|
|||||||||
|
|||||||||
|
|||||||||
|
|||||||||
This study will develop a decision rule (or guideline) for physicians to use to determine which patients with transient ischemic attack, ("TIA", or "mini-stroke") seen in emergency departments (ED), are at high risk of a life threatening or disabling stroke with 7 days of their initial visit. During a TIA, an individual experiences a sudden loss of neurological function with complete resolution over minutes to hours. TIAs are common. 20,000 patients per year are diagnosed with TIA in Canadian EDs.
Most patients do well in the short term following a TIA; however, about 4 in 100 TIA patients have a stroke within 1 week of their TIA. Many of these patients who suffer a stroke die or have permanent neurological deficits. Stroke is the leading cause of adult disability and the 3rd leading cause of death in Canada. Some of these strokes could be prevented with medical therapy, if the cause of the TIA is identified. Defining the cause requires evaluation by a stroke physician and a sequence of investigations selected on the basis of patient characteristics. Unfortunately, with limited health care resources available most centres do not have a stroke specialist see all patients, nor do they perform all tests while patients are in the ED (i.e. head CT scan, ultrasound of the heart and neck arteries, EKG, blood work). Knowing who is likely to have a stroke is critical in deciding the priority of testing.
Currently, there are no well-developed guidelines to help physicians decide which TIA patients require urgent investigations. This study will derive a clinical decision rule to help provide physicians with criteria to determine high-risk TIA patients. High-risk patients would then be fully investigated, assessed by a specialist, and started on maximal medical management very quickly, while low risk patients could be evaluated safely over a longer time period. The results would be expected to improve patient outcomes through the efficient use of existing resources. This study is currently collecting data to derive a rule at 8 Canadian sites. We anticipate completing the rule derivation by April 2010.
Principal Investigator: Dr. Jeffrey Perry
Co-Investigators: Dr. Michael Sharma, Dr. Ian Stiell, Dr. Marco Sivilotti, Dr. George Wells, Dr. Andrew Worster, Dr. Marcel Emond
Funding Agency: Canadian Institutes of Health Research (CIHR)
Duration of Study: 4 years
For more information, please contact Jane Sutherland, Study Coordinator.
Background
Transient ischemic attack (TIA) is a clinical diagnosis defined as a focal neurological deficit
of presumed vascular origin lasting less than 24 hours.
The occurrence of a TIA identifies individuals at high risk for stroke.
Stroke is one of the leading causes of death and disability with approximately
50,000 new stroke patients every year in Canada.
While effective interventions are available for stroke prevention,
no clear criteria have been developed to identify which TIA patients are
most likely to benefit from early intervention.
Gladstone and colleagues
(CMAJ 2004)
reported univariate risk factors for TIA and
a week stroke risk of 4-6% using record linkage but did not generate a decision rule.
Rothwell and colleagues
(Lancet 2005)
generated a score to identify individuals
at high risk of stroke following TIA using a small cohort study from 1981-1986.
The cohort was too small to allow for multivariate analysis and
patients were not managed following current practice for stroke prevention.
Johnston and colleagues
(JAMA 2000)
used a Health Maintenance Organization (HMO) database and
chart review to generate risk factors for stroke following TIA.
This data is subject to the problems and potential biases of retrospective analyses.
The derived indicators ignore investigations (i.e. CT head) available in the ED.
Johnston, Rothwell and colleagues
(Lancet 2007)
attempted to validate a new rule, but
only had a sensitivity of 83% (95%CI 81-84) for stroke within 7 days.
This proposed rule does not provide the sensitivity required (i.e. close to 100%) to be used by clinicians.
An international survey of emergency physicians in 2005, conducted by our group,
identified a need for a clinical decision rule to guide investigations with TIA.
Our group has successfully derived and validated several clinical decision rules including:
Ottawa Ankle Rules, Ottawa Knee Rule, Canadian Cervical Spine Rule,
Canadian CT Head Rule for minor head trauma and the Canadian SAH Rule.
No prospectively derived and validated clinical decision rule exists to identify TIA patients
requiring urgent intervention for early stroke.
We conducted a prospective cohort pilot study in which we enrolled 150 patients with
a final ED diagnosis of TIA in 3 months, representing 86% of eligible patients.
4 of these patients had a stroke within 7 days of initial ED visit.
Objectives
The goal of this study is to derive a decision rule to allow emergency physicians
to identify TIA patients at high risk for impending stroke and
who require rapid investigation, neurologist assessment and maximal stroke prevention treatment.
The specific objectives are to:
i) Apply standardized clinical assessments to patients with TIA;
ii) Determine the interobserver reliability of the clinical information;
iii) Determine the statistical association between the clinical findings and
the diagnosis of stroke within 7 days of TIA; and
iv) Use multivariate analysis to develop a highly sensitive clinical decision rule
for determining a high risk group for stroke within 7 days following diagnosis of TIA.
Methods
This phase I derivation study will be a 4 year prospective cohort study conducted at 9 Canadian EDs.
The study population will be all alert patients over 18 years of age with an ED diagnosis of TIA.
Standardized clinical variables including historical features, physical findings and
results of simple tests, will be collected by physicians in the ED prior to ED discharge/hospital admission.
Interobserver agreement by a second physician will be completed when possible.
The primary outcome measure is stroke, defined by a focal neurological deficit
of vascular origin lasting more than 24 hours.
An independent panel will adjudicate all endpoints.
The data analysis will include univariate and inter-observer kappa values
of clinical features for patients who have a subsequent stroke within 7 days of their diagnosis of stroke.
Variables with p-values <0.20 and kappa values >0.60 will be evaluated with
logistic regression and chi-squared recursive partitioning analysis to derive the rule.
5,550 patients with TIA will be enrolled including 150 patients with stroke within
7 days of ED diagnosis of TIA.
This sample size will allow a rule to be derived with 100% sensitivity (95% CI 98-100%).
Future studies will validate and test implementation of the rule.
Importance
This derivation study will develop a highly sensitive clinical decision rule
to determine which TIA patients are at high risk of early stroke within 7 days of TIA.
This rule will improve nationwide patient care and potentially lower health care costs
by identifying high-risk patients for stroke in the days following TIA.
This will allow physicians to prioritize care and direct finite resources
to the highest risk patients first to achieve maximal stroke prevention.
| |
| Terms of use and legal notices | |
| Copyright © 2005-2010 Ottawa Hospital Research Institute | Email Questions or Comments to emerg@ohri.ca |
Email Questions or Comments to emerg@ohri.ca
Copyright © 2005-2009 Ottawa Hospital Research Institute
An Institute of The Ottawa Hospital and the University of Ottawa
