---

Accuracy of duplex ultrasonography for the diagnosis of arterial occlusion of infrainguinal segment arteries
(Portuguese PDF version)

Carlos Alberto Engelhorn¹, Ana Luiza Engelhorn¹, Marco Antonio Lourenço¹, Renata Pullig¹, Emerson Ribas¹, Fausto Miranda Jr², Emil Burihan²

1. Hospital Santa Casa de Curitiba/ Pontifícia Universidade Católica do Paraná.
2. Universidade Federal de São Paulo - Escola Paulista de Medicina (School of Medicine).

Correspondence:
Carlos Alberto Engelhorn
Rua Deputado Heitor Alencar Furtado, 1720/901
CEP 81200-110 - Curitiba - PR
Tel.: +55 41 279.1241/ +55 41 362.0133/ +55 41 9985.4935
Fax: +55 41 362.0133
Email: engelhor@bsi.com.br

---

ABSTRACT

Objectives: The present study aims at assessing the correlation between a new ultrasonographic imaging method and digital arteriography in the investigation of occluded infraguinal arteries in adult patients with severe chronic obstructive disease.

Methods: Fifty lower limbs of 29 patients were prospectively studied by duplex ultrasonography (DUS). The results of this assessment with DUS were compared, in a double-blind, with those of arteriography. Sensitivity, specificity, positive predictive value, negative predictive value and general accuracy of DUS were evaluated. The correlation between DUS and arteriography was also analyzed with the aim of identifying arterial obstruction and extension, and determining the level of arterial reconstruction distal to the obstruction.

Results: By considering the arterial segments in which both vascular imaging methods were compared and in which the presence or absence of arterial obstruction was determined, DUS showed sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 100%, 100%, 100%, 100% and 100%, respectively, for the femoropoplietal segment; 80%, 100%, 100%, 85% and 91% for the anterior tibial artery, and 90%, 96%, 90%, 96% and 94% for the posterior tibial artery. The fibular artery had sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 100%, 90%, 60%, 100% and 91%, respectively; however, comparison was only possible in a low number of cases. By analyzing our results, there was good correlation between DUS and arteriography in the investigation of occlusion of the femoropopliteal segment and in anterior and posterior tibial arteries. DUS was able to determine the presence, location and extension of distal reconstruction in occlusions of the femoropopliteal segment and anterior and posterior tibial arteries.

Conclusion: DUS and arteriography revealed some restrictions in the investigation of the fibular artery.

Key words: ultrasonography, angiography, arterial obstruction
Palavras-chave: ultra-sonografia, arteriografia, obstrução arterial.

J Vasc Br 2002;1(1):55-64.

---

INTRODUCTION

Successful arterial revascularizations in the infrainguinal segment depends on factors related to the natural history of arterial disease, surgical indication and technical factors associated with the procedure itself. Surgical indication depends on a comprehensive preoperative assessment of arterial flow by means of contrast and noninvasive vascular imaging methods.

Although arterial revascularization had been technically standardized in the past by the experimental studies conducted by Carrel & Guthrie,¹ and been clinically performed in isolated attempts, only in the 1950's, with the experience acquired in the war period, it turned out to be an efficient treatment. For that purpose, the assessment of the involved arteries was imperative; the assessment included routine preoperative arteriography. Nevertheless, since this exam is invasive, it has complication risks in terms of contrast medium and the procedure itself. Due to these potential risks, alternative noninvasive diagnostic methods have been used in the last two decades.

Technological advances in ultrasound imaging, especially with color flow mapping and Power Doppler, allowed for a reliable anatomical and hemodynamic assessment of arterial flow. This was an alternative to arteriography in cases of arterial revascularization of the infrainguinal segment. However, the presentation of conventional ultrasound images is restricted to small arterial segments, shown separately, which requires further knowledge about the method from vascular surgeons so that surgical indication can be safely made by using only color echo-Doppler.

To facilitate the interpretation of ultrasonographic vascular images, an imaging method called duplex ultrasonography with Power Doppler mapping appeared in the 1990's. This method allowed for the acquisition of real-time vascular images, with spatial presentation that resembles that of arteriography, that is, a sole ultrasound image of a long vascular segment invoving all anatomical changes that are present.

The aim of this study is to compare the anatomical information of duplex imaging of infrainguinal arteries with that offered by arteriography, to assess the accuracy of duplex ultrasonography in relation to arteriography, and to identify and determine the extension of arterial occlusion in patients with indication for arterial revascularization.

METHODS

A prospective, double-blind study was carried out at the Service of Angiology and Vascular Surgery of the Hospital da Santa Casa de Misericórdia, Curitiba, State of Paraná, Brazil, with the aim of checking the accuracy of DUS for the diagnosis of arterial occlusion of the infrainguinal segment, compared to arteriography, which is regarded as gold standard. The study was approved by the Ethics and Research Committee of the Hospital Santa Casa de Curitiba (PUCPR) and by the Universidade Federal de São Paulo.

Fifty lower limbs of 29 patients were assessed by DUS and digital arteriography: 27 right limbs and 23 left limbs. Of the 29 patients, 22 were men and seven were women aged between 46 and 81 years (on average 64 years), presenting severe chronic arterial insufficiency arterial and arterial occlusive disease of the infrainguinal segment.

Six of the patients showed restrictive intermittent claudication, with an average claudication distance of 85 meters; five patients had ischemic rest pain, and 18 patients showed trophic lesions.² Patients with mild to moderate intermittent claudication, stenosing or aneurysmal arterial disease, patients who were allergic to the iodinated contrast medium, chronic renal insufficiency and behavioral disorder were not included in the study. According to these criteria, eight infrainguinal segments of eight patients were excluded (two infrapatellar amputations and six normal contralateral lower limbs).

All the patients who were included in the study were submitted to duplex ultrasonography by the same ultrasonographer and to arteriography by the same vascular radiologist. Both the ultrasonographer and radiologist did not have any previous information about the diagnosis. The interval between ultrasonography and arteriography did not exceed five days.

Anatomical assessment by Duplex Ultrasonography

The anatomical assessment of arterial occlusion by DUS was performed on the femoropopliteal and popliteal-pedal segments with flow mapping with Power Doppler, according to the following technique:

a) DUS was preceded by simple ultrasound and conventional color echo-Doppler in order to investigate the arterial anatomy of the patient, to detect arterial occlusions and to guide the acquisition of duplex images;

b) Duplex images were obtained with 5 to 7 MHZ transducers, by real-time continuous mapping of the femoropopliteal segment and of the tibial anterior and posterior, and fibular arteries. The images were created by a data processor able to generate continuous images of more extensive areas, combined with previous static images, stored in the equipment's memory. Recognition and registering techniques were used to restore the movement and rotation of the transducer (Figure 1);

c) The depth of the image, the gain and scale of the Power Doppler were adjusted according to each examined artery.

Figure 1 - DUS image of femoro-popliteal segment.

The citeria used for the anatomical assessment by means of duplex ultrasonography were the following:

- in the femoropopliteal segment, the origin of the deep femoral artery was used as reference point for locating the onset of arterial occlusion, while the articular interline of the knee was used to determine distal swelling;

- in the popliteal-podal segment, the knee articular interline was used as reference point for locating the onset of arterial occlusion, while the intermalleolar line was used to determine distal swelling;

- The articular interlines of the knee and ankle were identified by conventional ultrasonography and marked on the patient's skin, as well as distal swelling. A millimeter ruler was used to measure the corresponding distance;

- The extension of arterial occlusion was measured on the duplex image itself by direct ultrasound equipment measurement.

Anatomical assessment by arteriography

The anatomical assessment of arterial obstructions by arteriography was performed on the femoropopliteal and popliteal-pedal segments with digital subtraction, selective arterial catheterization, by means of brachial dissection or femoral artery puncture. The same criteria for ultrasonography were used for arteriography. Arteriographic measurements were made with the same standard radiopaque ruler, positioned parallel to the arterial segments of the patients at the time of exam.

Documentation

Duplex ultrasound images were stored, in real time, on a videocassette and documented by color photographs printed by the equipment, whereas arteriographies were documented on x-rays with image subtraction.

Equipment

A Siemens Sonoline Elegra was used for the ultrasonographic assessment of the arteries and Siemens Angioscopy with digital subtraction was used for the arteriographies.

Statistical analysis

The DUS quality variables were the following: specificity, sensitivity, positive predictive value, negative predictive value and accuracy.

To compare the quantitative variables (location, extension and swelling), we used Student t test for paired samples. P values less than 0.05 were considered statistically significant. To determine the association between the two exams, we employed Kappa statistics (values between -1 and +1). A significance level of 5% was considered for all tests.

RESULTS

Assessment of the femoropopliteal segment

Of the 50 evaluated femoropopliteal arterial segments, it was possible to determine the presence or absence of arterial occlusion by both vascular imaging methods, without technical restrictions, in 49 (98%). Among these, 15 segments were considered to be patent by both methods and 34 segments occlusion was detected by DUS and arteriography. In the remaining cases, the vascular surgeon was not able to accurately define the origin of deep femoral artery, which made the assessment of measurement parameters impossible. Kappa statistics, estimated for the detection of patency or occlusion of the femoropopliteal segment, was 0.9539. The 95%CI was 0.8658.¹ The hypothesis test showed that there was no statistically significant difference (P<0.00001) between the two methods.

By considering only the cases (49/50) in which patency or occlusion of the femoropopliteal segment was accurately determined by both methods, DUS had specificity, sensitivity, positive predictive value, negative predictive value and accuracy of 100% when compared to arteriography. Examples of patency of the femoropopliteal segment, detected by both methods, are shown in Figures 2 and 3.

Figure 2 - Normal DUS of the femoro-popliteal segment.

Figure 3 - Normal arteriography of the femoro-popliteal segment. Same patient shown in Figure 2.

According to DUS, the occlusions of the femoropopliteal segment were often located 11.5 cm (+/-10.7 cm) below the origin of the deep femoral artery, with an average extension of 22.4 cm (+/-12.7 cm). At arteriography, occlusions usually appeared 11.9 cm (+/-11.3 cm) below the origin of the deep femoral artery, with an average extension of 22 cm (+/-13.3 cm). No statistically significant difference was observed between the two methods (P=0.722).

Of the 34 segments with occlusion, detected by both methods, 30 showed distal suprapatellar swelling, and four of them revealed infrapatellar swelling. The average distance of distal swelling at DUS was 10.3 cm (+/-6.0 cm) and 9.8 cm (+/-5.5 cm) at arteriography. Although DUS overestimated the extension of the occlusions in some millimeters, no statistically significant difference (P=0.3446) was observed between the two methods.

Assessment of the anterior tibial artery

Of 50 anterior tibial arteries, the investigation made by DUS and arteriography was considered to be poorly conclusive in six arteries. Other three arteries were considered to be free of hemodynamically significant lesions by DUS and occluded by arteriography (false-negatives). Arteriography was regarded as poorly conclusive by the vascular radiologist in 11 arteries. DUS showed six occluded arteries and five patent arteries.

Therefore, of 50 anterior tibial arteries, it was possible to obtain the same findings by both methods, without technical restrictions, in 30 of them (60%). Among these, 18 arteries were considered patent by both methods and 12 arteries were considered occluded by DUS and arteriography. The estimated Kappa statistics for the detection of patency or occlusion of the anterior tibial artery was 0.5783. The 95%CI was 0.4072; 0.7495. No statistically significant difference (P<0.00001) was found between the two methods.

By considering only the cases (33/50) in which patency or occlusion of the anterior tibial artery was determined, DUS revealed specificity of 100%, sensitivity of 80%, positive predictive value of 100%, negative predictive value of 85% and accuracy of 91%, when compared to arteriography.

To determine the correlation between both methods as to the identification of the arterial occlusion site, we considered the 12 arterial segments in which arterial occlusion was found by both exams. The occlusions of the anterior tibial artery were shown by DUS and arteriography to be respectively located on average 10.6 cm (+/-12.1 cm) and 8.3 cm (+/-12.3 cm) below the knee. No statistically significant difference (P=0.3409) was observed between the two methods.

Of the 12 anterior tibial arteries in which occlusion was detected in two arteries, we could not accurately determine the extension of the occlusion by either of the methods. As to the remaining 10 arteries, the occlusions of the anterior tibial artery showed an average extension of 33.8 cm (+/-23.6 cm) at DUS. At arteriography, the average extension of the occusions was 40.7 cm (+/-20.6 cm). Although DUS underestimated the extension of the occlusions, no statistically significant difference (P=0.2213) was found between the two methods.

Of the 12 anterior tibial arteries in which occlusion was detected in one artery, it was not possible to determine the accurate distal swelling, even though occlusion was detected by both methods. In the remaining 11 arteries in which we could determine distal swelling by both exams, the average distance of distal swelling above the intermalleolar line was 4.3 cm (+/-6.6 cm) at DUS and 2.4 cm (+/-3.9 cm) at arteriography. Although DUS overestimated distal swelling in the occlusions of the anterior tibial artery, no statistically significant difference (P=0.3446 ) was observed between the two methods.

Assessment of the posterior tibial artery

Among all the posterior tibial arteries studied here, DUS and arteriography failed to provide conclusive results in four arteries. Arteriography was considered to be poorly conclusive by the vascular radiologist in 10 arteries, while DUS detected five occluded arteries and five patent arteries.

In one of the arteries considered patent by DUS, arteriography regarded it as occluded (false-negative). One artery considered by DUS to be occluded was regarded as patent by arteriography (false-positive). The example of the false-negative case is shown in Figures 4 and 5.

Figure 4 - False-negative case of DUS involving the patency of the posterior tibial artery.

Figure 5 - Arteriographic aspect of false-negative case shown in Figure 4.

Therefore, of 50 posterior tibial arteries, in 34 of them (68%) both exams showed the same results, without technical restrictions. Among these, 25 arteries was considered patent by both methods and nine were regarded as occluded by DUS and arteriography. The estimated Kappa statistics for the detection of patency or occlusion of the posterior tibial artery was 0.5968. The 95%CI was 0.4209; 0.7727. The hypothesis test did not show any statistically signficant difference (P<0.00001) between the two methods.

By considering only the cases (36/50) in which patency or occlusion of the posterior tibial artery was determined by both methods, DUS had a specificity of 96%, sensitivity of 90%, positive predictive value of 90%, negative predictive value of 96% and accuracy of 94%, when compared to arteriography.

To assess the correlation between the methods as to the identification of the site and extension of the occlusion, we considered the nine arteries in which arterial occlusion was detected by both exams. According to DUS, the occlusions of the posterior tibial artery were often located 4.7 cm (+/-9.4 cm) below the knee, with an average extension of 55.1 cm (+/-17.2 cm). At arteriography, the occlusions were located on average 2 cm (+/-4 cm) below the knee, with an average extension of 46.7 cm (+/-20.1 cm). Although DUS overestimated the extension of the occlusions, no statistically significant difference as to the site of occlusion (P=0.5028) and extension (P=0.3688) was observed between the methods.

Therefore, considering the nine arteries in which we could determine distal swelling by both exams, the average distance of distal swelling above the intermalleolar line was a distância 1.1 cm (+/-3.5 cm) at DUS and 2.1 cm (+/-4.9 cm) at arteriography. Although DUS underestimated distal swelling in the occlusions of posterior tibial artery, no statistically significant difference (P=0.1692) between the methods.

Assessment of the fibular artery

Of the 50 investigated fibular arteries, 10 were considered to be poorly conclusive by both methods; 13 were regarded as poorly conclusive by arteriography, of which two were considered by DUS to be occluded and 11 to be patent; in three arteries, DUS was considered poorly conclusive, while arteriography detected occlusion in one of them and patency in the other two.

Two arteries considered occluded by DUS were regarded as patent by arteriography (false-positive). In these arteries with intense arterial calcification, DUS did not find any flow along their whole extension, consequently interpreting this as occlusion.

Therefore, among the 50 fibular arteries studied here, it was possible to find the same results by means of DUS and arteriography, without technical restrictions, in only 22 arteries (44%). Among these, 19 arteries were considered patent by both methods, and three of them showed occlusion at both DUS and arteriography. The estimated Kappa statistics was 0.3931. The 95%CI was 0.1815; 0.6048.

Considering only the cases (24/50) in which patency or occlusion of the fibular artery was detected by both methods, DUS had a specificity of 90%, sensitivity of 100%, positive predictive value of 60%, negative predictive value of 100% and accuracy of 91%.

The assessment of the correlation between the methods as to identification and extension of the occlusions, and determination of distal arterial swelling was limited to three arteries only. In addition, due to the small number of arteries, the comparison between the two methods was not considered.

DISCUSSION

Arteriography is a singular situation in the history of vascular surgery and preliminary instrumental knowledge in general. It was the preffered method for the investigation of peripheral arteries throughout the 20th century owing to its practical use and good correlation with surgical findings. In addition to this good correlation with surgical findings, the presentation of arteriographic images in the form of contiguous segments on x-ray films facilitates the interpretation by radiologists and vascular surgeons, which shows it can be used routinely with no difficulty.

However, some authors who studied the intra- and interobserver variation in the interpretation of arteriographic images showed that this interpretation is sometimes limited, even after the introduction of digital angiography with image subtraction.³,⁴,⁵ Prospective studies reveal an incidence of general and more severe complications related to arteriography between 0.5% and 2.9%.⁶,⁷,⁸,⁹

Despite its morbid potential, only in the late 1990's, with the improvement of vascular imaging techniques, the routine and indiscriminate use of arteriography, as well as its accuracy, began to be questioned. In general, scientific evidence confirm the generally good accuracy of color echo-Doppler for the investigation of infrainguinal arteries. However, the results obtained from color echo-Doppler and arteriography show some discrepancy as to the investigation of infragenicular arteries in relation to the femoropopliteal segment and to the assessment of fibular arteries in relation to tibial arteries. More recent literature reviews have assessed the accuracy of color echo-Doppler compared to arteriography in the diagnosis of lower-limb arterial diseases. The specific analysis of data on the occlusion of the femoropopliteal segment revealed an average variation of sensitivity and specificity of color echo-Doppler above 90%.¹⁰,¹¹ By focusing specifically on the infragenicular segment, studies have showed a better correlation of color echo-Doppler with arteriography for the investigation of occlusion of anterior and posterior tibial arteries (85 to 100 %) than of the fibular artery (69 to 94%).¹²,¹³

The accuracy of color echo-Doppler is a reference for the validation of new ultrasonographic techniques, while we do not have studies that assess the applicability and reliability of these new methods. DUS is a new form of presenting ultrasonographic vascular images, similar to arteriography, which is preceded by conventional color echo-Doppler. This preliminary assessment detects arterial occlusions and guides the acquisition of duplex images. Therefore, the accuracy of color echo-Doppler can directly interfere with the results of DUS.

In this study, the results of the comparison between DUS and arteriography used to detect femoropopliteal occlusions showed a perfect correlation between the methods (kappa=0.95). By considering the cases in which a comparative study of the infrainguinal arteries was possilbe, the overall accuracy of DUS was 100% for the femoropopliteal segment, 91% for the anterior tibial artery, 94% for the posterior tibial artery and 91% for the fibular artery. However, among the 50 assessments we made, it was possible to detect the presence or absence of occlusion by both methods, without technical restrictions, in 98% of the arteries of the femoropopliteal segment, 60% of the anterior tibial arteries, 70% of the posterior tibial arteries and only 45% of the fibular arteries. The assessment of fibular arteries showed that 26 arteries (52%) were considered to be poorly conclusive, of which 10 were poorly conclusive at both arteriography and DUS. Also, 13 arteries were poorly conclusive at arteriography alone and other three arteries at DUS, whihc shows that limitations were more frequent at arteriography than in DUS as far as the assessment of fibular arteries is concerned .

According to some authors, this occurs because of the limitations of arteriography as a gold standard due to the presence of lesions at multiple levels of the arterial tree, thus reducing the concentration of contrast medium in more distal arteries and the anatomical location of the fibular artery, which can restrict the investigation by color echo-Doppler due to their depth and to the presence of parietal calcifications.⁵,¹¹,¹⁴,¹⁵,¹⁶

In addition to the accuracy of color echo-Doppler in the investigation of arterial occlusion of the infrainguinal segment, it is also important to precisely locate the occlusion, establish its extension, and determine distal swelling.¹⁷,¹⁸,¹⁹ Among the scarce studies found in the literature on DUS, the one conducted by Kroger et al. tested the reproducibility and accuracy of DUS by means of repetitive measurements of linear distances with 150 mm in length. The authors concluded that DUS tends to underestimate the extension of occlusions.²⁰

In general, the present study allowed us to detect, measure arterial occlusions, and determine distal swelling properly, without statistically significant differences when compared to the results obtained from arteriography, by means of a standard technique (radiopaque ruler). By considering that the correction was made by means of standard deviation, DUS showed values similar to those of arteriography as to the location of arterial occlusions of the femoropopliteal segment, although it tended to overestimate the extension of occlusions in the anterior and posterior tibial arteries. In the assessment of the extension of arterial occlusions, DUS showed similar results to those of arteriography for the femoropopliteal segment and for the anterior tibial artery, although it tended to overestimate the results for the posterior tibial artery. On determining distal swelling, DUS revealed similar results to those of arteriography for the femoropopliteal segment, although it tended to overestimate the results for the anterior tibial artery and underestimate the values for the posterior tibial artery.

The variations of the values obtained for the location and extension of arterial occlusions by DUS can be partly explained by the scanning of the whole arterial extension in free-hand longitudinal sections, that is, speed and direction are controlled by the operator, not by the equipment. Therefore, a slight inclination of the transducer can cause curvatures on the images, which could affect the final distance.

At arteriography, the incidence of x-ray beams can affect the size of the assessed structures, that is, same-size structures, with different distances from the central beam, will be presented as different-sized on the x-ray.²¹

In addition to a presentation that is similar to that of arteriography, the accuracy and reproducibility of DUS can directly influence the selection of this new imaging technology as a preoperative evaluation method when compared to arteriography. In the mid-1990's, we had the first studies on the assessment of color echo-Doppler as a separate vascular imaging method for the indication and planning of infrainguinal arterial revascularizations, showing initial success rates for infrainguinal bypasses between 76 and 90% during the first months after the surgery.²²,²³,²⁴,²⁵,²⁶

In these studies that considered color echo-Doppler as a safe method for the indication of infrainguinal arterial revascularizations, it was necessary to draw an arterial map (diagram) based on the information provided by color echo-Doppler in order to guide the vascular surgeon, which show the limitations of the conventional color flow mapping technique.

DUS is a new technology that allows obtaining ultrasonographic vascular images similar to those at arteriography, and which offers easy interpretation of the results and practical clinical use. As every new technology, it should be extensively studied so that its practical use and reliability can be checked. On top of that, when compared to arteriography, DUS can cost 10 times less than digital angiography in some centers.²⁷

The authors conclude that DUS, based on preliminary information provided by color echo-Doppler, is a reliable method for the investigation of infrainguinal arterial occlusive disease that allows for continuous images similar to those of arteriography. In addition, it has a good correlation with arteriography in terms of assessing the location, extension, and distal swelling of arterial occlusions of the femoropopliteal segment and of the anterior and posterior tibial arteries. Because of this good correlation with arteriography, DUS is potentially useful as an evaluation method for the indication of infrainguinal arterial revascularizations, eliminating the regular use of arteriography.

REFERENCES

1. Carrel A, Guthrie CC. Reversal of the circulation in a limb. Ann Surg 1906;43:203.

2. Rutherford RB, Baker JD, Ernst CE, et al. Ad Hoc Committee of Reporting Standards SVS/ISCVS. Recommended standards for reports dealing with lower extremity ischemia: Revised version. J Vasc Surg 1997;26:517-38.

3. Smith TP, Cragg AH, Berbaum KS, Nakagawa N. Comparison of the efficacy of digital subtraction and film-screen angiography of the lower limb: Prospective study in 50 patients. Am J Roentenol 1992;158:431-36.

4. Leng GC, Whyman MR, Donnan PT, et al. Accuracy and reproducibility of duplex ultrasonography in grading femoropopliteal stenosis. J Vasc Surg 1993;17:510-7.

5. Larch B, Minar E, Ramazanali A, et al. Value of color duplex sonography for evaluation of tibioperoneal arteries in patients with femoropopliteal obstruction: A prospective comparison with anterograde intraarerial digital subtraction angiography. J Vasc Surg 1997;25(4):629-36.

6. Hessel SJ, Adams DF, Abrams HL. Complications of angiography. Radiology 1981;138:273-81.

7. Vaugh JR, Sacharias N. Arteriography complications in DAS era. Radiology 1992;182:243-6.

8. Sigstedt B, Lunderquist A. Complications of angiography examinations. AJR Am J Roentegenol 1978;130:455-60.

9. Egglin T, O'Moore PV, Feistein AR, Waltman AC. Complications of peripheral arteriography: A new system to identify patients at increased risk. J Vasc Surg 1995;22:787-94.

10. Kolemey MJW, Hartog D, Prins MH, Kromhout JG, Legemate DA, Jabobs MJHM. Diagnosis of arterial disease of the lower extremities with duplex ultrasonography. Br J Surg 1996;83:404-9.

11. Pemberton M, London NJM. Colour flow duplex imaging of occlusive arterial disease of the lower limb. Br J Surg 1997;84:912-9.

12. Moneta GL, Yeager RA, Antonovic R, et al. Accuracy of lower extremity arterial duplex mapping. J Vasc Surg 1992;15:275-84.

13. Caster JD, Cimmings CA, Moneta GL, Taylor LM, Porter JM. Accuracy of tibial artery duplex mapping. J Vasc Technol 1992;16(2):63-8.

14. Flaningan DP, Willians LR, Keifer T, Schuler JJ, Behrend AJ. Prebypass operative arteriography. Surgery 1982;92(4):627-33.

15. Patel KR, Semel L, Clauss RH. Extended reconstruction rate for limb salvage with intraoperative prereconstruction angiography. J Vasc Surg 1988;7(4):531-7.

16. Sayers RD, Thompson NM, London NJM, Bell, PFR. Critical ischaemia of the lower limb: femorodistal bypass in preference to amputation. Br Med J 1992;304:898-9.

17. 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 1989;10:522-9.

18. Whyman MR, Gillespe I, Ruckley CV, Allan PL, Fowkes FGR. Screening patients with claudication from femoropopliteal disease before angioplasty using Doppler colour flow imaging. Br J Surg 1992;79:907-9.

19. Collier P, Wilcox G, Brooks D, Laffey S, Dalton T. Improve selection of patients for angioplasty utilizing colour Doppler imaging. Am J Surg 1990;160:171-4.

20. Kroger K, Massalha K, Dobonicci G, Rudofsky G. Siescape: a new sonography dimension with fictive images. Ultrasound Med Biol 1998;24:1125-9.

21. Kamm KF, Onnasch DGW. X-ray radiography. In: Lanzer P(editor). Diagnostics of Vascular Diseases. Heidelberg: Springer Verlag; 1997. p. 65.

22. Pemberton M, Nydhal S, Hartshorne T, Naylor AR, Bell PR, London NJ. Colour-coded duplex imaging can safely replace diagnostic arteriography in patients with lower limb disease. Br J Surg 1996;83:1725-8.

23. Mazzariol F, Ascher E, Salles-Cunha SX, Gade P, Hingorani A. Values and limitations of duplex ultrasonography as the sole imaging method of preoperative evaluation for popliteal and infra-popliteal bypasses. Ann Vasc Surg 1999;13:1-10.

24. Engelhorn CA, Engelhorn ALV, Gosalan CJ. Revascularização arterial fêmoro-poplítea sem arteriografia. In: Nectoux JL (editor). Ultrasonografia Vascular. 1ª ed. Rio de Janeiro: Revinter; 2000. p.120-7.

25. Mazzariol F, Ascher E, Hingorani A. Lower-extremity revascularisation without preoperative contrast arteriography in 185 cases: Lessons learned with duplex ultrasound arterial mapping. Eur J Vasc Endovasc Surg 2000;19:509-15.

26. Proia RR, Walsh DB, Nelson PR, et al. Early results of infragenicular revascularization based solely on duplex arteriography. J Vasc Surg 2001;33:1165-70.

27. Wain RA, Berdejo GL, Delvalle WN, et al. Can duplex scan arterial mapping replace contrast arteriography as the test of choice before infrainguinal revascularization ? J Vasc Surg 1999;29 (1):100-9.