DVT and PE. I reviewed a number of cases of liposuction-associated fatal PE, and each was associated with systemic anesthesia. Any surgical procedure requiring 30 minutes or more of general anesthesia significantly increases the risk of pulmonary thromboembolism.29

Local anesthesia itself may protect against DVT. In elective hip replacement surgery by general anesthesia, when a lidocaine IV infusion was continued for 6 days postoperatively, the incidence of DVT was 14%, whereas control patients had a 78% incidence.30 Local anesthetics have been shown to inhibit platelet aggregation31 and platelet adhesion.32 Prolonged presence of low-concentration lidocaine with tumescent liposuction could offer an extra margin of safety compared with the tumescent technique using epinephrine but without lidocaine.

Surgical Complications


Hemodilution may act as an acquired form of thrombophilia. One study has shown that hemodilution by an IV infusion of either lactated Ringer’s (LR) solution or normal saline (NS) during surgery may predispose to DVT.33 Among patients given only minimal oral fluids after surgery, only 7% developed DVT; in contrast, 30% of patients who had an IV infusion of crystalloids developed a DVT.

Hemodilution can cause hypercoagulability.34,35 In vitro a 30% hemodilution with NS significantly increases coagulability.36 Hemodilution may produce abnormal hemostasis before any compromise of tissue oxygen delivery.37


Hypothermia is known to predispose to a hypercoagulable state. The delicate biochemical reactions that maintain procoagulant and anticoagulant homeostasis are altered, shifting reaction toward the procoagulant process.

Hypothermia associated with general anesthesia or chilled solutions of tumescent anesthesia may predispose to a consumptive coagulopathy. Hypothetically, the hypothermia resulting from general anesthesia might increase the risk of DVT.


Trauma is a known predisposing factor for thromboembolism.

Multiple Concomitant Procedures. Numerous cases of liposuction-related thromboembolism and death have been reported. Exposing a patient to the cumulative trauma and physiologic stress of multiple surgical procedures performed on a single day may increase the risk of thromboembolism. Liposuction surgeons must be aware of the risks associated with exposure to prolonged and extensive surgical trauma and with prolonged exposure to systemic anesthesia.

Serial liposuction is the practice of dividing an extensive amount of liposuction into two or more separate surgical procedures that are performed sequentially and several weeks apart. Small, individual liposuction procedures performed serially is safer than one traumatic, large-volume liposuction. Similarly, liposuction usually should not be combined with other, unrelated cosmetic surgical procedures.

Circumferential Thigh Liposuction. In my experience, every case of DVT or PE after liposuction has been associated with concomitant liposuction of the inner and outer thighs or with abdominoplasty.

Circumferential liposuction of the thigh with a single, extensive surgical procedure causes prolonged focal lymphedema and distal pedal edema. When this liposuction is separated into two or more serial procedures, the incidence of distal edema is dramatically reduced. The risk of DVT might be reduced by avoiding circumferential thigh liposuction, with its associated pedal edema and lower extremity venostasis.

Excessive Liposuction. Venous thromboembolism is a common complication of major trauma.38-40 The number of areas treated by liposuction and the volume of fat removed are linearly related to the degree of surgical trauma from liposuction, duration of exposure to general anesthesia, and duration of postoperative immobilization. These in turn are directly related to the risk of postoperative DVT, PE, and fatal PE.41,42

Surgeons who use systemic anesthesia for liposuction are faced with a difficult dilemma: is it more dangerous to subject a patient to one aggressive total-body liposuction or to multiple exposures of general anesthesia? No scientific studies support either decision. The dilemma can be avoided by doing liposuction totally by local anesthesia.

Thermal Trauma. Hip surgery is associated with the greatest incidence of DVT and subsequent PE. General anesthesia, direct trauma to veins, extensive exposure of capillary endothelium with local activation of the coagulation cascade, and thermal injury from the heat of polymerizing acrylic may precipitate proximal venous thrombosis.

Extensive liposuction surgery exposes the patient to similar insults. The thermal trauma associated with internal ultrasound-assisted liposuction may increase the risk of DVT and PE. Ultrasonic liposuction machines generate heat and increase the temperature of the subcutaneous tumescent fluid.

Relative Risks

Pregnancy. Pregnancy is a well-recognized risk factor for thromboembolism. The relative risk appears to return to normal 4 weeks after the woman gives birth.

Oral Contraceptives and Estrogens. Current pharmacoepidemiologic data indicate no clinically significant increased risk of thromboembolism with an estrogen dose of less than 35 μg/day. Therefore women can continue taking low-dose estrogen oral contraceptives (OCs) before liposuction.

Also at present, data are insufficient to support the discontinuation of postmenopausal estrogen replacement therapy before liposuction surgery. The doses of estrogens used for treating menopausal symptoms are typically less than the already low doses used for OCs. Most liposuction surgeons do not discontinue low-dose estrogen therapy before liposuction. If a patient has a personal or family history of DVT or thromboembolism, however, the surgeon might consider discontinuing estrogens 4 weeks before surgery.

The first OCs in the 1960s, which contained 150 μg of estrogen, and those of the early 1970s with 80 to 100 μg of estrogen were associated with an increased risk of thromboembolic disease.43-45 Low-dose OCs such as Triphasil contain 30 to 35 μg of estrogen, however, and are now recognized as posing little increased risk for thromboembolism. An odds ratio for stroke of 1.18 has been estimated for users of low-dose oral OCs.46 Studies of subarachnoid hemorrhage have found odds ratios of 0.89,47 1.5,48 and of 1.1 for fatal subarachnoid hemorrhage in low-dose OC users.49 Because stroke is more accurately diagnosed than DVT, epidemiologic data on strokes provide useful clinical information regarding the risk of DVT and PE with OCs.

The U.S. Food and Drug Administration (FDA) requires manufacturers to state that OC users have a twofold to sixfold risk of developing thromboembolic disorders and that OCs should be discontinued 4 weeks before and 2 weeks after elective surgery associated with thromboembolism. Based on the new data for low-dose OCs, the FDA information appears to be outdated.

Smoking. Smoking is believed to increase the risk for thromboembolism. Liposuction patients should be encouraged to discontinue smoking several weeks before surgery. Informed consent should include a warning that cigarette smoking increases the risk of blood clots in the legs, lungs, and brain.

Immobility. The patient should be encouraged to ambulate frequently before and after liposuction, especially during long trips in an airplane or automobile.


The term thrombophilia refers to a group of disorders characterized by an inherited or acquired biochemical, molecular defect that predisposes to thrombosis. Thrombophilia should be suspected in any patient with a personal or family history of an unusual thrombosis, such as thrombosis at an early age (45 years or younger), thrombosis in an unusual site (cerebral, mesenteric, axillary), or multiple recurrent thromboses (Box 10-3).

These patients may or may not have the other wellrecognized risk factors for thromboembolic disease (see Box 10-1). A number of thrombophilias have recently been identified, and new laboratory tests are available in specialized laboratories that can identify specific molecular defects.

The common causes of thrombophilia and the relative prevalence among thrombosis patients include antithrombin III deficiency (2% to 10%); protein C deficiency (2% to 10%); protein S deficiency (2% to 10%); the combined group of patients with either homocystinemia, antiphospholipid syndrome, occult malignancy, or heparin-induced thrombosis (5%); and activated protein C resistance (40%). The remaining 40% result from unknown causes. Preclinical cancer among middle-age patients is a risk factor for PE.50

When thrombosis occurs in a patient with a familial form of thrombophilia, the physician should endeavor to identify and reduce or eliminate the associated risk factors before surgery.

Thrombophilia is not an absolute contraindication for liposuction. Prospective patients with thrombophilia should be identified and provided with information that will permit an informed decision about the risks of liposuction. Liposuction risks can be minimized by minimizing the extent (amount of body surface area) and the degree (volume of supranatant fat) of liposuction. Similarly, concomitant cosmetic procedures, the use of systemic anesthesia, postoperative immobility, and postoperative inflammation should be avoided or minimized.

As mentioned, a number of genetic causes of thrombophilia have recently been identified (Box 10-4). Other identifiable hypercoagulable states are now recognized as surgical risks, including the antiphospholipid syndrome (lupus anticoagulant syndrome, anticardiolipin antibody syndrome51), and a few rare forms of dysfibrinogenemia.

Understanding the clinical indications for preoperative laboratory tests to screen for heritable thrombophilia requires up-to-date knowledge (Box 10-5). The annual incidence of recurrent venous thromboembolism is highest during the first years after a first episode, then declines over time.52 Thrombophilia interacts with predisposing environmental factors, resulting in thrombosis. A predisposing factor was observed in 84% of patients with factor V Leiden (Arg506Gln) mutation who developed thromboembolism.53

Antithrombin III Deficiency

Antithrombin III (ATIII) inhibits thrombin and activated factors IXa and Xa. ATIII is a natural anticoagulant glycoprotein that is greatly enhanced by heparin. This is the mechanism for the use of heparin in antithrombotic therapy.

Hereditary ATIII deficiency is inherited as an autosomal dominant trait associated with either a quantitative or a qualitative deficiency. Acquired ATIII deficiency can result from decreased synthesis and increased loss of protein. Acquired ATIII deficiency is associated with hepatic disease, disseminated intravascular coagulation (DIC), medications (l-asparaginase, heparin, estrogens), nephrotic syndrome, hemodilution (unnecessary IV fluids), and malnutrition.

Patients with ATIII deficiency have evidence of continuous factor Xa generation and activation of prothrombin in vivo, leading to high plasma concentrations of prothrombin fragment 1.2. Because of other antithrombotic mechanisms, such as protein C/protein S pathway, overt thrombosis is not as common with ATIII deficiency. When these pathways are suppressed by surgery, systemic anesthesia, or sepsis, however, sufficient thrombin is generated to cause thrombosis.

Women with an inherited ATIII deficiency, deficiencies of proteins C or S, or activated protein C (APC) resistance have an increased risk of pregnancy-associated venous thrombosis and an increased risk of fetal loss. This risk appears to be greatest for some types of ATIII deficiency.54

Proteins C and S

Protein C and protein S are produced by the liver. They are the principal molecules of the major anticoagulant system involved in maintaining the natural homeostatic balance between procoagulants and anticoagulants. Both protein C and protein S require vitamin-K–dependent posttranslational modification to function.

Protein C circulates in an inactive form. During the coagulation process, prothrombin (factor II) is converted to thrombin, which then interacts with an endothelial surface protein known as thrombomodulin. On forming the thrombin-thrombomodulin complex, the substrate specificity of thrombin changes from fibrin to protein C, resulting in the conversion of protein C to APC.

Protein S, when located on phospholipid surfaces, is a cofactor for APC. The APC–protein S complex inactivates coagulation factors Va and VIIIa by proteolytic cleavage, thus attenuating the clotting process.

Patients who are heterozygous for protein C deficiency or protein S deficiency have an increased risk of recurrent venous thrombosis at a young age (less than 40). Patients who are homozygous for protein C or S deficiency have neonatal purpura fulminans, manifested by generalized microvascular thrombosis.

These deficiencies may be either quantitative (decreased amount of normally functioning protein) or qualitative (normal amounts of dysfunctional protein). A genetic mutation can cause a qualititative protein deficiency. Vitamin K deficiency causes dysfunctional proteins C and S; thus preoperative vitamin K might prevent bleeding diatheses and help maintain anticoagulant-coagulant homeostasis.

These deficiencies may be either an autosomal dominant genetic trait or an acquired disorder associated with hepatic disease, surgery, inflammation, sepsis, and possibly estrogens. Familial protein S deficiency has been reported in association with decreased concentrations of free (unbound) protein S.55 Combined protein C and protein S deficiency increases the risk for thrombophilia.56

Acquired Deficiencies. An acquired deficiency of protein C or protein S can lead to an increased production of thrombin and thus an increased risk of thrombosis and thromboembolism.

The protein C pathway is affected by inflammation. Inflammation causes a decrease in the concentration of free protein S. Similar to drugs that are highly protein bound, only the free fraction of the total circulating protein is available to participate in biochemical reactions. Determination of protein S activity and antigens allows separation of qualitative and quantitative protein deficiencies.57

Pregnancy is another example of acquired decrease in the concentrations of protein C and protein S. Pregnancy reduces protein S by 40% to 50% of normal levels, but it is not certain whether this lowered protein S concentration is responsible for the increased risk of thromboembolism in pregnant women.58

A disease association has been reported in a patient with acquired immunodeficiency syndrome (AIDS), protein S deficiency, and intracranial thrombosis.59

The fraction of protein S bound to tissue or to circulating molecules is essentially a reservoir for protein S, which becomes available as the free fraction is consumed. A fraction of the circulating protein S is bound to circulating C4b binding protein (C4bBP), where C4b is a regulatory protein of the complement system. Increased inflammation augments the concentration of C4bBP, which binds a greater proportion of protein S and thus reduces the active unbound (free) fraction.

Sepsis has been reported in association with an acquired decrease in protein C and protein S in an 11-year-old girl heterozygous for the Arg506Gln mutation of factor Va (APC resistance).60 Endotoxins (sepsis) and cytokines (inflammation) inhibit the expression of protein C receptors on activated monocytes and endothelial cell surfaces, stimulating the expression of coagulation tissue factors. This shifts the biochemical equilibrium toward production of more thrombin.

Another study found that three of five children with steroid-resistant nephrotic syndrome had an acquired form of protein S deficiency with reduced free protein S in plasma.61

Since inflammation or anesthesia induces an acquired state of thrombophilia, the risk of surgical thromboembolism might be reduced by modifying surgical techniques to decrease perioperative inflammation and exposure to offending drugs.

Certain surgical practices might help induce a clinically significant deficiency of important procoagulant molecules. Likely factors include some forms of general or systemic anesthesia, excessive surgical trauma, hemodilution from excessive IV fluids, and postoperative inflammation. In particular, postliposuction inflammation from ultrasonic thermal tissue damage, as well as prolonged subcutaneous accumulation of blood-tinged tumescent fluid, might predispose to thrombosis.

Activated Protein C Resistance (Factor V Leiden)

APC resistance is either inherited (factor V Leiden) or acquired. In the absence of factor V Leiden, APC resistance has been observed in pregnancy, in patients taking OCs, in the presence of antiphospholipid antibodies, and in patients with ischemic stroke. A recent report suggests evidence for an acquired antibody against APC. Thus anti-APC antibodies may be a potential cause of thrombosis.62

High-density lipoprotein (HDL) enhances the anticoagulant protein C pathway in vitro, which may explain one of HDL’s beneficial effects.63 Also, heparin and APC might act synergistically to inactivate factor V.64

APC resistance was described as a predisposing factor for thromboembolism in 1993.65 This disorder might account for a significant proportion of spontaneous DVT and thromboembolic disease in young, healthy, cosmetic surgical patients. APC resistance is associated with a threefold to sevenfold increase in the risk for DVT, and venous thrombosis occurs in 40% of patients with APC resistance.

Most patients with APC resistance have an abnormality of coagulation factor V (Arg506Gln) that causes factor Va to resist degradation by APC. Genetic analysis of 15 patients with APC resistance revealed nine (60%) who were positive for the Arg506Gln mutation affecting factor V.66 The genetic mutation Arg506Gln is a nucleotide substitution in which arginine (Arg) is replaced by glycine (Gln) at position 506 of the gene encoding for coagulation factor Va.

The defective factor Va that results from Arg506Gln is referred to as factor V Leiden, after the city where it was discovered. Laboratory evaluation of factor V Leiden is commercially available. The risk of thromboembolic events is significantly higher in carriers of factor V Leiden than in patients without this abnormality.67 This mutation increases the risk of DVT by a factor of 10 in heterozygotes and 50 in homozygotes. Approximately 10% of patients with DVT and APC resistance do not have the factor V Leiden (Arg506Gln) mutation.

The cumulative risk of recurrent thrombosis in patients with APC resistance is 71% without long-term anticoagulation therapy. APC resistance affects 5% of the general population and 40% of Caucasians who have a familial form of thrombosis. Other ethnic groups (e.g., Africans, Asians) have a much lower incidence of the Arg506Gln mutation. APC resistance is the most prevalent abnormality of inherited thrombophilia.68

The prevalence of factor V Leiden in the general European population is estimated to be 5% to 6%.69 This mutation is significantly more frequent in patients with chronic venous insufficiency and venous leg ulcers.70


A novel sequence variation in the 3’-untranslated region of the prothrombin (factor II) gene, a G-to-A transition at nucleotide position 20210, has recently been identified as a risk factor for familial thrombophilia with DVT and PE. It is found in approximately 1% to 4% of subjects in Western Europe.71 An increased frequency of nucleotide 20210 G→A mutation is associated with factor V Leiden.72,73 The risk of recurrent DVT is similar among carriers of factor V Leiden and patients without this mutation. Carriers of both factor V Leiden and the G20210A prothrombin mutation have an increased risk of recurrent DVT after a first episode and are candidates for lifelong anticoagulation.74


Heterozygous hyperhomocystinemia is a significant independent risk factor for venous occlusive disease, in addition to its better known reputation as a predisposing factor for arterial disease.75 Patients with hyperhomocystinemia (95th percentile or greater) have twice the risk of DVT as normal controls.

Hyperhomocystinemia is caused either by genetic mutations in genes encoding for the enzymes for homocystine metabolism (e.g., cystathionine beta-synthetase, methylenetetrahydrofolate reductase) or by nutritional factors (e.g., folate deficiency).

The free sulfhydryl group on homocystine makes it a highly reactive amino acid that may be directly toxic to vascular endothelium. Therefore homocystine may inhibit thrombomodulin expression, inhibit protein C activation, suppress heparin sulfate expression, and impair expression of nitric oxide.

Antiphospholipid Antibodies

The antiphospholipid antibody syndrome affects a heterogeneous group of patients with thrombosis, recurrent abortion, or thrombocytopenia who have antiphospholipid antibodies that are either lupus anticoagulant (LA) or anticardiolipin (ACL) antibodies. Cell membranes contain procoagulant anionic phospholipids that are important in phospholipiddependent coagulation reactions.

The annexins are a family of proteins described in 1990 and characterized by repetitive homologous domains consisting of sequences of about 70 amino acids. The annexins share the property of binding calcium and phospholipids. Annexin V forms a protective shield over the procoagulant anionic phospholipids and inhibits phospholipid-dependent coagulation reactions. Patients with antiphospholipid antibody syndrome have immunoglobulin G (IgG) antibodies against phospholipid-binding protein annexin V.76 Without adequate functioning annexin V to shield the phospholipids on cell membranes, the exposed phospholipids are available to accelerate coagulation reactions. Thus the annexins may have a unique role in regulating coagulation.77

Antiphospholipid antibodies, which can be detected by coagulation tests and immunoassay methods, are considered an acquired risk factor for thromboembolism.

Case Studies

Recently I had a rather obese patient who was hospitalized overnight for diagnostic workup of a superficial thrombophlebitis of the distal greater saphenous vein after liposuction of the hips and outer thighs. The patient was treated with aspirin and discharged home. On questioning, this patient admittted having some pain in her leg for more than a week before surgery, but she did not think it was sufficiently bothersome to mention it to me.

As with this case example, Case Reports 10-1 to 10-4 emphasize the need for careful patient selection and monitoring during liposuction to avoid potential risk factors for DVT, PE, and fatal PE.

Fat Embolism (Embolus) Syndrome

The fat embolus syndrome is a poorly understood clinical entity that might have clinical relevance for the liposuction surgeon. Fat emboli have been reported in association with inhalational anesthesia.78 There are no reports of fat embolism syndrome when liposuction was the only surgical procedure; reports of the syndrome occurring with liposuction are usually associated with a concomitant abdominoplasty. To my knowledge, no case of fat embolism has occurred with tumescent liposuction totally by local anesthesia.

Fat embolism syndrome is an uncommon clinical entity that consists of pulmonary insufficiency, coagulopathy, neurologic impairment, and clinical evidence of circulating fat globules. The syndrome most often occurs with bone trauma and orthopedic surgery but has also been reported with hemorrhagic pancreatitis, carbon tetrachloride poisoning, extracorporeal circulation, rapid high-altitude decompression, liver trauma, blast concussion, bone marrow transplantation, and closed-chest cardiac massage.

Almost every fracture produces a degree of fat embolism. Only 1% patients with a single fracture develop clinical pulmonary distress, coagulopathy, and neurologic symptoms of fat embolism syndrome, whereas 5% to 10% of patients with pelvic or multiple long bone fractures develop symptoms.78 Symptoms can develop within 2 hours after the traumatic event but typically are delayed for 1 to 3 days, presenting as respiratory distress, lethargy, confusion, and other symptoms of brain injury. Mortality from fat embolism syndrome is estimated at 10% to 15%.

Unlike the syndrome, the incidence of fat embolism is an extremely common pathologic finding after fatal accidental trauma.79 The incidence of fat embolism in an autopsy study of 300 accident victims ranged from 80% to 100%, with the higher incidence occurring in patients who survived for 12 hours or more after injury. In other studies the incidence ranged from 26% with single fractures to 44% of patients with multiple fractures.


The source of the fat globules is controversial and not well understood. Several unrelated sources of intravascular lipid globules may exist, and several distinct clinical syndromes may simply be grouped under one name. In trauma victims, fat globules from the marrow of fractured bones may enter the circulation through lacerated vessels. When fat embolism is diagnosed in the absence of trauma or surgery, the fat globules may result from the coalescence of plasma chylomicrons or a disturbance of plasma lipid chemistry that precipitates liposome-like globules.

The ultimate source of tissue injury is also controversial.80,81 The theory that physical occlusion of the microvascular circulation is the source of pathologic tissue injury is overly simplistic. Tissue injury and platelet activation probably cause the release of significant circulating inflammatory mediators, including prostaglandins, cytokines, free fatty acids, and tissue lipases. Within limits, fatty acids in the blood are bound to albumin; when trauma or hypermetabolism generates excessive circulating free fatty acids, the carrying capacity of albumin may be exceeded, exposing tissues to a systemic inflammatory response.82 The ingestion of fat emboli by pulmonary macrophages may release free fatty acids and precipitate an intense inflammatory response.83

Patients who have sustained severe trauma are at risk for both DIC and fat embolism syndrome. Although blood coagulation factor XII is activated by saturated fatty acids in vitro, it is not clear if this represents a cause-and-effect relationship in the pathogenesis of DIC associated with fat embolism.84


Deep Venous Thrombosis

No evidence indicates that a prophylactic treatment of liposuction patients with low-molecular-weight heparin is safe or effective in preventing DVT. Although such prophylaxis has an intuitive appeal and its benefits appear plausible, routine heparin use in liposuction patients might be more dangerous than no treatment at all. Because of the extensive trauma to subcutaneous capillaries, liposuction cannot be considered analogous to most other surgical procedures where heparin might be used.

The safest means to prevent DVT is to minimize exposure to the greatest risk factors, such as general anesthesia, circumferential thigh liposuction, and too much liposuction.

Pulmonary Embolism

Three important strategies exist to reduce the risk of fatal PE among liposuction patients. One strategy is the early diagnosis of DVT and nonfatal PE, but this is both expensive and ineffective in preventing fatal PE. Another strategy is to identify patients at high risk for postoperative PE and treat prophylactically with subcutaneous low-dose heparin. As noted, however, heparin may be relatively contraindicated in liposuction because of the associated bleeding tendency. The only reasonable approach is prevention.

The most important aspect of prevention is prudent patient selection. Patients at significantly increased risk must be identified. The liposuction surgeon must ask every patient about a personal or family history of blood clots in the legs or lungs. Patients who have a suspicious history should have a hematologic consultation to identify better the degree of the relative risk.

The next step in prevention is the exercise of prudent surgical decisions. Many risk factors for iatrogenic thromboembolism are not well recognized. It seems prudent to minimize suspected risks for iatrogenic thrombotic disease, such as exposure to general anesthesia or to thermal injury associated with ultrasonic liposuction. Exceedingly large-volume liposuction procedures (too many areas or too much volume) are unnecessarily dangerous; again, sequential liposuctions are safer than a single, huge-volume procedure.

Box 10-6 lists other measures to reduce the risk of thromboembolism.


If surgeons have concerns about liposuction-related pulmonary emboli, they should learn about the proximal causes. If general anesthesia is a significant cause of DVT and fatal pulmonary emboli, surgeons and patients must be aware of this danger. Surgeons and anesthesiologists who use systemic anesthesia should cooperate with epidemiologists to investigate the hypothesis that general anesthesia increases the risk of fatal pulmonary emboli.

Surgeons should learn how to do liposuction totally by local anesthesia. If liposuction can be accomplished totally by local anesthesia, patients should be informed of the alternative choices for liposuction anesthesia.

Cosmetic surgeons ultimately want to assess all factors that predispose a liposuction patient to DVT. The immediate concern is not DVT, however, but the relative risk for postliposuction fatal PE, with and without general anesthesia.


  1. Rose SD: Prophylaxis of thromboembolic disease, Med Clin North Am 63:1205, 1979.
  2. International Multicenter Trial: Prevention of fatal postoperative pulmonary embolism by low doses of heparin, Lancet 2:45, 1975.
  3. Browse NC, Clemenson G, Croft DN: Fibrinogen detectable thrombosis in the legs and pulmonary embolism, Br Med J 1:603, 1972.
  4. Weinmann EE, Salzman EW: Deep-vein thrombosis, N Engl J Med 331:1630-1638, 1994.
  5. Gallus AS, Salzman EW, Hirsh J: Prevention of venous thromboembolism. In Colman RW, Hirsh J, Marder VJ, Salzman EW, editors: Hemostasis and thrombosis: basic principles and clinical practice, ed 3, Philadelphia, 1994, Lippincott.
  6. Hirsch DR, Ingenito EP, Goldhaber SZ: Prevalence of deep venous thrombosis among patients in medical intensive care, JAMA 274:335-337, 1995.
  7. Huisman MV, Buller HL, Cate JW, et al: Unexpected high prevalence of silent pulmonary embolism in patients with deep vein thrombosis, Chest 95:498-502, 1989.
  8. Moser KM, Fedullo PF, LitteJohn JK, Crawford R: Frequent asymptomatic pulmonary embolism in patients with deep venous thrombosis, JAMA 271:223-225, 1994.
  9. Hull RD, Pineo GF: Deep venous thrombosis and anticoagulation. In Ouriel K, editor: Lower extremity vascular disease, Philadelphia, 1995, Saunders.
  10. Sandler DA, Martin JF: Autopsy proven pulmonary embolism in hospital patients: are we detecting enough deep vein thrombosis? J R Soc Med 82:203-205, 1989.
  11. Sevitt S: Fatal road accidents: injuries, complications, and causes of death in 250 subjects, Br J Surg 55:481-505, 1968.
  12. Geerts WH, Code KI, Jay RM, et al: A prospective study of venous thromboembolism after major trauma, N Engl J Med 331:1602-1606, 1994.
  13. Carson JL, Kelley MA, Duff A, et al: The clinical course of pulmonary embolism, N Engl J Med 326:1240-1245, 1992.
  14. Douketis JD, Kearon C, Bates S, et al: Risk of fatal pulmonary embolism in patients with treated venous thromboembolism, JAMA 279:458-462, 1998.
  15. Skinner DB, Salzman EW: Anticoagulant prophylaxis in surgical patients, Surg Gynecol Obstet 125:741, 1967.
  16. Shephard RM, White HA, Shirkey AL: Anticoagulant prophylaxis of thromboembolism in postsurgical patients, Am J Surg 112:498, 1966.
  17. Clark C, Cotton LT: Blood-flow in deep veins of leg: recording technique and evaluation of methods to increase flow during operation, Br J Surg 55:211-214, 1968.
  18. Kemble JVH: The effect of surgical operation on leg venous flow measured with radioactive hippuran, Postgrad Med J 47:773-776, 1971.
  19. Kohro S, Yamakage M, Arakawa J, et al: Surgical/tourniquet pain accelerates blood coagulability but not fibrinolysis, Br J Anaesth 80:460-463, 1998.
  20. Sharma SK, Philip J: The effect of anesthetic techniques on blood coagulability in parturients as measured by thromboelastography, Anesth Analg 85:82-86, 1997.
  21. Jennings S, Cass AJ, Heather BP, Greenhalgh RM: Coagulation changes during major surgery and relationship to post-operative deep vein thrombosis, J Cardiovasc Surg 22:327-329, 1981.
  22. O’Brien JR, Tulevski VG, Etherington M, Madgwick T: Platelet function studies before and after operation and the effect of postoperative thrombosis, J Lab Clin Med 83:342-354, 1974.
  23. Payen J-F, Baruch M, Horvilleur E, et al: Changes in specific markers of haemostasis during reduction mammoplasty, Br J Anaesth 80:464-466, 1998.
  24. Modig J, Borg T, Karlstrom G, Sahlstedt B: Thromboembolism after total hip replacement: the role of epidural and general anesthesia, Anesth Analg 62:174-180, 1983.
  25. Modig J, Maripuu E, Sahlstedt B: Thromboembolism following total hip replacement: a prospective investigation of 94 patients with emphasis on efficacy of lumbar epidural anesthesia in prophylaxis, Reg Anesth 11:72-79, 1986.
  26. Thorburn J, Loudon JR, Vallance R: Spinal and general anesthesia in total hip replacement: frequency of deep vein thrombosis, Br J Anaesth 52:1117-1121, 1980.
  27. McKenzie PJ, Wishart HY, Gray I, Smith G: Effects of anesthetic technique on deep vein thrombosis: a comparison of subarachnoid and general anesthesia, Br J Anaesth 57:583-587, 1985.
  28. Jorgensen LN, Rasmussen LS, Nielsen PT, et al: Antithrombotic efficacy of continuous extradural anesthesia after knee replacement, Br J Anaesth 66:8-12, 1991.
  29. Moser KM: Pulmonary thromboembolism. In Wilson JD, Braunwald E, Isselbacher KJ, et al, editors: Harrison’s Principles of internal medicine, ed 12, New York, 1991, McGraw-Hill.
  30. Cooke ED, Lloyd MJ, Bowcock SA, Pilcher MF: Intravenous lignocaine in prevention of deep venous thrombosis after elective hip surgery, Lancet 2:797-799, 1977.
  31. Feinstein MB, Fiekers J, Fraser C: An analysis of the mechanism of local anesthetic inhibition of platelet aggregation and secretion, J Pharmacol Exp Ther 197:215-228, 1976.
  32. O’Brien JR: The adhesiveness of native platelets and its prevention, J Clin Pathol 14:140-149, 1961.
  33. Janvrin SB, Davies G, Greenhalgh RM: Postoperative deep vein thrombosis caused by intravenous fluids during surgery, Br J Surg 67:690-693, 1980.
  34. Rutmann TG, James MFM, Viljoen JF: Haemodilution induces a hypercoagulable state, Br J Anaesth 76:412-414, 1996.
  35. Ng KF, Lo JW: The development of hypercoagulability state, as measured by thromboelastography, associated with intraoperative surgical blood loss, Anaesth Intensive Care 24:20-25, 1996.
  36. Egli GA, Zollinger A, Seifert B, et al: Effect of progressive haemodilution with hydroxyethyl starch, gelatin, and albumin in blood coagulation, Br J Anaesth 78:684-689, 1997.
  37. McLoughlin TM et al: Profound normovolemic hemodilution: hemostatic effects in patients and in a porcine model, Anesth Analg 83:459-465, 1996.
  38. Geets WH, Code KI, Jay RM, et al: A prospective study of venous thromboembolism after major trauma, N Engl J Med 331:1601-1606, 1994.
  39. Shackford SR, Moser KM: Deep vein thrombosis and pulmonary embolism in trauma patients, J Intensive Care Med 3:87-98, 1988.
  40. Consensus Development Panel: Prevention of venous thrombosis and pulmonary embolism, JAMA 256:744-749, 1986.
  41. Borrow M, Goldson H: Postoperative venous thrombosis: evaluation of five methods of treatment, Am J Surg 141:245, 1981.
  42. Moser KM: Thromboembolic disease in the patient undergoing urologic surgery, Urol Clin North Am 10:101, 1983.
  43. Jordan WM: Pulmonary embolism, Lancet 2:1146-1147, 1961.
  44. Inman WHW, Vessey MP: Investigation of deaths from pulmonary, coronary, and cerebral thrombosis and embolism in women of child-bearing age, BMJ 2:193-199, 1968.
  45. Collaborative Group for the Study of Stroke in Young Women: Oral contraception and increased risk of cerebral ischemia or thrombosis, N Engl J Med 288:871-887, 1973.
  46. Petitti DB, Sidney S, Bernstein A, et al: Stroke in users of low-dose oral contraceptives, N Engl J Med 335:8-15, 1996.
  47. Longstreth WT, Nelson LM, Koepsell TD, van Belle G: Subarachnoid hemorrhage and hormonal factors in women: a population-based case-control study, Ann Intern Med 121: 168-173, 1994.
  48. Hannaford PC, Croft PR, Kay CR: Oral contraceptives and stroke: evidence from the Royal College of General Practitioners’ Oral Contraceptive Study, Stroke 25:935-942, 1994.
  49. Thorogood M, Mann J, Murphy M, Vessey M: Fatal stroke and the use of oral contraceptives: findings from a casecontol study, Am J Epidemiol 136:35-45, 1992.
  50. Sørensen HT, Mellemkjaer L, Steffensen FH, et al: The risk of a diagnosis of cancer after primary deep venous thrombosis or pulmonary embolism, N Engl J Med 338:1169-1173, 1998.
  51. Deschiens MA, Conard J, Horellou MH, et al: Coagulation studies, factor V Leiden, and anticardiolipin antibodies in 40 cases of cerebral venous thrombosis, Stroke 27:1719-1720, 1996.
  52. van den Belt AG, Sanson BJ, Simioni P, et al: Recurrence of venous thromboembolism in patients with familial thrombophilia, Arch Intern Med 157:2227-2232, 1997.
  53. Samama MM, Simon D, Horellou MH, et al: Diagnosis and clinical characteristics of inherited activated protein C resistance, Haemostasis 26(suppl 4):315-330, 1996.
  54. Walker ID: Congenital thrombophilia, Baillieres Clin Obstet Gynaecol 11:431-445, 1997.
  55. Gokcil Z, Odabasi Z, Vural O, Yardim M: Cerebral venous thrombosis in pregnancy: the role of protein S deficiency, Acta Neurol Belg 98:36-38, 1998.
  56. Lefebvre P, Lierneux B, Lenaerts L, et al: Cerebral venous thrombosis and procoagulant factors—a case study, Angiology 49:563-571, 1998.
  57. Aillaud MF, Pouymayou K, Brunet D, et al: New direct assay of free protein S antigen applied to diagnosis of protein S deficiency, Thromb Haemost 75:283-285, 1996.
  58. Paternoster DM, Simioni P, Girolami A: Protein S deficiency and pregnancy, Minerva Ginecol 49:447-448, 1997.
  59. Iranzo A, Domingo P, Cadafalch J, Sambeat MA: Intracranial venous and dural sinus thrombosis due to protein S deficiency in a patient with AIDS, J Neurol Neurosurg Psychiatry 64:688, 1998.
  60. Shavit I, Brenner B, Lanir N, et al: Coexistence of acquired protein S and protein C deficiency and the Arg 506Gln mutation in factor Va in a child with severe thromboembolic disease, Acta Paediatr 87:349-350, 1998.
  61. Hanevold CD, Lazarchick J, Constantin MA, et al: Acquired free protein S deficiency in children with steroid resistant nephrosis, Ann Clin Lab Sci 26:279-282, 1996.
  62. Zivelin A, Gitel S, Griffin JH, et al: Extensive venous and arterial thrombosis associated with an inhibitor to activated protein C, Blood 94:895-901, 1999.
  63. Griffin JH, Kojima K, Banka CL, et al: High-density lipoprotein enhancement of anticoagulant activities of plasma protein S and activated protein C, J Clin Invest 103:219-227, 1999.
  64. Petaja J, Fernandez JA, Gruber A, Griffin JH: Anticoagulant synergism of heparin and activated protein C in vitro: role of a novel anticoagulant mechanism of heparin, enhancement of inactivation of factor V by activated protein C, J Clin Invest 99:2655-2663, 1997.
  65. Svennson P, Dahlback B: Resistance to activated protein C as a basis for venous thrombosis, N Engl J Med 330:517-522, 1993.
  66. Gillespie DL, Carrington RL, Alving BM: Resistance to activated protein C: a common inherited cause of venous thrombosis, Ann Vasc Surg 10:174-177, 1996.
  67. Simioni P, Prandoni P, Lensing AWA, et al: The risk of recurrent venous thromboembolism in patients with an Arg506→ Gln mutation in the gene for factor V (factor V Leiden), N Engl J Med 336:399-403, 1997.
  68. Gaussem P, Siguret V, Aiach M: [Evaluation of hemostasis in venous thromboembolism pathology], Ann Biol Clin (Paris) 56:49-56, 1998.
  69. Ridker PM, Miletich JP, et al: Ethnic distribution of factor V Leiden in 4047 men and women, JAMA 277:1305-1307, 1997.
  70. Maessen-Visch MB, Hamulyak K, et al: The prevalence of factor V Leiden mutation in patients with leg ulcers and venous insufficiency, Arch Dermatol 135:41-44, 1999.
  71. Biousse V, Conard J, et al: Frequency of the 20210 G→A mutation in the 3’-untranslated region of the prothrombin gene in 35 cases of cerebral venous thrombosis, Stroke 29:1398-1400, 1998.
  72. Conard J, Mabileau-Brouzes C, et al: [Multigenic thrombophilia: genetic anomaly of factor II and mutation of factor V Leiden: study in a French family], Presse Med 26:951-953, 1997.
  73. Prisco D, Gori AM, et al: Factor II 20210 G→A polymorphism associated to factor V Leiden: a report of two thrombophilic families, Thromb Res 89:249-252, 1998.
  74. De Steffano V, Martinelli I, Mannucci PM, et al: The risk of recurrent deep venous thrombosis among heterozygous carriers of both factor V Leiden and the G20210A prothrombin mutation, N Engl J Med 341:801-806, 1999.
  75. Heljer MD, Koster T, et al: Hyperhomocystinemia as a risk factor for deep-vein thrombosis, N Engl J Med 334:759, 1996.
  76. Rand JH, Wu XX, Andree HAM, et al: Reduction of annexin-V (placental anticoagulant protein-I) on placental villi of women with antiphospholipid-antibodies and recurrent spontaneous abortion, Am J Obstet Gynecol 171:1566-1572, 1994.
  77. Menell JS, Cesarman GM, Jacovina AT, et al: Annexin II and bleeding in acute promyelocytic leukemia, N Engl J Med 340:994-1004, 1999.
  78. Shier MR, Wilson RF: Fat embolism syndrome: traumatic coagulopathy with respiratory distress, Surg Annu 12:139-168, 1980.
  79. Hiyama DT, Zinner MJ: Surgical complications. In Schwartz SI, Shires GT, Spencer FC, editors: Principles of surgery, ed 6, New York, 1994, McGraw-Hill.
  80. Gossling HR, Donohue TA: The fat embolism syndrome, JAMA 241:2740-2742, 1979.
  81. Hulman G: Pathogenesis of non-traumatic fat embolism, Lancet 1:1366-1367, 1988.
  82. Moylan JA, Birnbaum M, Katz A, Everson MA: Fat emboli syndrome, J Trauma 16:341-347, 1976.
  83. Peltier LF: Fat embolism. III. The toxic properties of neutral fat and free fatty acids, Surgery 40:665-670, 1956.
  84. Bradford DS et al: Coagulation alterations, hypoxemia, and fat embolism in fracture patients, J Trauma 10:307, 1970.