Neuroradiology (2000) 42: 444±447 Ó Springer-Verlag 2000 T. Neumann-Haefelin H. J. Wittsack F. Wenserski T. Q. Li M. E. Moseley M. Siebler H. J. Freund Received: 26 July 1999 Accepted: 20 September 1999 ) T. Neumann-Haefelin ( ) Department of Neurology (ZNN 95-1), Goethe-Universität, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany Tel.: ++49-69-630 16677 Fax: ++49-69-630 16842 e-mail: tnh@rz.uni-frankfurt.de T. Neumann-Haefelin × H. J. Wittsack × M. Siebler × H. J. Freund Department of Neurology, Heinrich-Heine University, Düsseldorf, Germany T. Neumann-Haefelin ´ T. Q. Li × M. E. Moseley Department of Radiology, Stanford University, Lucas MRS Center, MC: 5488, Palo Alto, CA 94305, USA H. J. Wittsack × F. Wenserski Institute of Diagnostic Radiology, Heinrich-Heine University, Düsseldorf, Germany D IA G N O S T I C NE UR OR A DI O LO G Y Diffusion- and perfusion-weighted MRI in a patient with a prolonged reversible ischaemic neurological deficit Abstract We report acute and follow-up diffusion- and perfusionweighted MRI (DWI, PWI) findings in a patient with a prolonged reversible ischaemic neurological deficit. PWI 12 h after the patient was last seen to be without symptoms revealed a large perfusion deficit in the left posterior MCA territory with a relatively inconspicuous and much smaller abnormality on DWI. Follow-up showed resolution of abnormalities on both DWI and PWI, and conventional MRI was normal, apart from a very slight abnormality, visible only on FLAIR images, at the centre of the initially DWI-positive region. These findings demonstrate the utility of PWI when be used in combination with DWI to investigate the pathophysiology of transient ischemic syndromes. Key words Magnetic resonance imaging, diffusion-weighted × Magnetic resonance imaging, perfusion × Stroke, acute Introduction Although there is a rapidly growing body of data on diffusion- and perfusion-weighted MRI (DWI, PWI) in acute stroke [1, 2], relatively little information is available on findings in transient ischaemic syndromes, in particular concerning PWI. In the largest study reported to date, Kidwell et al. [3] analysed DWI findings in patients with transient ischaemic attacks (TIAs). More than half the patients in this study had normal DWI and, in those with positive DWI studies the lesions were small and relatively subtle. This suggests that DWI alone may not be sufficient to identify the pathophysiology underlying transient ischaemic syndromes. We report a patient with a reversible ischaemic neurological deficit lasting over 24 h (prolonged reversible ischaemic neurological deficit; PRIND), with extensive abnormalities on PWI and only relatively minor changes on DWI, who made a full clinical and nearly complete tissue recovery on follow-up MRI. 445 Fig. 1 In the acute phase (top), diffusion-weighted imaging (DWI) revealed a relatively inconspicuous lesion (arrow) with a decrease in apparent diffusion coefficient, but no infarct on T 2-weighted images (T 2W). A perfusion deficit was observed on all acute perfusion-weighted imaging maps: MTT mean transit time, CBF cerebral blood flow, CBV cerebral blood volume. The perfusion deficit was substantially larger than the abnormality on DWI, in particular on the MTT and CBF maps. Follow-up at 1 week (7 d, bottom) showed complete resolution of the abnormalities on DWI and PWI Case report A 71-year-old man with a history of hypertension and mild hypercholesterolaemia was admitted 2 h after waking up with severe, predominantly sensory aphasia and mild right hemiparesis. The patient had last been seen to be without symptoms 12 h before presentation. His European Stroke Scale (ESS) score was 73/100. MRI was obtained 30 min after his arrival at the emergency department, using a 1.5-T system. The protocol included fast spinecho T 2-weighted images, a single-shot echo-planar (EPI) diffusion-weighted sequence (DWI) and perfusion-weighted imaging (PWI). DWI was performed using TE (echo time) 100 ms; b 0 and 1000 s/mm2; field of view (FOV) 240 mm; matrix 96 ” 128; slice thickness 5 mm, gap 1.5 mm; 20 slices. PWI was obtained using a gradient-echo EPI technique (TE 54 ms; FOV 240 mm; matrix 128 ” 128) with 40 consecutive measurements in each of 12 slices at intervals of 2 s during injection of 0.2 mmol/kg gadolinium-DTPA. PW images were processed to generate maps of relative mean transit time (rMTT), relative cerebral blood flow (rCBF) and relative cerebral blood volume (rCBV) after deconvolving the tissue concentration-time curve with an arterial input function, obtained by manually choosing pixels over the middle cerebral artery (MCA) of the unaffected hemisphere [4]. PWI revealed a large perfusion deficit in the left posterior MCA territory, in which abnormal regions could be identified on the rMTT, rCBF, and rCBV maps (Fig. 1). The volume of the abnormal regions, and the difference in signal intensity between the two sides were calculated for both affected cortical regions and white matter (Table 1). The abnormal region was slightly larger on the MTT maps than on the rCBF maps and substantially larger than on the rCBV maps. The perfusion deficit was clearly less severe in the cortex than in the underlying white matter region. DWI showed a relatively inconspicuous diffuse increase in signal, confined to white matter, within the perfusion deficit. The apparent diffusion coefficient (ADC) in the abnormal area was decreased to 60 % of that of the corresponding contralateral area. The T 2-weighted images were normal. Investigation revealed a patent foramen ovale, with right-toleft shunting, shown by both transesophageal echocardiography and transcranial ultrasound, indicating that a paradoxical embolism may have underlain the cerebral ischaemic event. There was no evidence of current deep venous thrombosis. Other cardiac investigations and extracranial Doppler ultrasound yielded normal results. Since the patient was not eligible for i. v. thrombolysis (due to the unknown time of onset) he received only standard therapy (i. v. heparin, blood pressure management) and no neuroprotective agents or other investigational drugs. Several hours after the initial MRI study he started to improve clinically. The next morning only minor amnesic aphasia was still detectable and 4 days later he had returned to normal. The follow-up MRI studies obtained at 7 days (T 2-weighted images, DWI and PWI) and 3 months (T 2-weighted spin-echo and FLAIR images, DWI) showed resolution of the DWI and PWI abnormalities and no residual lesion was seen on fast spin-echo T 2-weighted images. However, there was slightly increased signal on the FLAIR images at the centre of the initially DWI-positive region (Fig. 2). 446 Table 1 Analysis of abnormalities on MRI Diffusionweighted imaging Regional mean transit time Lesion volume (ml)a 6.3 76.2 Signal intensity ratio between cerebral hemispheresb Cortex 0.97 2.37 White matter 0.56 1.76 Regional cerebral blood flow Regional cerebral blood volume 51.3 12.7 0.45 0.27 0.97 0.53 a Determined by manually tracing the edge of the regions and multiplying the calculated area by the slice thickness plus interslice gap b Ischaemic region/corresponding region of unaffected hemisphere Fig. 2 At 30 days FLAIR images revealed a slight increase in signal in the area which appeared abnormal on acute DWI, but no frank infarct. No infarct was found on T 2-weighted images Discussion We demonstrate that PWI can be useful in addition to DWI for the assessment of the pathophysiology underlying reversible ischaemic syndromes. The acute perfusion abnormality in our patients was far more extensive than the relatively subtle DWI lesion. In fact, the latter ± although unequivocal ± was appreciated only with hindsight, and might have been overlooked without knowledge of the PWI results. Previous studies using either conventional MRI or CT in patients with clinically reversible ischaemic syndromes have shown that 2±48 % have relevant infarcts at follow-up, despite clinical recovery [3, 5]. It has been suggested that these conditions may be more appropriately classified as ªcerebral infarcts with transient symptoms (CITS)º [6]. In the remainder of patients both conventional MRI and CT remain normal. DWI is known to be much more sensitive to ischaemic tissue injury, but the percentage of patients with a TIA and abnormal findings on DWI within 3 days of the onset of symptoms has also been reported to be less than 50 %, and in some of those with abnormal DWI no relevant infarct was found at follow-up [3]. Our case suggests that combined PWI and DWI may be considerably more sensitive than DWI alone in transient ischaemic syndromes. There are theoretical considerations supporting this view. Abnormalities on DWI are generally thought to be a marker of severe ischaemic tissue injury, and in practice these regions usually evolve into infarcts [7±9]. PWI, on the other hand, has been shown to be sensitive enough to detect less severely hypoperfused regions, which may have impaired function, contributing to the neurological deficit, but are still viable [10]. It is therefore reasonable to assume that most patients with truly reversible ischaemic damage will show substantially more extensive abnormalities on PWI than on DWI (or no abnormality at all on DWI), while they are clinically affected. The acute DWI and PWI findings in our patient did not allow us to predict the benign clinical course with full recovery. Even patients with acute abnormalities on PWI, but not on DWI may develop relevant infarcts and permanent clinical deficits [11]. However, both of these patterns (PWI deficit > DWI lesion or no DWI lesion) appear to be compatible with potentially (almost) complete tissue recovery, provided spontaneous or treatment-induced reperfusion occurs early enough. Acknowledgement Dr. Neumann-Haefelin was sponsored by a grant from the DFG (Ne 569/2-1). 447 References 1. Baird AE, Warach S (1998) Magnetic resonance imaging of acute stroke. J Cereb Blood Flow Metab 18: 583±609 2. Fisher M, Albers GW (1999) Applications of diffusion-perfusion magnetic resonance imaging in acute ischemic stroke. Neurology 52: 1750±1756 3. Kidwell CS, Alger JR, Di Salle F, Starkman S, Villablanca P, Bentson J, Saver JL (1999) Diffusion MRI in patients with transient ischemic attacks. Stroke 30: 1174±1180 4. Ostergaard L, Johannsen P, Host-Poulsen P, Vestergaard-Poulsen P, Asboe H, Gee AD, Hansen SB, Cold GE, Gjedde A, Gyldensted C (1998) Cerebral blood flow measurements by magnetic resonance imaging bolus tracking: comparison with [(15)O]H2O positron emission tomography in humans. J Cereb Blood Flow Metab 18: 935±940 5. Fazekas F, Fazekas G, Schmidt R, Kapeller P, Offenbacher H (1996) Magnetic resonance imaging correlates of transient cerebral ischemic attacks. Stroke 27: 607±611 6. Waxman SG, Toole JF (1983) Temporal profile resembling TIA in the setting of cerebral infarction. Stroke 14: 433±437 7. Lecouvet FE, Duprez TP, Raymackers JM, Peeters A, Cosnard G (1999) Resolution of early diffusion-weighted and FLAIR MRI abnormalities in a patient with TIA. Neurology 52: 1085±1087 8. Marks MP, de Crespigny A, Lentz D, Enzmann DR, Albers GW, Moseley ME (1996) Acute and chronic stroke: navigated spin-echo diffusion-weighted MR imaging. Radiology 199: 403±408 9. Lansberg MG, O'Brien MW, Norbash AM, Moseley ME, Morell M, Albers GW (1999) MRI abnormalities associated with partial status epilepticus. Neurology 52: 1021±1027 10. Neumann-Haefelin TW, Wittsack HJ, Wenserski F, Siebler M, Seitz RJ, Moedder U, Freund H-J (1999) Diffusion- and perfusion-weighted MRI: the DWI/PWI mismatch region in acute stroke. Stroke 30: 1591±1597 11. Ay H, Buonanno FS, Rordorf G, Schaefer PW, Schwamm LH, Wu O, Gonzalez RG, Yamada K, Sorensen GA, Koroshetz WJ (1999) Normal diffusion-weighted MRI during strokelike deficits. Neurology 52: 1784±1792