Abstract

Background and objective: Many patients with brain infarction (BI) lack traditional risk factors, suggesting that other factors (including infectious agents) might contribute to stroke risk. We investigated Chlamydia pneumoniae infection in a large cohort of patients with BI according to aetiological subtypes and carotid atherosclerosis.

Methods: We measured serum IgG and IgA to C pneumoniae by microimmunofluorescence in 483 BI cases and 483 controls matched for age, sex and centre. IgG ⩾1/32 and IgA ⩾1/24 were considered positive. Cases with BI proven by magnetic resonance imaging were consecutively recruited and were classified into aetiological subtypes. Carotid atherosclerosis (intima-media thickness, plaques, stenosis) was evaluated by duplex ultrasonography in all subjects following the same method and with central reading.

Results: C pneumoniae IgG seropositivity was not associated with BI (adjusted odds ratio (OR) 1.10, 95% confidence interval (CI) 0.80–1.51) and did not increase the risk of any aetiological subtype. Overall, C pneumoniae IgA was not associated with BI (adjusted OR 1.54, 95% CI 0.84–2.81), but there was a significant interaction with hypertension. IgA seropositivity increased the BI risk in patients without hypertension (adjusted OR 2.79, 95% CI 1.15 to 6.74). When stratifying BI into subtypes, IgA seropositivity increased the risk of BI of unknown cause, but without significant heterogeneity. There was neither association with atherothrombotic, lacunar and cardioembolic BI nor with carotid intima-media thickness, carotid plaques or stenosis.

Conclusions: We found no evidence that C pneumoniae seropositivity is associated with carotid atherosclerosis and BI, regardless of aetiological subtype; but it might be associated with an increased risk of BI in normotensive patients.