AIM: To describe the association between a history of diabetes and outcome among patients suffering an in-hospital cardiac arrest. METHOD: All patients suffering an in-hospital cardiac arrest in whom cardiopulmonary resuscitation (CPR) was attempted at Sahlgrenska University Hospital in Goteborg between 1994 and 2006 and at nine further hospitals in Sweden between 2005 and 2006. RESULTS: In all, 1810 patients were included in the survey, 395 (22%) of whom had a previous history of diabetes. Patients with a history of diabetes differed from those without such a history by having a higher prevalence of previous myocardial infarction, stroke, heart failure and renal disease. They were more frequently treated with anti-arrhythmic drugs during resuscitation. Whereas immediate survival did not differ between groups (51.7% and 53.1%, respectively), patients with diabetes were discharged alive from hospital (29.3%) less frequently compared with those without diabetes (37.6%). When correcting for dissimilarities at baseline, the adjusted odds ratio for being discharged alive (diabetes/no diabetes) was 0.57 (95% CL 0.40-0.79). CONCLUSION: Among patients suffering an in-hospital cardiac arrest in Sweden in whom CPR was attempted, 22% had a history of diabetes. These patients had a lower survival rate, which cannot simply be explained by different co-morbidity.