Eyeworld Daily News

56 | EYEWORLD DAILY NEWS | MAY 18, 2020 DAILY NEWS ASCRS VIRTUAL ANNUAL MEETING All patients had a minimum of 5 years and a maximum of 10 years postop follow up, with refractive measurements, visual acuity, pachymetry, and serial topographies, he said. Visual acuity, topography, and mani- fest refractions were performed at months 1, 3, 6, and 12, and after 1 year were performed every 6 months for a minimum of 60 months. In all follow-up visits (up to 5 years), the maximum keratometry values improved significantly (the mean change at 5 years was –2.06 D with a p-value of 0.01). The average keratometry, uncorrected, and corrected distance visual acuity improved in all follow-up time- points, although not always with the level of statistical significance, Dr. Quesada said. In nine eyes (18%), keratoco- nus had progressed in terms of increased keratometry greater than 1 D at the last follow-up visit, despite the crosslinking treatment. In conclusion, he said that crosslinking appears to be safe and effective in the treatment of childhood keratoconus. Given the typical progression of keratoconus observed in pediatric patients, crosslinking should be considered as an im- portant part of the treatment, Dr. Quesada said. These and other paper pre- sentations on this topic can be viewed in SPS-315. Editors' note: Dr. Quesada has financial interests with AcuFocus, Clerio Vision, Johnson & Johnson Vision, Alcon, and PowerVision. Dr. Azzam has no financial interests related to his presentation. by Ellen Stodola Editorial Co-Director A paper session focusing on pediatric cornea covered topics like crosslinking, intracor- neal ring segments, femtosec- ond LASIK, and more in the pediatric population. Daniel Azzam, MD, Irvine, California, shared "A Novel Epidemiological Approach to Geographically Mapping Pop- ulation Dry Eye Disease in the United States through Google Trends." "Dry eye is one of the most commonly seen conditions in the eye clinic, accounting for up to a quarter of the patients coming in," he began. Tradi- tional dry eye epidemiology uses surveys, which require costly resources and only pro- vide a snapshot that is limited in time and space of the popu- lation being studied, Dr. Azzam said, adding that the 2017 TFOS DEWS reported that in the last 10 years there have been zero dry eye prevalence studies in the entire southern hemisphere. An alternative is digital epidemiology, using tools like Google Trends, Dr. Azzam said. This has suc- cessfully predicted infectious disease outbreaks in advance, he added. The primary objective of Dr. Azzam's study was to serve as a proof of concept for using internet epidemiological tools and geospatial data as a mapping technique for dry eye disease surveillance. A second- ary objective was to fill the spatiotemporal gaps in dry eye epidemiology by using Google Trends in order to geograph- ically map dry eye disease, analyze relationships between dry eye disease interest and environmental risk factors, and to forecast the future of dry eye disease epidemiology using predictive analytics. In Google Trends, the study looked at the following thematic categories that were analyzed: DED symptoms, DED therapies, and control terms. It also included specific words that were examined. Dr. Azzam said the study collected Google Trends data from 2004–2019 and also looked at Climate Data Collec- tion. Data was extracted and three different analyses were performed: geospatial analysis, environmental analysis, and forecasting analysis. The forecast of dry eye interest, Dr. Azzam said, found that time and season are sig- nificant predictors of dry eye disease. He said that the spring season particularly had the highest dry eye interest, while fall had the lowest. The geospatial analysis mapped dry eye interest, hu- midity, and temperature in the United States. With this, dry eye interest could be visualized in localized patches, especially in the coastal states, Dr. Az- zam said. Correlating it with high-temperature regions, we notice hot spots in these areas as well, he said, though there doesn't seem to be the same correlation in the driest areas. Meanwhile, the environ- mental analysis compared dry eye interest across the four U.S. Census regions and found no differences among the regions. When looking at temperature and dry eye interest, there was moderate correlation, but humidity and dry eye interest did not show a correlation. A number of other environmental variables were considered as well, Dr. Azzam said, including temperature, humidity, percentage of sun, sunshine hours, wind speed, rank in low pollution, coastal status, specific coastal regions (Atlantic coast, Pacific coast, Gulf coast, Great Lakes), and regions (west, midwest, south, northeast). The only ones that showed correlation were tem- perature, coastal status, and air pollution, he said. However, when performing a multivari- able regression analysis, they found that the only significant variables correlated with dry eye interest were temperature and coastal status. In conclusion, he said that this project showed that Google Trends can be used as a novel digital epidemiologic for geographically mapping the U.S. population of dry eye disease. Application of digital epidemiology as a population disease surveillance model and a driver of ophthalmology clinic logistics according to real-time changes in dry eye risk factors warrants further in- vestigation, he said. Dr. Azzam did note several limitations of the study, including that Google Trends data lacks the clinical information provided by traditional surveys, there is the possibility of external influ- ence on the data, there may be outliers, and DED search terms used might reflect other ocular surface conditions as well. Rodrigo Quesada, MD, San Salvador, El Salvador, shared a paper on corneal crosslinking as a treatment of keratoconus in pediatric patients. The study was a retrospec- tive case control series of 52 eyes in 50 patients, ages 2–17 years old. After 6 months of observation with topographic evidence of progressive ecta- sia, all eyes were crosslinked. Corneal crosslinking was performed with the IROC 1000 and IROC 2000 equipment, removing the epithelium and complying with the Dresden protocol. Paper session highlights crosslinking and dry eye disease

• Cover