Prediction of Early Reperfusion From Repeated Arterial Spin Labeling Perfusion Magnetic Resonance Imaging During Intravenous Thrombolysis
Background and Purpose—There are few in vivo data on the pathophysiology of reperfusion during systemic thrombolysis. We monitored the time course of cerebral perfusion changes in patients during thrombolysis with repeated arterial spin labeling perfusion magnetic resonance imaging.
Methods—Ten patients with proximal arterial occlusion within 4.5 hours after symptom onset were prospectively enrolled. All patients received intravenous thrombolysis during the magnetic resonance imaging examination. Repeated arterial spin labeling perfusion images were acquired during the 60-minute therapy and at follow-up after 24 to 72 hours. Clinical data, magnetic resonance imaging features, and cerebral perfusion changes were analyzed.
Results—Before thrombolysis, arterial spin labeling hypoperfusion and fluid-attenuation inversion recovery vascular hyperintensity in the territory of the occluded arteries were observed in all patients. In 5 patients, extensive arterial transit artifacts (ATA) developed in the hypoperfused area. The ATA corresponded with fluid-attenuation inversion recovery vascular hyperintensities. All 5 patients who developed extensive ATA in the hypoperfused area had complete reperfusion after thrombolysis, whereas the 5 without extensive ATA showed no or only partial reperfusion (P<0.01). The development of ATA preceded the normalization of tissue perfusion.
Conclusions—The development of ATA during thrombolysis is associated with early reperfusion after thrombolysis. arterial spin labeling assessment during intravenous thrombolysis has the potential to guide subsequent therapeutic strategies in patients with acute stroke.
The response to systemic thrombolysis varies across individuals, and there are few in vivo data on the sequence of events during thrombolysis. Therefore, it is difficult to estimate early on who will benefit from thrombolysis. Furthermore, although recent clinical trials have demonstrated the efficacy of mechanical thrombectomy in acute stroke,1 it is not yet established how to select patients for additional mechanical thrombectomy. Arterial spin labeling (ASL) is a noninvasive magnetic resonance imaging (MRI) method for measuring cerebral perfusion without contrast agent or radiation exposure. Thus, it can be performed repeatedly in a short period of time. Using repeated ASL perfusion imaging, we examined the dynamic perfusion process of revascularization during and early after intravenous thrombolysis. In addition, we explored whether early reperfusion can be predicted by perfusion patterns of ASL images during thrombolysis.
Materials and Methods
This study was performed at UniversitätsMedizin Mannheim, Germany. The local ethics committee approved the study. All subjects gave their written informed consent. Between February 2011 and January 2013, 27 patients with acute stroke eligible for systemic thrombolysis were enrolled. The study protocol was described previously.2 In brief, the patients were transferred into the MRI scanner after clinical examination and prepared for systemic thrombolysis by applying an MR-compatible monitoring and infusion system. In patients eligible for systemic thrombolysis, the therapy was carried out inside the MRI scanner. During the 60-minute infusion, repeated ASL perfusion images using Q2TIPS-FAIR with 3D-GRASE readout3 were acquired at 5-minute intervals (acquisition time, 1 minute 36 s). Follow-up MRI scans were performed at 60 minutes and at 24 to 72 hours after the initiation of thrombolysis. All ASL images were analyzed using 3DSRT software (Fujifilm RI Pharma, Japan) as described previously4 (details are available in the Methods section in the online-only Data Supplement). Significance was assessed by Fisher exact test.
Of the 27 patients enrolled, 10 patients with proximal arterial occlusion were analyzed. Flowchart is given in Figure I in the online-only Data Supplement. The Table shows the patients’ characteristics. Baseline MRI revealed fluid-attenuation inversion recovery vascular hyperintensities in the distal portion of the occluded arteries and ASL hypoperfusion in the corresponding territory in all patients.
Two patients showed a reperfusion during the 60-minute infusion (cases 1 and 2). Figures 1A, 1B, 2A, and 2B demonstrate cerebral perfusion changes during thrombolysis. At the beginning of thrombolysis, a marked tissue hypoperfusion was observed in the territory of occluded arteries. After ≈30 minutes, arterial transit artifacts (ATA) emerged in the hypoperfused area. ATA corresponded well with baseline fluid-attenuation inversion recovery vascular hyperintensities (Figures 1C and 2C) and extended gradually from the proximal to the distal portions of the corresponding fluid-attenuation inversion recovery vascular hyperintensities and then disappeared before reperfusion. Normalization of tissue perfusion was observed after the disappearance of ATA (Figures 1A and 2A). The follow-up MR angiography just after thrombolysis showed complete recanalization (Figures 1D and 2D). These 2 patients had neurological recovery early after thrombolysis, and there was no expansion of the infarct into the hypoperfused area.
Other 3 patients (cases 3–5) showed regional tissue hypoperfusion with extensive ATA throughout the period of infusion (Figure II in the online-only Data Supplement). They exhibited no reperfusion during thrombolysis; however, complete reperfusion and recanalization was observed at 24 to 72 hours after thrombolysis. Further 5 patients (cases 6–10) showed regional tissue hypoperfusion with no or only partial ATA during thrombolysis. They showed no or only partial reperfusion in the follow-up MRI examinations. Taken together, 5 patients with extensive ATA had complete reperfusion after thrombolysis, whereas the other 5 with no or partial ATA had no or only partial reperfusion (P<0.01).
Our study demonstrates for the first time the dynamic process of reperfusion during thrombolysis. The presence of ATA seems to be a useful marker of the onset of the reperfusion process.
The presence of ATA suggests delayed arrival of tagged blood to the affected vascular tissue.5 It has been reported that ATA was observed proximal or distal to the occluded artery in patients with acute stroke6,7 and moyamoya disease8 and may represent the presence of leptomeningeal collateral flow or anterograde residual flow that could augment recanalization by delivering a thrombolytic agent to the distal end of the clot. In this study, 2 patients showed a transient ATA development before reperfusion. This phenomenon indicates incomplete microcirculatory reperfusion before complete tissue reperfusion. After opening of the occluded artery with thrombolysis, clots could move downstream to obstruct distal arterial branches and capillaries.9 This microvascular obstruction model can explain the slow extension of ATA and subsequent tissue reperfusion.
Our study has limitations. First, we could not analyze the associations between ASL perfusion patterns and clinical outcome because of the small sample size. Second, because we used the same inversion time in the pulsed ASL sequence for all patients, perfusion measurements might be biased toward arteries and arterioles in patients with reduced cardiac output.10 In addition, ATA also affected the perfusion measurements. Therefore, we did not perform a quantitative cerebral blood flow analysis and focused on the individual cerebral perfusion changes during and after thrombolysis.
In conclusion, our data suggest that ASL imaging is a promising strategy for early prediction of response to systemic thrombolysis. ATA could be a good marker of early recanalization by intravenous thrombolysis. ASL imaging during intravenous thrombolysis might be useful for the selection of patients for additional mechanical thrombectomy.
Sources of Funding
This study was supported by grants from the Federal Ministry of Education and Research, Germany (01EV0702), and from SENSHIN Medical Research Foundation, Japan.
Dr Kern received speaker’s honoraria and travel funding from Boehringer Ingelheim. The other authors report no conflicts.
Guest Editor for this article was Liping Liu, MD, PhD.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.115.011482/-/DC1.
- Received September 11, 2015.
- Revision received September 11, 2015.
- Accepted September 29, 2015.
- © 2015 American Heart Association, Inc.
- Iwanaga T,
- Harada M,
- Kubo H,
- Funakoshi Y,
- Kunikane Y,
- Matsuda T.
- Chalela JA,
- Alsop DC,
- Gonzalez-Atavales JB,
- Maldjian JA,
- Kasner SE,
- Detre JA.
- Yoo RE,
- Yun TJ,
- Rhim JH,
- Yoon BW,
- Kang KM,
- Choi SH,
- et al
- Zaharchuk G,
- Do HM,
- Marks MP,
- Rosenberg J,
- Moseley ME,
- Steinberg GK.
- Dalkara T,
- Arsava EM.