OBJECTIVES:The Infectious Disease Society of America recommends that all patients with candidemia undergo a dilated retinal exam to exclude endogenous Candida endophthalmitis. Our objective was to determine if there are significant risk factors in candidemic patients for developing endogenous Candida endophthalmitis METHODS: We conducted a retrospective study of all candidemic patients at three academic medical centers between 2012 and 2017. We extracted risk factors for Candida endophthalmitis based on prior literature and compared them between patients with and without endophthalmitis. We then built a multivariate logistic regression model to assess which ones were significant. RESULTS:We found 771 patients with candidemia. 120 (15.6%) of these patients were diagnosed with Candida endophthalmitis. In our logistic regression analysis, central venous catheter presence (OR 8.35), intravenous drug use (OR 4.76), immunosuppression (OR 2.40), total parenteral nutrition recipient (OR 2.28), race (OR 1.65), age (OR 1.02), and gender (OR 0.57) were risk factors for developing Candida endophthalmitis. Additionally, Candida albicans was more likely to result in Candida endophthalmitis (OR 1.86). CONCLUSIONS:This cohort represents the largest study of risk factors for candidemic patients who developed endogenous Candida endophthalmitis. Based on our findings, clinicians should develop targeted and cost-effective strategies for endophthalmitis screening.

BACKGROUND:Approaches for quantifying physical resilience in older adults have not been described.METHODS:We apply two conceptual approaches to defining physical resilience to existing longitudinal data sets in which outcomes are measured after an acute physical stressor. A "recovery phenotype" approach uses statistical methods to describe how quickly and completely a patient recovers. Statistical methods using a recovery phenotype approach can consider multiple outcomes simultaneously in a composite score (eg, factor analysis and principal components analysis) or identify groups of patients with similar recovery trajectories across multiple outcomes (eg, latent class profile analysis). An "expected recovery differential" approach quantifies how patients' actual outcomes are compared to their predicted outcome based on a population-derived model and their individual clinical characteristics at the time of the stressor.RESULTS:Application of the approaches identified different participants as being the most or least physically resilient. In the viral respiratory cohort (n = 186) weighted kappa for agreement across resilience quartiles was 0.37 (0.27-0.47). The expected recovery differential approach identified a group with more comorbidities and lower baseline function as highly resilient. In the hip fracture cohort (n = 541), comparison of the expected recovery differentials across 10 outcome measures within individuals provided preliminary support for the hypothesis that there is a latent resilience trait at the whole-person level.CONCLUSIONS:We posit that recovery phenotypes may be useful in clinical applications such as prediction models because they summarize the observed outcomes across multiple measures. Expected recovery differentials offer insight into mechanisms behind physical resilience not captured by age and other comorbidities.

Abstract Background The Infectious Disease Society of America recommends that all patients with candidemia undergo a dilated retinal examination to exclude invasive ocular candidiasis. However, it remains unclear whether there are patients with candidemia who do not warrant routine surveillance because the risk of ocular infection is low. Methods We conducted a retrospective cohort study of all patients with candidemia diagnosed at three academic medical centers (Duke, University of North Carolina and University of Virginia) from 2012 to 2017. We collected risk factors for invasive ocular candidiasis based on previous literature and compared them between patients with and without invasive ocular candidiasis. We then built a multivariate logistic regression model to assess which risk factors were significant for developing invasive ocular candidiasis. Results Overall, 942 patients were diagnosed with candidemia over the study period. The mean age was 55.9 years, 56% were men, 25% were non-White. Among these patients, 120 (13%) were also diagnosed with invasive ocular candidiasis, 10% with chorioretinitis and 3% with vitreous involvement. In our logistic regression analysis, central venous catheter presence [OR 8.35 (3.53, 19.77)], intravenous drug use [OR 5.02 (2.63, 9.58)], immunosuppression [OR 2.40 (1.55, 3.70)], total parenteral nutrition [OR 2.28 (1.42, 3.66)], non-White race [OR 1.65 (1.07, 2.55)], older age [OR 1.02 (1.01, 1.03)],and female gender [OR 0.57 (0.37, 0.89)] were risk factors for developing invasive ocular candidiasis. In addition, we found that persons with candidemia due to C. albicans were more likely to have invasive ocular candidiasis [OR 1.86 (1.22, 2.85)]. Conclusion This cohort represents the largest study of patients with candidemia who developed invasive ocular candidiasis to date. Based on our findings, clinicians should develop targeted and cost-effective strategies for endophthalmitis screening. Disclosures All authors: No reported disclosures.

A number of scoring tools have been developed to predict illness severity and patient outcome for proven pneumonia, however, less is known about the utility of clinical prediction scores for all-cause acute respiratory infection (ARI), especially in elderly subjects who are at increased risk of poor outcomes.

Despite advances in molecular techniques the etiology of acute respiratory infections (ARIs) is often difficult to differentiate either at the point of care or with advanced microbiological techniques. There is growing interest in host biomarker assays, including those based on gene expression patterns in circulating cells, to aid in differentiation of viral and bacterial diseases. However, there are concerns about how such tests perform in vulnerable aging populations where host responses are often muted. In order to assess performance of gene expression-based biomarkers, we enrolled patients presenting to the emergency department with clinical ARI and selected 184 individuals aged ≤25 and ≥60 years old with proven viral or bacterial ARI. Gene expression in peripheral blood was measured with Affymetrix microarrays. Published viral and bacterial signatures were applied to the data and Bayesian approaches were used to develop novel discriminative models. We noted a marked decline in signature performance between younger and older individuals in both viral (AUC 0.90 vs. 0.64) and bacterial (AUC 0.91 vs. 0.50) infections. Incorporation of age-related genomic changes was able to restore much of the signature performance in older individuals. When examining the genomic differences driving the drop in signature performance, we found marked perturbations in expression of immunoglobulin genes and pathways driving known immunoregulatory mechanisms that provide novel insights into an age-related decline in ARI-focused immunity. Pathogen class-specific host-based gene expression signatures offer great promise as diagnostic tools. However, altered immune responses in vulnerable populations such as the elderly are also manifested at the genomic level and can affect diagnostic signature performance. Age-specific alterations in the components of a diagnostic signature can minimize much of this effect, however this work highlights the need for consideration of age during biomarker development for infectious diseases. Furthermore, studies of age-related differences in biomarker performance can lead to important breakthroughs in our understanding of age-associated alterations in immunity. All authors: No reported disclosures.

Abstract Background Elderly individuals experience increased morbidity and mortality from acute respiratory infections (ARI), which are complicated by difficulties defining etiologies of ARI and risk-stratifying patients in order to guide care. A number of scoring tools have been developed to predict illness severity and patient outcome for proven pneumonia, however less is known about the use of such metrics for all causes of ARIs. Methods We analyzed risk factors, clinical course and major outcomes of individuals ≥60 years of age presenting to the emergency department with a clinical diagnosis of ARI over a 5-year period. Results Of the enrolled individuals 40 had proven viral infection and 52 proven bacterial infections, but 184 patients with clinically adjudicated ARI (67%) remained without a proven microbial etiology despite extensive workup. Age (71.5 vs. 65.9 years, P < 0.001) and presence of cancer and heart failure were strongly predictive of illness severe enough to require hospital admission as compared with treatment in the outpatient setting. Of those with proven etiology, individuals with bacterial infection were more likely to require hospital and ICU admission (P < 0.001). When applied to this study, a modified PORT score was found to correlate more closely with clinical outcome measures than a modified CURB-65 (r, 0.54 vs. 0.39). Jackson symptom scores, historically used for viral illness, were found to inversely correlate with outcomes (r, −0.34) and show potential for differentiating viral and bacterial etiologies (P = 0.02). Interestingly, a multivariate analysis showed that a novel scoring tool utilizing sex, heart rate, respiratory rate, blood pressure, BUN, glucose and presence of chronic lung disease and cancer was highly predictive of poor outcome in elderly subjects with all-cause ARI. Conclusion Elderly subjects are at increased risk for poor clinical outcomes from ARI and their clinical management remains challenging. However, modified PORT, CURB-65, Jackson symptom score, and a novel scoring tool presented herein all offer some predictive ability for all-cause ARI in elderly subjects. Such broadly applicable scoring metrics have the potential to assist in treatment and triage decisions at the point of care. Disclosures All authors: No reported disclosures.