A 56-year-old man with no history of trauma or neck manipulation presented with progressive right hemiparesis. Digital subtraction angiography showed a long-segment, smoothly tapered stenosis of the left cervical internal carotid artery (Figure 1A). This appearance does not by itself separate dissection from atherosclerotic disease. ^{Reference Debette and Leys1} Intravascular optical coherence tomography (OCT) performed before intervention showed a signal-poor mural mass with a sharply demarcated inner border, expanding the vessel wall and compressing the true lumen, the characteristic appearance of an intramural hematoma (Figure 1D–F). ^{Reference Jones, Attizzani and Given2,Reference Shen, Zhao, Mao and Nan3} The hematoma occupied a different circumferential sector at each imaged level so that the axis of luminal compression rotated progressively along the artery. A focal discontinuity of the luminal surface was seen distally (Figure 1E, triangle); tangential signal dropout at this eccentric catheter position is an alternative explanation that cannot be excluded. ^{Reference Tearney, Regar and Akasaka4} In spontaneous dissection, the mural hematoma may originate from an intimal entry tear or from rupture of the vasa vasorum, and the two cannot always be separated on imaging. ^{Reference Debette and Leys1} No distal re-entry was identified, so on either origin, the dissection plane behaved as a functionally blind-ended compartment. Because the hemiparesis continued to worsen despite dual antiplatelet therapy, a stent was deployed across the lesion. Control angiography showed a new severe stenosis in the previously patent proximal segment (Figure 1B), which resolved after a second overlapping stent (Figure 1C).

Figure 1.

Optical coherence tomography of a helical intramural hematoma of the left cervical internal carotid artery and its displacement after stenting. (A) Digital subtraction angiography before treatment shows a long-segment, smoothly tapered stenosis of the left cervical internal carotid artery (arrow). (B) After deployment of the first stent, a new severe stenosis appears in the previously patent proximal segment (arrow). (C) After a second overlapping stent, luminal patency is restored. (D) Three-dimensional rendering of the optical coherence tomography pullback; arrows indicate the levels shown in (E), numbered 1 (distal) to 5 (proximal). (E) Cross-sectional images at levels 1–5. Circles mark the IMH, which occupies a different circumferential sector at each level. The triangle marks a focal discontinuity of the luminal surface. FL denotes the false lumen at level 1; at this level, the distinction from a purely intramural collection could not be made with certainty. Asterisks mark the guidewire artifact. (F) Longitudinal reconstruction of the same pullback; the scale is in millimeters. *Wire artifact. FL = false lumen; IMH = intramural hematoma.

We interpret this new proximal stenosis as axial displacement of the hematoma. Blood is incompressible, and the compartment had no distal outlet, so radial force from the stent could redistribute the hematoma along the vessel axis but not decompress it, and the displaced volume accumulated immediately proximal to the stented segment. Stent-edge dissection and vasospasm are alternative explanations that we cannot formally exclude because no post-stenting pullback was obtained.

The case illustrates what OCT adds to angiography. Angiography is a luminogram: it showed only the residual channel and could not separate mural hematoma from plaque, define the longitudinal extent of the hematoma or reveal that the compressive vector rotated along the vessel, a geometry that makes apparent stenosis severity dependent on projection angle. The practical consequence is that stenting a closed intramural hematoma redistributes it rather than resolves it, and the second stent here was a bailout rather than a planned maneuver. The distance it travels cannot be predicted, but OCT defines the length of vessel that contains hematoma, and that whole length is at risk of new obstruction, not only the segment that appears stenotic on angiography.