| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the Departments of Physiology and Biophysics (S.M., E.L.R.), the Department of Internal Medicine, Divisions of Cardiovascular Diseases (T.R.B., A.L.) and Hypertension (L.O.L.), the Section of Biostatistics (A.L.W., V.S.P.), and the Department of Diagnostic Radiology (P.F.S.), Mayo Clinic and Foundation, 200 First St SW, Alfred 2-409, Rochester, MN 55905. Received May 25, 2000; revision requested July 12; final revision received March 5, 2001; accepted March 23. Supported in part by National Institutes of Health research grant HL-43025. Address correspondence to E.L.R. (e-mail: elran@mayo.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Electron-beam CT–based intramyocardial blood volume and perfusion and Doppler ultrasonography (US)–based intracoronary blood flow were obtained in 13 pigs at baseline and again 3 months later. Measurements were obtained at rest and after the administration of adenosine. The short-term variation during 30 minutes of electron-beam CT measurements was assessed in nine additional pigs.
RESULTS: Short-term variation of blood volume and perfusion averaged 8% and 9%, respectively, and was similar for both weight groups at rest and after adenosine administration. At rest, intracoronary blood flow, blood volume, and perfusion remained unchanged from baseline to follow-up. Long-term increases (percentage change with adenosine relative to that at rest) in blood volume and perfusion reserves were consistent with increasing intracoronary blood flow reserves. Despite these long-term changes in intracoronary blood flow, blood volume, and perfusion, the blood volume–to-perfusion relationship suggests a similar blood volume distribution among different microvascular functional components in normal porcine myocardium at both weight groups.
CONCLUSION: Electron-beam CT may be of value for quantifying long-term changes in intramyocardial microvascular function.
Index terms: Blood, volume • Computed tomography (CT), electron beam, 51.12119 • Coronary vessels, CT, 54.12119 • Coronary vessels, flow dynamics, 54.12119 • Myocardium, 511.12119
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
We have previously demonstrated that minimally invasive computed tomography (CT) can be used to obtain quantitative indexes of intramyocardial microcirculatory function (5,6). In particular, by virtue of its ability to quantify regional intramyocardial microvascular blood volume and perfusion, electron-beam CT represents a promising tool for evaluating microvascular function in healthy and diseased myocardium (7). Its sensitivity in the quantification of long-term changes in intramyocardial blood volume and perfusion, however, is yet to be shown.
The aim of this study was to examine the ability of electron-beam CT to help detect and quantify subtle long-term changes in microvascular function that occur in the normal porcine heart and compare it with intracoronary Doppler ultrasonography (US).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
All animals were intubated and ventilated with room air, given oxygen when necessary, and placed supine in a molded cast for the duration of both the electron-beam CT and Doppler US studies. The external and internal jugular veins were exposed by means of cutdown and cannulated with a 7-F sheath. A bipolar pacing catheter was positioned with its tip in the coronary sinus to permit atrial pacing when necessary and to monitor coronary sinus pressures. A pigtail catheter was placed in the superior vena cava for administering contrast material. The external carotid artery was cannulated with an 8-F sheath. After intravenous injection of 10,000 U of heparin (followed by intravenous infusion of 1,000 U of heparin per hour), an 8-F Judkins left coronary guiding catheter was placed in the left main coronary artery for selective intracoronary injections of contrast material and for monitoring proximal coronary artery pressure. A 2.2-F dual-lumen infusion catheter (Ultrafuse; Boston Scientific Scimed, Maple Grove, Min) was advanced into the proximal left anterior descending artery (LAD) for selective infusion of drugs and performance of intracoronary Doppler US studies.
Electron-Beam CT
The technical features of the electron-beam CT unit (C-150; Imatron, South San Francisco, Calif) that are relevant to myocardial imaging have been described in detail elsewhere (8). Complete electron-beam CT and Doppler US studies were performed on the same day in alternating order for sequential pigs. All animals were positioned in the electron-beam CT scanner so that the heart was centered in the imaging field. The diameter of the reconstructed image field of view was 21–30 cm, depending on the size of the animal (pixel size, 0.34–0.69 mm2; section thickness, 7 mm; voxel volume, 2.38–4.83 mm3). The scanning table was rotated and tilted to obtain short-axis tomographic images at midventricular cross sections of the left ventricular chamber and the surrounding myocardium; papillary muscles were used as landmarks. All images were obtained in late diastole (at 80% of the RR interval) with an image acquisition time of 50 msec.
Initially, nonionic contrast material (iopamidol [Isovue; Bracco Diagnostics, Princeton, NJ]) was injected selectively into the LAD (5 mL over 1.3 seconds) through the coronary infusion catheter to highlight the region of myocardium that is perfused by the LAD downstream to the catheter tip. Then, two CT flow studies were performed at 20-minute intervals, each after a rapid injection of a bolus of the same contrast material (0.33 mL/kg over 2 seconds) into the superior vena cava. Each sequence included 40 scans obtained every one to two heart cycles over 35–50 seconds, depending on the heart rate. Hemodynamic data were recorded just before each electron-beam CT study. The first image in each sequence was obtained before the injection of contrast material to provide background myocardial image intensity. The first scanning sequence was performed during resting conditions after selective intracoronary infusion of normal saline for 5 minutes at a rate of 1 mL/min. The second scanning sequence was performed after selective intracoronary infusion of adenosine (Sigma Chemical, St Louis, Mo; 100 µg/kg/min) for 5 minutes. Each study was followed by a 20-minute recovery period with selective intracoronary infusion of saline (1 mL/min) to allow for washout of the contrast material from the myocardium and the systemic circulation.
Intracoronary Doppler US and Coronary Angiography
A 0.014-inch Doppler guide wire (Cardiometrics FloWire; EndoSonics, Rancho Cordova, Calif) was advanced through the infusion catheter and positioned in a long straight proximal segment of the LAD. All drugs were subsequently infused selectively into the LAD at a rate of 1 mL/min through the coronary infusion catheter. As with the electron-beam CT study, 100 µg/kg/min of adenosine was infused continuously for 5 minutes before Doppler data were recorded.
Hemodynamic measurements were recorded before and after the infusion of drugs. The average peak velocity (APV, in centimeters per second) was obtained from an online analysis of Doppler parameters of intracoronary flow velocities and recorded just before selective angiography was performed. Selective coronary angiography was performed for quantitative analysis of end-diastolic videoangiographic frames to measure the luminal diameter of the coronary artery. Coronary artery diameters were measured by one author (S.M.) offline 3–5 mm distal to the tip of the Doppler wire by using a previously described quantification software program (9). Intracoronary blood flow (in milliliters per minute) was calculated from Doppler US–derived time-velocity integrals (in centimeters per second) and vessel diameters (in centimeters) as previously described (10): intracoronary blood flow = 0.5 · 
· APV · 60 · (0.5 · coronary artery diameter)2.
Image Data Display and Analysis
Electron-beam CT scans were reconstructed by using the vendor-supplied algorithm. The resulting tomographic images were transferred to a UNIX-based workstation (Sun Microsystems, Palo Alto, Calif). Images were evaluated by using an image analysis software package (ANALYZE version 7.5; Biomedical Imaging Resource, Mayo Foundation, Rochester, Minn). On tomographic images showing peak attenuation of the left ventricular (LV) chamber, one author (S.M.) manually outlined one region of interest in the anterior cardiac wall and a second region of interest in the LV chamber. Because of the increase in heart size with increasing body weight, regions of interest in the anterior cardiac wall were slightly smaller at baseline compared with those at follow-up (306 mm2 ± 65 vs 399 mm2 ± 89, respectively). For the region of interest that outlines the LAD perfusion territory, we used the image sequence in which contrast material was selectively injected into the LAD. This region of interest was applied to all subsequent studies in which contrast material was injected into the superior vena cava. Average pixel intensities were calculated for the region of interest in the LV chamber and in the LAD-perfused anterior cardiac wall on every image of the sequence. These time-intensity data were plotted to generate indicator dilution curves (Fig 1), which are needed to obtain indexes of myocardial perfusion (in milliliters per gram per minute) and myocardial blood volume (in milliliters per gram), as previously described (7).
|
CSA2. Because the length of these vessels does not change much relative to the change in cross-sectional area, the intraluminal volume of these nonrecruitable vessels (BVnr) should be proportional to the cross-sectional area. Hence, it should follow that F BV2nr or BVnr
. The other microvascular component consists of functionally recruitable vessels (presumably mostly capillaries, small arterioles, and venules), which are assumed not to change in diameter so much as to the change in the number of vessels perfused. The vascular resistance in these vessels is thus inversely related to the number of recruited vessels. Because the intraluminal blood volume of the recruitable vessels (BVr) should be proportional to the number of recruited vessels, it should follow that F BVr. Because the recruitable vessels and nonrecruitable vessels are in series, the following relationship should hold: blood volume = BVr + BVnr = aF + b
, where a and b are coefficients to be established experimentally.
Statistical Analysis
Values are reported as means ± SD. A logarithmic transformation was applied to some of the outcome measurements (APV, intracoronary blood flow, perfusion, and blood volume in the recruitable component) to stabilize the variance before the analysis. To evaluate the effect of body weight level (baseline vs follow-up) and adenosine use (resting vs adenosine) on the outcome measurements, three separate two-factor repeated measures multivariate analysis of variance (ANOVA) models were fit. Each multivariate ANOVA model enabled the examination of a separate set of measurements as outlined within the Results section. In addition, a three-factor multivariate ANOVA model was fit to evaluate the hemodynamic parameters by considering type of modality (electron-beam CT vs Doppler US) as an additional factor. In each multivariate ANOVA model, the type I error level (
) for assessing the results of the F tests was set at 0.0125 after applying a Bonferroni correction to adjust for the four separate analyses. In the presence of a statistically significant result based on a multivariate ANOVA test for a particular factor (eg, weight level, adenosine use, or their interaction), we then proceeded to evaluate the univariate F tests from a two-factor repeated measures ANOVA model for each separate outcome measurement at a type I error level of 0.05 with assurance that the experiment-wise error rate will be close to 5% (11). If the results of the univariate F test for the interaction effect were statistically significant at the .05 level, individual comparisons across the four levels of the two factors were then evaluated by using the Fisher protected least significant difference test. The protected least significant difference test uses the overall sample variance estimated from an ANOVA model and controls the overall experiment-wise error rate at 5% because it examines only pair-wise differences for factors with statistically significant F test results.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
= .0125). Adenosine induced a statistically significant increase in heart rate (P = .012) and a statistically significant decrease in mean and systolic blood pressure (P < .001) compared with that at rest (Table 1). Neither adenosine nor the increased weight had an effect on the rate-pressure product. The response of the hemodynamic parameters to adenosine relative to rest was similar at baseline and follow-up. Hemodynamic parameters were also similar in the Doppler and electron-beam CT studies at both baseline and follow-up (Table 1).
|
|
|
|
|
|
was fitted separately to all baseline and follow-up data. On the basis of this analysis, we found that blood volume and perfusion values obtained at rest and after adenosine administration followed the same curvilinear line at both baseline and follow-up despite differences in maximal blood volume and perfusion at the different body weights (Fig 4).
|
Short-term Variation of Blood Volume and Perfusion
Multiple blood volume and perfusion estimates were obtained from sequential scans obtained in a subset of animals in the low-weight group (n = 4) and the high-weight group (n = 5) to assess the within-animal variation. Blood volume and perfusion estimates obtained from sequential scans varied little during a period of up to 2 hours (Fig 5). Coefficients of variation were similar at rest and with adenosine in both weight groups, with an average coefficient of variation of about 8% for blood volume and 9% for perfusion (Table 3). Variation of measurements among different animals (between-animals evaluation) was generally lower in the high-weight group (Table 3).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Long-term Changes at Rest
During the follow-up period, blood pressures at rest tended to increase and heart rates tended to decrease; this is consistent with expected body weight–related changes in porcine cardiovascular hemodynamics (12). Epicardial coronary artery diameters increased with a concomitant trend toward increased intracoronary blood flow, whereas the mean APV remained unchanged. These findings are consistent with previous observations that the cross-sectional area of an epicardial coronary artery lumen is linearly related to the volume of myocardium it perfuses (13–15). Thus, with more myocardial tissue to be perfused after 3 months, intracoronary blood flow through epicardial coronary vessels is expected to increase, which is also consistent with observations in children, where coronary sinus blood flow has been shown to increase linearly with LV mass (16). In contrast to epicardial flow parameters, electron-beam CT–based intramyocardial blood volume and perfusion, because they are related to myocardial mass, did not change during 3 months during resting conditions, which is consistent with the finding of fairly constant capillary and arteriolar densities throughout growth (1,17–19).
Microvascular Responses to Adenosine and Long-term Changes in Microvascular Functional Reserve
As expected, adenosine induced a significant increase in Doppler US–based intracoronary blood flow and electron-beam CT–based intramyocardial blood volume and perfusion. Our data suggest that these changes are attributable to both the recruitable and nonrecruitable components. This is consistent with only about half of capillaries and arterioles being perfused during resting conditions (20), although adenosine has been shown to induce recruitment of both preterminal arterioles and capillaries in the rabbit and canine heart (21,22) and vasodilation of conducting and resistance vessels (23). Both Doppler US–based intracoronary blood flow and electron-beam CT–based intramyocardial blood volume and perfusion quantification revealed a significantly higher microvascular functional reserve in the porcine heart after 3 months compared with that at baseline. The increased microvascular functional reserve is consistent with findings by Hamaoka et al (24), who reported a significantly higher intracoronary Doppler US–based coronary flow reserve in older children compared with younger children. Buss et al (25) also observed similar changes in the rabbit heart, which, in their study, were attributable to an immature coronary responsiveness to exogenous adenosine in younger animals. The decrease in the variation of measurements during the follow-up period observed in our study may also be related to a change toward a more mature and more uniform microvascular responsiveness to exogenous adenosine among different animals. Despite the changes in microvascular functional reserve, the electron-beam CT–based blood volume–to-perfusion relationship indicates a consistent curvilinear pattern in the healthy porcine heart with a shift toward higher maximum blood volume and perfusion values after 3 months. Our data suggest that the relative distribution of recruitable (ie, primarily capillaries and preterminal arterioles) and nonrecruitable (primarily larger conducting microvessels) microvessels remains fairly constant in the normal heart. Furthermore, both the recruitable and nonrecruitable components contribute to the increased functional reserve of blood volume and perfusion after 3 months, which is suggestive of a link between preterminal arteriolar vasomotor function and capillary recruitment. The underlying mechanism that may be responsible for this regulation among these functional components, however, must still be identified.
Methodologic Considerations
Quantification of long-term changes in intramyocardial microvascular functional reserve by virtue of estimating electron-beam CT–based blood volume and perfusion was consistent with long-term changes in intracoronary Doppler US–based intracoronary blood flow. Although these indexes alone helped identify an abnormal microvascular functional reserve in low-weight compared with high-weight pigs, the combined evaluation of simultaneously measured blood volume and perfusion showed that all values followed one curvilinear line that describes the blood volume–to-perfusion relationship in our normal pigs. Deviations from this pattern may allow us to differentiate normal (but functionally immature) from truly diseased myocardium, which may improve the evaluation of microvascular abnormalities in children (16,24) and during senescence (26).
We also applied a previously developed model that relates intramyocardial blood volume to perfusion to help differentiate the functional behavior of recruitable and nonrecruitable microvessels. Intramyocardial microvessels, however, are not visually resolvable with electron-beam CT. Other investigators have used more invasive techniques to directly visualize microvessels on myocardial surfaces in vivo (27,28), and these techniques could potentially be used as independent reference techniques to quantify intramyocardial blood volume distribution in response to vasoactive drugs in normal (or diseased) myocardium. Nevertheless, electron-beam CT–based quantification of blood volume distribution in the two components were similar to results obtained with other methods (29), and the functional behavioral trend of the components, as inferred from the relationship between myocardial blood volume and perfusion, is consistent with previously described effects of adenosine on coronary microvessels (23,30).
In conclusion, long-term changes in microvascular functional reserve that occur in normal porcine hearts during a period of 3 months can be quantified with electron-beam CT. Compared with these changes, the short-term variation of blood volume and perfusion measurements is relatively small, which is important for differentiating true long-term changes from the random variation of measurements. This technique may thus provide an opportunity to identify physiologically or clinically meaningful alterations in microvascular functional reserve either during a disease process or in response to therapy longitudinally in the same subject. The relationship of simultaneously measured blood volume and perfusion follows a consistent curvilinear pattern in the normal porcine heart despite body weight- or maturation-related changes in microvascular functional reserve. Deviations from this pattern may potentially be of value for differentiating normal from pathologic intramyocardial microvascular function.
Practical application: In this study, we demonstrated the feasibility of using electron-beam CT–based indicator dilution techniques to longitudinally quantify long-term changes in microvascular functional reserve in the normal porcine heart. Evaluation of long-term microvascular functional changes may be extended to quantify the effect of physiologic stimuli such as aging and exercise on normal coronary microvascular functional reserve (17,31). The approach may be of value in the assessment of long-term microvascular alterations in abnormalities that affect intramyocardial microvessels (eg, diabetes mellitus, arterial hypertension, hypercholesterolemia, or cardiomyopathies) (2). It may further be used to monitor the efficacy of therapeutic strategies that target the coronary microcirculation such as transmyocardial laser revascularization (3) or gene therapy to induce myocardial angiogenesis in ischemic myocardium (4). Simultaneous quantification of intramyocardial blood volume and perfusion may be applicable in humans, too, by using either electron-beam CT or other minimally invasive techniques that provide simultaneous estimates of blood volume and perfusion (32,33). Finally, our findings are consistent with previously reported changes in microvascular functional reserve that occur with maturation of the heart (24,25). The results of this study thus support the concept (24,26,34) that assessment of coronary functional reserve may require adjustments for age and body weight.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, E.L.R.; study concepts and design, all authors; literature research, S.M., T.R.B.; experimental studies, S.M., L.O.L., K.A.S., P.E.L.; data acquisition, S.M., T.R.B.; data analysis/interpretation, all authors; statistical analysis, S.M., A.L.W., V.S.P.; manuscript preparation and definition of intellectual content, all authors; manuscript editing, D.C.D., J.M.P.; manuscript revision/review and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. T. George, C. Silva, M. A.S. Cordeiro, A. DiPaula, D. R. Thompson, W. F. McCarthy, T. Ichihara, J. A.C. Lima, and A. C. Lardo Multidetector Computed Tomography Myocardial Perfusion Imaging During Adenosine Stress J. Am. Coll. Cardiol., July 4, 2006; 48(1): 153 - 160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. J. Schoepf, C. R. Becker, B. M. Ohnesorge, and E. K. Yucel CT of Coronary Artery Disease Radiology, July 1, 2004; 232(1): 18 - 37. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X.-Y. Zhu, M. Rodriguez-Porcel, M. D. Bentley, A. R. Chade, V. Sica, C. Napoli, N. Caplice, E. L. Ritman, A. Lerman, and L. O. Lerman Antioxidant Intervention Attenuates Myocardial Neovascularization in Hypercholesterolemia Circulation, May 4, 2004; 109(17): 2109 - 2115. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rodriguez-Porcel, J. Herrman, A. R. Chade, J. D. Krier, J. F. Breen, A. Lerman, and L. O. Lerman Long-Term Antioxidant Intervention Improves Myocardial Microvascular Function in Experimental Hypertension Hypertension, February 1, 2004; 43(2): 493 - 498. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rodriguez-Porcel, A. Lerman, J. Herrmann, R. S. Schwartz, T. Sawamura, M. Condorelli, C. Napoli, and L. O. Lerman Hypertension exacerbates the effect of hypercholesterolemia on the myocardial microvasculature Cardiovasc Res, April 1, 2003; 58(1): 213 - 221. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Mohlenkamp, P. E Beighley, E. A Pfeifer, T. R Behrenbeck, P. F Sheedy II, and E. L Ritman Intramyocardial blood volume, perfusion and transit time in response to embolization of different sized microvessels Cardiovasc Res, March 1, 2003; 57(3): 843 - 852. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. MOHLENKAMP, L. O. LERMAN, Z. BAJZER, P. E. LUND, and E. L. RITMAN Quantification of Myocardial Microcirculatory Function with X-ray CT Ann. N.Y. Acad. Sci., October 1, 2002; 972(1): 307 - 316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. RULE, Z. BAJZER, E. L. RITMAN, and L. O. LERMAN Renal Handling of X-ray Contrast Media Imaging and Exploration with Electron Beam CT Ann. N.Y. Acad. Sci., October 1, 2002; 972(1): 317 - 324. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. O. Bonetti, S. H. Wilson, M. Rodriguez-Porcel, D. R. Holmes Jr, L. O. Lerman, and A. Lerman Simvastatin preserves myocardial perfusion and coronary microvascular permeability in experimental hypercholesterolemia independent of lipid lowering J. Am. Coll. Cardiol., August 7, 2002; 40(3): 546 - 554. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGR |
, adenosine = 
) and at follow-up (resting conditions = 
, adenosine = 
). Despite a different microvascular functional response to adenosine at baseline and follow-up, all values follow the same curvilinear line, which suggests that the relative distribution of recruitable and nonrecruitable vessels remained fairly constant in normal porcine hearts during the 3-month follow-up period (see text for details).