MONTREAL—Neurologists usually consider stroke and transient ischemic attack (TIA) to indicate a similar need for anticoagulation, but these events may not entail equivalent risks, according to a presentation given at the 11th World Stroke Congress.
An analysis of two-year follow-up data from the Global Anticoagulant Registry in the Field (GARFIELD-AF), which included more than 52,000 patients with newly diagnosed atrial fibrillation, showed that while patients with a history of stroke had significantly elevated rates of all-cause mortality and stroke, those with a history of TIA alone had rates of mortality and stroke that were virtually identical to those of patients with atrial fibrillation with no history of a cerebrovascular event.
Should Risk Calculators Be Revised?
“A history of TIA [alone] is not a reliable predictor of an increased risk for events,” said Werner Hacke, MD, Professor and Chair of Neurology at the University of Heidelberg in Germany. “A history of TIA should be removed from scores estimating the risk for stroke and systemic embolism in patients with atrial fibrillation.
“The weak predictive power of a history of TIA is probably caused by the relatively low reliability of establishing the diagnosis of TIA,” especially when the diagnosis is made by someone who is not a neurologist. “It is a fuzzy diagnosis,” even for a neurologist, and it consistently confounds other clinicians, he said. “I would be careful about deciding to anticoagulate a patient [with atrial fibrillation] based on a history of TIA. I am convinced that most people with a history of TIA [in the GARFIELD-AF registry] actually never had a TIA.”
Dr. Hacke has been unable to find the reason that TIA began to be considered to entail similar risks as stroke. “I asked all the old atrial fibrillation guys, ‘When did TIA start coming in and why?’ And none of them could remember,” he said. “At first, they talked about a history of cerebrovascular events, but then that became stroke and TIA, and it was as if it was one word” always said in the same breath. The CHADS2 score and the CHA2DS2-VASc score make a history of stroke or TIA, as well as thromboembolism, coequal risk factors that count for 2 points when calculating the thrombotic risk score for a patient with atrial fibrillation.
An Examination of Registry Data
To test whether this convention was appropriate, Dr. Hacke and his associates examined the consequences of a history of stroke alone, compared with those of a history of a TIA alone. They used data collected in GARFIELD-AF, a multinational registry with 51,670 patients newly diagnosed with atrial fibrillation who were followed for two years. All participants had complete information on their stroke and TIA history. This information included 5,617 patients with a history of at least one diagnosed cerebrovascular event, including 3,362 diagnosed with stroke alone, 1,788 diagnosed with TIA alone, and the remaining patients diagnosed with both events.
When compared with patients with atrial fibrillation without a history of any type of cerebrovascular event, those with a history of a stroke alone had a statistically significant 29% increased rate of all-cause death and a 2.3-fold higher rate of stroke after adjustment for baseline demographic and clinical differences. In contrast, the patients with a history of TIA alone had mortality and stroke rates during follow-up that did not differ significantly from the comparator group.
—Mitchel L. Zoler