Diagnose AMD earlier with dark adaptation testing
Patients with abnormal results can be kept under closer observation, and preventive measures can be taken.
An immediate referral to a retina specialist due to a bleed is a tense circumstance for all involved. However, I have learned through experience that this situation is avoidable.
Up to 78% of patients with age-related macular degeneration have substantial, irreversible vision loss at first treatment (Olsen et al., Cervantes-Castañeda et al.), but almost all of my patients fall on the minority side of this grim statistic.
Identifying AMD early
We installed the first commercial dark adaptometer in our practice in 2014. The AdaptDx (MacuLogix) is a functional testing device that has been shown in clinical trials to identify patients with the earliest stages of AMD even when they have no other structural signs of AMD. It does this by revealing impaired dark adaptation (DA) function associated with early AMD up to 3 years before it becomes clinically evident. In other words, DA is a biomarker for disease (Owsley et al., 2016).
Before bringing this technology into my practice, I would see patients with some small drusen but would be hesitant to deliver a definitive diagnosis of AMD. Now, when I see something mildly suspicious but seemingly innocuous, I know for certain that it is AMD if impaired DA confirms the diagnosis. Conversely, if DA is normal, I have greater confidence and worry less about the patient’s prognosis over the next 12 months. Patients who are concerned about developing AMD due to family history or other risk factors value this technology. On the one hand, it is great when you can ease a patient’s concerns and let them know that their drusen is nonpathologic. On the other hand, if DA is abnormal, it enables me to initiate closer follow-up and monitor them with OCT and other testing on a more regular basis so they do not have to race to a specialist because I had not seen them for 12 months. Consider this: If a patient develops choroidal neovascularization (CNV) that is left to progress for 12 months, that patient will likely lose about five lines of visual acuity, in my experience. But if I already know that the patient has abnormal DA, I will monitor them closely so those five lines can perhaps be saved. The patient will also be more proactive about their disease management; knowing you have AMD can make a big difference in terms of compliance. Suddenly, with the possibility of going blind, recommendations to eat well, wear sun protection and quit smoking are taken much more seriously. How DA confirms early AMD
Until the commercialization of automated DA testing, AMD grading scales were limited by our ability to identify and characterize small drusen. We might know they were there, but we did not know whether they were harbingers of AMD. Now we do. Several peer-reviewed studies have shown that DA function is impaired from the earliest stages of AMD, with increasing impairment as the disease progresses (Owsley et al. 2001, Curcio et al.). AMD is present before drusen are clinically visible. Histopathological studies have shown that the retinal pigment epithelium (RPE) cells deposit locally generated cholesterol beneath the RPE cell layer and in Bruch’s membrane before drusen are formed (Pikuleva et al., Curcio et al.). This cholesterol accumulation impairs normal transport of vitamin A across Bruch’s membrane and causes oxidative stress and inflammation, which, in turn, predisposes patients to the downhill course of vision loss. Although these panmacular deposits will not become visible drusen for several years after their formation, they are most certainly causing a localized vitamin A deficiency that results in poor night vision – an early symptom of this chronic, progressive disease. Rod intercept time
The AdaptDx measures a patient’s rod intercept (RI) time, the number of minutes it takes for the eye to adapt from bright light to darkness at a standard threshold stimulus level. RI, as measured by the AdaptDx, provides a clear and objective measurement of retinal function with 90% sensitivity and specificity (Jackson et al.). An RI of less than 6.5 minutes indicates normal DA consistent with healthy photoreceptor function. An RI greater than 6.5 minutes indicates impaired DA, most often due to AMD in patients older than 60 years, unless there is a pre-existing hereditary retinal degeneration or significant vitamin A deficiency, which is rare in the U.S.
How I use DA test results
Patients with drusen are an at-risk population, but without knowledge about their AMD status as revealed by their RI through DA testing, most often they will be seen annually. On the contrary, impaired DA tells me that a patient has AMD and, therefore, requires more frequent monitoring. I do not race these patients off to the retina specialist. I treat them and test them regularly — usually for many years before they ever need the services of a specialist. DA does not change why I refer; it changes when I refer. When a patient has abnormal DA, I see them at least twice per year for OCT testing, DA testing, photography and electrodiagnostic assessment. I also start them on supplements and counsel them on the importance of diet, blue light filtering protection and smoking cessation.
By proactively monitoring for both structural and functional changes in my AMD patients, I am able to catch CNV much earlier. As a result, I am able to refer my patients with advanced AMD to my local retina specialists for treatment while they still have good best corrected visual acuity in both eyes. We have several cases of patients starting anti-VEGF treatment while maintaining 20/20 and 20/25 vision. The past 5 years of using DA testing in my practice has changed my outlook on this formerly devastating disease. High rate of missed diagnoses
Prior to the advent of automated DA testing, our profession as a whole was doing a terrible job at diagnosing AMD in a timely manner. A study published in JAMA Ophthalmology in 2017 revealed that optometrists and ophthalmologists failed to diagnose AMD about 25% of the time — even when they knew their findings would later be reviewed (Neely et al.). The cross-sectional study included 1,288 eyes of 644 adults. Each patient in the study had digital color fundus photos taken, which were reviewed by masked, trained graders who determined the presence or absence of AMD findings according to the Clinical Age-Related Maculopathy Staging system (Seddon et al.). The results revealed that one of four eyes studied was not diagnosed with AMD during the dilated fundus examination, despite these eyes having macular characteristics indicative of AMD in the fundus photos.
We need to stop relying on our ability (or inability) to manually detect this catastrophic disease. Currently, clinical AMD is more prevalent than glaucoma and diabetic retinopathy combined, and by the year 2050 it is estimated to double, according to research published in Archives of Ophthalmology (Rein et al.). Grading scale criteria are outdated, created before we knew about the link between impaired DA function and AMD. Consider that we overlooked central corneal thickness in glaucoma patients not long ago, but now we know better. Management of disease evolves with our understanding of how it works.
Pairing DA functional testing with structural exams and imaging makes us better doctors and much better diagnosticians. More importantly, it helps us buy time for our patients by keeping them under close observation while taking preventive measures as needed.
Cervantes-Castañeda RA, et al. Eye. 2007;doi:10.1038/sj.eye.6702691.
Jackson GR, et al. Invest Ophthalmol Vis Sci. 2014;doi:10.1167/iovs.13-13745.
Neely DC, et al. JAMA Ophthalmol. 2017;doi: 10.1001/jamaophthalmol.2017.0830.
Olsen TW, et al. Ophthalmology. 2004;doi:10.1016/j.ophtha.2003.05.030.
Owsley C, et al. Ophthalmology. 2001;doi.org/10.1016/S0161-6420(01)00580-2.
Owsley C, et al. Ophthalmology. 2016;doi:10.1016/j.ophtha.2015.09.041.
Pikuleva IA, et al. Prog Retin Eye Res. 2014;doi:10.1016/j.preteyeres.2014.03.002.
Rein DB, et al. JAMA Ophthalmol. 2009;doi:10.1001/archophthalmol.2009.58.
Seddon JM, et al. Ophthalmology. 2006;doi:10.1016/j.ophtha.2005.11.001.
For more information:
Glenn S. Corbin, OD, is in private practice at Wyomissing Optometric Center with offices in Wyomissing, Douglassville and Myerstown, Pa., and chief of the section of optometry at Penn State Health St. Joseph Medical Center, Reading, Pa. He can be reached at: email@example.com.
Disclosure: Corbin reports he is a consultant/speaker for Alcon, Maculogix, MiBo Medical Group, PRN and Takeda-Shire.