Special Section Sponsored By CZM

Exploring Advances in OCT Imaging

Applying this emerging technology to glaucoma and retina care

By Mark T. Dunbar, OD, FAAO

The days of looking into a patient's eye with a direct ophthalmoscope and straining to discern the details of pathology will soon be behind us. With the growing prominence of OCT technology, we can clearly see and evaluate 3-D image changes at the foveal level — 2 to 3 µm in size — after a scan that takes only a few seconds to acquire.

As we embrace OCT, it's important that we understand all of the capabilities it has to offer.

CHOICES

The biggest change in OCT has occurred in scanning speed. Time-domain OCT, which uses 500 axial scans per second to achieve a resolution of 10 µm, was once at the vanguard of the technology. But now we can take advantage of spectral domain OCT (SD-OCT), which uses between 27,000 and 40,000 axial scans per second to achieve a resolution of 5µm.

With its latest software enhancements, SD-OCT provides a high-resolution evaluation of retinal anatomy, facilitating diagnoses of macular conditions that are difficult to establish with biomicroscopy. The technology also provides a quantitative assessment of retinal anatomic alterations and the vitreoretinal interface, allowing us to objectively monitor disease progression and therapeutic response.

OCT USE IN GLAUCOMA

The most common application of OCT for most of us is in daily glaucoma care. IOP monitoring, clinical observation of the optic nerve and interpretation of visual fields are no longer enough to treat glaucoma patients. OCT provides crucial information to help us determine if a patient has glaucoma, if it's progressing or if therapy should be modified.

Consider the case of a 46-year-old female with pressures of 19 mm Hg and 18 mm Hg. We conducted a same-day serial analysis with the Stratus time-domain OCT, including four scans while the patient remained in the chair. We printed the scans and plotted them on the Stratus' serial analysis program, which confirmed reliability. In both eyes, I saw an obvious nerve fiber layer defect inferotemporally, including congruity between inferior and superior notching and a nasal visual field defect. The patient had obvious fiber layer thinning, most significantly in the left eye. We treated her accordingly.

Another patient, a 61-year-old female, was struggling with visual fields, losing fixation on 13 of 15 attempts. An inferotemporal view of the nerve showed what appeared to be a nerve fiber layer defect, inferiortemporally. In years past, her pressures had ranged from 16 mm Hg to 19 mm Hg. Her mother also had a history of glaucoma.

Her Stratus OCT (Carl Zeiss Meditec, Inc.) images looked good. We plotted four scans taken the same day, finding minimal variability in the serial analysis. Two years later, after following her periodically, we saw variability that corresponded with our clinical findings, raising suspicions, especially when we considered her family history. We treated her conservatively.

A patient such as this would benefit from the advanced optic nerve head analysis software of Cirrus HD-OCT (Carl Zeiss Meditec, Inc.), which now calculates the vertical cup-to-disc ratio and rim area. The latest software provides a normative database based on the retinal nerve fiber layer (RNFL) symmetry, rim area, disc area and average cup-to-disc ratio (Figure 1).

FIGURE 1. Retinal thickness measurements are shown taken by the Cirrus HD-OCT.

DIAGNOSING GLAUCOMA WITH OCT

One study, which evaluated 73 eyes with glaucoma and 146 age-matched normals, sought to determine if we could diagnose glaucoma with SD-OCT.¹ Researchers found the Cirrus unit was able to discriminate between normal eyes and eyes with glaucoma or even mild glaucoma by using optic nerve head parameters. There was no difference in the ability of the OCT's parameters and RNFL thickness measurements to distinguish between normal and glaucomatous eyes. So, one could argue that you can diagnose or rule out glaucoma with OCT.

This case demonstrates what is possible. We followed a 62-year-old African-American male as a glaucoma suspect for nearly 10 years. He was receiving latanoprost (Xalatan, Pfizer) and brimonidine (Alphagan, Allergan), and his pressures were 12 mm Hg in both eyes. His right optic nerve showed a deep, vertically long cup, different from the nerve in the left eye, which also revealed signs of glaucomatous damage. Through the years, however, he had tested with normal fields.

A Cirrus scan was normal. Knowing that the RNFL was normal, compared to a normative database, I discontinued his medications and followed him closely to make sure he didn't have any IOP spikes. Since then, he has not progressed at all.

Another patient had nonproliferative diabetic retinopathy and poorly controlled blood sugar for 10 years. In his right eye, I saw a large cup and inferior thinning. His pressures were 21 mm Hg and 19 mm Hg. Imaging showed that he had inferotemporal thinning. His visual fields weren't bad, but his total deviation was a little worse in his right field. Clinical examination showed a superior arcuate defect.

OCT detected an inferonasal nerve fiber layer defect. As demonstrated by this case, the vertical average cup-to-disc ratio is very helpful as you image the patient 3, 4 and 12 months out. OCT helps you determine if the patient is getting worse. Here is where the normative database of the Cirrus software can help correlate OCT findings with cup-to-disc ratios.

PRECONCEIVED NOTIONS

In some cases, OCT can keep us from getting trapped by preconceived notions. Case in point: A 44-year-old Hispanic female's pressures ranged from 18 mm Hg to 24 mm Hg over a number of years. Her central corneal thickness, at 610 µm, was thicker than the average thickness of 550 µm. Cirrus, Stratus and Heidelberg Spectralis scans all produced normal images. The cup-to-disc ratio was 0.45 to 0.5.

From the Ocular Hypertension Treatment Study, we knew that the 5-year risk of developing glaucoma decreased to 13% if the cornea was thin and the cup was small.² Despite this patient's low risk, however, an OCT scan finally detected a nerve fiber layer defect. A comparison of old and new scans determined that significant thinning of the RNFL had occurred over 2 years, a finding that was also reflected on the patient's visual fields.

The newer software from Cirrus will automatically make this determination on one printout, plotting probability and rate of progression based on its glaucoma progression analysis software. This feature saves us from the more time-consuming and error-prone task of seeking these trends manually. We also obtain a view of the optic disc in the OCT analysis, which will essentially accomplish the same goal.

Ganglion cell analysis with a normative database and Enhanced Depth Imaging (EDI) just released by ZEISS will help further improve our monitoring and follow-up abilities.

Continuing advances such as these may some day make aspects of our clinical exams obsolete. For example, stereoscopic photographs and Cirrus HD-OCT images of the optic disc, cup, and neural rim were compared in a study of 23 glaucomatous eyes and seven suspect eyes at Bascom Palmer in Miami.³

Designation of the cup and optic disc boundaries by the Cirrus was found to be within the range of variable designations by human readers of color stereoscopic photographs. Cirrus measurements demonstrated better repeatability than the human readers, suggesting a promising use in the diagnosis and monitoring of progression.

IMPROVING RETINA CARE

As is the case with glaucoma, OCT can elevate our treatment of patients requiring retina care. Here is a case that demonstrates my point.

A 57-year-old female patient came to me with symptoms of poor vision associated with a long-term amblyopic right eye. She had a posterior staphyloma, even though she wasn't myopic.

The Stratus revealed a small amount of macular schisis in each eye. When scanning her with SDOCT, I could see the higher-resolution image of the macula sitting on a bit of a hill, which explained her poor vision. This tendency is well reported in patients with high myopia.⁴

In the patient's left eye, the Stratus showed an outer retinal defect on oblique section, however it wasn't until I did a vertical slice with the Stratus that I was able to see a full thickness defect. She had lost photoreceptors and had developed a full thickness macular hole, stage III. Of course this was immediately seen with the Cirrus HD-OCT. What impressed me was that I was still able to make the diagnosis with the Stratus, although the Cirrus clearly did a better job.

MISSING PHOTORECEPTOR LINES

A patient presented with 20/400 vision. I checked for evidence of subretinal hemorrhage or elevation and found none. Cirrus HD-OCT imaging ruled out vitreous pulling and tugging, cystoid macular edema and fluid beneath the retina. However, on close inspection, I was able to see that the patient was missing the photoreceptor integrity line in the center of the macula. The thickness map was abnormal. Fundus autofluorescence, which normally functions to help track progression in macular degeneration, suggested some degree of RPE atrophy or abnormalities within the sensory retina.

This finding was consistent with the disrupted photoreceptor integrity line, representing an old case of central serous. Most cases of central serous subside without treatment. In this patient, however, fluid had migrated inferiorly over time because of gravity, causing a neurosensory detachment.

Meanwhile, a 57-year-old male patient with 20/60 vision and a 25-year history of decreased vision presented with a yellowish change in the center of his right fovea. He also had a loss of the photoreceptor integrity line in the center of the fovea.

Color vision testing and an ERG were normal. Not until we took a better history did we pinpoint the source of his problem, which was related to solar maculopathy. The cause became apparent when he described that he had once used optical devices to look at the sun.⁵

BENEFITTING FROM TECHNOLOGY

If you're planning to purchase an SD-OCT, spend some time reviewing the different models available at optometric meetings. Take sample scans and evaluate the quality of the images. Factor in costs and consider that you need a unit that will remain viable when software upgrades become available.

As I've demonstrated, SD-OCT instruments are very effective, producing high-resolution images that offer great clinical value. They can elevate our level of care and vastly improve our ability to manage glaucoma and retinal conditions.

REFERENCES

1. Mwanza JC, Oakley JD, Budenz DL, et al. Ability of Cirrus HD-OCT optic nerve head parameters to discriminate normal from glaucomatous eyes. Ophthalmology. 2011;118(2):241-248.
2. Kass MA, Heuer EJ, Higginbotham EJ, et al; The Ocular Hypertension Treatment Study: A Randomized Trial Determines That Topical Ocular Hypotensive Medication Delays or Prevents the Onset of Primary Open-Angle Glaucoma. Arch Ophthalmol. 2002;120:701-713.
3. Sharma A, Oakley JD, Schiffman JC, Budenz DL, Anderson DR. Comparison of automated analysis of Cirrus HD OCT spectral-domain optical coherence tomography with stereo photographs of the optic disc. Ophthalmology. 2011;118(7):1348-1357.
4. Robichaud JL, Besada E, Basler L, Frauens BJ. Spectral domain optical coherence tomography of myopic traction maculopathy. Optometry. 2011;82(10):607-613.
5. Klemencic S, McMahon J, Upadhyay S, Messner L. Spectral domain optical coherence tomography as a predictor of visual function in chronic solar maculopathy. Optom Vis Sci. 2011;88(8):1014-1019.