Variety of algorithms can guide diagnosis, treatment of dry eye
Click Here to Manage Email Alerts
An algorithm is a set of instructions for solving a problem or accomplishing a task. One common example of an algorithm is a recipe, which consists of specific instructions for preparing a dish or meal.
– Investopedia
Once upon a time, dry eye disease (DED) was one of the simplest, most easily defined and, of course, least well treated diseases in all of eye care. At least at the point of care. My memory of our DED education during my mid-80s residency can be summed up as succinctly as, “If it’s dry, wet it.” It is as if ophthalmology and optometry channeled “old school” dermatology and called it a day. As I shared in my column on DED definitions, the first real efforts at defining DED harken back to 1995 with definitions from the National Eye Institute in the United States and the Japan Dry Eye Society.
- National Eye Institute: “Dry eye is a disorder of the tear film due to tear deficiency or excessive evaporation, which causes damage to the interpalpebral ocular surface and is associated with symptoms of ocular discomfort.”
- Japan Dry Eye Society: “Ocular surface epithelial damage caused by qualitative or quantitative abnormalities of tears.”
Interestingly, neither of these earliest landmark efforts at definition seem to have led to protocols to diagnose and treat DED, let alone a recipe for success. This distinction belongs to the authors of the 2006 Delphi report in which DED was referred to as dysfunctional tear syndrome. The Delphi panel stated that DED severity should be determined by cataloging signs and symptoms; ancillary tests were deemed unnecessary. A detailed history with emphasis on the quality of vision is taken. The exam called for is rather spartan: visual acuity; a slit lamp exam in which specific note is taken of the condition of the tear film and lid margins; fluorescein staining of the cornea and conjunctiva supplemented by only tear breakup time; and unanesthetized Schirmer test.
Four levels of increasing severity were then defined, from which level-specific recommendations were made. For example, treatments for level 1 (mild to moderate severity) would include mitigating causative environmental factors and the use of preserved artificial tears. Level 2 (moderate to severe) with signs of inflammation called for the short-term use of steroids or long-term therapy with cyclosporine A. With the exception of autologous serum tears, there was not much available to treat level 3 (severe) or level 4 (disabling) when this earliest algorithm was published.
The original Tear Film and Ocular Surface Society Dry Eye Workshop (DEWS) was highly influenced by the findings of the Delphi panel. DEWS I generated a severity-based treatment algorithm using the Delphi scale with little change. While the DEWS I panel was philosophically more welcoming to ancillary tests, none were included in the DEWS algorithm. Similar to the Delphi algorithm, the patient’s history (augmented by a validated questionnaire) is followed by a rather basic examination: a slit lamp exam with emphasis on tear meniscus and lid morphology (including gland expression), fluorescein staining and tear breakup time, supplemented by a Schirmer test without anesthesia. Treatment recommendations in the DEWS I algorithm are nearly identical to those in Delphi.
This historical review of the earliest known algorithms sets the stage for us to examine the three most current offerings: DEWS II, CEDARS and the American Society of Cataract and Refractive Surgery presurgical algorithm. Unlike their predecessors, there are clear differences between these modern algorithms. DEWS II continues to use a severity-based approach, whereas the CEDARS algorithm is driven by ocular surface disease subtype. True to its origin, the ASCRS algorithm is purpose driven (presurgical care), although it, too, can be used as the basis for a general DED protocol.
It is a daunting task to winnow down the encyclopedic DEWS II treatise on DED. It is best described as a staged algorithm dividing the dry eye world into two simple categories: aqueous deficient dry eye (ADDE) and evaporative dry eye (EDE). Once again, the evaluation begins with a questionnaire (Ocular Surface Disease Index is favorably reviewed) as well as a detailed inquiry about risk factors, external causes such as systemic disease and medications. DEWS II encourages point-of-care tests (tear osmolarity, MMP-9). The exam follows the familiar pattern of lid margin, gland expression and tear meniscus at the slit lamp. Tear breakup time is tested with fluorescein, and staining can be done with any or all of the vital dyes.
Treatment in DEWS II is similar to DEWS I. Options to treat mild to moderate disease now include dietary supplementation with omega fatty acids and the use of oil-based artificial tears for EDE. Newer anti-inflammatory treatments such as lifitegrast and intense pulsed light are now options for moderate to severe disease. Punctal plugs are recommended in stage 2 rather than stage 3. Long-term use of topical steroids is now discussed in stage 4 as well.
A different approach was taken by the CEDARS-led group in its January 2017 opus on dysfunctional tear syndrome. Here, the main thrust is to characterize the disease state in any given patient as classic ADDE, meibomian gland dysfunction with or without evaporation, goblet cell abnormality or exposure. A detailed history and questionnaire lead off as in the other systems. In-office testing of tear osmolarity and MMP-9 levels is an integral part of the workup, as is Schirmer testing with and without anesthesia. Notably, the CEDARS exam calls for both fluorescein and lissamine green staining of the cornea and conjunctiva. Additional tests such as corneal topography, meibomian gland imaging and cultures are discussed throughout.
Unlike the DEWS algorithms, treatment within a disease category is linear, beginning with the simplest intervention and moving largely in order to more intense or complex treatments if necessary. The authors highlight an important therapeutic principle, that of “incomplete resolution” of disease as opposed to “failed treatment.” If an early-level treatment improves signs and/or symptoms but does not completely resolve them, next-in-line treatment is added to the initial therapy.
Last, but far from least among the modern algorithms, we come to the ASCRS presurgical algorithm. The ASCRS model is a purpose-driven approach designed for quick diagnosis and speedy resolution of “visually significant ocular surface disease (OSD).” Anyone who performs preop workups for cataract or refractive surgery will recognize the workflow. The SPEED questionnaire is added to the intake history, and almost all testing (tear osmolarity, MMP-9, keratometry, topography) takes place before the patient is seen by a doctor. A positive screening sends the patient for additional testing (gland imaging, tear film lipid measurement, OCT, higher-order aberrations) to determine if the OSD is visually significant.
In addition to what we now know as standard operating procedure for a dry eye exam, the protocol asks us to “lift” the upper lid looking for floppy eyelid syndrome, “pull” on the lower lid to uncover signs of conjunctival dysfunction or exposure, and “push” on the lower lid to evaluate the quality of meibomian gland secretions. If there is no evidence of OSD or if it is not visually significant, the patient proceeds to surgery.
Patients with visually significant OSD are referred for treatment. Because these are surgical candidates, the treatment part of the ASCRS algorithm places a premium on a speedy resolution of visually significant OSD. Short-term pulse steroid therapy is a primary option, and immunomodulators such as cyclosporine A and lifitegrast are added if longer-term postop dryness is a concern. Treatments reserved for severe disease (eg, amniotic membrane therapy) are utilized earlier to accelerate improvement.
So, which one of these five DED recipes should you be using in your office? If you are a doctor who has a tertiary care DED practice in which you see and treat all levels of dry eye, it is my opinion that the CEDARS approach should form the backbone of your strategy. The linear approach of layering treatments depending on severity has been highly effective for us. For everyone else, I would look to the ASCRS algorithm as the foundation of your protocols. In most practices, DED is a disease of intermittent episodes of symptoms (flares) interspersed between longer periods of relative calm. A purpose-driven approach to either discomfort or blurred vision would be a nice fit.
Everyone should have a DED algorithm in their clinic. After all, we are all DED doctors.
- References:
- Behrens A, et al. Cornea. 2006;doi:10.1097/01.ico.0000214802.40313.fa.
- Jones L, et al. Ocul Surf. 2017;doi:10.1016/j.jtos.2017.05.006.
- Milner MS, et al. Curr Opin Ophthalmol. 2017;doi:10.1097/01.icu.0000512373.81749.b7.
- Ocul Surf. 2007;doi:10.1016/s1542-0124(12)70081-2.
- Starr CE, et al. J Cataract Refract Surg. 2019;doi:10.1016/j.jcrs.2019.03.023.
- For more information:
- Darrell E. White, MD, can be reached at SkyVision Centers, 2237 Crocker Road, Suite 100, Westlake, OH 44145; email: dwhite@healio.com.
Collapse
Read more about
Click Here to Manage Email Alerts