Catheter-based intravascular ultrasound imaging of chronic thromboembolic pulmonary disease

MISCELLANEOUS

Catheter-Based Intravascular Ultrasound Imaging of Chronic Thromboembolic Pulmonary Disease Francois Ricou, MD, Pascal H. Nicod, MD, Kenneth M. Moser, MD, and Kirk L. Peterson, MD

Pulmonary thromboendarterectomy is now the treatment of choice for pulmonary hypertension due to chronic pulmonary thromboemboli. A precise assessment of location and extension of these thrombi is important because only proximal chronic pulmonary thromboemboli are accessible to surgery. Because intravascular ultrasound imaging can assess not only arterial luminal size, but also wall thickness, its value as a complement to angiography was assessed in 11 patients aged 35 to 64 years with severe pulmonary hypertension (systolic pulmonary artery pressure, mean f standard deviation 70 f 19 mm Hg; pulmonary artery resistance, 609 f 297 dynes-s=cm-5). Intravascular ultrasound was obtained in 10 of 11 patients and no complication occurred. Intravascular ultrasound identified 10 segments with suspected chronic pulmonary thromboemboli in 7 patients, all confirmed at operation. Nine segments were considered normal, all of which (except 1) were free of chronic pulmonary thromboemboli at operation. Image quality was highly dependent on pulmonary artery size and position of the catheter. Therefore, intravascular ultrasound of pulmonary arteries is feasible and safe in patients with pulmonary hypertension. It may help to assess the location and extension of the pathologic process involving pulmonary arteries. (Am J Cardiol 1991;67:749-752)

urgical thromboendarterectomyis now the treatment of choice for severepulmonary hypertension due to chronic thromboembolic obstruction of the major pulmonary arteries. This operation has been shown to substantially improve pulmonary artery pressure and resistance,’ cardiac geometry2and functional class.1,3Despite recent advancesin surgical techniques, pulmonary thromboendarterectomy still carries a substantial risk with a perioperative mortality of 13%in the most favorable series.4A precisepreoperativeevaluation of the extent, the location and thickness of the chronic thrombi is essential,therefore, to optimize the surgical result and miriimize the operative risks. Pulmonary angiography is the more reliable diagnostic procedure in these patients, but its interpretation is problematic becauseof the highly variable pattern of recanalized chronic thromboemboli.5JjOther diagnostic proceduressuch as angioscopy,7computerized tomography* and magnetic resonanceimaging9 also have been used to complement pulmonary angiography. Catheterbased intravascular ultrasound is a new technique allowing 2-dimensional,cross-sectionalimaging of vessels. It can give information not only on luminal size, but also on vesselwall thickness. Preliminary studies have mostly describedits use in vitro and in peripheral arteries.l”-14 Its use in normal pulmonary arteries has also been recently described.l5 In this study, we report our experiencewith intravascular ultrasound imaging in 11 patients with pulmonary hypertension due to suspected chronic thromboembolic obstruction. Our early findings indicate that intravascular ultrasound can be done safely in these patients and may help to assessthe location and extension of the pathologic processinvolving pulmonary arteries.

S

METHODS Patients: Eleven patients (9 men and 2 women aged

From the Cardiology and Pulmonary Division, University of California, San Diego Medical Center, San Diego. Manuscript received August 27, 1990; revised manuscript received and accepted December 17, 1990. Address for reprints: Kirk L. Peterson, MD, Division of Cardiology, H811-A, University of California, San Diego Medical Center, 225 Dickinson Street, San Diego, California 92103-1990.

35 to 64 years [mean age f standard deviation 52.4 f 10.51) were evaluated in the University of California, San Diego Medical Center, for pulmonary hypertension and suspectedchronic thromboemboli of large pulmonary vessels.All patients underwent thorough clinical assessmentand various noninvasivetests including chest x-ray, electrocardiogram,echocardiography,V/Q scan, spirometric testing and arterial blood gas analysis as previously described.4J6 Cardiac catheterization: Before cardiac catheterization, all patients had to be clinically stable, normoten-

THE AMERICAN JOURNAL OF CARDIOLOGY APRIL 1, 1991

749

TABLE I Hemodynamic

Pt No. 1 2 3 4 5 6 7 8 9 10 11

Mean f SD PAP = pulmonary

artery pressure;

Data in Patients with Suspected Chronic Thromboembolic

(n = 11)

Mean PAP (mm Hg)

Mean RAP (mm Hg)

Cardiac output (liters/min)

Pulmonary Resistance (dynes.s.cm-5)

85 70 74 60 85 80 85 96 40 38 58 7o;t 19

35 45 44 35 58 46 45 50 23 28 36 41 f. 10

3 14 6 6 8 5 14 7 5 6 9 7.5dc3.7

5.6 2.5 4.6 4.5 4.1 5 3.3 4.1 4.5 2.7 6.5 4.3*

343 1,214 597 515 917 569 880 702 270 416 283 609 rt 297

RAP = right atrial pressure; SD = standard

1.2

deviation.

sive and free of significant arrhythmias. For accessto the right heart, the right internal jugular vein was used to avoid the risk of dislodging thrombi from veins in the legs. After insertion of an 8Fr vascular sheath, a 7Fr thermodilution balloon flotation catheter was advanced to the pulmonary artery to obtain standard measurements of pressuresand cardiac output. After completion of the hemodynamic studies,the initial catheter was replaced by an 8Fr Berman catheter. Pulmonary angiograms of the right and the left pulmonary arteries were performed sequentially as previously described.l6 Intravascular ultrasound imaging: A 0.032~inch guidewire was advanced to the pulmonary artery through a 7Fr Critikon balloon catheter. After removal of the catheter, an 8Fr Mullins transseptal sheath was advanced together with the balloon catheter over the wire and left in place to the level of the main right or left pulmonary arteries. The guidewire and the balloon catheter were then removedand the intravascular ultrasound imaging catheter was advanced through the sheath to the right or left main pulmonary artery. A 20MHz ultrasound imaging system (Cardiovascular Imaging System, Inc., Sunnyvale, California) is mounted on an 8Fr catheter, 65 cm in length. A mechanically rotating mirror located at the tip of the catheter radiates the ultrasound energy and collects the reflecting signals. The signal is then processedto create an image of the artery in a 360’ cross-sectionon a video screen. The penetration of the ultrasound beam using a frequency of 20 MHz is about 1 to 2 cm. The position of the intravascular ultrasound catheter was changedusing fluoroscopic guidance. Images were recorded on a videocassetterecorder and printed using a Sony UP-100 Videographic Printer. All patients signed an informed consentbefore the ultrasound procedure,which was approved by the Institutional Review Board at the University of California at San Diego. Eight patients, 7 with suspectedchronic thromboembolic pulmonary hypertension and 1 with suspectedpulmonary artery tumor, underwent surgery by techniques previously described.l7 Findings at operation were compared with the location of chronic thromboemboli predicted by intravascular ultrasound. 750

Pulmonary Hypertension

Systolic PAP (mm Hg)

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 67

RESULTS Hemodynamic findings: All patients evaluated had significant pulmonary hypertension as describedin Table I. Angiographic findings: Pulmonary angiogramswere obtained in all patients. The diagnosisof chronic thromboembolic pulmonary hypertension was suspectedin 9 of 11 patients. In 1 patient, primary pulmonary hypertension was diagnosed. In another, a primary pulmonary artery tumor was suspected. Intravascular ultrasound findings: Real time crosssectional images of the pulmonary arteries were obtained in 10 of 11 patients. In 1 patient, the catheter could not be advanced beyond the right ventricle becausean acute angle at that level led to kinking of the sheath. In 3 patients, both right and left pulmonary arteries could be imaged; in 7 patients, only the right pulmonary artery could be imaged while the left pulmonary artery could not be engagedbecauseof an acute angulation at its origin from the main pulmonary artery. The procedure was tolerated well by all patients. No complications occurred. In the segmentsconsiderednormal by intravascular ultrasound, the arterial lumen was circular or oval (Figure 1B). Segmentswith an internal diameter ranging from 6 to 31 mm were imaged. Branches could be detected easily (Figure 2C). In patients with angiographic suspicion of chronic thromboemboli, 7 had abnormalities on their ultrasound images. In some patients, a marked thickening of the vessel wall was noted (Figure lC), sometimeswith a “crescentic layer” appearance(Figure 2C). In the patient suspectedof having a tumor, an echogenic mass was seen adjacent to the vessel wall (Figure 3B). At operation, a fibrous histiocytoma was found extending from the pulmonary valve into the right and left pulmonary arteries. Correlation with angiographic findings: Ultrasound imaging identified 10 areasin 7 patients that indicated the presenceof organized thromboemboli or of tumor. At pulmonary angiography, abnormalities were noted in eachinstance (Figure 1A). However, in somecasesonly mild tapering of the vesselwas noted (Figure 2A).

Correlation with findings at surgery: Sevenof the 8 patients with diagnosis of suspectedchronic major vessel thromboembolic obstruction and the 1 with suspected tumor underwent surgery. In 1 patient with suspected thromboemboli, surgery was not undertaken because only mild pulmonary hypertension was present (pulmonary arterial resistance at rest: 280 dynes+cm-5) and the patient was stable clinically. She was treated by calcium antagonists and will be reevaluated after 6 months. In the surgically treated patients, the diagnosis of chronic thromboembolic obstruction or tumor of pulmonary arteries was confirmed at operation, Chronic organized thromboemboli or tumor were found in each of 10 segmentsjudged abnormal by ultrasound imaging (Figure 4). Nine segmentsthat were judged normal by ultrasound were free of chronic thromboemboli at surgery except 1 caseof a clot located in the right proximal pulmonary artery.

DISCUSSION

In this study, we report the initial results of intravascular ultrasound imaging of pulmonary arteries in a unique population with chronic thromboemboli. Our experience indicates that intravascular ultrasound images can be obtained safely and provide information about arterial luminal size and, particularly, vesselwall thickness that cannot be obtained from pulmonary angiograms. Chronic recanalized thromboemboli appear as areas of marked wall thickening or have a crescentic layer appearance.Thus, the technique may serve to localize accurately chronic organized thromboemboli in the pulmonary tree and may help to select the most suitable patients for surgical thromboendarterectomy. Pulmonary angiography is the most useful method for assessingthe extent of chronic pulmonary thromboemboli. However, in many cases,recanalized throm-

FIGURE 1. A, angiographic images of the right pulmonary artery with a normal appearance of the main pulmonary artery (large arrow) and tapering of the inferior lobs branch (small arrow) suggesting chronic clots. R, intravascular ultrasound images of the main right pulmonary artery showing a normal appearance of the lumen and of the vessel wall (2 mm between dots). C, intravascular ultrasound images of the inferior lobe right pulmonary artery showing thichening of the vessel wall (arrows) (2 mm betweeen dots).

FIGURE 2. A, angiographic image of the left pulmonary artery showing a mild tapering of the inferior branch (arrow). B, the intravascular ultrasound imaging catheter is positioned in the loft inferior tobe pulmonary branch (arrow). C, intravascular ultrasound image of the left inferior lobe pubnonary branch showing a “‘crescentic layer” appearance of the vessel wall (small arrows) (1 mm behveeen do&). X = presence of a side branch.

FIGURE 3. A, angiographic image of the right pulmonary artery showing a tapering of the intermediate branch (arrow). R, intravascular ultrasound image of the intermediate pulmonary artery showing an echolucent mass adjacent to the vessel wall (small arrows) (2 mm betweeen dots). At operation, a fibrous histiocytoma was found.

THE AMERICAN JOURNAL OF CARDIOLOGY APRIL 1, 1991

751

some loss of information. The ultrasound catheter we used had a frequency of 20 MHz. This allows imaging at a depth of about 1 to 2 cm, which limits visualization of the very large proximal pulmonary arteries often presentin this disease.This may explain why we missed a proximal clot in the right pulmonary artery, 28 mm in diameter. The use of catheters with a lower ultrasonic pulse frequency should provide images of larger crosssectional areas. Acknowledgment: We thank Maurice Buchbinder, MD, for his technical contribution to the study and David Sahn, MD, for his professional advice and encouragement.

REFERENCES

FIGURE 4. Surgical specimen of chronic pulmonary thrombosmboli of the right inferer lobs pulmonary branch removed by surgical thromboenderterectomy (same patient as Figure 1).

boemboli are present in areas with only subtle changes on pulmonary angiograms, such as mild or smooth tapering of the artery. This is particularly common when chronic emboli are present in only the lobar or segmental branchesand distinct cut-offs or websare not seen.‘* In such patients, definition of the thickness of the chronic thrombotic material is of particular importance. Even given optimal angiographic/angioscopic information, such distinctions are not possiblein somepatients. Ultrasound imaging, therefore, could prove valuable in defining the surgical accessibility of these chronic thromboemboli. Identification of tumor massesmay also be possibleusing intravascular ultrasound technology, as suggestedby our caseof fibrous histiocytoma. Study limitations: The ultrasound catheter has several important technical limitations. The catheters we have used were stiff and had limited steerability. Imaging was therefore limited to the right and left main pulmonary arteries and descending (lower lobe) lobar arteries. Development of a more flexible over-the-wire system may allow imaging of more distal branches. Also, an over-the-wire system may allow the imaging device to remain in the center of the artery to optimize the quality of the image. Currently, the catheter is often eccentric becauseof the large size and pulsatility of the pulmonary artery, which results in image distortion and

752

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 67

1. Moser KM, Daily PO, PetersonKL, Dembitsky W, Vapnek JM, Shure D, Utley Y, Archibald C. Thromboendarterectomyfor chronic, major vesselthromboembolicpulmonary hypertension.Ann Intern Med 1987;107:560-565. 2. Dittrich HC, Nicod PH, Chow LC, ChappuisFP, Moser KM, PetersonKL. Early changesof right heart geometryafter pulmonary thromboendarterectomy. J Am Coil Cardiol 1988;11:931-943. 3. Rich S, Levitsky S, Brundage BH. Pulmonary hypertension from chronic pulmonary thromboembolism.Ann Intern Med 1988;108:425-434. 4. Moser KM, Auger WR, Fedullo PF. Chronic, major vesselthromboembolic pulmonary hypertension.Circulation 1990;81:1735-1743. 5. Moser KM, Houk VN, JonesRC. Chronic, massivethrombotic obstructionof the pulmonaryarteries: analysisof four operatedcases.Circulation 1965;32:377385. 6. Moser KM, SpraggRG, Utley .I, Daily PO. Chronic thrombotic obstructionof major pulmonary arteries: result of thromboendarterectomyin 15 patients.Ann Intern Med 1983;99:299-305, 7. ShureD, GregoratosG, Moser KM. Fiberoptic angioscopy:role in the diagnosisof chronic pulmonaryarterial obstruction.Ann Intern Med 1985;103:844-850. 6. BrundageBH, Rich S, SpigosD. Computedtomographyof the heart and great vessels:presentand future. Ann Intern Med 1984;101:801-809. 9. FischerMR. Higgins CB. Central thrombi in pulmonary arterial hypertension detectedby MR imaging. Radiology 1986;159:223-226. 10. Yock PG, JohnsonEL, Linker DT. Intravascular ultrasound: development and clinical potential. Am J Cardiac Imaging 1988;2:525-536. 11. Yock PG, Linker DT, AngelsenBAJ. Two dimensionalintravascular ultrasound:technicaldevelopmentandinitial clinical experience..I Am Sot Echocurdiography 1989;2:296-304. 12. Nishimura RA, Edwards WD, WarnesCA, ReederGS, Holmes DR, Taiik JA, Yock PG. Intravascular ultrasoundimaging:in vitro validation and pathologic correlation. J Am CONCardiol 1990;16:145-154. 13. NissenSE, Grines CL, Gurley JC, Sublett K, Haynie D, Diaz C, BootheDC, DeMaria AN. Application of a new phased-arrayultrasoundimagingcatheter in assessment of vasculardimensions:in viva comparisonto cineangiography.Circulation 1990;81:660-666. 14. Tobis JM, Mallery JA, GessertJ, Griffith J, Mahon D, BessenM, Moriuchi M, McLeay L, McRae M, Henry W. Intravascular ultrasound cross-sectional arterial imaging before and after balloon angioplasty in vitro. Circulation 1989;80:873-882. 15. Pandian N, Weintraub A, Kreis A, Schwartz S, Konstam M, Salem D. Intracardiac, intravascular, two-dimensional,high-frequencyultrasoundimaging of pulmonary artery and its branches in human and animals. Circulation 1990;81:2007-2012. 16. Nicod P, PetersonKL, Levine M, Dittrich H, Buchbinder M, Chappuis F, Moser K. Pulmonaryangiographyin severechronic pulmonaryhypertension.Ann Intern Med 1987;107:565-568. 17. Daily PO, Dembitsky WP, PetersonKL, Moser KM. Modification of techniquesand early resultsof pulmonarythromboendarterectomyfor chronic pulmonary embolism.J Thorac Cardiooasc Surg 1987;93:221-233. 16. Peterson KL, Fred HL, Alexander JK. Pulmonary arterial webs: a new angiographicsignof previousthromboembolism.N Engl J Med 1967;277:33-35.