Real-time assessment of cardiac perfusion, coronary angiography, and acute intravascular thrombi using dual-channel near-infrared fluorescence imaging

Abstract

Objectives: We have developed an image-guided surgical system based on invisible near-infrared fluorescent light. Presently, the only clinically available near-infrared fluorophore is indocyanine green, which fluoresces at approximately 800 nm and is used for coronary angiography. Our objective was to determine whether methylene blue, already US Food and Drug Administration approved for other indications, has useful near-infrared fluorescence properties for image-guided cardiac surgery.

Methods: The optical properties of methylene blue were measured after dissolution in 100% serum. Biodistribution and clearance were quantified in organs and tissue after intravenous bolus injection of 2 mg/kg methylene blue in 3 rats. Coronary arteriography and cardiac perfusion were imaged in real time after intravenous bolus injection of 1 mg/kg methylene blue in 5 pigs with coronary obstructions. Coronary angiography and acute thrombi were assessed by using 800-nm fluorophores, indocyanine green, and IR-786-labeled platelets, respectively.

Results: The peak absorbance and emission of methylene blue as a near-infrared fluorophore occur at 667 nm and 686 nm, respectively. After intravenous injection, methylene blue provides highly sensitive coronary angiography. A lipophilic cation, methylene blue is extracted rapidly into tissue, with myocardium displaying unusually high uptake. Methylene blue permits real-time visualization and quantitative assessment of myocardial perfusion. Because of absent spectral overlap, use of 2 independent fluorophores in our imaging system permits simultaneous quantification of perfusion, venous drainage, and/or intravascular thrombi.

Conclusions: Methylene blue is an effective near-infrared fluorophore that provides direct visualization of coronary arteriography and cardiac perfusion. In conjunction with approximately 800-nm near-infrared fluorophores, important functional assessments during cardiac surgery are also possible.

Figure 1. Image-Guided Surgery System Equipped with Two NIR Fluorescence Channels

Schematic of the imaging system showing light paths for white light, NIR channel #1 excitation (670 nm) and emission (700 nm), and NIR channel #2 excitation (760 nm) and emission (800 nm). All images are acquired and displayed simultaneously, and in real-time.

Figure 2. Chemical and Optical Properties of Methylene Blue

A. Chemical structure and molecular weight (Da) of methylthioninium chloride (methylene blue). B. Absorbance (left axis) and fluorescence (right axis; 654 nm excitation) for 2 µM methylene blue in 100% FBS.

Figure 3. In Vivo Biodistribution and Clearance of Methylene Blue

A. Kinetics of 700 nm MB NIR fluorescence in the myocardium of rat. Shown are mean ± SEM signal intensity from N = 3 independent animals. B. Ratio of NIR fluorescence intensity of the myocardium relative to various vital organs and tissues, after intravenous injection of 2 mg/kg MB in rats (N = 3). Shown are values at 0 min, at the time (parentheses) of peak fluorescence in the organ/tissue under study, and at 60 min.

Figure 4. Real-Time, Simultaneous Assessment of Myocardial Perfusion and Coronary Angiography using MB and ICG

A. Arterial phase (12 sec), venous phase (1 min), and late phase (10 min) after intravenous bolus injection of 1 mg/kg MB into pigs. A diagonal branch of the LAD was clamped immediately prior to MB injection, with the perfusion defect easily detected (arrows). Shown are color video (top row), NIR fluorescence (middle row), and a pseudo-colored (lime green) merge of the two (bottom row). NIR fluorescence images were acquired with a 100 msec exposure time and are displayed with identical normalizations. Images are representative of N = 5 independent animals. B. Arterial phase (12 sec), venous phase (1 min), and late phase (10 min) after de-clamping of the diagonal branch and intravenous bolus injection of 0.06 mg/kg ICG into the pig from (A). Arrows mark the location of the previous perfusion defect seen with MB. Shown are color video (top row), NIR fluorescence (middle row), and a pseudo-colored (lime green) merge of the two (bottom row). NIR fluorescence images were acquired with a 100 msec exposure time and are displayed with identical normalizations. Images are representative of N = 5 independent animals. C. Quantitative analysis of the ratio of myocardial NIR fluorescence to coronary artery NIR fluorescence, 10 min after intravenous bolus injection of either ICG or MB. Shown are the mean ± S.E.M. for N = 5 independent animals, along with the statistical comparison of the two NIR fluorophores at each time point.

Figure 4. Real-Time, Simultaneous Assessment of Myocardial Perfusion and Coronary Angiography using MB and ICG

A. Arterial phase (12 sec), venous phase (1 min), and late phase (10 min) after intravenous bolus injection of 1 mg/kg MB into pigs. A diagonal branch of the LAD was clamped immediately prior to MB injection, with the perfusion defect easily detected (arrows). Shown are color video (top row), NIR fluorescence (middle row), and a pseudo-colored (lime green) merge of the two (bottom row). NIR fluorescence images were acquired with a 100 msec exposure time and are displayed with identical normalizations. Images are representative of N = 5 independent animals. B. Arterial phase (12 sec), venous phase (1 min), and late phase (10 min) after de-clamping of the diagonal branch and intravenous bolus injection of 0.06 mg/kg ICG into the pig from (A). Arrows mark the location of the previous perfusion defect seen with MB. Shown are color video (top row), NIR fluorescence (middle row), and a pseudo-colored (lime green) merge of the two (bottom row). NIR fluorescence images were acquired with a 100 msec exposure time and are displayed with identical normalizations. Images are representative of N = 5 independent animals. C. Quantitative analysis of the ratio of myocardial NIR fluorescence to coronary artery NIR fluorescence, 10 min after intravenous bolus injection of either ICG or MB. Shown are the mean ± S.E.M. for N = 5 independent animals, along with the statistical comparison of the two NIR fluorophores at each time point.

Figure 4. Real-Time, Simultaneous Assessment of Myocardial Perfusion and Coronary Angiography using MB and ICG

A. Arterial phase (12 sec), venous phase (1 min), and late phase (10 min) after intravenous bolus injection of 1 mg/kg MB into pigs. A diagonal branch of the LAD was clamped immediately prior to MB injection, with the perfusion defect easily detected (arrows). Shown are color video (top row), NIR fluorescence (middle row), and a pseudo-colored (lime green) merge of the two (bottom row). NIR fluorescence images were acquired with a 100 msec exposure time and are displayed with identical normalizations. Images are representative of N = 5 independent animals. B. Arterial phase (12 sec), venous phase (1 min), and late phase (10 min) after de-clamping of the diagonal branch and intravenous bolus injection of 0.06 mg/kg ICG into the pig from (A). Arrows mark the location of the previous perfusion defect seen with MB. Shown are color video (top row), NIR fluorescence (middle row), and a pseudo-colored (lime green) merge of the two (bottom row). NIR fluorescence images were acquired with a 100 msec exposure time and are displayed with identical normalizations. Images are representative of N = 5 independent animals. C. Quantitative analysis of the ratio of myocardial NIR fluorescence to coronary artery NIR fluorescence, 10 min after intravenous bolus injection of either ICG or MB. Shown are the mean ± S.E.M. for N = 5 independent animals, along with the statistical comparison of the two NIR fluorophores at each time point.

Figure 5. Real-Time, Simultaneous Assessment of Acute Coronary Artery Thrombi and Coronary Artery Patency

A coronary artery thrombus was induced in a diagonal branch of the LAD by treatment with FeCl₃ (center of dotted rectangle). Thrombus size and location was quantified by intravenous injection of IR-786-labeled autologous platelets (800 nm fluorescence; pseudo-colored in magenta in merged image). Coronary artery patency (and tissue perfusion) was quantified by intravenous bolus injection of 1 mg/kg MB (700 nm fluorescence; pseudo-colored in lime-green in merged image). Shown are color video (1^st column), 800 nm NIR fluorescence (2^nd column), 700 nm NIR fluorescence (3^rd column) and a pseudo-colored merge of the three (4^th column) at 12 sec post MB injection. The area within the dotted rectangle in the top row is shown magnified 4-fold in the bottom row. NIR fluorescence images were acquired with a 100 msec exposure time and are displayed with identical normalizations. Images are representative of N = 4 independent animals.