Skip to main content

  • Frailty Raises the Risk of Endogenous Endophthalmitis

    By Lynda Seminara
    Selected by Russell N. Van Gelder, MD, PhD
    Add to My Bookmarks

    Journal Highlights

    Ophthalmology, December 2022

    Download PDF

    As the proportion of aging adults con­tinues to grow, frailty is soon expected to affect all areas of medicine. In a population-based study of patients with septicemia, Henry et al. looked at the effect of frailty on the development of endogenous endophthal­mitis (EE) and explored the clinical outcomes of patients hospitalized for septicemia. They found that among inpatients with bacteremia and EE, more than 22% had at least one characteristic of frailty.

    For this retrospective cohort study, the authors in­cluded adult inpatients from the National Inpatient Sam­ple (2002-2014) with a di­agnosis of bacterial septicemia. The EE diagnosis was determined from ICD-9 codes, and the validated Johns Hopkins assessment tool was used to classify pa­tients as frail or not frail. Multivariable logistic regression was applied to deter­mine ORs for rates of EE development and in-hospital mortality according to frailty status. The authors also explored links between frailty and organism class (proven by blood culture), length of hospital stay, and charges billed to insurance providers.

    EE occurred in 9,294 (.05%) of 18,470,658 inpatients with bacteremia. Of these, 2,102 (22.6%) had at least one frailty-defining feature. Malnutrition was the most common (68%), followed by decubitus ulcer (38%). The risk of EE development was 16.7% higher for frail patients (OR, 1.167) after con­trolling for age, sex, race, concomitant HIV/AIDS, pyogenic liver abscess, infectious endocarditis, cirrhosis, and chronically complicated diabetes. The likelihood of a frail patient with EE dying while hospitalized was 27.9% higher than for healthier patients, inde­pendent of age, sex, race, or comorbid­ity score. Frail patients also had higher rates of methicillin resistant Staph­ylococcus aureus bacteremia (14.3% vs. 10.9%; p = .000016), gram-nega­tive bacteremia (7.6% vs. 4.9%; p = .000003), and concurrent candidemia (10.4% vs. 7.0%; p = .0000004). Hos­pital stays were significantly longer for frail patients, and their insurance bills were significantly higher.

    The authors recommend incorpo­rating frailty into the existing knowl­edge base of EE risk factors and into future studies of EE risk stratification.

    The original article can be found here.