PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Am Coll Cardiol. Author manuscript; available in PMC 2010 June 23.
Published in final edited form as:
PMCID: PMC2763280
NIHMSID: NIHMS130098

Skeletal muscle microvascular flow in progressive peripheral artery disease: Assessment with continuous arterial spin-labeling perfusion magnetic resonance imaging

Structured Abstract

Objectives

We present the novel application of continuous arterial spin-labeling (CASL) magnetic resonance imaging (MRI) for the measurement of calf muscle perfusion in subjects with progressive peripheral arterial disease (PAD).

Background

PAD is largely considered to be a disease of conduit vessels. The impact of PAD upon microvascular flow in the end-organ, muscle, remains unknown. CASL is a noninvasive MRI method capable of measuring microvascular flow and may assist in our understanding of the impact of PAD upon the microvasculature.

Methods

Forty subjects with varying degrees of PAD and seventeen age-matched PAD-free subjects were recruited and underwent measurement of the ankle-to-brachial index (ABI) and CASL. Peak hyperemic flow (PHF) and time-to-peak (TTP) were computed and assessed as a function of ABI and calf muscle group.

Results

ABI dependence was found in both PHF (p = 0.04) and TTP (p < 10−4). While TTP responded almost immediately to increasing PAD severity, PHF was preserved until subjects fell into cateogory-2 and even longer in the soleus muscle.

Conclusions

CASL flow measurements correlate with disease state as measured by ABI, and also demonstrate preserved microvascular flow reserve in the presence of early to intermediate vascular disease.

Keywords: Arterial spin-labeling, peripheral arterial disease, skeletal muscle, magnetic resonance imaging

Introduction

Peripheral artery disease (PAD) refers to disorders of the circulatory system outside the brain and heart as a consequence of narrowing and/or obstruction of peripheral arteries that carry blood to the extremities (1). Recent studies demonstrate that the atherosclerotic process in lower extremity PAD is not confined to conduit vessel but also affects skeletal muscle flow reserve (2), metabolism (3), endothelial and muscle mitochondrial function (4), gene transcription (5), and apoptosis (6). Several methods have been applied to measure skeletal muscle perfusion, such as positron emission tomography (PET) (7), dynamic susceptibility magnetic resonance imaging (MRI) (8), and ultrasound (9), although each involves either ionizing radiation and/or the administration of exogenous contrast agents.

Arterial spin labeling (ASL) (10) is a noninvasive MRI technique that offers quantitative perfusion measurements. A typical ASL experiment comprises two scans: the ‘tag’ image is acquired after the protons in arterial blood are ‘tagged’ by radiofrequency (RF) pulses, most commonly by inverting the blood proton magnetization, whereas the ‘control’ image is obtained without net magnetization perturbation in arterial blood. Blood flow is then computed from the signal difference between the tag and control images. Currently, the gold standard for measuring perfusion in skeletal muscle is the microsphere experiment (11) which is highly invasive as it requires arterial injection and subsequent arterial sampling as well as tissue sampling. A comparison study of ASL and microsphere methods (12) has shown a clear agreement in rat leg muscle while ASL offers improved spatial and temporal resolution.

ASL has also been successfully applied to the measurement of blood flow in the extremity muscle of healthy humans (1316). We present the first attempt to employ a continuous version of ASL (CASL) to measure calf muscle perfusion in subjects with various degrees of PAD.

Materials

1. Subjects and MR imaging

The Institutional Review Board approved the study protocol and written consent was obtained from all subjects. A cohort of 40 subjects diagnosed with PAD and 17 age-matched PAD-free subjects were recruited in this study (age = 26–86 years, 24 females and 33 males).

Subjects first underwent measurement of ankle-to-brachial index (ABI). An ABI between 0.90 and 1.30 is generally considered to be normal, with an ABI between 0.90 and 1.00 indicating borderline PAD, an ABI between 0.50 and 0.90 can represent a wide range of moderate disease, and an ABI below 0.50 usually indicates severe PAD (17). We therefore defined four clinical categories of ABI in the following manner: category-0: 0.90 ≤ ABI ≤ 1.30, category-1: 0.70 ≤ ABI < 0.90, category-2: 0.50 ≤ ABI < 0.70, and category-3: ABI < 0.50, and stratified subjects accordingly. Fourteen of the subjects in category-0 were also included in a previous study (16).

MR imaging was conducted on a 3.0 Tesla Siemens Trio system (Erlangen, Germany) with a transmit/receive knee coil (Nova Medical, Inc., Wakefield, MA). Subjects were imaged supine. A single-slice version of the CASL sequence was used (18) with a single-shot gradient-echo echoplanar readout: field-of-view (FOV) = 22 cm, in-plane matrix size = 64×64, slice thickness = 1 cm, TR = 4 sec, TE = 17 msec, flip angle = 900, tagging duration = 2 sec, post-labeling delay = 1900 msec (15). The labeling plane was 6 cm apart from the imaging slice, proximally for the tag scan and distally for the control (15).

An ischemic-hyperemic paradigm was chosen to create a uniform challenge across muscle groups and across subjects. A Zimmer 1000 (Warsaw, IN) nonmagnetic tourniquet system, with thigh cuff, was used to create a 5-min period of ischemia at 250 mmHg, followed by a period of hyperemic flow. CASL imaging commenced upon cuff inflation and ended 3 min after cuff deflation. A two-dimensional spoiled gradient-echo sequence (TR/TE = 50/3.4 msec, FOV = 220 mm, in-plane matrix size = 256×256, flip angle = 500, with 4 averages) was used to acquire a high-resolution anatomic image of the slice where CASL imaging was performed.

2. Data processing and image analysis

Reconstructed magnitude images were analyzed off-line using VOXBO (http://www.voxbo.org/) and IDL (RSI, Boulder, CO). CASL signals were generated by pair-wise subtraction of tag and control images and two adjacent data points in time were averaged resulting in an effective temporal resolution of 16 sec. CASL signal (ΔM) was then converted to quantitative flow (f) in ml/100g/min:

f=λΔM2αT1M0[exp(PLDT1)exp(PLD+TT1)]
[1]

where M0 is the fully-relaxed blood signal and α is the tagging efficiency (0.80 (19)). We assumed that T1/T2* = 1600/100 msec for arterial blood at 3.0 Tesla and the blood/tissue partition coefficient (λ) = 0.9 ml/g, comparable to the value measured in the brain (20). For further details regarding the model, please refer to reference (16).

Four muscle groups in the mid-calf were analyzed, representing the vascular distributions of the three major branches of the popliteal artery. The soleus muscle (Sol) receives a dual vascular supply from both the posterior tibial and the peroneal arteries. The medial gastrocnemius muscle (Gstrc-M) is supplied by the posterior tibial artery. The anterior compartment (AC) contains extensor muscles, including the extensor digitorum longus, extensor hallicus longus, tibialis anterior, and peroneus muscles, and is supplied by the anterior tibial artery. Finally, the lateral compartment (LC) contains the peroneus longus and peroneus brevis muscles, and is supplied by the peroneal artery. Regions of interest (ROI’s) were hand-drawn on the high-resolution spoiled gradient-echo anatomic image (Figure 1). Two flow indices were computed as defined below (Figure 2):

  1. Peak Hyperemic Flow (PHF) (ml/100g/min): The peak flow observed in hyperemic period which spans from the time (tbeg) when flow increases above a threshold (fT) to the time (tend) when flow returns to fT. Here, tbeg is later than the time of cuff deflation and fT is chosen to be one standard deviation of the ‘flow’ during the center 3-min occlusion when ideally flow is zero.
  2. Time-To-Peak (TTP) (sec): Measured from the time when cuff is deflated to the time when hyperemic flow peaks.
Figure 1
Cross-sectional images of the mid-calf in a representative subject
Figure 2
Schematic illustration of the ischemic-hyperemic paradigm and the flow indices measured in this study

3. Statistical Analysis

A random effects model (21) was employed to test the dependence of PHF upon the variables muscle group and ABI, as well as for any interaction effects between muscle group and ABI (Muscle x ABI). Since TTP was a count of the number of 16-sec epochs and followed a Poisson distribution, the TTP measure was modeled using generalized estimating equation (22) with a Poisson family assumption. ABI was treated as a continuous measure in all analyses.

Results

Table 1 gives a breakdown of demographics for the population studied by ABI category. Overall, the groups are well balanced by age, disease characteristics, and gender. The relatively fewer number of subjects in category-3 reflects the decreased frequency of PAD of this severity and the difficulty of enrolling subjects with more severe disease.

Table 1
A demographic summary of the patients recruited in this study.

Table 2 offers a breakdown of mean±SD measurements for PHF and TTP by muscle group and ABI category. In Figure 3, the average of subjects’ individual flow-time curves is plotted for different muscle groups, as well as the PAD stages defined by ABI. Qualitatively it can be seen that, with increasing disease severity PHF decreases, TTP increases, and the hyperemic period is broadened. Figure 4 graphs both mean TTP and PHF by ABI category and muscle group. From Table 2 and Figure 3 and Figure 4, there is evidence of a fall in PHF in all muscle groups with increasing disease severity as reflected in the ABI category. For Sol, the decrease in PHF appears to be delayed until category-3. In contrast, TTP increase in response to increasing disease severity appears to occur in a nearly parallel fashion for all muscle groups. Thus, for most muscle groups falling PHF as well as increasing TTP occur with early stages of disease, yet uniquely for Sol, increasing TTP may be a more sensitive indicator of early vascular disease.

Figure 3
Perfusion-time curves for different muscles and categories of the ankle-to-brachial index (ABI)
Figure 4
Relationship between flow indices ((a) peak hyperemic flow, (b) time-to-peak) and two main effects
Table 2
A summary of the measured flow indices expressed in Mean±SD.

Shown in Figure 5 is a scatter-plot with mean TTP versus mean PHF. The data show that TTP is inversely associated with PHF. TTP of 60 sec in combination with PHF of 63 ml/100g/min provides an approximate demarcation between categories 0–1 and 2–3, with an exception of category-2 Sol (see the arrow). This may indicate somewhat better flow preservation with severe PAD in Sol than in other muscle groups.

Figure 5
Relationship between time-to-peak and peak hyperemic flow

Statistical analysis of the dependence of PHF and TTP upon ABI and muscle groups is summarized in Table 3a and Table 3b, respectively. PHF was dependent upon muscle group (p < 0.0001) and ABI (p = 0.04). Within the muscle groups studied, PHF for Sol demonstrated relative resistance (coefficient = 26, p < 0.0001) and LC relatively greater sensitivity (coefficient = −14, p < 0.0001) to the effect of increasing ABI, all relative to the Gstrc-M reference group. TTP was found to be dependent upon both muscle group (p = 0.001) and ABI (p < 0.0001), with TTP increasing with increasing severity of disease. Although TTP was dependent upon muscle group, only AC appeared to demonstrate significant deviation in the behavior of TTP from the reference Gstrc-M group.

Discussion

Peripheral artery disease (PAD) is usually associated with discrete lesions within one or several vessels. The traditional understanding of PAD is one of progressive blood flow impairment in conduit vessels that, through the creation of chronic ischemic environment, ultimately affects the end organ, skeletal muscle (2). While artery supply varies between muscle groups, flow heterogeneity between muscle groups is thought to be a consequence of different composition of myofibril type and/or metabolic profile (7, 16). It is therefore conceivable that muscle groups may be differentially affected by PAD.

In the present study, we measured two indices of post-ischemic reactive hyperemia, peak hyperemic flow (PHF) and time-to-peak (TTP), in healthy subjects and in subjects with a range of PAD using a completely noninvasive MRI methodology, CASL. PHF was noted to decrease and TTP to increase with decreasing ABI and in conjunction with increasing PAD severity. While TTP appears to respond almost immediately to PAD progression, PHF is relatively well preserved until subjects fall into cateogory-2 and appears to be preserved even longer in the soleus muscle.

We and others have previously demonstrated that PHF is higher in soleus muscle than in all other calf muscle groups at baseline in healthy subjects (8, 16). The relative resilience of hyperemic flow to advancing PAD in soleus muscle, as compared to other muscle groups, is however revealed for the first time. Soleus muscle is composed of approximately 70–80% slow-twitch type I fibers with a higher capillarity and oxidative capacity, which enables soleus muscle to convert and utilize energy more effectively and thus better endure insufficient supply of flow, oxygen and nutrients accompanying disease progression (i.e., greater resistance to hypoxia). By contrast, gastrocnemius is more evenly composed of type I (50–60%) and type II fibers. Myofibril composition and differences in metabolic profiles may also account for the relative resistance of the soleus muscle to the presence of PAD.

We also observed an early prolongation in TTP with disease progression that increased in a parallel fashion amongst all the muscle groups studied (Figure 4b). This greater similarity in behavior of TTP across muscle groups is not surprising in that we have previously demonstrated that TTP is independent of muscle group (16). We also found an inverse correlation between TTP and PHF (Figure 5). While TTP increased early in disease, PHF did not decrease immediately, suggesting that microvascular reactivity partially compensates early on for the narrowing in large feeding vessels. In comparison with PHF, TTP appears more sensitive to early disease progression. A combination of the two flow indices may offer more comprehensive information for the grading and/or diagnosis of PAD.

In conclusion, CASL flow measurements correlate with disease state as measured by ABI, and demonstrate preserved microvascular flow reserve in the presence of early to intermediate vascular disease (categories 0–2). The ABI, an indirect measure of large vessel stenosis, tracks with TTP, preceding perfusion changes and countered by a preserved microvascular flow reserve until late stage in disease progression. The ASL approach may have several unique properties in the diagnosis and study of PAD, offering an extrinsic “contrast free” approach to assess end-organ microvascular results of chronic disease and therapy.

Acknowledgements

The authors thank Temitope Olufade, Bs, MPH, for research coordination, Jiongjiong Wang, PhD for technical consultation, and Lee J. Milas, BS for assisting with data acquisition. We also acknowledge the support from the NIH grant R01HL075649.

Abbreviation list

ABI
sankle-to-brachial index
ASL
arterial spin labeling
CASL
continuous arterial spin-labeling
MRI
magnetic resonance imaging
PAD
peripheral arterial disease
TR
repetition time
TE
echo time
T1
longitudinal relaxation time constant
T2
transverse relaxation time constant
PHF
peak hyperemic flow
TTP
time-to-peak

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

1. Mohler ER., 3rd Peripheral arterial disease: identification and implications. Arch Intern Med. 2003;163:2306–2314. [PubMed]
2. Bragadeesh T, Sari I, Pascotto M, Micari A, Kaul S, Lindner JR. Detection of peripheral vascular stenosis by assessing skeletal muscle flow reserve. J Am Coll Cardiol. 2005;45:780–785. [PubMed]
3. Isbell DC, Berr SS, Toledano AY, et al. Delayed calf muscle phosphocreatine recovery after exercise identifies peripheral arterial disease. J Am Coll Cardiol. 2006;47:2289–2295. [PMC free article] [PubMed]
4. Davidson SM, Duchen MR. Endothelial mitochondria: contributing to vascular function and disease. Circ Res. 2007;100:1128–1141. [PubMed]
5. Bosch-Marce M, Okuyama H, Wesley JB, et al. Effects of aging and hypoxia-inducible factor-1 activity on angiogenic cell mobilization and recovery of perfusion after limb ischemia. Circ Res. 2007;101:1310–1318. [PubMed]
6. Mitchell RG, Duscha BD, Robbins JL, et al. Increased levels of apoptosis in gastrocnemius skeletal muscle in patients with peripheral arterial disease. Vasc Med. 2007;12:285–290. [PubMed]
7. Kalliokoski KK, Kemppainen J, Larmola K, et al. Muscle blood flow and flow heterogeneity during exercise studied with positron emission tomography in humans. Eur J Appl Physiol. 2000;83:395–401. [PubMed]
8. Thompson RB, Aviles RJ, Faranesh AZ, et al. Measurement of skeletal muscle perfusion during postischemic reactive hyperemia using contrast-enhanced MRI with a step-input function. Magn Reson Med. 2005;54:289–298. [PMC free article] [PubMed]
9. Weber MA, Jappe U, Essig M, et al. Contrast-enhanced ultrasound in dermatomyositis- and polymyositis. J Neurol. 2006;253:1625–1632. [PubMed]
10. Detre JA, Leigh JS, Williams DS, Koretsky AP. Perfusion imaging. Magn Reson Med. 1992;23:37–45. [PubMed]
11. Laughlin MH, Korthuis RJ, Sexton WL, Armstrong RB. Regional muscle blood flow capacity and exercise hyperemia in high-intensity trained rats. J Appl Physiol. 1988;64:2420–2427. [PubMed]
12. Pohmann R, Kunnecke B, Fingerle J, von Kienlin M. Fast perfusion measurements in rat skeletal muscle at rest and during exercise with single-voxel FAIR (flow-sensitive alternating inversion recovery) Magn Reson Med. 2006;55:108–115. [PubMed]
13. Lebon V, Carlier PG, Brillault-Salvat C, Leroy-Willig A. Simultaneous measurement of perfusion and oxygenation changes using a multiple gradient-echo sequence: application to human muscle study. Magn Reson Imaging. 1998;16:721–729. [PubMed]
14. Frank LR, Wong EC, Haseler LJ, Buxton RB. Dynamic imaging of perfusion in human skeletal muscle during exercise with arterial spin labeling. Magn Reson Med. 1999;42:258–267. [PubMed]
15. Wu WC, Wang J, Detre JA, Ratcliffe SJ, Floyd TF. Transit delay and flow quantification in muscle with continuous arterial spin labeling perfusion-MRI. J Magn Reson Imaging. 2008;28:445–452. [PMC free article] [PubMed]
16. Wu WC, Wang J, Detre JA, et al. Hyperemic flow heterogeneity within the calf, foot, and forearm measured with continuous arterial spin labeling MRI. Am J Physiol Heart Circ Physiol. 2008;294:H2129–H2136. [PMC free article] [PubMed]
17. Hiatt WR. Medical treatment of peripheral arterial disease and claudication. N Engl J Med. 2001;344:1608–1621. [PubMed]
18. Alsop DC, Detre JA. Reduced transit-time sensitivity in noninvasive magnetic resonance imaging of human cerebral blood flow. J Cereb Blood Flow Metab. 1996;16:1236–1249. [PubMed]
19. Roberts DA, Detre JA, Bolinger L, Insko EK, Leigh JS., Jr Quantitative magnetic resonance imaging of human brain perfusion at 1.5 T using steady-state inversion of arterial water. Proc Natl Acad Sci U S A. 1994;91:33–37. [PubMed]
20. Herscovitch P, Raichle ME. What is the correct value for the brain--blood partition coefficient for water? J Cereb Blood Flow Metab. 1985;5:65–69. [PubMed]
21. Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics. 1982;38:963–974. [PubMed]
22. Zeger SL, Liang KY. Longitudinal data analysis using generalized linear models. Biometricka. 1986;73:13–22.