NSAIDs Use During Herpes Zoster Infection and Stroke Risk: A Nationwide Case-Crossover Study

Herpes zoster (HZ) infection and long-term non-steroidal anti-inflammatory drugs (NSAIDs) use are established risk factors for stroke and other cardiovascular diseases. Given the paucity of evidence regarding an association between NSAIDs use and HZ on stroke risk, this case-crossover study, utilizing a nationwide, population-based cohort, aimed to investigate the effect of HZ infection and concurrent NSAIDs use on the incidence of stroke.

Using Taiwan's National Health Insurance database (2014-2020), we identified 336,075 patients with incident stroke. A case-crossover design comparing exposure to HZ and NSAIDs between the focal period (1-30 days before stroke) and referent period (366-395 days before stroke) was employed. Conditional logistic regression estimated adjusted odds ratios (aORs) for stroke risk associated with NSAIDs use during HZ episodes. Pre-planned subgroup analyses further examined such effects on stroke subtypes (ischemic stroke, hemorrhagic stroke, and transient ischemic attack [TIA]), across age groups (< 50, 50-64, ≥ 65 years) and in patients with various comorbidities, including immunocompromised and autoimmune diseases, cardiometabolic risk factors, and renal and liver diseases.

Combined HZ infection and NSAIDs use was associated with doubled stroke risk (aOR 2.05, 95% confidence interval [CI] 1.80-2.33) compared with periods without either exposure. For specific stroke types, the aORs were 1.94 (95% CI 1.65-2.29) for ischemic stroke, 1.81 (95% CI 1.34-2.43) for hemorrhagic stroke, and 2.81 (95% CI 2.06-3.85) for TIA. HZ episodes without NSAIDs (aOR 1.70, 95% CI 1.45-2.00) and NSAIDs use alone (aOR 1.42, 95% CI 1.40-1.44) showed lower but significant risk increment. In age-stratified analyses, individuals aged 65 years and older exhibited a significantly elevated stroke risk while concurrently utilizing NSAIDs during HZ episodes (aOR 2.19, 95% CI 1.92-2.62). Subgroup analyses demonstrated consistent elevated risks in patients with pre-existing comorbidities, particularly immunocompromised conditions (aOR 3.07, 95% CI 1.95-4.81) and renal disease (aOR 4.30, 95% CI 2.20-8.41).

Our findings demonstrate a significant association between HZ infection and NSAIDs use on stroke risk, particularly among individuals aged 65 years and older or those with pre-existing immunocompromised, cardiometabolic, and chronic conditions. The optimization of pain management strategies during HZ episodes is paramount to mitigate the risk of stroke while ensuring effective management of HZ-associated pain.