Surgical Embolectomy 333 patients (Fig. 3b) and only 20% could be attributed to the history of pulmonary embolism—two following perioperative neurological injury, two from recurrent pulmonary embolism, and one from anticoagulation-related hemorrhage. Four deaths from unknown causes may have been related to pulmonary embolism. The majority of deaths, however, were from unrelated causes, mostly cancer (36%) and trauma (16%). Four studies report actuarial survival data (65,66,71,85) These survival curves all show a steep initial decline reflecting the 20 to 40% operative mortality and a very flat portion thereafter reflecting the excellent mid- and long-term sur- vival in operative survivors (Fig. 4). Medium-term (4–5 year) survival was 65 to 80% (65,66,71,85) and long-term (8–10 year) survival was 62 to 71% (65,66). Even patients who had suffered preoperative cardiac arrest and survived to dis- charge had excellent long-term survival (66). 2. Recurrent Pulmonary Embolism Clinically significant recurrent pulmonary embolism appears to be rare. Two pa- tients, noted above, died from recurrent embolism. There was only one additional report of a nonfatal embolism (71). Although these retrospective studies might underestimate the incidence, they are consistent with the relatively low rate of recurrent thromboembolism in patients treated medically after pulmonary embo- lism (86). In addition, most, but not all, centers routinely performed inferior vena Figure 4 Actuarial survival after pulmonary embolectomy with and without cardiopul- monary resuscitation (CPR). (From Ref. 66.) 334 Aklog caval interruption to protect against recurrent embolism. The three patients who did suffer recurrent embolism had not undergone inferior vena caval interruption either as a matter of policy (75) or for technical reasons (71). 3. Functional Status Six studies document New York Heart Association functional class in 176 hospi- tal survivors at a mean follow-up of 71 months (65,66,71,73,83,85). Nearly all (98%) were in Class I or II (Fig. 5). One nonrandomized study from the mid- 1980s compared late functional status among patients treated with heparin, strep- tokinase, or embolectomy (85). Although the surgical patients were more ill at presentation, their late functional status was better with 45% in Class I compared to 17% of medical patients. No surgical patient was in Class III or IV compared to 10% of medical patients. A recent study sought to correlate late functional status with objective diag- nostic studies in 19 of 21 survivors of embolectomy at a mean follow-up of 8.4 years (83). All patients underwent physical examination, ventilation-perfusion scintigraphy, echocardiography, and pulmonary function testing. Class III pa- tients also underwent right heart catheterization and pulmonary arteriography. Using an explicit set of criteria, three-fourths of patients had at least one minor abnormality and one-fourth had a major abnormality. The presence of minor or major abnormalities correlated well with the patient’s functional status, with ma- jor abnormalities occurring in only 12% of Class I or II patients but all Class III patients. Eighty percent of patients with no abnormalities were in Class I. Figure 5 Late functional status after pulmonary embolectomy in those major surgical series listed in Table 2 that report New York Heart Association (NYHA) functional class. Surgical Embolectomy 335 Table 3 Late Mortality Among Survivors of Pulmonary Embolectomy Total Number of patients 298 patients Follow-up period Mean 71 months Range 16–127 months Number of late deaths 25 deaths Overall late mortality rate 8% Linearized late mortality rate 1.4% per year Actuarial survival Medium-term (4–5 years) 65–80% Long-term (8–10 years) 62–71% Interestingly, the three Class III patients were found to have abnormalities unre- lated to their history of pulmonary embolism. Another study found a similar incidence of minor abnormalities in 10 of 12 survivors of embolectomy (84) that the authors attributed to incomplete embo- lectomy, intraoperative trauma with thrombosis, or small recurrent emboli. Two patients, both with a preoperative history of prior embolism, had major residual perfusion defects and pulmonary hypertension. They conclude that patients with underlying, chronic pulmonary thromboembolic disease have significantly worse long-term outcome. III. PREOPERATIVE EVALUATION A. Surgical Indications and Timing 1. Overall Strategy There is no consensus on the indications for surgery in acute pulmonary embolism (Table 4). The decision is fairly straightforward at either extreme. A stable patient with multiple peripheral perfusion defects is clearly not a surgical candidate. On the other hand, a patient in refractory cardiogenic shock with a documented sad- dle embolus, and a contraindication to thrombolysis, is certainly best served by emergency embolectomy. Most patients, however, fall between these two ex- tremes and require careful individualized assessment to determine the optimal management. The indications for surgery should be part of a multidisciplinary strategy that seeks to minimize overall morbidity and mortality using all available modalities. Embolectomy should be considered one tool which, if applied before severe hemodynamic compromise, can provide excellent short- and long-term results. A strategy that positions it as a treatment of last resort, to be avoided 336 Aklog Table 4 Factors to Consider in the Decision to Proceed with Surgical Pulmonary Embolectomy Contraindications to thrombolysis Possible indications for embolectomy Absolute Relative Embolus in the central pul- Active internal bleeding Age over 75 monary arteries Over 50% obstruction of Intracranial neoplasm, vas- Recent (less than 10 days) pulmonary vasculature cular malformation, or major surgery, puncture aneurysm of noncompressible puncture, or organ bi- opsy Refractory cardiogenic Neurosurgical procedure Pregnancy or early post- shock or stroke within 2 partum months Right atrial or ventricular Severe uncontrolled hyper- Recent trauma, including thrombus tension chest compressions ?Severe right heart dysfun- Known bleeding diathesis Recent GI bleeding or ac- ction tive ulcer disease (less than 10 days) ?Elevated cardiac troponin Known allergy to thrombo- Known coagulation de- lytic agents fects, including anticoag- ulant therapy and sig- nificant liver dysfunction ?Moderate hemodynamic High likelihood of left compromise or right heart thrombus (e.g., mi- heart dysfunction with tral stenosis with atrial contraindication to fibrillation) thrombolysis unless other options have failed or are unavailable, is doomed, in my opinion, to provide suboptimal results. 2. Location of Pulmonary Emboli The classic anatomical criteria for embolectomy have been over 50% obstruction of the pulmonary vascular tree by pulmonary angiography. The location of the emboli, however, may be as important as the degree of obstruction. The ideal patient has large emboli in the central pulmonary arteries (at times accompanied by right atrial thrombus) that can be completely removed in a few pieces resulting in normalization of the pulmonary artery pressures and restoration of right heart Surgical Embolectomy 337 function. A patient who has showered multiple emboli may also demonstrate vascular obstruction approaching 50% but is not as good a surgical candidate. Complete embolectomy may not be possible in these patients without traumatic maneuvers such as massaging the lungs, blind passage of instruments into the periphery, or retrograde perfusion of the lungs through the pulmonary veins. 3. Hemodynamic Status As we have seen from the results of the major surgical series, a specific center’s results will primarily depend on its hemodynamic threshold for intervening surgi- cally. If surgery is limited to those patients in extremis, one can expect mortality rates over 50%. If surgery is applied more liberally with every effort to intervene before the onset of shock or cardiac arrest, one can expect mortality rates as low as 10%, comparable to patients undergoing thrombolysis (87). This dichotomy is understandable given that the leading causes of operative death—right heart failure, neurological injury, and multisystem organ failure—are a result of the patient’s preoperative hemodynamic condition and not to the procedure itself. Pulmonary embolectomy, if applied with attention to minimizing pulmonary ar- tery trauma, is a relatively simple procedure that should carry little inherent risk of death. Cardiopulmonary bypass times should be short; cardioplegic arrest should be unnecessary; and bleeding complications should be minimal. That some patients will always present with sudden cardiovascular collapse may make it impossible to lower the overall mortality rate to less than 5 to 10% unless patients in extremis are simply not offered surgery. However, as Trende- lenburg observed nearly a century ago, most patients have between 15 min and several hours before severe hemodynamic compromise ensues. Utilizing this time period effectively is the most critical challenge, with high stakes on both sides. Delaying surgery for a trial of medical therapy may avoid a major operation but may risk converting a stable patient with an operative mortality of 5 to 10% to one in severe cardiogenic shock or cardiac arrest with a mortality of 40 to 60%. Sasahara’s classic hemodynamic criteria for embolectomy include refrac- tory cardiogenic shock (systolic pressure less than 90 mmHg) and oliguria (less than 20 mL/h) despite maximal medical therapy or when thrombolytic therapy is contraindicated (88). These stringent criteria, however, may withhold a poten- tially life-saving procedure until irreversible cardiac and end-organ injury has occurred. In my opinion, all patients with a hemodynamically significant central embolus should be considered for surgery. If the patient stabilizes with volume and low-dose inotropic support, then thrombolytic therapy may be appropriate, but the surgical team must be kept on alert and notified early if the patient deterio- rates. Invasive monitoring of cardiac output may permit intervention before end- organ injury. The use of high-dose pressors during thrombolytic therapy to avoid surgery at all costs is inappropriate. 338 Aklog 4. Right Ventricular Function Goldhaber (2) and others (89) have clearly demonstrated that right ventricular hypokinesis without systemic hypotension is a strong, independent risk factor for early mortality and that this finding can be used to triage patients to more aggres- sive treatment. In a large registry report, the 14-day mortality among the 40% of patients with right ventricular hypokinesis was double that in those without this finding (2). This risk stratification has typically been applied in selecting patients for thrombolysis but may also be used in recommending embolectomy in those patients with contraindications to thrombolysis, even if they are hemody- namically stable. Evidence of severe right ventricular dysfunction might justify proceeding with embolectomy even if thrombolytic therapy is not contraindi- cated. 5. Right Atrial or Ventricular Thrombus Mobile thrombus or ‘‘in-transit’’ emboli within the right atrium or right ventricle may occur in up to 10 to 15% of patients with pulmonary embolism (90,91). Even modest-sized thrombi can be fatal if they embolize in a patient with moder- ate degrees of vascular obstruction and right heart dysfunction. Because they are associated with a high (Ͼ40%) mortality rate, their presence is generally consid- ered a strong indication for aggressive intervention. Although thrombolysis has been advocated for this condition (90,92,93), embolectomy may be the most ap- propriate treatment. These thrombi can embolize at any time (94) and thromboly- sis may require several hours to achieve its maximal effect and might actually promote embolization. In my opinion, the presence of mobile thrombi in the right heart in a patient with documented central pulmonary emboli is a strong indica- tion for emergency surgery. 6. Cardiac Enzymes A recent report demonstrated that an elevated cardiac troponin T was present in 32% of patients with pulmonary embolism, correlated well with the severity of presentation, and was a strong predictor of in-hospital mortality—44% vs. 3% (95). On multivariate analysis, it was the only remaining independent predictor of death (adjusted odds ratio 15.2). Although these results need confirmation, cardiac troponins may supplement echocardiography in triaging patients to ag- gressive treatment. 7. Contraindications to Thrombolytic Therapy Thrombolytic therapy has been shown to accelerate the resolution of emboli, the normalization of pulmonary artery pressures and right heart function, and lower mortality in selected patients (96–98). The risk of intracranial hemorrhage is 1 Surgical Embolectomy 339 to 3% and the risk of other major hemorrhage is 10 to 20%. Some patients, however, have significantly higher risks of complications and thrombolytic ther- apy is contraindicated in them (Table 4). Patients with central pulmonary emboli who have contraindications to thrombolytic therapy should be considered for surgical embolectomy. B. Diagnostic Studies 1. Overall Strategy A streamlined diagnostic approach, focusing on documenting central emboli as rapidly as possible, is critical to achieving good results in patients being consid- ered for embolectomy. Once this has been done, the patient should be transported to the operating room immediately even if relatively stable. Echocardiography, pulmonary angiography, and CT scanning all play important roles, but the spe- cific tests performed depend on the patient’s clinical presentation, location, and what is available. Although many patients will have undergone ventilation-perfu- sion scintigraphy at some point, this study is not particularly helpful in this set- ting. Large perfusion defects may suggest central emboli but their actual location cannot reliably be determined without a chest CT scan, pulmonary angiogram, or echocardiogram that shows thrombus in the main pulmonary arteries. 2. Chest CT Scanning Contrast-enhanced spiral chest CT has become increasingly popular in evaluating pulmonary embolism (99,100). Its accuracy has been well documented in recent years and is comparable to other noninvasive tests. It may, in fact, be the ideal study in Emergency Department patients with suspected major pulmonary embo- lism. Although spiral CT is usually necessary to document peripheral emboli, large central emboli are usually well visualized with a standard chest CT scan. CT scanning is usually readily available, often in the Emergency Department itself, and can be performed in 10 to 15 min. If central emboli are visualized in the main or proximal branch pulmonary arteries, and if the patient is a candidate for surgical embolectomy by clinical criteria, no further confirmatory testing is necessary, and the patient can be immediately transported to the operating room. 3. Echocardiography Echocardiography, transthoracic and transesophageal, is also playing an increas- ing role in the diagnosis and risk stratification of patients with pulmonary embo- lism (6,7,99,101,102). Although the pulmonary arteries themselves are often not well visualized, echocardiography provides important prognostic information to help determine whether aggressive treatment should be considered. Right heart 340 Aklog function can be assessed, pulmonary artery pressures can be estimated, and mo- bile thromboemboli in the right atrium or ventricle can be visualized. Right ven- tricular hypokinesis is a strong predictor of poor outcome and may justify aggressive treatment—thrombolysis or surgery—even if the patient is hemo- dynamically stable. Echocardiography may also be helpful as an initial study in a patient with some hemodynamic compromise in whom the diagnosis of pulmonary embolism is being considered. If the echocardiogram is suggestive but not definitive, a confirmatory CT scan or, more likely, pulmonary angiogram would be indicated. A tenuous patient, with clear-cut echocardiographic findings of a major pulmo- nary embolism, who is considered a candidate for surgical embolectomy, could be taken to the operating room without further testing. Intraoperative transesopha- geal or surface echocardiography could then confirm the presence of central em- boli. Intraoperative TEE is also the only available test in the occasional patient who suffers a major pulmonary embolism on the operating room table during another operation. Evidence of a major embolism by TEE may justify immediate sternotomy and embolectomy. 4. Pulmonary Angiography Although pulmonary angiography remains the gold standard for diagnosing pul- monary embolism, its role in evaluating patients for embolectomy should be lim- ited. With spiral CT and echocardiography, most embolectomy patients will not need preoperative angiography. Clinically stable patients, with intermediate- to high-probability ventilation-perfusion scans, may undergo confirmatory angiog- raphy where unsuspected, massive central pulmonary embolism requiring surgery is discovered. Other patients, with echocardiographic findings suggestive of a large embolism, might need angiography to define the anatomical distribution of emboli and to determine whether surgical embolectomy is appropriate. IV. INTRA- AND POSTOPERATIVE MANAGEMENT A. Basic Principles Patients selected for embolectomy should be immediately transported to the op- erating room even if they ‘‘appear’’ relatively stable. In our hospital, patients go directly from the radiology suite or ward to the operating room without stopping in the intensive care unit for ‘‘stabilization.’’ Resuscitation can be performed en route and in the operating room. Patients from outside hospitals are transferred directly to the operating room by helicopter. In the operating room, teamwork is critical and only essential preoperative maneuvers are performed so the patient can be placed on cardiopulmonary bypass without delay. Surgical Embolectomy 341 B. Anesthesia Patients with large central pulmonary emboli are very tenuous, even if awake and appearing relatively clinically stable. The failing right ventricle is dependent on elevated preload to maintain flow through the obstructed pulmonary vascula- ture. The underfilled left ventricle is dependent on increased peripheral vascular tone to maintain systemic blood pressure. Although compensatory mechanisms can maintain normal blood pressure and tissue perfusion, small hemodynamic perturbations, such as venous and arterial vasodilatation from general anesthesia, can reverse these mechanisms and lead to hemodynamic collapse. This has been well appreciated in the earliest days of surgical embolectomy when femoral can- nulation for cardiopulmonary bypass under local anesthesia was quite popular. That is now rarely necessary with modern anesthetic techniques, including awake fiberoptic intubation and the use of new agents with minimal hemodynamic ef- fects. Nonetheless, anesthesia should not begin until the perfusionist is present with the heart–lung machine primed and ready. Prepping and draping the patient before induction may also be prudent as this will allow the surgeon to open the chest and be on bypass within 60 to 90 s. Although an arterial monitoring line and large-bore venous access are nec- essary prior to induction, central venous access can be deferred until after the patient is on bypass. A pulmonary artery catheter is critical for weaning from bypass and postoperative management but the dilated right heart can make ad- vancing the catheter difficult and time-consuming. The introducer sheath can be inserted and the catheter advanced with the surgeon’s guidance after embolec- tomy and prior to weaning from bypass. C. Transesophageal Echocardiography Intraoperative transesophageal echocardiography (TEE) is a useful tool during embolectomy. Although placement of the probe should not be a high priority in an unstable patient, imaging of the right atrium and ventricle prior to cannulation for bypass is helpful. Visualization of mobile thromboemboli in the right heart can help determine whether right heart exploration is necessary and guide venous cannulation. Assessment of right heart function can help anticipate difficulties with weaning from bypass. TEE can often visualize emboli in the central pulmo- nary arteries in a patient who has come to the operating room without good visualization of these structures. Surface scanning of the right heart and pulmo- nary arteries is very simple and can be used in the occasional patient who cannot undergo TEE or in whom adequate images cannot be obtained (103). Persistent right heart dysfunction can make postbypass hemodynamic management difficult. TEE can assess the degree of residual right heart dysfunction as well as ventricu- 342 Aklog lar volumes to help determine whether to treat the patient with volume, inotropic agents, pulmonary vasodilators, or systemic vasoconstrictors. D. Surgical Technique 1. Incision and Cardiopulmonary Bypass Embolectomy is performed through a full median sternotomy that provides good exposure of the pulmonary arteries, vena cavae, and right heart chambers. After pericardiotomy and full heparinization, the patient is cannulated for cardiopulmo- nary bypass. Arterial cannulation is usually performed in the distal ascending aorta. The most common technique for venous cannulation is to place separate cannulae in the superior and inferior vena cavae with tourniquets to prevent air from entering the venous line when the pulmonary artery or right heart is opened. The availability of intraoperative TEE has allowed us to simplify our venous cannulation technique. We explore the right atrium and ventricle only if the TEE (or surface echo) demonstrates mobile thromboemboli. In addition, we routinely employ vacuum-assisted venous drainage that is not affected by moderate amounts of air in the venous line. This allows us to rapidly cannulate the right atrium through the appendage using a standard two-stage venous cannula. This technique works even in situations where right heart exploration is necessary. With vacuum-assisted venous drainage, the right atrium can be explored without separately controlling the vena cavae. Although some authors report using aortic cross-clamping and cardioplegic or fibrillatory arrest, this is, in my opinion, never necessary and always potentially detrimental. Pulmonary embolectomy and right heart exploration can easily be performed with the heart beating under normothermic conditions. Some degree of right heart stunning is almost always present and aortic cross-clamping and cardiac arrest add an additional ischemic insult that risks further stunning and right heart dysfunction. Performing the operation on the unloaded, well-perfused beating heart not only avoids any further ischemic insult but also provides time for the heart to recover and regenerate its depleted energy stores. 2. Exploration of the Right Atrium and Ventricle If echocardiography demonstrates mobile thromboemboli in the right atrium or right ventricle, exploration of these chambers is mandatory and should be per- formed prior to embolectomy. Cannulating the right atrium and inferior vena cava in patients with right atrial thrombus must be performed with caution to prevent dislodging the thrombus or aspirating it into the venous line. In these situations we cannulate the right atrium while visualizing it and the thrombus by TEE. The cannula is advanced just far enough to achieve adequate venous drain- [...]... in acute MI, 113 reduced-dose fibrinolysis plus IIb/ IIIa inhibition, 114–115 types of GP IIb/IIIa inhibitors, 102 IIb/IIIa inhibition during PCI angioplasty, 102 106 abciximab, 102 103 eptifibatide, 103 105 primary PCI, 105 106 tirofiban, 105 106 IIb/IIIa inhibition in unstable angina and non-ST-elevation MI, 106 109 abciximab, 108 109 eptifibatide, 108 lamifiban, 109 tirofiban, 107 108 use of LMWH combined... Acta Chir Belg 1986; 86:118–122 Index Abciximab, 102 use in PCI angioplasty, 102 103 use in unstable angina and non-STelevation MI, 108 109 Acquired risk factors for venous thromboembolism (VTE), 266 Acute arterial occlusion of the lower extremity, 170 Acute atherothrombosis, inflammation and, 3–6 Acute coronary syndromes, hs-CRP for risk assessment in, 10 12 Acute infarction rampril efficacy (AIRE) study,... postmenopausal hormone therapy and, 68–74 association of hormones with lower risk of CHD, 73–74 combined hormone therapy, 71, 72–73 duration of hormone use, 70 primary prevention, 68–70 secondary prevention, 71–72 C-reactive protein (CRP), 6–7 hs-CRP and prevalent coronary heart disease, 7 hs-CRP and risk of future cardiovascular events, 7 10 hs-CRP and therapeutic interventions, 16–18 hs-CRP as inflammatory markers... FRISC-1 (controlled trial), 81, 82–84 FRISC-II (controlled trial), 81, 86 Genetic causes of elevated homocysteine levels, 40 Genetic risk factors for VTE, 266 Gibrafiban, 102 Glycoprotein (GP) IIb/IIIa inhibitors, 101 –124 angiographic observations, 110 111 mechanism of action, 101 102 oral IIb/IIIa inhibitors, 115–119 current status, 118–119 EXCITE trial, 117 OPUS-TIMI-16 trial, 116–117 Symphony I and. .. patients, 256 venous thromboembolism and, 216– 217 Index CAPRIE trial (clipidogrel compared to aspirin), 137, 138, 140 Cardiopulmonary bypass embolectomy and, 325, 344 versus inflow occlusion, 328 Cardiovascular disease: increased homocysteine levels and risk of, 41–42 morbidity and mortality, hs-CRP and risk of, 7 10 risk assessment algorithm using hsCRP and lipid screening, 15, 16 Catheter-directed lysis... contraceptives and, 218 venous thromboembolism and, 215– 216 Ticlopidine, 126, 132–136, 173 adverse effects of, 136 cardiovascular disease and, 134–135 coronary artery disease and, 133–134 peripheral arterial disease and, 135– 136 Ticlopidine aspirin stroke study (TASS), 134, 136 TIMI-11B (controlled trial), 81, 85–86 Tinzaparin sodium injection (Innohep), 93 Tirofiban, 102 , 175 use in PCI angioplasty, 105 use... Cardiovasc Surg 1982; 30 :103 108 Lund O, Nielsen TT, Ronne K, Schifter S Pulmonary embolism: long-term followup after treatment with full-dose heparin, streptokinase or embolectomy Acta Med Scand 1987; 221:61–71 Heit JA, Mohr DN, Silverstein MD, Petterson TM, O’Fallon WM, Melton LJ, III Predictors of recurrence after deep vein thrombosis and pulmonary embolism: a population-based cohort study Arch... 238–239 transthoracic, for PE patients, 249– 250 Elderly, increased homocysteine levels and mortality risk for, 42–43 Enzyme deficiencies and mutations, elevated homocysteine levels associated with, 37 EPILOG trial, 176–177 Eptifibatide, 102 efficacy in PCI angioplasty, 103 105 efficacy in unstable angina and nonST-elevation MI, 108 ESSENCE (controlled trial), 81, 85 Estrogen receptor modulators, risk of VTE... 181–182 low-molecular-weight heparin, 181 prevention of restenosis after PCI with, 182–183 unfractionated heparin, 179–181 Aortoiliac obstruction, 191–192 Argatroban (Novastan), 92 Arterial thromboembolism (ATE), APLA and, 51–55 background and historical aspects, 51 diagnostic pathway, 54 evidence-based treatment recommendations, 51–53 unanswered questions, 54–55 Index Arterial thrombosis and inflammation,... 6–21 high-sensitivity C-reactive protein, atherosclerosis, and atherothrombosis, 6–12 inflammatory cytokines, 12–13 inflammatory markers in clinical practice, 14–21 pathophysiological correlation, 13– 14 laboratory and pathologic evidence, 1–6 inflammation and acute atherothrombosis, 3–6 inflammation in atherogenesis, 1–3 Aspirin, 126–133, 173 adverse effects of, 131–132 cardiovascular disease and, 128–130 . reflecting the excellent mid- and long-term sur- vival in operative survivors (Fig. 4). Medium-term (4–5 year) survival was 65 to 80% (65,66,71,85) and long-term (8 10 year) survival was 62 to. trauma and protecting the right heart. An operative mortality of 10% or less and excellent long-term outcomes can be expected if the procedure is per- formed, prior to cardiovascular collapse, as part. Wilson RE, and Rudolf LE. A new look at pulmonary embolectomy. Surg Gynecol Obstet 1958; 107 :214–220. 14. Romaine-Davis A. John Gibbon and his heart-lung machine. Philadelphia: Univer- sity of