Clinical Predictors of Intracranial Bleeding in Older Adults Who Have Fallen: A Cohort Study
- Additional Document Info
- View All
OBJECTIVES: Emergency department (ED) visits among older adults are frequently instigated by a fall at home. Some of these patients develop intracranial bleeding. The aim of this study was to identify the incidence of intracranial bleeding and the associated clinical features in older adults who present to the ED after falling. DESIGN: Prospective cohort study. SETTING: Three Canadian EDs. PARTICIPANTS: A total of 2 176 patients age 65 years or older who presented to the ED with a fall were assessed, and 1753 were included. Inclusion criteria were a fall on level ground, off a bed, chair, or toilet, or from one or two steps within 48 hours. MEASUREMENTS: Emergency physicians recorded predefined clinical findings on initial assessment. The primary outcome was intracranial bleeding, diagnosed either by computed tomography at the index visit or within 42 days. Associations between baseline clinical findings and the presence of intracranial bleeding were assessed with multivariable logistic regression. RESULTS: A total of 1753 patients (median age = 82 y) were enrolled, of whom 39% were male, 35% were on antiplatelet therapy, and 25% were on an anticoagulant. The incidence of intracranial bleeding was 5.0% (95% confidence interval [CI] = 4.1-6.1). Overall, 76 patients were diagnosed at the index ED visit, and 12 were diagnosed during follow-up. Multivariable regression identified four clinical variables that were independently associated with intracranial bleeding: new abnormalities on neurologic examination (odds ratio [OR] = 4.4; 95% CI = 2.4-8.1), bruise or laceration on the head (OR = 4.3; 95% CI = 2.7-7.0), chronic kidney disease (OR = 2.4; 95% CI = 1.3-4.6), and reduced Glasgow Coma Scale from normal (OR = 1.9; 95% CI = 1.0-3.4). CONCLUSION: The incidence of intracranial bleeding in our study was 5.0%. We found significant associations between intracranial bleeding and four simple clinical variables. We did not find significant associations between intracranial bleeding and antiplatelet or anticoagulant use. J Am Geriatr Soc 68:970-976, 2020.
has subject area