We investigated the percentage and characteristics of patients with keratoconus who were already not indicated for CXL, and we evaluated general ophthalmic examination parameters that could be used to detect patients with keratoconus who were indicated for CXL. We found that at the time of their first visit to our hospital, 30% of the patients were already not indicated for CXL. The results of our analyses demonstrated that the patients with UCDVA (logMAR) <1.22 (converted to decimal visual acuity: ≥0.06) and those with CDVA (logMAR) <0.52 (converted to decimal visual acuity: ≥0.3) had an MCT of ≥400 μm and were likely to be ready for CXL.
A considerable number of studies have investigated the possibility of the early detection of keratoconus [5], and the following parameters have been reported to be useful: the corneal epithelial basal cell density [7], epithelial thickness [8], posterior corneal elevation [9], ratio of the anterior and posterior corneal surface areas [10], corneal light intensity distribution [11], anterior corneal higher-order aberrations [12], and total ocular higher order aberrations [12, 13]. These parameters are calculated from images taken using specialized equipment such as AS-OCT, topography, and Scheimpflug imaging. We have found no published studies of the early detection of keratoconus from two perspectives: the detection of keratoconus patients by a general ophthalmic examination, and the detection of keratoconus patients for whom CXL is indicated. It is significant that our present study identified a threshold value based on general ophthalmic examination parameters that could be used to determine how long an ophthalmic clinic (without corneal topography or AS-OCT) that treats patients with keratoconus should follow the patients.
Of the patients referred to our hospital for treatment of keratoconus, 30% were already not indicated for CXL at the time of their first visit. We defined an MCT <400 μm as not suitable for CXL according to the Dresden protocol [2]. The following modifications to the conventional protocol that has been used to make CXL possible for thin corneas have been reported: hypo-osmolar riboflavin [14], transepithelial CXL [15], iontophoresis-assisted CX [16, 17], a customized epithelial debridement technique [18], lenticule-assisted CXL [19], contact lens-assisted CXL [20], and individualized CXL [21]. However, the evidence regarding the safety and efficacy of these modifications is limited, and long-term follow-ups and large-scale studies are desirable [22]. From the viewpoint of screening to avoid delay, we believe that the criterion of an MCT <400 μm is appropriate at this time.
We propose the UCDVA and CDVA as general ophthalmic examination parameters that can be used to identify keratoconus cases indicated for CXL. It is well known that as the cornea becomes thinner, the visual acuity decreases due to increased irregular astigmatism; however, our present analyses revealed that CDVA with a hard contact lens was not a useful indicator for CXL. This may be because the average visual acuity in our <400-μm MCT group was also good at 0.7 (decimal visual acuity), and there were some cases in which the patient's CDVA with a hard contact lens exceeded 1.0 (decimal visual acuity). If the progress of patients with keratoconus is not monitored closely because they have good vision with correction by a hard contact lens and do not have problems in their daily life, we may miss the time point at which CXL is indicated.
We acknowledge several study limitations. There was a lack of consistency in how and when keratoconus was diagnosed at the referring clinics. However, when dealing with a disease such as keratoconus for which early detection methods are still being debated, it is difficult to exclude the possibility that technological advances such as the introduction of corneal topography and AS-OCT may provide a patient selection bias. In addition, our analyses were of a relatively small number of patients treated at a single institution. There is a possibility of double organ bias. A large-scale multicenter study is needed to further explore general ophthalmic examination parameters that can be used to identify keratoconus cases indicated for CXL.
In summary, it is important not to miss the time point at which CXL is possible by referring patients with keratoconus to a specialized facility for keratoconus when the following conditions is observed: (ⅰ) UCDVA (decimal visual acuity) ≥0.06, and (ⅱ) CDVA (decimal visual acuity) ≥0.3.