1605. Candida Endophthalmitis/Chorioretinitis: Incidence and Clinical Characteristics in a University Hospital
Poster Abstract Session: Mycology - There's a Fungus Among Us: Epidemiology
Friday, October 28, 2016
Room: Poster Hall
Background: There is paucity of data regarding the risk factors associated with endogenous ocular candidiasis. This study described and compared variables that may be associated with this entity. Methods: This was a case-control, retrospective cohort study from 7/1/2009 to 10/1/2014, at a university hospital and a VA-affiliated hospital, in patients with definite fungal infection (at least 1 positive blood culture for Candida
species and clinical signs of infection). Adults (age 18+) with a dilated eye exam performed by ophthalmology within 4 weeks of diagnosis of candidemia as an inpatient were included. Cases of Candida
Endophthalmitis/Chorioretinitis were defined as possible (Comorbidities with ocular disease)or probable (deep focal white infiltrates in the retina in addition to others) along with Controls with candidemia and no Candida eye disease. Patients whose clinical records were not available and those with a previous diagnosis/treatment of candidemia with ocular involvement done outside of these facilities were excluded. Chi-square, Fisher’s Exact, and Wilcoxon Rank Sum tests were used to compare demographic, clinical, and epidemiological variables between cases and controls. Multivariable logistic regression was used to determine independent risk factors for disease.
Results: 254 total non-duplicate candidemia cases were diagnosed during the study period (160 at the university hospital and 94 at the VA-affiliated hospital). 13 cases were identified proven (1) or possible (12) eye disease (5.1%) and matched in a 3:1 ratio with 39 controls. There was no difference in gender or age between the two groups. Cases were had higher APACHE 2 scores and were more likely to have a nasogastric tube. There was a trend towards significance for TPN as a risk factor 62% vs 31%, p=0.05 Controls were more likely to be Hispanic and had higher Charlson comorbidity scores. When these four variables were entered into a multivariable model, patients with a NGT were significantly more likely to have ocular fungal disease (OR=3.48, 95%CI=0.02-461.32). Though not statistically significant, broad-spectrum antibiotics and solid organ transplant were more prevalent in cases than controls.
Conclusion: NGT and TPN were independently associated with Candidaendophtalmitis/chorioretinitis in adult patients hospitalized with a fungal infection.
J. Cadena Zuluaga,
M. Lopez Vazquez,
C. R. Frei,
Findings in the abstracts are embargoed until 12:01 a.m. CDT, Wednesday Oct. 26th with the exception of research findings presented at the IDWeek press conferences.