Syphilis and the risk of stroke: a nationwide, population-based matched cohort study

OBJECTIVES: Syphilis has re-emerged globally recently. However, valid data on stroke risk following syphilis infection is limited. We aimed to assess the risk of stroke among patients with syphilis and to account for family related factors as well as confounding by test indication. METHODS: We conducted a nationwide, population-based matched cohort study in Denmark using linked registry and laboratory data from 2000-2022. Individuals aged 16-80 years with a positive syphilis serology test were matched 1:9 on sex and date of birth to individuals from the general population. To address confounding by test indication, a syphilis test-negative cohort was matched 1:1 to the syphilis-positive cohort. To assess familial and shared environmental factors, sibling cohorts corresponding to each primary cohort were included. Stroke incidence was ascertained through hospital registries. Hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated using Cox regression, and cumulative incidence functions were used to estimate absolute risk. RESULTS: The study included 3, 593 syphilis test-positive individuals, 3,590 syphilis test-negative individuals and 32,629individuals from the general population contributing over 300,000 person-years of follow-up. Median age was 36 years, 81% were male. The 10-year cumulative risk of stroke among syphilis-positive individuals was 1.8% (95% CI: 1.3-2.4). Stroke risk was markedly higher among syphilis-positive individuals compared with the general population (HR 5.4 (95% CI: 3.7-7.7), but not compared with syphilis-negative individuals (HR 1.5 (95% CI: 0.9-2.4). Siblings of syphilis-positive individuals also had an increased risk of stroke compared with siblings from the general population (HR 2.2; 95% CI 1.4-3.6), but not compared with siblings of syphilis-negative individuals. CONCLUSION: Individuals tested for syphilis - regardless of test result- have a higher risk of stroke than the general population. These findings suggest that confounding by test indication and shared familial factors, rather than syphilis infection itself, contribute substantially to the observed association.