Abstract
Many studies demonstrate positive associations between infections and Alzheimer disease (AD), suggesting that brain and/or systemic inflammation may impact AD pathogenesis. However, studies of meningitis and AD risk have been limited to animal models or small human cohorts in the USA. The objective of this study was to examine the relationship between incident AD and three different types of infections (meningitis, pneumonia, and urinary tract infections [UTIs]) using a population-based sample of US Medicare beneficiaries.
We created a case-control dataset by frequency matching 4:1 (control:case) by age group, sex, and month/year of the date of AD diagnosis or control selection date. We identified 52,628 newly diagnosed AD cases and 210,512 population-based controls ≥67 years of age using comprehensive Medicare claims data from 2016 to 2018. We classified infections using ICD-9-CM and ICD-10-CM diagnosis codes. We used logistic regression to calculate adjusted odds ratios (ORs) and 95% confidence intervals (CIs) to evaluate the association between AD and each infection separately. We lagged exposures up to 18 months and examined hospitalization or comorbid sepsis as a proxy for infection severity. Covariates included age, sex, race/ethnicity, and health care utilization.
AD was positively associated with meningitis in individuals hospitalized without superimposed sepsis with a 6-month lag (OR = 2.713, 95% CI: 1.277-5.764), and UTIs without superimposed sepsis with an 18-month lag (OR = 1.231, 95% CI: 1.101-1.376), and with superimposed sepsis with an 18-month lag (OR = 1.388, 95% CI: 1.050-1.835). There was no association between AD and pneumonia in individuals hospitalized with or without superimposed sepsis. When examining infections that occurred in the outpatient setting, the association between AD and UTI remained positive yet attenuated at all time points, however, the association became inverse between AD and pneumonia.
More severe infections, particularly meningitis, may be associated with a higher risk of AD, due to either unmasking of prodromal AD or acceleration of AD pathogenesis in susceptible individuals.