Ultrasonic-pathologic comparison of postangioplasty myointimal hyperplasia and primary atheroma of the superficial femoral artery

Copyright

Ultrasound in Med. & Biol., Vol. 22. No. 7. pp. 815-821. 1996 0 1996 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/96 $15.00 + .oO

PII: SO301-5629( 96)00101-9

ELSEVIER

@Original Contribution ULTRASONIC-PATHOLOGIC COMPARISON OF POSTANGIOPLASTY MYOINTIMAL HYPERPLASIA AND PRIMARY ATHEROMA OF THE SUPERFICIAL FEMORAL ARTERY IRIS BAUMGARTNER,

* FALAH REDHA, RALF WERNER DO-DAI Do and FELIX MAHLER

BAUMGARTNER,

Department of Medicine, Division of Angiology and Department of Neurology, University Hospital, Bern, Switzerland; and *Research Division, Department of Surgery, University Hospital, Zurich, Switzerland (Received

2 October

1995; in final

form

26 March

1996)

Abstract-The purpose of this study was to characterize postangloplasty myointhnal hyperplssia as compared to primary atheroma of superficial femoral arteries using color-coded duplex sonography (CCD), and to correlate sonographic findings with the histopathology of samples obtained from these lesions by catheter atherectomy (Redha-cut device). Preinterventionally, homogeneity, echogeneity, and the surface of plaques were described using CCD in nine cases with secondary stenoses after percutaneous transhuninal angloplasty and in seven cases with primary atheroma. Myointlmal hyperplasia of femoral restenoses showed a homogeneous (7 of 9 vs. 1 of 7) and hypoechogenlc (7 of 9 vs. 0 of 7) wall thickening compared to primary atheromas (p < 0.05). Primary atherosclerotic plaques showed a rather heterogeneous, hypoand hyperechogenic ultrasonic appearance with or without echo shadowing in slx of seven cases. The surface of restenoses was more often regular than that of primary atherosclerotic lesions, but thls Iindlng dld not reach statistical signlticance (6 of 9 vs. 2 of 7, p = 0.14). Thrombotic material appeared homogeneousand hypoechogenlcin three of five casesand could not be discriminated from intimal hyperplasia. In summary, postangloplasty lntimal hyperplasia is characterlsed by a hypoechogenic, homogeneous,rather regularly conilned vesselwall thickening and can be differentiated from primary atheroma at CCD. Key Words: Arteries, Transluminal angloplasty, Atherectomy, Femoral artery stenosis,Neointima, Ultra-

sound, Duplex scanning, Ultrasound tissue

characterization.

INTRODUCTION

Neointimal hyperplasia continues to be a major problem following percutaneous translmninal angioplasty (PTA) of peripheral arterial lesions. Restenosis of the treated arterial segment is believed to result from either myointimal cellular proliferation, thrombosis, or elastic recoil, or a combination of these three (Chesebro et al. 1987; Faxon et al. 1987). Although restenosis correlates well with recurrence of symptoms, these two features are not identical. Symptoms may correspond to restenosis at the treated vessel segment or to a progression of arteriosclerosis with relevant cross-sectional narrowing remote from the treated segment. In this context, ultrasound imaging combined with color Doppler flow measurements is an attractive completion to the conventional noninvasive follow-up evaluation, Address for correspondence: of Medicine, Division of Angiology, CH-3010 Bern, Switzerland.

Dr. I. Baumgartner, Department University Hospital, Inselspital,

because it is able to discriminate restenosis of the treated arterial segment from that distant from the intervention site (Henderson et al. 1994; Mewissen et al. 1992; Sacks et al. 1990). Moreover, it has been demonstrated that, by using sonographic appearance, it may be possible to characterize tissue with regard to histopathology of primary plaques from carotid arteries (O’Donnell et al. 1985; Reilly et al. 1983; Weinberger et al. 1987; Widder et al. 1990). So far, there are no results on ultrasonic features of secondary postinterventional stenoses attributed mostly to myointimal hyperplasia. Only by the possibility of noninvasive distinction of atheroma or hyperplasia and monitoring of progress of such changes, research on the evolution of peripheral arterial occlusive disease and the influence of reconstructive and pharmacologic interventions is possible. The purpose of this study was to characterize postangioplasty myointimal hyperplasia compared to primary atheroma of superficial femoral arteries using CCD, and

816

Ultrasound in Medicine and Biology

to correlate sonographic findings with histopathology of plaque specimens obtained from these lesions by atherectomy (Johnson et al. 1990; Redha et al. 1992). MATERIALS

AND METHODS

Patients Twenty-one patients with stenoses of the superficial femoral artery were consecutively recruited for this prospective correlation study during a period of 6 months. There were 15 patients, in whom tissue could be removed as a sufficient block by catheter atherectomy to evaluate plaque histopathology and correlate it with preinterventional ultrasonography. CCD of eligible patients was performed within 2 weeks before catheter atherectomy. There were seven superficial femoral arteries in six patients with primary atherosclerotic lesions, without earlier angioplasty or vascular surgery. Nine superficial femoral arteries in nine patients showed a recurrent lesion within 12 months after previous PTA (range 3 - 11 months). Demographic data for the patient population are given in Table 1. Ultrasound examination Doppler flow measurement and B-mode imaging were performed with commercially available high-resolution ultrasound equipment ( Acuson 128, XP 5, Mountain View, CA, USA) with the use of 5or ~-MHZ linear-array transducers. All examinations were performed by one investigator experienced in ultrasound and color Doppler flow imaging techniques, blinded for histopathology (I.B.) . Parameters used to determine and find relevant stenoses were the peak intrastenotic systolic and diastolic velocities calculated by the Doppler equation and the ratio of peak systolic intra- and prestenotic velocities (JSiger et al. 1985; Ranke et al. 1992). In two cases with multifocal stenoses of the superficial femoral artery, the most proximal lesion suitable for catheter atherectomy was analyzed. High-resolution images of the region-of-interest were recorded in multiple longitudinal and transverse scans. Plaques, defined

Table 1. Demographic

data of study population.

Characteristic

Primary atheroma (n = 7)

Recurrent stenosis (?I = 9)

Age (Y) Gender (M/F) Creatinine (~mol/L) Hypertension Diabetes Smoking history Hypercholesterolemia

69 + 12 413 85 2 14 5 1 4 3

72 + 5 41.5 91 2 15 4 2 3 6

Volume 22, Number 7, 1996

as intima-media thickening above 1.5 mm, were analyzed with delimitation of the flow lumen using color-coded duplex sonography. Plaques were described regarding homogeneity, echogeneity, and endoluminal surface irregularities. Echogeneity was classified into three categories: hypoechogenic; moderately echogenic; and brightly echogenic with or without shadowing. Homogeneity was classified as homogeneous or heterogeneous. A heterogeneous lesion was defined as a mixture of areas with different echogeneity within a plaque. A plaque was described as homogeneous if there was a more than 1.5 mm of endoluminal wall thickening of similar echogeneity. The endoluminal surface of a plaque was described as regular or irregular in good-quality longitudinal and transverse images, with a supplementary category for nonvisible, nondefined surface characteristics. Percutaneous catheter atherectomy The atherectomy device (Redha-cut device, Sherine Medical AG, Uettligen, Switzerland) was designed for the percutaneous transluminal resection of atheromatous material in segmental arterial stenoses. The device is a flexible catheter made of surgical-grade stainless steel carrying a hollow blunttipped cylinder at the end. It covers two cutting blades which, by moving the cylinder mechanically back and forth, can be opened and closed like an umbrella. The cutting edges are bent slightly inward to avoid perforation of the vessel walls while cutting. The edge-to-edge distance is 5 mm when the device is open, and 2.65 mm when it is closed. The device is introduced through a hemostatic sheath (8F), and advanced through the stenosis either alone or over a guidewire (0.012 in. [0.3 mm]). By withdrawing the catheter with open blades through the stenosis, pieces of the occluding material were cut away, retained by the device, and brought out of the body. Atherectomized tissue was determined as sufficient for histopathology when there was a minimum of one sample more than 3 mm in length. Histopathologic examination Tissue samples were carefully removed from the collection area of the device and fixed in 4% buffered formalin. The samples were photographed macroscopitally to document the size of the specimens (Fig. 1) . Tissues were processed routinely for histology study and embedded in paraffin. Tissue sections (5 pm thick) were cut and stained with hematoxylin-eosin, elastic van Gieson’s, and Masson’s trichrome. Von Kossa’s stain for calcium was also used in some cases. The histologic slides were examined under a microscope

Ultrasonic-pathologic

comparison 0 I. BAUMGARTNER et al.

817

Fig. 1. Myxoid, white-gray tissueatherectornizedfrom a femoral postangioplastyrestenosiswith a bright brown atheroscleroticplaqueportion in one of the three tissuesamples(white arrow). by one investigator (F.R.), who was blinded to the clinical characteristics or the ultrasonographic findings of the patients. Cellularity, type and organization of collagen fibers, tbrombotic material and calcification of the tissue sampleswere analyzed semiquantitatively.

groups, the differences in mean values were calculated with use of a nonparametric test (Mann-Whitney U test). Statistical tests were two-sided, with p < 0.05 as the criterion to indicate statistical significance.

Statistical analysis

Gross pathology

Statistical analysis was carried out with the Systat software package (Evanston, IL, USA). Between

The resected specimensconsisted of concave tissuefragments. On average, two to four specimenswere

RESULTS

Fig. 2. Color-codedduplex sonographyfrom a primary atheromawith macrocalcificationsshowinga brightly echogenicareawith an irregularsurface echogenicareawith shadowing(white arrow) anda small heterogeneous, adjacent (black arrow).

818

Ultrasound in Medicine and Biology

Table 2. Histopathology of nine postangioplasty restenoses and seven primary atheromas excised by catheter atherectomy from superficial femoral arteries.

Histopatbology

Postangioplasty restenosis (n)

Primary atberoma (4

Hypercellular Mixture of hyper-/hypocellular Hypocellular Thrombotic material Microcalcification Macrocalcification

5 4 0 3 3 0

0 3 4 2 1 4

P” O.Olb 0.02’ 0.84 0.39 0.01

’ Difference between both groups (Mann-Whitney U test). b Hypercellular and hyper-Arypocellular tissue samples. ’ Hypocellular and hyper-fiypocellular tissue samples.

removed from each of the 16 atherectomized lesions. The gross morphology of samples ranged from translucent shreds of tan-white tissue to opaque white-yellow fragments. Gritty nodules representing calcified areas were seen within specimens. Samples removed from sites of restenosis showed white-gray parts of tissue, in some cases combined with white-yellow to bright brown atherosclerotic plaque portions (Fig. 1) . Specimens removed from restenoses frequently appeared to have two layers, with white-gray myxoid tissue overlying a white-yellow atherosclerotic plaque. Microscopic pathology Semiquantitative results of histopathology fi-om postangioplasty restenoses and primary atherosclerotic stene ses are given in Table 2. Hypercellular, myointimal proliferation within a loose connective tissue matrix with sparse collagen fibrils was observed in all nine samples of postangioplasty restenoses (Fig. 2). Four tissue samples from restenoses showed a mixture of hyper- and hy=pocellularity

Volume 22, Number 7, 1996

due to myointimal neointima as a discrete layer overlying the residual, hypocellular atherosclerotic plaque. In the seven cases with primary atheroma, hypocellular dense fibrous or calcified plaques were seen, three of which were incompletely covered by a small hypercellular layer. Myointimal hyperplasia and primary atheroma were significantly different in cellularity and macrocalcifications (p < 0.05). Ultrasonic-pathologic comparison Comparison of postangioplasty myointimal hyperplasia and primary atheroma with hi&pathologic findings are summarized in Tables 3 and 4. Sonographically, postangioplasty restenosesdiffered from primaty atheroma by showing significantly more homogeneous (7 of 9 vs. 1 of 7, p = 0.015) and hypoechogenic (7 of 9 vs. 0 of 7, p = 0.003) endoluminal wall thickenings (Fig. 3). The CCD image allowed for sufficient resolution of the endoluminal surface in all nine restenoses, 6 of them regular, whereas 2 primary atheroma cases showed a regular endoluminal surface (6 of 9 vs. 2 of 7, p = 0.14). Ultrasonic patterns of primary atherosclerotic plaque samples showed no uniform appearance. Shadowing and reverbemtions were seen frequently. Complex lesions with portions of thrombotic material could not be differentiated in two of the three myointimal restenoses and in both instances of primary atheroma. DISCUSSION The present ultrasonic-pathologic correlation study demonstrates that CCD provides a reliable differentiation of the postangioplasty myointimal hyperplasia from primary atheroma in the superficial femoral artery. Although there are difficulties in interpreting results from a semiquantitative rather than a true quan-

Table 3. Comparison of histopathology and color-coded duplex (CCD) findings in nine postangioplasty restenoses of the superticial femoral artery. (n) 4 1 1 1 1 1

Histopatbology Hypercellular (myointimal + hypocellular portion, + hypocellular portion, + thrombotic material + hypocellular portion, + hypocellular portion, microcalcification

hyperplasia) microcalcification thrombotic material microcalcification tbrombotic material,

Ultrasonic findings Homogeneous, hypoechoge& I Heterogeneous, moderate echogenic Heterogeneous, mainly hypoechogenic

CCD allowed for sufficient resolution of the endohuninal surface in all nine restenoses, six of them regular and three irregular. The surfaces of those four restenoses with exclusively hypercellular findings at histopathologic assessment were judged regular at CCD. ’ Predominately hypercellular, myointimal proliferation was observed in all nine samples of postangioplasty restenoses, but some hypocellular portions and/or thrombotic material were present in five additional samples. b More than 1.5 mm of thickness.

Ultrasonic-pathologic

comparison 0 I. BAUMGARTNER

et al.

819

Table 4. Comparisonof histopathologyand color-codedduplex (CCD) findings

in sevenprimary atheromasof the superficialfemoral artery. (n)

Histopathology

Ultrasonic findings

3 1

Hypocellular” + macrocalcification + macrocalcification, thrombotic material

1 1 1

HyperJhypocellular mixtureb + microcalcification + thrombotic material

Heterogeneous, hyperechogenic I Heterogeneous, moderate echogenic I

Homogeneous, moderate echogenic

CCD allowed a sufficientresolution of theendoluminal surfacein four primaryatheromas, two regular and two irregular, respectively. a Four atherectomy samples of primary atheroma were predominately hypocellular with additional macrocalcifications in three and thrombotic material in one case, respectively. b A mixture of hypo- and hypercellular portions with additional microcalcifications and thrombotic material in one case, respectively, was seen in three atherectomy samples of primary atheroma.

titative method, myointimal hyperplasia could be discriminated from primary atherosclerotic plaques by its homogeneous,hypoechogenic ultrasonic appearancein seven of nine casescompared to one of seven cases in primary atheroma (p < 0.05). The endoluminal surface of postangioplasty restenoseswas more often regularly bordered compared to primary atberosclerotic stenoses,but this difference was not significant (6 of 9 vs. 2 of 7, p = 0.14). Thrombotic material associated with primary atheroma or postangioplasty restenosis was described as homogeneous and hypoechogenic in three out of the five casesand could not be differentiated from myointimal hyperplasia. Because imaging quality beyond the Hunter’s canal often becomes insufficient, the study was limited to the superficial femoral artery. Our investigation by high-resolution CCD imaging provided a direct and semiquantitative assessmentof peripheral vessel wall characteristics and local hemodynamic alterations. Multiplane examinations were mandatory both in longitudinal and cross-sections,becausethe stenoseswere frequently eccentric or irregular and difficult to record in longitudinal view only. Moreover, with increasing stagesof arteriosclerosis, the quality of the ultrasound image was poor becauseof shadowing and reverberation (Weinberger et al. 1987; Widder et al. 1990). High-density reflections produced echo shadowseither due to compact layers consisting of condensed fibrous material or of microcalcifications, not distinguishable by the ultrasound image. By comparison, the postangioplasty intimal hyperplasia represented a homogeneous, hypoechogenic layer of more than 1.5mm thickness, often regularly narrowing the vessel lumen. These ultrasonic findings of postangioplasty restenoses seem independent of the time of appearance, at least within the first 12 months after intervention. Hyperechogenic areas superimposed by the homogeneous, hypoechogenic layer corresponded histologically to

calcified and fibroatheromatous lesions compressed and fragmented by previous angioplasty and covered by neointima. The ultrasonic appearance of a neointima could be confused with thrombotic material. While the tbrombus has a tendency to bulge into the lumen to produce a crescent-shapedor slit-like surface, myointimal hyperplasia tends to form a regular, elongated stenosis (Hennerici et al. 1988). However, the small number of caseswith tbrombotic portions within the atherectomized tissue samplemakesa valid conclusion about ultrasonic findings of hypercellular, myointimal hyperplasia compared to thrombotic material impossible. This study does not provide a complete pathologic description of primary atheroscleroticstenosesand restenoses,becauseonly strip samplesof arterial wall, not completecrosssectionsof arteries,were evaluated (Johnsonet al. 1990). This limits our ability to analyze certain featuressuch as the precisequantity of thrombus and the distribution of myointimal hyperplasia around the lumen. Even the description of the surface of a plaque has no strongcorrelation with the incomplete atherectomy specimens, and whether myointimal hyperplasia shows tendency to form a more regular, endoluminal surfaceborder than primary atheroma remainsto be proved. There are some other limitations which should be mentioned to assess this correlation study. The CCD imaging provides no quantitative assessment of plaque characteristics;it is rather a subjective description of different echogeneity and homogeneity. Correspondingly, sonographerperformance representsa critical component of measurement assessment and reproducibility. Therefore, with the concrete question of myointimal hyperplasia, this study of CCD is hardly useful outside the femoral artery, although other artery regionsmay have beenthe site of an intervention. CCD doesnot replacephysiologic noninvasive studiessuch asankle-bra&al index measurementor oscillometric recording standardly usedto follow-up percutane-

820

Ultrasound in Medicine and Biology

Fig.

Volume 22, Number 7, 1996

Jltrasonic and histopathologic appearance of a femoral postangioplasty intimal hyperplasia. (Tc )p) Horn lypoechogenic layer with regular endoluminal surface border (CCD) (Bottom) Histopatholog ;ically h; cellular neointima (H&E.).

{fir-

Ultrasonic-pathologic

comparison 0 I. BAUMCARTNER

ous catheter interventions. However, to discriminate local restenoses, CCD or angiography are mandatory and the discrimination of myointimal hyperplasia from atherosclerotic progression may have advantages for further treatment decisions (Treiman et al. 1994). In conclusion, our results indicate that the ultrasonic finding of homogeneous, hypoechogenic lesions indicates myointimal hyperplasia involved in restenoses. It may be argued that our patient population is too small to draw conclusions on the basis of these results, but pathophysiologic reflection and comparison with B-mode findings from other hypercellular tissues, e.g., lymphoma, fit well with the present results. The ability of CCD to discriminate myointimaI hyperplasia from sclerotic lesions at various times during follow-up may render this method a useful tool in secondary prevention research. REFERENCES Chesebro JH, Lam JYT, Badimon L, Fuster V. Restenosis after arterial angioplasty: a hemorrheologic response to injury. Am J Cardiol 1987;60:10B-16B. Faxon DP, Sanbom TA, Haudenschild CC. Mechanism of angioplasty and its relation to restenosis. Am J Cardiol 1987;605B-9B. Henderson J, Chambers J, Jeddy TA, Chamberlain J, Whittingham TA. Serial investigation of balloon angioplasty induced changes in the superhcial femoral artery using colour duplex ultrasonography. Br J Radio1 1994;67:546-551. Hennerici M, Steinke W. Three-dimensional ultrasound imaging for the evaluation of progression and regression of carotid atherosclerosis. Workshop. Gtitersloh: Karger Basel: 1988: 115 - 132. Jlger K, Ricketts HJ, Strandness DE. Duplex scanning for the evalu-

et al.

821

ation of lower limb arterial disease. In: Bernstein EF, ed. Noninvasive diagnostic techniques in vascular disease. St. Louis, MO: Mosby, 1985:619-631. Johnson DE, Hinohara T, Sehnon MR, Braden LJ, Simpson JB. Primary peripheral arterial stenoses and restenoses excised by transluminal atherectomy: a histopathologic study. J Am Co11 Cardiol 1990; 15:419-425. Mewissen MW, Kinney EV, Bandyk DF, Reifsnyder T, Seabrook GR. The role of duplex scanning versus angiography in predicting outcome after balloon angioplasty in the femoropopliteal artery. J Vast Surg 1992; 15:860-866. O’Donnell TF Jr, Erdoes L, Mackey WC, McCullough J, Shepard A. Correlation of B-mode ultrasound imaging and arteriography with pathologic findings at carotid endarterectomy. Arch Surg 1985; 120:443-449. Ranke C, Creutzig A, Alexander K. Duplex scanning of the peripheral arteries: correlation of the peak velocity ratio with angiographic diameter reduction. Ultrasound Med Biol 1992; 18:433440. Redha F, Uhlschmid GK, Antonucci F, Zollikofer CL. New device for peripheral atherectomy: experimental results. Radiology 1992;185(suppl):229. Reilly LM, Lusby RJ, Hughes L, Ferrell LD, Stoney RJ, Ehrenfeld WK. Carotid plaque histology using real-time ultrasonography. Clinical and therapeutic implications. Am J Surg 1983; 146:188193. Sacks D, Robinson ML, Marinelli DL, Perlmutter GS. Evaluation of the peripheral arteries with duplex US after angioplasty. Radiology 1990; 176:39-44. Treiman GS, Ichikawa L, Treiman RL, Cohen JI, Cossman DV. Treatment of recurrent femoral or popliteal artery stenosis after percutaneous transluminal angioplasty. J Vast Surg 1994; 20~577-587. Weinberger J, Marks SJ, Gaul JJ, Goldman B, Schanzer H. Atherosclerotic plaque at the carotid artery bifurcation. J Ultrasound Med 1987;6:363-366. Widder B, Paulat K, Hackspacher J, Hamann H, Hutschenreiter S. Morphological characterization of carotid artery stenoses by ultrasound duplex scanning. Ultrasound Med Biol 1990; 16:349354.