Proteinuria and risk of ocular motor cranial nerve palsy: a nationwide population-based study

Understanding the association between dipstick-detected proteinuria and oculomotor cranial nerve palsy (CNP) could have significant implications for understanding the mechanism of CNP development and for developing preventive strategies against CNP development in patients with proteinuria. This study aimed to determine the relationship between dipstick-determined proteinuria and ocular motor CNP using National Sample Cohort (NSC) database from Korea's National Health Insurance Service (NHIS). A nationwide population-based cohort study was conducted using data from the NSC database of Korea's NHIS. These data were collected from 2009 to 2018. A one-year time lag was established to prevent a situation in which the causal link was inverted. Participants aged 20 years or more who were diagnosed with proteinuria in 2009 were included. Individuals with specific pre-existing CNP, missing data, and those who were newly diagnosed with CNP or who died within one year of being tested were excluded. The study population was classified into six groups according to the degree of proteinuria (negative, trace, or between 1 + and 4 +) based on the urine dipstick test. A Cox proportional hazard regression analysis was performed to determine the linkage between the degree of proteinuria and ocular motor CNP. A total of 5,807 (0.14% of subjects) with ocular motor CNP were assigned to the ocular motor CNP group and 4,047,205 subjects were assigned to the control group. After full adjustment of comorbidities, hazard ratios (HRs) for 1 + , 2 + , 3 + and 4 + proteinuria groups were 1.449 (95% confidence interval [CI] 1.244-1.687), 2.081 (1.707-2.538), 1.96 (1.322-2.904), and 3.011 (1.507-6.014), respectively, for developing ocular motor CNP compared to the proteinuria-negative group. In subgroup analysis, the HR of patients with proteinuria for the development of ocular motor CNP was higher in the younger age group (less than 40 years) (P = 0.0242) and the group with DM (P = 0.04). Our population-based cohort study demonstrated a significant association between proteinuria and the incidence of CNP, suggesting that urine protein level could be a new clinical marker for predicting the development of CNP.