ISSN 0017-8748 doi: 10.1111/head.13472 Published by Wiley Periodicals, Inc. Headache  © 2019 American Headache Society Brief Communications Migrainous Infarction in a Patient With Sporadic Hemiplegic Migraine and Cystic Fibrosis: A 99mTc-HMPAO Brain SPECT Study Valentina Mancini, MD ; Giulio Mastria, MD; Viviana Frantellizzi, MD; Patrizia Troiani, MD, PhD; Stefania Zampatti, MD; Stefania Carboni, MSc; Emiliano Giardina, MSc, PhD; Rosa Campopiano, MSc, PhD; Stefano Gambardella, MD, PhD; Federica Turchi, MD; Barbara Petolicchio, MD, PhD; Massimiliano Toscano, MD, PhD; Mauro Liberatore, MD; Alessandro Viganò, MD, PhD; Vittorio Di Piero, MD, PhD Genetic mutations of sporadic hemiplegic migraine (SHM) are mostly unknown. SHM pathophysiology relies on cortical spreading depression (CSD), which might be responsible for ischemic brain infarction. Cystic fibrosis (CF) is caused by a monogenic mutation of the chlorine transmembrane conductance regulator (CFTR), possibly altering brain excitability. We describe the case of a patient with CF, who had a migrainous stroke during an SHM attack. A 32-year-old Caucasian male was diagnosed with CF, with heterozygotic delta F508/unknown CFTR mutation. The patient experiences bouts of coughing sometimes triggering SHM attacks with visual phosphenes, aphasia, right-sided paresthesia, and hemiparesis. He had a 48-hour hemiparesis triggered by a bout of coughing with hemoptysis, loss of consciousness, and severe hypoxia-hypercapnia. MRI demonstrated transient diffusion hyperintensity in the left frontal-parietal-occipital regions resulting in a permanent infarction in the primary motor area. Later, a brain perfusion SPECT showed persistent diffuse hypoperfusion in the territories involved in diffusion-weighted imaging alteration. Migrainous infarction, depending on the co-occurrence of 2 strictly related phenomena, CSD and hypoxia, appears to be the most plausible explanation. Brain SPECT hypoperfusion suggests a more extensive permanent neuronal loss in territories affected by aura. CF may be then a risk factor for hemiplegic migraine and stroke since bouts of coughing can facilitate brain hypoxia, triggering auras. Key words: cortical spreading depression, cystic fibrosis, aura, SHM, CFTR Abbreviations: CF cystic fibrosis, CSD cortical spreading depression, DWI diffusion-weighted imaging, ED emergency department, FHM familial hemiplegic migraine, HM hemiplegic migraine, NSAID nonsteroidal antiinflammatory drugs, PFO patent foramen ovale, RCVS reversible cerebral vasoconstriction syndrome, SHM sporadic hemiplegic migraine (Headache 2019;0:1-6) From the Department of Human Neurosciences, Sapienza University of Rome, Rome, Italy (V. Mancini, G. Mastria, F. Turchi, B. Petolicchio, M. Toscano, A. Viganò, and V. Di Piero); Department of Radiological, Oncological and Anatomopathological Sciences, Sapienza University of Rome, Rome, Italy (V. Frantellizzi and M. Liberatore); Department of Pediatrics, Sapienza University of Rome, Rome, Italy (P. Troiani); Genomic Medicine Laboratory UILDM, IRCCS Santa Lucia Foundation, Rome, Italy (S. Zampatti, S. Carboni and E. Giardina); Molecular Genetics Center, IRCCS Neuromed, Pozzilli, Italy (S. Zampatti, R. Campopiano, and S. Gambardella); Department of Biomedicine and Prevention, University of Rome “Tor Vergata,” Rome, Italy (E. Giardina). Address all correspondence to V. Mancini, Department of Human Neurosciences, Sapienza University of Rome, Viale dell’Università 30 Rome 00185, Italy, email: v.mancini92@gmail.com Accepted for publication October 29, 2018. 1 2 INTRODUCTION Hemiplegic migraine (HM) is a form of migraine with an aura including motor weakness that can be either familial (FHM), thereby inherited with autosomal dominant genetics, or sporadic (SHM), the transmission of which remains mostly unclear.1 FHM forms are caused by mutations in membrane channels genes, such as CACNA1A (Cav2.1), ATP1A2 (Na+/K+ ATPase pump), and SCN1A (Nav1.1).1 The inheritance of sporadic forms may be due to a ­typical FHM mutation or to the development of a new u ­ nknown mutation, either located in the same or other genes.2 Only a few cases of SHM have typical mutations, whereas in the majority of the cases the g­enetic alteration is unknown. SHM patients with typical FHM mutations, especially those with CACNA1A mutation, tend to have an earlier onset, a worse phenotype, and a higher rate of episodes of impaired consciousness and stroke.3 Although many aspects of the HM-stroke association remain unexplained, the principal mechanisms have been analyzed in an elegant review by Mawet et al.4 In brief, cortical spreading depression (CSD), due to its metabolic and hemodynamic deregulation on the brain tissue, can be a predisposition factor to stroke. Migrainous infarction is an ischemic stroke associated with typical migraine aura symptoms and neuroimaging evidence of infarction in the brain territory involved in aura symptoms, mostly affecting the posterior cerebral circulation.4 Alternatively, an increased risk can be derived from migraine treatments or comorbidities (eg, depression, reversible cerebral vasoconstriction syndrome – RCVS) and common genetic predisposition to other independent risk factors for stroke (eg, cervical artery dissection or hypercoagulability and endothelial dysfunction) that are frequent in migraineurs.4 Cystic fibrosis (CF) is the most common autosomal recessive disorder worldwide. It is a life-limiting disease, caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene on chromosome 7. Alterations in CFTR result in a Conflict of Interest: The authors have no conflict of interest related to the study. Patient Consent: The patient gave his informed consent in written form. Funding: None. Dr. Viganò and Prof. Di Piero equally supervised. Month 2019 dysfunction of the anion (Cl−) transport and impaired mucociliary clearance, determining mucus retention and chronic infections with subsequent inflammation in the airways. CF-related morbidity and mortality are due to respiratory impairment, deriving from bronchiectasis, hemoptysis, recurrent respiratory infections, and small airways obstruction.5 As the concurrence of unrelated genetic diseases is unlikely to happen by pure statistical chance, disentangling the relationship between HM and CF is somewhat challenging. CASE REPORT We present the case of a 32-year-old Caucasian male patient diagnosed with cystic fibrosis at birth due to a heterozygotic mutation of the CFTR gene: he had a typical delta F508 mutation on one allele and an unknown mutation on the other. He progressively developed chronic obstructive pulmonary disease (COPD) with forced expiratory volume less than 40% and pulmonary hypertension (pulmonary artery pressure 29 mmHg), so that he needs oxygen therapy (1 L/min O2) during nighttime and physical efforts. Moreover, he developed diabetes for pancreatic insufficiency and had recurrent episodes of pneumonia occasionally associated with hemoptysis. Since 2012, his bouts of coughing sometimes triggered transient episodes characterized by a quick succession of phosphenes, one-sided paresthesia, motor aphasia, and homolateral hemiparesis. The symptoms always affected the right half of the body and usually had a short length of some minutes except for hemiparesis that could last up to some hours. After that, he usually experienced migraine headaches. During diagnostic work-up for these transient episodes, a large diameter patent foramen ovale (PFO) was found and surgically closed. However, he continued to have similar attacks once or twice per year, with hemiparesis that worsened in duration and subsequent leg involvement. In April 2017, due to the aggravation of COPD, the patient experienced a bout of coughing longer than usual, later accompanied by copious hemoptysis. Concomitantly, he had a typical migraine ­attack with aura, culminating in a 15-minute long loss of consciousness. Hemiparetic symptoms lasted 48 hours, even after the recovery of the patient. He was promptly Headache 3 Fig. 1.—(A,B) – DWI-MRI, performed at admission (A) and at day 10 (B), shows a diffuse asymmetrical altered signal involving bilateral occipital areas and parietal-frontal regions on the left. (C,D) – FLAIR (fluid-attenuated inversion recovery) MRI shows a subacute lesion in the left primary motor area at admission (C) and its permanence at day 10 (D). transferred to the emergency department, where an arterial blood gas test showed severe ­hypoxia with hypercapnia so that oxygen therapy was administered. No seizures were reported by the observers and paramedics. He immediately underwent an MRI with arterial angiographic sequences (MRA), revealing a wide area of diffusion restriction in ­ diffusion-weighted imaging (DWI), encompassing left occipital-­parietalfrontal regions (Fig. 1A,B). No arterial abnormalities were reported on MRA. During hospitalization, he underwent a second MRI (10 days later) showing a marked resolution of the DWI alterations, except for 4 a persistent ischemic lesion in the cortical motor areas (Fig. 1C,D). The stroke work-up included a CT scan of epiaortic vessels, venous Doppler ultrasound to exclude thromboembolism, transthoracic echocardiography, and hypercoagulable screening, all of which resulted negative. In addition, PFO was found to be unmodified even after closure. Two months later, he was admitted to our Headache Center and was diagnosed with HM. The patient reported no previous use of triptans, dihydroergotamine, or illicit substance. A venous MRI performed to complete the diagnostic process was unremarkable. Neurological examination revealed mild rightside hyperesthesia, mild weakness with increased right bicipital and tricipital reflexes, and homolateral Babinski sign. Finger tapping test was slower with the right hand. Moreover, he was under antihemorrhagics, such as vitamin K and tranexamic acid, so the routine use of nonsteroidal antiinflammatory drugs was not possible. No other episodes occurred until now; however, preventive therapy with calcium channel blockers has been proposed. Genetic screening for ATP1A2, CACNA1A, and SCN1A genes was performed with MiSeq platform (Illumina, San Diego, CA, USA). Annotation and characterization of the variants were performed with Variant Studio (Illumina). Sequence data were manually examined and visualized with an Integrative Genomics Viewer v.2.3 (Broad Institute, Cambridge, MA, USA). The analysis showed no FHM mutations. Due to the volume discrepancy between the area altered at the DWI and the final lesion, the patient underwent perfusion brain single photon emission computed tomography (BP-SPECT) with 740MBq of 99mTc-HMPAO (Ceretec™, GE Healthcare, Oslo, Norway) to highlight a possible selective neuronal damage with glial preservation in the parieto-occipital area. BP-SPECT was performed 15 minutes later by a dual-head gamma-camera (Infinia®, GE Healthcare, Chicago, IL, USA) equipped with LEHR collimators. Image reconstruction was performed using filtered back projection with a low-pass Butterworth filter and a ramp filter. The BP-SPECT showed significant permanent hypoperfusion encompassing a large part of the left fronto-parieto-occipital brain region involved in prolonged aura (see Fig. 2), indicating a selective neuronal loss larger than the visible damage. Month 2019 DISCUSSION Although extremely rare, the linkage between cystic fibrosis and sporadic HM has already been reported in the literature.6 A previous report showed 2 unrelated patients diagnosed with cystic fibrosis experiencing HM symptoms after severe bouts of paroxysmal cough, similar to our patient. Notably, in one of these cases, diffuse ischemic watershed lesions were found in bilateral parietal and occipital and right frontal regions.6 The most probable cause of the stroke in our patient is migrainous infarction, as alternative explanations seem less plausible. We ruled out paradoxical embolism because PFO resulted closed at echocardiography, no epiaortic vessels abnormalities were found, and RCVS was excluded as no sign of vasospasm was detected during hospitalization. Mitochondrial diseases were unlikely since patients had none of the typical features, such as optic neuropathy and sensorineural deafness. Finally, our patient has secondary diabetes, which might predispose to stroke; however, the location and the volume of the ischemia are not usual for a small vessel disease. Migrainous infarction is a stroke occurring in brain regions involved in aura symptoms during a persistent aura. In our case, an acute MRI showed DWI restriction, compatible with CSD,7 which reversed in 10 days, except for a small eloquent lesion in the primary motor area. Moreover, beyond the visible lesion, diffuse hypoperfusion was detected in brain regions affected by CSD, strengthening the idea that neuronal loss entirely involved penumbra territories covered by aura progression.8 Although we cannot completely rule out hypoxia as a primary cause of stroke, accumulating evidence suggests that migraine-related hyperexcitability may sensitize brain tissue to ischemic damage, so that the migraineurs’ brain is more prone to ischemia, ­resulting in a higher rate of CSD and brain infarction ­occurring after mild hypoxia.4 Not only is hypoxia a well-known trigger factor for aura,9 but in mice with CACNA1 gene mutation, even mild hypoxia can readily prompt a ­cerebral infarction, and the expansion of the core infarct is far more rapid than in wild-type mice.10 Therefore, we propose that hypoxia subsequent to bouts of coughing facilitated the occurrence of CSD and, due to its extended length, also migrainous infarction. 5 Headache Fig. 2.— 99 mTc-HMPAO perfusion brain SPECT scans in transaxial (A), sagittal (B), and coronal (C) reconstructions. SPECT images show brain hypoperfusion in the left occipital region (A), left frontal region (B), and left parietal region (C). The genetic relationship between cystic fibrosis and HM remains elusive since the gene research for FHM was negative in our patient, like the majority of patients with sporadic forms of HM. Moreover, a combined inheritance of CF and FHM seems unlikely because the respective genes are located on separate chromosomes and the former is an autosomal recessive disease, while the latter is a dominant one. However, even though the influence of CFTR on brain excitability is supported by little evidence, CFTR regulates a chloride channel, which is widely expressed in the brain cortex, diencephalon, midbrain, pons, and medulla oblongata.11 It has further been demonstrated that CFTR blockade produces impairment of glycine-mediated inhibition of motor neurons, suggesting that CFTR may indeed regulate cortical inhibitory activity.12 Therefore, its alteration might produce a state of brain hyperexcitability and then generate a favorable ground for CSD occurrence. Taken together, our findings suggest that CF patients might have an increased CSD proneness, both directly, due to the CFTR effect on brain excitability, and indirectly for the effect of hypoxia as a CSD trigger. Therefore, CF may be considered as an additional risk factor for aura and HM attacks, even without a family history of FHM. Moreover, these patients might have a higher chance of developing migrainous infarction due to the reciprocal causality relationship between CSD, hypoxia, and ischemia. STATEMENT OF AUTHORSHIP Category 1 (a) Conception and Design Valentina Mancini, Giulio Mastria, Alessandro Viganò, and Vittorio Di Piero 6 (b) Acquisition of Data Viviana Frantellizzi, Mauro Liberatore, Patrizia Troiani, Stefania Zampatti, Stefania Carboni, Rosa Campopiano, Federica Turchi, and Barbara Petolicchio (c) Analysis and Interpretation of Data Valentina Mancini, Giulio Mastria, Alessandro Viganò, Viviana Frantellizzi, Mauro Liberatore, Stefania Zampatti, Emiliano Giardina, Stefano Gambardella, Patrizia Troiani, Massimiliano Toscano, and Vittorio Di Piero Category 2 (a) Drafting the Manuscript Valentina Mancini, Giulio Mastria, and Alessandro Viganò (b) Revising It for Intellectual Content Alessandro Viganò, Valentina Mancini, Giulio Mastria, Viviana Frantellizzi, Patrizia Troiani, Stefania Zampatti, and Vittorio Di Piero Category 3 (a) Final Approval of the Completed Manuscript Valentina Mancini, Giulio Mastria, Viviana Frantellizzi, Patrizia Troiani, Stefania Zampatti, Stefania Carboni, Emiliano Giardina, Rosa Campopiano, Stefano Gambardella, Federica Turchi, Barbara Petolicchio, Massimiliano Toscano, Mauro Liberatore, Alessandro Viganò, and Vittorio Di Piero Acknowledgments: The authors want to thank Dr. Irene Turinese and Dr. Marta Puma for their help in the clinical management of the patient. REFERENCES 1. Russell MB, Ducros A. Sporadic and familial hemiplegic migraine: Pathophysiological mechanisms, clinical characteristics, diagnosis, and management. Lancet Neurol. 2011;10:457-470. Month 2019 2. Thomsen LL, Eriksen MK, Roemer SF, Andersen I, Olesen J, Russell MB. A population-based study of familial hemiplegic migraine suggests revised diagnostic criteria. Brain. 2002;125(Pt 6):1379-1391. 3. Riant F, Ducros A, Ploton C, Barbance C, Depienne C, Tournier-Lasserve E. De novo mutations in ATP1A2 and CACNA1A are frequent in early-­ onset sporadic hemiplegic migraine. Neurology. 2010;75:967-972. 4. Mawet J, Kurth T, Ayata C. Migraine and stroke: In search of shared mechanisms. Cephalalgia. 2015;35:165-181. 5. Elborn JS. Cystic fibrosis. Lancet. 2016;19;388: 2519-2531. 6. Rao DS, Infeld MD, Stern RC, Chelimsky TC. Cough-induced hemiplegic migraine with impaired consciousness in cystic fibrosis. Pediatr Pulmonol. 2006;41:171-176. 7. Busch E, Hoehn-Berlage M, Eis M, Gyngell ML, Hossmann KA. Simultaneous recording of EEG, DC potential and diffusion-weighted NMR imaging during potassium induced cortical spreading depression in rats. NMR Biomed. 1995;8:59-64. 8. Baron J-C, Yamauchi H, Fujioka M, Endres M. Selective neuronal loss in ischemic stroke and cerebrovascular disease. J Cereb Blood Flow Metab. 2014;34:2-18. 9. Britze J, Arngrim N, Schytz HW, Ashina M. Hypoxic mechanisms in primary headaches. Cephalalgia. 2017;37:372-384. 10. Eikermann-Haerter K, Lee JH, Yuzawa I, et al. Migraine mutations increase stroke vulnerability by facilitating ischemic depolarizations. Circulation. 2012;125:335-345. 11. Guo Y, Su M, McNutt MA, Gu J. Expression and distribution of cystic fibrosis transmembrane conductance regulator in neurons of the human brain. J Histochem Cytochem. 2009;57:1113-1120. 12. Morales FR, Silveira V, Damián A, Higgie R, Pose I. The possible additional role of the cystic fibrosis transmembrane regulator to motoneuron inhibition produced by glycine effects. Neuroscience. 2011;17:138-147.