Abstract Background Subarachnoid haemorrhage (SAH) is a devastating disease, in the majority of cases caused by a rupture of an arterial intracranial aneurysm. The effect of systemic low-grade inflammation on incidence of SAH is not known. The purpose of this study was to evaluate the relationship between leukocyte count, a marker of systemic inflammation, and incidence of SAH in a large cohort study. Methods Leukocyte count and other cardiovascular risk factors were measured in 19,794 individuals (17,083 men and 2,711 women, mean age 44 years) participating in a health screening program between 1974 and 1981. Incidence of SAH in relation to baseline leukocyte concentration was studied during a mean follow-up of 27 years in participants free from previous stroke. Results Ninety-five participants had a SAH, corresponding to an incidence of 22 per 100,000 in women and 17 per 100,000 in men. The hazard ratio for SAH per one standard deviation (2.01 × 10 9 cells/L) increase of leukocyte concentration was 1.26 (95% CI 1.05-1.53, p = 0.014) after adjustment for several potential confounding factors including smoking. In sensitivity analysis, there was a significant association in smokers but not in non-smokers. Conclusions High leukocyte count at baseline was associated with increased incidence of SAH, although this relationship might be restricted to smokers. The results support the view that low-grade systemic inflammation could be involved in the pathogenesis of SAH, or constitute an early risk marker for the disease.