Journal of Applied Physiology Virginia Commonwealth University
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 92: 1709-1716, 2002; doi:10.1152/japplphysiol.00445.2001
8750-7587/02 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fischman, A. J.
Right arrow Articles by Alpert, N. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fischman, A. J.
Right arrow Articles by Alpert, N. M.
Vol. 92, Issue 4, 1709-1716, April 2002

Regional measurement of canine skeletal muscle blood flow by positron emission tomography with H215O

Alan J. Fischman1,2,4, Hongbing Hsu1,4, Edward A. Carter2,3,5, Yong M. Yu2,3,5, Ronald G. Tompkins2,3,5, J. Luis Guerrero3, Vernon R. Young2,3, and Nathaniel M. Alpert1

1 Division of Nuclear Medicine, Department of Radiology, and 3 Surgical Service, Massachusetts General Hospital, and 2 Shriners Burns Institute, Boston 02114; and Departments of 4 Radiology and 5 Surgery, Harvard Medical School, Boston, Massachusetts 02115


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Positron emission tomography (PET) with H215O was used as an in vivo, relatively noninvasive, quantitative method for measuring regional blood flow to hindlimb skeletal muscle of anesthetized dogs. A hydrooccluder positioned on the femoral artery was used to reduce flow, and high-flow states were produced by local infusion of adenosine. Three to four measurements were made in each animal. Approximately 40 mCi of H215O were injected intravenously, and serial images and arterial blood samples were acquired over 2.5 min. Data analysis was performed by fitting tissue and arterial blood time-activity curves to a modified, single-compartment Kety model. The model equation was also solved on a pixel-by-pixel basis to yield maps of regional skeletal muscle blood flow. After each PET determination, flow was measured with radioactive microspheres. Results of the PET measurements demonstrated that basal flow to hindlimb skeletal muscle was 3.83 ± 0.36 ml · min-1 · 100 g-1 (mean ± SE). This value was in excellent agreement with the microsphere data, 3.73 ± 0.32 ml · min-1 · 100 g-1 (P = 0.69, not significant). Adenosine infusion resulted in flows as high as 30 ml · min-1 · 100 g-1, and the PET and microsphere data were highly correlated over the entire range of flows (r2 = 0.98, P < 0.0001). We conclude that muscle blood flow can be accurately measured in vivo by PET with H215O and that this approach offers promise for application in human studies of muscle metabolism under varying pathophysiological states.

hindlimb; microspheres; Doppler flow probe; dogs


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

ALTERATIONS IN TISSUE PERFUSION are common in a variety of pathophysiological states, including neoplasia, acute inflammation, and peripheral obstructive arterial disease (17, 44). Furthermore, changes in blood flow and, consequently, the supply of substrates and nutrients to tissues and organs and removal of metabolites must be considered as an important feature in metabolic regulation (10, 12). Because muscle metabolism changes with aging (51), is altered by many disease processes (40), and is involved in the adaptation of body energy and protein metabolism (49, 50), an accurate and precise method for the in vivo, relatively noninvasive quantification of blood flow in this tissue would be of considerable value.

Numerous tracer techniques, including use of radioactive inert-gas clearance (5, 8, 9, 11, 16, 28) and radioactive microspheres (13, 30, 42, 48), have been applied in measurements of muscle blood flow. Most of these measurements are based on the Kety-Schmidt principle (22, 23); namely, the rate of washout of a diffusible tracer from tissue is proportional to blood flow. Although several of these techniques have been used successfully in human metabolic and clinical studies, problems such as tissue specificity, requirements of prolonged measurements, and tissue sampling remain. Hence, we studied the validity of the positron emission tomography (PET) H215O method in a canine model, before carrying out physiological studies in both healthy subjects and hospitalized patients, for this purpose.

Intravenous bolus injection of H215O was used in the present study to measure regional blood flow with PET in dog skeletal muscle. Multiple measurements were performed in six postabsorptive dogs. A modified Kety model, which takes into account the time delay of the arterial bolus and dispersion of blood radioactivity, was used to fit the tissue and blood time-activity curves (TACs). The results of the PET measurements were compared with microsphere data [a "gold standard" for tissue perfusion (3)] acquired after each H215O injection.


    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Materials. Microspheres (15 µm) radiolabeled with 113Sn, 103Ru, 96Nb, 46Sc, or 141Ce and suspended in 0.9% saline and 0.01% Tween 80 to reduce aggregation were obtained from Dupont Radiopharmaceutical Division (Billerica, MA). Adenosine was purchased from Sigma Chemical (St. Louis, MO), and pentobarbital sodium (65 mg/ml) was purchased from Anthony Products (Arcadia, CA).

Preparation H215O. 15O2 was produced by the 14N2(d,n)15O2 reaction by using N2 gas with 1% O2 carrier in a 250-ml target at 68 psi pressure. A 6-min irradiation at 40 µA produced ~2 Ci of 15O2. The 15O2 gas was mixed with H2 and passed over a palladium catalyst heated to 450°C to produce H215O by oxidation. The water vapor was collected in a 10-ml vial containing 2 ml of isotonic saline. An aliquot of this solution was drawn through a 0.22-µm filter into a sterile, disposable syringe for injection.

Animal preparation. Mongrel dogs, weighing 25-30 kg, were obtained from a commercial supplier (Buckshire, Perkasie, PA) and housed in the Massachusetts General Hospital (MGH) animal farm. The animals were cared for in accordance with the guidelines set forth by the Committee on Laboratory Resources, National Institutes of Health Council [DHEW (DHHS) Publication No. (NIH) 78-23, DRR/NIH, Bethesda, MD 20892], and the protocol was approved by the MGH Committee on Animal Research. Before the tracer studies, the dogs were fasted overnight, and surgery was performed on the following morning by using aseptic techniques. The anesthesia used was an intravenous bolus of pentobarbital sodium (15 mg/kg) followed by a constant infusion (1 mg · kg-1 · h-1). The surgical procedures included implantation of venous, arterial, and atrial catheters, hydrooccluders, and Doppler flow probes. Polyethylene catheters (PE-90 or PE-260; Clay Adams, Parsippany, NJ) with Silastic tips (Dow Corning, Midland, MI) were inserted into the left jugular vein, left carotid and brachial arteries, and left atrium, as previously described (18). For placement of Doppler flow probes, incisions were made in the groin regions, and 1.5-cm segments of the femoral arteries were surgically isolated. The left femoral artery was instrumented with a 6-mm hydrooccluder (Biomedical Products, Silver Spring, MD). Doppler flow probes (size 2R, Transonic System, Ithaca, NY) were placed on both femoral arteries ~3 mm distal to the position of the occluder. A 0.7-mm cannula for adenosine infusion was inserted into the lumen of the left femoral artery just proximal to the occluder. Precise placement of this cannula was achieved by insertion via the deep femoral artery.

PET imaging. Anesthetized and mechanically ventilated dogs were positioned and stabilized in the gantry of a PC-4096 plus PET camera (Scanditronix, Uppsala, Sweden). The primary imaging characteristics of the PC-4096 camera are in-plane and axial resolutions of ~6 mm full width at half-maximum (FWHM), 15 contiguous images (PET slices) of 6.5-mm separation, and a sensitivity of ~5,000 cycles · s-1 · µCi-1 (41). In all animals, a 9.5-cm segment of muscle in the upper hindlimb (upper limit ~3 cm from the hip joint) was imaged.

At baseline flow (occluder open and adenosine infusion off) ~40 mCi of H215O were injected as a bolus through the left jugular vein catheter, and sequential images were acquired in 2-s frames over a period of 2.5 min. Arterial blood was sampled from the left carotid artery with a rotary pump (flow rate ~8 ml/min), and 15O radioactivity was recorded at 0.5-s intervals for 2.5 min with two pairs of on-line coincidence detectors. At the conclusion of imaging and arterial blood sampling, radioactive microspheres were injected via the left atrial catheter, and skeletal muscle blood flow was determined as described below. After the microsphere measurement was completed, flow was reduced with the occluder or increased by adenosine (1.0 mg/kg) infusion at variable rates through the femoral arterial catheter. After a stable flow was achieved for 5 min, the PET and microsphere measurements were repeated; three to four measurements were performed in six animals. During the PET and microsphere procedures at each flow value, flow through the femoral arteries was monitored with the Doppler probes, and conditions with varying flow were excluded from further analysis. A continuous intravenous saline drip was used to restore plasma volume lost during the blood-sampling procedures. Throughout the study, the animals were mechanically ventilated, and PCO2 and pH were maintained in the ranges 38-44 Torr and 7.35-7.45, respectively. A schematic diagram of the experimental protocol is illustrated in Fig. 1.


View larger version (7K):
[in this window]
[in a new window]
  		Fig. 1.   Schematic diagram of the experimental protocol used for positron emission tomography (PET) and microsphere measurements of blood flow to canine hindlimb skeletal muscle.

The PET images were reconstructed with a standard filtered back-projection algorithm, a Hanning filter with a width of 4 mm. Data for attenuation correction were measured with a 68Ga rotating pin source. All projection data were corrected for nonuniformity of detector response, dead time, random coincidences, and scattered radiation. The PET camera was cross-calibrated to a well scintillation counter by comparison of the camera response from a uniform distribution of a 18F solution in a 20-cm cylindrical phantom with the response of the well counter to an aliquot of the same solution.

The sequence of PET images was summed to allow for visualization of the distribution of H215O in muscle. At each flow condition, four to six irregular-shaped regions of interest (ROIs) were placed on selected PET slices (35). In positioning the ROIs, particular caution was taken to exclude bone and major vascular structures. The ROIs were propagated over all of the time frames of the dynamic PET acquisitions, and the concentrations of radioactivity were averaged to yield composite TACs. The upper and lower borders of the segment of hindlimb that was imaged were located with the camera gantry lasers (which are registered to the PET slices), and their positions were marked on the skin with ink. Additional ink marks were placed around the circumference of the limb. After the animals were killed, these markers were used to produce sketches of the cross section of the imaged muscle, which were compared with the PET ROIs to guide excision of tissue samples for measurement of microsphere radioactivity.

Flow probe measurements. To determine the reliability of using the flow probe for setting specific (reduced and increased) flow states, a preliminary study was performed with five dogs. In this investigation, simultaneously with PET measurements, blood flow through the left femoral artery was measured with the flow probe over a range of constrictor settings with previously reported methods (14); however, microsphere measurements were not performed. For these PET studies, larger ROIs (total 20-30 cm3 of tissue) were used for constructing the TACs. At the conclusion of each PET study, the flow probe was recalibrated by controlled volumetric collection of blood drained from the femoral artery. The combined PET-microsphere studies were performed with the same procedures.

Microsphere measurements of blood flow. Microsphere measurements of skeletal muscle blood flow were performed with the reference sample technique, assuming 100% first-pass extraction, as previously described (14). The microspheres were checked for size uniformity and absence of aggregation and fragmentation by light microscopy with a calibrated reticle. Approximately 2.5 × 107 spheres (113Sn, 103Ru, 96Nb, 46Sc, or 141Ce) were injected into the left atrium for each measurement of skeletal muscle blood flow. This quantity was selected to ensure that at least 1,000 microspheres were contained in the reference and tissue samples. With this number of spheres, the distribution variability of flows is expected to lie within 10% of the mean distribution at the 95% confidence level (14). Freshly drawn heparinized blood (20 ml) was used to flush each sample of microspheres from a specially designed chamber into the left atrium over 20 s. Before injection, the chamber containing the microspheres was vortexed for 5 min to ensure uniform suspension. Radioactivity in the chamber was measured before and after administration to precisely determine the injected dose. Beginning 15 s before injection of the microspheres, a reference blood sample was collected from the brachial artery with a Harvard infusion pump (Harvard Apparatus, South Natick, MA) at a constant rate (10 ml/min) over 2 min into preweighed heparinized vials. Volume was replaced with 0.9% saline via the jugular venous catheter. Tissue samples were excised as described above and weighed, and radioactivity was measured with a well-type gamma counter with computerized corrections for spillover between photopeaks (CompuGamma 1282, LKB-Wallac, Turku, Finland). Blood flow was calculated by using the relationship
F = S · [(A<SUB>o</SUB>/m<SUB>o</SUB>)/A<SUB>s</SUB>] (1)
where F is skeletal muscle blood flow (ml · min-1 · g-1), S is the rate of arterial blood withdrawal after microsphere injection (ml/min), Ao is the microsphere radioactivity trapped in a muscle sample (counts/min), As is the microsphere radioactivity in the blood sample (counts/min), and mo is the weight of the muscle sample (g).

Kinetic modeling. The configuration of the compartmental model used for data analysis is illustrated in Fig. 2. F is in milliliters per minute per 100 g, k2 is washout rate per minute, and C represents tissue concentration of H215O. The arterial blood concentration of H215O (Ca) represents the true input function to the tissue and can be calculated from the measured blood concentration (Cs) as follows.


View larger version (4K):
[in this window]
[in a new window]
  		Fig. 2.   Kinetic model for H215O transport by skeletal muscle. F, blood flow (ml · min-1 · 100 g-1); k2, washout rate (per minute); Ca and Ct, arterial and tissue concentration of H215O, respectively.

The simple "Kety flow model" can be expressed mathematically by
C(<IT>t</IT>) = F<IT>e</IT><SUP>−(<IT>k</IT><SUB>2</SUB>+&lgr;)<IT>t</IT></SUP> ⊗ C<SUB>a</SUB>(<IT>t</IT>) (2)
where t is time and lambda  is radioactivity decay constant of 15O.

Because arterial blood was sampled from the carotid artery, a small time delay and dispersion were expected to occur. We used a simple exponential dispersion function to approximate these effects, as expressed in the following equation (27)
C<SUB>s</SUB>(<IT>t+&dgr;</IT>)<IT>=pe</IT><SUP>−(<IT>p</IT>+&lgr;)<IT>t</IT></SUP> ⊗ C<SUB>a</SUB>(<IT>t</IT>) (3)
where delta  and p are time delay and dispersion parameters, respectively.

After Eqs. 2 and 3 were combined and the integration was performed, the final mathematical model can be summarized by the following equation
C(<IT>t</IT>) = <FR><NU>F</NU><DE><IT>p</IT></DE></FR> <FENCE>C<SUB>s</SUB>(<IT>t+&dgr;</IT>)<IT>−e</IT><SUP>−(<IT>k</IT><SUB>2</SUB>+&lgr;)<IT>t</IT></SUP>C<SUB>s</SUB>(&dgr;) </FENCE> (4)

<FENCE>+ (<IT>p − k</IT><SUB>2</SUB>)<IT>e</IT><SUP>−(<IT>k</IT><SUB>2</SUB>+&lgr;)(<IT>t</IT>+&dgr;)</SUP><LIM><OP>∫</OP><LL>&dgr;</LL><UL><IT>t</IT>+&dgr;</UL></LIM><IT>e</IT><SUP>(<IT>k</IT><SUB>2</SUB>+&lgr;)&eegr;</SUP>C<SUB>s</SUB>(&eegr;)d&eegr;</FENCE>
F, k2, p, and delta  were estimated by nonlinear least square fitting by using the MATLAB software package. Because trial fittings demonstrated that dispersion was negligible for the present study, the dispersion parameter p was fixed at 30 per minute, which corresponds to minimal dispersion. In the fitting procedure, instead of calculating k2, 1/V was determined; V is the distribution volume or partition coefficient of water in muscle. The relation between k2 and 1/V can be expressed by following equation
k<SUB>2</SUB>=<FR><NU>1</NU><DE>V</DE></FR><IT> · </IT>F (5)
Flow maps were computed with fixed delay delta  and dispersion p by applying weighted least squares in a voxel-by-voxel fashion. First, the value of delta  was determined by nonlinear least squares fitting of Eq. 4 to the data extracted from a large ROI placed over muscle. Dispersion was set at 30 per minute. After delay and dispersion are taken into account, the Kety flow model for an elemental volume was integrated to yield
C(<IT>t</IT>) = <FR><NU>F</NU><DE><IT>p</IT></DE></FR> <FENCE>C<SUB>s</SUB>(<IT>t+&dgr;</IT>)<IT>+</IT>(<IT>p+&lgr;</IT>)<LIM><OP>∫</OP><LL>0</LL><UL><IT>t</IT></UL></LIM>C<SUB>s</SUB>(<IT>u+&dgr;</IT>)d<IT>u</IT></FENCE> (6)

<IT>−k</IT><SUB>2</SUB><LIM><OP>∫</OP><LL>0</LL><UL><IT>t</IT></UL></LIM>C(<IT>u</IT>)d<IT>u</IT>
Eq. 6 is mathematically equivalent to Eq. 4 and can be solved for F and k2 on a pixel-by-pixel basis by weighted linear least squares. In the present study, we used a weighting inversely proportional to the total counts recorded for each frame. The kinetic modeling procedures yielded values for blood flow in units of milliliters per minute per milliliter of tissue. To express the results as milliliters per minute per 100 g of tissue, these values were multiplied by the density of skeletal muscle (1.04 g/ml) × 100 (46).

Statistical analysis. As discussed in more detail below, statistical comparisons of H215O-PET and microsphere measurements are complicated by the fact that both methods are subject to unknown biases and statistical fluctuations. Rigorous statistical methods do not exist for characterizing the differences in a definitive manner.

Two approximate analyses were performed. In the first analysis, data obtained with PET and microspheres in the basal state were compared by using the paired Student's t-test. A second analysis was performed by using linear regression, assuming that the microsphere data were both unbiased and measured with negligible random errors. Additional simulation studies were performed to assess the likelihood that data satisfying the null hypothesis could yield results similar to those found in this study. The simulation assumed that the expected value of both PET and microsphere measurements was identical. Normally distributed random noise was added to the expected values for both methods, and the results were analyzed as above to yield a slope and intercept. Values similar to those found from the measured data occurred once in 10,000 simulations, again suggesting some difference between the methods. Further characterization of differences that appear to be on the order of 5-10% in the normal physiological range is hampered by a lack of detailed knowledge of the bias in the methods.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Figure 3 shows a least squares fit of a TAC, for the average concentration of H215O in hindlimb muscle, to the kinetic model presented in Fig. 2. All values have been corrected for radioactive decay. The time dependence of the residuals derived from the least squares procedure is also illustrated. These results show that the time dependence of the concentration of H215O in skeletal muscle is well described by the model. This is further supported by the random nature of the residuals.


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3.   Least squares fit of a PET time-activity curve for the average concentration of H215O in hindlimb muscle based on the kinetic model shown in Fig. 2. Inset: time dependence of the residuals derived from the least squares procedure.

The results of the preliminary studies to establish the utility of the Doppler flow probe setting skeletal muscle blood flows to approximate values for the combined PET and microsphere studies are illustrated in Fig. 4. Because the flow probe measures total flow to the limb, blood flow to muscle was estimated by assuming that ~80% of total limb blood flow perfuses this tissue (6, 34). As illustrated in Fig. 4, after this correction was made, flow probe and PET measurements of muscle blood flow were not significantly different. Furthermore, hindlimb muscle blood flows determined by the two methods were highly correlated (r2 = 0.92, P < 0.001). Least squares fitting of the hindlimb TACs acquired after injection of H215O yielded values for the volume of distribution of labeled water and the time delay from the site of blood sampling (carotid catheter) to the muscle bed that were 0.23 ± 0.073 ml/g and 0.012 ± 0.018 min, respectively.


View larger version (15K):
[in this window]
[in a new window]
  		Fig. 4.   Correlation between skeletal muscle blood flow determined by PET and the Doppler flow probe technique.

Figure 5 shows the values for basal hindlimb blood flow determined by microsphere techniques and PET with H215O. The results of the PET measurements demonstrated that basal flow to hindlimb skeletal muscle was 3.83 ± 0.36 ml · min-1 · 100 g-1 (mean ± SE). This value was in excellent agreement with the microsphere data, 3.73 ± 0.32 ml · min-1 · 100 g-1 (P = 0.69, not significant), and values reported in the literature (6, 29, 36, 43). By use of the occluder and adenosine infusion, flows over the range of ~2-30 ml · min-1 · 100 g-1 were achieved. Adenosine infusion did not affect flows recorded with the contralateral flow probe. As illustrated in Fig. 6, over this ~15-fold range of blood flows, the PET (FlowPET) and microsphere (FlowµSph) data were highly correlated (r2 = 0.98, P < 0.0001), and regression analysis yielded the following equation (r2 ~ 0.98)
Flow<SUB>PET</SUB> = 0.911 × Flow<SUB>&mgr;Sph</SUB> + 0.448 (7)
The regression coefficients analyzed with Student's t-test (degrees of freedom = 26) did not support the hypotheses that the intercept was zero (t = 2.74) and the slope was unity (t = 5.58). However, even though the results are not identical and may even differ by a small amount, the systematic differences are unlikely to be of any importance in actual practice. In addition to deriving flows for specific muscle groups, an even more important characteristic of the PET method is its ability to produce high-resolution quantitative maps of blood flow. Examples of flow maps produced under different flow conditions by using Eq. 6 are illustrated in Fig. 7.


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 5.   Values for blood flow to canine hindlimb skeletal muscle determined by the microsphere technique and PET with H215O. Individual values (solid circles) and means ± SE (bars) are illustrated. Data from 16 paired measurements performed in 6 animals are illustrated.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 6.   Correlation between microsphere and PET measurements of hindlimb skeletal muscle blood flow in dogs (r2 = 0.98, P < 0.0001). The relationship between the 2 measurements was FlowPET = 0.911 × FlowµSph + 0.448, where µSph is microsphere. Data from 28 paired measurements in 6 dogs are illustrated.



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 7.   Examples of blood flow maps (coronal projection) derived by pixel-by-pixel solution of Eq. 6. 1. Low flow, baseline map. 2. High-flow map derived after local infusion of adenosine in the right thigh. Blue area in upper right of both panels represents bone marrow.

The effect of acquisition time on the flow measurements was evaluated by comparing the model estimates that were obtained with 1, 1.5, 2, and 2.5 min of data. Compared with the average flows determined from 1.0 min of data, flows estimated from 1.5-, 2.0-, and 2.5-min acquisitions were 98.7 ± 5.4, 97.0 ± 5.5, and 96.3 ± 5.8% of this value, respectively. These data indicate that blood flow estimations from least squares fitting have a minor dependency on sampling time, with a difference of <10% in most cases. Although average estimated flow was slightly lower when data acquired over longer collection periods were analyzed, fitting errors were also lower. These effects were slightly greater at higher flows and might explain the bias illustrated in Fig. 6.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Most measurements of muscle blood flow in intact animals and conscious humans have been performed by plethysmography (21) or dilution techniques (26, 37). However, these techniques only provide information about whole limb flow, which includes contributions from bone, skin, and fat. Although muscle blood flow can be directly quantitated by 133Xe clearance (5, 8, 9, 11, 16) and hydrogen electrode techniques (31, 32), both procedures have significant technical limitations. Because of the long half-life of the tracer, 133Xe clearance is not applicable for repeated measurements after physiological interventions. In the hydrogen electrode procedure, the requirement of placing electrodes in each muscle group to be evaluated limits sampling, particularly in human studies. PET has been used extensively for measuring regional cerebral blood flow (rCBF). The most commonly used procedures for these measurements are bolus injection of H215O and continuous inhalation of C15O2 (which is converted to H215O in the lungs). Both techniques have been carefully studied and validated, and they are now routine procedures at most PET facilities (2, 7, 15, 19, 20, 24, 38, 47).

Application of these techniques for measuring regional blood flow to skeletal muscle is simpler and potentially more accurate compared with measurements of rCBF. Because skeletal muscle is macroscopically a homogeneous tissue compared with brain, mixing of different tissues, such as gray and white matter, is less of a problem. The resolution of present PET cameras allows partial volume effects to be ignored for muscle measurements. Furthermore, although effects of vascular volume can introduce errors in rCBF (25), this is less important for muscle because vascular space accounts for only ~1.5% of total muscle, compared with ~4.5% for brain (39, 45, 52).

The values for basal blood flow to skeletal muscle reported here (3.83 ± 0.36 ml · min-1 · 100 g-1) are in excellent agreement with previously reported results [3.12 ± 1.55 ml · min-1 · 100 g-1 (43), 3.7 ± 0.6 ml · min-1 · 100 g-1 (36)]. Also, the present technique yields flow values comparable to those based on PET studies by using continuous inhalation of C15O2 (44).

The agreement with microsphere measurements made at essentially the same time as the PET measurements was excellent. As noted above, standard regression methods assume that the reference measurement has negligible error, compared with the test observations. At basal flow, paired t-tests support the hypothesis that PET and microsphere values are indistinguishable. When standard linear regression is performed over a much larger range, hypothesis testing indicates small but statistically significant bias. Whether this is really the case is debatable. In any event, these systematic errors are unlikely to be of importance in physiological or clinical investigation of human subjects, the area in which PET is used.

Although the PET and microsphere techniques yielded similar blood flows, the partition coefficient for water in muscle determined by PET was considerably lower than the value assumed in continuous inhalation studies or measured by equilibration of tritiated water in tissue and arterial blood. This discrepancy is not surprising and is likely due to limitations of the kinetic model, which must estimate partition coefficients by extrapolating the predicted H215O PET curves to infinite time. Other sources of bias were noted above. For example, data acquired 1, 1.5, 2, and 2.5 min after injection yielded similar values for blood flow, within 10% in most cases. However, the estimated flows were generally lower when more data were used for fitting, and this is consistent with previous findings for cerebral blood flow. Although use of less data appears to yield better estimates of flow (less bias), the results are associated with larger estimation errors. From these considerations, we propose that a 2.5-min imaging time is a good compromise for measuring muscle blood flow. Although the bias associated with this measurement time is not critical for most clinical applications, when extremely accurate measurements of blood flow are required for research studies, corrections that used calibration curves such as that illustrated in Fig. 6 can be applied to the PET estimates.

Compared with conventional methods for measuring skeletal muscle blood flow, PET with H215O has several distinct advantages: 1) the method is relatively noninvasive and requires only imaging and arterial blood sampling; 2) cannulation of specific vessels is not required and flow to any muscle region can be measured; 3) because of the short physical half-life of 15O, repetitive measurements are possible; and 4) whole body imaging is possible. Recently developed PET cameras have much larger fields of view than the PC-4096, namely, 16-25 cm. With these devices, most of the musculature can be imaged with five to six sequential injections of H215O. Furthermore, when these devices are operated in three-dimensional (3D) mode, sensitivity is increased five- to sixfold, and the dose of tracer can be reduced proportionately. As illustrated by the blood flow maps in Fig. 7, these types of studies can yield detailed quantitative blood flow maps that are not available by any other technique.

Although our results validate the utility of PET with H215O for measuring blood flow to skeletal muscle, the study had several limitations. Because skeletal muscle varies greatly in fiber type and previous investigations have demonstrated that blood flow is not uniform within and between muscle of various types (3, 4, 33), not making muscle- or muscle group-specific measurements was a limitation of the study. However, by acquiring a contrast computerized tomography (CT) scan of the region imaged by PET, these measurements could have been performed. By coregistering the PET and CT data sets (1), ROIs confined to anatomically specific types of muscle can be outlined on the CT data and transferred to the PET images to calculate these flows. Application of this technique is somewhat limited by the relatively low resolution and lack of 3D imaging capabilities of the PC 4096 PET camera that was used in our studies. However, most newer PET imaging devices have superior resolution (~4.5 mm FWHM) and sensitivity when operated in 3D mode. Also, PET cameras with even higher resolution (~2.0 mm FWHM) will soon be introduced into clinical practice. In addition, hybrid PET-CT imaging systems have recently become available. Precise measurements of blood flow to specific types of muscle will be greatly facilitated with these instruments because PET and CT data sets can be acquired without moving the subject. When these techniques are applied to humans, the larger size of the specific muscles to be studied will greatly enhance the quality of the data. Another limitation of H215O PET is the requirement of an arterial line for acquiring the input function for kinetic modeling. This problem can also be resolved by the use of high-resolution PET-CT devices. With these instruments, CT images can be used to localize large arteries and construct ROIs, which can be transferred to PET data sets for defining the arterial input function. Because large blood pool structures can be included in the imaging field, this approach will be particularly useful for measuring blood flow to upper extremity and paraspinal musculature. Overall, these limitations are related to specific details of our study and not to the general approach of using PET with H215O for measuring blood flow to skeletal muscle.

Conclusions. Regional muscle blood flow determined by PET with H215O was in excellent agreement with microsphere measurements and consistent with previously reported results with other techniques. Because of the precision and accuracy of this technique, it should be of considerable value in future applications for human metabolic and clinical studies. The short physical half-life of 15O allows extension of the technique to pre- and postintervention studies and/or time course studies in which frequent, intermittent measurements are required.


    ACKNOWLEDGEMENTS

The authors acknowledge the excellent technical support from Stephen Weise and Avis Loring, as well as the efforts of the MGH Cyclotron Laboratory staff.


    FOOTNOTES

This work was supported in part by grants from the Shriners Hospitals for Children and National Institute of General Medical Sciences Grants P50-GM-21700, T32-GM-07035, and T32-CA-09362.

Address for reprint requests and other correspondence: A. J. Fischman, Division of Nuclear Medicine, Massachusetts General Hospital, Boston, MA 02114 (E-mail: Fischman{at}pet.mgh.harvard.edu).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

10.1152/japplphysiol.00445.2001

Received 10 May 2001; accepted in final form 11 December 2001.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Alpert, NM, Berdichevsky D, Levin Z, Morris ED, and Fischman AJ. Improved methods for image registration. J Neuroimaging 3: 10-19, 1996.

2.   Alpert, NM, Eriksson L, Chang JY, Bergstrom M, Litton JE, Correia JA, Bohm C, Ackerman RH, and Taveras JM. Strategy for the measurement of regional cerebral blood flow using short-lived tracers and emission tomography. J Cereb Blood Flow Metab 4: 28-34, 1984[ISI][Medline].

3.   Armstrong, RB. Distribution of blood flow in the muscles of conscious animals during exercise. Am J Cardiol 62: 9-14, 1988.

4.   Armstrong, RB, and Laughlin MH. Blood flows within and among rat muscles as a function of time during high speed treadmill exercise. J Physiol (Lond) 344: 189-208, 1983[Abstract/Free Full Text].

5.   Bell, G, and Short DW. The measurement of blood flow through muscle from the clearance of radioactive xenon. Surg Gynecol Obstet 127: 61-65, 1968[ISI][Medline].

6.   Biolo, G, Gastaldelli A, Zhang XJ, and Wolfe RR. Protein synthesis and breakdown in skin and muscle: a leg model of amino acid kinetics. Am J Physiol Endocrinol Metab 267: E467-E474, 1994[Abstract/Free Full Text].

7.   Carson, RE, Huang SC, and Phelps ME. Optimization of the O-15 water bolus injection techniques for quantitative local cerebral blood flow measurements using positron emission tomography. J Cereb Blood Flow Metab 3, Suppl1: S11-S12, 1983.

8.   Christensen, NJ. The significance of work load and injected volume in xenon 133 measurement of muscular blood flow. Acta Med Scand 183: 445-447, 1968[ISI][Medline].

9.   Chung, SY, Kim I, Roy UY, Maskin C, and Pinsky S. Blood-flow measurement in muscle with Xe-133. Radiology 165: 571-572, 1987[Abstract/Free Full Text].

10.   Clark, MG, Colquhoun EQ, Rattigan S, Dora KA, Eldershaw TP, Hall JL, and Ye J. Vascular and endocrine control of muscle metabolism. Am J Physiol Endocrinol Metab 268: E797-E812, 1995[Abstract/Free Full Text].

11.   Corman, LA, Flickinger FW, Sokoloff J, and Nodine JH. Radioactive xenon tissue clearance: standardization for measurement of peripheral blood flow. J Nucl Med 11: 233-238, 1970[Free Full Text].

12.   Elia, M. General integration and regulation of metabolism at the organ level. Proc Nutr Soc 54: 213-232, 1995[ISI][Medline].

13.   Hales, JR. Radioactive microsphere measurement of cardiac output and regional tissue blood flow in the sheep. Pflügers Arch 344: 119-132, 1973[ISI][Medline].

14.   Harringer, W, Fernandez-Del Castillo C, Rattner DW, Guerrero JL, Warshaw AL, and Vlahakes GL. Evaluation and validation of microsphere technique for determination of pancreatic blood flow. Am J Physiol Gastrointest Liver Physiol 265: G587-G594, 1993[Abstract/Free Full Text].

15.   Herscovitch, P, Markham J, and Raichle ME. Brain blood flow measured with intravenous H215O. I. Theory and error analysis. J Nucl Med 24: 782-789, 1983[Abstract/Free Full Text].

16.   Hinsenkamp, M, d'Hollander A, Coussaert E, Rasquin C, and Schoutens A. Measurement of blood flow in peripheral muscles using Xe-133. Angiology 31: 58-65, 1980.

17.   Hlastala, MP, Malvin GM, Quartararo C, and Gronlund L. Inert gas washout measurement of muscle blood flow distribution-roles of hypoxia and diffusion limitation. Adv Exp Med Biol 317: 745-750, 1992[Medline].

18.   Hsu, H, Yu YM, Babich JW, Burke JF, Livni E, Tompkins RG, Young VR, Alpert NM, and Fischman AJ. Measurement of muscle protein synthesis by positron emission tomography with L-[methyl-11C]methionine. Proc Natl Acad Sci USA 93: 1841-1846, 1996[Abstract/Free Full Text].

19.   Huang, SC, Carson RE, Hoffman EJ, Carson J, MacDonald N, Barrio JR, and Phelps ME. Quantitative measurement of local cerebral blood flow in humans by positron computed tomography and O-15 water. J Cereb Blood Flow Metab 3: 141-153, 1983[ISI][Medline].

20.   Huang, SC, Carson RE, and Phelps ME. Measurement of local blood flow and distribution volume with short-lived isotopes: a general input technique. J Cereb Blood Flow Metab 2: 99-108, 1982[ISI][Medline].

21.   Jackson, RA, Peters N, Advani U, Perry G, Rogers JJ, Brough WH, and Pilkington PR. Forearm glucose uptake during the oral glucose tolerance test in normal subjects. Diabetes 22: 442-458, 1973[ISI][Medline].

22.   Kety, SS. Theory of blood-tissue exchange and its application to measurement of blood flow. Methods Med Res 8: 223-227, 1960.

23.   Kety, SS, and Schmidt CF. The nitrous oxide method for the quantitative determination of cerebral blood flow in man. Theory, procedure and normal values. J Clin Invest 27: 476-483, 1948.

24.   Koeppe, RA, Holden JE, and Ip WR. Performance comparison of parameter estimation techniques for the quantitation of local cerebral blood flow by dynamic positron computed tomography. J Cereb Blood Flow Metab 5: 224-234, 1985[ISI][Medline].

25.   Koeppe, RA, Hutchins GD, Rothley JM, and Hichwa RD. Examination of assumptions for local cerebral blood flow studies in PET. J Nucl Med 28: 1695-1703, 1987[Abstract/Free Full Text].

26.   Laakso, M, Edelman SV, Brechtel G, and Baron AD. Decreased effect of insulin to stimulate skeletal muscle blood flow in obese man. A novel mechanism for insulin resistance. J Clin Invest 85: 1844-1852, 1990.

27.   Lammertsma, AA, Martin AJ, Friston KJ, and Jones T. In vivo measurement of the volume of distribution of water in cerebral grey matter: effects on the calculation of regional cerebral blood flow. J Cereb Blood Flow Metab 12: 291-295, 1992[ISI][Medline].

28.   Lassen, NA, and Larsen OA. Measurement of blood flow with freely diffusible indicators as inert gases, antipyrine, labelled water and rubidium. Acta Endocrinol 158, Suppl: 95-111, 1972.

29.   Lochs, H, Williams PE, Morse EL, Abumrad NN, and Adibi SA. Metabolism of dipeptides and their constituent amino acids by liver, gut, kidney, and muscle. Am J Physiol Endocrinol Metab 254: E588-E594, 1988[Abstract/Free Full Text].

30.   Maki, Y, Breidenbach WC, and Firrell JC. Evaluation of a local microsphere injection method for measurement of blood flow in the rabbit lower extremity. J Orthop Res 11: 20-27, 1993[ISI][Medline].

31.   Misra, SK. Measurement of local skeletal muscle blood flow in animals by the hydrogen electrode technique. Muscle Nerve 3: 285-288, 1980[ISI][Medline].

32.   Misra, SK. Measurement of local skeletal muscle blood flow in normal humans by hydrogen clearance. Muscle Nerve 3: 289-292, 1980[ISI][Medline].

33.   Musch, TI, Friedman DB, Pitetti KH, Haidet GC, Stray-Gundersen J, Mitchell JH, and Ordway GA. Regional distribution of blood flow of dogs during graded dynamic exercise. J Appl Physiol 63: 2269-2277, 1987[Abstract/Free Full Text].

34.   Padberg, FT, Franco CD, Kerr JC, Lynch TG, Burns WF, and Hobson RW. Acute ischemia-reperfusion injury in the canine hindlimb. J Card Surg 30: 925-931, 1989.

35.   Palmer, EJ, Scott JA, and Strauss HW. Practical Nuclear Medicine, Instrumentation and Radiopharmaceuticals. Philadelphia, PA: Saunders, 1992, chapt. 2, p. 27-69.

36.   Peltoniemi, P, Yki-Jarvinen H, Oikonen V, Oksanen A, Takala TO, Ronnemaa T, Erkinjuntti M, Knuuti MJ, and Nuutila P. Resistance to exercise-induced increase in glucose uptake during hyperinsulinemia in insulin-resistant skeletal muscle of patients with type 1 diabetes. Diabetes 50: 1371-1377, 2001[Abstract/Free Full Text].

37.   Rabinowitz, D, and Zierler KL. Forearm metabolism in obesity and its response to intra-arterial insulin. Characterization of insulin resistance and evidence for adaptive hyperinsulinism. J Clin Invest 41: 2173-2181, 1962.

38.   Raichle, ME, Matin WRW, Herscovitch P, Mintun MA, and Markham J. Brain blood flow measured with intravenous H215O. II. Implementation and validation. J Nucl Med 24: 790-798, 1983[Abstract/Free Full Text].

39.   Raitakari, M, Knuuti MJ, Ruotsalainen U, Laine H, Makea P, Teras M, Sipila H, Niskanen T, Raitakari OT, Iida H, Harkonen R, Wegrlius U, Yki-Jarvinen H, and Nuutila P. Insulin increases blood volume in human skeletal muscle: studies using [15O]CO and positron emission tomography. Am J Physiol Endocrinol Metab 269: E1000-E1005, 1995[Abstract/Free Full Text].

40.   Rennie, MJ. Muscle protein turnover and the wasting due to injury and disease. Br Med Bull 41: 257-264, 1985[Abstract/Free Full Text].

41.   Roto Kops, E, Herzog H, Schmid A, Holte S, and Feinendegen LE. Performance characteristics of an eight-ring whole body PET scanner. J Comput Assist Tomogr 14: 437-445, 1990[ISI][Medline].

42.   Rudolph, AM, and Heymann MA. Measurement of flow in perfused organs, using microsphere techniques. Acta Endocrinol 158, Suppl: 112-127, 1972.

43.   Ruotsalainen, U, Raitakari MM, Nuutila P, Oikonen V, Sipila H, Teras M, Knuuti MJ, Bloomfield PM, and Iida H. Quantitative blood flow measurement of skeletal muscle using oxygen-15-water and PET. J Nucl Med 38: 314-319, 1997[Abstract/Free Full Text].

44.   Senda, M, Fischman AJ, Weise S, Alpert NM, Correia JA, Rubin RH, and Strauss HW. Regional perfusion, oxygen metabolism, blood volume and immunoglobulin G accumulation at focal sites of infection in rabbits. Eur J Nucl Med 19: 166-172, 1992[ISI][Medline].

45.   Shadle, OW, Zukof M, and Diana J. Translocation of blood from isolated dog's hindlimb during levarterenol infusion and sciatic nerve stimulation. Circ Res 6: 326-333, 1958[Abstract/Free Full Text].

46.   Snyder, WS, Cook MJ, Nasset ES, Karhausen LR, Howells GP, and Tipton IH. Report on the Task Group on Reference Man. New York: Pergamon, 1975, p. 284-285.

47.   Subramanyam, R, Alpert NM, Hoop B, Brownell GL, and Taveras JM. A model for regional oxygen distribution during continuous inhalation of 15O2, C15O and C15O2. J Nucl Med 19: 48-53, 1978[Abstract/Free Full Text].

48.   Triffitt, PD, and Gregg PJ. Measurement of blood flow to the tibial diaphysis using 11C radioactive microspheres. A comparative study in the adult rabbit. J Orthop Res 8: 642-645, 1990[ISI][Medline].

49.   Waterlow, JC. Metabolic adaptation to low intakes of energy and protein. Annu Rev Nutr 6: 495-526, 1986[ISI][Medline].

50.   Young, VR. The role of skeletal and cardiac muscle in the regulation of protein metabolism. In: Mammalian Protein Metabolism, edited by Munro HN.. New York: Academic, 1970, vol. IV, p. 585-674.

51.   Young, VR, and Sanchez M. Albumin, skeletal muscle, lean body mass as functional predictors in the elderly. In: Nutritional Assessment of Elderly Populations, edited by Rosenberg JH.. New York: Raven, 1995, p. 63-73.

52.   Zilkha, E, Ladurner G, Iliff LD, Du Boulay GH, and Marshall J. Computer subtraction in regional cerebral blood-volume measurements using the EMI-scanner. Br J Radiol 49: 330-334, 1976[Abstract].

J APPL PHYSIOL 92(4):1709-1716
8750-7587/02 $5.00 Copyright © 2002 the American Physiological Society


This article has been cited by other articles:


Home page
 The OncologistHome page
A. J. de Langen, V. E. M. van den Boogaart, J. T. Marcus, and M. Lubberink
Use of H215O-PET and DCE-MRI to Measure Tumor Blood Flow
Oncologist, June 1, 2008; 13(6): 631 - 644.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
I. Heinonen, S. V. Nesterov, J. Kemppainen, P. Nuutila, J. Knuuti, R. Laitio, M. Kjaer, R. Boushel, and K. K. Kalliokoski
Role of adenosine in regulating the heterogeneity of skeletal muscle blood flow during exercise in humans
J Appl Physiol, December 1, 2007; 103(6): 2042 - 2048.
[Abstract] [Full Text] [PDF]

Home page
 J Ultrasound MedHome page
M.-A. Weber, H. Krakowski-Roosen, S. Delorme, H. Renk, M. Krix, J. Millies, R. Kinscherf, A. Kunkele, H.-U. Kauczor, and W. Hildebrandt
Relationship of skeletal muscle perfusion measured by contrast-enhanced ultrasonography to histologic microvascular density.
J. Ultrasound Med., May 1, 2006; 25(5): 583 - 591.
[Abstract] [Full Text] [PDF]

Home page
 J Ultrasound MedHome page
M. Krix, M.-A. Weber, H. Krakowski-Roosen, H. B. Huttner, S. Delorme, H.-U. Kauczor, and W. Hildebrandt
Assessment of Skeletal Muscle Perfusion Using Contrast-Enhanced Ultrasonography
J. Ultrasound Med., April 1, 2005; 24(4): 431 - 441.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
M. S. Laaksonen, K. K. Kalliokoski, H. Kyrolainen, J. Kemppainen, M. Teras, H. Sipila, P. Nuutila, and J. Knuuti
Skeletal muscle blood flow and flow heterogeneity during dynamic and isometric exercise in humans
Am J Physiol Heart Circ Physiol, March 1, 2003; 284(3): H979 - H986.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fischman, A. J.
Right arrow Articles by Alpert, N. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fischman, A. J.
Right arrow Articles by Alpert, N. M.