Background
Peripheral arterial disease (PAD) affects more than 8 million adults in the U.S. [
1]. Those affected with PAD have impaired exercise tolerance, progressive functional decline, reduced quality of life [
2] and increased cardiovascular mortality [
3]. PAD extends beyond large vessel stenosis to impaired skeletal muscle perfusion [
4]. The ankle-brachial index (ABI) is a clinical non-invasive measure of PAD with excellent diagnostic and prognostic value [
5]. However, abnormal blood flow to the lower extremity measured by an ABI < 0.90 does not correlate well with time to initial claudication or maximum claudication distance [
6].
Our group has shown that contrast-enhanced perfusion cardiovascular magnetic resonance (CMR) of the calf muscle at peak exercise using a plantar-flexion ergometer reproducibly discriminates PAD patients from healthy volunteers independent of exercise workload [
7] and that it correlates with walking distance [
8]. One important limitation of contrast enhanced CMR using gadolinium chelates is the risk of nephrogenic systemic fibrosis in patients with advanced chronic kidney disease (CKD)(Glomerular filtration rate < 30 ml/min/1.73 m
2) [
9], which commonly coexists with PAD [
10].
Arterial spin labeling (ASL) is a CMR technique capable of quantifying perfusion in a spatially and temporally resolved fashion on a ml/min-100 g basis without the use of exogenous contrast [
11,
12]. Its application to calf perfusion has been validated against venous occlusion plethysmography [
13]. Wu et al. used a continuous ASL technique in combination with cuff occlusion to measure post-ischemic reactive hyperemia in the calf using the endpoints of peak hyperemic perfusion and time to peak perfusion [
14]. Our group recently showed that peak exercise pulsed ASL differentiates PAD patients with claudication from healthy (NL) subjects with excellent test-retest and inter-observer reproducibility [
15]. Similar to our findings with CMR perfusion, the difference remained even when healthy subjects matched PAD patients’ exercise time.
Cuff studies would facilitate quantitative perfusion measurements in those unable to exercise due to critical limb ischemia (CLI), severe claudication, or arthritis. Additionally, cuff studies allow the assessment of hyperemic response in muscles not typically used during ankle flexion exercise. Compared to exercise, cuff occlusion hyperemia is an effort-independent stress, which may further improve reproducibility and maximize the hyperemic response. We hypothesized that Cuff would yield higher perfusion values and improved reproducibility in PAD and healthy subjects compared to Exercise.
Discussion
ASL CMR enables the measurement of calf muscle perfusion in healthy volunteers and PAD patients without the use of exogenous contrast in combination with exercise or cuff occlusion stress paradigms. We found that calf muscle perfusion is significantly higher in controls compared to PAD patients independent of stress modality. Perfusion was similar to exercise levels when Cuff perfusion was averaged. However, Cuff yielded greater maximal calf muscle perfusion than exercise within subject groups; an effect driven by higher perfusion values in the soleus. Soleus perfusion in the PAD-Cuff group was similar to exercise perfusion in normal subjects. When the same subjects underwent both stressors, perfusion was similar. Interobserver agreement was excellent for both methods, while test-retest reproducibility and area under the ROC curve were superior with Cuff.
Cuff occlusion has been reported to induce substantial increases in perfusion to all calf muscle groups [
14,
18,
19]. However, there is significant heterogeneity in the magnitude of hyperemia with higher perfusion observed in the soleus muscle. In this study we show similar findings, with maximal Cuff perfusion occurring in the soleus in 83% of subjects. Using CE- and ASL-CMR we previously showed that supine ankle flexion-dorsiflexion exercise induces localized hyperemia in the anterior and lateral compartments and gastrocnemius muscles [
7,
15]. Consequently, we compared exercise hyperemia to maximal and average Cuff perfusion. Both parameters differentiated controls and PAD. Maximal perfusion values were greater with Cuff, driven by significantly larger hyperemic responses in the soleus muscle. Using average calf perfusion, perfusion within subject groups was similar. Furthermore, Cuff
mean
showed higher reproducibility and better discriminatory properties.
Our findings confirm those of Wu et al. [
14] indicating that in PAD patients reactive hyperemia is relatively preserved in the soleus and, although lower than NL-Cuff soleus, is similar to maximal NL-Ex perfusion. The mechanisms underlying this phenomenon are incompletely understood. The soleus receives dual blood supply from the posterior tibial and peroneal vessels. Additionally, it is a predominantly red muscle composed of Type I oxidative fibers. Animal studies have shown that oxidative muscles have increased capillary density and increased capillary to muscle fiber ratio compared to Type II glycolytic muscles such as the anterior tibialis [
20]. Histopathological studies evaluating the soleus capillary anatomy in PAD patients is limited, but Askew et al. [
21] have shown decreased Type I fibers and capillary density in the gastrocnemius muscle of patients with claudication. Thus, a combination of baseline differences between the soleus and other calf muscle capillary density, accentuated by adaptive responses in PAD, may be responsible in part for this finding.
The perfusion time course analysis indicates that the time to peak perfusion occurs relatively soon after cuff release for both normal subjects and PAD patients, albeit somewhat longer in PAD patients. This is consistent with prior studies using strain gauge plethysmography. Despite similar temporal resolution, these findings do not agree with those of Wu et al. who reported mean TTP of 45 – 55 s in NL, 65 – 76 s in moderate PAD and up to 86 – 99 s in those with ABI < 0.5 [
14,
18] when measured with CASL. Similarly, Grozinger et al. reported TTP of around 60s in claudicants with moderate to severely reduced ABI using pseudo-continuous ASL (PCASL) [
22]. The reasons for the discrepancy in TTP results are unclear, but may be due in part to differences in experimental design and/or type of ASL used. We did not image throughout the occlusion period which may have caused a short delay in what we determined to be time zero.
There are multiple versions of ASL that fall into three primary types of preparation techniques to magnetically label inflowing blood protons: pulsed (PASL), continuous or pseudo-continuous (CASL, PCASL), and velocity-selective ASL [
23]. In our study we used PASL utilizing the PICORE and Q2TIPS techniques. Advantages of PASL over CASL include lower radiofrequency (RF) power deposition, and the absence of need for continuous RF transmit hardware [
17,
24], which is not readily available in most centers. PCASL does provide higher signal-to-noise ratio and does not require specialized hardware. Hence, it has been recommended as the preferred mode for brain perfusion imaging [
23]. Its application to calf perfusion appears promising as demonstrated in a study calf perfusion changes before and after percutaneous angioplasty in a pilot study of 10 claudicants [
22].
The ability to quantitatively measure calf perfusion without the use of gadolinium is an important advantage over CMR because of the high prevalence of advanced CKD [
25] in PAD patients who are most vulnerable to nephrogenic systemic fibrosis [
9]. Another benefit of a contrast-free technique is the elimination of the need for blood chemistry tests and intravenous line placement. Lastly, with this particular sequence, perfusion maps are instantly reconstructed online, and easily analyzed at the MR console.
Optimal stress modality
Because skeletal muscle perfusion at rest is extremely low in both controls and PAD patients, ASL cannot be used to quantify rest perfusion as perfusion values are at the level of noise in the images. Hence, ASL requires a physiological challenge to measure skeletal muscle tissue perfusion. Exercise and cuff occlusion are commonly used methods, but it has been unclear if one is superior to the other. Our results would indicate that cuff occlusion improves upon the test-retest reproducibility of exercise. The superior reproducibility of Cuff may be mediated by the elimination of the effort-dependent component of exercise stress. Interobserver agreement was excellent with both provocative tests.
Although exercise stress has somewhat lower reproducibility, it has close clinical correlates. Perfusion CMR at peak exercise has been shown to correlate with walking distance [
8] and time to claudication [
26]. Studies correlating functional capacity and Cuff have not been performed. However, PAD patients with CLI or arthritis, may be unable to adequately exercise. Furthermore, cuff occlusion elicits hyperemia beyond the muscles used during supine ankle exercises, which would allow studies of regional perfusion throughout the calf.
The optimal stress modality may ultimately be dependent on the patient cohort and the question at hand. Exercise may be useful for studies of therapies targeting improved perfusion in ambulatory patients with claudication; especially those utilizing exercise training. On the other hand, trials of therapies targeting CLI in patients who are generally unable to exercise may be best served by Cuff. In addition, a better discriminatory power and reproducibility suggests that Cuff
mean
may be the parameter of choice to study populations at risk such as young smokers and diabetics with borderline ABI.
Limitations
Exercise and cuff occlusion hyperemia are physiologically different entities and interpretation of the results is dependent on the analytic approach. The exercise used for this study does not elicit hyperemia in the soleus; therefore it is not known how exercise-induced soleus perfusion compares to cuff occlusion. Secondly, the primary outcome measures were obtained comparing exercise to Cuff in two different patient cohorts of similar clinical classification. However, results were similar in a subgroup that underwent testing with both stressors. Third, our PAD cohort was made up of patients with mild to moderate disease and a study of the feasibility of Cuff in patients with CLI is needed. In addition, Cuff occlusion is not applicable in patients with superficial femoral artery stents. Finally, we did not perform angiography or segmental pressures to localize disease.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
Drs. Kramer, Meyer, and Epstein receive research support from Siemens Healthcare.
Authors’ contributions
DL helped design the study, performed most of the experiments, analyzed all of the data, and drafted the manuscript. AWP recruited the exercise patients and performed all of those experiments and edited the manuscript. CHM helped design the study, develop and apply the pulsed ASL sequence, aided in the experiments, and edited the manuscript. FHE helped design the study, develop and apply the pulsed ASL sequence, aided in the experiments, and edited the manuscript. LZ helped optimize the pulsed ASL sequence, helped perform the studies, and edited the manuscript. AJP recruited some of the patients and controls, performed analysis, including the interobserver variability. RJ recruited some of the patients, performed some of the exercise experiments, and critically reviewed the manuscript. JRK helped recruit the patients and volunteers, shepherded them through the studies, and reviewed the manuscript. JMD helped with data analysis, database maintenance, and manuscript review. JMC performed all of the MR studies, helped with analysis and reviewed the manuscript. CMK planned the studies, wrote the grant that funded the studies, supervised all aspects of the study and critically edited the manuscript. All authors read and approved the final manuscript.