Abstract
Purpose :
To use a national database for understanding endogenous endophthalmitis in America and identify risk factors for admission and mortality.
Methods :
The National Emergency Department Sample (NEDS) was queried from 2006 to 2017 with details encompassing emergency department (ED) visits: diagnostic codes, procedures, patient demographics, payment sources, total monetary charge, and hospital characteristics. Patients with diagnoses of endophthalmitis and septicemia were required for inclusion using International Classification of Diseases codes for precisely identifying cases of endogenous endophthalmitis, and to characterize patient comorbidities. P value of 0.05 was defined as statistically significant for all analyses.
Results :
A total of 6,400 patients with endogenous endophthalmitis were identified. Incidence increased from 0.10 (95% CI: 0.07-0.12) per 100,000 in the American civilian population in 2006 to 0.25 (95% CI: 0.21-0.30) in 2017 (p<0.05). Most were female (55.4%), had Medicare (53.5%), were in the first income quartile (29.3%) [top 25% of income bracket], lived in the southern region (40.5%), and presented to a metropolitan teaching hospital (66.6%) (p<0.05). There were no seasonal trends in admission. The median ED visit cost increased from $1,229 (interquartile range [IQR]: $690-$1,563) in 2006 to $2,529 (IQR: $1,538-$4,038) in 2017 (p<0.05). Mortality increased from 8.6% in 2006 to 13.8% in 2017 (p=0.94). On multivariate analysis, factors that predicted admission included older age, and the following comorbidities: pneumonia, endocarditis, renal/urinary tract infection (UTI), and intravenous drug use (IVDU). Factors associated with increased mortality included: human immunodeficiency virus infection (HIV)/immune deficiencies, heart failure, pneumonia, renal/UTI, and hepatic infections/cirrhosis. Patients with diabetes had a decreased odds ratio for mortality.
Conclusions :
Endogenous endophthalmitis has increased in incidence throughout America. Factors predicting increased hospitalization include older age, pneumonia, endocarditis, renal/UTI, and IVDU. Factors associated with mortality include HIV/immune deficiencies, heart failure, pneumonia, renal/UTI, and hepatic infections/cirrhosis. Additional exploration of the potential protective effect of diabetes from mortality in this context is needed.
This is a 2021 ARVO Annual Meeting abstract.