To the Editor:

I am writing in response to Dr. Scott D. Smith's article, "Does Better OCT Mean Better Glaucoma Care?" (January 2009, Glaucoma Management, p. 58). Dr. Smith provides us an excellent summary of the differences between time and spectral domain technologies. Referring to  the Zeiss Cirrus HD-OCT, he is concerned that there is no statistical guidance as to how large or deep a defect needs to be before it is really abnormal and states that interpretations are subjective. He does not discuss the capability of the instruments to monitor for glaucomatous progression, but this certainly is a concern of many users.


It is refreshing to have an article that expresses a healthy skepticism as to whether or not the additional information afforded by a new technology is really that advantageous.


My experience with the Cirrus differs from Dr. Smith's. In my practice, the objective data from the Cirrus printout often serves as a game-changer in terms of whether to treat or observe. For instance, in patients with suspicious cupping and IOPs under 22, a normal nerve fiber layer study will almost always point me in the direction to observe. On the other hand, an abnormal scan invariably points to treatment, particularly if the eye with the higher IOP has more thinning on the thickness, deviation or symmetry map(s). Moreover, there is statistical guidance, as the deviation map lists normal distribution percentages, 0 to 1 percent, 1 to 5 percent, and so forth.


Dr. Smith also comments that the information outside the Cirrus 3.4-mm diameter auto-centration circle—while manifest in the scan—requires subjective interpretation. Based on my experience I believe that deviations found outside the circle are of clinical significance only if they are extensions of deviations within the circle; isolated outside-the-circle deviations carry much less import.


Granted as Dr. Smith discusses, there are some patients with abnormal scans who do not have glaucoma and are undoubtedly false positives.

In younger patients, highly symmetrical defects on the deviation map often tell me that the patient is simply an outlier from the normative data- base. The disk or NFL may be rotated to account for the "dropout." In older patients with multiple pathologies such as cataract, PCO or corneal opacities, irregular or asymmetric deviations on the scan can occur even if the quality reading of the scan is acceptable. In addition there is an occasional patient with irregular nerve fiber layers in an analogous way that some patients have funny- looking cups. These cases, while problematic, comprise only a small percentage of studies. As always, clinical correlation with other data is advisable.


Finally, each eye effectively serves as a "control" for the other eye, regardless of which spectral domain instrument one has. If the higher IOP eye has the more abnormal OCT scan, consider the diagnosis of glaucoma until proven otherwise.


Lawrence
Stone, MD

Chicago



Author reply:

I thank Dr. Stone for his interest in my article. Like Dr. Stone, I am an enthusiastic user of this technology and find it to be extremely useful, particularly in helping to determine whether patients suspected of having glaucoma have early NFL loss indicative of glaucomatous optic nerve damage. On a daily basis, I find my task of identifying glaucoma in its early stages to be easier, and I feel more confident in monitoring glaucoma suspects and ocular hypertensives without treatment on the basis of normal NFL thickness measurements.


However, I often find myself faced with test results that are more difficult to interpret. When NFL thickness measurements are not completely normal, but there are shallow "defects," does this truly represent disease, or is it a false positive result? I find myself wondering whether the result is abnormal enough to be definitive, whether a patient's degree of myopia may be affecting the result, etc. When viewing the plot of NFL thickness measurements in the peripapillary retina, I find myself asking how many pixels need to be abnormal, and how deep does a defect need to be or how much asymmetry needs to be present for me to consider the result to be abnormal. The suggestions given by Dr. Stone for clinical interpretation of these types of scans can be quite helpful, but this assessment remains subjective, and different clinicians could come to different conclusions using the same test results. The fact that he has needed to devise a system for judging the test results indicates that improvements in the normative database and analysis software are needed. In addition, the judgment of glaucoma progression remains subjective, as there is currently no statistical guidance offered as to how much change is needed to be outside the range of expected variability of the measurement itself.


The accumulation of a larger, more detailed normative database that accounts for anatomic and demographic factors that can affect NFL thickness (age, race, disk size, refractive error and possibly others) will greatly reduce the subjectivity involved in the clinical interpretation of NFL thickness measurements. In addition, I envision a system analogous to those we are familiar with for visual field interpretation that assist us in objectively determining how much asymmetry is expected, and whether the degree of asymmetry present in a scan is normal or abnormal. An objective system for analyzing pointwise changes in NFL thickness is also needed to make the determination of glaucoma progression more objective. Certainly the manufacturers are aware of this need, and are working to address these shortcomings. Once this has happened, we will be able to make even better use of the NFL thickness measurements that these instruments are capable of making.


Scott D. Smith, MD, MPH

Director, Glaucoma Service, Columbia University

New York City



To the Editor:

To the thorough January article on Thyroid Orbitopathy ("Update on Thyroid Eye Disease," Plastic Pointers, p. 61) should be added the use of Guanethidine Sulfate topical solution. In my experience, it has been very effective when used in 5% solution four times daily, to reduce lid retraction, cosmetically, and to treat secondary exposure keratopathy. Many of these patients cannot close their eyes during sleep because of Müller's muscle contracture. Guanethidine sulphate relaxes that muscle and allows the lids to close, thus providing cover for the cornea. Guanethidine sulphate is also helpful to minimize the appearance of proptosis cosmetically.


Before surgery is contemplated, I would urge our colleagues to try Guanethidine sulphate for few days, q.i.d., and then for maintenance two times or once (preferably at bedtime) daily.


Heskel M. Haddad, MD

New York City

 

Author reply:

We thank Dr. Haddad for pointing out this therapeutic option for patients with thyroid eye disease and lid retraction. Although not widely used, guanethidine sulfate is a sympatholytic causing relative ptosis, with side effects of miosis, conjunctival injection and ocular irritation. The authors do not have personal experience with this intervention. However there is scientific literature to support its efficacy and the authors encourage medical prior to surgical interventions in this disease.


Chih-Hsien Hsiao MD, C. Robert Bernardino, MD FACS


1. Gay AJ, Wolkstein MA. Topical guanethidine therapy for endocrine lid retraction. Arch Ophthalmol. 1966;76(3):364-7.