LAUSR

Descemet stripping automated endothelial keratoplasty (DSAEK) has undergone significant advancements in techniques, indications, and outcomes in recent years. Traditionally, DSAEK grafts were approximately 150 to 200 mm; however, surgeons have recently begun to use thinner grafts. In 2012, Busin et al1 coined the term “ultrathin DSAEK (UT-DSAEK)” to refer to donor tissue grafts thinner than 100 mm. Multiple studies have reported that UT-DSAEK provides better visual acuity outcomes while offering comparable levels of graft failure, endothelial cell loss, and graft dislocation as traditional DSAEK.2 However, there is significant variance in the exact thickness of donor grafts, with studies including lenticule thicknesses ranging from 61 mm to 165.6 mm under the category of UT-DSAEK.3,4 Therefore, the analysis of pooled data of UT-DSAEK studies has often included grafts that would be better classified as DSAEK.2 More recently, there has been a drive toward thinner grafts that would offer a balance between ease of tissue handling and offer postsurgical visual acuity comparable with Descemet membrane endothelial keratoplasty (DMEK). Cheung et al3 introduced the term “nanothin DSAEK (NTDSAEK)” as a description for DSAEK grafts less than 50 mm. These investigators reported the results of NT-DSAEK compared with DMEK and found that although patients with DMEK achieved faster initial visual recovery after surgery, visual acuity outcomes and complication rates between patients with NT-DSAEK and DMEKwere similar after 1 month. These types of results, in conjunction with the increased availability of consistently thinner donor tissue, have led to a wave of interest in the viability of NT-DSAEK as a procedure comparable with DMEK.5 Large randomized clinical trials comparing NT-DSAEK with DMEK are an area of interest among many researchers. As corneal surgeons are routinely performing the full spectrum of DSAEK, UT-DSAEK, NT-DSAEK, and DMEK, it is becoming clear that more standardized nomenclature is needed to accurately reflect the type of endothelial keratoplasty (EK) being performed. Until this point, researchers have varied on their definition of quantitative cutoffs for a procedure to be considered DSAEK, UT-DSAEK, or NT-DSAEK. In addition, this has surgical importance because a 100 mm EK graft, for example, will have a difference in insertion, ease of manipulation, and attachment as compared to a 40-mm graft. Therefore, what 1 surgeon may consider UTDSAEK may be considered as NTDSAEK by another surgeon. Clearly, a need for a common language exists to more accurately identify the type of EK procedure being referenced. We believe that the current nomenclature system has several issues: First, UT-DSAEK was originally described to reflect graft thicknesses less than 100 mm. However, the range of tissue thickness that is currently considered UT-DSAEK is variable, including thicknesses much higher than 100 mm and therefore not originally meant to be under this classification. Therefore, we suggest that the term “Sub100-DSAEK (,100 mm)” be used to more accurately reflect this type of procedure. We believe this is important for consistent comparison, especially if we are to use the UT-DSAEK results for comparison with other EK procedures. Reanalyzing the reported results of Sub100-DSAEK with the exclusion of .100 mm grafts may yield interesting observations about the outcomes of each procedure. Second, we posit that the term “NTDSAEK” is misleading and inaccurate. “Nano” implies nanometer, which is a unit of measurement equivalent to one-billionth of a meter, whereas graft thickness is customarily measured in micrometers, which is equivalent to one-millionth of a meter.6,7 The term “nano” therefore references a unit of measurement that is inaccurate by a factor of a thousand. Furthermore, different surgeons may have different cutoffs for what they personally accept as NTDSAEK, including tissue less than 60 mm rather than 50 mm.8 Marketing terms such as NT-DSAEK should be replaced by accurate anatomical nomenclature, especially before widespread adoption of an inaccuracy as seen with “nanophthalmos.” Therefore, we suggest that the term “Sub50-DSAEK (,50 mm)” can be used to refer to DSAEK grafts less than 50 mm. This allows the originally intended thickness of NT-DSAEK, as described by the initial authors, to be maintained as well. Third, pre-Descemet endothelial keratoplasty (PDEK) may represent a classification dilemma because it may be thought of as a subset of DSAEK. As originally described, the average PDEK graft was 28 6 5.6 mm; other studies have reported a thickness of approximately 30 mm.9,10 Although it may make sense to refer to this procedure as “Sub30-DSAEK,” given that the creation of the PDEK lenticule involves pneumodissection rather than the use of an automated microkeratome, we believe that PDEK is uniquely different than DSAEK and the term is specific enough as it currently stands. Ideally, a multinational panel of expert corneal specialists should discuss these limitations and agreeable nomenclature as a “working group” or at an international ophthalmology symposium. Other topics such as ideal methods to measure graft thickness (ultrasound vs. optical coherence tomography) and stratification according to preopertaive versus postoperative thickness may also be considered. To start the conversation, we propose a tentative classification system of stratifying various DSAEK procedures into the following categories (Table 1). We acknowledge that the introduction of additional terminology may initially cause further confusion. For example, surgeons may potentially reject precut tissue offers because it may be a few microns thicker or thinner than their preferred size. Others may believe that the current system, despite its inaccuracies, is sufficient for their personal practice. However, we believe that classifying DSAEK into these categories will allow for accurate comparison among these subsets and against PDEK/