First experiences of local pulse wave velocity measurements in 4D-MRI in focally stented femoropopliteal arteries
Abstract
Summary:Background: In peripheral arterial disease (PAD) the femoropopliteal (FP) artery is the most frequently recanalized lower limb artery. Stent-based interventions change the biomechanical properties of FP arteries. However, no clinical tool for functional imaging is established for quantitative measurements in vivo. Four-dimensional-flow magnetic resonance imaging enables a detailed evaluation of the hemodynamics of the central and – more challenging – the peripheral arteries. The present study aimed to determine the feasibility of assessing pulse wave velocities (PWV) as a marker of vessel stiffness in PAD patients with multiple spot stents and to compare the values with age-matched subjects and young-adult healthy subjects. Patients and methods: Contrast-free 4D-flow MRI was performed in seven PAD patients with focally stented FP arteries, five age-matched subjects after exclusion of PAD, and five young, healthy adults. PWV values were calculated from flow curves by using the foot-to-foot method. Results: Four-D-flow MRI sequences offering high spatial and temporal resolution enables quantification of flow velocity measurements and estimation of PWVs. Assessment of segmental PWV as a surrogate of vascular stiffness in focally stented femoral arteries is feasible. PWV values across all groups were 15.6±5.2 m/s, 13.3±4.1 m/s, and 9.9±2.2 m/s in PAD patients, senior-aged volunteers, and young-adult volunteers respectively. PWV values in PAD patients were similar with those in the senior-aged volunteers group (15.6±5.2 vs. 13.3 ±4.1 years, p=0.43). However, when compared to the young-adult volunteers, PAD patients had a statistically significantly higher mean local PWV (15.6±5.2 m/s vs. 9.9±2.2 m/s, p<0.05). Conclusions: Calculating segmental PWV in the femoral arteries is feasible in PAD patients with focally stented FP arteries. PWV values in PAD patients were similar to those in senior-aged volunteers, both of which were higher than in young-adult volunteers.
Introduction
Peripheral arterial disease (PAD) affects around 237 million people worldwide [1]. A stenosis or occlusion of the femoropopliteal (FP) artery is the most common lesion in intermittent claudication [2] and the FP artery is by far the most common recanalized lower limb artery [3]. According to the current guidelines, percutaneous intervention is the preferred first-line treatment modality for patients with symptomatic FP lesions [4, 5, 6, 7].
A special feature of the FP artery is the high deformation stress. When the limbs are moved, there is significant bending, shortening and twisting [8, 9]. The biomechanical forces on the FP artery are made largely responsible not only for the occurrence of degenerative vascular changes, but also for the failure of recanalizing therapies with and without stent deployment in middle- and long-term follow-up, in particular in long complex lesions [9, 10, 11, 12]. The incidence of restenoses after an initially successful vascular intervention is between 5% and 70% within one year, depending on endogenous and procedural factors [13]. In stent-based FP interventions, stents are deployed planned (“primary stenting”) or as a “bailout”, when elastic recoil and intimal dissection after balloon-dilation is managed with mechanical scaffolding in order to stabilize the vessel lumen. However, the mechanical properties of the metal scaffolds used are not identical to the biomechanical properties of the FP artery. In a perfused human cadaver model, however, none of seven different stents was able to match all FP deformations without changing the deformations of the vessel inside or outside the stent [14]. The modern interventional strategy of implanting foreign bodies “as less as reasonably achievable” [15] aims to minimize traumatic interactions between the implant and the vascular wall, to avoid stent fractures and instent-restenosis and to enable bypass anastomoses on stent-free vessel segments, if necessary. The concept of “spot stenting” (SS) has demonstrated good clinical results in conceptual research [16], dedicated registries [17, 18] and a randomized-controlled trial [19] and favourable patency rates compared to the conventional “full-metal jacket” strategy [16, 19]. Although the results seem to be promising, the hemodynamic and biomechanical consequences of multiple short stents are unclear [20]. A better understanding of the dynamic environment of the FP artery is crucial to optimise treatment strategies in the future. SS opens up diagnostic options. While long-segment metallic stents can severely disturb diagnostic vascular imaging [21], SS significantly reduces metal burden leaving 38 – 60% of the total lesion length “stent-free” [16, 17, 19, 22]. In this study, we aimed to assess the feasibility of 4D-flow MRI of assessing the local PWV of the stented FP arteries.
Patients and methods
Study design and setting
This study was a prospective, single-center pilot study, conducted at the University Medical Centre Mannheim. Contrast-free 4D-flow MRI measurements were performed in PAD patients with focally stented FP arteries, age-matched subjects after exclusion of PAD and without known cardiovascular morbidities and young healthy adults.
PAD patient cohort
The patient group consisted of seven consecutive PAD patients after SS using the VascuFlex® Multi-LOC stent system (B.Braun Melsungen, Germany). The interventional procedures had been performed at the Diakonissenkrankenhaus Mannheim in Germany, a teaching hospital of the University of Heidelberg. All patients of the study cohort had initially been entered in the LOCOMOTIVE EXTENDED registry (Clinical-Trials.gov Identifier: NCT02900274) the data from which have been published previously [17]. Briefly, the VascuFlex® Multi-LOC system is a multiple stent system with six self-expanding Nitinol stents with a length of 13 mm each. Stents had been deployed as a bailout in case of insufficient acute results (flow limiting dissections or persisting stenosis>30%) following antegrade plain old angioplasty and/or drug coated balloon dilatation of symptomatic de novo FP lesions. Balloon and stent diameters were based on the reference lumen diameters, avoiding oversizing.
PAD eligibility criteria
In order to minimize 4D-flow MRI measurement errors associated with modifications in the featuring shape of pulse waveforms acquired from different arterial sites [23], the following eligibility criteria were defined: 1) Duplex sonographic exclusion of relevant steno-occlusive lesions in the (aorto-iliac) inflow and (popliteo-crural) outflow of the superficial femoral artery. 2) Biphasic flow profile waveform with fast systolic upstroke (acceleration time<100 ms; similar acceleration time in the proximal and distal superficial femoral artery) along the whole superficial femoral artery. 3) Crural two- or three-vessel run-off without high-grade stenosis (peak-velocity ratio<2.0) in pulse wave doppler sonography, to minimize the effect of severe arterial wave reflections in distal blood pulse waves. Furthermore, PAD patients had to have a normal ankle-brachial index (ABI>0.9) after the intervention. In a previous population-based study, a low ABI was associated with low values of lower-limb PWV [24] and, possibly due to “falsely” low PWV values, attributable to the stenosis reducing blood flow and distending pressure downstream [25]. The control group volunteers were considered healthy if they had no history of cardiovascular disease and had not taken any cardiovascular drugs. Prior to MRI, PAD was excluded by means of pulse palpation and the ABI was between 0.9 and 1.3. The local ethics committee approved the protocol and all subjects gave written informed consent.
MR sequence parameters
All measurements were performed on a 3T MRI scanner (Magnetom Skyra; Siemens Heathineers, Erlangen, Germany) with an 18-channel receiver body coil array wound around the target limb. A 3D non-contrast magnetic resonance angiography (NATIVE, TOF) sequence was acquired to localize the superficial femoral artery. Four-dimensional flow phase contrast was acquired without contrast agent using the following scanning parameters: FOV=256×256 mm2, matrix size (in-plane base resolution)=192×192 (1.3×1.3 mm2), 12–18 slices, slice thickness (1.5 mm), TR=20–40 ms, TE=3 ms, bandwidth=490 Hz/px, and flip angle=7°. The velocity encoding (VENC) was set to 100 cm/s in the vessels of interest, according to the maximum flow velocity of the superficial femoral artery measured by previous ultrasound examinations.
The R-wave of the electrocardiogram was used to retrospectively trigger the 4D flow measurement with 14–22 timeframes acquired in the R–R interval dependent on the heart rate. The total time of the MRI examination including patient preparation took approximately 20 min. The total scan time of the 4D flow took approximately 7.8±2.1 min depending on the subject’s heart rate. The approximate post processing time was 5 minutes.
Image processing
The data were analyzed using the CVI42 platform and MATLAB softwares (The MathWorks; Natick, MA, USA) to generate hemodynamic parameters. First, the artery was semi-automatic segmented using maximal intensity projections on the phase images for every slice. The maximal velocity of every slice in the superficial femoral artery is calculated and within a 9×9 neighborhood, all pixels with a velocity higher than 50% of the local maximal velocity were taken for artery segmentation. The segmentation is represented as a vectorized line, which is then vectorially multiplied with the velocity field, which leads to the velocity over the distance along the artery.
The PWV is defined as the distance between two positions in the artery over their arrival time of one bolus. The PWV was estimated as the average along the entire segment of the superfical femoral artery assessed with the MRI of approximately 20 cm as shown in Figure 1. The arrival time was estimated by the foot-to-foot method. The foot-to-foot method is fitting the upslope of the wave profile by the tangential intersection to estimate the “foot” of the wave. The relation between arrival time and position in the artery results in the global (whole artery) and the local (step-wise division) PWV derived by linear fitting.
Ethics
This study was approved by the local institutional review board (identifier: 2019-740 N), and written, informed consent was obtained prior to scanning from all subjects. The study was conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonisation guideline. Data protection was in accordance with the EU Data Protection Directive.
Statistical analysis
Demographic and clinical characteristics of the study population are reported as means and standard deviations (SD) or as medians and (interquartile) ranges for continuous variables, according to their distribution. Categorical data are given as counts and percentages. The Mann-Whitney U-test was applied to determine significant differences between non-parametric continuous variables (PWV). p-values <0.05 were considered statistically significant. All statistical analyses were performed using SPSS version 23 (IBM, Munich, Germany).
Results
Baseline characteristics of the study population
Seven PAD patients (four male, three female) were enrolled in this study (mean age 72±10). Baseline characteristics of the study participants are presented in Table I. At baseline, all patients had intermittent claudication in Fontaine stage IIb. The most frequent comorbid conditions and cardiovascular risk factors were: dyslipidemia (100%), hypertension (86%), and diabetes (57%); four patients had a history of nicotine consumption (57%), and one patient was active smoker. All patients received statins, and antiplatelet drugs, and one patient in the PAD group received oral anticoagulant therapy. According to lesion characteristics, the mean lesion length was 18.0±6.8 cm, and two patients had chronic total occlusions. The mean number of “spot” stents implanted was 5.0±0.8 per lesion.
Comparison of the femoral PWV among groups
PWV was calculated using the time-to-foot method, examples are shown in Figure 2. PWV values across all groups were 15.6±5.2 m/s, 13.3±4.1 m/s, and 9.9±2.2 m/s in PAD patients, senior-aged volunteers, and young-adult volunteers respectively (Figure 3).
PWV values in PAD patients were similar to those in the senior-aged volunteer group (15.6±5.2 vs. 13.3 ±4.1, p=0.43). When compared to the young-adult volunteers, the senior-aged volunteers had a numerically higher mean local PWV (13.3± 4.1 vs. 9.9 ± 2.2, p=0.13). Despite the relatively small sample size, the differences between the young-adult volunteers and the PAD patients were statistically significant (p<0.05).
Discussion
4D-flow MRI is an emerging tool for the evaluation of vascular hemodynamics. It measures the flow velocity in three directions for every voxel directly in vivo and, together with the distance measurement along the femoral artery, enables the calculation of the local PWV as an expression of the stiffness of an arterial segment.
Our study showed that 4D-flow MRI for estimation of segmental PWV in focally stented femoral arteries is feasible and that PAD patients with focally stented FP arteries but also senior-aged volunteers have higher PWV than young adult volunteers.
Feasibility
Generally, a small vessel diameter and high blood flow velocities are major challenges for quantification of flow velocity measurements and estimation of PWVs by means of functional MRI. This can be overcome by MRI sequences offering high spatial and temporal resolution, facilitating local PWV measurements [26]. Further, metallic vessel scaffolding makes it difficult to adequately assess the vascular lumen for blood flow analysis [21]. This additionally complicates the biomechanical assessment of the FP artery after stent-based interventions, especially after full-metal jacket stenting. The SS technique was developed with the aim of influencing the FP biomechanical properties as little as possible. To record the latter in vivo is challenging. Today, a substantial part of the findings on how stents change the biomechanics inside or outside the stent segments is based on perfused human cadaver models [14]. The SS strategy offers MRI “windows” in the inter-stent segments. In the present study, the feasibility of estimation of hemodynamic parameters and local PWV as a surrogate of vascular stiffness in FP arteries after SS was shown for the first time. Employing 4D flow MRI systems offer several advantages over other techniques for local PWV assessment: It is non-invasive, free from ionizing radiation and provides a direct, 3-dimensional measurement of the path length of pulse wave trajectory, even from anatomically deep and tortuous arteries.
PWV values
Generally, the values of local PWVs for peripheral arteries (e.g. the femoral artery) are higher compared to the central arteries (e.g. the aorta and carotid artery). Causes for this encompass the muscular nature of peripheral artery walls together with their smaller cross-sectional area and the influence of wave reflections [23]. While PWV estimates from central arteries have emerged as a powerful independent predictor of cardiovascular outcomes [27, 28], local PWV estimates from peripheral arteries in PAD patients might be an early step towards a deeper understanding of the diseased vessels’ characteristics and the impact of stent deployment. Further studies to systematically compare local PWV measures in patients after FP stenting and in patients after non-stent based treatment strategies are planned. In principle, MRI provides the possibility to visualise and measure even more biomarkers such as peak systolic velocity and wall shear stress [29].
Until today, no methodological standardization has been established globally for determining the gold-standard of local vessel stiffness measurements of individual arteries. Extensive comparative studies on the accuracy of the available methods are lacking. Generally, invasive techniques for local PWV measurement – capturing hemodynamic signals from the target artery – are considered to provide the gold-standard estimate. Similar to non-invasive methods of arterial pulse waveform assessment for PWV measurement using applanation tonometry, cuff-based or echo-tracking techniques and current 4D-flow protocols, there is a lack of standardization of local PWV measurement and reference values for lower-extremity arteries [26].
In our study on superficial femoral arteries, the mean PWV in the young-adult group (9.9 m/s) aged between 19 and 28 years was in close agreement with those reported by Bustin et al. (mean 9.8 m/s) for the<29 years group [30]. The mean segmental PWV values of the PAD patients (15.6 m/s) were approximately 50% higher. According to the Bramwell–Hill derived equation (Compliance=k/PWV2; [31]) an increase of PWV in that range means more than doubling arterial stiffness. Previously reported data regarding the aging of the lower-extremity arteries are discrepant. Overall the stiffness of lower-extremity arteries seems to be affected to a lesser extent by age compared to central arteries [32]. In an epidemiological study in more than 5,000 healthy individuals, femoral stiffness was shown to remain constant over many years, but to increase significantly between the sixth and eighth decade, which matches our results [33]. The mean segmental PWV values of the senior-aged control group (13.3 m/s) were not far from that of the PAD patients (15.6 m/s).
Limitations
Aside from the small number of subjects studied, our study has several limitations concerning this issue. First, the study groups were small in number, not extensively pre-examined cardiovascular or heterogeneous. Our control subjects were neither screened for cardiovascular risk factors nor was vascular calcification assessed. The latter is known to be associated with age and can begin at a young age and has to be distinguished from atherosclerotic occlusive disease [34]. However, given the lack of consensus on the definition of calcification and classification of its burden, especially in the healthy, we dispensed with further examinations in this feasibility study. All PAD patients were treated for major cardiovascular risk factors, including statins, which are known to reduce lower-limb PWV [35, 36]. First, reliable local PWV measurement techniques are needed to achieve accurate performance under in-vivo conditions. Second, larger studies are required to make deductions concerning local PWV associated with PAD. Finally, future research is warranted to standardize and validate the method and evaluate the clinical utility of local PWV.
Conclusions
Contrast-free 4D-flow MRI provides a high spatiotemporal resolution that allows the assessment of hemodynamics in focally stented as well as in native FP arteries. Calculating segmental PWVs in the FP artery segment is feasible in PAD patients with multiple spot stents. The PWV values in PAD patients were similar to those in senior-aged volunteers, both of which were higher than in young-adult volunteers.
The authors would like to thank the patients and the volunteers for their contribution to the study. We would like to thank Dr. Hannelore Müller-Mürtz, Dr. Christel Krahl, Marianne Hoffmann, Svetlana Pausch and Stefanie Schönfelder for their time and support regarding the care of patients.
References
1 Global, regional, and national prevalence and risk factors for peripheral artery disease in 2015: an updated systematic review and analysis. Lancet Glob Health. 2019;7(8):e1020–30.
2 . Clinical practice. Intermittent claudication. N Engl J Med. 2007;356(12):1241–50.
3 Recent Trends in Clinical Setting and Provider Specialty for Endovascular Peripheral Artery Disease Interventions for the Medicare Population. J Vasc Interv Radiol. 2020;31(4):614–21.e2.
4 2017 ESC Guidelines on the Diagnosis and Treatment of Peripheral Arterial Diseases. Eur Heart J. 2018;39(9):763–816.
5 Society for Vascular Surgery practice guidelines for atherosclerotic occlusive disease of the lower extremities: management of asymptomatic disease and claudication. J Vasc Surg. 2015;61(3 Suppl):2s–41s.
6 2016 AHA/ACC guideline on the management of patients with lower extremity peripheral artery disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017;69(11):e71–126.
7 An Update on Methods for Revascularization and Expansion of the TASC Lesion Classification to Include Below-the-Knee Arteries: A Supplement to the Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II). Ann Vasc Dis. 2015;8(4):343–57.
8 Three-dimensional bending, torsion and axial compression of the femoropopliteal artery during limb flexion. J Biomech. 2014;47(10):2249–56.
9 . Limb flexion-induced axial compression and bending in human femoropopliteal artery segments. J Vasc Surg. 2018;67(2):607–13.
10 Finite element analysis for fatigue behaviour of a self-expanding Nitinol peripheral stent under physiological biomechanical conditions. Comput Biol Med. 2019;104:205–14.
11 . Unique demands of the femoral anatomy and pathology and the need for unique interventions. J Cardiovasc Surg. 2013;54(2):191–210.
12 Full Drug-Eluting Stent Jacket: Two-Year Results of a Single-Center Experience With Zilver PTX Stenting for Long Lesions in the Femoropopliteal Arteries. J Endovasc Ther. 2018;25(3):295–301.
13 . Restenosis after endovascular revascularization in peripheral artery disease. VASA. 2015;44(4):257–70.
14 Stent Design Affects Femoropopliteal Artery Deformation. Ann Surg. 2019;270(1):180–7.
15 . Less is more: the “As Less As Reasonably Achievable Stenting” (ALARAS) strategy in the femoropopliteal area. J Cardiovasc Surg. 2018;59(4):495–503.
16 Outcomes of spot stenting versus long stenting after intentional subintimal approach for long chronic total occlusions of the femoropopliteal artery. JACC Cardiovasc Interv. 2015;8(3):472–80.
17 Provisional focal stenting of complex femoropopliteal lesions using the Multi-LOC multiple stent delivery system - 12-month results from the LOCOMOTIVE EXTENDED study. Vasa. 2021;50(3):209–16.
18 Focal Stenting of Complex Femoropopliteal Lesions with the Multi-LOC Multiple Stent Delivery System: 12-Month Results of the Multicenter LOCOMOTIVE Study. Cardiovasc Intervent Radiol. 2019;42(2):169–75.
19 Comparison of Spot versus Long Stenting for Femoropopliteal Artery Disease. Ann Vasc Surg. 2019;58:101–7.
20 . 12-Month Results from the Multicenter LOCOMOTIVE Study: Should We All Jump on this Bandwagon? Cardiovasc Intervent Radiol. 2019;42(2):176–7.
21 MR imaging in the presence of vascular stents: A systematic assessment of artifacts for various stent orientations, sequence types, and field strengths. J Magn Reson Imaging. 2000;12(4):606–15.
22 Spot stenting versus full coverage stenting after endovascular therapy for femoropopliteal artery lesions. J Vasc Surg. 2019;70(4):1166–76.
23 . Mcdonald’s blood flow in arteries: theoretical, experimental and clinical principles. (6th ed). London: Hodder Arnold; 2011.
24 Lower-extremity arterial stiffness vs. aortic stiffness in the general population. Hypertens Res. 2013;36(8):718–24.
25 . Lower-limb pulse wave velocity: correlations and clinical value. Hypertens Res. 2013;36(8):679–81.
26 . Local Pulse Wave Velocity: Theory, Methods, Advancements, and Clinical Applications. IEEE Rev Biomed Eng. 2020;13:74–112.
27 Prognostic value of aortic pulse wave velocity as index of arterial stiffness in the general population. Circulation. 2006;113(5):664–70.
28 Aortic pulse wave velocity improves cardiovascular event prediction: an individual participant meta-analysis of prospective observational data from 17,635 subjects. J Am Coll Cardiol. 2014;63(7):636–46.
29 State-of-the-art aortic imaging: part I - fundamentals and perspectives of CT and MRI. Vasa. 2013;42(6):395–412.
30 . Noninvasive assessment of mechanical properties of peripheral arteries. Ann Biomed Eng. 1990;18(5):549–66.
31 . The velocity of pulse wave in man. Proc. R. Soc. Lond. B. 1922;93:298–306. Available from https://royalsocietypublishing.org/doi/10.1098/rspb.1922.0022
32 . Comparing the Heart-Thigh and Thigh-Ankle Arteries with the Heart-Ankle Arterial Segment for Arterial Stiffness Measurements. Vasc Health Risk Manag. 2020;16:561–70.
33 Reference values for local arterial stiffness. Part B: femoral artery. J Hypertens. 2015;33(10):1997–2009.
34 . Prevalence of Calcification in Human Femoropopliteal Arteries and its Association with Demographics, Risk Factors, and Arterial Stiffness. Arterioscler Thromb Vasc Biol. 2018;38(4):e48–57.
35 . Simvastatin improves arterial compliance in the lower limb but not in the aorta. Atherosclerosis. 2001;155(1):245–50.
36 Effect of atorvastatin on regional arterial stiffness in patients with type 2 diabetes mellitus. J Atheroscler Thromb. 2005;12(4):205–10.
