J Clin Ultrasound 20369374, JulyiAugust 1992 0 1992 by John Wiley & Sons, Inc.
CCC 0091-2751/92/060369-06 $04.00
Color Flow Doppler Ultrasonography: Comparison with Peripheral Arteriography for the Investigation of Peripheral Vascular Disease J.F. Whelan, MD,* M.H. Barry, MD,t and J.D. Mob, MDt
Abstract: Using arteriography as the gold standard, Color flow Doppler ultrasonography was evaluated with regard to its ability to detect peripheral vascular occlusive disease and hemodynamically significant stenosis in patients having peripheral arteriography. One hundred legs in 51 patients were compared at seven arterial segments for disease. Color flow Doppler ultrasonography correctly detected 84 occluded segments, and demonstrated a sensitivity and specificity for patency vs occlusive disease of 95% and 99%,respectively. One hundred and thirty hemodynamically significant lesions (occlusions plus significant stenosis) were correctly identified with color flow Doppler ultrasonography, with a sensitivity and specificity of 92% and 97%, respectively. Color flow Doppler ultrasonography is a safe, inexpensive, and noninvasive method of accurately documenting significant peripheral arterial disease and offers a new first-line investigation for patients presenting with symptoms of peripheral arterial insufficiency. Indexing Words: Color flow Doppler ultrasonography Peripheral arterial disease Occlusions * Stenosis Arteriography
-
Arteriography has been the gold standard for the investigation of peripheral vascular disease. However, it is invasive, costly, and associated with a small morbidity and mortality. Many referring physicians investigate patients presenting with rest pain, night pain, pregangrene, or gangrene. Noninvasive methods such as systemic blood pressure measurements, Doppler analog waveforms, and pulse volume readings have been used to evaluate these patients; however, these examinations are not commonly performed within a diagnostic imaging department. Highresolution anatomical information and movement of blood flow can accurately be detected by color flow Doppler (CFD) ultrasonography .’ CFD ultrasound consists of three components: realtime ultrasonography for road mapping, pulse Doppler ultrasonography for waveform analysis From *Renforth and iftothesay, New Brunswick, Canada. For reprints contact J.F. Whelan, MD, 23 Elizabeth Parkway, Renforth, New Brunswick, Canada.
-
and peak velocity estimation, and CFD for assessing vessel patency and placement of the pulse Doppler. The purpose of this prospective study was to compare the efficacy and accuracy of CFD ultrasonography to peripheral arteriography and to determine if CFD ultrasonography would be a worthwhile examination to perform in a diagnostic imaging department. MATERIALS AND METHODS The lower extremity was scanned with CFD ultrasonography from the common femoral artery to the adductor’s canal in the transverse plane to allow a quick assessment of patency occlusion. CFD ultrasonography with continuous spectral Doppler sampling was then performed in the longitudinal plane assessing hemodynamically significant disease. Finally, with the patient in the prone position, the popliteal artery from adductar’s canal to the trifurcation (origin of the anterior tibia artery and the posterior tibioperoneal 369
310
WHELAN ET AL.
FIGURE 1. Occlusion of right superficial femoral artery. (A) Color flow Doppler ultrasound image demonstrates abrupt end of color at level of occlusion. (B) Reconstitution of right popliteal artery on CFD ultrasonography with a monophasic waveform pattern.
FIGURE 2. Hemodynamically significant stenosis of right popliteal artery. (A) Color flow Doppler ultrasound image of stenosis of right popliteal artery. (B) Spectral waveform analysis shows marked increase in velocity at stenosis t o 244 cmis. (C) Waveform analysis below stenosis shows spectral broadening and decreased velocity t o 67 cm/s.
JOURNAL OF CLINICAL ULTRASOUND
COLOR ULTRASONOGRAPHY VS. ARTERIOGRAPHY
FIGURE 1. Occlusion of right superficial femoral artery. Peripheral arteriogram shows occlusion at adductor canal and reconstruction of right popliteal artery.
FIGURE 2. Hernodynamically significant stenosis of right popliteal artery. Arteriogram of right popliteal artery shows severe stenosis. VOL. 20, NO. 6.JULYIAUGUST 1992
371
artery distal to the origin of the anterior tibia artery) was examined in the transverse and longitudinal planes with CFD ultrasonography and spectral Doppler sampling. The common and external iliac arteries were not scanned directly with CFD ultrasonography. However, Doppler waveform and peak velocity estimation were interpreted in the common femoral artery to infer disease in these arteries. Fifty-one patients were investigated and 100 extremities were scanned. All patients having peripheral arteriography for the investigation of peripheral vascular disease from June 1, 1990 to November 25, 1990, were included in the study, except the patients with previous peripheral vascular surgery. All CFD ultrasound examinations were performed on the day of the arteriography. Only one was performed after the arteriogram. A Toshiba SSA 270 A ultrasound scanner with a 5-MHz linear array probe was used. The peripheral vascular arteriogram was performed by another radiologist, and the results of the CFD ultrasonography and arteriography were compared for patency, stenosis greater than 50%,and/or occlusions at the following segments: iliac artery, common femoral artery, superficial femoral artery (SFA) proximal, SFA distal, popliteal proximal, popliteal distal, and trifurcation. Lesions detected with CFD ultrasonography were noted on anatomical chart and the length of each examination was documented. The interpretation of both the CFD ultrasonography and the arteriography was performed independently by different radiologists, and the results were compared at a later date. The origin of the profunda femoral artery was patent in all cases but was not included in the statistical data. Only the origin of the anterior tibia, posterior tibia, and peroneal arteries were examined. Patency of a vessel was determined by a normal triphasic waveform pattern and color saturation demonstrated throughout the lumen of the artery. Occlusion of an artery was diagnosed when no color saturation and no Doppler waveform was seen in the artery (Figure 1).A hemodynamically significant stenosis (stenosis greater than 50% was diagnosed when the peak systolic velocity estimation was over 200 cm/s at the stenotic lesion and the waveform changed from triphasic to monophasic below the stenosis.2 If the critical stenosis was beyond another stenosis or was in a reconstituted segment, the criteria of the ratio of the velocity at the stenosis to the velocity just distal to the stenosis of greater than 2: was to a lesion2 (Figure 2).
WHELAN ET AL.
372 TABLE 1 Arterial Occlusion Disease Arteriography Present
Absent
Total
CFD Positive Negative
84 4
9 599
93 603
Total
88
608
696
RESULTS
The length of the CFD ultrasound examination varied between 30 minutes and 45 minutes depending on the severity of the patient’s peripheral vascular disease. Four trifurcation segments were inadequately scanned due to flexure contractures at the knee in 2 extremities and due to the patient’s inability to turn to the prone position in the other 2 extremities. Forty-three of 51 patients (82%) had peripheral vascular disease, with 31 of 51 having multisegmental disease. When comparison between CFD ultrasonography and arteriography for occlusive disease was performed (Table 1 and 21, the sensitivity was 95%, specificity 99%,accuracy 98%,positive predictive value 90%,and negative predictive value 99%.3 When the comparison was made between CFD ultrasonography and arteriography for hemodynamically significant lesions, (Tables 3 and 41, the sensitivity was 92%, specificity 97%, accuracy 96%, positive predictive value 90%, and negative predictive value 98%.3All 23 occlusions in the proximal SFA were correctly identified as
to site of origin and their reconstitution. In the distal SFA there were 6 false positive occlusions for positive predictive value of 84%.The causes of false positives were as follows: (1) Two long segments of severe stenosis misidentified as occlusions. (2) Two adductor segments were inadequately visualized and called occluded. A normal triphasic wave form was evident in the popliteal arteries. (3) Two short-segment stenoses were diagnosed as short occlusions. All 40 hemodynamically significant lesions were identified in the proximal SFA. In the distal SFA, 3 short segmental stenoses were not identified. The sensitivity for detecting hemodynamically significant lesions in the distal SFA was 94%. DISCUSSION
CFD ultrasonography allows precise anatomical and physiological information of peripheral arterial anatomy. It can quickly localize and identify arteries in the leg. CFD ultrasonography can detect flow abnormalities qualitatively and can measure flow abnormalities quantitatively. The normal Doppler waveform of a peripheral artery is a triphasic pattern: high-velocity peak from cardiac systole, brief reversal due to high arterial peripheral resistance, and low peak velocity from late cardiac diastole.2 High-grade stenosis (50%to 99%)shows a marked increase in systolic velocities with extensive spectral broadening and loss of reverse flow component.’ Distal to a severe stenosis or occlusion, the Doppler wave-
TABLE 2 CFD Ultrasonography vs Afieriagraphy- Occlusive Disease” Blood Vessel Iliac CFA SFA P SFA D Pop P Pop D Trifa
Sens
Spec
PPV
NPV
Acc
% 717
% 93193 100/100 77/77 61/67 91/91 90191 87189
% 717
% 93/93 1001100 77i77 61/62 91191 90191 87/89
% 1oolloo 1001100 1001100 931100 1001100 981100 92196
-
23123 32/33 919 819 517
-
23/23 32138 919 a19 517
Sens: sensitivity Spec: specificity PPV: positive predictive value NPV: negative predictive value ACC: accuracy CFA: common femoral artery SFA P: superficial femoral artery, proximal SFA D: superficial femoral artery, distal Pop P: popliteal, proximal Pop D: popliteal, distal Trif: trifurcation ”Four segments inadequately scanned. JOURNAL OF CLINICAL ULTRASOUND
COLOR ULTRASONOGRAPHY VS. ARTERIOGRAPHY TABLE 3 HemodynamicallySignificant Lesions Arteriography Present
Absent
Total
CFD Positive Negative
130 11
15 541
145 552
Total
141
556
697
form becomes monophasic with reduced peak velocity (Figure 2). Highvelocity jets and turbulence associated with arterial stenosis are damp~ ened out within a very short d i ~ t a n c e .Hence CFD makes the placement of the pulsed Doppler easy and accurate for proper measurement of peak velocity estimation. The tables illustrate that CFD ultrasonography is accurate for the detection of significant stenosis and occlusive disease of the peripheral vascular tree. This prospective study correlates well with the results presented by Cossman et al.3 and Pollack et al.5 dealing with CFD ultrasonography and arteriography in the lower extremities. The 6 false-positive occlusions in the distal SFA illustrates some important points. In 4 cases, 2 short segmental stenoses and 2 long multisegmental stenoses, the slow flow in the stenotic lesions resulted in no detectable Doppler signal. The remaining 2 cases illustrate misinterpretation of the data. The distal SFA segTABLE
4
CFD Ultrasonography vs Arteriography Hemodynamically Significant Lesions”
Blood Vessel
Sens
Spec
PPV
NPV
Acc
%
% 79179 92197 59161 50153 85/87 89/90 87/89
% 17117 2i7 40142 44147 13115 9/10 5i7
%
Yo
79183 92193 59159 50153 85185 89/90 87/89
961100 941100 991101 941100 981100 981100 92/96
~
Iliac
CFA SFA Pa SFA D Pop P Pop D Trifb
17121 213 40140 44147 13113 9/10 517
Sens: sensitivity Spec: specificity PPV: positive predictive value NPV: negative predictive value ACC: accuracy CFA: common femoral artery SFA P: superficial femoral artery, proximal SFA D: superficial femoral artery, distal Pop P: popliteal, proximal Pop D: popliteal, distal Trif: trifurcation eOne segment had 2 stenoses. bFour segments inadequately scanned. VOL. 20, NO. 6, JULYIAUGUST 1992
373
ments were not well visualized and called occluded, whereas with a normal triphasic Doppler waveform in the popliteal arteries, these segments should have been interpreted as being patent. CFD ultrasound evaluation of hemodynamically significant lesions gave high sensitivity and specificity. In the proximal SFA and the popliteal segments, the results were excellent, with no significant lesions missed. However, in the distal SFA segment a slight decrease in sensitivity was seen, with 3 short segmental stenotic lesions not being identified in the adductor canal. This is the most difficult segment to visualize due to the course of the artery and its deep location within the muscle. Time must be taken to properly assess this region in longitudinal and transverse planes. If the entire artery cannot be examined, then the Doppler waveform and peak velocity estimation of the popliteal artery must be assessed to verify that no segmental occlusions or significant stenoses exist in the adductor canal region. The iliac artery’s patency is inferred from Doppler waveform analysis and peak velocity estimation obtained in the common femoral artery. As seen in the results from hemodynamically significant lesions, 4 stenotic lesions were unidentified with CFD ultrasonography in the iliac arteries. At the common femoral artery, 5 lesions deemed insignificant with arteriography were called hemodynamically significant lesions with CFD ultrasonography. Arteriography dogs have its limitations for it is not biplanar in most institutions and does not give morphological details of plaques. Hence, the significance of these lesions detected with CFD ultrasonography is difficult to assess and may represent significant disease. Image directed ultrasound scanning requires one to two hours with an experienced technologist for accurate evaluation of the lower leg arterial system.2,6The average length of the CFD ultrasound examination in our institution was slightly longer than that given in a recent publication, which gave an average of 29 minutes per e ~ a m i n a t i o nThirty-one .~ of our 51 patients had multisegmental disease. In addition, this study was undertaken while our technologists were on their learning curve under the supervision of a radiologist. CONCLUSION
CFD ultrasonography is an inexpensive, fast, mobile, repeatable, and noninvasive procedure for the investigation of peripheral vascular dis-
WHELAN ET AL.
374
ease. From these data, it appears to be accurate and effective for the diagnosis of hemodynamically significant disease in lower extremity arteries. Thus CFD ultrasonography could be used as a first-line investigation in patients presenting with peripheral vascular disease to a diagnostic imaging department. ACKNOWLEDGMENT
We thank Toshiba Medical Canada (Toronto) for their financial support.
REFERENCES 1. Mitchell DG: Color Doppler imaging: Principles, limitations and artifacts. Radiology 177:l- 10, 1990.
2. Jager KAA, Philips DJ, Martin RL, et al: Noninvasive mapping of lower limb arterioal lesions. Ultrasound Med. BioZ11:515-521, 1985. 3. Cossman DV, Ellison JE, Wagner WH, et al: Comparison of contrast arteriography to arterial mapping with color flow duplex imaging in the lower extremities. J Vasc Surg 55522-529, 1989. 4. Evans DH, MacPherson DS, Asher MJ, et al: Changes in Doppler ultrasound sonagrams at varing distances from stenosis. Cardiovasc Res 16~631-636,1982. 5. Polak JF, Karmel MI, Mannick JA, et al: Determination of the Extent of lower-extremity peripheral arterial disease with color-assisted duplex sonography. AJR 1441085-1089, 1990. 6. Kohler TR, Nanee DR, Cramer MM, et al: Duplex scanning for diagnoses af avitoiliae and femoropopliteal disease: A prospective study. Circulation No. 5, 1074-1080, 1987.
JOURNAL OF CLINICAL ULTRASOUND
CCC 0091-2751/92/060369-06 $04.00
Color Flow Doppler Ultrasonography: Comparison with Peripheral Arteriography for the Investigation of Peripheral Vascular Disease J.F. Whelan, MD,* M.H. Barry, MD,t and J.D. Mob, MDt
Abstract: Using arteriography as the gold standard, Color flow Doppler ultrasonography was evaluated with regard to its ability to detect peripheral vascular occlusive disease and hemodynamically significant stenosis in patients having peripheral arteriography. One hundred legs in 51 patients were compared at seven arterial segments for disease. Color flow Doppler ultrasonography correctly detected 84 occluded segments, and demonstrated a sensitivity and specificity for patency vs occlusive disease of 95% and 99%,respectively. One hundred and thirty hemodynamically significant lesions (occlusions plus significant stenosis) were correctly identified with color flow Doppler ultrasonography, with a sensitivity and specificity of 92% and 97%, respectively. Color flow Doppler ultrasonography is a safe, inexpensive, and noninvasive method of accurately documenting significant peripheral arterial disease and offers a new first-line investigation for patients presenting with symptoms of peripheral arterial insufficiency. Indexing Words: Color flow Doppler ultrasonography Peripheral arterial disease Occlusions * Stenosis Arteriography
-
Arteriography has been the gold standard for the investigation of peripheral vascular disease. However, it is invasive, costly, and associated with a small morbidity and mortality. Many referring physicians investigate patients presenting with rest pain, night pain, pregangrene, or gangrene. Noninvasive methods such as systemic blood pressure measurements, Doppler analog waveforms, and pulse volume readings have been used to evaluate these patients; however, these examinations are not commonly performed within a diagnostic imaging department. Highresolution anatomical information and movement of blood flow can accurately be detected by color flow Doppler (CFD) ultrasonography .’ CFD ultrasound consists of three components: realtime ultrasonography for road mapping, pulse Doppler ultrasonography for waveform analysis From *Renforth and iftothesay, New Brunswick, Canada. For reprints contact J.F. Whelan, MD, 23 Elizabeth Parkway, Renforth, New Brunswick, Canada.
-
and peak velocity estimation, and CFD for assessing vessel patency and placement of the pulse Doppler. The purpose of this prospective study was to compare the efficacy and accuracy of CFD ultrasonography to peripheral arteriography and to determine if CFD ultrasonography would be a worthwhile examination to perform in a diagnostic imaging department. MATERIALS AND METHODS The lower extremity was scanned with CFD ultrasonography from the common femoral artery to the adductor’s canal in the transverse plane to allow a quick assessment of patency occlusion. CFD ultrasonography with continuous spectral Doppler sampling was then performed in the longitudinal plane assessing hemodynamically significant disease. Finally, with the patient in the prone position, the popliteal artery from adductar’s canal to the trifurcation (origin of the anterior tibia artery and the posterior tibioperoneal 369
310
WHELAN ET AL.
FIGURE 1. Occlusion of right superficial femoral artery. (A) Color flow Doppler ultrasound image demonstrates abrupt end of color at level of occlusion. (B) Reconstitution of right popliteal artery on CFD ultrasonography with a monophasic waveform pattern.
FIGURE 2. Hemodynamically significant stenosis of right popliteal artery. (A) Color flow Doppler ultrasound image of stenosis of right popliteal artery. (B) Spectral waveform analysis shows marked increase in velocity at stenosis t o 244 cmis. (C) Waveform analysis below stenosis shows spectral broadening and decreased velocity t o 67 cm/s.
JOURNAL OF CLINICAL ULTRASOUND
COLOR ULTRASONOGRAPHY VS. ARTERIOGRAPHY
FIGURE 1. Occlusion of right superficial femoral artery. Peripheral arteriogram shows occlusion at adductor canal and reconstruction of right popliteal artery.
FIGURE 2. Hernodynamically significant stenosis of right popliteal artery. Arteriogram of right popliteal artery shows severe stenosis. VOL. 20, NO. 6.JULYIAUGUST 1992
371
artery distal to the origin of the anterior tibia artery) was examined in the transverse and longitudinal planes with CFD ultrasonography and spectral Doppler sampling. The common and external iliac arteries were not scanned directly with CFD ultrasonography. However, Doppler waveform and peak velocity estimation were interpreted in the common femoral artery to infer disease in these arteries. Fifty-one patients were investigated and 100 extremities were scanned. All patients having peripheral arteriography for the investigation of peripheral vascular disease from June 1, 1990 to November 25, 1990, were included in the study, except the patients with previous peripheral vascular surgery. All CFD ultrasound examinations were performed on the day of the arteriography. Only one was performed after the arteriogram. A Toshiba SSA 270 A ultrasound scanner with a 5-MHz linear array probe was used. The peripheral vascular arteriogram was performed by another radiologist, and the results of the CFD ultrasonography and arteriography were compared for patency, stenosis greater than 50%,and/or occlusions at the following segments: iliac artery, common femoral artery, superficial femoral artery (SFA) proximal, SFA distal, popliteal proximal, popliteal distal, and trifurcation. Lesions detected with CFD ultrasonography were noted on anatomical chart and the length of each examination was documented. The interpretation of both the CFD ultrasonography and the arteriography was performed independently by different radiologists, and the results were compared at a later date. The origin of the profunda femoral artery was patent in all cases but was not included in the statistical data. Only the origin of the anterior tibia, posterior tibia, and peroneal arteries were examined. Patency of a vessel was determined by a normal triphasic waveform pattern and color saturation demonstrated throughout the lumen of the artery. Occlusion of an artery was diagnosed when no color saturation and no Doppler waveform was seen in the artery (Figure 1).A hemodynamically significant stenosis (stenosis greater than 50% was diagnosed when the peak systolic velocity estimation was over 200 cm/s at the stenotic lesion and the waveform changed from triphasic to monophasic below the stenosis.2 If the critical stenosis was beyond another stenosis or was in a reconstituted segment, the criteria of the ratio of the velocity at the stenosis to the velocity just distal to the stenosis of greater than 2: was to a lesion2 (Figure 2).
WHELAN ET AL.
372 TABLE 1 Arterial Occlusion Disease Arteriography Present
Absent
Total
CFD Positive Negative
84 4
9 599
93 603
Total
88
608
696
RESULTS
The length of the CFD ultrasound examination varied between 30 minutes and 45 minutes depending on the severity of the patient’s peripheral vascular disease. Four trifurcation segments were inadequately scanned due to flexure contractures at the knee in 2 extremities and due to the patient’s inability to turn to the prone position in the other 2 extremities. Forty-three of 51 patients (82%) had peripheral vascular disease, with 31 of 51 having multisegmental disease. When comparison between CFD ultrasonography and arteriography for occlusive disease was performed (Table 1 and 21, the sensitivity was 95%, specificity 99%,accuracy 98%,positive predictive value 90%,and negative predictive value 99%.3 When the comparison was made between CFD ultrasonography and arteriography for hemodynamically significant lesions, (Tables 3 and 41, the sensitivity was 92%, specificity 97%, accuracy 96%, positive predictive value 90%, and negative predictive value 98%.3All 23 occlusions in the proximal SFA were correctly identified as
to site of origin and their reconstitution. In the distal SFA there were 6 false positive occlusions for positive predictive value of 84%.The causes of false positives were as follows: (1) Two long segments of severe stenosis misidentified as occlusions. (2) Two adductor segments were inadequately visualized and called occluded. A normal triphasic wave form was evident in the popliteal arteries. (3) Two short-segment stenoses were diagnosed as short occlusions. All 40 hemodynamically significant lesions were identified in the proximal SFA. In the distal SFA, 3 short segmental stenoses were not identified. The sensitivity for detecting hemodynamically significant lesions in the distal SFA was 94%. DISCUSSION
CFD ultrasonography allows precise anatomical and physiological information of peripheral arterial anatomy. It can quickly localize and identify arteries in the leg. CFD ultrasonography can detect flow abnormalities qualitatively and can measure flow abnormalities quantitatively. The normal Doppler waveform of a peripheral artery is a triphasic pattern: high-velocity peak from cardiac systole, brief reversal due to high arterial peripheral resistance, and low peak velocity from late cardiac diastole.2 High-grade stenosis (50%to 99%)shows a marked increase in systolic velocities with extensive spectral broadening and loss of reverse flow component.’ Distal to a severe stenosis or occlusion, the Doppler wave-
TABLE 2 CFD Ultrasonography vs Afieriagraphy- Occlusive Disease” Blood Vessel Iliac CFA SFA P SFA D Pop P Pop D Trifa
Sens
Spec
PPV
NPV
Acc
% 717
% 93193 100/100 77/77 61/67 91/91 90191 87189
% 717
% 93/93 1001100 77i77 61/62 91191 90191 87/89
% 1oolloo 1001100 1001100 931100 1001100 981100 92196
-
23123 32/33 919 819 517
-
23/23 32138 919 a19 517
Sens: sensitivity Spec: specificity PPV: positive predictive value NPV: negative predictive value ACC: accuracy CFA: common femoral artery SFA P: superficial femoral artery, proximal SFA D: superficial femoral artery, distal Pop P: popliteal, proximal Pop D: popliteal, distal Trif: trifurcation ”Four segments inadequately scanned. JOURNAL OF CLINICAL ULTRASOUND
COLOR ULTRASONOGRAPHY VS. ARTERIOGRAPHY TABLE 3 HemodynamicallySignificant Lesions Arteriography Present
Absent
Total
CFD Positive Negative
130 11
15 541
145 552
Total
141
556
697
form becomes monophasic with reduced peak velocity (Figure 2). Highvelocity jets and turbulence associated with arterial stenosis are damp~ ened out within a very short d i ~ t a n c e .Hence CFD makes the placement of the pulsed Doppler easy and accurate for proper measurement of peak velocity estimation. The tables illustrate that CFD ultrasonography is accurate for the detection of significant stenosis and occlusive disease of the peripheral vascular tree. This prospective study correlates well with the results presented by Cossman et al.3 and Pollack et al.5 dealing with CFD ultrasonography and arteriography in the lower extremities. The 6 false-positive occlusions in the distal SFA illustrates some important points. In 4 cases, 2 short segmental stenoses and 2 long multisegmental stenoses, the slow flow in the stenotic lesions resulted in no detectable Doppler signal. The remaining 2 cases illustrate misinterpretation of the data. The distal SFA segTABLE
4
CFD Ultrasonography vs Arteriography Hemodynamically Significant Lesions”
Blood Vessel
Sens
Spec
PPV
NPV
Acc
%
% 79179 92197 59161 50153 85/87 89/90 87/89
% 17117 2i7 40142 44147 13115 9/10 5i7
%
Yo
79183 92193 59159 50153 85185 89/90 87/89
961100 941100 991101 941100 981100 981100 92/96
~
Iliac
CFA SFA Pa SFA D Pop P Pop D Trifb
17121 213 40140 44147 13113 9/10 517
Sens: sensitivity Spec: specificity PPV: positive predictive value NPV: negative predictive value ACC: accuracy CFA: common femoral artery SFA P: superficial femoral artery, proximal SFA D: superficial femoral artery, distal Pop P: popliteal, proximal Pop D: popliteal, distal Trif: trifurcation eOne segment had 2 stenoses. bFour segments inadequately scanned. VOL. 20, NO. 6, JULYIAUGUST 1992
373
ments were not well visualized and called occluded, whereas with a normal triphasic Doppler waveform in the popliteal arteries, these segments should have been interpreted as being patent. CFD ultrasound evaluation of hemodynamically significant lesions gave high sensitivity and specificity. In the proximal SFA and the popliteal segments, the results were excellent, with no significant lesions missed. However, in the distal SFA segment a slight decrease in sensitivity was seen, with 3 short segmental stenotic lesions not being identified in the adductor canal. This is the most difficult segment to visualize due to the course of the artery and its deep location within the muscle. Time must be taken to properly assess this region in longitudinal and transverse planes. If the entire artery cannot be examined, then the Doppler waveform and peak velocity estimation of the popliteal artery must be assessed to verify that no segmental occlusions or significant stenoses exist in the adductor canal region. The iliac artery’s patency is inferred from Doppler waveform analysis and peak velocity estimation obtained in the common femoral artery. As seen in the results from hemodynamically significant lesions, 4 stenotic lesions were unidentified with CFD ultrasonography in the iliac arteries. At the common femoral artery, 5 lesions deemed insignificant with arteriography were called hemodynamically significant lesions with CFD ultrasonography. Arteriography dogs have its limitations for it is not biplanar in most institutions and does not give morphological details of plaques. Hence, the significance of these lesions detected with CFD ultrasonography is difficult to assess and may represent significant disease. Image directed ultrasound scanning requires one to two hours with an experienced technologist for accurate evaluation of the lower leg arterial system.2,6The average length of the CFD ultrasound examination in our institution was slightly longer than that given in a recent publication, which gave an average of 29 minutes per e ~ a m i n a t i o nThirty-one .~ of our 51 patients had multisegmental disease. In addition, this study was undertaken while our technologists were on their learning curve under the supervision of a radiologist. CONCLUSION
CFD ultrasonography is an inexpensive, fast, mobile, repeatable, and noninvasive procedure for the investigation of peripheral vascular dis-
WHELAN ET AL.
374
ease. From these data, it appears to be accurate and effective for the diagnosis of hemodynamically significant disease in lower extremity arteries. Thus CFD ultrasonography could be used as a first-line investigation in patients presenting with peripheral vascular disease to a diagnostic imaging department. ACKNOWLEDGMENT
We thank Toshiba Medical Canada (Toronto) for their financial support.
REFERENCES 1. Mitchell DG: Color Doppler imaging: Principles, limitations and artifacts. Radiology 177:l- 10, 1990.
2. Jager KAA, Philips DJ, Martin RL, et al: Noninvasive mapping of lower limb arterioal lesions. Ultrasound Med. BioZ11:515-521, 1985. 3. Cossman DV, Ellison JE, Wagner WH, et al: Comparison of contrast arteriography to arterial mapping with color flow duplex imaging in the lower extremities. J Vasc Surg 55522-529, 1989. 4. Evans DH, MacPherson DS, Asher MJ, et al: Changes in Doppler ultrasound sonagrams at varing distances from stenosis. Cardiovasc Res 16~631-636,1982. 5. Polak JF, Karmel MI, Mannick JA, et al: Determination of the Extent of lower-extremity peripheral arterial disease with color-assisted duplex sonography. AJR 1441085-1089, 1990. 6. Kohler TR, Nanee DR, Cramer MM, et al: Duplex scanning for diagnoses af avitoiliae and femoropopliteal disease: A prospective study. Circulation No. 5, 1074-1080, 1987.
JOURNAL OF CLINICAL ULTRASOUND
