The disparity in long-term survival after a first stroke in patients with and without diabetes persists: the Northern Sweden MONICA study.

BACKGROUND Diabetes is an established risk factor for stroke. Compared to nondiabetic patients, diabetic patients also have an increased risk of new vascular events and death after stroke. We analyzed how differences in long-term survival between diabetic and nondiabetic stroke patients have changed over time, and if differences varied with respect to sex and age. METHODS This population-based study included 12,375 first-ever stroke patients, 25-74 years old, who were registered in the Northern Sweden MONICA Stroke Registry 1985-2005. Uniform diagnostic criteria for stroke case ascertainment were used throughout the study period. The diagnosis of diabetes was based on medical records or diabetes diagnosed during the acute stroke event. Patients were separated into four cohorts according to year of stroke and followed for survival until August 30, 2008. RESULTS The diabetes prevalence at stroke onset was 21%, similar in men and women, and remained stable throughout the study period. The diabetic patients were an average of 2 years older, more often nonsmokers and more likely to have antihypertensive treatment, antithrombotics, atrial fibrillation, and a history of myocardial infarction or transient ischemic attack than the nondiabetic patients. The total follow-up time was 86,086 patient-years during which a total of 1,930 (75.7%) of the diabetic patients and 5,744 (58.5%) of the nondiabetic patients died (p < 0.001). Median survival was 60 months (95% CI: 57-64) in diabetic patients and 117 months (113-120) in the nondiabetic patients. Survival improved significantly in both groups (p < 0.001). A Cox regression, adjusting for possible confounders (age, sex, antihypertensive medication, antithrombotics or other thrombolytic agents, history of myocardial infarction, type of stroke, diabetes, cohort and the diabetes-by-sex, diabetes-by-age and diabetes-by-cohort interactions), showed a hazard ratio of 1.67 (1.58-1.76) comparing survival in diabetic versus nondiabetic patients. The reduced survival in diabetic stroke patients was more pronounced in women (p = 0.02) and younger patients (p < 0.001). There was a tendency that the difference in survival decreased between the earlier cohorts and the 2000-2005 cohort, but the test for interaction did not reach statistical significance (p = 0.08). CONCLUSION Long-term survival after a first stroke has improved in both diabetic and nondiabetic patients. Survival is markedly lower in diabetics, especially in women and younger patients, and the disparity persisted over 24 years. Decreasing the disparity in stroke survival is a challenge for stroke and diabetes care. New treatment methods in combination with intense secondary prevention in diabetic patients, especially in younger women, are needed.