Nationwide incidence of clinically diagnosed central retinal artery occlusion in Korea, 2008 to 2011.

PURPOSE: To define the incidence and demographics of clinically diagnosed central retinal artery occlusion (CRAO) in Korea. DESIGN: Nationwide population-based retrospective study using data entered into the Korean national health claims database from 2007 to 2011. PARTICIPANTS: Data of the entire Korean population (N = 47,990,761, based on the 2010 census) were analyzed. METHODS: We used the national health claims database to identify patients diagnosed with CRAO. Incident cases were those with no claims related to CRAO in 2007 and were included once on the earliest claims related to CRAO in the years 2008 to 2011. Incident cases had a disease-free period before diagnosis of at least 1 to 4 years. The average incidence rate of CRAO was estimated according to the entire Korean population. MAIN OUTCOME MEASURES: The person-time incidence rates of clinically diagnosed CRAO in Korea, including the age- and sex-specific incidence rates, were estimated. RESULTS: A total of 3464 CRAO cases (59.1% men) were identified. The incidence rate of clinically diagnosed CRAO during the study period was 1.80 per 100,000 person-years (95% confidence interval [CI], 1.74-1.86). The incidence rate among men and women was 2.15 (95% CI, 2.05-2.24) and 1.47 (95% CI, 1.39-1.54) per 100,000 person-years (male-to-female ratio, 1.47), respectively. The age-specific male-to-female ratios were constant between the ages of 30 and 89 years (range, 1.51-2.10 years). The highest incidence of 10.08 (95% CI, 8.80-11.35) per 100,000 person-years was observed in those aged 80 to 84 years (14.65 [95% CI, 11.90-17.40] and 8.00 [95% CI, 6.63-9.37] per 100,000 person-years for men and women aged 80-84 years, respectively). The incidence rate of CRAO increased exponentially with age until the 9th decade of life. CONCLUSIONS: To our knowledge, this is the first nationwide epidemiologic study of CRAO in individuals of all ages. The incidence rate of CRAO in Korea increased exponentially with increasing age and was highest among those aged 80 to 84 years. Moreover, the incidence rate in men was 1.47 times higher than that in women.

Machine learning-based prediction of acute kidney injury after nephrectomy in patients with renal cell carcinoma.

The precise prediction of acute kidney injury (AKI) after nephrectomy for renal cell carcinoma (RCC) is an important issue because of its relationship with subsequent kidney dysfunction and high mortality. Herein we addressed whether machine learning (ML) algorithms could predict postoperative AKI risk better than conventional logistic regression (LR) models. A total of 4104 RCC patients who had undergone unilateral nephrectomy from January 2003 to December 2017 were reviewed. ML models such as support vector machine, random forest, extreme gradient boosting, and light gradient boosting machine (LightGBM) were developed, and their performance based on the area under the receiver operating characteristic curve, accuracy, and F1 score was compared with that of the LR-based scoring model. Postoperative AKI developed in 1167 patients (28.4%). All the ML models had higher performance index values than the LR-based scoring model. Among them, the LightGBM model had the highest value of 0.810 (0.783-0.837). The decision curve analysis demonstrated a greater net benefit of the ML models than the LR-based scoring model over all the ranges of threshold probabilities. The application of ML algorithms improves the predictability of AKI after nephrectomy for RCC, and these models perform better than conventional LR-based models.

Retinal vein occlusion and pregnancy, pre-eclampsia, and eclampsia: the results from a nationwide, population-based study using the national claim database.

OBJECTIVE: To investigate the incidence of retinal vein occlusion (RVO) in pregnant women and in the subpopulation of pregnant women with pre-eclampsia/eclampsia compared to that in the age-matched general female population to determine if there is increased risk of RVO in pregnancy. DESIGN: Nationwide population-based retrospective study using data entered into the Korean national health claims database from 2007 to 2011. SETTING AND PARTICIPANTS: Of the incident RVO cases in the database, RVO cases that occurred during the pregnancy-associated period, which spanned a 52-week period from 40-weeks-before to 12-weeks-after childbirth, were identified. Of these cases, the presence of pre-eclampsia/eclampsia was determined. MAIN OUTCOME AND MEASURE: The standardized incidence ratios (SIRs) of RVO in the general pregnant population and in the pregnant population with pre-eclampsia/eclampsia were determined with respect to the age-matched general female population. RESULTS: Pregnancy-related RVO was identified in 33 cases from the 1.8 million women who experience childbirth during the study period, while the expected number of cases calculated by the direct standardization to the age-matched general population was 113. Of the 33 patients, 12 patients (36.4%) had pre-eclampsia or eclampsia. The SIR for the general pregnant population in reference to the age-matched general female population was 0.29 (95% CI, 0.20-0.41). In contrast, the SIR for the pregnant population with pre-eclampsia/eclampsia in reference to the age-matched general female population and the age-matched general pregnant population was 67.50 (95% CI, 34.88-117.92) and 246.50 (95% CI, 127.37-430.59), respectively. CONCLUSIONS AND RELEVANCE: The results suggest that pre-eclampsia/eclampsia is a risk factor for RVO, while pregnancy itself may not be a risk factor for RVO.