Abstract WMP55: Risk Factors and Composite Risk Score for Predicting Stroke After Sepsis

    loading  Checking for direct PDF access through Ovid


Background and Purpose: Sepsis has been identified as a risk factor for stroke, however, the underlying mechanisms remain unclear and risk factors that may predispose a patient to an increased stroke risk post-sepsis are unknown. The study aims to identify risk factors associated with post-sepsis stroke and subsets of the population that are most vulnerable to stroke after sepsis.Methods: The 2007 - 2009 California State Inpatient Database from the Health Care Utilization Project (HCUP) was used. Patients who were over the age of 18 and hospitalized with sepsis, defined by ICD-9 codes, were included. Patients who died during their sepsis hospitalization were excluded. The outcome of interest was a primary diagnosis of stroke, including ischemic and hemorrhagic, as defined through ICD-9 codes. Associations between risk factors and stroke were analyzed using multivariable logistic regression. A composite risk score was generated to assess the accuracy of prediction among the post-sepsis population.Results: Of the 114679 patients with diagnosis of sepsis, 0.5% (N=572) had a primary diagnosis of stroke within a year of their sepsis hospitalization. Significant predictors for stroke are listed in Table 1. A score was generated from these risk factors with points assigned based on beta coefficients. The ROC for the score is 0.6840. As the composite score increases, the risk of stroke increases (OR=1.46, 95% CI 1.40-1.53 for each point increase in the score). Post-sepsis patients with a score greater than or equal to 1 (N=35773) have the highest prevalence (prevalence = 31.19%) and are more likely to have a stroke. (OR= 3.18, 95% CI 2.69, 3.76)Conclusion: Independent risk factors for stroke after sepsis include valvular diseases, congestive heart failure, coagulopathy, lymphoma, peripheral vascular diseases, pulmonary circulation disorders, and renal failure.

    loading  Loading Related Articles