Visual assessment of stroke scans can have negative impact on patient care

Eighteen physicians with experience using MRI for acute stroke imaging often visually estimated infarct volumes incorrectly, according to a multi-center study published in the Journal of NeuroInterventional Surgery. The approach could lead to a significant number of incorrect clinical decisions.

The clinicians assessed 32 MR scans over two independent sessions; they were asked to visually estimate the diffusion-weighted imaging (DWI) infarct volume in each case. Sensitivity, specificity and accuracy were based on available RAPID measurements for various volume cut-off points.

Overall, the raters achieved a mean accuracy of less than 90 percent for all volume cut-points, reported corresponding author, Robert Fahed, MD, with the department of interventional neuroradiology at Foundation Rothschild Hospital in Paris, and colleagues. Mean accuracy was lower for low volume categories (less than 21 milliliters [mL] and less than 31 mL) in which raters achieved a sensitivity of 61 percent and 82 percent, respectively.

Inter-rater agreement was “below substantial” for all DWI infarct volume cut-off points, and intra-rater agreement depended on the selected cutoff points, but was “substantial” in 55 to 83 percent of raters, according to the authors.

“In other words, it (visual assessment of DWI infarct volume) proved to be operator dependent and therefore lacking strong objectivity for its use as a pivotal clinical decision making tool for an invasive therapy,” wrote Fahed, and colleagues.

More importantly, the authors noted, using DAWN criteria instead of RAPID-generated measurements would have led to wrong thrombectomy decisions in 19 percent of cases. Two groups of patients in particular would have been denied the procedure despite meeting the necessary criteria, the authors noted.

The DAWN trial (Diffusion weighted imaging or CT perfusion Assessment with clinical mismatch in the triage of Wake-up and late presenting strokes undergoing Neurointervention with Trevo) has proven the benefits of thrombectomy in those with unknown or late onset strokes, with RAPID software-generated infarct volume measurements. However, such software is not available in all institutions. And in centers that use brain MRI for acute screening, clinicians are required to visually estimate DWI infarct volumes if RAPID is not available.

Fahed and colleagues did expect that visual assessment alone would not be accurate and reliable, and suggested other techniques, such as the semi-visual ABC/2 method, could be used as an alternative due to its proven reliability.

“Our study suggests that even with the use of visual reference aids, the visual assessment of DWI infarct volume lacks accuracy and reproducibility, and could result in a high proportion of erroneous thrombectomy decisions,” the authors concluded.