Monitoring Glaucoma and Retinal Disease
Monitoring Glaucoma and Retinal Disease
In addition to never-before-seen detail, new technologies provide guidance for the difficult task of progression analysis.
Diana Shechtman, O.D., F.A.A.O. & Joseph Sowka, O.D., F.A.A.O., DIPL. Fort Lauderdale, Fla.
Recent advances in imaging technology allow practitioners to view the posterior segment of the eye in greater detail than ever before. To the obvious benefits of increased visualization, device manufacturers more recently have added computer software programs that help you interpret this information and offer database-derived analysis on whether disease is present or progressing. These programs exist for both glaucoma and retinal pathologies.
Having had extensive experience with technologies in both fields, we describe changes that occur in glaucoma and retinal disease and the newest ways of monitoring those changes, including structural analysis of retinal diseases and combining structural and functional analyses of glaucoma.
Part 1: glaucoma
Three broad undertakings comprise the whole of glaucoma management: diagnosis, treatment and monitoring for stability or progression. All three are challenging, but treatment tends to be the least challenging. Pharmacological and surgical strategies can usually lower and control intraocular pressure effectively. Diagnosing glaucoma is notoriously complex. But monitoring known cases of glaucoma and determining whether the disease is progressing — and whether that advancement threatens vision — can be extraordinarily difficult.
Progression may be measured by either changes in function (visual field deterioration) or structure (optic disc and retinal nerve fiber layer [RNFL] appearance). Further, progression can be evaluated by means of “event analysis” or “trend analysis.” Event analysis compares baseline measurements to the most recent test data to determine, according to preset criteria, whether change (progression) has occurred. Thus, the “event” has either happened or not. It does not quantify the magnitude of change or the rate of change, but merely whether change has occurred based upon preset criteria. Trend analysis looks at the significance of the rate of change over a period of time by arranging the data points of multiple tests in a linear regression formula (it is sometimes called regression analysis). Because progression rate is the most predictive factor in estimating risk of future visual impairment, trend analysis can be very helpful. However, its weakness lies in its assumption of linear calculations, since glaucoma does not necessarily progress linearly. The more focused event-based analysis offers better sensitivity to small instances of localized progression. Each computation has its advantages and disadvantages, but generally speaking, we need as much data as possible from both.
A great deal of heated academic debate has occurred regarding whether structural damage precedes functional damage or vice versa. For this article, we set aside that argument and assume some patients show structural changes first (probably the majority, in our view) while others show functional changes first. Today's standard of care calls for both functional testing and structural measurements. One alone is seldom enough.
However, as a practical matter, visual field testing presents well-documented challenges and limitations. For starters, patients tend to feel threatened by visual field testing: It takes a long time; they cannot always understand it; and some have physical limitations that preclude enduring the entire test. Also, visual field testing's reliability and reproducibility are challenges. Finally, in most cases, significant damage must occur before measurable functional loss registers.
Guided Progression Analysis
Years ago, another major drawback of visual field testing was the highly subjective interpretation of results. At that time, a technician would print out a series of visual field test results. Arranging the sheets of paper chronologically on a long table, the clinician would then study them to determine whether progression had occurred. Needless to say, such guesswork made for considerable imprecision and variability.
In 2003, Carl Zeiss Meditec introduced Guided Progression Analysis (GPA), a software program designed for use with the company's Humphrey Field Analyzer. The software helps to solve the biggest challenge of assessing multiple visual field results, that is, differentiating clinically significant progression from normal test variability. It accomplishes this by drawing on data collected in conjunction with the Early Manifest Glaucoma Trial. In preparation for this landmark study, hundreds of glaucoma patients at multiple study sites underwent field testing numerous times in a single month. Because one month is not enough time for glaucoma to progress, an amalgam of the data gives the computer a fair range of the average glaucoma patient's test variability parameters. Thus, unless visual field test results exceed these parameters, the GPA does not regard them as a genuine change.
The GPA begins by averaging two visual fields. This forms the baseline to which future data is compared. Then, the software performs an automated point-by-point analysis of subsequent exams. It identifies consistent and repeated patterns of visual loss while also correcting for ocular media effects. GPA can be used in conjunction with SITA Standard and SITA Fast strategies (older full threshold fields can be used as baseline, but not for ongoing analyses). Visual fields that repeatedly and consistently show changes exceeding what is known to represent typical variability are identified as having “possible” or “likely” progression. Trend analysis can be plotted through a three- to five-year period, which helps to predict whether established progression might jeopardize vision.
Initial baseline readings must be of the highest quality, as an inaccurate baseline will cause skewing of subsequent data. Conveniently, the software allows the baseline to be “reset” at any time by repeating the visual field testing. This also permits setting a new baseline following a major pharmacological or surgical intervention.
GPA for OCT
Similar statistical analyses can be performed on images generated by retinal scanning technologies that track structural changes in the optic nerve head and RNFL. For this article, we focus on optical coherence tomography (OCT), but similar software is available on other platforms as well. The dynamics remain the same as in visual fields. The program seeks changes that are similar, repeated and significant. When change occurs, normative data-bases help determine whether it is within acceptable tolerances for variation or exceeds normal variation. Finally, trend analysis indicates what the future may hold.
Again, two exams form the baseline. To ensure accurate comparison, follow-up exams are registered to the baseline by matching the patient's blood vessels. Maps of the optic nerve head and RNFL are combined on one printout. On the third exam, a sub-pixel map denotes change with yellow pixels. On the fourth exam, change identified in three of the four comparisons is indicated by red pixels. As in visual fields, change is defined as that which exceeds the known variability of a study population.
A display called the “Summary Parameter Trend Analysis” charts the rate and significance of the change in the RNFL. Again, yellow denotes change from both baseline exams while red denotes change from three of four comparisons. Confidence intervals, shown by a gray band, estimate how much RNFL loss will occur if current trends continue.
Part 2: retinal conditions
The widespread use of OCT has changed the way we think about the posterior segment in many ways. Diagnosis of occult pathologies, evaluation of the intricate relationships among ocular structures, the understanding of various ocular conditions and better monitoring capabilities are now possible with the use of OCT.
For example, a 50-year-old patient presents with no known refractive error and 20/25 visual acuity OD. Health exams, including ophthalmoscopy, are unremarkable. The OCT reveals vitreomacular traction (VMT) associated with macular edema, corresponding to the mild decrease in visual acuity. (Once believed rare, VMT is commonly found today in a routine eye exam using an OCT.)
Advance visualization capabilities have helped in the evaluation of various retinal conditions. Using new computer software, we can sequester a particular “slab” of a retinal layer to evaluate only the portion of the retina that we want to assess. We can select and view a specific structure, such as the internal limiting membrane (ILM), in vitreoretinal conditions, such as vitreomacular traction syndrome.
A retinal pigment epithelial (RPE) slab, in conditions such as age-related macular degeneration (AMD), has also been proven useful. Different scan views, like C-scan (en face) imaging, have been practical in the assessment of choroidal lesions and in fact have been tantamount to a “virtual” fluorescein angiography. In addition, we can view retinal thickening in three dimensions, allowing us to evaluate and monitor patients more effectively than before.
OCT has also been invaluable in enabling us to monitor posterior segment ocular disease. The rise in our ability to monitor and track change in the retina was evident with the advent of anti-VEGF therapy in AMD patients, (PrONTO study). While these new therapies are highly beneficial, they have a limited duration of effect. Thus, proper monitoring is necessary to determine the need for more therapy, and OCT is ideally suited for this.
Software included in the OCT allows us to effectively track retinal changes. The following example illustrates its utility:
A patient who was diagnosed and treated for choroidal neovascular membrane associated with exudative AMD returns for a follow-up exam in two months, demonstrating what appears to be thickening of the macula. The question becomes: Is the thickening associated with disease reoccurrence?
Using the Macular Change Analysis software in the Cirrus HD-OCT, we can compare the two retinal thickness maps side-by-side to accurately determine retinal thickness change over time. In this case, there was a quantitative measurement increase of 100 microns. The software's primary benefit is ensuring that the precise same locations are compared between scans. This is accomplished via the Automatic Fovea Finder, which precisely aligns the fovea of each image. The Automatic Fovea Finder can also be helpful for AMD patients whose central scotomas prevent them from focusing on the red light while the scan is performed.
Tracking dry AMD
Evaluation of drusen and its affect on the RPE can be visualized using an RPE slab. Associated retinal thinning, which may cause functional change, can be delineated on the retinal thickness map. Typically, this is described as a blue-colored area.
We can use these quantitative and qualitative measurements to monitor patients and make decisions regarding more frequent monitoring, intervention with nutritional supplements, life-style modifications and finally prompt referral when treatment is a necessity.
Subtle cases of VMT may be unperceivable with the use of ophthalmoscopy alone.
For example, an 83-year-old retired O.D. came to our clinic with a chief complaint of “anisocoria.” His visual acuity was 20/20 OD, OS, with no significant differences in refractive error between the two eyes. The dilated fundus exam revealed a subtle epiretinal membrane (ERM) of the right macula. No obvious pathology was noted in the left macula following the dilated fundus examination.
Using the HD-OCT scans on the left eye, we noted VMT associated with multiple areas of attachments and detachments. In addition, the right eye was remarkable for an ERM associated with VMT, causing retinal distortion. This was further analyzed using the ILM slab, which showed substantial disruption in the ILM for only the right eye, hence attributing to the initial chief complaint.
The use of OCT has led to an increased recognition of VMT. A recent study on a series of patients who have various maculopathies is illustrative. VMT was noted in 8% of patients when using ophthalmoscopy alone, compared to 30% in the same patient group when they were evaluated in conjunction with OCT images.
Since the advent of OCT, the staging of macular holes has been revised, with the addition of a stage 0 macular hole. We also have a better understanding regarding the pathogenesis of such holes. A stage 0 macular hole is a partial posterior vitreous detachment with associated oblique insertion of the posterior hyaloid into the fovea (typically not associated with macular distortion or symptomology). Yet, further traction from the posterior hyaloid can lead to the formation of a full-thickness macular hole. Researchers speculate tractional forces are the contributing factor to the formation of a full-thickness idiopathic macular hole.
Evaluation of the presence of a partial or the stage of the hole is especially useful in management of the patient. Staging is crucial in making management decisions with regard to treatment intervention.
Protocols in transition
New imaging and computer technologies offer optometrists valuable tools that aid in the diagnosis and comanagement of ocular diseases. In glaucoma, enhanced, computer-assisted interpretation of both structure and function has revolutionized our ability to detect glaucomatous change and predict risk of future disability.
In retinal disease, the explosion in new therapies aimed at treating degenerative disorders, such as AMD and diabetic retinopathy, has created an equally large need to monitor and evaluate a whole new patient population. Hence, the widespread acceptance of OCT could not have occurred at a more opportune time, as it is ideally suited to observe patient response to these therapies. In addition, the ease, speed and accuracy of OCT technology opens unexplored diagnostic opportunities for optometry. Routine exams can now consistently reveal complex pathologies that not long ago would have remained occult. OM
Neither Dr. Shechtman nor Dr. Sowka have a financial interest in any of the products mentioned in this article.
|Progression Technology in SD-OCT and Perimetry Devices|
Compiled by the OM Editorial Staff
Below is a sampling of technologies offered by spectral domain OCT and perimetry device manufacturers to aid clinicians in their efforts to monitor the progression of posterior disease. Contact the manufacturer for additional information.
Carl Zeiss Meditec Cirrus HD-OCT — Guided Progression Analysis (GPA) determines whether statistically significant change has occurred to the RNFL. Macular Change Analysis software provides qualitative and quantitative comparison of the macula compared with successive exams.
Heidelberg Spectralis — The AutoRescan feature automatically places follow-up scans in precisely the same location as the baseline scan, enhancing the clinician's ability to observe true change through time rather than change resulting from alignment error.
Opko Spectral OCT/SLO — The Spectral OCT/SLO offers an “Auto-Compare” of multiple topography maps (progression/regression plot) including location of retinal thickness changes overlaid on the SLO (scanning laser ophthalmoscope) fundus image.
Optovue RTVue — The RTVue offers Ganglion Cell Complex (GCC) analysis for enhanced confidence in retina and glaucoma assessment. An OCT fundus image can be overlaid with a colorcoded thickness map that provides database-derived thickness values.
Topcon 3D OCT-2000 — The 3D OCT-2000 glaucoma module allows automated disc topography, normative database comparison and total progression analysis (trend analysis). The module is complemented by ganglion cell analysis and anterior chamber angle measurements. The 3D OCT-2000 also includes a fundus camera.
Carl Zeiss Meditec Humphrey Field Analyzer — Guided Progression Analysis (GPA) software differentiates clinically significant progression from normal test variability.
Octopus Haag-Streit International EyeSuite Progression Analysis — With EyeSuite Perimetry, you perform trend analysis right on your computer screen with specific and comprehensive graphs. Change is identified by color codes. The software includes a unique structure-function correlation and cluster analysis.
Kowa AP-5000C — This perimeter offers the ability to overlap the ocular fundus image and perimetry result, thus helping to shed light on the structure-function relationship.
Paradigm Autoperimeter Model 500 — An updated version of the Dicon LD 400 Perimeter, this device is supported with an improved version of the Advanced FieldView software. New threshold algorithms improve the reliability of threshold levels and measure patient response time.
||Dr. Shechtman is an associate professor at Nova Southeastern University College of Optometry, where she is an attending optometric physician at the eye institute and diabetic/macula clinic.|
||Dr. Sowka is a professor at Nova Southeastern University College of Optometry where he's chief of the advanced care center and director of the glaucoma service. He is a Diplomate of the American Academy of Optometry's Disease Section, Glaucoma Subsection.|
Optometric Management, Issue: February 2011