Surgical Management of Unresolved Pulmonary Embolism

Surgical Management of Unresolved Pulmonary Embolism* A Personal Series of 72 Patients Gerald Simonneau, MD; Reza Azarian, MD; Frant;ois Brenot, MD; Philippe G. Dartevelle, MD; Dominique Musset, MD; and Pierre Duroux, MD

Chronic thromboembolic pulmonary hypertension (CT-E PH) is a rare and aberrant outcome of acute pulmonary embolism. Because it has become a potentially curable form of pulmonary hypertension, the frequency of recognized cases has increased. We report a case series of 72 patients with CT-E PH evaluated in our institution between 1984 and 1993, and discuss diagnostic clues and therapeutic approaches. All patients complained of dyspnea on exertion, a history of acute thromboembolic event, and lung murmurs were found in 60% and 17% of patients, respectively. The presence of a disorder of coagulation was found in 30% of the patients tested, the most common abnormality being lupus anticoagulant. 'Fhe key noninvasive study for diagnosis was the lung perfusion scan which showed at

least one segmental or larger perfusion defect in all patients. Pulmonary angiography confirmed the diagnosis in all cases, and sometimes associated to intravascular ultrasound imaging, established the feasibility of thromboendarterectomy. Medical therapy included the use of long-term oral anticoagulant, and in the case of lower limb venous thrombosis, inferior vena cava filtration. Finally two surgical procedures were discussed in selected patients: thromboendarterectomy and lung transplantation. Since 1988, eight patients have benefited from lung transplantation (six patients are still alive), and 11 patients underwent thromboendarterectomy which was successful in 9 patients with a dramatic functional and hemodynamic improvement. (Chest 1995; 107:52S-55S)

an acute pulmonary embolism, some patients fail to resolve their thrombi and may develop chronic thromboembolic pulmonary hypertension (CT-E PH) . Until recently, this entity was considered to be a rare autopsy curiosity. However, because of advances in diagnostic approaches and surgical management, this disease can be now considered not only more common than anticipated but also accessible to surgical treatment. There are currently two different surgical procedures that can be proposed to patients suffering from CT -E PH: pulmonary thromboendarterectomy (PTE) and lung transplantation. The purpose of this article is to report a French series of 72 patients and to discuss the respective indications of PTE and transplantation.

and measurement of the diffusing capacity for CO; analysis of arterial blood gases; lower limb venography; perfusion lung scan; right heart catheterization; and pulmonary angiography. CT-E PH was defined as a mean resting pulmonary artery pressure >25 mm Hg during right heart catheterization with a mean pulmonary wedge pressure below 12 mm Hg, associated with at least one segmental or larger perfusion lung scan defect and an angiographic pattern of chronic thromboembolic pulmonary obstruction. Patient selection criteria for PTE were those defined by the San Diego groupJ symptomatic patients with calculated pulmonary vascular resistance (PVR) at rest greater than 300 dynes· s- 1 • em-s; thrombus location in the main, lobar, or segmental arteries; the absence of severe associated diseases; and the willingness of the patient to accept the surgical risks. Selection criteria for lung transplantation were derived from our selection criteria for patients with primary pulmonary hypertension, 2 ie, patients belonging to functional class 3 or 4 as defined by the New York Heart Association, a mean pulmonary artery pressure above 40 mm Hg associated with a cardiac index below 2.5 L/ min / m 2 , and PVR>900 dynes· s- 1 . cm- 5 . The choice between heart-lung transplantation and double lung transplantation depended on donor's availability and some other surgical and medical considerations. In all cases, patients were maintained on a regimen of permanent war-

~ter

MATERIAL AND METHODS

Between 1984 and 1993, 72 patients were evaluated at Antoine Bidere Hospital, Universite ParisSud. This evaluation included the following: medical history: physical examination; electrocardiogram; two-dimensional echocardiogram and Doppler; measurements of antithrombin III, protein C, protein S, and lupus anticoagulant; pulmonary function tests *From the Service de Pneumologie e t de Re animation Respiratoire and Radiologic, Hopi tal Antoine Beclere, Clamart, France, and the Service de Chirurgie Thoracique, Centre Chirurgical Marie Lannelongue, Le Plessis Robinson , France. Reprint requests: Dr. Simonneau, Hopital Antoine Beclere, 157 Rue de la Porte de Trivaux, 92141 Clamart, France 52S

Surgical Management of Unresolved PE (Simonneau eta/)

Table !-Clinical Characteristics of All Patients* Age, yr Sex ratio, M/ F Functional class, % 1-2 3-4 History of DVT or PE,% Time between onset of symptoms and diagnosis, mo

52±15 1/ 3 56 44 63 28±43

*DVT-deep venous thrombosis; PE=acute pulmonary embolism.

farin therapy. An inferior vena cava filter was placed only in patients with abnormal lower limb venography. RESULTS

The major clinical characteristics of the 72 patients are listed in Table 1. Twenty-six patients (36%) had no history of deep vein thrombosis or acute pulmonary embolism and were referred because of dyspnea on exercise. A disorder of coagulation was found in about 30% of the tested patients, the most common abnormality being the presence of a lupus anticoagulant (Table 2). Pulmonary function test results and hemodynamic values are shown in Table 3. Pulmonary spirometry was normal in most patients. The diffusing capacity for carbon monoxide was often within the normal range. Arterial Pa02 at rest was normal or slightly decreased, except for patients with a right to left intracardiac shunt due to a patent foramen ovale. Right heart catheterization revealed a markedly elevated pulmonary arterial pressure with a normal wedge pressure. Cardiac output was normal or decreased. In all patients, there was at least one segmental or larger perfusion defect on perfusion lung scan. In all cases, pulmonary angiography showed characteristic patterns of chronic thromboembolic pulmonary obstruction, including webs, bands, pouchings, and some degrees of intimal irregularities, or abrupt vascular narrowing and complete obstruction. All patients had long-term oral anticoagulation in order to obtain an international normalized ratio between 2 and 3. An inferior vena cava filter was placed in 33 (46%) patients. Eleven patients (15%) were selected for PTE. Two of them (18%) died postoperatively because of reperfusion pulmonary edema in one and pulmonary hypertensive lesions in Table 2-Coagulopathies in Patients With CT-E PH

Antithrombin Ill deficiency Protein C deficiency Protein S deficiency Lupus anticoagulant Plasminogen deficiency

Tested Patients, n

Abnormalities, %

47 39 39 38 21

2 8 0 24 5

Table 3-Pulmonary Function and Hemodynamics in All Patients at the Time of Diagnosis* TLC,% pred FEV 1,% pred bco,% Pa02, mm Hg Mean RAP, mm Hg Mean PAP, mm Hg Mean PWP, mm Hg CI, L/ min/ m2 PVR, IU/ m 2 PVR, dynes·s-l·cm-5

90±14 81±13 63±20 68±15 10±4 48±18 10±6 2.6±0.7 20±10 919±447

*TLC and FEV1, respectively, total lung capacity and forced expiratory volume in 1 s, both expressed as percent of predicted values. RAP=right atrial pressure; PAP=pulmonary artery pressure; PWP=pulmonary wedge pressure; CI=cardiac index; PVR=pulmonary vascular resistance.

the other. They both had preoperative PVR of 1,500 and 1,700 dynes·s- 1 ·cm- 5 and a duration of symptoms until diagnosis of 3 to more than 15 years, respectively. In the nine survivors, PTE led to a dramatic functional and hemodynamic improvement (Table 4). Eight patients (11%) were selected for heart-lung or lung transplantation. As shown in Table 5, they were very severely compromised patients. One patient died on the first postoperative day; another died of obliterative bronchiolitis 2 years after surgery. The other six patients are alive with a follow-up of 5 months to 5 years. DISCUSSION

Chronic thromboembolic pulmonary hypertension is a rare disease that occurs in approximately 0.01% of patients with acute pulmonary embolism.3 Since it has become potentially treatable, because of advances in lung transplantation and cardiac surgery, the antemortem recognition of this disorder has become a concern. The natural history and the clinical characteristics of the disease have been well described by Moser et al 1 on the basis of their experience of nearly 400 patients. We report a French series of 72 patients with almost the same findings as Moser et al, except for some points that are now discussed. The mechanism of this aberrant outcome of acute Table 4-Functional and Hemodynamic Results of Pulmonary Thromboendarterectomy

NYHA 1-2, n NYHA 3-4, n mPAP, mm Hg CI, L/ min/ m 2 PVR, IU / m 2 PVR, dynes·s-l·cm-5

Before PTE

After PTE (3 mo)

4 5 39±11 2.6±0.9 18±10 823±505

9 0 26±8 3.3±0.8 8±4 393±161

*NYHA =New York Heart Association functional class; mP AP=mean pulmonary artery pressure; CI=cardiac index; p<0.05. CHEST /107/1 I JANUARY, 1995/ Supplement

53S

pulmonary embolism is unclear. The absence of resolution of thrombi may be due to endogenous fibrinolytic system abnormalities, but such a mechanism has not yet been confirmed. 4 Another hypothesis implies recurrent unrecognized pulmonary embolism. This is supported by the finding of 30% coagulopathies in our patients, a rate that is much higher than in the general population (2.2%), in patients presenting with deep vein thrombosis (8.3%) 5 and surprisingly higher than in the patients of Moser et aP That coagulopathies have not been studied in all patients could explain the low rate reported by Moser et al. 1 In patients with CT-E PH, pulmonary hypertension is not only due to the mechanical obstruction by chronic thrombi, but also to pulmonary hypertensive lesions (Eisenmenger-type lesions) of small arteries that may develop in the "open" pulmonary vascular bed during the "honeymoon" period.1·6 Histopathologic findings in patients with CT-E PH are similar to those described in patients with primary pulmonary hypertension.6 This explains why the hemodynamic worsening can occur without any change in lung scan pattern in most patients. It also could be one of the main causes of failure after thromboendarterectomy, suggesting that efforts have to be made to operate on patients before the "honeymoon" period ends. Diagnostic approaches in CT-E PH have been well defined by Moser et at,l the two key studies being cardiac echo-Doppler, which evaluates pulmonary artery pressure, and perfusion lung scan, which suggests an embolic etiology. These findings lead to right heart catheterization to confirm pulmonary hypertension and to pulmonary angiography to confirm the diagnosis and define the location of chronic thrombotic material and its accessibility to surgery. The angiographic patterns of chronic thromboembolic pulmonary artery obstruction have been well described? "pouching" defects, webs, and bands, abrupt vascular narrowing, intimal irregularities, and complete vascular obstruction.

In our experience, most patients were not considered as candidates for surgery and only required medical treatment that included long-term anticoagulation (to obtain an international normalized ratio between 2 and 3), and inferior vena cava filter placement when lower limb venous thrombosis was supported by venography or echography. The benefit of anticoagulant therapy in patients with pulmonary hypertension has been demonstrated previously .8 However, the prognosis of CT-E PH remains poor with a 5-year survival rate of 30% when the mean pulmonary artery pressure exceeds 40 mm Hg.9 The only curative alternatives are lung transplantation and PTE. The selection criteria for PTE, as established by Moser et al, 1 include the following: (1) thrombi defined as accessible to surgical removal; (2) a willingness of the patient ; (3) a PVR of 300 dynes·s- 1·cm-5; and (4) the absence of significant associated disease. Pulmonary thromboendarterectomy is obviously preferred to transplantation because of lower postoperative mortality rate and a better long-term survival rate without requiring any immunosuppressive therapy .10•11 Thus, the main goals of the evaluation of such patients are, first to determine whether major pulmonary vessels are involved, and, second, to detect any associated disease that would exclude PTE. Perfusion lung scanning underestimates the degree of proximal obstruction 12 and angiography may fail to define the precise thrombus location. Computed tomography (CT) can sometimes show proximal chronic thrombi. Its main advantage is that it is noninvasive. 13·14 Intravascular ultrasound imaging may help to analyze the main pulmonary arteries. In one study that demonstrated its feasibility and safety, the results were not convincing.l5 However, the authors used a 20-MHz ultrasound imaging system that is less appropriate than the 12-MHz system for large-vessel analysis. The latter is currently being evaluated in our institution. Angioscopy allows the direct visualization of pulmonary arteries and can define precisely the site of chronic thrombotic material. 3 However, the angio-

Table 5-Results of Lung Transplantation in Patients With CT-E PH* Patient No. / Age, yr

NYHA Class

Mean PAP, mm Hg

CI, L / min / m 2

PVR, IU/ m 2

PVR, dyneS·S-l·Cffi-S

Transplantation Type

Follow-up

1/ 29 2/48 3/ 27 4/34 5/52 6/ 48 7/37 8/39

3 3 3 3 3 4 4 4

53 70 53 59 53 45 56 48

2.2 3.3 1.4 2.2 1.6 1.5 2.0 2.3

24 21 38 27 33 30 28 21

946 921 1600 1443 1631 1286 1179 960

DLT HLT DLT HLT HLT HLT HLT HLT

5 mo (alive) 4 yr (alive) 3 yr (alive) 4 yr (alive) 3 yr (alive) 1 d (dead) 5 yr (alive) 2 yr (dead)

*NYHA =New York Heart Association; P AP=pulmonary artery pressure; CI = cardiac index; DL T=double lung transplantation; HL T= heart-lung transplantation.

54S

Surgical Management of Unresolved PE (Simonneau eta/)

scope used in San Diego is an experimental device that is unfortunately not available elsewhere, and other angioscopy devices are not suitable for such patients with high pulmonary artery pressure. The second main goal in the evaluation of such patients is the search for any comorbid disorder that would rule out PTE. However, there are no clearly defined absolute exclusion criteria for PTE. Daily et aP 6 did not find any associated disease (including coronary artery disease , renal failure, diabetes, or peripheral vascular disease) as significant risk factors for postoperative ventilator dependency and hospital mortality; the only significant preoperative factor that influenced ventilator dependency was ascites, reflecting end-stage right heart failure . Chitwood et aP 0 stated that small distal emboli, massive obesity, and severe right heart failure were contraindications to PTE. In this report , the two deaths after PTE were probably related to an advanced stage of the disease and the high level of preoperative PVR . To improve the results of PTE, it is necessary to prevent the postoperative reperfusion pulmonary edema, but it is also important to better identify patients with associated pulmonary hypertensive lesions. In our series, only 15% of the patients were selected for PTE. This is in contrast with the 90% PTE candidates in San Diego. The reason forthis difference is probably that patients who were referred to San Diego were already "preselected" for surgery . Actually, most of our patients were not candidates for PTE because thrombi were considered too distal or patients were too severely compromised. To better identify PTE candidates, we need to improve the assessment of thrombus definition by intraluminal echography or angioscopy, or both. When PTE is not possible, the only curative therapy is lung or heart-lung transplantation. Considering the current results of lung transplantation, with an actuarial survival rate of 70% at 1 year and 40% at 4 years, 17 this alternative should be considered only in very severely hemodynamically compromised patients. The selection criteria for lung transplantation for CT-E PH in our institution were derived from those for primary pulmonary hypertension: (1) functional class 3 to 4 (New York Heart Association); (2) mean pulmonary artery pressure >40 mm Hg, cardiac index <2.5 L /min / m 2 , and PVR >20 IU / m 2 (900 dynes·s- 1 -cm- 5 ) ; (3) the absence of general contraindication to lung transplantation; (4) distal

thrombi not accessible to PTE ; and (5) the willingness of the patient. The respective indications of heartlung, double lung, and single lung transplantation remain to be evaluated. REFERENCES

2 3

4

5

6

7

8

9

10 11

12

13

14

15

16

17

Moser KM, Auger WR, Fedullo PF, et al. Chronic thromboembolic pulmonary hypertension: clinical picture and surgical treatment. Eur Respir J 1992; 5:334-42 Brenot F . Primary pulmonary hypertension: case series from France. Chest 1994; 105(suppl):33S-6S Moser KM , Auger WR, Fedullo PF. Chronic major-vessel thromboembolic pulmonary hypertension. Circulation 1990; 81:1735-43 Olman MA, Marsh JJ, Lang IM, et al. Endogenous fibrinolytic system in chronic large-vessel thromboembolic pulmonary hypertension. Circulation 1992; 86:1241-48 Heijboer H , Brandjes DPM, Buller HR, et al. Deficiencies of coagulation-inhibiting and fibrinolytic proteins in outpatients with deep-vein thrombosis. N Eng! J Med 1990; 323:1512-16 Moser KM , Bloor CM. Pulmonary vascular lesions occurring in patients with chronic major vessel thromboembolic pulmonary hypertension. Chest 1993; 103:685-92 Auger WR, Fedullo PF, Moser KM, et al. Chronic major-vessel thromboembolic pulmonary artery obstruction: appearance at angiography. Radiology 1992; 182:393-98 Fuster V, Steele PM, Edwards WD, et al. Primary pulmonary hypertension: natural history and the importance of thrombosis. Circulation 1984; 70:580-87 Riedel M, Stanek V, Widimsky J, et al. Longterm follow-up of patients with pulmonary thromboembolism. Chest 1982; 81 :151-57 Chitwood WR, Lyerly HK, Sabiston DC. Surgical management of chronic pulmonary embolism. Ann Surg 1985; 201:11-26 Moser KM, Daily PO, Peterson K, et al. Thromboendarterectomy for chronic, major-vessel thromboembolic pulmonary hypertension. Ann Intern Med 1987; 107:560-65 Ryan KL, Fedullo PF, Davis GB, et al. Perfusion scan findings understate the severity of angiographic and hemodynamic compromise in chronic thromboembolic pulmonary hypertension. Chest 1988; 93:1180-85 Falaschi F, Palla A, Formichi B, et al. CT evaluation of chronic thromboembolic pulmonary hypertension. J Comput Assist Tomogr 1992; 16:897-903 Tardivon A, Musset D, Maitre S, et al. Role of CT in chronic pulmonary embolism: comparison with pulmonary angiography. J Comput Assist Tomogr 1993; 17:345-51 Ricou F, Nicod PH, Moser KM, et al. Catheter-based intravascular ultrasound imaging of chronic thromboembolic pulmonary disease. Am J Cardiol1991 ; 67:749-52 Daily PO, Dembitsky WP, Iversen S, et al. Risk factors for pulmonary thromboendarterectomy. J Thorac Cardiovasc Surg 1990; 99:670-78 Chapelier A, Dartevelle P, Vouhe P, et al. Comparative outcome of heart-lung and lung transplantation for pulmonary hypertension: report of 30 cases. J Thorac Cardiovasc Surg 1993; 106:299-307

CHEST / 107 / 1 I JANUARY, 1995 / Supplement

555