A new microscope system for the continuous observation of the coronary microcirculation in the beating canine left ventricle

MICROVASCULAR

RESEARCH

28, 387-394 (1984)

TECHNICAL REPORT A New Microscope System for the Continuous Observation the Coronary Microcirculation in the Beating Canine Left Ventricle

of

KOUICHI ASHIKAWA, HIROSHI KANATSUKA, TOSHIMI SUZUKI, AND TAMOTSU TAKISHIMA First Department

of Internal

Medicine,

Tohoku University

Received

School of Medicine,

Sendai 980, Japan

October 18. 1984

A microscope system was designed using a new type of objective lens which makes possible the direct and continuous observation of the coronary microcirculation throughout the entire cardiac cycle in the beating canine heart. The microscope system consists of a standard microscope and a floating objective system which is composed of a pair of convex lenses and transmits a real image of the coronary microcirculatory bed to a standard microscope without any change in magnification. The convex lens facing the heart is supported by a weight-adjusting coil spring and low-resistance ball bearings which allow the lens to move perpendicularly in unison with cardiac motion. To reduce excessive cardiac movement, two 24-gauge needles connected to the animal table by a needle holder are horizontally inserted through the midmyocardium of the left ventricle beneath the area of interest. The epimyocardium of the left ventricle is transilluminated by means of a light pipe and a xenon-arc lamp. The distance between the floating lens and the cardiac surface is kept constant using a spacing device connected to the light pipe holder to prevent the compression of the tissue in the microscopic field of view. This improvement in the microscope system combined with high-speed cinematography greatly facilitates the continuous analysis of the coronary microcirculation in the beating left ventricle throughout the entire cardiac cycle, and may provide a useful approach to the understanding of the regulation mechanism of the coronary circulation. D 1984 Academic press, hc.

INTRODUCTION Direct and continuous observation of the microcirculation has been quite useful in the study of the structure and function of the microcirculatory bed, in which both the transfer of nutrients and oxygen and the removal of metabolic waste products occur. The coronary microcirculation differs greatly from other microcirculatory areas in its continuous exposure to myocardial contraction. Several methods for the direct observation of the coronary microcirculation have been previously reported (Bing et al., 1972; Hellberg et al., 1971; Martini and Honig, 1969; Steinhausen et al., 1978; Tillmanns et al., 1974). In these, the development of a transillumination technique of the atrium (Bing et al., 1972; 387 00%2862/&l $3.00 Cowrinht Q 1984 bv Academic Press. Inc. All righk of reproduction in any form res&ed. Printed in U.S.A.

388

TECHNICAL

REPORT

Hellberg et al., 1971) and the ventricle (Tillmanns et al., 1974) provided new approaches for the study of the coronary microcirculation. However, the alteration of the distance between the heart and the microscopic objective which results from the constant movement of the heart prevented the continuous observation of the coronary microcirculation throughout the entire cardiac cycle, especially during rapid heart rate. Therefore, we have designed and built a new microscope system which allows for the direct and continuous observation of the coronary microcirculation throughout the entire cardiac cycle in the in situ beating heart. The purpose of this paper is to present the theoretical design considerations, a description of the system, and the pattern of red cell velocities in microvessels throughout the entire cardiac cycle in the beating canine left ventricle. THEORETICAL CONSIDERATIONS AND DESIGN OF THE MICROSCOPIC OBJECTIVE SYSTEM The most difficult aspect of microscopic observation of the coronary microcirculation is the alteration of the distance between the microscopic objective and the heart during cardiac contraction. This makes the direct and continuous microscopic observation of the coronary microcirculation almost impossible. The principle of our objective system is based on the transmission of the image through a pair of convex lenses (Fig. 1). That is, the light from the focus of convex lens A becomes parallel to the light axis after passing through lens A. The parallel light is gathered by the opposite convex lens B to produce a subsequent image plane which is not affected by a change in distance between these two convex lenses. Essentially, the real image on the front focus of lens A is transmitted to the back focus of lens B without any change in magnification. This real transmitted image is then observed by the objective lens C of a standard microscope. In fact, there was no change in magnification and no blurring of a microscale image on film for changes in floating objective lens (lens A) position by more than 15 mm in vertical distance. Lens B, CF 20 x (chromatic aberration-free objective; numerical aperture 0.40) (Nikon, Tokyo), was fixed on the microscopic stage. Lens A, on the other hand,

FIG. 1. A schematic illustration of the principle of image transmitting objective lens. for details.

See

text

TECHNICAL

REPORT

389

was designed to move up and down in unison with the myocardial surface during cardiac contraction. Lens A, CF 20 x (numerical aperture 0.40) (Nikon, Tokyo), was mounted in a thin aluminum tube (17 cm in length, 14 mm in outer diameter) to reduce the weight. The end of this tube was covered with an aluminum cap which had a small hole in the center (2 mm in diameter) that was shielded with a cover glass to prevent moisture on the lens. The distance between lens A and the cardiac surface was adjusted to the focal distance of this lens by turning the end cap to change vertical displacement of the lens. This aluminum tube, which included lens A, weighed 19.5 g and was supported by three pairs of lowresistance ball bearings and by a weight-adjusting coil spring so that the pressure on the cardiac surface was minimized (to approximately 3.5 g resting weight). In our image transmission system, the mass of the movable part was so light (19.5 g) that the inertial force for vertical movement was minimum. This minimal inertial force allows the floating lens to follow the cardiac motion. Although a CF 20 x (numerical aperture 0.40) (Nikon, Tokyo) was generally used for the objective lens C, a CF 10x (numerical aperture 0.25) (Nikon, Tokyo) was used in selecting a microscopic field for motion picture recordings,

TRANSILLUMINATION

OF EPIMYOCARDIUM

A 20-gauge light-transmitting needle (Fig. 2) was inserted into the subepicardial muscle layer of the left ventricle by using a micromanipulator. Tillmanns et al. (1974) have reported that this size needle scarcely disturbs the coronary microcirculation. A glass fiber (0.6 mm in diameter and 10 cm in length) was introduced through the lumen of the needle to transmit the light to the needle tip. The tip of the glass fiber was polished to give a mirrored reflecting surface of 45” angle (Nidek Co. Ltd., Tokyo) which directed light upward through a small hole 0.6 mm in diameter (Fig. 2). Thus, the superficial layers of the left ventricle were transilluminated with heat-filtered light. The tip of the needle and the space between the glass fiber and the hole of the needle through which light was transmitted were sealed with balsam (Matsunami Glass Co. Ltd., Osaka) to prevent blood and tissue fluid from entering the needle. The needle was fixed to a needle holder which was connected to a micromanipulator in a manner which allowed the tip of the needle to move up and down in union with cardiac motion (Fig. 3). TO MICROSCOPE f GLASS FIBER

i

\ ‘20 GAUGE NEEDLE

45” MIRRORED SURFACE FIG. 2. Illustration of the light-transmitting needle used for the transillumination of the epimyocardium. A glass fiber (0.6 mm in diameter) was introduced into a 20-gauge needle. The tip of the fiber was polished to give a mirrored reflecting surface of 45” angle to transmit the light toward the objective lens.

390

TECHNICAL

REPORT

JECTIVE LIFTER

FIG. 3. A schematic illustration of the needle holder and the lifter of the floating objective lens. The needle holder was operated three-dimensionally by a micromanipulator. The tip of the needle was designed to move vertically in unison with the cardiac motion. The lifter was operated by another micromanipulator which positioned the floating objective lens just above the surface of the epicardium.

Moreover, to prevent any disturbance of the microcirculation by the compression of the tissue in the microscopic field of view, the objective lens was lifted by an “arm” that was connected to the needle holder (Fig. 3). Since this “arm” could be moved perpendicularly by means of oil pressure in a microsyringe connected to a micromanipulator, the objective lens was lightly positioned just above the surface of the heart under the microscopic observation to avoid the compression of the tissue in the microscopic field of view. The thickness of transilluminated muscle was about 150 to 250 pm. The light source was a highpressure xenon-arc lamp (UXL 500D-0, USHIO INC., Tokyo). EXPERIMENTAL

PROCEDURE

Young mongrel dogs of both sexes, weighing 4.0-7.0 kg, were anesthetized with an intravenous injection of urethane (500 mg/kg) and chloralose (60 mglkg). If necessary, additional doses were given to maintain anesthesia. Following endotracheal intubation, ventilation was controlled by a positive-pressure respirator (Harvard, Type NSH-34RH). Positive end-expiratory pressure of 3 to 5 cm HZ0 was introduced to prevent atelectasis of the lung. Arterial blood gases were monitored and arterial PO, was maintained between 80 and 100 mm Hg. Metabolic acidosis during anesthesia was prevented by an intravenous infusion of sodium bicarbonate which maintained arterial pH at about 7.40. Aortic pressure was measured in the aortic root with a catheter that had been passed through the right carotid artery. A polyvinyl catheter was introduced into the superior vena cava via the external jugular vein for a drip infusion. A lead II ECG was monitored. A thoracotomy was performed in the fifth left intercostal space and the heart was suspended in a pericardial cradle. A plastic wrapping (Saran) was used to separate the lung from the anterior aspect of the heart. A 16-gauge Teflon tube was passed into the left ventricle through the apex for recording left ventricular pressure. Heart rate was kept constant at 140 beats per minute by means of right atria1 pacing after sinus node block. The preparation was kept moist during microscopic observations by continuously dripping warm

TECHNICAL

REPORT

391

Krebs-Ringer solution (37°C pH 7.400) on the cardiac surface. Rectal temperature was maintained at about 37°C by a heat blanket. The ECG, and aortic and left ventricular pressures were recorded on a Rectigraph (San Ei Sokki, Type 8k 12IS-ME, Tokyo) at a paper speed of 100 mm/set as needed. To reduce excessive movement of the heart, two 24-gauge steel needles, 5 cm in length, were horizontally inserted through the midmyocardium of the left ventricle at beneath the area of microscopic observations. The average distance between the two needles was 5 mm. The end of each needle was fixed to the plastic holders which were held with coil springs and were connected to the animal table via a steel holder to reduce excessive vertical movement of the heart (Fig. 4). This apparatus allowed the heart to move perpendicularly, but it limited excessive horizontal movements in order to hold the transilluminated area in the microscopic field of view. Moreover, the precautions which we took to avoid damage to larger coronary arteries lead us to believe that there was minimal disturbance of the microcirculation in the area of the epimyocardium which we observed. Thus, mechanical factors which could disturb the coronary microcirculation were eliminated as much as possible. By means of these techniques, the red cell velocities in capillaries had not significantly changed for at least 1.5 hr in preliminary experiments. The beam-splitter view of the microscopic image was monitored by means of a rotary shutter camera (shutter speed of l/l800 set; RSC 3000A, Sony) and a video motion analyzer (SVM-1110, Sony). Precise focusing was done by monitoring the TV and turning the knob for the fine vertical displacement of lens B. After focusing, the motion pictures were taken at 500 frames per second with a 16 mm high-speed motion picture camera (Milliken DBM-SD) (Fig. 5). High-speed panchrome film (Eastman Kodak 4-X) and Ektachrome film (Eastman Kodak 7251) were used with IOO-psecexposure times. Timing flashes at IO-msec intervals and signals that were synchronized with the R wave of the ECG were simultaneously recorded on each edge of the film. The optical magnification was 200x and was reduced to 50 x on the film as verified by a reference scale. This film magnification

LVP

FIG. 4. A schematic illustration of the heart holder designed to reduce excessive movement. See text for details.

392

TECHNICAL

REPORT

HIGH SPEED CAMERA

TV CAMERA BEAM SPLITTER

/

MICROSCOPIC SYSTEM M

EPI MYOCARDIUM XENON ARC LAMP

FIG. 5. A diagram of the coronary microcirculation observation system. See text for details.

was sufficient to identify individual erythrocytes on the projection screen. The smallest length scale which could be reliably resolved on the screen was less than 1 ,um as verified by a reference scale. The velocity of red cells in coronary microvessels was calculated every 20 to 40 msec from the distance of cell progressions on a projection screen and the number of frames needed. The optical magnification on the projection screen was verified by a reference scale. The film speed (500 frames/set) was confirmed by timing flashes on the edge of the film. RED CELL VELOCITY DURING A CARDIAC CYCLE Figure 6 shows a representative red cell velocity pattern in the capillaries of a beating canine left ventricle. With our technique it was possible to determine red cell velocities throughout the cardiac cycle, even during rapid cardiac contractions (heart rates of 140/min). The red cell velocities were measured every 20 to 40 msec. The velocity curve was then correlated to the ECG, and to aortic and left ventricular pressures. The peak capillary red cell velocity was reached in the ejection phase and decreased during diastole. During the isovolumetric contraction phase, a momentary reverse flow was observed. Continuous observation is essential for precise analyses of red cell velocity pattern throughout the entire cardiac cycle. However, there has been no report of the continuous observation of coronary microcirculation throughout the cardiac cycle in a beating mammalian heart. Tillmanns et al. (1974) have reported on the phasic changes of red cell velocities in the microcirculatory bed of the turtle ventricle. However, their ability to analyze the flow pattern of the entire cardiac cycle during rapid heart rates was limited by technical difficulties. By means of our floating objective lens system and high-speed cinematography, we could observe the coronary microcirculation directly and continuously throughout the entire cardiac cycle in beating canine left ventricle. Recently, techniques for the measurement of pressure and diameter in the

TECHNICAL REPORT

3ooo

393

CAP

2oocJlOOOo-lOOO-

-2000’ 10Gsec FIG. 6. A representative red cell velocity pattern in capillaries in beating canine left ventricle (epimyocardium). The top graphs illustrate the simultaneously recorded ECG, and aortic and left ventricular pressures. The right atrium was electrically paced at 140 beats/min. St = electrical stimulation, AP = aortic pressure, LVP = left ventricular pressure, CAP = capillary.

small coronary vessels of beating mammalian hearts have been reported by others (Nellis et al., 1981; Tillmanns er al., 1981). There is no doubt that blood flow velocity, as well as pressure and diameter, in coronary microvessels is an important variable in determining the hemodynamics of coronary microcirculation. We believe our new method offers considerable advantages For the study of the coronary microcirculation and may be very useful in reaching a better understanding of the regulation mechanism of coronary circulation. ACKNOWLEDGMENTS We are grateful to Dr. Richard J. Bing who first introduced coronary microcirculation research to us. We also greatly appreciate Dr. Harold Wayland for his helpful comments and advice. This work was supported, in part, by a grant from the Aging Research Council of Kowa Pharmaceutical Company Ltd., Tokyo, Japan.

REFERENCES R. J., WAYLAND, H., RICKART, A., AND HELLBERG, K. (1972). Studies on the coronary microcirculation by direct visualization. In “Myocardial Blood Flow in Man,” (A. Maseri, ed.), Minerva Medica, Torino, Italy. pp. 23-34. HELLBERG, K., RICKART, A., WAYLAND, H., AND BING, R. J. (1971). The coronary microcirculation in the potassium chloride arrested heart. J. Mol. Cell, Curdiol. 2, 221-230. MARTINI, J., AND HONIG, C. R. (1969). Direct measurement of intercapillary distance in beating rat heart in situ under various conditions of 0, supply. Microvasc. Res. 1, 244-256. NELLIS, S. H., LIEDTKE, A. J., AND WHITESELL, L. (1981). Small coronary vessel pressure and diameter in an intact beating rabbit heart using fixed-position and free-motion techniques. Circ. Res. 49, 342-353. BING,

394

TECHNICAL REPORT

M., TILLMANNS, H., AND THEDERAN, H. (1978). Microcirculation of the epimyocardial layer of the heart. I. A method for in vivo observation of the microcirculation of superficial ventricular myocardium of the heart and capillary flow pattern under normal and hypoxic conditions. Pfuegers Arch. 378, 9-14. TILLMANNS, H., IKEDA, S., HANSEN, H., SARMA, J. S. M., FAUVEL, J.-M., AND BING, R. J. (1974). Microcirculation in the ventricle of the dog and turtle. Circ. Res. 34, 561-569. TILLMANNS, H., STEINHAUSEN, M., LEINBERGER, H., THEDERAN,H., ANDK~BLER,W. (1981). Pressure measurements in the terminal vascular bed of the epimyocardium of rats and cats. Circ. Res. 49, 1202-1211. STEINHAUSEN,