It is generally accepted that structural damage precedes functional change in glaucoma. This is supported by the observations that 25-40% of optic nerve fibers can be lost while still retaining a normal visual field. This has led to the development of nerve fiber layer analyzers and optic nerve head topographers to aid in the early diagnosis and detection of progression for glaucoma.
Optical coherence tomography (OCT) is one of the principle technologies used for this purpose. The principle of OCT is analogous to that of ultrasound, with the reflection of waves (light instead of sound) from tissue structures. The reflectance pattern is analyzed and the delay in reflected signals measured and converted to depth information. An image is achieved by scanning the wave laterally and combining a series of axial scans.
Zeiss OCT1 debuted at 100 axial scans per second. The Zeiss Stratus quadrupled the speed in 2002. 400 axial scans per second was sufficient to make OCT a standard for the diagnosis of many retinal diseases and glaucoma. The new generation of Fourier-domain technology enables the collection of data at 26,000 axial scans per second, a 65-fold advance over the “time-domain” OCT. The FD-OCT scanners also offer a 2-fold advance in resolution to 5 micron, close to the “ultrahigh-resolution” level. There are currently at least seven FD-OCT scanners on the market, including models from Optovue, Topcon, Heidelberg, Zeiss, OTI, Optopol and Bioptigen.
What is the difference between the older OCT technology and FD-OCT? In time domain OCT, a mirror in the reference arm of the interferometer is moved to match the delay in various layers of the sample. The resulting interference signal is processed to give the axial scan data. The reference mirror must move one cycle for each axial scan. The need for mechanical movement limits the speed of image acquisition. In Fourier domain OCT, the reference mirror is kept stationary. The interference between the sample and reference reflections is split into a spectrum and captured by a line camera. The spectral interferogram is Fourier transformed to provide an axial scan. The absence of moving parts allows the image to be acquired very rapidly. FD-OCT can capture 2000 pixels simultaneously, while TD-OCT captures one pixel at a time. So in the time it take TD-OCT to form one single axial scan, FD-OCT can capture an entire image. The higher speed and resolution of FD-OCT allows higher definition, or more pixels per image. Because the FD-OCT picture is captured in a small fraction of a second, there is no motion artifact that is commonly seen in conventional OCT images.
Anterior Segment Imaging
The current generation anterior segment OCT (ASOCT) is the Zeiss Visante system. It utilizes a 1.3 micron wavelength scanning laser, with 2000 axial scans per second enabling up to 8 microns resolution. Corneal applications include pachymetry mapping, refractive surgery flap and stromal bed measurements, and anterior chamber biometry. Glaucoma applications include the assessment of narrow angles and post surgical changes in angle surgery. The angle evaluation compares favorably with gonioscopy. While gonioscopy is quick and useful in the clinical setting and allows for compression gonioscopy, it also requires light and is therefore prone to artificial opening of the angle due to pupillary dilation as well as indentation errors. ASOCT, however, requires less expertise, is non-contact, and does not use visible light. The latter feature allows for mydriatic provocative testing of a narrow angle. It is currently being evaluated as a screening parameter to help identify patients at risk for angle closure.
ASOCT allows evaluation of the angle and anterior segment through corneal opacities. We have published data showing the ability to assess tube patency and position, and IOL position in glaucoma patients with opaque corneas, thereby helping to guide diagnosis and surgical approach. One disadvantage is the inability to image structures posterior to the iris, such as the ciliary sulcus or ciliary processes.
The only FD-OCT model currently equipped with anterior segment imaging is the RTVue by Optovue. The FD-OCT system has higher resolution (5 microns), allowing for detection of small anatomic details such as trabecular meshwork, and Schlemm’s canal. This does come at the cost of poorer penetration through sclera and limbus, leading to the loss of detail in the peripheral angle (such as iris root insertion) in some cases.
Macular Imaging
Imaging of retinal thickness in the macula is relatively insensitive for glaucoma diagnosis, with power less than circumpapillary nerve fiber layer thickness. There are two main limitations to using macular thickness measurements. The first is an under-sampling of this tissue using the standard TD-OCT with six radial line scans. The second is that glaucoma preferentially affects certain layers of the retina, not the entire macular retinal thickness. Our group at the Doheny Eye Institute, led by David Huang, has developed a macular scanning pattern with FD-OCT that utilizes a 5mm grid composed of 34 lines and over 19,000 axial scans. In addition, we are able to employ retinal layer segmentation and measure the layers most affected by glaucoma. This is comprised of the nerve fiber layer, ganglion cell layer and inner plexiform layer, and is termed the Ganglion Cell Complex (GCC). Thickness of the macular GCC can be plotted as a reduced average thickness, fractional loss map, and fractional deviation map, which shows percentage thinning compared to an average normal retina. Preliminary results from the Advanced Imaging in Glaucoma Study (AIGS) suggests that macular GCC can increase the power to detect glaucoma when used alone or in combination with circumpapillary nerve fiber layer measurements.
Retinal Blood Flow
Interest in measuring retinal blood flow stems from the observation that poor perfusion to the optic nerve head and retina may be a factor for glaucoma progression. Current technologies have limitations in this measurement, however. Fluorescein angiography cannot measure volume flow. Ultrasound does not have enough resolution to see small retina vessels. Laser Doppler flowmetry does not accurately measure flow in a vessel because it cannot directly measure the flow profile and relies on the assumed relationship between maximum Doppler shift and the average flow. Further more, the probe beam and scan direction must be individually aligned with each retinal vessel and thus it is not possible to quickly measure the total retinal blood flow.
In Doppler OCT, the probe beam is back scattered from flowing blood in a vessel at an incidence angle alpha. The scattered light has a Doppler frequency shift delta nu that is proportional to the total flow velocity and the cosine of the angle alpha. Thus, to measure the flow velocity, one needs to measure both the Doppler shift and the relative angle between the OCT beam and the blood vessel. To determine the angle between the scanning beam and blood flow, we developed a dual scanning plane method. The blood vessel is imaged by 2 parallel cross-sectional scans spaced a small distance apart along the length of the blood vessel. The relative position of the blood vessel in the 2 scan planes determines the direction of flow relative to the probe beam. Integration of the velocity profile over the vessel area provides the flow measurement. We hope that these retinal blood flow measurements will help our understanding of the role of perfusion of the optic nerve and retina in glaucoma. It will also enable the measurement of medical and surgical treatments on retinal blood flow.
In summary, FD-OCT provides information that is unique in imaging. Many glaucoma parameters are measureable with FD-OCT, including circumpapillary nerve fiber layer, optic nerve topography, macular ganglion cell complex thickness, anterior segment, and retinal blood flow. The summation and combination of these parameters may lead to greater sensitivity and specificity of glaucoma diagnosis and detection of progression. Retinal blood flow measurements may lead to new diagnostic parameters, theories of glaucoma causation, and guide therapeutics.
Special thanks to David Huang, MD, PhD, Vikas Chopra MD, Farnaz Memarzadeh, MD and Rohit Varma, MD, MPH
Have a question or comment on this article? Use the “Comment” link below to leave your thoughts, and the author will respond.