Time-varying association between herpes zoster infection and subsequent occurrence of stroke

Objectives: The aim of this study was to investigate the shape of the time-varying relationship between herpes zoster infection, nominally shingles, and the occurrence of stroke. Study design: Retrospective cohort study. Methods: Using the Italian Health Search Database, a cohort of patients aged ≥18 years who were registered between 2002 and 2021 was selected. In this cohort, a nested case-control analysis was used to model the time-varying distance (in months) between the dates of shingles and post-herpetic stroke, using a regression cubic spline, based on the odds of the occurrence of stroke compared with those without shingles. Results: The dataset comprised 42,513 cases (51.1% males; mean age [stanndard deviation {SD}]: 71.0 [11.8] years) and 425,124 related controls (51.1% males; mean age [SD]: 70.9 [12] years). In the first 12 months following shingles diagnosis, a rapid increase in the risk of stroke was observed, reaching an odds ratio of 1.31 (95% confidence interval: 1.21-1.41); subsequently, there was some risk reduction and a new symmetric increase within the first 4.2 years of follow-up, thus shaping a bimodal distribution. Then, a new increase in the stroke risk was reported, although less steep, which was followed by a regular risk reduction (still 10% higher compared with those without shingles), resulting in a right-skewed relationship between the time from the shingles diagnosis and the occurrence of stroke. This association was no longer statistically significant 13.1 years after shingles diagnosis. Conclusions: This study demonstrated that the risk of post-herpetic stroke has a short- and long-term association according to a risk continuum relationship. These findings confirm the relevance of vaccination coverage for herpes zoster.