Abstract: Objectives. To assess the relationship between the resting (RG) and hyperemic (HG) translesional peripheral gradients, with the functional and anatomic parameters before and after an infrainguinal endovascular procedure. Background. RGs and HGs are objective tools in defining the hemodynamic significance of an arterial stenosis. Methods. In 25 subjects with infrainguinal arterial stenosis, RG and HG were measured via a pressure wire before and after angioplasty. Before and after the procedure, all subjects had an ankle-brachial index (ABI) and Duplex ultrasound evaluation, recording prelesion and in-lesion peak systolic velocity (PSV-L), and calculating a peak systolic velocity ratio (PSV-R). A Pearson R correlation coefficient was calculated. Results. The mean age was 73 ± 12 years, 70% were men, median Rutherford class 3. At baseline and after angioplasty, mean ABI was 0.78 ± 0.2 and 0.99 ± 0.1, mean PSV-L was 459 ± 110 cm/s and 126 ± 35 cm/s, and mean PSV-R was 6.7 ± 4 and 1.2 ± 0.5, respectively. RG and HG significantly improved (P<.001) from baseline to after angioplasty (28.7 ± 20.5 mm Hg to 5 ± 13 mm Hg and 40.2 ± 21.4 mm Hg to 10 ± 13 mm Hg, respectively). RG before and after the procedure correlated well with ABI (r = -0.58; r= -0.41), PSV-L (r= 0.40; r = 0.52), and PSV-R (r = 0.46; r = 0.42). An improvement of 9 mm Hg in RG predicted a change of 0.1 in ABI. Conclusions. Improvement in RG during endovascular intervention in superficial femoral artery correlates well with the improvement in ABI, PSV-L, and PSV-R. A postprocedural decrease in RG of 9 mm HG predicts an improvement in ABI of 0.1.
J INVASIVE CARDIOL 2015;27(10):483-488
Key words: peripheral vascular disease, hemodynamics, physiology
The endovascular treatment of femoropopliteal disease is regarded as very effective, but currently we lack objective, standardized methods to document and report intraoperative functional outcomes once the procedure is completed. Angiography is currently being utilized as the endpoint for procedure success, but is inadequate for detection of residual, hemodynamically significant stenosis after percutaneous transluminal angioplasty (PTA). Feasibility of translesional gradient (TLG) measurement has been established for superficial femoral artery (SFA) and iliac lesions, showing good correlation to preprocedural subjective scores (walking impairment questionnaire scores) and objective measures (rest and exercise ankle-brachial index [ABI]) of lower-extremity claudication,1 and provides better assessment of the effect of treatment.2-4
The main purpose of this study was to assess the relationship between the resting (RG) and hyperemic (HG) translesional peripheral gradients, with the functional and anatomic parameters before and after an endovascular procedure.
Patient population. Subjects with claudication (Rutherford class 2-3) and documented peripheral vascular disease (ABI <0.9 at rest, and in-lesion peak systolic velocity >250 cm/s) who were scheduled for a potential endovascular intervention were included in the study after signing a consent form. The study protocol was approved by the Western institutional review board. Subjects were enrolled if angiography confirmed a hemodynamically significant stenosis (>50%) in the femoral or popliteal artery and patent inflow arteries.
Non-invasive assessment. At baseline, all subjects had the following: demographics, Rutherford class assessment, resting ABI, and standardized Duplex ultrasound. The Duplex ultrasound was performed by a single operator (JS) within 2 weeks prior to the procedure and included in-lesion peak systolic velocity (PSV-L), and pre-lesion PSV-P velocity; a ratio (PSV-R) of in-lesion PSV and pre-lesion PSV was calculated offline. The Bernoulli-predicted peak resting gradient was derived using the velocity data in the modified Bernoulli equation: the pressure gradient ΔP = 4(PSV-L2 – PSV-P2).5 All exams were performed on a Phillips Ie33 ultrasound system utilizing a L9-3 MHz transducer. All subjects underwent the same evaluation within 1 week of procedure completion.
Invasive assessment. Standard lower-extremity angiography of the entire vascular tree was performed. The Combo Wire (Volcano Corporation) was normalized at the level of common femoral artery, and then placed in the distal popliteal artery, just proximal to the take-off of the anterior tibial artery. The RG was recorded. A blood pressure cuff was positioned at the calf level and inflated for 1 minute at 10 mm Hg above the systolic blood pressure. After rapid cuff deflation, the pressure measurements were recorded. Once the procedure was completed, the pressure wire was repositioned in the same segment of the popliteal artery, and all measurements were repeated. The decision to perform or defer endovascular treatment of the target SFA lesion was based on the presence of an angiographically severe stenosis (50%) or a resting gradient >10 mm Hg. The procedure was ended if there was angiographic residual stenosis <30% or residual RG <10 mm Hg in the absence of a flow-limiting dissection.
PC 2.1 ComboMap software (Volcano Corporation) is an offline tool developed to take an existing case file, separate the baseline and hyperemic pressure runs, calculate the fractional reserve from the baseline runs, and prepare a report summarizing the values. Case files were collected and archived on the ComboMap console (Volcano Corporation).
Statistical analysis. The strength of association between the RG and hyperemic gradient with the resting ABI, in-lesion peak systolic velocity (PSV-L), and pre-lesion/in-lesion PSV ratio (PSV-R), and Bernoulli-derived ΔP before and after intervention was assessed with the Pearson coefficient correlation.
Twenty-five subjects completed the study. The mean age was 73 ± 12 years, 70% were men, median Rutherford class 3. All lesions were femoropopliteal. The mean lesion stenosis was 86.6 ± 10%. Two lesions (8%) were short chronic total occlusions (52 mm and 80 mm in length). The mean lesion length was 142 mm (range, 40-255 mm). Twelve lesions (48%) were <100 mm in length. Five subjects had 2 tandem lesions. Severe calcification was noted in 4 cases (16%). One patient had single-vessel run-off, 13 patients had double-vessel run-off, and 11 patients had triple-vessel run-off. The median Rutherford class after the procedure was 1. The mean residual stenosis post procedure was 17.6 ± 10%. At baseline and after the angioplasty, mean ABI was 0.78 ± 0.2 and 0.99 ± 0.1, mean PSV-L was 459 ± 110 cm/s and 126 ± 35 cm/s, and mean PSV-R was 6.7 ± 4 and 1.2 ± 0.5, respectively. RG and HG significantly improved (P<.001) from baseline to after angioplasty (28.7 ± 20.5 mm Hg to 5 ± 13 mm Hg and 40.2 ± 21.4 mm Hg to 10 ± 13 mm Hg, respectively). The changes in ABI, PSV, RG, and HG before and after the procedure are shown in Figure 1. The correlations between the TLGs (RG and HG) and PSV-L, PSV-R, Bernoulli-derived ΔP, and ABIs, as well as the changes in TLGs vs the changes in PSV-L, PSV-R, and ABI before and after angioplasty are shown in Table 1.
This study demonstrates that both RGs and HGs obtained via a pressure wire before and after an endovascular procedure correlate very well with the ABIs performed prior and after the procedure. These findings provide operators with a quantifiable tool to assess the functional improvement in the degree of claudication while the patient is still in the angiography suite.
Resting and hyperemic systolic gradients. Fluid mechanical analyses have provided insight into flow alterations and pressure drops associated with idealized flow patterns and geometric settings.6 Physiologically, the most important variables affecting pressure drops are blood-flow velocity and the degree of cross-sectional narrowing. The influence of a stenosis becomes more significant as flow increases. Therefore, a particular critical stenosis value depends on velocity or flow. As opposed to the coronary flow, the highest flow in the peripheral arteries occurs in the systolic phase of the cardiac cycle; for that reason, the systolic pressure represents the most sensitive measure of hemodynamic significance. Furthermore, exercise, by means of increasing flow rates, unmasks subcritical stenosis. Studies have shown that up to 80% of segments previously classified as normal on the basis of resting pressure measurements were reclassified as abnormal with use of induced reactive hyperemia.3 Early surgical studies have shown that reactive hyperemia during lower-extremity cuff inflation increases the flow in femoral arteries by an order of 1.6-7.0 times above baseline.7 Although pharmacological hyperemia is more objective, our choice of reactive hyperemia using cuff inflation was related to its ease of use and avoidance of interaction with vasodilator agents potentially delivered during the endovascular intervention. Indeed, in our series of claudicants, reactive hyperemia increased the systolic gradient on average by an order of 1.4 (1.1-9.0 times baseline). Lower RGs led in general to larger increases in the HG in our population of symptomatic patients (Figure 2). This phenomenon is potentially explained by the fact that subjects with lower gradients have more preserved endothelial function, which allows more vasodilation after cuff deflation. By the same token, the presence of severely impaired, calcified tibial vessels might limit the benefit of mechanical hyperemia.
The utility of TLGs has been documented in the early vascular surgery studies, especially after iliac artery interventions. More recently, Banerjee et al1 proved the utility of invasive TLG measurements using vasodilation for determining the functional and hemodynamic significance of SFA lesions. In that study of 19 patients, the authors found that the threshold of TLGs >11 mm Hg post 100 µg adenosine identified patients with symptomatic peripheral arterial disease and ABI ≤0.7, which was 100% specific and 71% sensitive.
Based on the data currently available, the use of RGs would suffice in the majority of cases.8 In symptomatic patients with RGs <10 mm Hg, either cuff hyperemia or pharmacologic vasodilation should be performed. This method will unmask the minority of symptomatic patients with normal resting TLGs. In those cases, improvement in HG should be documented at the end of the procedure.
Therapeutic endpoints and assessment of functional improvement. Functional impact of revascularization procedures may be best assessed by alleviation of systolic TLGs. Kaufman et al2 have shown that hemodynamic pressure gradients are more accurate than angiography for determining satisfactory endpoints for angioplasty. Our study also demonstrates the role of pressure gradient measurement as an endpoint of angioplasty. This is reflected in the excellent correlation seen between the change in the TLGs before and after intervention (ΔTG) and the change in ABI (ΔABI) (r = 0.527). Furthermore, this view is supported by the weak correlation between angiographic stenosis severity and the ABI, both at baseline and after intervention.
The lack of a perfect correlation between the ABI and TLGs is explained by the fact that ABI reflects the entire limb perfusion, while TLG is a reflection of the femoropopliteal segment. Stronger correlations would be expected if the pressure wire would be positioned at the ankle. Unfortunately, this approach is clinically impractical.
The primary endpoint of any revascularization procedure for claudicants is improved leg function. Traditionally, ABI has been used as the main objective method to document hemodynamic improvement and clinical benefit. Nevertheless, the results of ABI are only available at follow-up visits. The ability of an intraoperative tool to predict postprocedural ABI would be of most importance in guiding procedure termination. Our data suggest that an improvement in RG of 8.6 mm Hg would translate to an improvement in ABI of 0.1. This knowledge will allow clinicians to make informed decisions regarding procedural termination, particularly when tandem lesions are treated, or different anatomic segments are contemplated.
Angiography and TLGs. Given the complex features of each stenosis, conventional angiography is limited in its ability to give clinicians a complete understanding of a lesion’s anatomy. With proper angulation, angiography can be used to measure the length and minimal luminal diameter of a stenotic arterial segment. However, other features such as flow entrance, exit angle, orifice shape, and degree of turbulence are not well evaluated angiographically,9 but are incorporated within pressure-gradient measurements. Furthermore, compared with interobserver agreement on pre-PTA angiograms (r = 0.51-0.65), agreement on residual stenosis after angioplasty varies considerably (r = 0.36-0.45)3.
Brewster et al10 showed a poor correlation between angiographic and hemodynamic assessment of stenosis significance. In their study, femoral artery pressure gradients were most useful in patients with moderate stenosis on angiography.
Our study confirms the findings of the prior surgical experiments: the correlations between the resting and post-hyperemic gradients and angiographic assessment of luminal stenosis were only moderate (r = 0.38, P=.05 for RG; r = 0.26, P=.22 for HG). This is explained by the fact that a focal, severe stenosis has the same hemodynamic impact as a moderate but long, diffuse lesion. In this regard, the direct measurement of pressure gradient across a lesion, in vivo, is likely to be the most helpful guide.
In the case example illustrated in Figure 3, resting peak systolic gradient was 14 mm Hg and increased to 40 mm Hg post hyperemia. After orbital atherectomy and balloon angioplasty, there was angiographic improvement; however, the RG remained elevated at 12 mm Hg and postprocedure HG at 29 mm Hg. Based on gradient findings, the decision was for additional intervention with stent placement and the final RG was reduced to 0 mm Hg and HG to 13 mm Hg. This is one example of how both RG and HG could be used in the clinical practice. In fact, in 10 of our 25 cases (40%), the knowledge of the final RG influenced our therapeutic decision process. In 6 cases, further therapy was performed although the angiographic result seemed acceptable, while in 4 cases the procedure was ended based on the achievement of RG <10 mm Hg and an acceptable angiographic result. Of note, all these cases were atherectomy followed by angioplasty.
A benefit of angiography is the immediate real-time visualization of a lesion post intervention, and it is currently being utilized as an endpoint for procedure success both in clinical practice and clinical trials. However, this is an inadequate surrogate marker for functional improvement. Although much more investigation is needed, our study sets the basis for the use of the TLGs as an important complement to angiography in assessing anatomical and functional improvement after endovascular procedures.
Duplex ultrasound and TLGs. The lack of a strong correlation between PSV and TLG is likely explained by the presence of tandem lesions or moderately severe diffuse disease. Studies of tandem lesions have indicated that multiple subcritical stenoses can act in a critical fashion.11,12 PSV is a reliable indicator of hemodynamic significance of stenosis severity in highly occlusive, focal segments; the value of PSV diminishes in moderate stenosis, tandem lesions, and diffusely diseased vessels.
Our data show a marginal correlation of preprocedural and postprocedural PSV-L, or PSV-R with RGs, likely due to the inability of the ultrasound measurements to assess the severity of long diffuse lesions. A better correlation was seen between the RG and the Bernoulli-predicted pressure gradient. The Bernoulli-predicted peak resting gradient, which is derived using the velocity data in the modified Bernoulli equation: the pressure gradient ΔP = 4(PSV-L2 – PSV-P2). It takes into consideration both the velocity proximal to the lesion as well as the in-lesion velocity.5
Of interest, the PSV-L and PSV-R correlated better with the RG than with the ABI. Again, the ABI reflects the hemodynamic status of the entire limb, and therefore is not the ideal method to evaluate the short-term or long-term success of treatment of a specific arterial segment. In a study by Lotfi et al, although pressure gradients were not reported, the baseline PSV-L was significantly associated with preintervention peripheral fractional flow reserve (r = -0.77, P<.001).13 Interestingly, postintervention peripheral fractional flow reserve <0.95 predicted a more rapid increase in PSV-L over time, which is a reasonably accepted surrogate for restenosis.
This study addressed the feasibility of using translesional hemodynamic data as an adjunct to infrainguinal angiography in clinical decision making. The preprocedural and postprocedural RGs measured via the pressure wire showed a very good correlation with the resting ABI. Moreover, the improvement in RG after angioplasty strongly correlated with the improvement in PSV-L, PSV-R, and ABI. These findings indicate that employing physiologic measurements in conjunction with angiography is clinically useful not only in assessing the functional significance of peripheral lesions, but also in predicting the degree of clinical improvement.
- Banerjee S, Badhey N, Lichtenwalter C, Varghese C, Brilakis ES. Relationship of walking impairment and ankle-brachial index assessments with peripheral arterial translesional pressure gradients. J Invasive Cardiol. 2011;23:352-356.
- Kaufman SL, Fara JW, Udoff EJ, Harrington DP, White RI Jr. Hemodynamic effects of vasodilators across iliac stenoses in dogs. Invest Radiol. 1979;14:471-475.
- Tetteroo E, Haaring C, van der Graaf Y, van Schaik JP, van Engelen AD, Mali WP. Intraarterial pressure gradients after randomized angioplasty or stenting of iliac artery lesions. Dutch Iliac Stent Trial Study Group. Cardiovasc Interv Radiol. 1996;19:411-417.
- Kinney TB, Rose SC. Intraarterial pressure measurements during angiographic evaluation of peripheral vascular disease: techniques, interpretation, applications, and limitations. AJR Am J Roentgenol. 1996;166:277-284.
- De Smet AA, Tetteroo E, Moll FL. Non-invasive evaluation before and after percutaneous therapy of iliac artery stenoses: the value of the Bernoulli-predicted pressure gradient. J Vasc Surg. 2000;32:153-159.
- Yongchareon W, Young DF. Initiation of turbulence in models of arterial stenoses. J Biomech. 1979;12:185-196.
- Brener BJ, Raines JK, Darling RC, Austen WG. Measurement of systolic femoral arterial pressure during reactive hyperemia. An estimate of aortoiliac disease. Circulation. 1974;50:II259-II267.
- Archie JP Jr, Feldtman RW. Intraoperative assessment of the hemodynamic significance of iliac and profunda femoris artery stenosis. Surgery. 1981;90:876-880.
- Hannawi B, Lam WW, Younis GA. Pressure wire used to measure gradient in chronic mesenteric ischemia. Tex Heart Inst J. 2012;39:739-743.
- Brewster DC. Clinical and anatomical considerations for surgery in aortoiliac disease and results of surgical treatment. Circulation. 1991;83:I42-I52.
- Flanigan DP, Tullis JP, Streeter VL, Whitehouse WM Jr, Fry WJ, Stanley JC. Multiple subcritical arterial stenoses: effect on poststenotic pressure and flow. Ann Surg. 1977;186:663-668.
- Karayannacos PE, Talukder N, Nerem RM, Roshon S, Vasko JS. The role of multiple noncritical arterial stenoses in the pathogenesis of ischemia. J Thorac Cardiovasc Surg. 1977;73:458-469.
- Lotfi AS, Sivalingam SK, Giugliano GR, Ashraf J, Visintainer P. Use of fraction flow reserve to predict changes over time in management of superficial femoral artery. J Interv Cardiol. 2012;25:71-77.
From New York University Medical Center, Cardiovascular Medicine
New York, and Cardiovascular Associates, Medicine, New York, New York.
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 January 20, 2015, provisional acceptance given February 12, 2015, final version accepted April 20, 2015.
Address for correspondence: Cezar S. Staniloae, MD, New York University Medical Center, Cardiovascular Medicine
New York, Cardiovascular Associates, Medicine, 275 7th Ave 3rd Floor,
New York, NY 10001. Email: Cezar.firstname.lastname@example.org