Color Doppler Evaluation of the Pedal Arteries as a Predictive Tool in Peripheral Arterial Disease
Background. Early diagnosis of peripheral arterial disease (PAD) is important. Screening tests like pulse palpation and ankle-brachial pressure index (ABPI) measurement have varying levels of objectivity, accuracy, and reproducibility. Objective. To evaluate the ability of color Doppler study of the dorsalis pedis (DP) and posterior tibial (PT) arteries to predict the presence of significant PAD in that limb. Methods. This was a cross-sectional study done in a tertiary hospital, which involved patients who had undergone color Doppler studies of the lower limbs during a 6-month period. Color Doppler study images of the DP and PT arteries were arranged in a random manner and independently evaluated by 2 investigators who were unaware of the status of PAD in these limbs. Each investigator evaluated the images for peak systolic velocity and waveform pattern, trying to predict the possibility of PAD in that limb. Results. The 2 investigators showed excellent agreement in predicting PAD, which was slightly better with PT (92.31%) than with DP (90.57%). The first investigator showed a higher sensitivity while the second investigator showed a higher specificity in the prediction of PAD. Both investigators showed excellent agreement regarding the waveform pattern in the DP and PT arteries, which was 86.8% and 82.7%, respectively. Conclusion. Color Doppler evaluation of the pedal arteries could be a good screening tool for PAD and could have some potential to replace other screening tests. Further evaluation of this tool is required in a large-scale community-based setting.
VASCULAR DISEASE MANAGEMENT 2011;8(10):E172–E174
Key words: predicting significant PAD, color Doppler, pedal arteries, peripheral arterial disease
Peripheral arterial disease (PAD) involving the lower limbs is seen in 3–10% of the population and can be even higher (up to 15–20%) in people aged over 70 years.1 Most people are asymptomatic. However, intermittent claudication is common and can progress to develop severe symptoms in acute or chronic critical limb ischemia (CLI). PAD is a common cause of lower limb amputation and can significantly impair the quality of life, making its early identification very important.
This study was performed to evaluate the ability of the color Doppler study of the dorsalis pedis (DP) and posterior tibial (PT) arteries to predict the presence of significant PAD in that limb.
Materials and methods
This cross-sectional study was done in a tertiary level hospital. The institutional ethics committee had given their approval for the study. Patients who had undergone a complete color Doppler evaluation of the arteries of the lower limbs (either unilateral or bilateral) for suspected PAD over a 6-month period were included in the study. Cases were excluded if the documentation was incomplete, if the color Doppler images of the dorsalis pedis artery (DPA) or the posterior tibial artery (PTA) were unavailable on the picture archival and communication system (PACS), or if there was a previous history of vascular surgery or endovascular treatment for the limb.
The color Doppler evaluation of the patients had been done earlier as per the usual protocol by experienced radiologists. The DPA on the dorsum of the foot and the PTA at the level of the ankle had been examined at rest. Grey scale images and waveform patterns had been documented to demonstrate each artery and the blood flow within.
The color Doppler report was used as the reference standard to characterize each leg as having PAD or not. PAD was considered present if there was any stenotic or occlusive lesion involving the arteries, whether hemodynamically significant (> 50%) or not (< 50%), irrespective of the number of lesions.
Each color Doppler image was demonstrated as a picture slide of either the DPA or the PTA showing the waveform pattern and mentioning the name of the artery (Figure 1). All patient identifiers were removed and each slide was given a code number. These slides were then rearranged in a randomized manner using a computer generated randomization sequence and sent as an email attachment to 2 consultant radiologists for independent evaluation. The initial investigator who collected, rearranged, and sent the images was the only person having knowledge of the patient data and the status of PAD in each image. The other 2 investigators were unaware of the patient data and interpreted the images independently. After describing the waveform pattern (mono, bi, or triphasic) in each slide, they stated whether they felt that that limb had PAD or not. The impressions of the investigators were collected and analyzed to calculate the ability of pedal artery Doppler to predict PAD. The agreement between the investigators was also calculated.
The measures of diagnostic accuracy (sensitivity, specificity, positive and negative predictive values) were calculated to evaluate the ability of pedal artery color Doppler to detect PAD. The inter-investigator agreement was calculated using the kappa statistic, which corrects for chance.2 This was interpreted depending on the kappa value: poor (kappa = 0 to 0.25), moderate (0.25 to 0.5), good (0.5 to 0.75), and excellent (0.75 to 1). All data analysis was done using STATA 10.0 (StataCorp LP, College Station, Texas).
The first investigator showed a higher sensitivity in picking up PAD, which was higher with PTA images (sensitivity 74.4%, specificity 92.3%) than with DPA images (sensitivity 69.2%, specificity 64.3%). The second investigator showed a higher specificity in ruling out PAD, which was higher with PTA images (sensitivity 66.7%, specificity 100%) than with DPA images (sensitivity 61.5%, specificity 78.6%). Both investigators showed a higher sensitivity with PTA images than with DPA images. There was excellent agreement between the investigators for a diagnosis of PAD, the correlation being higher for PTA images (92.3%, kappa value of 0.85) than for DPA images (90.6%, kappa value of 0.81). Both investigators showed excellent agreement on the waveform pattern, which was higher with DPA images (86.8% with a kappa value of 0.8) than with PTA images (82.7%, with a kappa value of 0.73).
The opinions of the investigators on waveform patterns in the DPA and PTA are given in Table 3. There was no disagreement when an artery was occluded. Of the DPA images, 5 cases were considered monophasic by the first investigator, 3 of them were regarded as biphasic, and 2 of them regarded as triphasic by the second investigator. Also, 2 cases, which were considered biphasic by the first investigator, were regarded as triphasic by the second investigator. Of the PTA images, 2 cases, which were regarded as monophasic by the first investigator, were regarded as biphasic by the second investigator. Six cases that were regarded by the first investigator as biphasic were regarded as triphasic by the second investigator. One case that was considered triphasic by the first investigator was considered biphasic by the second investigator.
It is important to identify PAD when it is asymptomatic because this is when the treatment would be effective both in terms of results as well as cost. Lifestyle modification and medications could be planned. Further tests could be planned to evaluate the rest of the vascular system such as the coronaries, the cerebral circulation, and other limbs.
Screening tests for PAD are used in an outpatient department or a community setting along with clinical examination. Palpation of foot pulses is commonly used, but is known to have a poor sensitivity and specificity, and high interobserver variation.3,4 Ankle-brachial pressure index (ABPI) is widely used in screening programs because it provides an overall assessment of blood supply to a limb, is noninvasive, easy to perform, and requires only basic Doppler equipment.5,6 Sensitivity, specificity, positive, and negative predictive values are being estimated in the ongoing ABPI collaboration of 16 studies worldwide.7 Wall calcification of the distal arteries, as commonly occurs in diabetes, can result in falsely raised ankle pressures, thus underestimating the extent of disease. ABPI is known to have a poor sensitivity in low-grade disease.5,8,9 Using ABPI measurements, the prevalence of asymptomatic PAD in a population aged between 55 and 74 years was found to be 8%.10
Color Doppler evaluation is also noninvasive, but expensive and requires a trained operator. Scanning of all the arterial segments is resource intensive and technically demanding. A better screening test for PAD should be easy to use even by people with limited training. Doppler provides anatomic and physiologic information in terms of flow velocity and waveform evaluation at the vascular segment being studied. A study comparing color Doppler and resting ABPI found an agreement of 83% (kappa 0.66) between color Doppler and resting ABPI.8 A resting ABPI of less than 0.6 provided 100% agreement with color Doppler. When ABPI was normal (0.9), the agreement was only 76%.8 Thus ABPI detected more serious disease.
There could be a difference in waveform interpretation during Doppler procedures because of different operators. Because color Dopplers were personnel-specific, this might have affected the quality of representation of findings on the image.
A good screening test should be better than ABPI and overcome its disadvantages, while at the same time being not as time or resource intensive as a full color Doppler evaluation of the lower limbs. With the good sensitivity and specificity demonstrated by this study, color Doppler evaluation of the pedal arteries could have potential to be an effective screening tool, which will be tested further in a community setting.
A limitation of this study is that the prevalence of patients affected with PAD in a tertiary hospital would be different from that in the community or in an outpatient clinic. Further plans to evaluate the applicability of this screening tool in a community setting is under way.
Color Doppler evaluation of the DPA and PTA was found to have a good sensitivity and specificity in predicting PAD. Since this is less technically demanding than a complete color Doppler study of the lower limbs, it could be a screening test for PAD and requires further evaluation in a community setting.
- Selvin E, Erlinger TP. Prevalence of and risk factors for peripheral arterial disease in the United States: Results from the National Health and Nutrition Examination Survey, 1999-2000. Circulation. 2004 Aug 10;110(6):738-743.
- Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977 Mar;33(1):159-174.
- Lundin M, Wiksten JP, Peräkylä T, et al. Distal pulse palpation: Is it reliable? World J Surg. 1999 Mar;23(3):252-255.
- Brearley S, Shearman CP, Simms MH. Peripheral pulse palpation: An unreliable physical sign. Ann R Coll Surg Engl. 1992 May;74(3):169-71.
- Nuzzaci G, Giuliano G, Righi D, Baroncelli T, Lotti A, Marinoni M. A study of the semeiological reliability of dorsalis pedis artery and posterior tibial artery in the diagnosis of lower limb arterial occlusive disease. Angiology. 1984 Dec;35(12):767-772.
- Doobay AV, Anand SS. Sensitivity and specificity of the ankle-brachial index to predict future cardiovascular outcomes: A systematic review. Arterioscler Thromb Vasc Biol. 2005 Jul;25(7):1463-1469.
- Fowkes FG, Murray GD, Butcher I, et al; for the Ankle Brachial Index Collaboration. Ankle brachial index combined with Framingham Risk Score to predict cardiovascular events and mortality: A meta-analysis. JAMA. 2008 Jul 9;300(2):197-208.
- Allen J, Oates CP, Henderson J, et al. Comparison of lower limb arterial assessments using color-duplex ultrasound and ankle/brachial pressure index measurements. Angiology. 1996 Mar;47(3):225-232.
- Stein R, Hriljac I, Halperin JL, Gustavson SM, Teodorescu V, Olin JW. Limitation of the resting ankle-brachial index in symptomatic patients with peripheral arterial disease. Vasc Med. 2006 Feb;11(1):29-33.
- Fowkes FG, Housley E, Cawood EH, Macintyre CC, Ruckley CV, Prescott RJ. Edinburgh Artery Study: Prevalence of asymptomatic and symptomatic peripheral arterial disease in the general population. Int J Epidemiol. 1991 Jun;20(2):384-392.
From the Christian Medical College Vellore, Department of Radiology, Vellore, Tamil Nadu, India.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.
Manuscript submitted July 14, 2011, provisional acceptance given August 3, 2011, final version accepted August 8, 2011.
Address for correspondence: Dr. Chiramel George Koshy, MD, Christian Medical College Vellore, Department of Radiology, Christian Medical College Hospital Campus, Ida Scudder Road, Vellore, Tamil Nadu 632004, India E-mail: [email protected]