For patients who receive their first scan within 3 hours of symptom onset, authors of a new Japan-based study recommend repeat DWI assessment.
Diffusion-weighted imaging (DWI) performed within 3 hours of the onset of stroke symptoms may provide a false-negative result for transient ischemic attack (TIA), according to a small Japanese study. The authors recommend that imaging of suspected stroke patients should still be undertaken as soon as possible but that some may require a follow-up DWI to confirm the absence of TIA.
“In Japan, MRI is widely available, and access to MRI in early stages of stroke is relatively easy, lead authors Kenji Shono, MD, and Junichiro Satomi, MD, PhD (Tokushima University, Tokushima, Japan), told Neurovascular Exchange in an email. DWI can sometimes be provided so rapidly, they said, that it is too soon for evidence of TIA to be apparent.
“In all the TIA cases at our institution, the average latency from stroke onset to an MRI examination was 3.1 hours; in cases that were subsequently detected to be false negative, the average latency was 1.7 hours,” reported the authors, whose findings were published online ahead of the July 2017 issue of Stroke.
To find out the minimum window of time needed for detecting TIA, the investigators identified 73 consecutive patients with TIA who underwent an initial DWI on admission and again 24 hours after admission. They divided the patients into two groups: Group 1 consisted of the 9 patients (12%) for whom the initial DWI findings were not indicative of TIA but the follow-up DWI were (ie, the false negative group). Group 2 consisted of the remaining 64 patients, who showed evidence of TIA during both scans.
They determined that the latency from TIA onset to initial DWI was 1.7 ± 0.6 hours for the false negative group (range, 1-2.8 hours) and 3.3 ± 2.6 hours for the second group (range, 35 minutes to 12 hours). The probability of false-negative findings on initial DWI decreased as the latency from TIA to initial DWI increased, disappearing completely after the latency reached 3 hours.
Probability of False-Negative TIA by Time From Symptom Onset to DWI
|
After 1 Hour |
25% |
|
After 2 Hours |
21% |
|
After 3 Hours |
7% |
|
After > 3 Hours |
0 |
Out of the 73 cases in their study, reported the authors, 45 (62 %) underwent an MRI examination within 3 hours of stroke onset, and 9 out of these 45 cases (20 %) were false negatives.
Multivariate analysis revealed that short latency (i.e., ≤ 2 hours) from TIA onset to initial DWI was an independent risk factor for false-negative findings (odds ratio 13.11, 95% CI 1.07-161.38, P = 0.045).
Follow-Up DWI May Be Needed
The authors emphasized that MRI should still be conducted as soon as possible following suspected stroke symptoms, in order to identify patients who might benefit from tPA. But “it is essential to perform a repeat MRI examination in cases where TIA is suspected,” they advised, adding, “It is known that cases of TIA with infarction confirmed [on] imaging [have] a higher risk of recurrent stroke. The current results provide important insight [for not missing] the detection of positive findings in the initial imaging examination.
“Inter-rater variability is known to be large in TIA diagnosis,” Shono and Satomi continued. “In particular, diagnosing TIA can be challenging for the physicians in the emergency rooms who are not specialized in stroke.” Their results, they said, “are very clear-cut and would support an improved quality of TIA examinations by emergency room physicians and residents.”
The researchers also noted that as imaging technology continues to improve, the definition of TIA will evolve. “We are presently experiencing a transition from time-based TIA to tissue-based TIA,” they said. “Patients with evidence of infarctions in an imaging examination [will likely] be diagnosed with cerebral infarction in the future, and hence the findings of DWI examinations would become even more important.”
An important next step in research, according to Shono and Satomi, is to determine whether false-negative findings are associated with a different prognosis.
Source:
Shono K, Satomi J, Tada Y, et al. Optimal timing of diffusion-weighted imaging to avoid false-negative findings in patients with transient ischemic attack. Stroke. 2017;Epub ahead of print.
Disclosures:
Shono and Satomi report no relevant conflicts of interest.
Related Stories:
Clinical-Diffusion Mismatch Predicts Infarct Growth Best in Ischemic Stroke
