Deterioration following improvement with tPA therapy: Carotid thrombosis and reocclusion

Case report.

A 42-year-old, right-handed woman was seen 80 minutes after the acute onset of right hemiplegia, global aphasia, eye deviation to the left, and a right homonymous hemianopsia (NIHSS score 24). Her medial history included smoking, non–insulin-dependent diabetes mellitus, and peripheral vascular disease requiring bilateral femoral–popliteal bypasses. She had no history of cardiac or cerebral ischemia, and was not taking any antiplatelet treatment. A head CT scan showed a hyperdense left middle cerebral artery (MCA) and no hemorrhage.

At 90 minutes from symptom onset, a TCD was performed according to a published protocol,4 using a single-channel, 2-MHz portable unit (Multigon 500M, Yonkers, NY), and head-frame fixation (Marc 500, Spencer Technologies, Seattle, WA).

The TCD was consistent with a proximal M1 MCA and A1 anterior cerebral artery (ACA) occlusion ( figure, frame 1) followed by rapid progression to a terminal internal carotid artery (ICA) occlusion (see the figure, frame 2). Within 5 minutes, the patient became drowsy (NIHSS score 26). IV tPA was started at 120 minutes from symptom onset using a standard dose of 0.9 mg/kg (10% bolus, 90% infusion over 1 hour, maximal dose 90 mg).

• Download figure
• Open in new tab
• Download powerpoint

Figure. Sequential sonographic and clinical findings during IV tissue plasminogen activator (tPA) infusion. M1 middle cerebral artery (MCA) waveforms were obtained at 55 mm (upper row) and terminal internal carotid artery (TICA) at 68 mm (lower row) by switching between the depths at a constant angle of insonation. White arrows indicate high intensity, short duration microembolic signals (MES). The middle row is a graphic interpretation of transcranial Doppler (TCD) findings. The corresponding NIH Stroke Scale (NIHSS) scores are provided below each frame. Frame 1: Pre-tPA. Minimal flow signals in the M1 MCA, TICA, and A1 anterior cerebral artery (ACA) segments. Frame 2: Bolus. Progressive TICA occlusion with decreased flow signals in the A1 segment. Frame 3: After 15 minutes of infusion, minimal flow signals are seen with MES in the M1 MCA and the beginning of TICA/A1 ACA recanalization. Frame 4: After 35 minutes of infusion, complete M1 MCA recanalization is seen with low resistance flow, stenotic mean flow velocities (104 cm/s) in the TICA, and low resistance flow with MES in the A1 ACA. Frame 5: After 42 minutes of infusion, are seen developing M1 MCA and TICA reocclusion with dampened flow signals and decreased mean flow velocities compared to frame 4. Digital subtraction angiography (DSA): DSA immediately after IV tPA infusion showed TICA “T”-type occlusion (lateral view) and a clot in the proximal ICA with a residual stenosis. Peak = peak systolic velocity; mean = mean flow velocity; PI = pulsatility index; ED = end diastolic flow; RI = resistance index.

At 10 minutes after tPA bolus, the TCD study suggested terminal ICA recanalization with resumption of end-diastolic flow in the A1 ACA, followed shortly by improvement in the patient’s level of consciousness. At 15 minutes of infusion, microembolic signals were heard in the M1 MCA accompanied with proximal M1 segment recanalization and resumption of low resistance end-diastolic flow toward the lenticulostriate perforating arteries (see the figure, frame 3). Clinically, the patient’s right leg began to move, followed by antigravity strength in the distal arm and improved facial weakness (NIHSS score 18).

The TCD study indicated a continuing recanalization of the A1 ACA and proximal M1 MCA at 20 minutes, followed by resolution of gaze preference and continued improvement in right-sided weakness by 30 minutes (NIHSS score 15). At 35 minutes, the TCD image was consistent with complete M1 MCA recanalization with multiple microembolic signals suggesting continuing clot dissolution (see the figure, frame 4). By 37 minutes, the patient could lift her arm with a mild drift, verbalize simple words, and follow axial and extra-axial commands (NIHSS score 8).

At 42 minutes of infusion, the TCD image was consistent with a reocclusion of the M1 MCA and dampening of the terminal ICA flow (see the figure, frame 5). At 44 minutes, the patient rapidly became drowsy and resumed her eye deviation, global aphasia, and right hemiplegia (NIHSS score 24). At the end of tPA infusion, the TCD image was compatible with a terminal ICA “T”-type occlusion (signal deterioration similar to frame 2). A post-tPA CT scan was unchanged.

Immediate angiography revealed a severe stenosis in the proximal ICA and a complete terminal ICA T-type occlusion with no flow in the M1 MCA and A1 ACA segments (see the figure, frame 5, digital subtraction angiography). Under an approved experimental protocol, the patient was given an additional 6 mg of intra-arterial tPA with mechanical clot disruption leading to complete distal ICA, proximal M1 MCA, and A1 ACA recanalization with a remaining distal M1 MCA occlusion and a proximal ≥ 80% ICA stenosis. Diagnostic workup showed no other cause for stroke. At 2 weeks, the patient’s major deficits included aphasia and arm plegia (NIHSS 18).

Discussion.

Although limited to a single case, continuous patient monitoring during tPA infusion suggested that arterial recanalization and reocclusion preceded all changes in the neurologic examination. Gradual MCA or terminal ICA recanalization preceded a stepwise resolution of clinical deficits (leg, followed by arm and face, then eye deviation), only to have the symptoms return in the reverse order when the terminal ICA reoccluded. The brief delay in clinical worsening may be attributed to a failure of tenuous collateral flow or a progression from near occlusion to complete occlusion, which may be below the resolution of TCD.

In the absence of another identifiable cause, the most likely stroke mechanism for this patient was artery-to-artery embolization from a proximal ICA thrombus. Microembolic signals can originate from carotid sources,5 suggesting possible reocclusion by pieces of a proximal clot, possibly liberated by tPA.

In our studies, at least 15% of all patients having a stroke experience deterioration,4 almost always in the setting of persisting arterial occlusion or reocclusion.6 Other causes of deterioration such as reperfusion, edema, or hemorrhage3 may preclude subsequent intra-arterial rescue, whereas patients with reocclusion may comprise a target group for intra-arterial thrombolysis or mechanical clot disruption.

In one trial of intraarterial thrombolysis, 56% of patients with a NIHSS range of 10 to 14 did not have an angiographically identifiable clot in the arterial distribution relative to their symptoms.7 Therefore, the ability to predict a clot and its location clinically may benefit from noninvasive vascular screening when considering patients for intraarterial therapy. Although TCD is operator dependent and not widely used in the acute setting, our bedside TCD study rapidly and reliably located a clot without delay or risk.4,6⇓ Our case showed the feasibility of noninvasive monitoring of tPA therapy that links reocclusion to clinical deterioration.