Presepsis depressive symptoms are associated with incident cognitive impairment in survivors of severe sepsis: a prospective cohort study of older Americans

Abstract

Objectives: To test the hypothesis that presepsis depressive symptoms are associated with risk of new cognitive impairment in survivors of severe sepsis.

Design: Prospective longitudinal cohort study.

Setting: Population-based cohort of older U.S. adults interviewed as part of the Health and Retirement Study (1998-2006).

Participants: Four hundred forty-seven individuals with normal presepsis cognition who survived 540 hospitalizations for severe sepsis and completed at least one follow-up interview.

Measurements: Severe sepsis was identified using a validated algorithm in Medicare claims. Depressive symptoms were assessed prospectively using a modified version of the Center for Epidemiologic Studies Depression Scale. Cognitive function was assessed using versions of the Telephone Interview for Cognitive Status (TICS). Logistic regression with robust standard errors was used to examine associations between substantial depressive symptoms at any interview before sepsis and incident cognitive impairment (mild or moderate to severe cognitive impairment) at any interview after sepsis.

Results: The prevalence of substantial depressive symptoms in participants with normal cognition before sepsis was 38% (95% confidence interval (CI) = 34-42%). After severe sepsis, 18% (95% CI = 15-20%) of survivors had incident cognitive impairment. In unadjusted analyses, presepsis substantial depressive symptoms were associated with postsepsis incident cognitive impairment (odds ratio (OR) = 2.56, 95% CI = 1.53-4.27). After adjustment for demographics, health-risk behaviors, clinical characteristics of the sepsis episode, and presepsis TICS scores, substantial presepsis depressive symptoms remained the strongest factor associated with postsepsis incident cognitive impairment (OR = 2.58, 95% CI = 1.45-4.59).

Conclusion: Substantial presepsis depressive symptoms are independently associated with incident postsepsis cognitive impairment. Depressed older adults may be particularly at risk of developing cognitive impairment after a serious medical illness.

Figure 1. Point prevalence of cognitive impairment among severe sepsis survivors with normal cognition before sepsis

(Grouped by timing of HRS Interview Wave) Abbreviations (in alphabetical order): CI = confidence interval; IQR = interquartile range. Interpretive Example: The point prevalence of cognitive impairment was 17% (95%CI: 14%, 20%) at the first HRS interview after severe sepsis.