For the population in this study, the observation that shortening of the SFA due to leg flexion was greater in the bottom third as compared to the top two-thirds (P < 0.005) is likely due to the fact that knee flexion was significantly greater than hip flexion (P < 0.001). In addition, the greater variability of arc length shortening in the top third of the SFA is likely due to the lesser musculoskeletal constraints around the hip as compared to the knee. The non-significant trend for greater twisting in the inferior sections of the SFA could be due to greater flexion in the knee as already mentioned, as well as the more complex musculoskeletal influences on the geniculate branches as described in Cheng et al. (25
The spatially-resolved curvature metrics for the SFA reveal markedly non-uniform deformations. In the straight leg position, the SFA appears to be relatively straight for the entire length, however, with hip and knee flexion, the curvature values become disparate (bottom > top > middle), and the variability of the bottom third curvature is greater than the top and middle thirds. This can be explained by the fact that the adductor canal provides more muscular and membranous constraint proximal to the adductor hiatus (22
). In addition, the vastoadductor membrane creates an increasingly constrained space progressing distally along the adductor canal (34
). Therefore, increase in vessel curvature is greatest in the bottom third, adjacent to the greatest joint flexion (at the knee) and where there is the least constraint, and least in the middle third, not close to any joint flexion and where the adductor canal is highly constrained.
While the superficial femoral arteries remain visually smooth and relatively straight in younger subjects with hip and knee flexion, older subjects exhibit substantial curvature and buckling with flexion (). This may indicate that younger subjects retain tension in the SFA with hip and knee flexion, while older subjects do not. This is consistent with documented variation in arterial tension with age. While body growth persistently stretches arteries through adolescence, once maximum body height is reached, the chronic tension that arteries experience should decrease in the absence of further stretching, and with the degradation of elastin in arteries with age (7
), older adults have longer and less elastic arteries than young adults.
Figure 6 Maximum intensity projections of magnetic resonance angiograms of a young adult (top row) and older adult (bottom row) in the supine position (left column) and in the leg-flexed position (right column). Note that while the superficial femoral arteries (more ...)
Comparing the SFA deformations presented in this study with those of a younger population (age: 27 ± 5 years) reported in Cheng et al. (25
), some notable differences in biomechanical behavior were found. Note that there was significantly less flexion in the older population as compared to the younger population at the hip (older: 39 ± 6 °, younger: 120 ± 9°, P < 0.001) and knee (older: 86 ± 6°, younger: 134 ± 3°, P < 0.001), and that the data for the younger subjects were averaged along the entire length of the SFA rather than resolved into thirds. The variation in length change of the young population (−13 ± 11 %) was greater than that of all three sections of the older SFAs (P < 0.001). Also, there was greater variability in SFA twist in the young as compared to the older adults (young: 2.8 ± 1.7 °/cm, older: 0.7 ± 0.5 °/cm, P < 0.001). Furthermore, the older subjects exhibited greater maximum change in curvature than the young subjects in the bottom third (young: 0.04 ± 0.16 cm−1
, older: 0.41 ± 0.22 cm−1
, P < 0.001).
Along with visual angiographic evidence in , the above comparisons support the postulate that aging decreases stretch and elasticity in the SFA. For example, the far lower variability in arterial shortening due to leg flexion for the older adults compared to the young adults indicates that the percentage shortening of the older SFAs is more narrowly distributed in the population. This narrow distribution may indicate that the population of older subjects all shortened their SFAs to their minimum slack length, corresponding to zero tension and strain. This hypothesis is strengthened by the fact that the older subjects had significantly lower hip and knee flexion angles as compared to the young subjects, and the minimum SFA slack lengths were reached nonetheless. Similarly, the variability in overall SFA degrees of twist per cm was significantly greater in the younger as compared to the older population. A more narrow distribution of arterial twisting angles in the older population also supports less compliant arteries. The most definitive evidence, however, is that the bottom third of the older SFAs increase their maximum curvature more dramatically than the bottom third of the younger SFAs, which indicates that the older SFAs shortened past their point of slack and then buckled off-axis, while the younger SFAs did not.
As mentioned in the Introduction, it may be beneficial to characterize the axial strain of a target vessel. While younger SFAs do not experience significant increases in curvature when flexing from supine to fetal position (25
), the older SFAs curved significantly for the top, middle, and bottom thirds. Assuming that large increases in curvature indicated off-axis buckling in the older subjects (), it can be deduced that the arteries were straight and in tension while in the supine position, and past the point of slack when the legs were flexed. Therefore, the percent shortening of the SFA from supine to flexed position is the approximate axial tensile strain that the SFA experiences in a straight leg. Analogously, in the younger subjects, since the arteries remain straight even with full fetal position flexion and the point of slack is never exceeded, we can logically deduce that the chronic axial tensile strain of the SFA in a straight leg is greater than or equal to the percent shortening quantified.
The evidence presented here shows that with hip and knee flexions approximately commensurate with walking, the distal and proximal portions of the SFA are subject to greater longitudinal and bending deformations than the middle of the SFA, with the distal SFA deforming the most. In addition, we hypothesize that there is decrease in baseline strain in femoral arteries with aging. Combined with the natural decrease in compliance of vessels with age, the relative fixed musculoskeletal geometry of the adult anatomy, and the shortening of the SFA path length with leg flexion, it follows that repetitive arterial deformations may vary qualitatively and quantitatively with age. Namely, with decreased vessel stretch and increased vessel stiffness, the SFAs of older people tend to shorten less due to reaching the point of slack, and buckle off-axis, while younger SFAs remain straight. With the presence of atherosclerotic disease, arteries tend to be even stiffer, have longer equilibrium lengths, and lower longitudinal strain, probably resulting in more exacerbated off-axis buckling with repetitive, lower extremity movements.
There were limitations to this study that warrant mention as well as future investigation. Due to variations in vascular anatomy, the arterial branches were not consistent between subjects, necessitating the use of averaging the deformation data into top, middle, and bottom thirds of the SFA. In addition, the subjects included in this study did not have lower extremity vascular disease. In the presence of disease, which could cause non-homogeneous vessel properties along the length of the SFA, the motions and deformations could be qualitatively and quantitatively different from those of these healthy subjects. Ultimately, deformation analysis should be performed pre- and post-treatment, to better evaluate how different treatments and implants may change the properties of the vessel, as well as provide insight into how to treat the artery back to a healthy biomechanical state.
The results and concepts of this study can be used to improve current device testing, development of future devices, as well as guide therapeutic strategies. The longitudinal, twisting, and bending deformations of the SFA can be used to refine computational and bench top durability tests for industry and FDA evaluation. Although flexibility is not the only factor in efficacy, benchtop and clinical experience with SFA stents have shown that greater flexibility, and greater uniformity of flexibility, are correlated to fracture resistance. For example, along the spectrum of open vs closed-cell designs, the more open the design, the greater the axial, twisting, and bending flexibility, resulting in lower fracture rate (24
). In addition, evidence shows that longer stented regions correlate with higher stent fracture rates, especially in the presence of stiffness non-uniformities caused by stent overlap (36
). We can postulate that the low fracture rate of stent grafts may be due to their flexibility as well as the stiffness homogenizing effect of the graft material on the device. Not only can the results from this study extend the understanding of fracture mechanisms of SFA stents, they can also motivate the invention of next generation devices and therapies that better consider the biomechanical environment of the SFA.
Perhaps most importantly, the measurement of in vivo
longitudinal strain provides mechanical tissue data that can be used to help evaluate the biomechanical health of a vessel, potentially enhancing the diagnostic process (8
), as well as providing information to enable subject-specific devices and therapies. For example, implants and treatments can be designed to be more harmonious with the biomechanical environment by matching the strain state of an implant with that of the target vessel (e.g. implanting a stretched stent in the stretched vessel of a straight leg, or implanting a neutral stent in the non-strained vessel of a flexed leg). In addition, with vessel strains derived from these techniques and vessel stiffness derived from pulse propagation studies, vessel stresses and forces can be computed, creating more opportunities for anatomically-focused design.