Incorporating New Technologies into a Glaucoma Practice
EXPLORING The Latest Techniques and Treatments for Glaucoma
Incorporating New Technologies into a Glaucoma Practice
Experts discuss how the emergence of spectral domain OCT, new scanning lasers and other devices has changed their approach to patient management.
J. James Thimons, OD, FAAO: With the wealth of new instrumentation available for glaucoma care, clinicians must decide which devices are essential and how best to incorporate them into practice. Our panelists, all of whom have used the latest optical coherence tomography (OCT) units, perimeters, laser scanning devices and photography systems, will share their insights on the newest technologies and how they're using them to improve patient care.
Let's begin with your basic tools. What devices do you rely on day-to-day and why?
Peter Libre, MD: My mainstay perimeter is the Humphrey visual field analyzer. I rely principally on its standard achromatic perimetry capability, although I use the blue-yellow fields occasionally. The newer electronic model ties into my network, allowing me to access fields from the computer in my examination room, which is helping us move toward a truly paperless office, which will be great for environmental reasons and reducing labor costs.
I also rely on OCT. This technology is especially beneficial for evaluating the retina, which in my practice is invaluable. I wouldn't want to practice without OCT.
Larry Alexander, OD, FAAO: Visual fields continue to disappoint me on a daily basis. Regardless of the visual field I'm using — white-on-white, blue, yellow — I always must rely on the patient. That would be fine if I had only 40-year-old, responsive patients, but most glaucoma patients are not. Therefore, I rely very heavily on imaging and my direct observations of the optic nerve head.
I agree with Dr. Libre about the utility of OCT for imaging the retina. If you're caring for glaucoma patients, you need to consider the macula. In a patient with large or unusual discs, imaging helps me confirm or rule out disease. I can see if ganglion cells and retinal structures have been altered but would appreciate better technology to do this in a more quantifiable fashion. This is now becoming available and will significantly affect our ability to monitor our patients.
Gonioscopy is also very important. I'm continually surprised by how often I find nerve head damage and then see a narrow angle that appears to have precipitated subacute attacks. Cost-effective imaging technology to assist in this endeavor also will significantly impact on the management of glaucoma.
||The data we're acquiring with spectral domain technology is so phenomenal that we don't even know what to do with it yet. Ten years ago, most of us hadn't even heard of OCT. Now, most of us can't practice without it … We're still only scratching the surface of its potential.|
— Larry Alexander, OD, FAAO
Robert Dunphy, OD, FAAO: In our practice, we use the GDx VCC scanning laser polarimeter (Carl Zeiss Meditec, Dublin, Calif.) and the Stratus time domain OCT (Carl Zeiss Meditec). Most of our experience, however, is with the Heidelberg HRT2 retinal tomograph (Heidelberg Engineering, Vista, Calif.). We also recently acquired an RTVue-100 spectral domain OCT unit (Optovue Inc., Fremont, Calif.) and a Spectralis HRA+OCT (Heidelberg).
I agree that assessment of macular structure is an important aspect of glaucoma management. Our time domain device hasn't enabled us to segment the macular sublayers that we need to obtain accurate thickness measurements.1 My preliminary experience with the RTVue strongly suggests that the newer algorithms incorporated into this machine will be helpful in these areas. I've seen good correlation between our nerve head clinical analysis and the corresponding RTVue macular thickness evaluation, as well as good correlation between clinical nerve head assessment and RTVue nerve fiber thickness assesssment.
The HRT2's ability to model the nerve head is also vital from my perspective. This device enhances our ability to see the disc as a 3-dimensional volumetric structure.2 With HRT2's longitudinal analysis, we can see structural change or stability over time and therefore better assess the efficacy of our therapy. Likewise, the image registration and very detailed scan images of the Spectralis yield highly reliable data for cross-sectional and longitudinal analysis. Spectralis also can reveal structural details deep within the optic nerve head and choroid. I'm seeing similar deeper tissue information from the RTVue as well. I think that ongoing research will yield important information from these regions.
Dr. Thimons: I use the HRT, the Stratus and the RTVue. Although the other systems have been more than sufficient, I've found in my short period of observation that the RTVue spectral domain system offers remarkable functionality and image quality.
Spectral Domain Technology Assessment
Dr. Thimons: Is spectral domain OCT simply a replacement, providing better visualization, or does it provide clinically significant advantages over other technologies?
Dr. Alexander: Spectral domain OCT algorithms will be critical in defining the technology's utility. The trend analysis and progression analysis functions are like a book that no one can quite interpret yet. The data, including all of its nuances, are rich in quality.
One challenge, however, is that we're currently involved in retinal studies that won't allow us to switch to spectral domain OCT. Three of these devices are available, and it becomes an economic burden, especially now, to add three instruments without eliminating three other instruments.
Dr. Thimons: Dr. Libre, if cost weren't a factor, would you upgrade to spectral domain OCT?
Dr. Libre: I certainly would upgrade. The retinal imaging is far superior and would enable me to use macular thickness and ganglion cell layer thickness as additional parameters when following glaucoma patients.3
Dr. Alexander: An important point to keep in mind is that this technology provides pieces of the puzzle that inform our decisions. I make those decisions based on what will benefit the patient. As long as we're not losing revenue, then we must pursue the use of imaging.
I'd add that if we focus on imaging as a tool for managing glaucoma patients only, we'll miss out on widespread benefits. About 70% of my imaging occurs during retinal care, for example.
|Addressing Coding and Reimbursement Challenges|
|Dr. Thimons: Are there differences in reimbursement, cost savings or efficiencies when comparing imaging technologies?|
Bud O'Leary, OD: Imaging technologies can be roughly divided into three categories: posterior segment photography, anterior segment photography and scanning computerized ophthalmic diagnostic imaging. These generalized categories are codified in CPT and recognized by CMS as 92250, 92285 and 92135, respectively. Each category has utilization criteria, supporting ICD-9/10 diagnosis codes and documentation requirements specific to the management of glaucoma.
Reimbursement challenges, equipment costs and other issues certainly delay the adoption of new technologies. However, the enhanced ability to diagnose and effectively manage eye disease that, just a few years ago, could have led to greater organ morbidity and loss of vision, must be favorably considered. I like the analogy of compound interest: If you can get the problem under control earlier and in a less costly place of service, then the healthcare system doesn't have to pay the compounded interest to treat the more advanced disease state later on. Although some payors may try to refute the value of some new technologies and thereby limit utilization as a method of cost control, I feel theirs is a selfish argument.
Dr. Thimons: I believe the power of this new technology will defeat this initiative because the technology allows for much more sophisticated and accurate analysis of early disease and progression.
Dr. O'Leary, OD: Accurate, earlier diagnosis of any sight-threatening disease process can't be dismissed. The imaging technologies to make this early diagnosis possible are available now. I couldn't imagine practicing optometry without these tools.
||The retinal imaging [of spectral domain OCT] is far superior [to time domain OCT] and would enable me to use macular thickness and ganglion cell layer thickness as additional parameters when following glaucoma patients.|
— Peter Libre, MD
Dr. Thimons: Dr. Dunphy, you practice in the Veterans Affairs (VA) system. What impact has this new technology had in your setting?
Dr. Dunphy: The longer wavelength light used in the scanning lasers of the HRT and OCT devices provides improved imaging of the nerve head and macular structure, essentially functioning as another form of ophthalmoscopy. In my patient population, which presents with a high prevalence of media opacity that doesn't yet require surgery, I obtain excellent raw scan data on the nerve head and macular structure and change over time. This leads to better interventions.
Many of my glaucoma patients are in moderate or early-advanced stages of disease. Structural changes at this point in the disease often are associated with conversion from pre-perimetric glaucoma into frank perimetric glaucoma, potentially affecting the patient's lifestyle. My ability to identify patients at risk for these conversions is a great clinical asset. Using the technology to model the nerve head complements the assessments I make through nonimaging-based evaluations.
Dr. Alexander: I believe some of these instruments eventually will have capabilities beyond their current use. Already, we can use some of them to image the anterior chamber angle and assess the relative anatomy. Eventually, with enhanced software, the technology will help with blood flow analysis and evaluation of functions throughout the body.4
Dr. Thimons: Considering the apparent association between Alzheimer disease and papillo-macular bundle decline,5,6 I was fascinated to find that the RTVue can distinguish between a normal and non-normal macular bundle level. Several instruments in this new technology group have potential applications related to neurosensory evaluation of the eye, relative to cognitive dysfunction and other issues. Dr. Libre, what is the importance of imaging related to the macula?
Dr. Libre: OCT effectively guides how I manage patients postoperatively. After surgery, a patient may experience cystoid macular edema, or his vision may be impaired due to epiretinal membranes. OCT shows me immediately if I should be treating a patient more aggressively with corticosteroids and non-steroidal anti-inflammatory drugs, or if his suboptimal acuity is due to some other factor, such as corneal astigmatism from trabeculectomy sutures or posterior capsule opacity.
As we discuss the value of OCT technology for glaucoma management, I want to emphasize there's no substitute for direct ophthalmoscopic observation of the optic nerve head. Even the most advanced technology never will relieve us of the duty to become skilled examiners of the optic nerve head. Nor should doctors feel that the absence of advanced technology substantially limits their ability to take good care of their patients.
Dr. Thimons: Do you see technology as an adjunct to direct observation?
Dr. Libre: Yes, I do. OCT and visual fields help confirm my impressions about the optic nerve head. For example, if I think the inferior temporal rim looks damaged and OCT confirms that this portion of the nerve fiber layer is thinner than the same area on the other side, and also perhaps thinner than other sectors in the same eye, this confirmatory information helps me decide if I should initiate treatment.
Dr. Thimons: I believe we all agree that the basic examination of a glaucoma patient must include a thorough, comprehensive evaluation of all systems by the clinician, aided by adjunctive information from visual fields and optic nerve imaging. I use this analysis system every day, and I feel comfortable that the system gives me a better view of a patient at 6 months and 1 year than my observation of the disc can. I believe the best application for spectral domain OCT is to monitor progression, not to make an initial diagnosis. Does anyone care to comment?
Dr. Dunphy: Every clinician wants a device that will tell him whether or not glaucoma is present and, once a diagnosis is made, whether or not a therapy is stabilizing the disease process. Over time, clinicians came to expect these devices essentially to make the diagnosis for them, and this is why I believe Dr. Libre's point regarding nerve head clinical assessment deserves to be repeated again and again.
Clinicians need to look at the nerve head in conjunction with their assessment of the patient as a whole and then evaluate for change over time. There's no question that these devices will be more useful for monitoring change than for discriminating normal vs. abnormal on initial assessment. Optic nerve head initial assessment is less certain because of the inherent variability in normal nerve heads. The most valuable devices are those that effectively register images and reliably produce information that indicates either stability or change over time.
We have been discussing OCT technology extensively. Remember, however, that the foundation of image registration and the extensively validated process for following patients for change over time is found in the HRT platform.
Dr. Thimons: Another important consideration, though, is that spectral domain systems also have nerve head registration. This is a significant step beyond what's possible with time domain technology, which doesn't register nerve position and shape as effectively.
||Our time domain device hasn't enabled us to segment the macular sublayers that we need to obtain accurate thickness measurements. My preliminary experience with the RTVue strongly suggests that the newer algorithms incorporated into this machine will be helpful in these areas.|
— Robert Dunphy, OD, FAAO
Dr. Alexander: Glaucoma is becoming much more complex. Suddenly, we're returning to the concept that Felix Lagrange of Bordeaux, France, expressed years ago — that glaucoma is a sick eye in a sick body.7 We must consider so many nuances when making a differential diagnosis and managing a patient's therapy. Besides lowering IOP, we have to determine where the visual system has gone awry. Any help that I can get from advanced technology is beneficial. However, we can't expect a device to produce the answer.
Dr. Thimons: Anyone who takes care of patients whose cognitive dysfunction prevents the use of visual fields appreciates the data provided by these devices. Progression analysis via imaging may be our only means for accurately assessing function in these patients.
Dr. Libre: Another consideration is that OCT saves time when a patient who appears to have normal optic nerves and physiologically large cups struggles to complete an accurate visual field. Obtaining a normal OCT scan of the retinal nerve fiber layer often is sufficiently reassuring; I don't feel obliged to evaluate visual fields further.
Visual fields are still essential, however, for managing confirmed glaucoma, providing critical information about function that's not available through imaging. Ultimately, we're trying to preserve function to prevent visual disability.
Role of Photography
Dr. Thimons: I use stereo photography on every new glaucoma patient to establish baseline images, and then on a continuing basis to watch for hemorrhage and other issues when the view isn't particularly clear and I don't want to dilate an eye. I also use photography annually to record the stereo status of the optic nerve.
Does stereo photography still have a role in your practice?
Dr. Dunphy: Photography is important and easily overlooked when discussing newer technologies. With the newer systems, we can easily acquire excellent high-resolution photographs. In addition, we can identify important features and changes in the nerve fiber layer and rim over time. This is no substitute for an ophthalmoscopic assessment, but one can review the nerve head and make excellent observations of key features and/or change over time when the nerve head image is the size of a dessert plate on your monitor.
Bud O'Leary, OD: One study8 demonstrated that about 85% of Drance hemes detected using stereophotographs weren't identified upon direct funduscopic examination. The importance of the technological advantage is compelling.
Dr. Alexander: Stereo photography is critical. I'd choose imaging before stereo photography, but I wouldn't want to practice without my photography system. Practitioners who are forced to choose because of economic reasons, however, should prioritize, relying first on visual fields and then on some type of imaging.
Dr. Libre: I still find optic nerve photography extremely useful, and I take simultaneous stereo disc photos on almost all of my glaucoma patients. Most doctors like to record images or photographs or both, as well as visual fields, every year on every patient. I use a photograph of the optic nerve as a baseline. I don't necessarily repeat photography every year because that creates more data than I need. I believe the likelihood of progression during 1 year is quite low and taking photographs every 4 to 5 years is more cost-effective. I place the baseline stereo photograph next to my slit lamp as I perform a stereoscopic evaluation of the optic nerve, and I visually compare the nerve head and the photograph to identify any changes.
Meanwhile, I think the frequency of nerve fiber layer analysis should be determined by the likelihood of progression in a year. Performing analysis every 2 to 3 years is reasonable in low-grade glaucoma suspects and some patients with mild glaucoma and pressures consistently at target.
|Tips for Improving Glaucoma Patient Care|
|When diagnosing and managing patients with glaucoma, these five pearls may be helpful.|
• Pearl #1: Don't underestimate the information you can obtain from confrontation fields.
Particularly in severe cases of glaucoma, simply asking patients to fix on your forehead and count your fingers can help you stage disease. Test each quadrant to determine whether the patient counts fingers well or poorly, or sees only hand motion. This test is far less sensitive than automated perimetry,1 and won't become abnormal until roughly three-quarters of the nerve is dead.
• Pearl #2: Consider the implications of intertest fluctuation on visual fields.2 If you identify a zone of abnormality, this would then trigger an abnormal pattern standard deviation. A positive pattern standard deviation doesn't mean anything if the zone of abnormally clustered spots shifts from one quadrant to another upon retesting. A repeatable defect in the same quadrant, even if the density varies, establishes a more reliable sign of functional impairment.
I stage neuronal loss primarily by considering the mean deviation (MD), the average amount by which every point is depressed from normal. 1+ glaucomatous optic neuropathy is undetectable by visual fields. Minus 5 decibel (dB) MD roughly corresponds to 2+ damage. Minus 10 dB reflects 3+ severe damage, and –20 dB is just about end stage. The patient might still have central acuity, but he's in deep trouble. The worst you can get is –28 to 30 dB.
• Pearl #3: Put a drop of artificial tears on the cornea just before performing optical coherence tomography (OCT). Credit for this tip goes to Dr. Thimons. The tears can help improve signal strength by one or two points.
• Pearl #4: Understand the real implications of OCT thickness readings. Although a reading of 50 microns is just half of normal, don't be deceived. A finding of 70 microns ±10 represents moderate loss, while a reading of 50 microns indicates severe loss.
• Pearl #5: If you see an intact temporal rim and the patient's acuity is poor, suspect a tumor. If the rim is very thinned out, I worry because the patient's central acuity is at risk, probably due to glaucoma. If the temporal rim is pale and intact but acuity is reduced, a tumor is the first thing to consider. I have seen only two tumors in the last 3 years, but vigilance is always important.
— Peter Libre, MD
- Bass SJ, Cooper J, Feldman J, Horn D. Comparison of an automated confrontation testing device versus finger counting in the detection of field loss. Optometry. 2007;78:390-395.
- Advanced Glaucoma Intervention Study. Visual field test scoring and reliability. Ophthalmology. 1994;101:1445-1455.
||With spectral domain OCT, after initiating the acquisition, you simply watch the data evolve. You don't have to be ambidextrous in order to run the machine and watch the image capture. The spectral domain systems have elevated information acquisition to a new level.|
J. James Thimons, OD, FAAO
Dr. Thimons: Do you think some patients would benefit from nerve fiber layer analysis more than once a year, given progressive disease or apparent lack of control, as determined by either visual field analysis or optic nerve assessment?
Dr. Libre: I wouldn't want to perform imaging more than once a year. I perform visual fields more than twice a year only in a small percentage of my patients. In patients at special risk for progression in whom I've previously seen disc hemorrhages, I'd examine the optic nerve every 3 months to check for disc hemorrhage.
Dr. Thimons: The confluence of various technologies has caused me to adjust my frequency of photography and testing throughout the years. I now take photographs yearly to compare the images. I like to review them on-screen at the same time to perform a serial analysis. For my patients who can't successfully complete visual fields, I use photography more than once a year. As a consultant to several nursing homes, I have many patients who are either minimally verbal or cognitively impaired. Photographs are an excellent way to assess a patient's status over time without the use of visual fields. Additionally, I use a digital software management system that markedly enhances my ability to manipulate data quickly and efficiently.
Dr. Dunphy: We have to remember that we're involved with glaucoma patients on a long-term basis. One advantage of nerve head imaging, using photography and/or a scanning laser, is that we can demonstrate these findings to our patients. In my experience, the better patients understand stability vs. change, the more likely they are to participate in their care and follow my instructions. Providing patients with feedback for what seems to them to be a symptom-free disease is helpful and engages them in the continuing care process.
Dr. Thimons: One of the basic concerns about a new technology is if we're adequately preparing ourselves to interpret the data that it produces. Do you have any pearls to share about the advantages and limitations associated with interpretation?
Dr. Libre: I've found the most useful data point provided by traditional OCT is the average thickness of the retinal nerve fiber layer. This parameter best describes the overall health of the nerve, allowing me to gauge progression.
The most significant limitation is related to varying signal strengths when monitoring progression.9 You need to know the signal strength used for each scan. If the second scan has declined by more than 2 since the first scan, that factor alone likely will explain any perceived thinning of the nerve fiber layer.
Dr. Thimons: For every single-digit decrease in signal strength, a 3-micron decrease in thickness can be attributed to the device, not the disease. A 10-micron drop is significant if it is truly pathologic. However, if it's just a reflection of differences in signal strength, I recommend repeating the measurement at another time and supplementing your assessments with adjunctive measures.
Dr. Alexander: The fallacy associated with this instrumentation is that we believe it's objective, rather than subjective. I'd argue that imaging is subjective, from the perspective of the patient and the technician who's acquiring the data. This is the classic "garbage-in-garbage-out" principle. If you don't concentrate on getting a good image, you'll be in trouble. This is made easier because data acquisition with the new instruments has been simplified.
||Accurate, earlier diagnosis of any sight-threatening disease process can't be dismissed. The imaging technologies to make this early diagnosis possible are available now. I couldn't imagine practicing optometry without these tools.|
Bud O'Leary, OD
Dr. Dunphy: I agree that it's important for us to think that imaging devices generate data that we interpret. Talking about hardware and technologies is easy. Patients come to us for an opinion, which contains elements of a subjective thought process. Patients are best served when doctors have excellent information to use in conjunction with their experience with and awareness of various problems. Subjectivity is acceptable, as long as the doctor is using reliable objective information in the development of an opinion.
Furthermore, when using these technologies, we need to think about them from an ophthalmoscopic perspective. Consider the raw scan information. If you can't look into the data set, you can't trust what's coming out of the machine.
We need to resist the temptation to look only at the report output. In fact, sometimes, we need to perform the scan ourselves to make sure the data set is valid. We need to evaluate devices on their ability to generate high-quality images. We also need to think about software variability across different imaging platforms.
Dr. Thimons: Twenty years ago, when quantified perimetry entered the glaucoma arena, the biggest challenge was learning how to read output from the devices, including the validity of the data, the significance of the images and when to take action. Absorbing that informational system and making it part of our daily management routine took more than a decade. I think we'll make the transition to the new OCT technology much more quickly.
With spectral domain OCT, after initiating the acquisition, you simply watch the data evolve. You don't have to be ambidextrous in order to run the machine and watch the image capture. The spectral domain systems have elevated information acquisition to a new level.
Scratching the Surface
Dr. Alexander: The data we're acquiring with spectral domain technology is so phenomenal that we don't even know what to do with it yet. Ten years ago, most of us hadn't even heard of OCT. Now most of us can't practice without it, whether we're focusing on the retina or the optic nerve. We're still only scratching the surface of its potential. The extent of the utilization of the data acquired with this technology is only limited by the ability to create software to analyze it.
Dr. Thimons: We've seen parallel technological advances for the anterior segment. We've gone from simple topography with basic systems to more advanced systems, such as the Atlas. We moved on to the Orbscan corneal topographer, the Pentacam and now we have wavefront analysis.
Our colleagues have significantly improved outcomes with each advancement. I think spectral domain technology will produce the same type of improvement — significant, not incremental — in our ability to manage glaucoma over time. I'm delighted that we've made this quantum leap, which is on the same scale as the arrival of wavefront analysis a few years back. Thank you all for illuminating these important issues, which I think probably will change our entire profession. OM
- 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. 2008 Apr 30. [Epub ahead of print].
- Zangwill LM, Weinreb RN, Beiser JA, et al. Baseline topographic optic disc measurements are associated with the development of primary open angle glaucoma: The Confocal Scanning Laser Ophthalmoscopy Ancillary Study to the Ocular Hypertension Treatment Study. Arch Ophthalmol. 2005;123:1188-1197.
- Tan O, Li G, Lu AT, Varma R, Huang D. Advanced Imaging for Glaucoma Study Group. Mapping of macular substructures with optical coherence tomography for glaucoma diagnosis. Ophthalmology. 2008;115:949-956.
- Barlis P, Serruys PW, Gonzalo N, van der Giessen WJ, de Jaegere PJ, Regar E. Assessment of culprit and remote coronary narrowings using optical coherence tomography with long-term outcomes. Am J Cardiol. 2008;102:391-395.
- Danesh-Meyer HV, Birch H, Ku JY, Caroll S, Gamble G. Reduction of optic nerve fibers in patients with Alzheimer disease identified by laser imaging. Neurology. 2006;28:1852-1854.
- Iseri PK, Altinaş O, Tokay T, Yüksel N. Relationship between cognitive impairment and retinal morphological and visual functional abnormalities in Alzheimer disease. J Neuroophthalmol. 2006;26:18-24.
- Lagrange, F. Du Glaucoma et de l'Hypotonie, Paris, 1922.
- Uhler TA, Piltz-Seymour J. Optic disc hemorrhages in glaucoma and ocular hypertension: Implications and recommendations. Curr Opin Ophthalmology. 2008;19:89-94.
- Cheung CY, Leung CK, Lin D, Pang CP, Lam DS. Relationship between retinal nerve fiber layer measurement and signal strength in optical coherence tomography. Ophthalmology. 2008;115:1347-1351.