Risk of herpes zoster among adults in the United States initiating immunosuppressive therapy: A retrospective cohort study.

Abstract

Patients immunosuppressed because of disease and/or therapy are at increased risk of developing herpes zoster (HZ) and related complications. However, HZ risk due to immunosuppressive medication remains underexplored across a broad spectrum of medication classes and conditions. To estimate HZ risk among US adults initiating immunosuppressive therapy, across underlying conditions. This retrospective study used administrative claims from the Optum Research Database (October 1, 2015, to December 31, 2022) to identify incident HZ diagnoses and the proportion of HZ with complications. Included patients were aged 18 years or older and had at least a 12-month baseline period of continuous enrollment with medical and pharmacy benefits. Analyses produced incidence rates (IRs) of HZ during periods of immunosuppression and nonimmunosuppression and compared patients initiating immunosuppressive therapy (immunosuppressed subcohort) vs matched controls who did not (nonimmunosuppressed subcohort). Cox regression analyses compared HZ hazards and a logistic regression model compared the difference in the occurrence of postherpetic neuralgia (PHN) following HZ diagnosis. In patients initiating immunosuppressive medications (N = 528,283), the overall HZ IR was 18.2 (95% CI = 17.9-18.4) per 1,000 person-years. Patients with the highest IRs were those treated with Janus kinase inhibitors (30.5 [95% CI = 25.0-36.9] per 1,000 person-years), rituximab (27.8 [95% CI = 25.2-30.6] per 1,000 person-years), or cyclophosphamide (27.5 [95% CI = 19.3-38.1] per 1,000 person-years). Patients had a higher HZ risk when using immunosuppressive medications (adjusted hazard ratio 2.86 [95% CI = 2.76-2.95]) and with increasing numbers of immunosuppressive medication classes (one vs none: 2.45 [95% CI = 2.37-2.54]; at least 2 vs none: 4.04 [95% CI = 3.87-4.21]). The immunosuppressed subcohort had a higher HZ hazard (adjusted hazard ratio: 1.39 [95% CI = 1.35-1.43]), higher frequency of HZ complications (21.4% vs 18.8% of HZ cases developed PHN, <i>P</i> < 0.001), and increased odds of PHN (adjusted odds ratio: 1.22 [95% CI = 1.14-1.31]) compared with the nonimmunosuppressed subcohort. Immunosuppressive medication use was associated with a higher risk of HZ and HZ-related complications. Measures to prevent HZ before immunosuppressive treatment initiation should be considered according to Advisory Committee on Immunization Practices recommendations and other professional organizations.