Open Access
Add to BibSonomyAbstract
Simultaneous imaging of very early embryonic heart structure and function has technical limitations of spatial and temporal resolution. We have developed a gated technique using optical coherence tomography (OCT) that can rapidly image beating embryonic hearts in four-dimensions (4D), at high spatial resolution (10-15 um), and with a depth penetration of 1.5 - 2.0 mm that is suitable for the study of early embryonic hearts. We acquired data from paced, excised, embryonic chicken and mouse hearts using gated sampling and employed image processing techniques to visualize the hearts in 4D and measure physiologic parameters such as cardiac volume, ejection fraction, and wall thickness. This technique is being developed to longitudinally investigate the physiology of intact embryonic hearts and events that lead to congenital heart defects.
©2006 Optical Society of America
1. Introduction
Imaging modalities such as computed tomography, ultrasound, and magnetic resonance are used to diagnose pathology of the human heart. Gated cardiac imaging, defined as image acquisition synchronized to the heart cycle, is commonly used to mitigate motion artifact due to heart motion during image acquisition. Over many heart cycles, gated cardiac imaging can acquire sufficient data to produce 3D images of the heart that can be used to accurately calculate various physiological parameters such as ejection fraction and stroke volume [1–3]. Unfortunately, these imaging modalities provide limited resolution when investigating embryonic cardiac development in animal models.
Defects in developmental mechanisms during embryogenesis can result in congenital cardiac anomalies. Our understanding of normal mechanisms of heart development and how abnormalities can lead to defects has been hindered by our inability to simultaneously detect anatomic and physiologic changes in these small (<2mm) organs during development. Optical coherence tomography (OCT) [4] uses back-reflected near-infrared light to image internal structures of the living embryonic heart with high-resolution in two- and three-dimensions [5]. OCT offers higher resolution (2-15 μm axial, 5-20μm lateral) than other modalities that have been used to image embryonic hearts, such as high-resolution ultrasound (28 μm axial, 63 μm lateral) [6] and magnetic resonance microscopy (25 μm axial, 31 µm lateral) [7]. In addition, OCT imaging does not require acoustic index matching, as in ultrasound imaging. OCT can image in real time, whereas magnetic resonance microscopy’s temporal resolution ranges from minutes to hours [8]. Laser scanning confocal fluorescence microscopy has been used to image embryonic hearts with higher resolution than OCT [9]. However, confocal microscopy typically cannot image deeper than 200 μm in living tissue [10]. OCT routinely achieves depth penetration of 1-2 mm and does not require extrinsic contrast agents.
An important application of in vivo imaging of embryonic hearts is to measure physiological parameters such as ventricular volume, ejection fraction, stroke volume, and wall thickness. The ejection fraction (EF) represents the percent change in the ventricular volume from end diastole (relaxation) to end systole (contraction), while stroke volume is the magnitude of the change. A commonly used method [11,12] estimates stroke volume in embryonic hearts by extrapolating a 3D volume in the shape of an ellipsoid, representing the left ventricle, from 2D in vivo video tracings of the external surface of the heart. This assumption introduces artifact as the embryonic ventricle is not an ellipse. Currently there is no method available to directly measure these physiological parameters in 3D.
OCT has been demonstrated for high-resolution imaging in several animal models for developmental biology including Brachydanio rerio (zebrafish) [13], Rana pipiens (American leopard frog) [13], and Xenopus laevis (South African clawed frog) [14,15]. OCT has been used to estimate ejection fraction using 2D images of a Xenopus laevis heart [15]. In that work, a 3D volume in the shape of an ellipsoid, representing the left ventricle, was extrapolated from the 2D images. OCT has been used to record 3D images of excised, non-beating chick embryo hearts [5]. The investigators recorded 3D images from fixed, non-viable stage 15 and 18 hearts, and 2D slices were compared to histologic slices. They provided 2D real-time images of a beating stage 15 heart. Doppler OCT imaging of blood flow has also been demonstrated in the Xenopus laevis heart [16,17]. Yazdanfar et al [17] used retrospective cardiac gating with a slow speed time domain OCT system to produce a 2D Doppler OCT movie of the beating Xenopus heart. Retrospective gating refers to post-facto parsing of data recorded asynchronously with respect to the heart cycle. Images (or lines (A-scans) in reference 17) are parsed into bins approximately representing phases of the cardiac cycle.
This work represents, to our knowledge, the first demonstration of directly (acquisition synchronized with the heart cycle) gated OCT imaging [18,19]. We show four-dimensional representations of the early embryonic mouse and chick hearts. Image processing methods permitted digital sectioning of the beating heart, enabling visualization of three-dimensional internal embryonic structures of the heart from any desired orientation during contraction and relaxation. Finally, internal volumes and myocardial wall thicknesses can be directly measured and quantified during embryonic cardiac contraction and relaxation permitting direct measurements of embryonic ejection fraction at several stages of development. This is the first measurement of an embryonic heart ejection fraction from volumetric data, as opposed to extrapolation from 2D data. This technology will provide developmental cardiology research with an imaging tool that has adequate temporal resolution, spatial resolution, and field of view to assess early embryonic cardiac function.
2. Methods
2.1 Gated OCT imaging
Fig. 1. Gated Reconstruction. The figure depicts B-scans captured for 4 different phases in the cardiac cycle before moving to the next position to capture B-scans in the same phase as in the previous position. 3D images are reconstructed at each phase.
To obtain 3D volumetric images of a chicken/mouse embryo heart in a sequence of specific temporal phases through the cardiac cycle, we gated OCT image acquisition to the stimulus used to pace the heart. Images were acquired over many heart cycles as the OCT field of view was translated through the heart volume. A single 3D image was constructed by bringing together all of the B-scans acquired at a given phase in the cardiac cycle. This resulted in a 3D image that was a ‘snap shot’ of the heart at that phase. By generating 3D images at many sequential phases in the cardiac cycle, a 3D movie was assembled and 3D measurements of ejection fraction were made. Figure 1 shows an illustration of the gated imaging technique. The figure depicts, horizontally, B-scans captured at four different phases in the cardiac cycle before the field of view is translated to the next position to capture B-scans at the same four phases in the cardiac cycle.
Precise timing instrumentation was developed to pace the heart, control the beam-scanning optics, and generate the data acquisition synchronization signal. Figure 2 shows a block diagram of the gated OCT instrumentation. The pacing pulse, the frame save synchronization signal, and the scanning optics in the slow-scan direction (y-direction) were controlled using software-generated signals and a digital-to-analog converter (DAC) board in a PC. The pacing pulse from the DAC stimulated the heart, while simultaneously triggering a first arbitrary waveform generator (AWG) that generated a series of pulses (in this work 16). These pulses triggered a second AWG that generated signals to drive the fast scanning optics (x-direction) and the data acquisition synchronization signal. This system takes five minutes to capture 4800 B-scans, which constitute 16 3D volume sets (an entire day 2-3 chick embryo heart or a day 13.5 mouse embryo heart).
Fig. 2. Gated OCT Setup. Arbitrary waveform generators control timing elements (scanning optics and frame synchronization signal) in the OCT setup. We paced the heart at 1 Hz and are capable of capturing 4800 B-scans in five minutes. AWG 1 - arbitrary waveform generator 1, AWG 2 - arbitrary waveform generator 2.
2.2 Model preparation and imaging protocol
Embryonic chick hearts (Gallus gallus) between days 2 and 6 of development (n=15), and wild-type embryonic mouse hearts (day 12.5 and 13.5 dpc (days post coitum), n=6) were used. Fertilized eggs were incubated in a humidified, forced draft incubator (GQF Manufacturing, Savannah, Georgia) at 39°C. Pregnant female mice (C57Bl6Jax) underwent cervical dislocation and embryonic hearts were immediately harvested. These species were chosen because they are commonly used to investigate developmental mechanisms of normal and abnormal cardiac embryogenesis and the role they play in formation of congenital heart defects.
An electrode chamber was developed to hold and pace the embryonic heart. This consisted of two platinum sheets 4 cm apart embedded into a silicone pad, thus pacing the heart with field stimulation. Field stimulation was selected for ease of use to develop the imaging technology, not to preserve physiologic function. Methods under development will gate from the electrocardiogram and pacing will not be necessary. The heart was excised from the embryo in oxygenated Tyrode’s solution and immediately pinned to the silicone between the platinum sheets and bathed in an oxygenated Tyrode’s solution. The temperature of the bath was kept at room temperature to reduce the intrinsic heart rate from 200-180 beats per minute to below 60 beats per minute, which allowed us to pace the heart at 1 Hz. The pacing signal was a 50 ms square wave relayed to the platinum sheets through an isolator. The isolator generated biphasic pulses to prevent the pacing chamber from becoming polarized. Our microscope-integrated OCT scanner allowed us to visually verify that the embryonic heart was being paced appropriately [20].
We used a time domain system capable of capturing 4000 A-scans per second. The coherence function of our light source (which determines the axial resolution) was 14 μm full width at half maximum (FWHM) in air, while the intensity profile of the incident beam (which determines the transverse resolution) was measured as 10 µm FWHM. B-scans were acquired at 20 μm intervals to reduce imaging time. Four frames were recorded at each spatial location at each cardiac phase, and then averaged to reduce noise. We captured 16 equally spaced phases in the cardiac cycle resulting in 16 3D volumes. The entire 4D data set was recorded in less than 5 minutes.
2.3 Image analysis
Two essential post-acquisition image processing tasks that we performed were 3D data visualization (mouse and chick) and the measurement of physiological parameters (chick only). To reduce background noise and speckle noise, four frames at each position and phase were temporally averaged to produce one frame to be used for 3D visualization and measurements. The images were then cropped around the region of interest to reduce the size of the data set and therefore, processing time. Because the images are characterized by an intensity fall-off in depth, which is approximately exponential in nature, an exponential correction was applied in the axial direction [21]. Speckle noise was further reduced by applying a median filter. Finally, the heart was segmented from the background by thresholding and seeded region growing. The image sets with background set to zero were then ready for further visualization steps and for measurements to be made.
To create 3D renderings of the image data for visualization purposes, the aspect ratio was first corrected. Then, Gaussian filters in x, y, and z were used to smooth the borders of the heart. A marching cubes algorithm transformed the volume data to surface data for visualization with appropriate lighting and view angle. Movies were visualized by displaying the several 3D frames in sequence with the appropriate time lag (1/16 second) between frames. For cut-away 3D images and movies, the corresponding gray-scale 2D image was superimposed on the cut-away surface.
The wall thickness was measured from transverse optical sections through the embryonic chick heart. To segment the inner volume from the myocardium in a range of B-scans for a specific volume of scans (a single phase of the cycle), we applied thresholding and seeded region growing in 2D to demarcate the myocardium from the cardiac chamber. The segmentations were reviewed by an expert and edited manually where needed. Next, radial lines were drawn from the center point of the section. The “wall thickness” was defined as the distance from the intersection of the radial line with the inner surface to the intersection of the radial line with the outer surface. The wall thickness measurements reported here were taken from 2D sections, as opposed to ventricular volume and ejection fraction measurements, which were taken from 3D data.
In order to make an accurate physiological measurement of ejection fraction, it is important to consider structural, electrophysiological and motility characteristics of the early-stage hearts under stimulation. Because the embryonic hearts are field stimulated, the terms systole (representing end-contraction in a normally beating heart) and diastole (representing end-relaxation in a normally beating heart) are not appropriate. Instead, we use the terms end-contraction and end-relaxation, respectively. In the chick embryo the ventricle does not divide into a separate left and right ventricle until day six, and blood flow is peristaltic from initiation of the heart beat around stage 9 until approximately stage 15.
The ventricular volume was manually segmented. To aid in determining the true ventricular cavity surface during manual segmentation, we referred to a scanning micrograph and 2D movies of the slices (B-scans) in question. The ventricular volume was defined as the entire volume of the combined ventricle, since ventricular septation had not yet occurred at the stages of development investigated. When the entire combined ventricular volume was used to estimate the ejection fraction of the peristaltically beating heart tube (stage 13 and 15), however, much lower values than expected were obtained (results not shown). This was because one region of the ventricle was contracted during the peristaltic wave while a smaller region containing the forward moving fluid bolus was relaxed. Therefore, the ejection fraction was measured from a partial volume of the ventricular chamber for the earlier stage hearts (13 and 15) corresponding to a bolus volume. The stage 20 heart was calculated from the full volume. This partial volume was identified as an area of ventricular cavity that was maximally contracted after pacing, but maximally expanded after the contraction ceased. Ejection fraction was calculated as follows: [(End-relaxation volume - End-contraction volume)/End-relaxation volume]. All image processing was done using a combination of standard software suites in Analyze 4.0 (Mayo Clinic) and custom coding in Matlab 7.0 (Mathworks, Inc).
3. Results
3.1 Embryonic mouse
Fig. 3. Three-dimensional reconstruction of a 13.5 dpc embryonic mouse heart, ventral view (top left panel). A digital image of the same mouse heart is below the rendered image. The scale bar applies to all of the images. Panels A-D are samples of transverse 2D OCT “slices” through the heart at levels represented by the labeled lines in the reconstruction. The smooth walled outflow tract (oft) can be distinguished from the trabeculated right and left ventricles (panels B and C) right ventricle - RV, left ventricle - LV. The “corrugated” inner surfaces of the atria are also visible (panels A and B, left and right atria: LA and RA)
Optical coherence tomography was used to image a day 13.5 wild-type embryonic mouse heart in 4D (Fig. 3 and 4). At day 13.5 of mouse development, the heart has just completed ventricular septation and has achieved the structure of a four-chambered heart. OCT allowed discrimination of trabeculations within both the right and left ventricles (Fig. 3(B) and 3(D)). Thicker portions of the embryonic heart were not fully penetrated by the OCT probe light, as can be seen in B-scans (Fig. 3(B) and 3(D)).
In a cutaway view of the anterior portion of the heart in which the dorsal portion of the heart was digitally removed (Fig. 4), the irregular trabecular surface, as well as the smooth right atrial endocardial surface was clearly revealed. The surfaces of the endocardial cushions were also visible. This cut-away view of the three-dimensional heart was analyzed during contraction and relaxation. By clicking on Fig. 4(B), one can see a movie of the beating mouse heart.
Fig. 4. (A) Digitally “sliced” open embryonic mouse heart (same heart as in Figure 3). The scale bar applies to all of the images. The grey regions represent the “cut” surface, the red areas are the reconstructed portion. The rough ventricular trabecular surface can be easily distinguished from the smooth endocardial surface of the atria, and endocardial cushions lining the lumen are easily seen. Note that the OCT slices used to create the reconstruction in Figure 3 were transversely oriented, however once the reconstruction is made, the heart can be viewed from any angle, section, or view. (B) (321 KB) Three-dimensional reconstruction of a 13.5 dpc embryonic mouse heart, ventral view. The black line indicates the slice for the heart in (A) [Media 1].
3.2 Embryonic chick
OCT was able to resolve detailed aspects of the internal architecture of the beating tubular embryonic chick heart. Figure 5 shows images recorded from a stage 13 chick embryo heart. At this stage of development (day 2) the heart is tubular and blood flow is peristaltic. Figure 5(A) displays a photograph of the heart, and Fig. 5(B) represents a 3D OCT image of the same heart at a single phase of the cardiac cycle (3 phases prior to end-relaxation). Figure 5(C) shows a cutaway 3D image of the embryonic heart in end-relaxation, while Fig. 5(D) shows the heart in end-contraction. One can see that the OFT is open in end-contraction and closed in end-relaxation, as would be expected. Figure 5(G) shows OCT B-scans of a cross-section through the loop of the tubular heart. On the left the heart is shown in end-relaxation, while on the right the heart is in end-contraction. The segmented ventricular cavity is indicated by yellow and light green outlines (smaller at end-contraction). The yellow outline indicates the area used in the calculation of EF, while the green lines delineate sections of ventricular chamber that were not used in the computation. Panel E and F show a scanning electron micrograph of the same heart presented in the earlier panels, while panel H and I display sections through a different stage 13 embryonic heart stained with H&E and anti-sarcomeric actin (a marker for myocardium). The unlabeled arrow points to invaginated endothelial tissue, absent in panel I, as endothelial tissue does not contain actin. By clicking on Fig. 5(B) (3D rotating chick heart), 5(C) (3D beating chick heart), and 5(G) (2D beating chick heart), one can see OCT movies.
Figure 6 shows images from 4D datasets of stage 15 and stage 20 embryonic chick hearts. The top panel presents a 3D reconstruction of one of the phases from a stage 15 embryonic chick heart and a corresponding subset of 2D OCT images used in its construction (A-D). The bottom panel shows a 3D reconstruction of one of the phases from a stage 20 embryonic chick heart and a corresponding subset of 2D OCT images used in its construction (E-H). The outflow tract cushions (asterisks in panel E) can easily be distinguished, as can the atrioventricular endocardial cushions (EC in panel G). The interventricular foramen between the primitive left and right ventricles can also be distinguished from the atrioventricular cushions (panels F and G).
Fig. 5. Stage 13 chick embryo heart. All images in panel come from the same heart except for panel H and I. (A) Photograph of heart taken after fixation. (B) (752 KB) 3D OCT image taken with the 4D-OCT setup, providing a ‘snapshot’ of one of the phases in the cardiac cycle [Media 2]. (C & D) (370 KB) Embryonic chick heart in end-relaxation and end-contraction obtained with the 4D-OCT scanner. The yellow arrows point to a closed then open OFT [Media 3]. (E & F) Scanning micrographs cut with the same orientation as the 2D scans in panel G. The arrows point to the endothelial invaginations. (G) (567 KB) Two 2D images from the same position (position 16), but different phases (end-relaxation and end-contraction) in the cardiac cycle [Media 4]. Outlines delineate the ventricular chamber (smaller at end-contraction). The yellow outline indicates area used in the calculation of ejection fraction. The yellow arrow points to trabeculae in the ventricle. (H & I) Sections through the embryonic heart stained with H&E (panel H & I show low and high power images of the same section, respectively) and anti-sarcomeric actin (panel I, red). The unlabeled arrow points to the endothelial invagination, absent in panel I. CM -compact myocardium, CJ - cardiac jelly, D - dorsal, V - ventral, Pos 16 - position 16, Vent - ventricle, OFT - outflow tract, End-Relax - end-relaxation, and End-Cont - end-contraction.
High resolution imaging performed on a beating embryonic heart permitted physiologic measurements not possible by imaging modalities that require fixation for optimal resolution. It was possible to obtain measures of intracardiac volume, ejection fraction, and wall thickness. End-relaxation volumes were predictably larger than that at end-contraction.
Wall thickness measurements of the beating hearts were made. Fig. 7, panel A, shows the outline of a 2D slice through the ventricle of a stage 15 heart at end-relaxation. Radial lines were drawn from a center point and the difference between the endocardial and epicardial surfaces was determined during contraction and relaxation of this heart. Panel B is a plot of measurements made from this sample. Wall thickness during contraction was greater than during relaxation of the ventricle, as would be expected. These measurements, however, were not statistically significant.
Fig. 6. Top panel: 3D reconstruction of a stage 15 embryonic chick heart and a corresponding subset of 2D OCT images used in its construction (A-D). Bottom panel: 3D reconstruction of a stage 20 embryonic chick heart and a corresponding subset of 2D OCT images used in its construction (E-H). Panels A-D: The red outline marks endocardium of the heart tube that is in the process of contraction, green lines indicate that part of the heart tube in the process of relaxation. Panels E-H: The outflow tract cushions (asterisks in panel E) can easily be distinguished, as can the atrioventricular endocardial cushions (EC in panel G). The interventricular foramen between the primitive left and right ventricles can also be distinguished from the atrioventricular cushions (panel F). The scale bar applies to all of the images. oft - outflow tract, IVF - interventricular foramen, EC - endocardial cushions, LA - left side of atrium, LV - left ventricle, RV - right ventricle.
We were able to observe peristaltic contraction of the endocardial surface in the stage 13 looped hearts. Peristaltic contraction made identification of “contraction” and “relaxation” phases difficult and hindered calculation of a simple ejection fraction for this stage of development. Figure 8 shows static images from each phase of the cardiac cycle of a stage 13 heart in which the endothelial lining was outlined in red. Image 4 (end contraction) shows the bolus of saline forced into the outflow tract, whereas image 10 (end relaxation) shows a closed outflow tract with a maximally dilated ventricle. By clicking on Fig. 8, one can see a 2D movie demonstrating peristaltic beating.
Fig. 7. Wall thickness measured in a stage 15 embryonic heart. Panel A indicates the technique employed (see text). Panel B graphically shows the wall thickness during relaxation of contraction for the same heart at the radial positions indicated in Panel A.
Fig. 8. (337 KB) Peristalsis in a stage 13 embryonic chick heart is demonstrated in these selected OCT slices. The heart (same one as in Fig. 5) was originally imaged in the transverse direction. After the reconstruction was made, sections through the ventricle (vent) and outflow tract (oft) were digitally prepared. This orientation permitted the very first visualization of peristalsis as seen from the endocardial surface. The red outline traces the endocardial surface. The heart is paced just prior to image 1, and the fluid within the heart is distributed evenly between the ventricle and the outflow tract (arrows) until image 3, when the bolus is ejected into the outflow tract (3 and 4). Image 4 has been termed “end-contraction”. Image 10 reflects “end-relaxation”, and the fluid bolus is primarily within the ventricle. [Media 5]
4 Discussion
The embryonic heart undergoes dramatic morphological changes during development that previously could only be viewed with high resolution in preparations that have been fixed, embedded, and sectioned. We have demonstrated that optical coherence tomography is a powerful tool that can be used to rapidly visualize internal structures of the beating embryonic heart with near histology-level resolution in living tissue. Furthermore, acquisition of anatomic data in a gated fashion from living tissue mitigates motion artifact and permits direct and accurate physiologic measurements such as wall thickness, ventricular volume, and ejection fractions. Acquisition of anatomic data from living tissue also eliminates histologic artifacts produced by fixation and dehydration processes. With this tool, cardiac structures can be digitally sectioned and visualized from all orientations and at any time during the cardiac cycle, particularly views difficult to obtain consistently with histologic techniques.
By permitting relatively non-invasive imaging of the beating embryonic heart with high-resolution, this technology fills an important niche in embryonic cardiovascular physiologic investigations. Ultrasound techniques including Doppler analyses of blood flow have been modified to successfully image the embryonic heart [22-24] and are valuable tools for rapid, high resolution scans of larger embryonic hearts. OCT is able to image embryonic structures at a resolution necessary for imaging subtle defects within tubular hearts as small as 200-250 µm across. Three-dimensional imaging with ultrasound is possible, but published literature is limited to analysis of older fetuses [25,26]. The high resolution of OCT, in addition to the ability to quantify changes in cardiac physiology from different time points in the cardiac cycle, and the ability to observe the heart from any plane after the data has been collected, make OCT an ideal technique for detailed studies of defects of embryogenesis that occur in the early heart tube. It should be noted that, at present, both technologies - ultrasound and OCT - require externalization of the mammalian uterine sac for detailed imaging of embryonic cardiovascular physiology due to limitations of depth penetration that occur with increasing resolution [23].
Magnetic resonance microscopy (MRM) techniques generally require a non-beating heart to achieve maximal resolution, and therefore have not been used for physiologic investigations [7,27]. Imaging times with MRM can also be lengthy, up to 30 hours, for maximal resolution [7,27]. Zhang et al took 29 hours and 10 minutes to collect 3D MRI data for a day 4 chick embryo [27]. Newer methods based on confocal laser microscopic techniques [9,28,29] have been used on living embryos with excellent resolution (1-5 microns) but are limited to less than 300 µm depth penetration. Confocal laser techniques generally require fluorescent labeling of blood or endothelial surfaces to provide contrast [9,29]. Presently multiple imaging techniques and overlays of images are necessary to obtain physiologic measurements that can be correlated to anatomic images [29]. Selective plane illumination microscopy [30] has also been used to image the living zebrafish heart with excellent resolution (6 μm) and better depth penetration than confocal microscopy (approximately 500 μm), but still not enough depth penetration to observe avian and rodent embryo hearts.
The ventricular volume measured with OCT was less than comparably staged embryos investigated with video microscopy [11,31]. Video microscopy relies on traces of the outer epicardial surface of the heart in systole and diastole from video images, followed by derivation of the inner area from mathematical models that assume the ventricular chamber is an ellipse [11,31]. The assumption that the embryonic ventricular cavity is an ellipse rather than the irregular tube that it appears to be is a source of error that could explain the smaller volumes measured with OCT. Furthermore, the high spatial resolution of OCT allowed detection of slender invaginations of endothelial cells into the cushion matrix, and exclusion of the majority of trabeculations from calculations of ventricular volume. The volume measured with OCT methods, therefore, is a direct measurement of the volume from a 3D image set, and is likely more accurate than video microscopy.
Table 1. Ventricular ejection fractions calculated from 2D OCT slices as described in the text. Ejection fraction values are calculated from regions of maximal contraction rather than the entire ventricular volume due to peristaltic activity of the embryonic heart.
Ejection fractions were calculated for several stages of the embryonic chick heart during its transition from a tubular to a looped segmented heart. The values obtained with our methods varied between 31% and 62% (see table 1). Published reports of the ejection fraction in similarly staged chick embryo hearts using the video-tracing method measured ejection fractions that were near 60% [11,31]. The variability and lower ejection fraction measurements were likely due to the non-physiologic experimental setup, as the hearts investigated were excised, paced, and studied at room temperature. Thus, the extent of myocardial contraction was less than what would be expected in vivo and at normal temperatures. Video microscopy may also overestimate ventricular volumes as described above. Techniques are being developed that will permit OCT imaging gated to the electrocardiogram of intact, normothermic hearts, and will likely produce more physiologic, less variable results.
Hemodynamics are critically important in the structural and physiologic development of the heart [11,32–37]. In previously published work, intracardiac hemodynamics were altered mechanically in the very early embryo (tubular stages) with the use of clips or sutures placed prior to the inlet of the heart (lowers intracardiac pressures), or at the outflow tract (raises intracardiac pressures). It is becoming clear that stretch on cardiomyocytes is important for cell-cell signaling and the turning on or off of important cellular events [33,37–39]. Until now, the effects of altered hemodynamics could not be measured acutely since there was no technology that could be used to observe and measure these changes in such small embryonic hearts. Effects could only be measured in later stages, after major structural changes have taken place. With OCT, it is now possible to investigate the detailed structural changes that take place in the very early embryo, before obvious structural defects appear when the effects may potentially be reversed. Sensitive and rapid assays such as OCT used to study early embryonic hearts will be important in determining the earliest stages when functional and/or morphological defects can be detected in experimentally manipulated embryos to track primary defects and differentiate them from secondary and tertiary defects.
Currently, we are working on steps to improve our gated OCT system. Our current configuration required excision of the heart and external pacing. The excised heart does not behave like a heart within the intact embryo due primarily to absence of preload. We are building an OCT system capable of in vivo recordings, gating from the ECG of the beating heart.
The normal heartbeat for a chick embryo is 3-4 Hz, which is faster than our current pacing frequency (1Hz). Our present system can handle a 3-4 Hz heartbeat frequency, but we would collect fewer phases (5 versus 16) in the heart cycle. To improve the temporal resolution of our system, we are building a gated spectral-domain OCT (SD-OCT) system [40] (5 times faster) and gating individual A-scans. This system will make it possible to perform non-invasive, high resolution longitudinal studies of the mechanics of heart development.
In summary, we have shown that gated OCT can image the beating embryonic heart rapidly and in fine detail to allow quantitation of morphologic changes that occur as a consequence of cardiac contraction and relaxation in the beating embryonic heart. This technology has the potential to expand our understanding of the genetic and epigenetic mechanisms that drive normal and abnormal heart development.
References and Links
1. M. Belohlavek, D. A. Foley, T. C. Gerber, T. M. Kinter, J. F. Greenleaf, and J. B. Seward, “Three and Four dimensional Cardiovascular Ultrasound Imaging: A new Era for Echocardiography,” Mayo Clin. Proc. 68, 221–240 (1993). [PubMed]
2. P. Lanzer, E. Botvinick, N. Schiller, L. Crooks, M. Arakawa, L. Kaufman, P. Davis, R. Herfkens, M. Lipton, and C. Higgins, “Cardiac imaging using gated magnetic resonance,” Radiology 150, 121–127 (1984). [PubMed]
3. K. Saito, M. Saito, S. Komatu, and K. Ohtomo, “Real-Time Four-dimensional Imaging of the Heart with Multi-Detector Row CT,” Radiographics 23, e8–8 (2003). [CrossRef] [PubMed]
4. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical Coherence Tomography,” Science 254, 1178–1181 (1991). [CrossRef] [PubMed]
5. T. M. Yelbuz, M. A. Choma, L. Thrane, M. L. Kirby, and J. A. Izatt, “Optical Coherence Tomography A New High-Resolution Imaging Technology to Study Cardiac Development in Chick Embryos,” Circulation 106, 2771–2774 (2002). [CrossRef] [PubMed]
6. C. K. L. Phoon, O. Aristizabal, and D. H. Turnball, “Spatial velocity profile in mouse embryonic aorta and Doppler-derived volumetric flow: a preliminary model,” Am. J. Physiol. Heart Circ. Physiol. 283, H908–H916 (2002). [PubMed]
7. T. M. Yelbuz, X. Zhang, M. A. Choma, H. A. Stadt, M. Zdanowicz, G. A. Johnson, and M. L. Kirby, “Approaching Cardiac Development in Three Dimensions by Magnetic Resonance Microscopy,” Circulation 108, 154–5 (2002). [CrossRef]
8. J. M. Tyszka, S. E. Fraser, and R. E. Jacobs, “Magnetic resonance microscopy: recent advances and applications.,” Curr. Opin. Biotechnol. 16, 93–99 (2005). [CrossRef] [PubMed]
9. E. A. V. Jones, M. H. Baron, S. E. Fraser, and M. E. Dickinson, “Measuring hemodynamic changes during mammalian development,” Am. J. Physiol. Heart Circ. Physiol. 287, H1561–H1569 (2004). [CrossRef] [PubMed]
10. J. A. Izatt, M. R. Hee, G. M. Owen, E. A. Swanson, and J. G. Fujimoto, “Optical coherence microscopy in scattering media,” Opt. Lett. 19, 590 (1994). [CrossRef] [PubMed]
11. S. Steckelenburg-De Vos, P. Steendijk, N. T. C. Ursem, J. W. Wladimiroff, R. Delfos, and R. E. Poelmann, “Systolic and Diastolic Ventricular Function Assessed by Pressure-Volume Loops in the Stage 21 Venous Clipped Chick Embryo,” Pediatr. Res. 57, 16–21 (2005). [CrossRef]
12. K. Tobita and B. B. Keller, “Maturation of end-systolic stress-strain relations in chick embryonic myocardium,” Am. J. Physiol. Heart Circ. Physiol. 279, H216–H224 (2000). [PubMed]
13. S. A. Boppart, M. E. Brezinski, B. E. Bouma, G. J. Tearney, and J. G. Fujimoto, “Investigation of developing embryonic morphology using optical coherence tomography,” Dev. Biol. 177, 54–64 (1996). [CrossRef] [PubMed]
14. S. A. Boppart, B. E. Bouma, M. E. Brezinski, G. J. Tearney, and J. G. Fujimoto, “Imaging developing neural morphology using optical coherence tomography,” J. Neurosci. Methods 70, 65–72 (1996). [CrossRef] [PubMed]
15. S. A. Boppart, G. J. Tearney, B. E. Bouma, J. F. Southern, M. E. Brezinski, and J. G. Fujimoto, “Noninvasive assessment of the developing Xenopus cardiovascular system using optical coherence tomography,” Proc. Natl. Acad. Sci. USA 94, 4256–61 (1997). [CrossRef] [PubMed]
16. V. X. D. Yang, M. L. Gordon, B. Qi, J. Pekar, S. Lo, E. Seng-Yue, A. Mok, B. C. Wilson, and I. A. Vitkin, “High speed, wide velocity dynamic range Doppler optical coherence tomography (Part II): Imaging in vivo cardiac dynamics of Xenopus laevis,” Opt. Express 11, 1650–1658 (2003), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-11-14-1650. [CrossRef] [PubMed]
17. S. Yazdanfar, M. Kulkarni, and J. A. Izatt, “High resolution imaging of in vivo cardiac dynamics using color Doppler optical coherence tomography,” Opt. Express 1, 424–431 (1997), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-1-13-424. [CrossRef] [PubMed]
18. M. W. Jenkins, C. J. Pedersen, R. S. Wade, V. Nikolski, Y. Cheng, I. R. Efimov, and A. M. Rollins, “Three-dimensional OCT imaging of Endocardial architecture,” in Coherence Domain Optical Methods and Optical Coherence Tomography in Biomedicine VIII, V. V. Tuchin, J. A. Izatt, and J. G. Fujimoto , eds. Proc. SPIE 5316, 62–65 (2004). [CrossRef]
19. M. W. Jenkins, F. Rothenberg, D. Roy, V. Nikolski, D. L. Wilson, I. R. Efimov, and A. M. Rollins, “4D optical coherence tomography of the embryonic heart using gated reconstruction,” in Coherence Domain Optical Methods and Optical Coherence Tomography in Biomedicine IX, V. V. Tuchin, J. A. Izatt, and J. G. Fujimoto, eds. Proc. SPIE 5690, 1–3 (2005). [CrossRef]
20. Z. Hu and A. M. Rollins, “Quasi-telecentric optical design of microscope-compatible OCT scanner,” Opt. Express 13, 6407–6415 (2005), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-13-17-6407. [CrossRef] [PubMed]
21. J. A. Izatt, A. M. Rollins, and R. Ung-arunyawee, “System Integration and Signal Processing,” in Handbook of Optical Coherence Tomography, B. Bouma and G. Tearney, eds. (Marcel Dekker, Inc., New York, 2002), pp. 143–174.
22. Y. Q. Zhou, F. S. Foster, D. W. Qu, M. Zhang, K. A. Harasiewicz, and S. L. Adamson, “Applications for multifrequency ultrasound biomicroscopy in mice from implantation to adulthood,” Physiol. Genomics 10, 113–126 (2002). [PubMed]
23. C. K. L. Phoon and D. H. Turnball, “Ultrasound biomicroscopy-Doppler in mouse cardiovascular development,” Physiol. Genomics 14, 3–15 (2003). [PubMed]
24. Y. Gui, K. K. Linask, P. Khowsathit, and J. Huhta, “Doppler echocardiography of normal and abnormal embryonic mouse heart,” Pediatr. Res. 40, 633–642 (1996). [CrossRef] [PubMed]
25. A. H. Bhat, V. N. Corbett, R. Liu, N. D. Carpenter, N. W. Liu, A. M. Wu, G. D. Hopkins, X. Li, and D. J. Sahn, “Validation of Volume and Mass Assessments for Human Fetal Heart Imaging by 4-Dimensional Spatiotemporal Image Correlation Echocardiography: In Vitro Balloon Model Experiments,” J. Ultrasound Med. 23, 1151–1159 (2004). [PubMed]
26. A. H. Bhat, V. Corbett, N. Carpenter, N. Liu, R. Liu, A. Wu, G. Hopkins, R. Sohaey, C. Winkler, C. S. Sahn, V. Sovinsky, X. D. Li, and D. J. Sahn, “Fetal Ventricular Mass Determination on Three-Dimensional Echocardiography: Studies in Normal Fetuses and Validation Experiments.,” Circulation 110, 1054–1060 (2004). [CrossRef] [PubMed]
27. X. Zhang, T. M. Yelbuz, G. P. Cofer, M. A. Choma, M. L. Kirby, and G. A. Johnson, “Improved Preparation of Chick Embryonic Samples for Magnetic Resonance Microscopy,” Magn. Reson. Med. 49, 1192–1195 (2003). [CrossRef] [PubMed]
28. T. F. H. Ota, Y. Ishihara, H. Tanaka, and T. Takamatsu, “In situ fluorescence imaging of organs through compact scanning head for confocal laser microscopy,” J. Biomed. Opt. 10, 024010 (2005). [CrossRef] [PubMed]
29. J. R. Hove, R. W. Koster, A. S. Forouhar, G. Acevedo-Bolton, S. E. Fraser, and M. Gharib, “Intracardiac fluid forces are an essential epigenetic factor for embryonic cardiogenesis,” Nature 421, 172–177 (2003). [CrossRef] [PubMed]
30. J. Huisken, J. Swoger, F. Del Bene, J. Wittbrodt, and E. H. K. Stelzer, “Optical Sectioning Deep Inside Live Embryos by Selective Plane Illumination Microscopy,” Science 305, 1007–1009 (2004). [CrossRef] [PubMed]
31. B. B. Keller, J. Tinney, and N. Hu, “Embryonic ventricular diastolic and systolic pressure-volume relations,” Cardiol. Young 4, 19–27 (1994). [CrossRef]
32. N. T. C. Ursem, S. Steckelenburg-De Vos, J. W. Wladimiroff, R. E. Poelmann, A. C. Gittenberger-de Groot, N. Hu, and E. B. Clark, “Ventricular diastolic filling characteristics in stage-24 chick embryos after extra-embryonic venous obstruction,” J. Exp. Biol. 207, 1487–1490 (2004). [CrossRef] [PubMed]
33. A. Shay-Salit, M. Shushy, E. Wolfovitz, H. Yahav, F. Breviario, E. Dejana, and N. Resnick, “VEGF receptor 2 and the adherens junction as a mechanical transducer in vascular endothelial cells.,” Proc. Natl. Acad. Sci. USA 99, 9462–9467 (2002). [CrossRef] [PubMed]
34. M. Reckova, C. Rosengarten, A. deAlmeida, C. F. Stanley, A. Wessels, R. G. Gourdie, R. P. Thompson, and D. Sedmera, “Hemodynamics Is a Key Epigenetic Factor in Development of the Cardiac Conduction System,” Circ. Res. 93, 77–85 (2003). [CrossRef] [PubMed]
35. T. Ishiwata, M. Nakazawa, W. T. Pu, S. G. Tevosian, and S. Izumo, “Developmental Changes in Ventricular Diastolic Function Correlate With Changes in Ventricular Myoarchitecture in Normal Mouse Embryos,” Circ. Res. 93, 857–865 (2003). [CrossRef] [PubMed]
36. N. Hu and E. Clark, “Hemodynamics of the Stage 12 to Stage 29 Chick Embryo,” Circ. Res. 65, 1665 (1989). [PubMed]
37. C. Hall, R. Hurtado, K. Hewett, M. Shulimovich, C. Poma, M. Reckova, C. Justus, D. Pennisi, K. Tobita, D. Sedmera, R. Gourdie, and T. Mikawa, “Hemodynamic-dependent Patterning of Endothelin Converting Enzyme 1 Expression and Differentiation of Impulse-Conducting Purkinje fibers in the Embryonic Heart,” Development 131, 581–592 (2004). [CrossRef] [PubMed]
38. J. Omens, “Stress and strain as regulators of myocardial growth.,” Progress in Biophysics & Molecular Biology 69, (1998).
39. Y. Kuramochi, C. C. Lim, X. Guo, W. S. Colucci, R. Liao, and D. B. Sawyer, “Myocyte contractile activity modulates norepinephrine cytotoxicity and survival effects of neuregulin-1-beta,” Am. J. Physiol. Cell Physiol. 286, C222–229 (2004). [CrossRef]
40. A. F. Fercher, C. K. Hitzenberger, G. Kamb, and S. Y. El-Zaiat, “Measurements of intraocular distances by backscattering spectral interferometry,” Opt. Commun. 117, 43–48 (1995). [CrossRef]
