INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The upper respiratory tract is the site of a wide range of disorders extending from infectious to neoplastic etiologies (1). Early detection of many of these disorders could greatly improve patient morbidity and mortality, especially the recognition of premalignant lesions. However, the sensitivity of current diagnostic techniques makes the detection and diagnosis of many diseases difficult if not impossible. For example, bronchial malignancies are not commonly detected at curable stages since precursor lesions are generally beyond the detection limit of current imaging modalities. The limited resolution of approaches such as conventional radiography, CT, and MRI generally prevents the detection of lesions less than 1 cm in diameter. Even the combination of bronchoscopy and excision biopsy suffers from substantial limitations since both have an inability to identify pathology below the tissue surface and a practical limit in the number of biopsies that can be performed diminishes the diagnostic potential. A technique that could perform real-time imaging to delineate tissue pathology with micron-scale resolution would be a very powerful tool in the diagnosis of disease. Optical coherence tomography (OCT) is a new imaging technology currently under development that enables fast, cross-sectional imaging of tissue microstructure (2). It allows in vitro or in vivo imaging of tissue in virtually any organ accessible via a catheter or endoscope, generating two-dimensional tomographic images of tissue microstructure or three-dimensional reconstructed volumes.

OCT was first applied to imaging optically transparent structures such as the anterior eye and retina (3, 4). Preliminary clinical investigations suggest that OCT is a promising technology for the detection and management of a variety of retinal diseases, including glaucoma and macular edema (5). Recent advances have led to the application of this modality to nontransparent tissue (6). Although the imaging depth of OCT is limited by the light-scattering and attenuation properties of tissue, penetration of 2 to 3 mm can be achieved even in heavily calcified samples. The resolution of OCT is on the micron scale, as much as two orders of magnitude higher than conventional ultrasound. In essence, imaging is being performed over the distance of a conventional biopsy and near the resolution of histopathology, which has led to the description of OCT as "optical biopsy." Previous studies with OCT have included the identification of pathology in the cardiovascular system, gastrointestinal tract, and skin, in addition to studies of normal urinary tract, nervous system, and reproductive tract (7). A small catheter-endoscope system has also been developed that allowed high resolution imaging to be performed internally in a rabbit model (10, 11).

In addition to its high resolution, several features of OCT suggest it will be a powerful imaging technology for the diagnosis of a wide range of pulmonary pathology (12). First, unlike ultrasound, OCT does not require a transducing medium, and it can be performed directly through air. This is particularly important for pulmonary imaging since the introduction of relatively small amounts of fluid can compromise the respiratory status of vulnerable patients. Second, unlike CT or MRI, OCT can be performed at or near real time, allowing information on tissue microstructure to be obtained within the dynamic environment of the bronchoscopy suite. Third, OCT is compact and portable, an important consideration for a clinically viable device. Finally, OCT is fiberoptic-based, allowing relatively easy integration with bronchoscopes without significant changes in device diameter.

This study has demonstrated the feasibility of OCT for ultrahigh resolution imaging of the upper respiratory tract by in vitro studies of human tissue from the epiglottis to the secondary bronchi. Images were obtained that demonstrated the ability of OCT to image microstructural features and demarcate tissue layers. Microstructure was correlated with histopathology to confirm the image interpretation and verified the ability of OCT to delineate features that could be used as pathologic markers. The ability of OCT to generate image resolution in the range close to that of histopathology in real time supports the hypothesis that this optical technology will become a powerful modality in the diagnosis and management of a wide range of clinical respiratory pathology.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

OCT is analogous to ultrasound imaging, but it is based on the detection of infrared light waves instead of sound, back-scattered (reflected) from different layers and structures within the tissue. Unlike ultrasound, however, the speed of light is very high, rendering electronic measurement of the echo delay time of the reflected light (time for the signal to return) impossible. Similar measurements can be performed, however, using a technique known as low coherence interferometry. Within the interferometer, the beam leaving the light source is split into two parts, a reference beam and a sample beam. The reference beam is reflected off a mirror at a known distance and returns to the detector. The sample beam reflects off different layers within the tissue and light returning from the sample and reference arms recombines. If the two light beams have traveled the same distances (optical path length), the two beams will interfer. OCT measures the intensity of interference obtained from different points within the tissue by moving the mirror in the reference arm, which changes the distance light travels in that arm. In Figure 1B and C, light from a coherent and low coherent source are shown illustrating how low coherence can be used to localize back-reflection sites and provide the desired high resolution. Two- or three-dimensional images are produced by scanning the beam across the sample and recording the optical backscattering versus depth at different transverse positions. The resulting data are a two- or three-dimensional representation of the optical backscattering of the sample on a micron scale. The logarithm of the backscattering signal is represented as a false color or gray scale image. A schematic of the complete OCT system is shown in Figure 2.

View larger version (26K): [in this window] [in a new window]   Figure 1.   Low coherence interferometry principles. Basic Michelson interferometer (A) and interference patterns, resulting from the reference arm mirror movement, using a coherent (B) and low coherent (C ) source. This technique measures the path length (distance traveled by light) to yield the optical backscattering in the specimen versus depth.

View larger version (15K): [in this window] [in a new window]   Figure 2.   Schematic of the fiberoptic implementation of the OCT System. OCT images are generated by performing successives measurements of optical backscattering versus depth at different transverse positions on the specimen.

The axial resolution (dz) with OCT is defined by the property referred to as the coherence length of the light source. A mathematical description yields:

(1)

where dz is the resolution,  is the wavelength, and is the bandwidth, i.e. the wavelength range, of the light source (13). The axial image resolution is inversely proportional to the bandwidth of the light source. The experiments reported here were performed using a short pulse chromium-forsterite laser operating at a wavelength of 1,280 nm with a bandwidth, , of 130 nm. The 1,280-nm wavelength in the near IR is close to a minimum in tissue absorption and thus allows deep penetration and imaging. The bandwidth of the light source yields a 6-µm axial resolution (14). The transverse resolution (dx and/or dy) is limited by the optical characteristics of the focusing delivery system (i.e., lens) on the specimen and more specifically the confocal parameter, i.e., the depth of focus, of the setup. The transverse resolution is determined by the following relationship:

(2)

where b is the confocal parameter and  is the wavelength of the source (15). The combination of optics and source used for the experiments reported here results in a 10- to 20-µm transverse resolution. The penetration is approximately 2 to 3 mm in scattering tissue and is limited by multiple scattering (16).

The tissue samples were obtained from cadavers and were refrigerated and maintained at 0° C in 0.1% sodium azide. After dissection, the tissue specimens were placed in a Petri dish and irrigated with isotonic saline to prevent dehydration during imaging. The acquisition of each image required between 10 and 30 s depending on the size (number of pixel elements) of the image. Because the OCT beam is invisible, tissue registration was performed with a visible-light guiding beam. The orientation of the imaging scan was marked on the specimen using the microapplication of India ink. The samples then underwent routine histologic processing. Samples were immersed in 10% buffered formalin for 48 h. The tissues were then processed for standard paraffin-embedding. Sections 5 µm thick were cut at the marked imaging sites and stained with hematoxylin-eosin or trichrome blue. The stained histologic sections enabled verification of tissue identity, and in most instances it allowed identification of sources of tissue contrast in the OCT images.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All the images are 6 by 3 mm and about 20 µm thick, and the scale bar (bottom left of each OCT image) is 500 µm, unless otherwise indicated. They were plotted in a logarithmic intensity scale with the least backscattering areas being white and the most backscattering areas being black as indicated by the intensity bar (bottom of each OCT image.)

An OCT image of a human epiglottis with associated histology is shown in Figure 3. The different layers of the tissue of the epiglottis are clearly visible. The epithelium, lamina propria, and cartilage are demarcated as well as are several glandular structures.

View larger version (149K): [in this window] [in a new window]   Figure 3.   In vitro OCT image of human epiglottis (A) and associated histology (B). The epithelial (e) and lamina propria (l ) layers are clearly identified as well as cartilaginous (c) and glandular (g) structures (image size: 6 × 3 mm × 20 µm).

The images of human vocal and ventricular chords (Figure 4) illustrate the ability of OCT to image the epithelial and lamina propria of tissue as well as characteristic structural features such as glands and vessels. Muscle layers appear as uniform amorphous regions in the OCT image. Higher resolution should enable identification of individual fiber bundles.

View larger version (72K): [in this window] [in a new window]   Figure 4.   In vitro OCT image of human vocal (A) and ventricular chords (B) and associated histology (C and D, respectively). Areas of epithelium (e) of different thickness can be identified as well as the lamina propria (l ) and the muscular (mm) and glandular (g) layers of the tissue (image size: 6 × 3 mm × 20 µm).

Trachea damaged from intubation was also imaged. (Figure 5A and B) The damage to both the epithelial and mucosal layers is visible in the images and associated histology. Several glands and gland ducts are also observed. Structures within the cartilaginous layer can also be identified, with layers such as the perichondrium and the interterritorial and territorial matrix clearly demarcated without any staining. In Figure 5C and D, an image of a secondary bronchus and the corresponding histology are shown. Cartilage, lamina propria, and glands are noted as well as the fact that OCT was able to image through the entire thickness of the bronchus. A closer look at an undamaged specimen reveals the presence of a highly scattering epithelial layer under mucus and debris deposits, which have possibly accumulated there in vitro (Figure 6A and B). The lamina propria is highly vascularized, which could be an indication of inflammation.

View larger version (78K): [in this window] [in a new window]   Figure 5.   In vitro OCT image of human trachea and bronchus (A and B ) and associated histology (C and D, respectively). The mucosal (l ) and cartilage (c) as well as the damage caused from the patient's intubation are visible in the first image. Several glands (g) and a gland neck (gn) can also be identified. The perichondrium (p) and the interterritorial (i ) and territorial matrix of the trachea cartilage are visible without staining. The epithelium (e), lamina propria (l ), and smooth muscle layer (m) are seen in the second image (C ). There is also some debris and mucus deposits (d ) on the surface (image size: 6 × 3 mm × 20 µm).

View larger version (88K): [in this window] [in a new window]   Figure 6.   In vitro OCT image of a secondary bronchus (A) and associated histology (B). Once again, cartilage (c), glands (g), and lamina propria (l ) are noted (image size: 6 × 2 mm × 20 µm).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This work has demonstrated the feasibility of OCT for intraluminal imaging of the upper respiratory tract. In addition to sharp differentiation of the epithelium, subsurface structures were identified, including glands, supportive tissue, and cartilage. The primary focus of this study was to demonstrate the feasibility of OCT imaging on relatively normal tissue of the respiratory tract. Whether OCT can reproducibly identify respiratory pathology both in vitro and in vivo is still to be determined. In particular, future work will need to examine the ability to assess different pathologies, including early neoplastic changes such as dysplasia and metaplasia, targets suggested by the capability of assessing epithelial structure.

Improvements in endoscope design, acquisition rates, and resolution are also necessary to transform the current OCT system into a viable clinical device. The OCT endoscope presently used consists primarily of an optical fiber and light-directing devices (lens, prism), making it relatively inexpensive. It is 1 mm or 2.9 French in diameter, but can likely be engineered to substantially smaller diameters, allowing imaging to be performed in distal portions of the conduit system. Furthermore, imaging is not limited to the airway. OCT endoscopes could be designed and used in conjunction with transbronchial or transcutaneous techniques. In addition, since imaging can be performed though the width of a normal artery, diagnostic imaging of the lung parenychema via the pulmonary artery is a realistic possibility.

The acquisition rate for images in this study ranged from 10 to 30 s per image. This is obviously too slow to prevent artifacts associated with the motion of the respiratory tract. Recently, OCT systems have been developed that can generate images at four to eight frames per second (11). This has allowed in vivo imaging to be performed of internal structures within a rabbit. Furthermore, with future modifications, acquisition rates at or near video rate are likely.

The 6-µm resolution of images used in this study allows imaging of pulmonary microstructure, but it does not allow subcellular imaging to be performed. The ability to identify individual cells and assess subcellular structures such as nuclei would be useful in the assessment of a wide range of pulmonary disorders. Recently, solid-state lasers have been used that allow resolutions of less than 5 µm. Althouth these sources are complex and not yet viable for a clinical instrument, robust sources with similar bandwidth characteristics will likely be available in the near future. Therefore, true subcellular in vivo imaging is a likely possibility of clinical OCT system.

Many applications of this high resolution technology can be envisioned, from the identification of early neoplastic changes, both of the larynx and conduit airways, to the assessment of non-neoplastic disorders such as tracheal malacia and chronic bronchitis. Furthermore, the ability to perform micron scale, real-time imaging will likely be directly extended to transbronchial and transcutaneous techniques, areas where the diagnosis of pathology is often unsuccessful because of limitations in the number of excisional biopsies that can be performed. The resolution of OCT, which is higher than any other clinical imaging technology, makes OCT an attractive technology for diagnostic imaging. The feasibility of OCT for the diagnosis of many disorders, including premalignant epithelial tumors, is further supported by the easy integration with small, relatively inexpensive, endoscopes, low cost, and lack of a need for a transducing medium.