Background Abdominal obesity is a well-recognized cardiovascular risk factor. Conflicting data concerning its significance with respect to stroke have been discussed in recent years. The objective of this study was to analyze the association between anthropometric parameters and the risk of stroke and transient ischemic attacks (TIAs) in German primary care. Methods Patient recruitment in this large-scale epidemiological study was performed in 3188 representative primary care offices in Germany. Among 6980 study participants, 1745 patients with a history of stroke or TIA were identified and matched for age and gender with 5235 regional controls. Associations between standard anthropometric measures such as body mass index (BMI), waist-to-hip ratio, waist circumference, waist-to-height ratio, and cerebrovascular risk were investigated using logistic regression analysis with adjustment for age, gender, and vascular risk factors. Results BMI showed no significant associations with the risk of stroke or TIA in any of the applied mathematical models. Markers of abdominal obesity were associated with an increased risk of stroke or TIA in the unadjusted model (waist circumference: odds ratio [OR] 1.15; 95% confidence interval [CI], 1.00-1.32; waist-to-hip ratio: OR 1.21; 95% CI, 1.05-1.38; waist-to-height ratio: OR 1.25; 95% CI, 1.09-1.44, comparisons between top and bottom tertiles). After adjustment for vascular risk factors, all associations were insignificant. Conclusions Abdominal obesity is a stronger predictor of risk of stroke or TIA than BMI. However, the association between abdominal obesity and the risk of stroke or TIA is not independent of other vascular risk factors. Stroke-related weight changes should be considered in longitudinal studies examining the role of obesity in cerebrovascular disease.