# CT perfusion lesions are present in most MRI confirmed lacunar strokes

## TL;DR

Most lacunar strokes confirmed by MRI show perfusion abnormalities on CT scans, but small or late-detected strokes may not.

## Contribution

The study reveals that CT perfusion detects perfusion lesions in 78% of lacunar strokes, despite limitations in current algorithms.

## Key findings

- CT perfusion detects focal perfusion abnormalities in 78% of lacunar strokes confirmed by MRI.
- Perfusion-negative strokes are smaller, less severe, and imaged later than perfusion-positive ones.
- Brainstem perforator strokes are less likely to show up on CT perfusion.

## Abstract

•Focal perfusion abnormalities are present in 78% of all lacunar strokes.•Low signal to noise ratio on subcortical CT perfusion limits interpretation.•Brainstem perforator strokes are less likely to be seen on CT perfusion.•Perfusion-negative lacunar strokes tend to be smaller and less severe.

Focal perfusion abnormalities are present in 78% of all lacunar strokes.

Low signal to noise ratio on subcortical CT perfusion limits interpretation.

Brainstem perforator strokes are less likely to be seen on CT perfusion.

Perfusion-negative lacunar strokes tend to be smaller and less severe.

Lacunar stroke remains diagnostically challenging in acute settings due to limited sensitivity of routine imaging. While CT perfusion (CTP) is widely used for evaluating stroke, its role in the identification of subcortical lacunar infarcts is less clear. Current core/penumbra algorithms used for CTP interpretation intentionally suppress small subcortical lesions rendering them unsuitable for lacunar stroke diagnosis. This retrospective, multicentre study evaluates the frequency of perfusion abnormalities in retrospectively confirmed lacunar stroke cases and characterizes clinical and radiological differences between cases with and without perfusion abnormalities.

We reviewed consecutive patients presenting to two comprehensive stroke centres between 2018 and 2022, who underwent acute multimodal CT and had a final diagnosis of acute lacunar stroke confirmed by diffusion-weighted MRI. Patients were included if no perfusion lesion was identified on algorithmic core/penumbra maps. CTP parameter maps were reviewed for the presence of focal perfusion abnormalities by stroke neurologists aware of the clinical scenario but blinded to MRI findings, followed by a second review with DWI co-registration to identify additional subtle perfusion abnormalities.

Of the 183 patients meeting inclusion criteria, 107 (58 %) demonstrated obvious perfusion abnormalities on blinded CTP review, an additional 36 (20 %) were found to have subtle lesions identified retrospectively with DWI guidance, resulting in an overall frequency of 78 %. Cases with no perfusion abnormality (22 %) had significantly smaller infarct volumes (median 0.7 ml vs 1.8 ml, p < 0.001), lower baseline NIHSS scores (median 2 vs 4, p = 0.005), and longer time from onset to imaging (median 8.4 h vs 4.5 h, p = 0.033).

Despite negative algorithmic CTP maps being negative, in DWI proven lacunar stroke, focal perfusion abnormalities are in fact present in most cases. CTP-negative cases appear to represent a distinct subgroup with smaller infarcts and milder deficits. These results highlight the influence of infarct volume and scan timing on lesion visibility and suggest that with advanced processing or machine learning assisted interpretation, CTP could play a greater role in the acute assessment of lacunar stroke.