Introduction

Over the years orthopedic surgeons have used modalities such as plain radiography, magnetic resonance imaging (MRI), and ultrasound to assess joints and other musculoskeletal structures. Each of these imaging modalities has its own advantages; however, there are still instances when these technologies do not possess adequate resolution to effectively assess the relevant pathology. An example is the current inability to assess cartilage during treatment for osteoarthritis.

We have developed a new technology, optical coherence tomography (OCT), for the assessment of articular cartilage, tendons, and ligaments. OCT is analogous to ultrasound, but measures the intensity of backreflected infrared light rather than sound^1-3. These efforts in orthopedic imaging have received several awards, including the Presidential Award in Science and Engineering from President Clinton in 1998.

OCT has several advantages for the assessment of musculoskeletal pathology. First, OCT has a resolution of 10 – 25 times that found in other clinical imaging technologies. Laboratory-based state-of-the-art OCT systems have attained resolutions as high as 4 µm⁴. Second, OCT has a faster speed of acquisition⁵. OCT can image with an acquisition rate of up to 16 frames per second, which could allow this technology to image surgical procedures in near real time. Third, since OCT is based on fiber optics, imaging instruments utilizing OCT technology can be built with cross-sectional diameters as small as 0.014 inches⁶. This opens the potential of designing OCT catheters to be incorporated into arthroscopic instruments or bedside needle-based devices. Fourth, the entire unit is compact, similar in size to an ultrasound unit, and can be readily transported into a surgical ward or clinic. Finally, since OCT is based on optics, it can be combined with other spectroscopic techniques to assess the optical and biochemical aspects of the tissue being imaged.

Technical Aspects of OCT

The details of OCT have been previously described^1-3. As stated, OCT is analogous to B-mode ultrasound, measuring the backreflection of near-infrared light rather than sound waves. Due to the high speed of light, the echo delay time cannot be measured electronically (as it is with ultrasound) and therefore OCT relies on a technique known as low coherence interferometry. Figure 1 depicts a schematic of a general OCT system and illustrates the principle of low coherence interferometry. The broad bandwidth light, which can be thought of as a series of pulses, is split into two separate arms, referred to the reference and sample arms. Light that passes down the reference arm is reflected back from a movable mirror. The sample arm directs the light toward the tissue being imaged. Once the light reaches the tissue it can be absorbed or scattered. Light backreflected from the tissue will ultimately be recombined with the light from the reference arm at the beam splitter. If the light has traveled the same path length in both arms, to within the coherence length (or in the context of our analogy, the pulse length), interference will occur when the light is recombined at the beam splitter. Therefore, OCT measures the intensity of this interference and uses it to represent backreflection within tissue. The beam in the sample arm scans the tissue to generate two- and three-dimensional images.

The resolution of OCT is dependent upon the bandwidth of the source (range of wavelengths within the beam). The wider the bandwidth, the greater the resolution. OCT is essentially an "optical biopsy" technique, allowing resolutions close to 2-10X that of microscopy, with a penetration depth slightly greater than a mechanical biopsy of approximately 3mm.

Imaging Osteoarthritic Cartilage

Recent research has indicated that the progression of osteoarthritis (OA) may be delayed or halted if treatment is initiated early in the course of the disease. In order to facilitate the treatment of OA, it will be necessary to image the articular cartilage at higher resolutions than are currently available. Unfortunately, the limited resolution of current imaging technologies does not allow the accurate monitoring of early cartilagenous changes. While MRI is effective for the macroscopic assessment of the joint, resolutions of current clinical devices are between 250 – 300 µm. With this limited resolution, it is difficult to detect fine changes in the articular cartilage. Furthermore, the high cost of MRI would be a limiting factor in its successful implementation as a screening tool. Specifically, it would be impractical to use this technology at many time points, particularly if multiple joints are involved.

Another powerful technique is arthroscopy. In addition to its ability to facilitate joint repair, arthroscopy allows for direct inspection of the surface of the cartilage and ligaments. However, both the inability to image below the cartilage surface and the expense prevent its use as a routine screening procedure.

OCT Imaging of Cartilage

Our group has investigated the capability of OCT to image the musculoskeletal system, with an emphasis on articular cartilage². In Figure 2, an OCT image of normal cartilage appears on the upper left. On the lower left, an image of the corresponding normal histology is shown, where "c" is cartilage and the arrow indicates the bone-cartilage interface. Due to the high resolution of OCT, the cartilage thickness in this image can be measured to within 10µm. On the upper right in Figure 2, an OCT image of the diseased cartilage of an osteoarthritic femoral head is shown. In this image, cartilage thinning is seen on the left side. In addition, a fibrous band (f) has developed on the surface , and disruption of the bone-cartilage interface has occurred (nb). On the lower right is the corresponding histology. This study has also revealed that OCT is capable of detecting articular cartilage defects such as microfibrillations and fibrosis².

Among the earliest changes in OA is the breakdown of collagen. We have developed OCT with polarization sensitivity (PS-OCT) to identify organized collagen⁷. In normal cartilage, the image changes with alterations in the polarization state of the incident light. Figure 3 compares images obtained from normal and OA cartilage with respect to polarization sensitivity⁷. On the upper left of the figure, OCT images (a,b,c) of normal cartilage are seen. In these images a smooth banding pattern is present that changes with the polarization state. It is important to note that these bands do not correspond to any specific structure, but rather arise from the birefringent (polarization) properties within cartilage. This birefrigence is due to the highly organized nature of collagen within healthy articular cartilage, which behaves like a polarization filter. The fourth image (d) is the corresponding histologic section stained with picrosirius, a specialized staining technique where increased brightness demonstrates highly organized collagen. In the three images to the right of the figure, the cartilage is thick but with evidence of disease (e,f,g). It can be seen that there are essentially no changes with the polarization state of the incident light in the OCT images. In the picrosirius stained histology (h), there is a dramatic attenuation in the brightness, representing a reduction in the organization of the collagen network. These results indicate that OCT can detect early degenerative changes in articular cartilage, before cartilage thinning and fibrillations occur.

In vivo studies using a hand-held probe during open knee surgery have also been completed⁸. Results similar to previous in vitro studies were obtained, indicating that OCT can be used during surgical procedures to assess the extent of articular cartilage damage. Currently we are designing a probe that could be used during arthroscopic procedures.

Animal Models of Osteoarthritis

The development of new therapeutics will require the use of animal models to assess the progression of osteoarthritis. Currently, most studies utilizing animals require a large number of animals, since animals need to be sacrificed at different time points. By incorporating OCT as a means of assessing cartilage, we have developed rat and rabbit models for sequentially following joint cartilage properties without the need for animal sacrifice⁹. These studies have been performed in both chemically and mechanically induced models of arthritis. We believe that the rat model may be the model of choice for ultimately assessing many therapeutic approaches by the research community. The benefits of this model include a reduction in costs, avoidance of difficulties in data analysis due to differences in the heterogeneity present within a population, and the use of smaller amounts of novel therapeutics (often available only in small quantities).

Assessing the Microstructure of Tendons and Ligaments

Abnormalities of tendons and ligaments can lead to significant morbidity. Examples include injuries of the Achilles tendon, anterior cruciate ligament (ACL), or patellar tendon. While there are many technologies capable of assessing tendons and ligaments with above 200 µm resolution, there are instances where a higher resolution would be of value. Since these tissues are composed of highly organized networks of collagen fibers, they also display birefringence, which allows for assessment by OCT. Any alteration in the normal organized arrangement of collagen fibers should result in an attenuation of the birefringent properties of the tissue. We have imaged both normal and diseased tendons and ligaments to determine if OCT can monitor changes in these properties. Figures 4a-4c depict an area of an ACL with no evidence of injury or disease imaged at different polarization states¹⁰. Due to the birefringence of the tissue, a banding pattern is present similar to that seen within normal cartilage. As the polarization state is altered, the position of the bands is shifted. This is consistent with the picrosirius image in figure 4d, demonstrating organized collagen. In contrast, OCT images from a section of disrupted ACL (Figures 4f-h) do not show the birefrigence or clear banding pattern. Future work is needed to determine if polarization-sensitive changes detected by OCT can indicate areas susceptible to injury or if these techniques can help to determine the etiology of patient discomfort, such as in Achilles tendinosis.

Other Work

In addition to the projects described above, our group also has several other ongoing projects. One clinically important focus is on developing OCT to assist in guiding small nerve and vessel repair in trauma and microsurgical flap reconstruction¹¹. In particular, these applications may allow one to distinguish between sensory and motor fibers in peripheral nerves. We also are investigating the utilization of OCT in the guidance of laser cartilage repair. Other basic work focuses on such technical issues as analysis of dispersion¹², reduction of system noise levels, absorption spectroscopy, improving our understanding of birefringence of collagen with SEM, and automated high speed quantification of cartilage thickness^13,14.

Conclusion

OCT represents a promising new technology for the assessment of the musculoskeletal system. In particular, the most important application will likely be the assessment of cartilage and the monitoring of its changes during therapeutic intervention.

Acknowledgements

Dr. Brezinski is currently funded by NIH-RO1-AR44812, NIH R01 AR46996, NIH R01- HL63953, NIH-1-R01-HL55686, and NIH R01 EB000419. Previous funding has also included the Whitaker Foundation, the Air Force Office of Scientific Research Contract F4920-98-1-0139, and the Navy. We would also like to acknowledge our ongoing collaborations with the Fujimoto group at MIT and King's College. In addition, we would like to recognize all the efforts of technicians, postdoctoral fellows, students, and collaborators in this work. Dr. Brezinski has sold his interest in Lightlab Inc.

Notes:

Debra L. Stamper PhD is an Instructor in Orthopedic Surgery at Harvard MedicalSchool and Scientist at Brigham and Women's Hospital.

Scott D. Martin MD is an Assistant Professor in Orthopedic Surgery at Harvard Medical School and Attending Physician, Brigham and Women's Hospital.

Mark E. Brezinski MD PhD is an Associate Professor in Orthopedic Surgery at Harvard Medical School and Senior Scientist at Brigham and Women's Hospital.

Correspondence should be addressed to:
Mark E. Brezinski, MD PhD
Brigham and Women's Hospital
Department of Orthopedic Surgery
75 Francis Street
Boston, MA 02115
(617) 525-6738
(617) 732-6705 (fax)

References:

1. Huang D, Swanson EA, Lin CP, Schuman JS, Stinson WG, Chang W, et al. Optical coherence tomography. Science. 199; 254(5035):1178-81.
2. Herrmann JM, Pitris C, Bouma BE, Boppart SA, Jesser CA, Stamper DL, et al. High resolution imaging of normal and osteoarthritic cartilage with optical coherence tomography. J Rheumatol. 1999; 26:627-35.
3. Tearney GJ, Brezinski ME, Boppart SA, Bouma BE, Weissman N, Southern JF, et al. Images in cardiovascular medicine. Catheter-based optical imaging of a human coronary artery. Circulation. 1996; 94:3013.
4. Boppart SA, Bouma BE, Pitris C, Southern JF, Brezinski ME, Fujimoto JF. In vivo subcellular optical coherence tomography imaging in Xenopus laevis: implicatications for the early diagnosis of neoplasms. Nat Med. 1998, 4:861-865.
5. Tearney GJ, Brezinski ME, Bouma BE, Boppart SA, Pitris C, Southern JF, Fujimoto JG. In vivo endoscopic optical biopsy with optical coherence tomography. Science. 1997, 276:2037-2039.
6. Lightlab Inc (Westford, Mass.). Available at: Lightlabimaging.com.
7. Drexler W, Stamper D, Jesser C, Li X, Pitris C, Saunders K, et al.. Correlation of collagen organization with polarization sensitive imaging of in vitro cartilage: implications for osteoarthritis. J Rheumatol. 2001;28:1311-8.
8. Li X, Martin S, Pitris C, Ghanta R, Spector M, Fujimoto J, Brezinski M. High-resolution optical coherence tomographic imaging of osteoarthritic cartilage during open knee surgery. Submitted.
9. Adams SB, Herz P, Roberts M, Bourquin S, Patel N, Hsiung P, et al. In vivo imaging of cartilage degeneration in the rat using optical coherence tomography. In prep.
10. Martin SD, Patel NA, Adams SB, Roberts MJ, Plummer S, Stamper DL, et al. New technology for assessing microstructural components of tendons and ligaments. Int Ortho. In press.
11. Brezinski ME, Tearney GJ, Boppart SA, Swanson EA, Southern JF, Fujimoto JG. Optical biopsy with optical coherence tomography: feasibility for surgical diagnostics. J Surg Res. 1997 Jul 15;71(1):32-40.
12. Niblack WK, Schenk JO, Liu B, Brezinski ME. Dispersion in grating based optical delay line for OCT. Applied Optics. In press.
13. Rogowska J, Bryant CM, Brezinski ME. Cartilage thickness measurements from optical coherence tomography. J Opt Soc Am A Opt Image Sci Vis. 2003; 20:357-67.14. Rogowska J, Brezinski ME. Image processing techniques for noise removal, enhancement and segmentation of cartilage OCT images. Phys Med Biol. 2002, 47:641-655.