Immunocompromise and herpes simplex virus type 1 encephalitis (HSVE): a nationwide cohort and case-control study
Alice Tang1, Shamik Bhattacharyya2, Edison Miyawaki2, Kazuki Yoshida3, Douglas Wilcox4, Karen Costenbader3, Daniel Solomon3, Hongshu Guan5, Hannah Lahey6
1Harvard, 2Brigham and Women'S Hospital, 3Harvard Medical School, 4Massachusetts General Hospital, 5Brigham and Women's Hospital, 6Quinnipiac University
Objective:

We aim to determine whether a relationship exists between immunocompromise as defined by autoimmune disease or immunosuppressive and immunomodulatory medications in patients and incident HSVE. 

Background:

Among infectious etiologies of encephalitis, herpes simplex virus type 1 (HSV-1) is commonest. The association between immune status and HSVE is unclear. Using a U.S. Medicaid database of 75.6 million persons, we evaluated the association between HSVE and autoimmune conditions, exposure to immunosuppressive and immunomodulatory medications, and other medical comorbidities. 

Design/Methods:

We used the U.S. Medicaid Analytic eXtract Database between the years of 2007–2010 from the 29 most populated American states. We first examined the crude incidence of HSVE in the population. We then age- and sex-matched adult cases of HSVE with a sufficient enrolment period to a larger control population without HSVE.  In a case-control analysis, we examined the association between HSVE and exposure to both autoimmune disease and immunosuppressive/immunomodulatory medications. Analyses were conducted with conditional logistic regression progressively adjusting for sociodemographic factors, Charlson comorbidity index, and non-autoimmune comorbidities.

Results:

Incidence of HSVE was ~3.01 per 105 person-years among adults. A total of 951 HSVE cases and 95,100 age- and sex-matched controls were compared. The HSVE population had higher rates of medical comorbidities than the control population. The association of HSVE and autoimmune conditions was strong, adjusted OR 2.6 (95% CI [2.2, 3.2]). The association of HSVE and immunomodulating medications had an OR of 2.2 [1.9, 2.6], also after covariate adjustment. When both exposures were included in regression models, the associations remained robust: 2.3 [1.9, 2.7] for autoimmune disease and 2.0 [1.7, 2.3] for immunosuppressive and immunomodulatory medications.

Conclusions:

In a large, national population, HSVE is strongly associated with pre-existing autoimmune disease and exposure to immunosuppressive and immunomodulatory medications.  The role of antecedent immune-related dysregulation may have been underestimated to date.

10.1212/WNL.0000000000202070