Downloaded from www.ajronline.org by 131.252.96.28 on 09/13/14 from IP address 131.252.96.28. Copyright ARRS. For personal use only; all rights reserved

1085

Determination of Lower-Extremity Arterial Disease Assisted Duplex Comparison

Joseph F. PoIak1 Mitchell I. Karmel1 John A. Mannick2 Daniel H. O’Leary1 Magruder C. Donaldson2 Anthony D. Whittemore2

Color-assisted

sonography motion

of

with Angiography

duplex

to rapidly

of flowing

the Extent Peripheral with ColorSonography:

sonography

survey

blood

combines

color

Doppler

flow

the vasculature.

Color

Doppler

sonography

on a sonogram

and

is used

to visualize

imaging

with

duplex

displays

directly

arterial

the lumen

narrowing of greater than 50%. Absence of Doppler signals signifies arterial occlusion. Preangiographic triage of patients for surgical or interventional therapy requires that a rapid and accurate survey be performed of the extent of arterial disease. Color-assisted duplex sonography was applied prospectively in 17 consecutive patients with the clinical diagnosis of peripheral arterial disease who had not previously undergone angiography. Results of angiography of the femoropopliteal arteries were graded and localized in one of

seven

approximately

equal

arterial

segments.

These

were

compared

with

similar

segmental maps made with sonography. For the detection of stenosis or occlusion in any of the 238 segments, the sensitivity was 0.88 (49/56), specificity 0.95 (173/182), and accuracy 0.93 (222/238). The average time taken to survey both limbs was 29 mm. We conclude that color-assisted duplex sonography is an accurate and rapid tool for the

noninvasive AJR

mapping

of the

155:1085-1089,

extent

November

of femoropopliteal

arterial

disease.

1990

Patients with peripheral arterial disease of the lower extremities who undergo angiography may have long stenoses and occlusions suitable for surgical bypass or shorter segmental lesions amenable to percutaneous intervention [1 -3]. It has previously not been possible to noninvasively predict the extent and severity of segmental arterial lesions [4, 5]. Increasing clinical experience with color Doppler flow sonography suggests this technique can facilitate the preangiographic determination of the nature and extent of segmental arterial disease in the legs. We propose that the subjective grading of narrowing of the arterial lumen by color flow imaging be combined with Doppler spectral analysis, a technique that by itself is accurate yet time-consuming [6, 7]. This combined approach can then map out the extent of segmental disease in symptomatic patients with suspected peripheral arterial disease.

Subjects

and

Seventeen Received April 2, 1990; accepted after June 1, 1990. 1 Department of Radiology, Harvard

revision Medical

School and Brigham and Women’s Hospital, 75 Francis St. , Boston, MA 021 1 5. Address reprint requests to J. F. Polak. 2 Department of Surgery. Harvard Medical School and Brigham ton, MA 02115.

and Women’s

0361-803X/90/1 555-1085 © American Roentgen Ray Society

Hospital,

Bos-

ography

Methods

consecutive

or surgical

prospectively women

before

21 -82

symptomatic

revascularization

arteriography.

years

old

patients

who

of the lower

(mean

The patient age,

62).

had

not

extremities

The

previously

were

undergone

examined

population

consisted

indications

for

with

arteri-

sonography

of 1 3 men

arteriography

and four

consisted

of

claudication in 1 0 patients, cellulitis in one patient, nonhealing foot ulcer in one patient, foot ulcer and claudication in one patient, foot ulcer and rest pain in one patient, acute rest pain (possible embolic disease) in two patients, and vasculitis in one patient. Each lower extremity was scanned 1 -4 hr before arteriography with the sonographer blinded

to the type

and

pressure

measurements.

knee

along

(i.e.,

location

the course

of symptoms

Imaging of the

as well

was

performed

femoral

and

as to the

from

popliteal

results

the groin arteries).

of previous

crease Each

to just

extremity

noninvasive

below was

the

divided

1086

POLAK

into seven arterial the

profunda

femoral

arteries;

Color

standard

consisting

artery;

and

Doppler

sonographic View, CA)

Downloaded from www.ajronline.org by 131.252.96.28 on 09/13/14 from IP address 131.252.96.28. Copyright ARRS. For personal use only; all rights reserved

segments

femoris

the

the

proximal

imaging

was

and

mid,

distal

performed

system (Acuson in conjunction with high-resolution

of the common

proximal,

distal

popliteal

with

Each segment lumen

percent

was imaged

with

narrowing

measurement Arterial

transversely,

respect

to

of

lumen

the

the

trans-

of peak systolic segments

were

to size the colored

arterial

walls

diameter,

velocity

it as

and

longitudinally,

for

from the Doppler

scanned

contiguously

transverse

by 1 -cm

Once imaging

parallel

down

were

mapped

to the vessel

gate in the lumen lumen.

increments

the vessels

was

the

used

to place

and to angle-correct

A Doppler

spectrum

was

leg.

in the transverse it parallel

then

plane,

the Doppler

obtained

color

sampling

to the flow

and

the

peak

in the systolic

map

was

(0-49%),

significant

duration

of

the

drawn

summarizing

the location

(50-99%),

stenoses color-assisted

duplex

recorded in minutes. Arteriography was performed nique

and

ceuticals).

injections When

of 77-84 completed,

of normal

and occlusions sonographic

by using a four-station ml of Hypaque-76

the

arteriograms

segments

(1 00%). evaluation

were

The was

cut-film

(Winthrop

tech-

Pharma-

evaluated

blindly

by consensus opinion of two angiographers. Each segment was graded as being without significant narrowing (0-49%), significantly narrowed (50-99%), or occluded (1 00%). The segments were graded on the basis of the most severe component. Of the 34 extremities

TABLE

1:

Summary

Arteriography

of the

Results

for the Detection

of Color-Assisted

of Arterial

Segment

Duplex

Sonography

Stenosis

True Positive

Common femoral Profunda femoris Superficial femoral Proximal Mid Distal Popliteal Proximal Distal Total

and

Compared

False Positive

Occlusion

with

and Occlusions

Artery

Stenosis

stenoses,

of occlusions

,

calculated from it. Doubling of the peak systolic velocity expressed as a peak systolic velocity ratio was used as the criterion for a significant stenosis. This ratio was determined with respect to a point with a normal color flow pattern in the lumen situated at least 4 cm proximally. After both extremities were scanned and before arteriography, an velocity

arterial

femoropopliteal

eight stenoses,

had and

occlunine

All 25 limbs that had at least one diseased segment were detected with sonography, whereas three examinations were false-positive in nine limbs without femoropopliteal disease, two for stenoses and one for an occluded segment, giving a sensitivity of 1 .0 (0.87, 1 .0 [95% confidence intervals]), a specificity of 0.66 (0.35, 0.87), and an accuracy of 0.9 (0.77, 0.97). The two examinations false positive for stenoses were distal to iliac artery occlusions and the one false positive for occlusion was in a limb with severely diseased tibioperoneal branch arteries. Three limbs had long isolated femoral artery stenosis (12, 1 1 and 30 cm). The average of the velocities measured in the corresponding six arterial segments was 0.54 m/sec (± 0.38 m/sec SD; range, 0.3-1 .3 m/sec). The average velocity in nine extremities without diseased femoral segments was 0.97 m/sec (±0.31 m/sec SD). Seven limbs had isolated short segmental stenosis of the superficial femoral arteries. The average velocity at the stenoses was 2.53 ± 0.93 m/sec (range, 1 .47-3.84 m/sec), whereas it was 0.77 ± 0.51 m/sec in the segment above. The average peak systolic velocity ratio was 5.3 ± 5.5 (range, 2.0-17.0). A total of 238 arterial segments were evaluated for peak systolic velocity and color flow characteristics. Twenty-two significant stenoses were predicted and confirmed on arteriography. Three (1 4%) were predicted by a doubling of the peak systolic velocity. Nine (41 %) were predicted by a marked narrowing of the transverse lumen seen with color Doppler

plane for assessment of lumen diameter. The transducer, although transverse to the vessel, was angled 30#{176} to the vertical plane in order to encode arterial color signals. Relative narrowing of the lumen, when present, was graded as percentage of lumen narrowing with respect to the visualized arterial walls by using the smallest lumen diameter either in the left-to-right or anteroposterior direction. The transducer was held static for one to two cardiac cycles, then displaced

significant a combination

Results

spectrum.

in the

had

1990

flow

express

and

1 1 had six

November

had no significant femoropopliteal disease. Sensitivity and specificity were calculated according to previous recommendations [8, 9]. The accuracy of the technique was determined both for the combined 238 arterial segments as well as for the 34 lower extremities. Confidence limits are given at the 95% level [9].

ducer. in the

AJR:155,

studied,

Mountain MHz) and

of a linear-array

AL.

sions,

a 5-MHz

Sonography, Doppler (3.5

by means

artery;

superficial

arteries.

prospectively

Computed both pulsed

sonography

femoral

and

El

Stenosis

Occlusion

True . Negative

1 1

0 0

0 0

0 0

32 30

4 7 4

5 5 7

1 1

0 0

24 20

0

4

17

3

4

2

0

25

2

6

i

0

25

22

27

5

4

173

Note-Results of seven examinations were false negative for stenosis. Three of these resulted from dense calcifications, one each in the common femoral and mid and distal superficial femoral arteries. Four lesions were distal to a more proximal high-grade stenosis or occlusion, with three in the profunda femoris artery and one in the distal superficial femoral artery. No examination results were false negative for occlusion.

Downloaded from www.ajronline.org by 131.252.96.28 on 09/13/14 from IP address 131.252.96.28. Copyright ARRS. For personal use only; all rights reserved

AJR:155,

November

COLOR

1990

DOPPLER

OF

LOWER-EXTREMITY

imaging; and 1 0 lesions (45%) had both increased peak systolic velocity and narrowing of the lumen on color-assisted duplex imaging. Five significant stenoses were predicted but not confirmed by arteriography. These predictions were all based on the subjective view that the lumen was narrowed on color flow images. A doubling of the peak systolic velocity was not found at any of these sites. Seven significant stenoses were found at arteriography but not predicted prospectively (Table 1 ). Of these seven lesions, two were immediately distal to occluded segments, two were immediately distal to tight stenoses, and three were in heavily calcified segments. No arterial occlusions were missed (Table 1). Twenty-seven occlusions were correctly predicted on the basis of absence of both color flow and duplex Doppler signals. Four occluded

ARTERIAL

DISEASE

1087

segments were predicted but not confirmed by arteriography. All four of these were at the adductor canal in the distal superficial femoral artery and two were also distal to more proximal occlusions of the superficial femoral artery. Incidental findings included two popliteal aneurysms, one thrombosed and the other with significant intramural thrombus. The arteriogram was not diagnostic in either case. One hundred seventy-three segments were correctly predicted to be patent and without stenosis on the basis of colored flow in the lumen of greater than 50% and absence of a doubling of peak systolic velocities in these segments. The sensitivity of the technique for detecting a segmental stenosis, exclusive of an occlusion, was 0.76 (22/29), for a specificity of 0.97 (1 73/1 78) and an accuracy of 0.94 (195/ 207). When both occluded and stenotic segments were ana-

Fig. 1.-Images showing ease with which color Doppler sonography helps in identifying normal arterial branches. This 21-year-old woman with vasculitis affecting the tibioperoneal arteries has, nevertheless, a normal femoropopliteal system. A, Longitudinal color flow image clearly shows popliteal artery (arrowheads) and smaller geniculate branch (arrow). B, Doppler spectrum is sampled from corresponding popliteal artery. This shows normal triphasic velocity profile with late systolic reversal (arrowhead). (Figure is printed with cephalad to reader’s right and caudad to reader’s left.) C, Corresponding arteriogram shows normal popliteal artery (arrowheads) and geniculate branch (arrow), which courses superficial to popliteal artery.

Fig. 2.-Images ample

of focal

showing lesion

typical cx-

amenable

to an-

gioplasty detected prospectively by color-assisted duplex sonography. This 76-year-old man had claudication. A, Longitudinal color flow scan of mid superficial femoral artery (right side is proximal, left is distal) shows marked narrowing of rowhead). Note also calcific plaque (open lesion, note turbulence sal encoded as blue to each other (curved B, Corresponding

flow channel (arshadowing from arrows). Distal to with flow reverand red adjacent arrows). arteriogram confirms high-grade stenosis (arrow) seen on color Doppler sonogram. Lesion was subsequently dilated with balloon an-

gioplasty.

Downloaded from www.ajronline.org by 131.252.96.28 on 09/13/14 from IP address 131.252.96.28. Copyright ARRS. For personal use only; all rights reserved

Fig. 3.-Images

showing

typical

example

of occluded

superficial

femoral

artery,

making

patient

an unlikely

candidate

for angioplasty.

Duplex

sonography

is used to confirm diagnosis of occlusion suggested by absence of color flow signals. This 59-year-old man had severe claudication. A, Longitudinal sonogram of distal superficial femoral artery with Doppler gate over artery reveals no detectable flow. Deflections above and below baseline represent noise due to high gain setting. B, Corresponding transverse color flow sonogram at same level shows flow in superficial femoral vein (curved arrow). No flow is seen in superficial femoral artery, which is immediately superficial to vein (arrowheads). Small flow channels encoded as red (straight arrows) are due to flow in multiple small collateral arteries. C, Arteriogram confirms

occlusion

of distal

superficial

femoral

artery

with multiple

lyzed, sensitivity was 0.88 (49/56), specificity 0.95 (173/182), and accuracy 0.93 (222/238). The average length of the duplex sonographic evaluation of both lower extremities in a given patient was 29.5 mm (± 4.9 mm SD; range, 20-38 mm).

Discussion We have used color-assisted duplex sonography, a technique that combines the subjective assessment of the lumen offlowing arterial blood with the selective sampling of Doppler spectra and the derived peak systolic velocity ratio, to rapidly evaluate lower-extremity arterial segments for stenoses, occlusions, or patency (Fig. 1). In our series of 1 7 patients, 238 arterial segments were evaluated, 29 of which had stenoses and 27 of which were occluded. Twenty-two of the 29 stenotic segments and all occluded segments were correctly predicted before arteriography (Figs. 2 and 3). In the stenotic segments, an isolated doubling of peak systolic velocity was present in three instances, diminished color flow in nine instances, and both criteria in 1 0 instances. In vessels that were heavily calcified, the only marker of a significant stenosis was the isolated increase of peak systolic velocity. In less calcified vessels with focal stenoses 0.5-4 cm long, an increase in peak systolic velocity and narrowing of the lumen were readily definable. In irregular stenotic segments longer than 5 cm, instead of an increase in velocity, there was in fact a velocity decrease. These lesions were identified only by a narrowed lumen seen on transverse color sonography. The false-positive diagnosis of stenosis was always made by a narrowed transverse colored lumen (five cases). This occurred contiguous to long segmental narrowings, where velocities are low. The size of the colored flow in the lumen was most likely underestimated because blood flow velocities dropped below the flow sensitivity of the device

collaterals.

A and B were

obtained

at level

of arrows.

near the wall of the arteries. We were careful to visualize each segment transversely, because nonsignificant (30 mm Hg) is measured between two cuffs at different locations in the extremity (i.e., upper thigh and lower thigh) [4, 5]. Accuracy of segmental Doppler pressures for femoropopliteal disease (>50% narrowing of lumen diameter) is moderate, with a sensitivity varying from 0.67 to 0.89 depending on the presence or absence of concomitant aortoiliac disease [1 1 -1 2]. The technique cannot determine the nature of lower-extremity arterial lesions (i.e., occlusion vs focal stenosis) or their respective lengths. Preangiographic color-assisted Doppler sonography influenced the diagnostic approach in six patients. Of the four patients in whom severely diminished inflow at the level of the common femoral artery was caused by severely diseased or occluded iliac arteries, two cases had not been predicted clinically. The two cases of common femoral artery disease were not obvious clinically. Focal stenoses of the superficial femoral artery in two patients were correctly predicted to be amenable to angioplasty. The angioplasties were performed after the diagnostic studies. An additional patient was found to have bilateral popliteal aneurysms, one of which was thrombosed. apparent thrombus

these

The popliteal on angiography

aneurysm

that was

because

of the

patent

amount

lining it. Negative arteriographic findings are thought to occur in up to 36% of diagnosed

was

not

of mural

such as popliteal

aneurysms [1 3]. This patient has now had arterial bypass surgery after ligation of the popliteal aneurysm. Our experience suggests color-assisted Doppler sonography should be performed with longitudinal color flow imaging to locate sites where the Doppler spectrum is to be sampled [1 4]. This approach may also decrease study variability caused by the sonographer [1 5]. Transverse color Doppler imaging would then be used only as an adjunct for long segmental stenoses. The choice of sonographic imaging device also is important. We believe that steering the color Doppler angle separately from the gray-scale image and a high frame rate make it feasible to perform the examination in a reasonable amount of time. In summary, color-assisted Doppler sonography is a noninvasive tool helpful in assessing the extent of peripheral arterial disease. Subjective evaluation of the colored flow in the lumen and determination of peak systolic velocity can be used to rapidly predict segmental stenoses and occlusions in the lower limb arteries.

ACKNOWLEDGMENTS We thank Vennetta Hanley for secretarial assistance; Polak for editorial aid; and the technological staff of both vasive laboratory and angiography suites, without whom would not have been possible.

Jo-Anne the noninthis study

REFERENCES 1 . Murray AR Jr, Hewes AC, White RI Jr, et al. Long-segment femoropopliteal stenoses: is angioplasty a boon or a bust? Radiology i987;162:473-476 2. Krepel VM, van Andel GJ, van Erp WF, Breslau PJ. Percutaneous transluminal angioplasty of the femoropopliteal artery: initial and long-term results. Radiology i985;1 56:325-328 3. McLean L, Jeans WD, Horrocks M, Baird RN. The place of percutaneous transluminal angioplasty in the treatment of patients having angiography for ischaemic disease of the lower limb. Clin Radiol i987;38: 157-160 4. Lynch TG, Hobson RW II, Wright CB, et al. Interpretation of Doppler segmental pressures in peripheral vascular occlusive disease. Arch Surg 1984;1 19:465-467 5. Cutajar CL, Marston A, Newcombe JF. Value of cuff occlusion pressures in assessment of peripheral vascular disease. Br Med J 1 973: 2 : 392-395 6. Kohler TA, Nance DR. Cramer MM, Vandenburghe N, Strandness DE Jr. Duplex scanning for diagnosis of aortoiliac and femoropopliteal disease: a prospective study. Circulation 1987;76: 1074-1080 7. Jager KA, Phillips DJ, Martin AL, et al. Noninvasive mapping of lower limb arterial lesions. Ultrasound Med Biol i985;1 1 :515-521 8. McNeil BJ, Keller E, Adelstein SJ. Primer on certain elements of medical decision making. N Engl J Med 1975;293:21 1-215 9. Berry CC. A tutorial on confidence intervals for proportions in diagnostic radiology. AJR 1990;1 54:477-480 10. Kasai C, Namekawa K, Koyano A, Omoto A. Real-time two dimensional blood flow imaging using an autocorrelation technique. IEEE Trans Sonics Ultrason i985;SU-32:458--463 1 1 . Bell G, Nielsen PE, Wolfson B, Ulrich J, Engell HC. Lassen NA. Measurements of systolic pressure in the limbs of patients with arterial occlusive disease. Surg Gynecol Obstet 1973;1 36:177-181 12. Heintz SE, Bone GE, Slaymaker EE, Hayes AC, Barnes RW. Value of arterial pressure measurements in the proximal and distal part of the thigh in arterial occlusive disease. Surg Gynecol Obstet i978;i46:337-343 13. MacGowan SW, Saif MF, O’Neill G, Fitzsimons P, Bouchier-Hayes D. Ultrasound examination in the diagnosis of popliteal artery aneurysms. Br J Surg i985;72:528-529 14. Polak JF, Donaldson MC, Dobkin GA, Mannick JA, O’Leary DH. Early detection of saphenous vein arterial bypass graft stenoses by colorassisted duplex sonography: a prospective study. AJR i990;1 54: 857-861 1 5. Polak JF, Dobkin GA, O’Leary DH, Wang A-M, Cutler SS. Internal carotid artery stenosis: accuracy and reproducibility of color-Doppler-assisted duplex imaging. Radiology 1989;173:793-798

Determination of the extent of lower-extremity peripheral arterial disease with color-assisted duplex sonography: comparison with angiography.

Color-assisted duplex sonography combines color Doppler flow imaging with duplex sonography to rapidly survey the vasculature. Color Doppler sonograph...
996KB Sizes 0 Downloads 0 Views