Eur J VascSurg 5, 527-534(1991)

Doppler Colour Flow Imaging After Carotid Endarterectomy Wolfgang Steinke, Michael Hennerici, Christof Kloetzsch and Wilhelm Sandmann 1 Departments of Neurology and 1Vascular Surgery, Heinrich-Heine University, Didsseldorf, Germany Sixty-six patients (72 carotid arteries) were examined after carotid endarterectomy (CEA) using Doppler colour flow imaging (DCFI). Examinations were performed 4-18 days (mean: 7 days) after surgery (32 patients, 34 arteries) or between 2 and 100 months (mean: 39 months) after CEA (34 patients, 38 arteries). Minor vessel wall abnormalities werefound in 36 % in the internal carotid artery (ICA) and in 55 % in the common carotid artery (CCA) or bifurcation. One patient had a minor residual ICA stenosis after surgery; two low-grade stenoses and three ICA-occlusions were diagnosed at follow-up. Altered flow patterns occurred most in CCA (90 %)and were predominantly located adjacent to the vessel wail and in dilated vascular segments of the CCA. Disturbed haemodynamics in the ICA were less marked (57%) and frequently found in the central vessel lumen or diffusely distributed. We conclude that surgically induced changes of the vessel geometry and residual or recurrent vascular wall abnormalities are often associated with distinct haemodynamic disturbances, which can reliably be detected by DCFI. Key Words: Carotid endarterectomy; Ultrasound; Doppler colour flow imaging.

Introduction Intra-operative assessment of the results of carotid endarterectomy (CEA) is essential to identify and exclude technical problems. The methods routinely used by many surgeons include angiography or noninvasive ultrasound. 1-7 Due to its inherent risks, postoperative intra-arterial angiography has rarely been used for regular follow-up examinations after CEA, 8-1° and in only a few studies have histopathological changes in endarterectomised vessels been studied in detail. 11-1s Furthermore, n u m e r o u s p o s t operative investigations after CEA, using various ultrasound methods, have focused on the incidence of significant recurrent obstruction as revealed by haemodynamic Doppler measurements. 16-23 Again morphological changes in the operated artery have been examined less extensively, 23-26 and for technical reasons, the direct interaction of postoperative vessel morphology and adjacent haemodynamics has not been analysed so far. Since the latter is supposed to provide important information about the long-term prognosis of surgically induced vascular alterations, our knowledge about the pathophysiology of early and late carotid obstructions is still very limited. The n e w ultrasound technique using Doppler colour flow imaging (DCFI) provides a simultaneous Please address all correspondence to: M. Hennerici MD, Dept. of Neurology, University of Heidelberg, Theodor-Kutzer-Ufer, D-6800Mannheim 1, Germany. 0950-821X/91/050527+08$03.00/0© 1991Grune & Stratton Ltd.

real-time display of both the spatial and temporal distribution of blood flow, which may be superimposed on a high-resolution B-mode echotomogram of tissue and vessel.wall structures. 27-31 The aim of the present study was to evaluate the usefulness of this technique in the assessment of local morphological changes and blood flow disturbances due to the altered vessel geometry immediately after CEA. The results are discussed with a view to the performance of future prospective long-term trials and the investigation of prognostic factors or the effects of drugs in the prevention of recurrent stenoses.

Patients and Methods All patients undergoing CEA in our institution were subjected to postoperative conventional Doppler studies some days after surgery and at routine followup examinations. Out of these patients, 50 men and 16 w o m e n with 72 CEAs were selected randomly and were examined 4-18 days (mean 7 days) after surgery (32 patients, 34 arteries) or at follow-up intervals of 2100 months (mean 39 months) after CEA (34 patients, 38 arteries). All patients had undergone a standard CEA performed under general endotracheal anesthetic. Autogenous saphenous vein patch angioplasty was carried out in 61 (85%) vessels, Dacron was used in seven (10%) and in four (5%) the arteries were closed

528

W. Steinke et aL

primarily. The intra-operative carotid flow velocity pattern was measured after restoration of the blood flow, using a pulsed Doppler system with the transducer placed directly on the external surface of the internal carotid artery (ICA) distal to the arteriotomy. In all cases, this test showed a good operative result as assessed by the pertubation index, s Two patients had suffered a minor stroke during the operation and two other patients developed a stroke ipsilateral to the previously performed endarterectomy during the follow-up period. All patients had a neurological examination, and by means of conventional continuous-wave Doppler sonography (MDV Delalande, D 480/K, Rauenberg, Germany; operating frequency 4MHz) and duplex-system analysis (DRF 400, Diasonics, Milpitas CA) the extracranial arteries supplying the brain were studied. DCFI examinations were performed with a Quantum angiodynograph (Philips QAD, Hamburg, Germany). The system uses a 7.5 MHz linear transducer for simultaneous display of a grey scale tissue echotomogram (B-mode) and the Doppler signal, which is colour-coded as a function of intravascular flow direction. Blood cell morion away from the transducer is coded in red, and towards the transducer in blue. The degree of colour saturation indicates the velocity of moving targets. Multiple longitudinal- and cross-sections in the neck were assessed with the transducer in the anterior- and posteriorlateral positions. The total examination procedure was stored on a digital video-tape for subsequent offline evaluation of the vessel morphology with special regard to local plaques, stenoses and thrombus formarion, as well as for the measurement of the maxim u m systolic diameters in the common carotid artery (CCA), the proximal 1-2 cm segment of the ICA and the bifurcation. Haemodynamic disturbances during the cardiac cycle were analysed with regard to the degree (minor, moderate, severe), location (central, adjacent to the vessel wall, diffuse) and flow direction (orthograde, retrograde). Semi-quantitative analysis of the degree of flow alteration included the spatial and temporal extent of flow reversal zones and mixed turbulence. Image quality was classified separately for the CCA, the bifurcation and the ICA as technically good if the normal and pathological vascular morphology as well as the colour-coded blood flow could be visualised adequately. Further technical details of the DCFI system and results of studies performed in normal and diseased carotid arteries are reported elsewhere. 27-31

Results

Performance Image quality was good or fair in all CCAs and bifurcations apart from one (1.4%) artery in which a Dacron patch had been used. Display was poor in seven (9.7%) ICAs, two of which were also Dacron patch angioplasties. Although the B-mode image of the vessel wall structures and the colour-coded flow signal were technically unsatisfactory in these arteries, normal single-gate pulsed Doppler spectra could be assessed at various sites within the vessel, thus excluding significant obstruction. The poor image quality was due to the inability to hyperextend the neck, high location of the bifurcation, postoperative swelling or subcutaneous haematoma and the large size of the transducer. On the other hand, in four cases with questionable results from conventional Doppler and duplex-analysis, the DCFI of the endarterectomised vessels was straightforward and markedly improved the diagnostic reliability.

Morphology Structural irregularities were common, only 44% of the CCA and 69% of the ICA with technically satisfactory studies showed normal wall surfaces (Table 1). Various types of pathological echotomograms could be distinguished. 1. Smooth, echointensive structures Table 1. Echomorphologic findings after carotid endarterectomy in the common carotid artery (CCA), bifurcation (BIF) and internal carotid artery (ICA) with technically satisfactory DCFI

CCA/BIF

ICA

Number

71

65

Normal

31 (44%)

45 (69%)

Local vessel wall abnormalities small, echointensive, smooth

31 (44%)

18 (28%)

heterogeneous, irregular surface

5 (7%)

4 (6%)

tissue-isodense, smooth surface

3 (4%)

1 (2%)

Thrombus formation

6 (8%)

0

Aneurysmatic dilatation

3 (4%)

0

Stenosis (>40%)

0

3 (5%)

Occlusion

0

3 (5%)

Total percentage exceeds 100%, since more than one morphologic abnormality was detected in some vessels. Eur J VascSurg Vol5, October1991

Fig. 1. Proximal end of the endarterectomy in the CCA. (a) B-mode echotomogram shows small echointensive plaques at the proximal end of the endarterectomy at the anterior and posterior vessel wall (arrows) 24 months after endarterectomy. (b) DCFI displays flow reversal close to the fibrous plaques in early systole. (c) Extension of flow separation into the centre of the vessel during diastole.

Fig. 2. Homogeneous smooth structure, probably representing residual media, at the posterior wall of the CCA 4 days after enclarterectomy (arrows). (a) Normal blood flow pattern during systole. (b) Proximal and distal turbulences (blue colour) during diastole. (c) Crosssection showing residual vascular wall structures (arrows) and associated turbulences.

Fig. 3. (a) B-mode display of a heterogeneous irregular plaque (arrows) at the posterior wall of the CCA 47 months after endartereetomy. (b) DCFI shows turbulences (blue colour) distal to the plaque. (c) Cross-section of the plaque (arrows). JV: jugular vein. Eur J Vasc Surg Vol 5, October 1991

530

W. Steinke e t a L

Fig. 4. Residual minor ICA stenosis 5 days after endarterectomy. (a) B-mode image shows a heterogeneous plaque at the posterior wall of the ICA (arrows). (b) Colour fading indicates increased flow velocity and minor turbulences due to the irregular surface of the plaque.

Fig. 6. DCFI of a carotid birfurcation 4 days after carotid endarterectomy. (a) Flow reversal (blue colour) in the proximal ICA. ECA stenosis, showing colour fading and post-stenotic mixed-coloured turbulences during systole. (b) Very slow blood flow in the ICA not detectable by DCFI (black vessel lumen) during diastole, persistent increased flow velocity in the ECA.

Fig. 5. DCFI of thrombus formations. (a) Large echolucent thrombus (arrows) at the posterior wall of the CCA 5 days after endarterectomy. (b) Corresponding cross-section. (c) Thrombus adherent to the anterior vessel wall with a smooth surface 16 months after surgery. Minor flow disturbances distal to the thrombus in the bifurcation. (d) Corresponding cross-section. Residual patent vessel lumen with colour-coded Doppler signal. Arrows indicate vascular wall.

Fig. 7. DCFI characteristics of flow disturbances after endarterectomy. (a) Flow reversal and separation (blue colour) near the vascular wall during systole. (b) Extension of flow separation into the centre of the vessel and to the opposite wall during diastole. (c) Minor turbulences are found at the origin of the ICA. (d) Severe diffuse flow disturbances in the CCA [same patient as in (c)].

w e r e f r e q u e n t l y o b s e r v e d i n t h e C C A (44%): m o s t o f these were located at the bifurcation, however 8% w e r e f o u n d a t t h e p r e s u m e d p r o x i m a l e n d of t h e e n d a r t e r e c t o m y (Fig. 1). C o r r e s p o n d i n g m o r p h o l o g i c a l f i n d i n g s w e r e s e e n i n t h e I C A i n 28% w i t h a p r e d o m i n a n t l o c a t i o n a t t h e d i s t a l e n d of t h e e n d a r -

t e r e c t o m y i n 12%. 2. S m o o t h , h o m o g e n e o u s , t i s s u e i s o d e n s e p l a q u e s w i t h a n e c h o g e n i c i t y s i m i l a r to normal vessel wall tissue were found in only four endarterectomised vessel segments a few days after s u r g e r y (Fig. 2). 3. H e t e r o g e n e o u s p l a q u e s w i t h a n i r r e g u l a r s u r f a c e o c c u r r e d in 7 % of t h e C C A a n d in

Eur J Vasc Surg Vol 5, October 1991

Doppler Colour Flow Imaging

6% of the ICA (Fig. 3). One patient had a minor residual ICA stenosis after the endarterectomy and two low grade ICA stenoses were found due to heterogeneous plaques at the origin of the ICA at followup examinations (Fig. 4). 4. Large thrombus formations were detected in the CCA only (8%) (Fig. 5). They were characterised by a heterogeneous mass of low echodensity with a smooth surface and were adherent to the vessel wall of the CCA extending into the bifurcation. The intravascular thrornbus formation was always associated with a dilation of the artery, thus the patent vessel lumen never decreased below presumed preoperative diameters. Dilatation after CEA was common, especially when the arteriotomy had been closed by using a patch angioplasty. The mean maximum diameter was wider in all measured segments of the arteries studied a few days after surgery compared with those examined at follow-up, however differences were not significant. Dilatation was considered to be aneurysmal (vessel diameter >18mm) in three (4%) of the CCAs. Two of these had had venous patch angioplasty and one had been closed primarily. In another three arteries, a pseudoaneurysmal niche was found at the proximal end of the endarterectomy, probably due to a very deep plaque removal (not included in Table 1). In addition, three total ICA occlusions were diagnosed at follow-up. One of these was bilateral in a patient with a Dacron patch angioplasty on both sides. External carotid artery (ECA) stenosis was found in six cases (Fig. 6) and occlusion of the ECA occurred twice. Findings in the ECA are not included in Table 1.

Haemodynamics Flow patterns showed minor or moderate alterations in most ICAs compared with the severe disturbances in most CCAs (Table 2). In the ICA, blood flow patterns were normal in 43%, but in the CCA only 10% demonstrated undisturbed blood flow. During systole, the flow separation was located typically near the vessel wall at the proximal site of the endarterectomy in the CCA extending into the centre of the vessel during diastole (70%) (Fig. 7). CCA flow separation was predominantly central in only three (4%) arteries and diffusely distributed turbulence occurred in 11 (16%) CCAs. Even in the presence of severe flow abnormalities in the CCA and the bifurcation, only minor or moderate haemodynamic alterations were found a short distance above the bifurcation in

531

Table 2. Colour-coded blood flow alterations after carotid endarterectomy in the common carotid artery (CCA), bifurcation (BIF) and internal carotid artery (ICA) with adequate visualisation in DCFI

Number

CCA/BIF

ICA

71

62*

Undisturbed blood flow

7 (10%)

27 (43%)

Degree of flowalteration minor

13 (18%)

19 (31%)

moderate

24 (34%)

10 (16%)

severe

27 (38%)

6 (10%)

50 (70%)

9 (15%)

central

3 (4%)

12 (19%)

diffuse

11 (16%)

14 (23%)

Locationof disturbance vessel wall and central

* Three ICA occlusionsnot induded.

the ICA. If flow disturbances occurred in the ICA, they were more frequently in the central vessel lumen (19%) or distributed diffusely (23%) rather than near the vessel wall (15%) (Table 2).

Morphologic-haemodynamicinteraction The degree of haemodynamic alterations correlated with the vessel diameter in the endarterectomised segment: flow reversal and separation were more pronounced in dilated arteries and were most severe in three cases with pseudoaneurysmal dilatation. If CEA had extended far distally to the bifurcation; marked reversed flow in the ICA was a persistent feature. Haemodynamic alterations were rarely associated with small, echointensive vessel wall irregularities, but in six patients flow separation occurred adjacent to these small plaques at the proximal end of the endarterectomy (Fig. 1). Turbulence was associated with five of nine heterogeneous, irregular plaques (Fig. 3) and with two of four tissue-isodense, smooth, wall structures (Fig. 2). Surprisingly, no or only minor flow changes were found in six arteries with large thrombus formations adherent to the vascular wall in the CCA, probably due to the smooth surface and the normal residual lumen (Fig. 5). The three low-grade ICA stenoses were characterised by long-segment colour fading during systole with minor additional turbulence at the surface of the plaque (Fig. 4). Eur J VascSurgVol5, October1991

532

W. Steinke et aL

Discussion

This report summarises morphological and haemodynamic changes after CEA assessed by DCFI. An initially performed comparison of the results between patients w h o were examined some days after CEA and those who were studied at later follow-up showed no significant differences apart from three ICA occlusions in the follow-up group. On the other hand, there were considerable differences in surgical techniques between the two subpopulations. As a result we focused on the analysis of the variety of morphologic and haemodynamic changes after CEA in the total study population rather than attempting to separate the two groups. Our results show that haemodynamic changes regularly occur in the CCA and extend into the ICA for only a short distance above the bifurcation. Turbulence and flow revergal are particularly severe in dilated segments but are only occasionally seen near minor vascular wall irregularities. These findings suggest that the diagnosis of minor, recurrent stenosis (