(Circulation. 2001;103:1999.)
© 2001 American Heart Association, Inc.
Basic Science Reports |
Index Matching to Improve Optical Coherence Tomography Imaging Through Blood
From the Department of Orthopedics, Harvard Medical School, Brigham and Women’s Hospital, Boston, Mass (M.B., K.S., C.J.); and the Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge (X.L., J.F.).
Correspondence to Mark Brezinski, MD, PhD, Massachusetts Institute of Technology, Bldg 36-345, 50 Vassar St, Cambridge, MA 02139. E-mail mebrezin{at}mit.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background—Most myocardial infarctions are caused by the rupture of small rather than large plaques in the arteries of the heart that are beyond the detection limit of current technologies.
Methods and Results—Recently, optical coherence tomography (OCT) has demonstrated considerable potential as a method for high-resolution assessment of vulnerable plaque. However, intravascular OCT imaging is complicated by the need to remove blood from the imaging field because blood results in substantial signal attenuation. This work examines index matching as a method for increasing penetration. Index matching is based on the hypothesis that the predominant source of scattering in blood is the difference in refractive index between the cytoplasm of erythrocytes and serum. By increasing the refractive index of serum to a value near that of the cytoplasm, or index matching, scattering can be substantially reduced. The concept was tested with a system that pumped blood in vitro through transparent tubing. The test compounds, dextran and intravenous contrast agent, both led to significant improvements in penetration (69±12% and 45±4%). No significant effect was seen with the saline control. For dextran, the effect could not be attributed to reductions of red cell number or volume because changes in these parameters were not different from the control. In the case of intravenous contrast, a small but significant relative reduction in red cell volume was seen.
Conclusions—This study demonstrates the feasibility of index matching for improving OCT imaging through blood. Future studies are required to identify compounds for effective index matching in vivo.
Key Words: tomography • plaque • myocardial infarction • blood • imaging
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Myocardial infarction, commonly known as heart attack, is the leading cause of death in the industrialized world.1 Most myocardial infarctions result from the rupture of small, thin-walled plaques in the coronary arteries. When these plaques rupture, they release lipid into the blood, a clot forms, and the vessel occludes.2 3 Most of these plaques are below the detection limit of currently available imaging technologies.4 Therefore, a true clinical need exists for an imaging technology capable of identifying these plaques before rupture.
A recently developed, high-resolution imaging technology, optical coherence technology (OCT), has demonstrated considerable potential as a method for imaging vulnerable plaque.5 6 7 8 OCT is analogous to ultrasound, measuring the intensity of back-reflected infrared light rather than sound.9 Initial in vitro OCT imaging of cardiovascular tissue was conducted with postmortem human aorta.5 In these studies, OCT was shown to identify structural features such as lipid collections, thin intimal caps, and fissures characteristic of plaque vulnerablity. OCT has also been directly compared with high-frequency intravascular ultrasound, the current clinical technology with the highest resolution.6 7 The superior resolution of OCT has been confirmed both quantitatively and qualitatively.6 7
Recently, in vivo OCT imaging has been performed of the rabbit aorta at 4 frames per second through a 2.9F catheter.8 Structure within the rabbit aorta wall was defined at 10-µm resolution, but saline infusion was required during imaging because blood led to significant attenuation of imaging. Eliminating the need for saline flushes would represent a substantial advance for intravascular OCT imaging.
Light attenuation in tissue (or blood) can arise from either
scattering or absorption. OCT imaging is typically performed at 1300 nm
in nontransparent tissue, where absorption is
low.10 Previous work has
demonstrated that scattering of light by human tissue occurs from both
the cells themselves and organelles within the cells, such as the
mitochondria and
nuclei.11 12
However, by far the predominate cell type of blood is the erythrocyte
or red blood cell (approximately 5x106/µL
versus 4x103/µL for the white blood
cell). Because mature erythrocytes contain no nucleus or organelles, it
can be hypothesized that the red cell volume is the major source of
scattering. Mies treatment of the red blood cell is consistent
with this hypothesis.13 The
reduced scattering coefficient, a measure of total scattering, has been
previously expressed through the approximation method of the
Mie theory
as14 µs'=(2.46/a)(
{1-})(2
a
nex/
)0.37({nin/nex}-1)2.09
where µs' is the
reduced scattering cross section, is the volume fraction of the
scattering particles relative to the total tissue volume, a is the
radius of the particle, is the wavelength of the scattered light,
nin is the intracellular refractive index, and
nex is the extracellular refractive index. This
equation can be divided into two components, the refractive index
dependent factor
({nin/nex}-1)2.09
and the scatter size–dependent factor ({1-})(2a
nex/)0.37. It can
be seen that as nin and
nex become closer, the scattering approaches
zero. The nin and nex are
not known at 1300 nm for the red cell, but in the visible region, they
are 1.456 and 1.33.15
Therefore, in the case of the erythrocyte, we hypothesize that by index
matching, or increasing the plasma-refractive index closer to that of
the intracellular refractive index, penetration will increase. In this
study, dextran and intravenous contrast will be used to
increase the refractive index of serum closer to that of the
erythrocyte cytoplasm, theoretically increasing penetration. A
feasibility will therefore be suggested for index matching as a method
of improving OCT penetration through blood in
vivo.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The principles behind OCT have been previously described.5 9 A superluminescent diode light source was used for OCT imaging. The diode had a median wavelength of 1300 nm and a spectral bandwidth of 50 nm. The resolution of images was 10 µm. The resolution was determined experimentally by measuring the axial point spread function with the use of a mirror, a standard technique for defining resolution. The lateral resolution was determined by the spot size or focusing power of the lens system and was measured to be 10 µm. Acquisition rates were
5 seconds, but systems have now been developed that can image at
video rate.16
A system was used that allows the blood to be circulated in
vitro through transparent tubing to reproduce coronary flow.
Human whole blood was obtained from discarded blood bank samples and
diluted to a hematocrit (Hct) of 35% with normal saline (actual
measured values for each group are listed in results section). As shown
in
Figure 1
, blood was pumped through a closed system of tubing
by a perfusion pump (Sigma). The flow rate was 200 mL/min, which is
approximately the peak flow in the coronary artery. The
diameter of the tubing was 6 mm, approximately the diameter of a
normal adult coronary. A reflector was placed in the tubing as
shown in
Figure 2. The section of the reflector imaged is 2
mm below the inner surface of the tubing. Once blood was introduced
into the system and circulated, OCT imaging of the reflector was
performed. An example OCT image of the reflector is shown in
Figure 3a. The red arrow is the reflector, the black arrow
is the inner surface of the tubing, and B is the blood.
|
|
|
The total intensity of the signal off the reflector is used to represent penetration. The more light is scattered by blood, the less the signal off the reflector. The intensity on each image was measured in 10 transverse points on the reflector (A-scans), equally spaced along the length of the reflector, for each image using IP Laboratory (Scanalytics, Inc). The 10 paired measurements were averaged to give a single measurement for the image. Ten measurements were taken because the metal reflector is heterogeneous, varying by an average of 12%, for example, for dextran. A mirror could not be used because of the angular dependence.
After baseline data had been obtained with blood, test
substances were added to the blood. The test substances were dextran
(10 g in 40 mL of normal saline), 40 mL of intravenous
contrast (Hexabrix, Mallinckrodt Inc), or normal saline (40 mL). The
active ingredients of intravenous contrast are ioxaglate
meglumine and ioxaglate sodium. The group refractive index for dextran
and intravenous contrast we measured to be 1.52 and
1.46, respectively. Hct and red cell concentrations were measured
before and after the experiments. The Hct was measured with capillary
tube centrifugation; red cell concentration was
measured with a hematocytometer. All substances added had a volume of
40 mL, which was added to a total volume of 260 mL.
In a separate series of experiments, the hypothesis that the predominate source of scattering is the intracellular/extracellular refractive index mismatch was qualitatively tested. A reflector was placed in the same circulating system as described above. The reflector was first imaged in the presence of saline. Then, after the saline was removed, blood (Hct 35%) was placed in the system and the intensity off the reflector measured again. Next, the lysed blood was added. To induce lysis, instead of diluting to a Hct of 35% with saline, the same volume of distilled water was added. Lysis to a Hct <1% was confirmed by light microscopy.
All values represent mean±SEM. Two-sample t procedures were used to determine significant differences between groups.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Figure 3
shows an example image generated in the presence
and absence of intravenous contrast. The reflector
intensity, shown by the red arrows, has increased in the second image
because of the presence of the intravenous contrast. In
Figure 4, the summary of effects of dextran,
intravenous contrast, and the saline control on penetration
are illustrated. For the saline control, a 7±3% increase in signal
intensity was noted, which was not a statistically significant effect.
A 69±12% increase was noted for dextran, which was statistically
different from the saline control
(P<0.005). For the
intravenous contrast, a 45±4% increase was noted, which
was significantly different from the control
(P<0.001).
|
A decrease in Hct and red cell concentration is expected in
all groups as the result of dilution (40 mL in 260 mL) alone, which
should be constant across each group. In
Figure 5, the influence of each test compound on Hct and red
cell concentration is examined to look for additional effects. Saline
resulted in a 15±2% decrease in Hct, whereas a 17±2% decease was
noted for dextran. Therefore, the effect is not due to tonicity. These
values were not significantly different. A 24±2% decrease in Hct was
noted for intravenous contrast, which was significantly
more than the saline control
(P<0.005). There was no
significant difference in the reduction in red cell concentration for
dextran (15±2%), intravenous contrast (19±2%), or the
normal saline control (18±2%). Initial Hcts were not significantly
different for saline (33±2%), dextran (31±1%), or
intravenous contrast (32±2%).
|
Figure 6 demonstrates imaging in the presence of saline,
blood (Hct 35%), and lysed blood (Hct <1%). It can be seen
that in the presence of blood, the reflector is difficult to
locate. However, when the red cells were lysed, signal intensity
returned to values not significantly different from saline. This is
consistent with the hypothesis that intracellular/extracellular
mismatch and not membranes or hemoglobin absorption is the main source
of near infrared attenuation by blood.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Previous work has suggested a link between changing refractive index and the optical properties of tissue. For one group, a possible correlation between blood glucose concentration and the reduced scattering coefficient of tissue was noted.17 In their study, the authors suggested that increasing glucose concentration in the extracellular fluid resulted in increasing refractive index. Their hypothesis was that increasing the extracellular refractive index to a level near that of tissue cellular constituents resulted in improved penetration in tissue. Another group demonstrated that solute concentration effected scattering in yeast and perfused liver through a similar mechanism.18 19 In our study, the feasibility of index matching, the raising of serum refractive index to reduce index mismatch, was demonstrated for the potentially important clinical problem of increasing OCT penetration through blood. Both dextran and intravenous contrast were found to significantly increase penetration, as demonstrated by an increase in signal intensity. In the case of dextran, the effect was not due to the small amount of dilution that occurred, because the same volume of saline vehicle had no significant effect. The effect was also not due to lysis or changes in red cell volume because changes in cell number and Hct were the same for both dextran and saline. The dextran effect is therefore consistent with index matching. With the intravenous contrast, a small but significant decrease in red cell volume was noted by a decrease in Hct but not the red cell number relative to the saline control. Therefore, although unlikely, some improvement in penetration may be due to a reduction in cell volume. The lack of improved penetration with the addition of normal saline (40 mL) may seem surprising. However, it is consistent with previous work that suggested dilution of the Hct to <10% was necessary before significant improvement in penetration was seen.20
Because the red cell has no organelles and hemoglobin should not absorb significantly at this wavelength, our hypothesis is that the predominant source of attenuation is the mismatch between the serum and cytoplasm. In this study, a qualitative defense of this hypothesis was performed by demonstrating that lysis returned penetration to a value not significantly different from saline. The fact that the cell membrane is not the major source of scattering is not surprising because they are too small relative to the wavelength to significantly scatter.21
In this study, fresh blood was not used for several reasons. First, the authors believed that although the mismatch and cell shape for blood bank cells may be somewhat different from fresh blood, blood bank samples were still sufficient for proof of concept. Second, because large volumes of blood were required, the blood bank appeared to be the optimal source.
The authors are not suggesting that dextran or intravenous contrast is the ultimate compound to be used for in vivo index matching. They were compounds that the laboratory possessed that had reasonable refractive indexes and that are already used intravenously. These experiments were intended only as feasibility studies for index matching. The current compounds did not increase penetration sufficiently for in vivo use and impractical volumes. Furthermore, considerations that need to be taken into account when considering an agent for in vivo use include toxicity, homogeneity of distribution, binding to serum constituents, rate of uptake by the cells, and rate of uptake into extravascular spaces. Although it may seem odd to use intravascular agents to reduce contrast, which is not currently done in clinical medicine, intravascular agents are widely used to increase contrast. This includes agents used in conventional radiography, CT, MRI, and nuclear medicine. Therefore, extension to contrast reduction does not represent a large change in clinical practice.
Conclusions
Index matching represents an attractive
approach to increasing penetration through blood for intravascular OCT
imaging. However, future studies are required to identify the
appropriate agents for in vivo
studies.
|
Acknowledgments |
|---|
This research was supported in part by National Institutes of Health contract R01-HL63953-01 (Dr Brezenski), NIH-9-RO1-EY11289-10 (Dr Fujimoto), NIH-1-RO1-CA75289-01 (Dr Brezinski), the Medical Free Electron Laser Program, Office of Naval Research contract grant N00014-97-1-1066 (Drs Brezinski and Fujimoto), Whittaker Foundation contract 96-0205 (Dr Brezinski), 1-RO1AR44812-01 (Dr Brezinski), and National Institutes of Health contract NIH-1-R29-HL55686-01A1 (Dr Brezinski).
Received July 26, 2000; revision received November 11, 2000; accepted November 6, 2000.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Herrick JB. Clinical features of sudden obstruction of the coronary arteries. JAMA.. 1912;59:2015–2020.
2. Richardson PD, Davies MJ, Born GVR. Influence of plaque configuration and stress distribution on fissuring of coronary atherosclerotic plaques. Lancet.1989;2:941–944.
3.
Falk E. Plaque
rupture with severe preexisting stenosis precipitating
coronary thrombosis: characteristics of coronary
atherosclerotic plaques underlying fatal occlusive thrombi.
Br Heart J. 1983;50:127–134.
4. Ambrose JA, Tannenbaum MA, Alexopoulos D, et al. Angiographic progression of coronary disease and the development of myocardial infarction. J Am Coll Cardiol. 1988;12:56–62.
5.
Brezinski ME,
Tearney GJ, Bouma BE, et al. Optical coherence tomography for optical
biopsy: properties and demonstration of vascular pathology.
Circulation. 1996;93:1206–1213.
6.
Brezinski ME,
Tearney GJ, Weissman NJ, et al. Assessing atherosclerotic plaque
morphology: comparison of optical coherence tomography and high
frequency intravascular ultrasound.
Heart. 1997;77:397–403.
7. Patwari P, Weissman NJ, Boppart SA, et al. Assessment of coronary plaque with optical coherence tomography and high frequency ultrasound. Am J Cardiol. 2000;85:641–644.[Medline] [Order article via Infotrieve]
8.
Fujimoto JG,
Boppart SA, Tearney GJ, et al. High resolution in vivo
intra-arterial imaging with optical coherence tomography.
Heart. 1999;82:128–133.
9.
Huang D, Swanson
EA, Lin CP, et al. Optical coherence tomography.
Science. 1991;254:1178–1181.
10. Parsa P, Jacques SL, Nishioka NS. Optical properties of rat liver between 350 and 2200 nm. Appl Opt. 1989;28:2325–2330.
11. Beauvoit B, Liu H, Kang K, et al. Characterization of absorption and scattering properties for various yeast strains by time-resolved spectroscopy. Cell Biophys. 1993;23:91–109.[Medline] [Order article via Infotrieve]
12. Dunn A, Richards-Kortum R. Sources of contrast in confocal reflectance imaging. Appl Opt. 1996;35:3441–3446.
13. Steinke JM, Shepherd AP. Comparison of Mic theory and the light scattering of red blood cells. Appl Opt. 1988;27:4027–4033.
14. Graaff R, Aarnoudse JG, Zijp JR, et al. Reduced light-scattering properties for mixtures of spherical properties: a simple approximation of Mic’s theory. Appl Opt.1992;31:1370–1376.
15. Weast RC. CRC Handbook of Chemistry and Physics. Cleveland, Ohio: CRC; 1975.
16. Rollins AM, Kulkarni MD, Yazdanfar S, et al. In vivo video rate of optical coherence tomography. Opt Express. 1998;3:219–230.[Medline] [Order article via Infotrieve]
17. Maler JS, Walker SA, Fantini S, et al. Possible correlation between blood glucose concentration and the reduced scattering coefficient of tissues in the near infrared. Opt Lett. 1994;19:2062–2064.[Medline] [Order article via Infotrieve]
18. Chance B, Lui H, Kitai T, et al. Effects of solutes on optical properties of biological materials: models, cells, and tissues. Anal Biochem. 1995;227:351–362.[Medline] [Order article via Infotrieve]
19. Liu HB, Beanvoit M, Kimura B, et al. Dependence of tissue optical properties on solute induced changes. J Biomed Opt. 1996;1:200–211.
20. Roggan A, Friebel M, Dorschel K, et al. Optical properties of circulating human blood in the wavelength range from 400–2500 nm. J Biomed Opt. 1999;4:36–46.
21. Bowman JJ, Senior TB, Uslenghi PL. Electromagnetic and Acoustic Scattering by Simple Shapes. New York, NY: Hemisphere Publishing Corp; 1987.
This article has been cited by other articles:
 |
|
 |
|
  M. U Farooq, A. Khasnis, A. Majid, and M. Y Kassab The role of optical coherence tomography in vascular medicine Vascular Medicine, February 1, 2009; 14(1): 63 - 71. [Abstract] [PDF]
|
|
|
|
 |
|
 D. Stamper, N. J. Weissman, and M. Brezinski Plaque Characterization With Optical Coherence Tomography. J. Am. Coll. Cardiol., April 18, 2006; 47(8S): C69 - C79. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A G. Podoleanu Optical coherence tomography Br. J. Radiol., November 1, 2005; 78(935): 976 - 988. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Nadkarni, B. E. Bouma, T. Helg, R. Chan, E. Halpern, A. Chau, M. S. Minsky, J. T. Motz, S. L. Houser, and G. J. Tearney Characterization of Atherosclerotic Plaques by Laser Speckle Imaging Circulation, August 9, 2005; 112(6): 885 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D.S. Celermajer Understanding the pathophysiology of the arterial wall: which method should we choose? Eur. Heart J. Suppl., September 1, 2002; 4(suppl_F): F24 - F28. [Abstract] [PDF]
|
|
|
|
|
|
P. R. Moreno, R. A. Lodder, K. R. Purushothaman, W. E. Charash, W. N. O'Connor, and J. E. Muller Detection of Lipid Pool, Thin Fibrous Cap, and Inflammatory Cells in Human Aortic Atherosclerotic Plaques by Near-Infrared Spectroscopy Circulation, February 26, 2002; 105(8): 923 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Villard, M. D. Feldman, J. Kim, T. E. Milner, and G. L. Freeman Use of a Blood Substitute to Determine Instantaneous Murine Right Ventricular Thickening With Optical Coherence Tomography Circulation, April 16, 2002; 105(15): 1843 - 1849. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||