Background Variant angina has unique clinical characteristics: geographical preponderence of its prevalence and development of the disease late in the middle-aged persons. Furthermore, chronic activation of T lymphocytes has been demonstrated in the patients with variant angina. So we hypothesized that infection might be involved in the pathogenesis of coronary spasm. Methods In 60 patients with variant angina and 100 control participants, immunoglobulin G antibody to cytomegalovirus, hepatitis A virus, hepatitis B virus, hepatitis C virus, Helicobacter pylori, Herpes simplex virus type 1 and 2, and Chlamydia pneumoniae were determined. Infection burden divided into 0 to 3 seropositivities and 4 to 8 seropositivities. Results Although the frequency of seropositivity for each pathogen tended to be higher in the patients with variant angina, any one did not reach statistical significance. Sixty four percentage of patients had been exposed to more than 4 pathogens, but analysis showed that increasing pathogen burden was not significantly associated with coronary spasm. Multiple logistic regression analysis using risk factors and seropositivity status to each pathogen showed that smoking was the only independent predictive factor for coronary spasm (odds ratio 2.83, 95% confidenced interval 1.25-6.41, P=0.008). Conclusion Our data suggest that infection might not play a role in the genesis of coronary spasm. However, subgroup analysis and further study including a larger number of patients will be needed.