Ultrasound in Med.& BioLVol. 18, No. 4, pp. 421-425, 1992
0301-5629/92 $5.(]0+ .00 © 1992 Pergamon Press Ltd.
Printed in the U.S.A.
OBrief Communication ULTRASOUND DOPPLER SPECTRAL ANALYSIS IN THE DIAGNOSIS OF OCCLUSIVE LESIONS OF THE CAROTID ARTERIES E. COOPERBERG Department of Surgical Treatment of Cerebral Circulation's Disorders, Bakulev Institute for Cardiovascular Surgery, Moscow, Russia (Received 30 April 1991; in final form 31 October 1991)
Abstract--To evaluate the accnracy of Ultrasound Doppler Spectral Analysis (USDA), 432 carotid arteries were examined [group I consisted of 145 eases of normal ICA (internal carotid artery), group II--187 eases of ICA stenosis and group III--100 eases of ICA occlusion]. The maximal systolic peak (Smax), maximal diastolic peak (Dmax), the spectral broadening index (SB) and the puisativity index (PI) were calculated and related to the degree of stenosis as determined by contrast arteriography. Our results indicate that the change of the value of Smax is the main diagnostic criterion: for carotid arteries' stenoses, it increases to over 3.2 kHz. A Smax over 6.0 kHz is a veritable index of ICA stenosis above 50%: in such eases as the stenosis increases by 10%, this corresponds to an approximately I kHz increase of Smax. In ICA occlusion, the main criterion was the absence of blood flow in the distal ICA segment. The UDSA sensitivity in hemodynamieally insignificant stenosis is 73.9% which is far higher than in the periorbital Doppler examination; in hemodynamieally significant stenosis and ICA occlusions, the sensitivity is 98.6 and 99% respectively.
Key Words: Ultrasound Doppler Spectral Analysis, Cerebral arteriography, Carotid artery.
INTRODUCTION
MATERIAL AND M E T H O D S
Progress in the noninvasive diagnosis of occlusive lesions of brachiocephalic arteries is largely due to the introduction of ultrasonic methods. The first to be used was periorbital Doppler (POD) which was widely used in the 1960s. POD was of value for the detection of hemodynamically significant carotid stenosis (Barnes et al. 1976; Gusev et al. 1977; Keller and Baumgartner 1973; Keller et al. 1976; Lees and Kistler 1986; Nikitin et al. 1980). At the same time, POD could not detect stenoses that did not reduce pressure and flow. In addition, it could not separate a stenosis of the internal carotid artery from a total occlusion (Eriksson et al. 1981; Zbornicova et al. 1982). The introduction of the method of Ultrasound Doppler Spectral Analysis (UDSA) from the carotid arteries using continuous-wave (CW) Doppler helped solve some of those diagnostic problems (Bandyk et al. 1985; Johnston et al. 1982; Keagy et al. 1982; Norrving et al. 1985). This study's goal consisted in the elaboration of the diagnostic criteria of the UDSA method on the basis of Doppler-angiographic comparisons.
The clinical material consisted of 216 patients with atherosclerotic lesions of carotid arteries examined during the last 3 years (211 males and 5 females, mean age--52.4 years). All the patients were examined by the UDSA method with subsequent arteriographic study of the carotid bifurcations in two planes. The degree of the stenosis was determined by the following formula: S = [1 -
d,d']D,D'], 100%,
where S = the stenosis percentage, d and d' = the ICA diameter in the most restricted site, and D and D' = the ICA diameter in the normal segment. As assessed by arteriography, the carotid lesion was not found in 29 cases; 41 had unilateral and 50 had bilateral ICA stenoses; in 45 cases there was a combination of occlusion and contralateral ICA stenosis. Unilateral ICA occlusion was found in 45 patients; bilateral occlusion was found in 6 cases. The patients were divided into 3 groups, on the basis of the arteriographic data. Group I consisted of 421
422
Ultrasound in Medicineand Biology
Volume18, Number 4, 1992
145 cases of normal carotid arteries (control group); group II consisted of 187 cases o f l C A stenosis; group III consisted of 100 cases of ICA occlusion (in total, 432 arteries in 216 patients). Group II was further subdivided into IIa--115 arteries--with the stenosis below 50% and Iib--72 arteries--with ICA stenosis above 50%.
UDSA of the carotid arteries was carried out with the help of Vasoscan (Berner Ross, England). The method consisted in the study of the spectral characteristics of the CW Doppler signal (ultrasound probe, 4 MHz) located over the carotid artery. The spectrogram consisted of many-colored pixels the total of which gives a spectrum of velocity rates in the trans-
GRIN 35
v"
4 MHz ,H
MRXR : ::, KHz MRXD ,, ~5KHz RP 7: ~i.YPI SB P. RCBS 2 -~
TIME S
0.33s
3.0" 11/02/86
4:01
pm
(1) 1 1.9~HZI GAIN35
~/
qMHz
HRXR ; K H z MRXD ~ , / K H z RP ~ , 3 PI i ~ SB !! RI S 3 11/02/86 4:03 pm
(2) Iol KH
GRIN35
,,/
4MHz 3 . 2 . . . < 6.0 kHz; C = Smax > 6.0 kHz.
423
The volume of blood which passes through the ICA is constant. The ICA stenosis results in an increase in the blood flow velocity, distal to the stenosis Smax falls which means that the regular laminar flow becomes irregular (turbulent). The diagnostic criteria of the UDSA are based on these features. We have analyzed the following diagnostic criteria of the UDSA: S m a x - - t h e maximal systolic peak, D m a x - - t h e maximal diastolic peak, index SB (spectrum broadening) and index PI (the pulsation index). The velocity signals were recorded from three points (on each side of the neck): near the lower border of the sternocleidomastoidal muscle (CCA), upper border of the thyroid cartilage (CCA bifurcation) and near the edge of the lower jaw (distal ICA segment). The spectrograms were registered with the help of a CANON printer (Japan). Diagnostic criteria of the UDSA were compared to the arteriographic data. RESULTS
verse arterial plane during one cardiac cycle. The position of a given point relative to the Y-axis (the scale of frequencies) corresponds to a certain linear blood flow rate (expressed in kHz), and its color corresponds to the specific weight of this frequency in the spectrum (maximal values indicate the point as red; minimal values show as blue). The form of the ICA and the external carotid artery (ECA) spectrograms are different: the ECA spectrogram has a sharp systolic peak and a low diastolic component; the ICA spectrogram has a broad systolic peak and a significantly higher diastolic component (Fig. 1). In doubtful cases, the ICA and ECA spectrograms were differentiated with the help of Russel's test (Russel et al. 1984).
At the level of the bifurcation and ICA, the normal values of Smax (group I) did not exceed 3.2 kHz in 97.2%. In ICA stenosis 50% (group IIb), Smax exceeded 3.2 kHz in 98.6% (Fig. 2). In normal subjects, the Dmax value did not exceed 1.0 kHz in 99.3%. At the same time, for ICA stenosis 1 kHz was only present in 25.7%, but in stenosis >50% (group IIb), it exceeded 1 kHz in 73.9% (Fig. 3). The SB index value in normal subjects (group I) varied widely; however, it almost never exceeded 40% (only in 2.2% was it more than 40%). In ICA stenosis
MAXIMAL DIASTOI,I(! PEAl(
SPI:X'TR[ M I~ROAI)I(NING
1(
100
/
90 80 70-
,.J
6050 40
z
30 20 10-
1
O-
Fig. 3. Dependence of Dmax values on ICA state: A = Dmax < 1.0 kHz; B =/)max > 1.0 kHz.
Fig. 4. Dependence of SB values on ICA state: A = SB < 40%; B = SB > 40%.
424
Ultrasound in Medicine and Biology
Volume 18, Number 4, 1992
12 "4-4-
1Dq
z
oj -I-
-E.
9 0
m
"4-
8
-P-I-
'4'- "4- _ p " 4 - + ~ ' ~ ++ / +
0
'4-~
4--t-
"4-
+
7cO
"4-
65-
so
O
.4_4 .
gs
45
go
15
io
go
gs
1 O0
OF STENOSIS (ARTERIOGRAPHIC DATA)
%
Fig. 5. Comparison of Smax values during location of the arteriographic data: vertical line = value of Smax in kHz; horizontal line = arteriographic data (% of stenosis of ICA).
0301-5629/92 $5.(]0+ .00 © 1992 Pergamon Press Ltd.
Printed in the U.S.A.
OBrief Communication ULTRASOUND DOPPLER SPECTRAL ANALYSIS IN THE DIAGNOSIS OF OCCLUSIVE LESIONS OF THE CAROTID ARTERIES E. COOPERBERG Department of Surgical Treatment of Cerebral Circulation's Disorders, Bakulev Institute for Cardiovascular Surgery, Moscow, Russia (Received 30 April 1991; in final form 31 October 1991)
Abstract--To evaluate the accnracy of Ultrasound Doppler Spectral Analysis (USDA), 432 carotid arteries were examined [group I consisted of 145 eases of normal ICA (internal carotid artery), group II--187 eases of ICA stenosis and group III--100 eases of ICA occlusion]. The maximal systolic peak (Smax), maximal diastolic peak (Dmax), the spectral broadening index (SB) and the puisativity index (PI) were calculated and related to the degree of stenosis as determined by contrast arteriography. Our results indicate that the change of the value of Smax is the main diagnostic criterion: for carotid arteries' stenoses, it increases to over 3.2 kHz. A Smax over 6.0 kHz is a veritable index of ICA stenosis above 50%: in such eases as the stenosis increases by 10%, this corresponds to an approximately I kHz increase of Smax. In ICA occlusion, the main criterion was the absence of blood flow in the distal ICA segment. The UDSA sensitivity in hemodynamieally insignificant stenosis is 73.9% which is far higher than in the periorbital Doppler examination; in hemodynamieally significant stenosis and ICA occlusions, the sensitivity is 98.6 and 99% respectively.
Key Words: Ultrasound Doppler Spectral Analysis, Cerebral arteriography, Carotid artery.
INTRODUCTION
MATERIAL AND M E T H O D S
Progress in the noninvasive diagnosis of occlusive lesions of brachiocephalic arteries is largely due to the introduction of ultrasonic methods. The first to be used was periorbital Doppler (POD) which was widely used in the 1960s. POD was of value for the detection of hemodynamically significant carotid stenosis (Barnes et al. 1976; Gusev et al. 1977; Keller and Baumgartner 1973; Keller et al. 1976; Lees and Kistler 1986; Nikitin et al. 1980). At the same time, POD could not detect stenoses that did not reduce pressure and flow. In addition, it could not separate a stenosis of the internal carotid artery from a total occlusion (Eriksson et al. 1981; Zbornicova et al. 1982). The introduction of the method of Ultrasound Doppler Spectral Analysis (UDSA) from the carotid arteries using continuous-wave (CW) Doppler helped solve some of those diagnostic problems (Bandyk et al. 1985; Johnston et al. 1982; Keagy et al. 1982; Norrving et al. 1985). This study's goal consisted in the elaboration of the diagnostic criteria of the UDSA method on the basis of Doppler-angiographic comparisons.
The clinical material consisted of 216 patients with atherosclerotic lesions of carotid arteries examined during the last 3 years (211 males and 5 females, mean age--52.4 years). All the patients were examined by the UDSA method with subsequent arteriographic study of the carotid bifurcations in two planes. The degree of the stenosis was determined by the following formula: S = [1 -
d,d']D,D'], 100%,
where S = the stenosis percentage, d and d' = the ICA diameter in the most restricted site, and D and D' = the ICA diameter in the normal segment. As assessed by arteriography, the carotid lesion was not found in 29 cases; 41 had unilateral and 50 had bilateral ICA stenoses; in 45 cases there was a combination of occlusion and contralateral ICA stenosis. Unilateral ICA occlusion was found in 45 patients; bilateral occlusion was found in 6 cases. The patients were divided into 3 groups, on the basis of the arteriographic data. Group I consisted of 421
422
Ultrasound in Medicineand Biology
Volume18, Number 4, 1992
145 cases of normal carotid arteries (control group); group II consisted of 187 cases o f l C A stenosis; group III consisted of 100 cases of ICA occlusion (in total, 432 arteries in 216 patients). Group II was further subdivided into IIa--115 arteries--with the stenosis below 50% and Iib--72 arteries--with ICA stenosis above 50%.
UDSA of the carotid arteries was carried out with the help of Vasoscan (Berner Ross, England). The method consisted in the study of the spectral characteristics of the CW Doppler signal (ultrasound probe, 4 MHz) located over the carotid artery. The spectrogram consisted of many-colored pixels the total of which gives a spectrum of velocity rates in the trans-
GRIN 35
v"
4 MHz ,H
MRXR : ::, KHz MRXD ,, ~5KHz RP 7: ~i.YPI SB P. RCBS 2 -~
TIME S
0.33s
3.0" 11/02/86
4:01
pm
(1) 1 1.9~HZI GAIN35
~/
qMHz
HRXR ; K H z MRXD ~ , / K H z RP ~ , 3 PI i ~ SB !! RI S 3 11/02/86 4:03 pm
(2) Iol KH
GRIN35
,,/
4MHz 3 . 2 . . . < 6.0 kHz; C = Smax > 6.0 kHz.
423
The volume of blood which passes through the ICA is constant. The ICA stenosis results in an increase in the blood flow velocity, distal to the stenosis Smax falls which means that the regular laminar flow becomes irregular (turbulent). The diagnostic criteria of the UDSA are based on these features. We have analyzed the following diagnostic criteria of the UDSA: S m a x - - t h e maximal systolic peak, D m a x - - t h e maximal diastolic peak, index SB (spectrum broadening) and index PI (the pulsation index). The velocity signals were recorded from three points (on each side of the neck): near the lower border of the sternocleidomastoidal muscle (CCA), upper border of the thyroid cartilage (CCA bifurcation) and near the edge of the lower jaw (distal ICA segment). The spectrograms were registered with the help of a CANON printer (Japan). Diagnostic criteria of the UDSA were compared to the arteriographic data. RESULTS
verse arterial plane during one cardiac cycle. The position of a given point relative to the Y-axis (the scale of frequencies) corresponds to a certain linear blood flow rate (expressed in kHz), and its color corresponds to the specific weight of this frequency in the spectrum (maximal values indicate the point as red; minimal values show as blue). The form of the ICA and the external carotid artery (ECA) spectrograms are different: the ECA spectrogram has a sharp systolic peak and a low diastolic component; the ICA spectrogram has a broad systolic peak and a significantly higher diastolic component (Fig. 1). In doubtful cases, the ICA and ECA spectrograms were differentiated with the help of Russel's test (Russel et al. 1984).
At the level of the bifurcation and ICA, the normal values of Smax (group I) did not exceed 3.2 kHz in 97.2%. In ICA stenosis 50% (group IIb), Smax exceeded 3.2 kHz in 98.6% (Fig. 2). In normal subjects, the Dmax value did not exceed 1.0 kHz in 99.3%. At the same time, for ICA stenosis 1 kHz was only present in 25.7%, but in stenosis >50% (group IIb), it exceeded 1 kHz in 73.9% (Fig. 3). The SB index value in normal subjects (group I) varied widely; however, it almost never exceeded 40% (only in 2.2% was it more than 40%). In ICA stenosis
MAXIMAL DIASTOI,I(! PEAl(
SPI:X'TR[ M I~ROAI)I(NING
1(
100
/
90 80 70-
,.J
6050 40
z
30 20 10-
1
O-
Fig. 3. Dependence of Dmax values on ICA state: A = Dmax < 1.0 kHz; B =/)max > 1.0 kHz.
Fig. 4. Dependence of SB values on ICA state: A = SB < 40%; B = SB > 40%.
424
Ultrasound in Medicine and Biology
Volume 18, Number 4, 1992
12 "4-4-
1Dq
z
oj -I-
-E.
9 0
m
"4-
8
-P-I-
'4'- "4- _ p " 4 - + ~ ' ~ ++ / +
0
'4-~
4--t-
"4-
+
7cO
"4-
65-
so
O
.4_4 .
gs
45
go
15
io
go
gs
1 O0
OF STENOSIS (ARTERIOGRAPHIC DATA)
%
Fig. 5. Comparison of Smax values during location of the arteriographic data: vertical line = value of Smax in kHz; horizontal line = arteriographic data (% of stenosis of ICA).
