Association between atopic dermatitis and serious cutaneous, multiorgan and systemic infections in US adults

Abstract

Background: Atopic dermatitis (AD) is associated with barrier disruption, immune dysregulation, and immunosuppressing treatments that can increase the association with an unusual number of infections.

Objective: To determine whether adults with AD have an unusually large number of serious infections and related outcomes.

Methods: Data from the 2002 to 2012 National Inpatient Sample were analyzed, including an approximately 20% sample of all US hospitalizations (n = 72,108,077 adults). Prevalence of serious infections in hospitalized patients with vs without AD, length of stay, cost of care, and inpatient mortality secondary to serious infections were determined.

Results: The prevalence of serious infections expressed as a percentage (95% confidence interval) was higher in adults hospitalized with than in those without AD (42.1% [41.6-42.6] vs 25.4% [25.2-25.6]; P = .0002). In logistic regression models with multiple predictors (multivariable logistic regression models), AD was associated with 32 of 38 infections examined. Associated cutaneous infections included eczema herpeticum (odds ratio [95% confidence interval] adjusted for other predictors: 67.93 [47.93-96.28]), erysipelas (11.15 [9.47-13.1]), and cellulitis (4.53 [4.42-4.64]). Associated respiratory infections included aspergillosis (1.51 [1.21-1.88]) and tuberculosis (1.57 [1.41-1.76]). AD was associated with extracutaneous, multiorgan, and systemic infections, including infectious arthropathy (2.01 [1.84-2.20]), endocarditis (1.25 [1.12-1.39]), encephalitis (1.65 [1.40-1.96]), and methicillin-resistant Staphylococcus aureus infections (3.29 [3.17-3.42]). Patients with AD hospitalized with vs without any serious infection had an increased geometric mean cost of inpatient care ($8,273 [8,126-8,423] vs $7,179 [7,052-7,307]) and length of stay (5.3 days [5.2-5.3] vs 3.9 [3.9-4.0]; P = .0002), with $11 to $228 million excess annual costs from hospitalization with serious infections in adults with AD.

Conclusion: Adults with AD had increased cutaneous, respiratory, multiorgan, and systemic infections, which were associated with a considerable cost burden.

Figure 1

Association between AD-E and serious infections. Survey logistic regression models were constructed with AD-E as the independent variable and the respective serious infection as the dependent variable. Models included age, gender, race/ethnicity, and insurance status as covariates. Adjusted odds ratios and 95% confidence intervals were estimated. Forest-plots of the adjusted odds ratios and 95% confidence intervals are presented.

Figure 2. Prevalence of serious infections in hospitalized patients with AD (squares) vs. without AD (circles)

Proportion of hospitalized patients with AD (squares) and without AD (circles) is plotted over time (2002–2003, 2004–2005, 2006–2007, 2008–2009, 2010–2012).