Limitations of and Lessons Learned from Clinical Experience of 1,020 Duplex Arteriography

Patients and Methods

At our institution, from January 1, 1998, to January 1, 2005, 906 patients (51% men) requiring lower extremity revascularization underwent 1,020 duplex arteriograms. An attempt to image from the distal aorta to the pedal arteries was made in all patients. The selection of optimal inflow and outflow bypass anastomotic sites and angioplasty sites was based on a schematic drawing following DA examination. For bypasses, inflow disease was also assessed by an intraoperative pressure gradient between donor and radial arteries. Completion arteriography of the runoff vessels was obtained. Indications for the examinations included tissue loss (409), rest pain (221), claudication (310), acute ischemia (74), popliteal aneurysm (45), superficial femoral artery (SFA) aneurysm (2), abdominal aortic aneurysm (10), and failing bypass (55).

Results

The ages ranged from 30 to 98 years, with a mean of 73 ± 11 (SD) years. Fifty percent of the patients were diabetics. Previous procedures had been performed in 262 patients. DA was performed by six technologists (four of whom have medical degrees). DA was performed intraoperatively in 205 patients and in the remainder preoperatively in preparation for open revascularization. ^46,47

The resultant procedures based on DA included bypass to the popliteal artery (262) and bypass to an infrapopliteal artery (325), endovascular procedures (363), thrombectomy (11), embolectomy (9), inflow bypass procedures to the femoral arteries (46), débridement (4), amputation (8), and no intervention (75). The areas not visualized well included the iliac (73), femoral (26), popliteal (17), and infrapopliteal (221) arteries. Additional imaging after DA was deemed necessary in 102 cases to obtain enough information to plan lower extremity revascularization. The reasons for this were extensive ulcers (6), edema (10), severe arterial wall calcification (64), uncooperative patient (6), low flow (8), obesity (12) multiple previous surgeries (15), poor visualization of the origin of anterior tibial artery (1), and very poor runoff (18).

Factors associated with an increased need to obtain CA included diabetes mellitus (p < .001), infrapopliteal calcification (p < .001), older age (p = .01), and limb-threatening ischemia (p < .001). Factors not associated with the need to obtain CA included which technologist performed the examination, whether the technologist has a medical degree, or whether the patient had undergone previous revascularization.

Completion arteriography of the runoff vessels, in general, correlated well with the preoperative findings. However, in a few cases, DA and CA were in disagreement. In one case, DA identified the anterior tibial artery as patent. CA demonstrated distal anterior tibial disease, and an extension bypass was necessary. In another case, distal peroneal stenosis was identified on CA but missed on DA, necessitating balloon angioplasty of the distal peroneal artery. Both of these sites were noted to have severely calcified tibial vessels, suggesting that, once severe calcification is noted, DA is no longer reliable. Five cases of poor iliac visualization were noted in which graft pressure measurements at the completion suggested an inflow lesion. Angiography confirmed these lesions, which were treated in the same setting. One preoperative duplex arteriogram suggested that the distal posterior tibial artery was open; very low flow was noted in the distal posterior tibial artery. Completion angiography revealed severe disease in the distal posterior tibial artery, necessitating a jump to the plantar artery. In another case, a very wide wire loop in the occluded SFA segment during attempted subintimal dissection was noted, and extravasation was suspected on DA. This was confirmed with angiography, and the procedure was aborted. In one case, rupture of a heavily calcified SFA after balloon deflation was suspected. A small amount of extravasation was confirmed by arteriography, which stopped after protamine administration. In another case of duplex-guided angioplasty, owing to the inability to advance the wire to the tibioperoneal trunk with calcification and poor visualization, we converted to angiography, and the tibioperoneal trunk could be cannulated.

Discussion

Our experience has documented the progression of preoperative DA evolving into an integral part of the revascularization procedure itself. This incorporation of DA into the revascularization procedure is predicated on the need to accurately assess the entire arterial tree from the aorta to the pedal vessels before, during, and at the end of the procedure. ^48–50 Although the results of these interventions have already been described, the purpose of this series was to focus on some of the lessons that we have learned during this process.

Although this experience clearly demonstrates the feasibility of DA, as with any technology, it is important to appreciate its limitations. When severe tibial vessel wall calcification or extremely low-flow situations are encountered (PSV < 20 cm/s or volume flow of < 20 cc/min), DA can be unreliable, and alternative imaging modalities need to be employed. The key difference between using DA and preoperative CTA or MRA is that DA can alert the technician and the surgeon that it has become unreliable, whereas we have not been able to identify the factors that suggest when CTA or MRA is no longer supplying reliable data. The technologist can identify the areas that have severe calcification and where adequate visualization was not obtained or if extremely low flow was encountered. Based on the present experience, we suggest that about 90% of lower extremity revascularization procedures can be performed based on preoperative DA alone.

The key to setting up this process is the training method for the technologist and the surgeon interpreting the results and applying the information obtained. It is crucial that the accuracy of the results of each technologist be ensured, with at least the first 50 examinations being confirmed with another imaging modality before relying on DA as a sole imaging modality. ⁵¹ The most experienced technologist in our laboratory performs the examination and draws out the mapping. This is also verbally discussed with the interventionalist, specifying which areas were not visualized well and why. These results are then compared with the contrast angiogram, and any differences, and the causes of the differences, are reviewed. By the fiftieth examination, the interventionalist and the technologist know whether the DA performed by this team is adequate. By this time, the length of the examination also decreases from 90 minutes to a range of 45 to 60 minutes.

Part of the process also includes training technologists by bringing them to the operating room to appreciate the anatomy, the details of the intended procedures, and what information needs to be obtained for the procedure to take place. During the process of setting up these protocols, the interventionalist also has to be familiar with the data obtained, to be sure of its accuracy and how to apply it. This training process necessitates a change in mentality for the technologist and the interventionalist. Rather than just describing the disease, the technologist needs to present the necessary information to suggest treatment options. Inherent in this technique is a continuous quality assurance process that needs to be maintained with the constant feedback between the surgeon and the duplex arteriographer of expected findings, the actual intraoperative findings, and what interventions were actually performed.

Some of the issues with DA may necessitate some nonstandard methods of resolution. When patients experience severe ischemic pain that precludes the completion of the examination, premedication with analgesics can be helpful for a repeat examination. Additionally, since the examination does take the cooperation of the patient, having a family member present during the examination of a confused elderly patient can sometimes allow the examination to be completed. Visualization of the iliac arteries can sometimes be better accomplished by having the patient fast. Leg elevation for 24 to 48 hours can decrease calf edema, allowing adequate visualization of the tibial vessels.

Although insonation in the area of open ulcers or excessive scarring may not be possible, this also would not be an area in which one would want to perform anastomosis. Therefore, even if there is local poor skin condition, severe obesity, edema, or excessive depth of the vessel (such as the tibioperoneal artery and the origin of the anterior tibial artery), sufficient accurate information, can, at times, still be obtained to complete the needed intervention. For example, if a claudicant is found to have a patent popliteal artery and one-vessel runoff, but the other tibial vessels were unable to be fully assessed, one may decide to perform femoropopliteal bypass even though the tibial vessels were not completely assessed. If a diabetic patient with gangrene is found to have a patent dorsalis pedis artery and an adequate conduit, and the anterior tibial artery is too calcified to visualize, one may choose to perform a bypass to the dorsalis pedis because the anterior tibial artery will probably be a much more difficult artery to perform the distal anastomosis, as these vessels identified to have severe calcification by DA were found to be rock-like on exploration. ⁵² If one prefers to perform a bypass to a clearly patent dorsalis pedis rather than to a peroneal artery in the presence of an adequate conduit, lack of information on the peroneal artery may not really change the treatment plan.

If the wall of the vessels where anastomosis is to be performed is found to be thickened or calcified with DA, one may choose to perform the anastomosis in another area. This vessel may be patent and be identified by CA, CTA, or MRA as an adequate vessel for the anastomosis. However, operative attempts at intervening at these sites can often lead to further unplanned exploration for an adequate site for the anastomosis.

It is important to note the advantages of the portability of the duplex machine itself. Because DA examinations can be performed at the bedside, in the operating room, or in the holding area, the time spent transporting the patient and the personnel required is significantly reduced. Additionally, obtaining the CA or the MRA and their interpretation can entail a delay in the definitive treatment of a severely ischemic limb in a debilitated patient, as well as take a toll on the operative team. With DA, once the patient is identified as needing urgent revascularization, the machine and the technician can be brought to any part of the hospital for an abbreviated directed examination.

Given that DA is not just a luminal technology, it can be used to assess the actual disease of the vessel. High-frequency duplex imaging can assess the luminal diameter and thickness of the wall down to approximately one-tenth of a millimeter. This can be important not only in the selection of appropriate balloons and stents but also, especially, with cutting balloons. In addition, DA has the ability to more accurately assess the chronic nature of the occlusion. It is possible to differentiate between an isolated chronic SFA occlusion and an acute embolism with little underlying disease or acute thrombosis with severe underlying atherosclerotic disease. Aneurysmal vessels with partial thrombosis may have little to no luminal dilatation and may be undetectable by CA. Similarly, ulcerated and irregular plaques that may be a source of embolization are also poorly assessed by CA. High-resolution DA more clearly visualizes these plaques. Consequently, we have found this imaging modality to be particularly valuable in determining patient management compared with other technologies.

Furthermore, the hemodynamic information obtained using DA may alter patient management. Volume flow measurements can help assess whether the visualized lesion is hemodynamically significant and determine whether repair of the lesion may be beneficial. For example, a poorly visualized iliac plaque with little change in the ratio of PSVs (< 2) may suggest that the lesion may not be of clinical significance. Lesions that are poorly visualized owing to severe calcification with elevated ratios distal to the obscured lesion suggest a hemodynamically significant lesion. Other luminal imaging modalities do not readily furnish the details of inadequately visualized lesions or of lesions of questionable hemodynamic significance. With CA, the hemodynamic significance of a moderate lesion is often judged subjectively rather than on objective data.

During DA, we routinely perform venous mapping to identify usable veins for harvest, thereby avoiding the additional time and energy needed in pursuit of veins of good quality and caliber. If no usable vein is identified and a bypass to the tibial vessels is required, the diameter of the tibial veins may be measured for a possible prosthetic bypass with a distal fistula. Finally, examination of the subclavian-axillary segment may be performed as a possible inflow source for debilitated patients with severe aortoiliac disease. This is accomplished without the risk of an additional thoracic aortogram or the time needed for an additional thoracic MRA.

Overall, our most common problem with DA has been with calcification. However, as we have indicated before, some techniques can be used to obtain the necessary information even with severely calcified vessels, such as using multiple projections and SonoCT (ATL, Phillips, Bothell, WA). ⁵³ Although each individual field of view is limited, we have demonstrated that the vessels from the aorta to the pedal vessels can be visualized using DA. Although the role of contrast-enhanced DA has not been explored, the data on the use of contrast-enhanced duplex examinations for the carotid arteries and aortic endoleaks make this a possible future adjunct. ^54–58 During our experience, we noted that the duration of the examination became much shorter with increased experience. In our initial experience, the entire examination took us as long as 90 minutes. By the end of our experience, the examination could take as little as 25 minutes.

These data are based on our previous experience with preoperative DA and represent an extension of our use of duplex imaging to become an integral part of the intervention itself. To perform duplex-guided angioplasties, one needs extensive experience with preoperative DA. Furthermore, since duplex imaging is also being used for duplex-guided thrombin injections of arterial pseudoaneurysms, endovenous interventions of the great saphenous vein, placement of inferior vena cava filters, carotid angioplasties, and lower extremity angioplasty, we propose that this new field be termed interventional ultrasonography. ^59–72 Based on this experience, we submit that DA can be used for evaluation of the lower extremity arteries in patients being evaluated for lower extremity revascularization. The advantages, limitations, and early results of the technique suggest that further study is warranted.