Abstract 4575: Stable Rates and Improving Survival Following Hospitalization for Subarachnoid Hemorrhage between 1986–2005
Introduction Primary prevention of cerebrovascular disease has improved in recent years as has the acute management of subarachnoid hemorrhage (SAH). It is unclear whether these changes have translated into reduced rates of SAH and improved short term prognosis at a population level. We describe trends in rates and 30 day case fatality following a first hospitalization with a SAH.
Methods Using the linked Scottish Morbidity Record and General Register Office databases, we identified all first hospitalizations and subsequent deaths in Scotland where subarachnoid hemorrhage was coded in the principal diagnostic position at discharge from 1986 –2005. Logistic regression was used to model deaths at 30 days adjusted for age, sex, comorbidities, socioeconomic deprivation and year of admission.
Results Between 1986 –2005, 3678 men (mean age 51 years) and 6435 women (mean age 56 years, P<0.05). From 1986 to 2005 unadjusted crude 30 day case fatality declined from 37%(95% CI 29 – 45) to 22%(16 –28) in men and 36%(30 – 42) to 27%(22–33) in women. Adjusted case fatality at 30 days increased with age, OR per year increase =1.033 (1.029 –1.036) and socioeconomic deprivation (deprived vs. affluent), OR=1.31 (1.13–1.51). Previous alcohol misuse (OR=1.46[1.1–2.0]), renal failure (OR=2.20[1.4 –3.4]) and cancer (OR=1.67[1.26 –2.21]) were also associated with poorer outcome. There was no difference between men and women OR=1.01(0.92–1.11). There was a significant reduction in adjusted 30 day case fatality over time OR=0.54(0.40 – 0.73), 2005 vs 1986.
Conclusion Population rates of SAH have remained stable over time. Short term case fatality has improved substantially in men and women. This is likely to reflect more aggressive acute management of SAH. However, marked age and socioeconomic inequalities persist in outcome following a SAH. Further research is required to understand why improvements in survival have not been realized equally in all groups.