24 Pulmonary Thromboendarterectomy for Chronic Thromboembolic Pulmonary Hypertension
Pulmonary arterial hypertension (PAH) resulting from chronic pulmonary embolism (PE) is a major cause of morbidity and mortality worldwide. It is much more common than generally appreciated and is certainly underdiagnosed. The incidence is difficult to estimate because of the uncertainty regarding the frequency of acute PE, as well as the percentage of patients in whom embolic residua fail to resolve. Chronic thromboembolic pulmonary hypertension (CTEPH) is a common variation of pulmonary hypertension. There are an estimated 2500 new cases in the United States,1 or a calculated prevalence of about 3 of 100 cases of PE (20 per 1 million). Pulmonary thromboendarterectomy (PTE) is the curative procedure for CTEPH. Medical management provides only limited and temporary relief of symptoms. Lung transplantation is the only other therapy for patients with pulmonary hypertension and in some centers is still the surgical therapy of choice for patients with thromboembolic disease. Transplant surgery, however, is not a reasonable choice for CTEPH patients when taking into account the risk for death on the waiting list, shortage of organ supply, expense, the risk of immunosuppressive agents, risk for infection, and rejection. The mortality rate for lung transplant has been reported to be 20% including the waiting period versus 4% for PTE at the University of California San Diego (UCSD) Medical Center. Therefore, PTE remains the only permanent curative surgical therapy for patients with CTEPH. PTE historically is how the procedure has been known; however, other terms for the procedure that have been used interchangeably are pulmonary artery endarterectomy (PAE) or pulmonary endarterectomy (PEA). The success of the operation revolves around endarterectomy of the organized fibrous thrombus in the intimal and part of the medial layers of the pulmonary vascular tree. CTEPH results from obstruction of the pulmonary arteries by nonresolving thromboemboli causing increased pulmonary vascular resistance (PVR), right-heart failure, and death if left untreated. Distinguishing CTEPH from other forms of pulmonary hypertension is imperative because CTEPH is potentially curable with PTE. This chapter focuses on information about the clinical history, diagnostic workup, surgical approach, anesthesia, and postoperative care for patients with CTEPH undergoing PTE surgery. Acute PE and PAH also are discussed in this chapter, concluding with three cases demonstrating the application of intraoperative transesophageal echocardiography (TEE) in patients with CTEPH.
Epidemiology and pathophysiology
Acute or recurrent PEs are thought to be the inciting event in the development of CTEPH. Incomplete resolution of the emboli followed by thrombus organization and fibrosis lead to partial or complete vessel obstruction. In addition, vascular remodeling in the distal pulmonary arteries (pulmonary arteriopathy) also may contribute to the increased PVR seen in CTEPH1–3 and is the cause of residual pulmonary hypertension seen in some patients after otherwise successful PTE. Unresolved PEs in the proximal pulmonary arterial tree cause vascular obstruction by two ways: canalization of the clot leading to multiple small endothelized channels separated by bands and webs or fibrin clot organization; the other way is absent canalization leading to dense fibrous connective tissue completely occluding the arterial lumen.4–6 This fibrous plug is firm and adherent to the arterial wall, and the challenge is to remove enough of the fibrous plug as one unit to reduce the vascular resistance without disrupting the arterial wall.
Generally, most patients with PE have complete resolution of the thromboembolic event. However, there is a small portion of patients in whom embolic resolution is incomplete, resulting in the development of CTEPH. The mechanism by which thromboembolic material remains unresolved and becomes incorporated in the pulmonary arterial wall is not exactly known. A variety of factors may play a role: The volume of the embolic substance may simply overwhelm the lytic system or total occlusion of a major arterial branch, preventing the lytic material from reaching and dissolving the embolus completely. The emboli may be made of substances such as well-organized fibrous thrombus, fat, or tumor that cannot be dissolved by normal mechanisms. Some patients may have tendencies for thrombus formation, a hypercoagulable state, or abnormal lytic mechanism. Rosenhek and collegues7 showed that subjects under normal physiologic conditions have greater levels of tissue plasminogen activator versus plasminogen activator inhibitor 1 expression in the PA compared with the aorta, leading to improved natural fibrinolysis. However, Olman et al8 and Lang et al9 were unable to demonstrate a reversal in the tissue plasminogen activator/plasminogen activator inhibitor 1 relation favoring incomplete thrombus resolution in patients with history of CTEPH.
CTEPH is a common, but under-recognized, cause of pulmonary hypertension, in which the actual incidence of CTEPH remains somewhat uncertain. Pengo et al10 observed 223 patients for a median of 94.3 months after an acute PE event in which 3.8% of patients experienced development of symptomatic CTEPH. Ribeiro et al11 examined echocardiograms in 278 patients surviving acute PE for 1 year and performed clinical follow-up for 5 years. Five percent of these patients experienced development of clinically significant CTEPH. Dentali et al12 noted similar findings in 91 patients examined at 6 months after acute PE. Pulmonary hypertension associated with residual perfusion defects were identified in eight patients (8.8%), four of whom were symptomatic. Related factors that predispose patients to development of CTEPH remain unclear. It can be argued that unresolved thrombi with recurrent asymptomatic PE may be the reason for the development of clinically significant CTEPH.13 However, another possibility is based on clinical observation of patients with a history of PE who received anticoagulant therapy and still developed CTEPH. Pengo et al’s10 study identified a younger age at presentation, larger perfusion defects at diagnosis, idiopathic thromboembolic disease, and a history of multiple PE events as risk factors for development of CTEPH after an acute PE.
Bonderman et al14 identified the following risk factors for CTEPH in a controlled retrospective cohort study: ventriculoatrial shunts, infected pacemakers, splenectomy, previous venous thromboembolism (VTE), recurrent VTE, blood group other than O, lupus anticoagulant/antiphospholipid antibodies, thyroid replacement therapy, and a history of malignancy. Despite being a risk factor for VTE, the prevalence of hereditary thrombophilic states (deficiencies of antithrombin III, protein C and protein S, and factor II and factor V Leiden mutations) is similar to that in normal control subjects or in patients with idiopathic pulmonary hypertension.15 In contrast, lupus anticoagulant/antiphospholipid antibodies can be found in up to 21% of CTEPH patients,15 and Bonderman et al16 demonstrated increased levels of factor VIII in 41% of CTEPH patients. Finally, small, preliminary studies suggest that there may be structural and functional abnormalities of fibrinogen in CTEPH patients, perhaps conferring resistance to fibrinolysis.17,18 Morris et al17 reported a relative resistance of fibrin to plasmin-mediated lysis caused by an alteration in fibrin(ogen) structure affecting accessibility to plasmin cleavage sites in patients with CTEPH.
Clinical manifestations
The initial symptoms of CTEPH typically are progressive exertional dyspnea and exercise intolerance, nonspecific symptoms that often are attributed to more commonly occurring medical conditions such as obstructive lung disease, cardiac disease, obesity, or deconditioning. These symptoms are due to an increased PVR limiting cardiac output (CO) and increased minute ventilation requirements because of increased alveolar dead space.1,3 Many patients have no history of a documented acute VTE, making the diagnosis even more challenging.14,19,20 As the disease progresses and the right heart fails, patients may experience development of ascites, early satiety, epigastric or right upper quadrant fullness, edema, chest pain, and presyncope/syncope. Other symptoms reported include nonproductive cough, hemoptysis, and palpitations. Left vocal cord dysfunction and hoarseness may arise from compression of the left recurrent laryngeal nerve between the aorta and an enlarged left main PA. Early in the disease process, the physical examination may be normal or may reveal an accentuated pulmonic component of the second heart sound, but this can be easily overlooked.1,3 Pulmonary flow murmurs are bruits heard over the lung fields and are caused by turbulent blood flow through partially occluded or recanalized thrombi. These flow murmurs are heard in 30% of CTEPH patients and are not found in idiopathic pulmonary hypertension.21 Late in the disease process when right ventricular (RV) function fails to meet normal resting metabolic needs, patients start to experience exertion-related syncope and resting dyspnea. The nonspecific nature of these complaints is indisputably the reason that most patients with chronic thromboembolic events experience a delay in diagnosis or are improperly labeled with a different diagnosis. Furthermore, the fact that some patients do not report a history of PE or deep venous thrombosis (DVT) makes proper diagnosis challenging. The physical signs are far from uniform, and the physical examination may be surprisingly unrewarding if right-heart failure did not ensue, even in the presence of severe dyspnea. Physical findings of RV failure such as jugular venous distention, RV lift, fixed splitting of the second heart sound, murmur of tricuspid regurgitation (TR), RV gallop, hepatomegaly, ascites, and edema may appear with later stages of the disease.1,3 Cyanosis, if present, may suggest right-to-left shunting through a patent foramen ovale (PFO) in patients with pulmonary hypertension. Pulmonary function tests often demonstrate minimal changes in lung volume and ventilation. Diffusing capacity often is reduced and may be the only abnormality on pulmonary function testing. Pulmonary arterial pressures are increased and are not infrequently suprasystemic. Resting COs and indices are lower than the reference range, and pulmonary arterial oxygen saturations are reduced. The majority of patients are hypoxic, with room-air arterial oxygen tension ranging between 50 and 83 mm Hg22; CO2 tension is slightly reduced and is compensated for by reduced bicarbonate. Dead space ventilation is increased, together with moderate ventilation/perfusion mismatch, which correlates poorly with the degree of pulmonary vascular obstruction.23
Diagnostic evaluation
Once the diagnosis of pulmonary vascular disease is entertained, the patient evaluation involves three goals: to establish the presence and severity of PAH, to determine the cause of the PAH, and if thromboembolic disease is present, to determine whether it is amenable to surgical correction. Standard diagnostic tests performed in the evaluation of dyspnea may be suggestive of pulmonary vascular disease.1 Chest radiography may be normal in the early stages of disease, but with worsening PAH, dilation of the central pulmonary arteries and enlargement of the right atrium and ventricle can be seen (Figure 24-1). CTEPH patients also may have irregularly shaped and asymmetrically enlarged pulmonary vessels.24 Electrocardiographic (ECG) findings such as right-axis deviation, right ventricular hypertrophy (RVH), right atrial enlargement, right bundle branch block, ST-segment abnormalities, and T-wave inversions in the precordial and inferior leads may be seen. Pulmonary function testing frequently is performed in the evaluation of dyspnea and is useful in excluding coexisting parenchymal lung disease. Normal lung volumes or mild to moderate restrictive defects because of parenchymal scarring from pulmonary infarction may be seen in CTEPH.25 A mild-to-moderate reduction in single-breath diffusing capacity of carbon monoxide also may be present in CTEPH, but a severe reduction should stimulate further evaluation for an alternative pulmonary process disrupting the distal pulmonary vascular bed.26 CTEPH patients may demonstrate a normal Pao2 at rest but exhibit a decline with exertion. Hypoxemia at rest occurs in the setting of severe RV dysfunction or a right-to-left shunt, such as a PFO.3
Figure 24-1 Chest radiograph demonstrating cardiomegaly and enlargement of the right pulmonary artery.
Transthoracic echocardiography (TTE) frequently is the first study to suggest the presence of pulmonary hypertension. Depending on the severity of the disease, echocardiography may demonstrate right-heart chamber enlargement, abnormal RV function, TR, leftward displacement of the interventricular septum (IVS), decreased left ventricular size, and abnormal ventricular systolic and diastolic function.27 PA systolic pressures can be estimated by Doppler evaluation of the tricuspid regurgitant envelope. Contrast injection may demonstrate a right-to-left shunt as a result of the increased right atrial pressures and a PFO. The echocardiogram is also useful in excluding left ventricular dysfunction, valvular disease, or congenital heart disease, which may cause pulmonary hypertension (see Chapters 1 and 2).
Radioisotopic ventilation-perfusion scanning is essential in distinguishing large-vessel occlusive disease from small-vessel pulmonary vascular disease (Figure 24-2). Patients with CTEPH invariably have one or more segmental or larger mismatched perfusion defects. This is in contrast to patients with idiopathic pulmonary arterial hypertension (IPAH) or other forms of small-vessel PAH in which the perfusion scans are normal or demonstrate subsegmental or mottled perfusion abnormalities.28,29 Notably, the magnitude of perfusion defects often underestimates the degree of vascular obstruction in CTEPH. This is due to organization and recanalization of clot resulting in partial obstruction of pulmonary arteries that allows limited passage of radiolabeled aggregated albumin resulting in “gray zones” in areas of hypoperfused lung on perfusion scan.30 Hence, even a single, mismatched, segmental perfusion abnormality should raise the question of CTEPH in a patient with pulmonary hypertension. Conversely, unmatched perfusion defects are not specific for thromboembolic disease because similar defects can be seen with pulmonary vascular tumors, large-vessel arteritis, extrinsic compression, and pulmonary veno-occlusive disease.31,32 Consequently, further imaging is required. Contrast-enhanced chest computed tomography (CT) imaging is assuming an increasingly important role in the evaluation of thromboembolic disease. Although a negative CT scan does not rule out the diagnosis of CTEPH, when used in conjunction with ventilation/perfusion scanning, it may aid in making the diagnosis and determining operability in patients with CTEPH. CT findings in CTEPH include chronic thromboembolic material in the central pulmonary arteries, mosaic perfusion of the lung parenchyma, central PA enlargement, variability in the size and distribution of pulmonary arteries, right atrial and ventricular enlargement, and increased bronchial collateral arteries33 (Figure 24-3). Chest CT scanning also is valuable in assessing the lung parenchyma in patients with coexistent emphysematous or fibrotic lung disease, as well as assessing patients who suffer from other diseases that can obstruct central pulmonary arteries and mimic CTEPH, such as mediastinal fibrosis, tumors, lymphadenopathy, and arteritis.1,3,34 Finally, the presence of lining thrombus in the central pulmonary arteries has been described in IPAH (previously referred to as primary pulmonary hypertension) and other end-stage lung disease.35,36 The radioisotope perfusion scan is helpful in distinguishing such patients from CTEPH, which is essential because PTE carries substantial risk in these patients and is unlikely to result in hemodynamic benefit.1,3 More recently, there is an increasing interest in the use of helical CT scanning,37 single-photon emission CT-CT fusion imaging,38 and magnetic resonance angiography39 in the evaluation of CTEPH40 patients, but further experience and research are needed.
Figure 24-2 Radioisotopic ventilation and perfusion scan.
The top row demonstrates ventilation wash-in, equilibrium, and wash-out of Xenon in the posterior view. There is some decreased ventilation of the right lower lobe corresponding to the elevated right hemidiaphragm in Figure 24-1. Perfusion images in multiple views are seen in the bottom row and demonstrate markedly diminished flow to the left lower lobe, lingula, right middle lobe, and right lower lobe.
Figure 24-3 A, Computed tomographic angiogram of the same patient in Figure 24-2. Lining thrombus can be seen in the right interlobar artery (thick arrow), and a web is noted in the left descending pulmonary artery (thin arrow). B, Right atrial and ventricular enlargement in the same patient.
Pulmonary angiography remains the gold standard in the evaluation of CTEPH and can be performed safely by experienced individuals, even in patients with severe hemodynamic impairment.41 The angiographic appearance of chronic thromboembolic disease is distinctly different from that of acute PE because of the organization and recanalization that take place during partial embolic resolution. Characteristic angiographic findings in CTEPH include vascular webs or bandlike narrowings, intimal irregularities, pouch defects, abrupt narrowing of vessels, and proximal obstruction of pulmonary arteries5 (Figure 24-4). Biplane imaging is optimal in providing anatomic detail because lateral images provide better detail of lobar and segmental vessels that overlap on anteroposterior images.1
Cardiac catheterization may be performed at the time of pulmonary angiography. Right-heart catheterization defines the severity of the PAH and degree of cardiac dysfunction. Typically, the right atrial (RA), RV, PA, and pulmonary artery occlusion pressures (PAOP); CO; and mixed venous oxygen saturation (Svo2) are measured. These hemodynamics are helpful in assessing risk of surgical intervention. Left-heart catheterization and coronary arteriography also are performed in patients at risk for coronary artery disease or in suspected left-heart dysfunction or valvular heart disease.3 Ascertaining the differential diagnosis between PAH and distal and small vessel with thromboembolic disease remains a challenge in about 10% to 15% of cases. Usually, these patients will not have a clear history of thromboembolism. Pulmonary angioscopy often is helpful in such cases. The pulmonary angioscope is a fiberoptic bronchoscope 120 cm in length and 3 mm in external diameter. The distal tip can flex and extend, and it has a balloon that is then filled with CO2 and pushed against the vessel wall, creating a bloodless field to better visualize the PA wall. This is easily accomplished through placement via a central venous catheter to gain access to the PA. The normal appearance of the vessel wall is of pink, white, smooth, and glistening material, whereas the classic appearance of chronic pulmonary thromboembolic disease by angioscopy consists of intimal thickening, irregularity, scarring, pitting, and the presence of pouches, bands, and webs across small vessels. These features are due to the organization of PE with recanalization and subsequent fibrosis. The bands and webs are thought to be the residue of the resolved thrombi occluding small and medium vessels and are important diagnostic findings.
Operability
Pulmonary endarterectomy offers the potential for cure and results in substantial improvement in hemodynamics, functional status, and long-term survival.42–45 Assessment of surgical candidacy should be performed at centers with expertise in the diagnosis and management of CTEPH. To be considered surgical candidates, patients must have surgically accessible chronic thromboembolic disease (Box 24-1). The definition of “surgically accessible” will vary depending on the experience and skill of the surgeon. In addition to the location of chronic thrombi, the extent of vessel obstruction and its correlation to hemodynamic compromise are important in determining candidacy for surgery. Most patients presenting for this operation are within New York Heart Association Class III or IV. Age ranges from 7 to 85 years. Many patients (20%) have a PVR greater than 1000 dynes.sec.cm−5, and it is not uncommon to have suprasystemic pulmonary artery pressures (PAPs).
BOX 24-1 Patient Selection Criteria for Pulmonary Thromboendarterectomy
The increase in PVR arises from not only central surgically accessible lesions but also distal, small-vessel arteriopathy. Patients with a significant component of small-vessel arteriopathy may not experience a significant decrease in PVR after PTE, leaving them at increased risk for short-and long-term consequences. Patients with a postoperative PVR greater than 500 dynes.sec.cm−5 have a perioperative mortality rate of 30%, compared with 1% in those with a postoperative PVR less than 500 dynes.sec.cm−5.19
Several techniques are being studied to attempt to partition the upstream (thromboembolic disease) from downstream (arteriopathy) resistance, but they remain investigational.46,47 Most patients who undergo PTE typically have PVRs greater than 300 dynes.sec.cm−5, and most are in the range of 700 to 1100 dynes.sec.cm−5. Surgery may be considered for patients with normal resting hemodynamics if they have significant obstruction of one main PA, those with a vigorous lifestyle who exhibit significant pulmonary hypertension with exercise, and those who are symptomatic from increased dead space ventilation.3 Patients with preoperative PVRs greater than 1000 dynes.sec.cm−5 have been shown to have a greater operative mortality rate,19 but a markedly increased preoperative PVR does not exclude the patient from being a surgical candidate. Comorbid diseases also must be assessed as part of the preoperative evaluation. Severe underlying parenchymal diseases, particularly involving regions of the lung that would be reperfused by PTE, are a contraindication to surgery.
Medical treatment of chronic thromboembolic pulmonary hypertension
Untreated, CTEPH has a poor prognosis, with more than half of patients with mean pulmonary artery pressure (mPAP) of more than 50 mm Hg not surviving beyond 1 year after diagnosis.48 Surgery is associated with increased survival rate, when significant reductions in pulmonary hemodynamics and PVR are achieved.19 Despite the advances achieved by PTE, up to 50% of patients are judged inoperable and more than 10% of them experience persistent or recurrent PAH after PTE. Medical therapy may be beneficial for select patients with CTEPH. There are four different groups in whom medical therapy can be beneficial (Box 24-2): patients in whom comorbidities are so significant that surgery is contraindicated or because of personal choice they opt to not proceed with PTE surgery; and patients with severe PAH and RV failure who exhibit distal limited resectable chronic thrombus and are unlikely to benefit from PTE surgery. For those patients, medical therapy may be the only possibility, with lung transplant reserved for the most severely ill individuals. Another group is high-risk patients with extremely poor hemodynamics, which includes those with functional Class IV symptoms, mPAP greater than 50 mm Hg, cardiac index less than 2.0 L/min/m2, and PVR greater than 1000 dynes.sec.cm−5. For these patients, intravenous epoprostenol may be considered as a therapeutic bridge to PTE. Although this approach may improve surgical success, medical therapy should not significantly delay PTE. Patients with residual pathology after PTE (about 10%) should be considered for medical therapy.
BOX 24-2 Groups to Consider for Medical Treatment
Vasodilator therapy in select patients with CTEPH deserves further evaluation. Ono et al49 evaluated the use of an oral prostacyclin analog, “beraprost,” in a small number of patients with inoperable CTEPH together with conventional therapy compared with a matched group undergoing conventional therapy alone. Fifty percent of patients who received beraprost experienced functional improvement compared with no functional improvement in the group who received the conventional therapy alone. During the same time, modest declines in the mPAP and total pulmonary resistance were observed and, throughout follow-up, improved survivorship was seen in patients treated with beraprost. Bonderman et al50 evaluated the use of bosentan, an endothelin-receptor antagonist, in 16 patients with inoperable CTEPH, showing an improvement in New York Heart Association functional status in 11 patients during a period of 6 months, a reduction in pro-brain natriuretic peptide levels, and an improvement in 6-minute walk (6MW) distance. The greatest results were observed in CTEPH patients who had undergone endarterectomy surgery. Sildenafil,51 a phosphodiesterase-5 (PDE-5) inhibitor, has dual effects, increasing RV inotropy and decreasing RV afterload, and it may be more advantageous than drugs that affect only the PA. An expert consensus document released in 2009 recommended the use of either sildenafil or endothelin-receptor antagonists as first-line therapy in patients with PAH, with functional Class II or early Class III disease.52 Alternative medical therapy includes prostacyclin analogs (e.g., iloprost, Flolan, Remodulin).49,53 It is recommended that patients on a prostacyclin analog continue their medications before surgery and not discontinue them abruptly, because these medications can lead to severe rebound hypertension, resulting in catastrophic events.54 Regarding patients presenting for surgery on epoprostenol or teprostenol infusions, it is recommended to discontinue the infusion during the prebypass period, and depending on the surgical results, it is either continued after bypass or after surgery in the intensive care unit (ICU). Further details of the pharmacotherapy for PAH are discussed in the Pulmonary Arterial Hypertension section later in this chapter.
The use of pulmonary vasodilator therapy as a bridge to PTE and its effects on postoperative outcome remain unknown and it should not delay surgical intervention. This is a mechanical condition that cannot be effectively treated by pharmacotherapy or angioplasty; it can only be removed by open operation. Figure 24-5 depicts the surgical specimen obtained at PTE from the patient depicted in the previous figures.
Operation
Historical Development
Chronic thromboembolic disease was not recognized as a distinct diagnostic entity until the late 1920s and has remained severely underdiagnosed since then. The first surgical attempt to remove the adherent thrombus from the pulmonary arterial wall was reported by Hurwitt et al in 1958.55 This was the landmark operation that provided the distinction between an endarterectomy rather than an embolectomy as the surgical procedure of choice for chronic thromboembolic disease. In 1961 and 1962, systemic hypothermia and CPB standby were used to perform two successful endarterectomies. A historical review of the world’s experience for PTE up to 1985 published by Chitwood et al56 revealed an overall perioperative mortality rate of 22% in 85 patients who underwent surgical endarterectomy. Moser et al23 published their experience of 42 patients with CTEPH who underwent PTE, with an in-hospital mortality rate of 16.6%. Of the 35 survivors (mean follow-up, 28 months), 16 had NYHA Class I disease, 18 Class II, and 1 Class III. PVR declined significantly after surgery from 897 ± 352 to 278 ± 135 dynes.sec.cm−5. This study confirmed the substantial improvements and functional ability experienced by these patients one year after surgery.23 Nina Braunwald commenced the UCSD experience with this operation in 1970, which now totals more than 2400 cases. The first patient was a 67-year-old man who underwent the operation on July 14, 1970. Through a right lateral thoracotomy, a PTE was performed using CPB. The patient was discharged from the hospital and returned to full activity. A total of seven surgeons have been involved with the program since the beginning. Drs. Braunwald, Utley, Daily, and Dembitsky, who together performed 188 procedures between 1970 and 1989, made progressive modifications to the surgical technique, including the use of a median sternotomy and hypothermic circulatory arrest. Drs. Jamieson and Kapelanski performed more than 1400 cases from 1989 to early 2000. Recently, Dr. Jamieson, together with Dr. Michael Madani, has expanded the total number of cases to more than 2400. UCSD is the world’s pioneer in PTE surgery, and most of the surgical experience in PTE to date has been reported from there; hence this section is based on their extensive experience (Figure 24-6).
Pulmonary Endarterectomy Procedure
When CPB is initiated, surface cooling with both the head jacket and the cooling blanket is begun. The blood is cooled with the pump-oxygenator. During cooling, a 10° C gradient between arterial blood and bladder or rectal temperature is maintained.57 Cooling generally takes 45 minutes to an hour. When ventricular fibrillation occurs, an additional vent is placed in the left atrium through the right superior pulmonary vein. This prevents atrial and ventricular distention from the large amount of bronchial arterial blood flow that is common with these patients.
There are four broad types of pulmonary occlusive disease related to thrombus that can be appreciated, and the authors use the following classification58,59 (Box 24-3):
BOX 24-3 Jamieson’s Classification
A modified cerebellar retractor is placed between the aorta and superior vena cava. When blood obscures direct vision of the pulmonary vascular bed, thiopental is administered (500 mg to 1 g) until the electroencephalogram becomes isoelectric. Circulatory arrest then is initiated, and the patient undergoes exsanguination. All monitoring catheters to the patient are turned off to prevent the aspiration of air. Snares are tightened around the cannulae in the superior and inferior vena cavae. It is rare that one 20-minute period for each side is exceeded. Although retrograde cerebral perfusion has been advocated for total circulatory arrest in other procedures, it is not helpful in this operation because it does not allow a completely bloodless field, and with the short arrest times that can be achieved with experience, it is not necessary (see Chapter 21).
Anesthetic management for patients undergoing pulmonary thromboendarterectomy
Preoperative Preparation
Patients are admitted before surgery for a full workup. Right-heart catheterization demonstrates the severity of pulmonary hypertension, typically with a PVR greater than 300 dynes.sec.cm−5 or more and mPAPs greater than 25 mm Hg. With disease awareness and more recognition of CTEPH surgical candidates, more advanced cases are presenting with PVRs well above 1000 dynes.sec.cm−5 and suprasystemic pulmonary hypertension. Hartz et al60 found that a preoperative PVR greater than 1100 dynes.sec.cm−5 and an mPAP greater than 50 mm Hg predicted a higher operative mortality. In a recent report of 500 surgical patients by Jamieson et al,19 a postoperative mortality rate of 10% was associated with a preoperative PVR above 1000 dynes.sec.cm−5, compared with 1.3% in patients with a preoperative PVR less than 1000 dynes.sec.cm−5. The preoperative right-heart catheterization focuses on specific data: RA pressures, RV pressures, with RV diastolic pressures more than 14 mm Hg, suggesting RV failure, PAPs, and PVR. CO and index also provide an insight to RV dysfunction. Before surgery, all PTE patients undergo inferior vena caval filter placement to prevent future PE after the endarterectomy. Transthoracic echocardiogram is a valuable diagnostic tool for PAH to exclude primary pathology of the LV, valvular disease, or intracardiac shunting as the cause for the clinical presentation. Preoperative TTE evaluates right heart chamber enlargement and function, paradoxical interventricular septal motion, and intracardiac shunting, and identifies any thrombus in the RA or pulmonary arteries. All these data are critical to identify before proceeding with induction.
Hemodynamic Consideration and Induction
The majority of patients with CTEPH presenting for PTE are without left ventricular pathology. Induction and decision making are, thus, centered on RV function. The right ventricle typically is hypertrophied and dilated, associated with a dilated right atrium. PTE patients have a relatively fixed PVR and concomitant RV dysfunction; therefore, any significant decrease in mean arterial pressure during induction may compromise RV perfusion, causing cardiovascular collapse and death. Maintenance of adequate SVR, adequate inotropic state, and normal sinus rhythm serve to preserve systemic hemodynamics, as well as RV coronary perfusion. Attempts to reduce PVR pharmacologically using nitroglycerin or nitroprusside should be avoided because they have minimal efficacy in treating the relatively fixed PVR and result in SVR reduction that compromises RV coronary perfusion and RV function, rapidly leading to hypotension and cardiovascular collapse. Hence administration of vasopressors, such as phenylephrine or vasopressin, is vital to maintain SVR and ensure adequate RV perfusion. Despite a relatively fixed PVR, attempts should be made to minimize conditions that increase PVR further such as avoiding episodes of hypoxia, hypercarbia, and acidosis, and any changes should be treated aggressively. The choice of anesthetic induction drugs depends on the degree of hemodynamic instability. Etomidate frequently is used because it maintains sympathetic tone and does not possess significant direct myocardial depressant effect. Succinylcholine or a rapid-sequence dose of rocuronium can be used to achieve a fast intubation environment and control of the airway. However, other nondepolarizing agents remain appropriate choices. It is recommended that titration of narcotics take effect after control of ventilation to avoid any chest rigidity and hypoventilation episodes, as well as any response to intubation. Inotropic support with an infusion of a catecholamine is used in patients who are at high risk for cardiovascular collapse (Box 24-4).
A pulmonary artery catheter (PAC) routinely is placed after induction rather than before because the hemodynamic status and goals are usually known. A PAC is vital to assess the impact of surgery on pulmonary vascular reactivity without delay. In addition, patients with advanced disease are unable to lie supine or in Trendelenburg position, which sometimes can lead to cardiorespiratory collapse. If the preoperative TTE reveals evidence of RA, RV, or main PA thrombi, TEE is performed immediately after induction and before placement of the PAC (Figure 24-11). In this instance, PAC placement is guided with TEE and the PAC is left in the superior vena cava (at 20 cm) until completion of the surgical procedure. PAC placement can be challenging because of the dilated right atrium and right ventricle, as well as significant TR. TEE guidance often assists in placement. Because all PTE patients undergo prolonged CPB and circulatory arrest, a femoral arterial catheter is placed after induction to monitor arterial pressure after CPB as a radial artery catheter significantly underestimates systemic arterial pressure with a gradient of as much as 20 mm Hg. This phenomenon has been observed by Mohr et al61 and Baba et al62; they proposed redistribution of blood flow away from the extremity as the cause (see Chapter 14).
SEDLine brain-function monitoring (Hospira, Lake Forest, IL) is used to monitor brain function. A four-channel processed electroencephalograph monitor provides monitoring of the isoelectric electroencephalogram and confirmation of minimal oxygen utilization of the brain before circulatory arrest. It also serves as a monitor for the level of consciousness during the entire procedure. The INVOS system (Somanetics Corporation, Troy, MI) is utilized to monitor cerebral oximetry. The device represents a balance between oxygen delivery and consumption by the brain. In a retrospective study of more than 2000 patients, Goldman et al63 confirmed that the institution of cerebral oximetry in their practice decreased the stroke rate in cardiac surgical patients. Yao et al64 observed an association between cerebral desaturation and neurocognitive dysfunction in 101 patients undergoing cardiac surgery. They found that patients with a cerebral oxygen saturation of less than 40% for longer than 10 minutes had an increased incidence of neurocognitive dysfunction (see Chapter 16).
Temperature monitoring is achieved in several ways during PTE to allow accurate quantification of thermal gradients and to ensure even cooling and rewarming. Bladder temperature and rectal probes are used for core temperature estimation. A tympanic membrane probe is used for brain temperature estimation, and the PAC measures blood temperature.65
TEE is used routinely in PTE patients to monitor hemodynamics, evaluate right and left ventricular function, identify any intracardiac thrombus or valvular pathology, and evaluate RV function and deairing after bypass. A thorough intra-atrial septal evaluation, including a bubble study with a Valsalva maneuver, is performed to rule out a PFO (Figure 24-12; see Chapters 12 and 13).
Figure 24-12 Algorithm for performance of intraoperative echocontrast study for pulmonary thromboendarterectomy patients.
PFO is present in 25% to 35% of PTE patients.66 Most PFOs are repaired intraoperatively if detected by color-flow Doppler or with the use of an agitated saline test because some patients may experience high right-sided pressures after surgery. In such cases, hypoxemia will ensue because of right-to-left shunting. In rare instances when results of the operation are not favorable and severe right-sided pressures are expected, the PFO is left open as a “pop-off” to improve RV function and increase CO at the expense of some hypoxemia. It has been suggested that closure of a PFO can be detrimental to clinical status by reducing LV filling and increasing filling of the noncompliant right ventricle.67
Because all PTE patients undergo circulatory arrest, the head is wrapped in a cooling blanket. The head-wrap system (Polar Care; Breg, Vista, CA) is composed of two items: the “Polar Care 500” cooling device (cooling bucket, pump, pump bracket, and AC power transformer), which is reusable, and the actual “wrap,” which is a one-time-use item. The “wrap” utilized was actually designed as a cold therapy pad for postoperative shoulder surgery, but it functions well as a head wrap (Figure 24-13). In a series of 55 patients in whom this device was used during circulatory arrest, they had a mean tympanic membrane temperature of 15.1° C.19 It is the impression that the head wrap provides sufficient cooling to the brain, wraps the whole head, and is far easier to use than ice bags. So far it has been used in more than 2400 cases without any complication.
Initiation of Cardiopulmonary Bypass
The prebypass time is typically short unless concomitant CABG is planned. The bypass pump is primed with 1100 mL of plasmalyte A, 100 mL of 25% albumin, 5–12 mL (100 units/kg) of heparin, 12.5 grams of mannitol, and 1 ampule of bicarbonate. Methylprednisolone, 30 mg/kg max, is given; 3g shortly after initiation of bypass and an additional 500 mg at rewarming. Methylprednisolone theoretically functions as a cell-membrane stabilizer and anti-inflammatory agent.68 Cooling begins immediately after initiation of bypass, using CPB temperature adjustment and the cooling blankets present under the patient, together with the head wrap. Allowing appropriate time to cool and warm the patient in each direction using rectal, bladder, tympanic, PAC, and perfusate temperatures with appropriate thermal gradients ensures even and thorough cooling and warming, respectively.
Phenytoin, 15 mg/kg of propofol, is administered by the perfusionist shortly after initiation of CPB for prophylaxis of postoperative seizures after DHCA. Thiopental, 500 mg, or 2.5 mg/kg of propofol, is administered immediately before DHCA to provide cerebral protection because of dose-dependant reductions in both cerebral blood flow and cerebral metabolic rate.69,70 Although no clear evidence supports the benefit of thiopental administration for DHCA and global cerebral ischemia,71,72 it may be beneficial in focal ischemia because brain cooling may be uneven or incomplete, cerebral emboli may occur as PTE is an open procedure, and in case of sparse electroencephalographic activity, thiopental will abolish any residual activity. Interestingly, Harris et al73 found that thiopental given during profound hypothermia was associated with a dose-related increase in mean arterial pressure of 32 ± 11 mm Hg (P < 0.0001), and flow may need to be decreased temporarily to prevent hypertension.
Rewarming Phase and Separation from Bypass
Separation from CPB follows the same guidelines as any other cardiac surgery requiring CPB, with a few minor exceptions. Communication with the surgeon is of paramount importance because surgical classification of the thromboembolic disease and how much organized clot was successfully removed will dictate how much inotropic and vasopressor support (if any) is needed to separate from bypass. Type I and type II Jamieson disease are likely to have improved hemodynamics with substantial reduction in PVR and improved RV function, which is revealed immediately postbypass with TEE74 (Figure 24-14). Blanchard et al75 demonstrated that the RV Tei index is abnormally increased in CTEPH patients and decreases substantially after PTE. Tei index is a useful parameter in estimating PVR independent of ventricular geometry, and it is monitored easily in CTEPH patients before and after PTE.
In contrast, with type III and IV Jamieson disease, the patient is less likely to have reduced PVR and improved RV function immediately after PTE because surgery is likely to be only partially successful at best. Anticipated need for more aggressive inotropic support (e.g., dopamine, 3 to 7 μg/kg/min, or epinephrine, 0.03 to 0.15 μg/kg/min), together with pulmonary vasodilators such as milrinone, inhaled prostacyclin, and nitric oxide (NO) usually are considered. NO frequently is used if surgery is partially successful because it exerts its effect on the pulmonary vasculature mediating vascular smooth muscle relaxation with minimal systemic effects. Atrial and ventricular epicardial pacing leads are placed routinely to improve RV and atrial function, together with moderate inotropic support to ensure adequate RV coronary perfusion. End-tidal carbon dioxide is a poor measure of ventilation adequacy in these patients both before and after CPB because dead space ventilation is an integral part of the disease process. The arterial-to-end-tidal carbon dioxide gradient will improve after successful surgery, but the response varies. Greater minute ventilation often is required to compensate for a metabolic acidosis that develops after prolonged periods of CPB, circulatory arrest, and hypothermia. Before separation from CPB, intracardiac air and right and left ventricular function are assessed with TEE. With successful operative results, immediate improvements of RV function with resolution of interventricular septal distortion and flattening are seen on intraoperative TEE. With the dramatic resolution of pulmonary hypertension after PTE, transmitral diastolic flow improves in a predictable manner. Not surprisingly, this correlates with improvement in the CO and cardiac index.76 Tricuspid annuloplasty rarely is performed, even if severe TR was documented before surgery, because tricuspid annular geometry is restored with remodeling of the right ventricle after PTE. Figure 24-15 demonstrates severe TR with color-flow and continuous-wave Doppler. It is of paramount importance that the anesthesiologist check the endotracheal tube before separation from CPB because the presence of frothy sputum or bleeding indicates either reperfusion pulmonary edema or airway bleeding, two of the most dreaded complications of the procedure.77 Reperfusion pulmonary edema can start as early as a few hours after the endarterectomy, when the endotracheal tube is suctioned, and escalating amounts of positive end-expiratory pressure (PEEP) are applied.
If frank blood is seen in the endotracheal tube, this signifies a mechanical violation of the airway barrier and stems from a technical error during the operation. Bleeding and pulmonary edema can be distinguished by inspection of the endotracheal tube but may coexist. The two main goals for the treatment of these bleeding episodes are to prevent exsanguination and ensure adequate gas exchange. Management depends on the severity of the bleeding. Often, conservative management consisting of PEEP, topical vasoconstrictors, reversal of heparin, and correction of coagulopathies will reduce the bleeding. If bleeding is severe, resumption of CPB is indicated, and diagnostic fiberoptic bronchoscopy is used to evaluate the source of bleeding, as well as to help isolate the bleeding segment. A double-lumen tube is the best method for lung isolation, and differential ventilation and PEEP can be used effectively. The only caveat with a double-lumen tube is the difficulty in using a large bronchoscope that limits suctioning and diagnostic capabilities, and exchanging the tube can be risky. A Univent tube is an alternative to a double-lumen tube. Treatment options (e.g., lung isolation, PEEP, topical vasoconstrictors such as vasopressin, phenylephrine) depend on the severity of bleeding.77 Risk factors for bleeding in these patients are difficult to identify before. They include the presence of friable vessels, residual pulmonary hypertension, spontaneous rupture, trauma, and a technically difficult procedure.
Pulmonary hemorrhage most likely manifests on reinstitution of pulmonary blood flow during weaning from CPB. Unfortunately, it is difficult to test the integrity of the PA during the endarterectomy because cardiac ejection of blood into the PA and appropriate systolic pressure are needed. The anesthesiologist involved with these procedures should be prepared to provide diagnostic and therapeutic maneuvers such as bronchial blockade, differential lung ventilation, endobronchial administration of vasoconstrictors, and pharmacologic control of residual pulmonary hypertension (Figure 24-16). Massive pulmonary hemorrhage after PTE, although a rare event, is the third most common cause for perioperative mortality in the PTE population, with residual pulmonary hypertension and reperfusion pulmonary edema as the leading two causes.
Postoperative Period and Adverse Effects
Reperfusion Pulmonary Edema
Most patients undergoing PTE will develop some degree of localized pulmonary edema or “reperfusion response.”78 Reperfusion response or reperfusion injury is a localized form of acute respiratory distress syndrome defined as a radiologic opacity seen in the areas of the lungs that underwent endarterectomy. It usually occurs within 24 to 72 hours, but in its most severe form, it is seen right after CPB separation after the endarterectomy, presenting as profound desaturation. Mortality from clinically significant reperfusion injury is observed in less than 10% of patients. One common cause of the reperfusion pulmonary edema is persistent high PAPs after a thorough endarterectomy, with a large part of the pulmonary vascular bed affected by type IV disease. However, the reperfusion phenomenon may be seen in patients after a complete resolution of high PAPs. One theory of the cause is reactive hyperemia after restoration of blood flow to segments of the PA bed after periods of no flow. Other contributing factors may include pulmonary ischemia and conditions associated with high-permeability lung injury. The incidence of this complication is much less common now secondary to the large experience acquired over the past ten years, leading to more complete and expeditious endarterectomies.
Management of the “Reperfusion Response.”
Aggressive measures should be taken to minimize the development of pulmonary edema with diuresis, maintenance of adequate hematocrit levels, and the early use of PEEP. Careful management of ventilation and fluid balance is required, and usually reperfusion pulmonary edema resolves. PEEP is initiated with a gradual transition from volume-limited to pressure-limited inverse ratio ventilation and the acceptance of moderate hypercapnia.78 Inhaled NO at 20 to 40 parts per million (ppm) has been used occasionally to improve gas exchange. Extracorporeal perfusion support (extracorporeal membrane oxygenation or extracorporeal carbon dioxide removal) has been limited to patients with hemodynamic improvement who are suffering from significant reperfusion response.79,80
Persistent Pulmonary Hypertension
In rare instances when certain patients will unlikely survive waiting for a lung transplant donor, the surgeons will operate on them despite considerable risk. More than one third of perioperative deaths in the most recent 500 patients who were operated on in this institution were directly attributable to inadequate relief of PAH. This was a diagnostic rather than an operative technical problem. Pharmacologic attempts to reduce high residual PVR levels with sodium nitroprusside, epoprostenol, or inhaled NO generally are not effective. Mechanical circulatory support or extracorporeal membrane oxygenation use in these patients is not appropriate.79,80
Pulmonary Vascular Steal
Pulmonary arterial steal is a postoperative phenomenon in which blood flow is diverted from the previously well-perfused segments to the newly endarterectomized segments.81 The steal entity was recognized when perfusion scans obtained after the operation showed regions of reduced perfusion suggestive of recurrent emboli.81 Whether the cause is failure of autoregulation in the newly endarterectomized segments or secondary small-vessel changes in the previously open segments is not clear. Preferentially perfused areas may be the site of reperfusion response in patients with steal phenomena and residual hypertension, leading to severe hypoxia from areas that are not effectively contributing to oxygenation. However, long-term follow-up has documented a decrease in pulmonary vascular steal in the majority of patients, suggesting a remodeling process in the pulmonary vascular bed.82
Other Postoperative Complications
Delirium had a greater incidence in the early experience before 1990. A prospective study of 28 patients showed that 77% of patients who underwent PTE experienced delirium, with peak incidence at 72 hours after surgery.83 Analysis attributed delirium to a circulatory arrest time greater than 55 minutes. The incidence rate declined to 11% and, in a recent cohort of 1000 patients, to 1.3%. More experience, expeditious operations resulting in shorter circulatory arrest time, and the use of a direct cooling jacket placed around the head have contributed to the substantial decline in neurologic complications. Postoperative confusion in PTE cases is now the same as with other cardiac surgery.