A population-based incidence of M2 strokes indicates potential expansion of large vessel occlusions amenable to endovascular therapy

Abstract

Background: M2 occlusions may result in poor outcomes and potentially benefit from endovascular therapy. Data on the rate of M2 strokes is lacking.

Methodology: Patients with acute ischemic stroke discharged over a period of 3 years from a tertiary level hospital in the 'stroke belt' were evaluated for M2 occlusions on baseline vascular imaging. Regional and national incidence was calculated from discharge and multicounty data.

Results: There were 2739 ICD-9 based AIS discharges. M2 occlusions in 116 (4%, 95% CI 3.5% to 5%) patients constituted the second most common occlusion site. The median National Institute of Health Stroke Scale (NIHSS) score was 12 (IQR 5-18). Good outcomes were observed in 43% (95% CI 34% to 53%), poor outcomes in 57% (95% CI 47% to 66%), and death occurred in 27% (95% CI 19% to 37%) of patients. Receiver operating characteristics curves showed the NIHSS to be predictive of outcomes (area under the curve 0.829, 95% CI 0.745 to 0.913, p<0.0001). An NIHSS score ≥9 was the optimal cut-off point for predicting poor outcomes (sensitivity 85.7%, specificity 67.4%). 71 (61%) patients had an NIHSS score ≥9 and 45 (39%) an NIHSS score <9. The rate of good-outcome was 22.6% for NIHSS score ≥9 versus 78.4% for NIHSSscore <9 (OR=0.08, 95% CI 0.03 to 0.21, p<0.0001). Mortality was 42% for NIHSS score ≥9 versus 2.7% for NIHSS score <9 (OR=26, 95% CI 3.3 to 202, p<0.0001). Infarct volume was 57 (±55.7) cm³ for NIHSS score ≥9 versus 30 (±34)cm³ for NIHSS score <9 (p=0.003). IV recombinant tissue plasminogen activator (rtPA) administered in 28 (24%) patients did not affect outcomes. The rate of M2 occlusions was 7 (95% CI 5 to 9)/100 000 people/year (3%, 95% CI 2% to 4%), giving an incidence of 21 176 (95% CI 15 282 to 29 247)/year. Combined with M1, internal carotid artery terminus and basilar artery, this yields a 'large vessel occlusion (LVO)+M2' rate of 31 (95% CI 26 to 35)/100 000 people/year and a national incidence of 99 227 (95% CI 84 004 to 112 005) LVO+M2 strokes/year.

Conclusion: M2 occlusions can present with serious neurological deficits and cause significant morbidity and mortality. Patients with M2 occlusions and higher baseline deficits (NIHSS score ≥9) may benefit from endovascular therapy, thus potentially expanding the category of acute ischemic strokes amenable to intervention.

Keywords: M2; incidence; stroke; thrombectomy.

Figure 1

Acute occlusion of a left superior division. The first column shows axial and sagittal multiple intensity projection images localizing the occluded superior division (white arrows). The second column shows abnormal time-to-peak perfusion maps (axial, coronal, sagittal) highlighting the ischemic territory corresponding to the occluded superior division. The third column shows non-contrast CT (axial, coronal, sagittal) images obtained after 24 hours showing a large acute infarct involving the superior division and corresponding to the time-to-peak maps.

Figure 2

Acute occlusion of a left inferior division. The first column shows axial and coronal multiple intensity projection images localizing the occluded inferior division (white arrows). The second column shows abnormal time-to-peak perfusion maps (axial, coronal, sagittal) highlighting the ischemic territory corresponding to the occluded inferior division. The third column shows volumetric diffusion-weighted images (axial, coronal, sagittal) from an MRI examination performed after 48 hours showing a large acute infarct involving the inferior division and corresponding to the time-to-peak maps.

Figure 3

The chart shows distribution of all vascular occlusion sites (n=619). The rate given above each bar is n (%, 95 CI) based on the entire sample (n=2739).