Imaging devices that scan the retinal space have generated new
histologic understanding of the back of the eye as well as of the structural
consequences of several macular pathologies. Latest-generation scanning
modalities employing spectral domain imaging are capable of faster and more
detailed scans and have contributed to greater understanding of macular
Spectral domain optical coherence tomography captures tens of thousands
of individual scans over a 6 mm × 6 mm × 2 mm area of the retina
that are reassembled using a Fourier equation, which is why the modality is
sometimes referred to as Fourier domain OCT. As with other OCT imaging devices,
light is introduced and split, with half entering the eye and half sent to a
reference mirror; the difference between the two beams, as measured by a
spectrometer, yields depth and opacity measurements, which are translated into
an image by computational interpretation.
Spectral domain OCT is differentiated from time domain OCT by a static
reference mirror, which allows higher scanning speed and more images taken in a
single pass. Clinical studies have speculated that spectral domain OCT improves
image acquisition 25 to 100 times over time domain OCT. In other words, with
equal light sources, spectral domain OCT and time domain OCT might produce
similar quality images, but spectral domain OCT will image in seconds what time
domain OCT would take longer to achieve.
To help improve image resolution, device manufacturers have also
introduced light sources with wider bandwidths on spectral domain machines that
effectively increase both axial and transverse resolution. Advanced software
produces histopathological pictures of the back of the eye that were previously
thought impossible to view. These advances, along with the quicker scanning
speeds, are moving imaging closer to obtaining a functional biopsy of the human
eye in situ.
|Carmen A. Puliafito, MD, MBA, dean of the Keck School of
Medicine of the University of Southern California, co-invented the technology
Image: Milici D
“With the segmentation algorithms and the data set that we have, we
can have a much more detailed view of the pigment epithelium, of the internal
limiting membrane, the retinal thickness map,” Carmen A. Puliafito, MD,
MBA, OSN Retina/Vitreous Section Editor, said in a presentation at Hawaiian Eye
2009. “It shows what’s going on in the macula in a very dramatic
Improvements in imaging
The improvement in imaging functionality with spectral domain OCT
produces clearer pictures that penetrate deeper into the retinal architecture
than time domain OCT. The addition of segmentation algorithms and 3-D imaging
in some devices allows output images that show not only the whole picture of
the back of the eye, but also detailed assessment of each part of the imaged
Spectral domain OCT may also deliver a more accurate measurement of
macular thickness. A study by Kakinoki et al reported that macular thickness
was up to 60 µm thicker when measured with spectral domain OCT compared
with time domain OCT.
The greater accuracy with spectral domain OCT has also been seen in the
measurement of foveal thickness. In the October issue of American Journal
of Ophthalmology, Amod Gupta, MD, said that studies by Leung et al and
Forooghian et al show that spectral domain OCT consistently measures a thicker
fovea than time domain OCT.
According to Dr. Gupta, this difference may be explained by the
orientation of the scans: Time domain technology produces an image of the fovea
by six-linear scan lines through the foveal center, but spectral domain
technology measures retinal thickness by an evenly distributed square with the
fovea near the center.
Another reason may be because spectral domain OCT gathers a reading from
the retinal pigment epithelium through the internal limiting membrane, whereas
time domain OCT only measures from the inner/outer segment junction to the
internal limiting membrane – a penetration difference of around 45
Although the clinical implications of macular and foveal thickness
remain uncertain, the deeper penetration into ocular tissue with spectral
domain OCT has direct clinical utility. One such application might be in
capturing macular hemorrhaging. While time domain OCT would also theoretically
show fluid and blood on the macula, it would not be able to depict with the
same accuracy at what retinal layer the buildup occurs.
According to Dr. Puliafito, the more precise measurements with spectral
domain OCT might be particularly useful in managing diabetic retinopathy,
especially in its proliferative form, and assist in clinical decision making.
“We can evaluate underlying pathology and monitor disease response,
retinal thickness, cystoid macular edema, vitreoretinal traction and track
subretinal fluid,” he said.
Spectral domain OCT is being used more often to visualize tractional
retinal detachment in proliferative diabetic retinopathy, Dr. Puliafito said.
Each vitreomacular traction detachment is unique, and the segmented images and
precise registration with spectral domain OCT allow a picture of not only the
retinal detachment, but also exactly where it is occurring.
The capabilities of spectral domain technology may make it a useful tool
for postoperative evaluation as well. Segmentation can show if and where the
retinal layers are reattaching, and because of point-to-point registration,
reattachment can be tracked over time over the course of multiple visits.
Spectral domain OCT is also capable of imaging through silicone oil.
“The level of precision for clinical decision making is
better” than what is possible with time domain OCT, Dr. Puliafito said.
“Spectral domain OCT offers a quantitative way of looking at clinically
significant macular edema, to identify its geography specifically in a
comprehensive way and to monitor the results of treatment.”
Faster scanning, more sensitivity
If there is a drawback of spectral domain OCT, it is that the image is
not centered exactly over the fovea, and so it is actually less sensitive to
discrete foveal pathology than time domain OCT. However, that deficiency is
compensated for by capturing images of a larger section of the retinal space.
“As retinal specialists, we know that what is going on in the fovea
is important for visual acuity, but everything else in that 6 mm × 6 mm
area is quite relevant,” Dr. Puliafito said.
Spectral domain OCT captures a picture of a larger space because it
scans at a faster speed, capturing tens of thousands of images in seconds over
a larger surface area than time domain OCT. And there are distinct advantages
to that increased speed. For instance, because spectral domain OCT takes
multiple pictures of the same spot, the software in the device can average out
the resulting image.
Peter K. Kaiser
“If you have multiple cuts through the same spot, anything that is
noise can be eliminated, and that makes the image look better,” Peter K.
Kaiser, MD, OSN Retina/Vitreous Board Member, said.
Time domain OCT could, theoretically, capture an image the same size as
spectral domain OCT, but it would take longer to capture an inferior image and
patients would have to remain still for a longer amount of time. From a
practice management perspective, scanning the peripheral retina with time
domain OCT could cause a bottleneck of patients.
From a clinical perspective, faster and more thorough imaging of the eye
means less information is interpolated, and therefore, results are less
variable. A better light source allows spectral domain OCT to image the
vitreous, for instance, and display anatomy down to a histopathological level.
This is an important consideration given the increased awareness of the role of
a healthy inner segment/outer segment junction and photoreceptor cells in
visual recovery after intervention for epiretinal membrane, macular edema and
other retinal-detaching pathologies.
“We are able to get a better view of areas of the retina that we
could not see before that are important to determine vision,” Dr. Kaiser
Spectral domain OCT is not only faster, but it is also more accurate.
Dr. Kaiser said that a study at his institution showed “in a relatively
large number of patients, spectral domain picked up leakage that was not
present on time domain, and so those patients needed anti-VEGF therapy that we
would not have given with time domain,” Dr. Kaiser said. “In other
words, fluid was there, but we could not see it with time domain.”
Role in AMD
Clinical trials looking at OCT-guided therapy have challenged popular
notions about the optimal anti-VEGF treatment regimen in AMD. In particular,
the PrONTO (Prospective OCT imaging of patients with neovascular AMD treated
with intraocular ranibizumab) trial introduced a variable treatment paradigm
that originated from a suggested sustained effect of treatment that was
observed in phase 3 and phase 3 extension trials.
In the PrONTO trial, patients were given a 3-month run-in with Lucentis
(ranibizumab, Genentech). Re-treatment was offered only if certain criteria
were met, including several factors based on OCT findings. At the end of 2
years, visual acuity gains were similar to those observed in the drug’s
phase 3 ANCHOR (Anti-VEGF antibody for the treatment of predominantly classic
choroidal neovascularization in AMD) and MARINA (Minimally classic/occult trial
of the anti-VEGF antibody ranibizumab in the treatment of neovascular AMD)
trials, but with fewer injections: about five per year compared with the
typical 12 that would be performed with monthly dosing.
Srinivas R. Sadda
“It would be one thing if the treatment was relatively trivial for
the patient to endure, for example, if it was just an eye drop to be instilled
once a day,” Srinivas R. Sadda, MD, said. “But to have to come into
the office, have your family take time off to bring you in, have to endure an
injection and bear the potential risks of the injection — that is a
The incorporation of spectral domain OCT into clinical practice proffers
to improve decision-making ability in AMD patients. Spectral domain OCT,
because it more densely samples the macula, will be more likely to image fluid
in its nascent stages, meaning treatment can be introduced early and before it
causes limitations to vision.
According to Dr. Sadda, currently available spectral domain machines
provide so much data that clinicians may be overloaded with information. The
segmentation algorithms for AMD are not as advanced as they are for pathologies
such as diabetic macular edema, he said, so relying exclusively on the
automated OCT measurements may not be advisable.
Still, the abundance of data may be a good problem to have. According to
a study he performed at his institution, Dr. Sadda said that spectral domain
OCT was more sensitive than time domain OCT for identifying factors that would
drive re-treatment with anti-VEGF therapy.
“Spectral domain would have changed our management decision at
least 10% of the time,” Dr. Sadda said.
The clearer pictures and faster scans possible with spectral domain OCT
make the device a valuable research tool, and in fact, the laboratory is where
spectral domain OCT first proved useful. Over time, however, the technology has
been adopted into clinical practice.
“It has become integral to clinical practice because it gives us a
quantitative, reproducible cross-section of macular pathology that is
essentially unattainable with other ancillary tests,” Jay S. Duker, MD,
OSN Retina/Vitreous Board Member, said.
For example, spectral domain OCT has proven particularly useful in
delineating fine vitreous structures and defining the pathogenesis of macular
hole, he said.
Research is under way on the next generation of imaging technology. Dr.
Duker said that his lab will soon receive a prototype ultrahigh-resolution
machine capable of resolution of 1 µm and 100,000 scans per second; by
comparison, most spectral domain machines produce images at around 4 µm
and take around 25,000 to 40,000 scans per second. The clinical and research
applications of next-generation OCT is still uncertain, but these kinds of
devices may be capable of imaging individual photoreceptor cells on the surface
of the retina.
Through media opacities
Back of the eye research with spectral domain OCT will continue to be
important as investigators attempt to close the gap on structure-function
incongruence in pathologies that alter the physical structure of the macula.
“For the diseases in the macula that we are using [spectral domain
OCT] for now, like diabetic macular edema and age-related macular degeneration,
we still have a lot to learn about following these patients with OCT,” Dr.
Another potential application of spectral domain OCT is in following
macular changes in patients with uveitis. A recent study showed that spectral
domain technology was superior to time domain technology in imaging through
vitreous haze secondary to uveitis.
According to Dr. Gupta, a co-investigator in the study, spectral domain
OCT identified three patients with epiretinal membrane that time domain OCT
missed, and so vitreous surgery was initiated. Likewise, spectral domain OCT
also identified epiretinal membrane in a patient who was thought to have only
macular edema after a time domain scan.
In the study, spectral domain OCT also proved useful in mitigating risk:
Spectral domain images showed no macular edema in some cases that would be
considered edema suspects after time domain OCT was incapable of producing a
“Time domain images could be interpreted only in media grade 1, ie,
clear media. Even minimal haze such as in grade 2 prevented acquisition of any
interpretable images,” Dr. Gupta said.
Compared with time domain OCT, spectral domain OCT “samples across
a defect or opacity with approximately 13 times more number of scans, and thus,
any opacity would have much less impact on the quality of image,” he said.
“Current high-definition OCTs provide useful information on the
impact of inflammation on the macula that includes cystoid macular edema,
epiretinal membrane or the subretinal neovascular membrane, each of which
requires different treatment strategy. Further, OCT may show irreversible
pathology such as foveal atrophy,” he said.
Imaging through inflammation secondary to uveitis may be critical not
only in delivering interventions in cases that otherwise might be missed, but
also in lowering exposure to risk from unnecessary steroid injections. –
by Bryan Bechtel
How much importance do
you put on spectral domain OCT findings in comparison to other diagnostic
modalities and clinical findings?
- Brown DM, Regillo CD. Anti-VEGF agents in the treatment of
neovascular age-related macular degeneration: Applying clinical trial results
to the treatment of everyday patients. Am J Ophthalmol.
- Forooghian F, Cukras C, Meyerle CB, Chew EY, Wong WT. Evaluation of
time domain and spectral domain optical coherence tomography in the measurement
of diabetic macular edema. Invest Ophthalmol Vis Sci.
- Fung AE, Lalawani GA, Rosenfeld PJ, et al. An optical coherence
tomography-guided, variable dosing regimen with intravitreal ranibizumab
(Lucentis) for neovascular age-related macular degeneration. Am J
- Gupta V, Gupta P, Singh R, Dogra MR, Gupta A. Spectral-domain
Cirrus high-definition optical coherence tomography is better than time-domain
stratus optical coherence tomography for evaluation of macular pathologic
features in uveitis. Am J Ophthalmol. 2008;145(6):1018-1022.
- Kakinoki M, Sawada O, Sawada T, Kawamura H, Ohji M. Comparison of
macular thickness between Cirrus HD-OCT and Stratus OCT. Ophthalmic Surg
Lasers Imaging. 2009;40(2):135-140.
- Leung CK, Cheung CY, Weinreb RN, et al. Comparison of macular
thickness measurements between time domain and spectral domain optical
coherence tomography. Invest Ophthalmol Vis Sci.
- Puliafito CA. A brief history of optical coherence tomography: a
personal perspective. Ophthalmic Surg Lasers Imaging.
- Roesel M, Henschel A, Heinz C, Spital G, Heiligenhaus A.
Time-domain and spectral-domain optical coherence tomography in uveitic macular
edema. Am J Ophthalmol. 2008;146(4):626-627; author reply 627-628.
- Jay S. Duker, MD, can be reached at New England Eye Center, 800
Washington St., Box 450, Boston, MA 02111-1533; 617-636-4604; fax:
617-636-4866; e-mail: email@example.com.
- Amod Gupta, MD, can be reached at Department of Ophthalmology, Post
Graduate Institute of Medical Education and Research, Sector 12, Chandigarh,
India 160012; e-mail: firstname.lastname@example.org.
- Peter K. Kaiser, MD, can be reached at Cole Eye Institute, Division
of Ophthalmology, A31, 9500 Euclid Ave., Cleveland, OH 44195; 216-444-6702;
- Carmen A. Puliafito, MD, MBA, can be reached at Office of the Dean,
Keck School of Medicine, University of Southern California, 1975 Zonal Ave.,
KAM 500, Los Angeles, CA 90033; 323-442-1900; e-mail:
- Srinivas R. Sadda, MD, can be reached at Keck School of Medicine,
University of Southern California, 10 Congress St., Pasadena, CA 91105;
626-395-0777; e-mail: email@example.com.